Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

New ichthyosporean protists and their biology as inferred from molecular genetic and light and electron… Marshall, Wyth 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2010_spring_marshall_wyth.pdf [ 5.24MB ]
Metadata
JSON: 24-1.0070884.json
JSON-LD: 24-1.0070884-ld.json
RDF/XML (Pretty): 24-1.0070884-rdf.xml
RDF/JSON: 24-1.0070884-rdf.json
Turtle: 24-1.0070884-turtle.txt
N-Triples: 24-1.0070884-rdf-ntriples.txt
Original Record: 24-1.0070884-source.json
Full Text
24-1.0070884-fulltext.txt
Citation
24-1.0070884.ris

Full Text

 NEW ICHTHYOSPOREAN PROTISTS AND THEIR BIOLOGY AS INFERRED FROM MOLECULAR GENETIC AND LIGHT AND ELECTRON MICROSCOPIC STUDY    by   WYTH MARSHALL  B.Sc., University of Victoria, 1998      A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES  (Botany)    THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)    September 2009   © Wyth Marshall, 2009 ii   Abstract Precipitated by unexpected discoveries, this thesis is dedicated to the study of the biology of ichthyosporeans. While searching for undiscovered opisthokonts living osmotrophically in marine invertebrate digestive tracts, I established 177 cultures of ichthyosporeans. Ichthyosporeans are one of six understudied unicellular lineages related to the multicellular animals and fungi. Ichthyosporeans are comprised of approximately 30 genera but, prior to this thesis, only four genera had been cultivated. I identified and described three new genera and six species using microscopy and molecular phylogenetic techniques. Two, which I named Abeoforma whisleri and Pirum gemmata, were most closely related to divergent clone sequences and had no known relatives. My other four species, each isolated between eight and 126 times, were related to single isolates Sphaeroforma arctica and Pseudoperkisus tapetis, also found in marine invertebrates. I described one as the new genus and species Creolimax fragrantissima because of its amoeboid reproductive and dispersal stage and fragrance. The other three species were closely related and morphologically indistinguishable. To delimit species, I sequenced three loci from multiple isolates and applied a genealogical concordance species concept. Once delimited, I was able to describe variations in life cycle, morphology and a possible difference in host preference. Rather than adapt cytological techniques to describe the life cycle of S. tapetis, the most abundant species, I used population genetics to work in reverse. Absence of heterozygotes provided evidence for haploidy. Phylogenetic incongruence and a lack of support for linkage between two loci signified a history of recombination consistent with a sexual cycle. I described the ultrastructure of five species using high-pressure frozen cells from healthy, luxuriantly growing cultures. The quality of preservation allowed me to describe features that were new to ichthyosporeans, such as, spindle pole bodies and tubular extensions of the cell that penetrated the cell wall. These features were found in both clades and may have been present in the ancestral ichthyosporean. By combining genetic evidence for sex, observations of asexual reproduction in culture and collection frequencies I proposed a life cycle involving infrequent recombination within a predominantly asexual organism that infected invertebrate hosts indiscriminately via asexual endospores. iii  Table of contents Abstract ........................................................................................................................................................ ii Table of contents ........................................................................................................................................ iii List of tables............................................................................................................................................... vii List of figures ............................................................................................................................................ viii List of terms and their usage ..................................................................................................................... xi Acknowledgements ................................................................................................................................... xii Dedication ................................................................................................................................................. xiii Co-authorship statement ......................................................................................................................... xiv Chapter 1. Introduction to the Ichthyosporea: a unicellular lineage within the opisthokonta. ........... 1 Introduction ............................................................................................................................................... 1 The unicellular opisthokonta ................................................................................................................. 1 The Ichthyosporea ................................................................................................................................. 2 Thesis theme and objectives ..................................................................................................................... 4 References ................................................................................................................................................. 6 Chapter 2. Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. 11 Introduction ............................................................................................................................................. 11 Methods .................................................................................................................................................. 13 Animal collection and isolation of strains ........................................................................................... 13 DNA extraction, PCR amplification and sequencing ......................................................................... 14 Phylogenetic analysis .......................................................................................................................... 15 Light and electron microscopy ............................................................................................................ 16 Results ..................................................................................................................................................... 17 Diagnoses ............................................................................................................................................ 17 Phylogenetic inference ........................................................................................................................ 18 Light and electron microscopy ............................................................................................................ 19 Discussion ............................................................................................................................................... 22 Phylogenetic relationships .................................................................................................................. 22 Host-symbiont relationships ............................................................................................................... 23 Comparative morphology of C. fragrantissima vegetative cells ......................................................... 24 iv  Light microscopy ................................................................................................................................ 24 Ultrastructure ...................................................................................................................................... 26 Conclusion .......................................................................................................................................... 27 Acknowledgements ................................................................................................................................. 27 References ............................................................................................................................................... 39 Chapter 3. Population level analyses indirectly reveal cryptic sex and life-history traits of Pseudoperkinsus sp., an ecological model for basal Opishthokont lineage Ichthyosporea. ................ 46 Introduction ............................................................................................................................................. 46 Genetic evidence of chromosome copy number ................................................................................. 47 Genetic evidence of recombination in microeukaryotes ..................................................................... 47 Life-cycle and relationship with host .................................................................................................. 48 Methods .................................................................................................................................................. 49 Recombination, effective population size, and frequency of sex ........................................................ 52 Clonality and population structure ...................................................................................................... 53 Results ..................................................................................................................................................... 54 Population variation and lack of heterozygosity ................................................................................. 55 Recombination, effective population size, and frequency of sex ........................................................ 56 Clonality and population structure ...................................................................................................... 57 Discussion ............................................................................................................................................... 58 Evidence for both recombination and clonality .................................................................................. 59 Frequency of sex ................................................................................................................................. 60 Population size and host relationship .................................................................................................. 61 Conclusions ......................................................................................................................................... 63 Acknowledgements ................................................................................................................................. 63 References ............................................................................................................................................... 75 Chapter 4. Genealogical concordance species recognition criteria show cryptic species among morphologically similar sympatric isolates of Sphaeroforma (Ichthyosporea, Opisthokonta). ......... 83 Introduction ............................................................................................................................................. 83 Species recognition in microorganisms .............................................................................................. 83 Methods .................................................................................................................................................. 86 v  Collection of isolates and genetic analysis.......................................................................................... 86 Light and electron microscopy ............................................................................................................ 90 Results ..................................................................................................................................................... 91 Phylogenetic species recognition and genealogical concordance ....................................................... 91 Light microscopy and asexual reproduction ....................................................................................... 92 Ultrastructure ...................................................................................................................................... 94 Host specificity and gene flow between east and west Vancouver Island .......................................... 97 Diagnosis............................................................................................................................................. 98 Discussion ............................................................................................................................................. 100 Comparative morphology ................................................................................................................. 102 Conclusions ....................................................................................................................................... 104 References ............................................................................................................................................. 128 Chapter 5. Facing unknowns: Living cultures (Pirum gemmata gen. nov., sp. nov., and Abeoforma whisleri, gen. nov., sp. nov.) from invertebrate digestive tracts represent an undescribed clade within the unicellular opisthokont lineage ichthyosporea (Mesomycetozoa). ................................... 132 Introduction ........................................................................................................................................... 132 Methods ................................................................................................................................................ 133 Isolation, DNA extraction and phylogenetic analyses ...................................................................... 133 Light and electron microscopy .......................................................................................................... 135 Results ................................................................................................................................................... 136 Phylogenetic inference ...................................................................................................................... 136 Light and electron microscopy .......................................................................................................... 137 Diagnosis........................................................................................................................................... 143 Discussion ............................................................................................................................................. 145 Comparative morphology ................................................................................................................. 147 Conclusions ....................................................................................................................................... 150 References ............................................................................................................................................. 174 Chapter 6. Concluding chapter ............................................................................................................. 178 Summary ............................................................................................................................................... 178 Comparative morphology and speculation on ancestral traits .............................................................. 179 vi  Relationship of ichthyosporeans to other unicellular opisthokonts .................................................. 179 Relationships among clades within the ichthyosporea ...................................................................... 181 Life history characteristics of the ichthyosporea .................................................................................. 183 Life cycles ......................................................................................................................................... 183 Ecology ............................................................................................................................................. 185 References ............................................................................................................................................. 188 Appendices ............................................................................................................................................... 192 Appendix A. Abstracts for published or submitted manuscripts .......................................................... 192 Abstract: Chapter 2. .......................................................................................................................... 192 Abstract: Chapter 3. .......................................................................................................................... 193 Appendix B. Supplementary table for chapter 2. .................................................................................. 194 Appendix C. Supplementary tables and figures for chapter 3. ............................................................. 201 Appendix D. Supplementary tables and figures for chapter 4. ............................................................. 208 Appendix E. Supplementary table for chapter 5. .................................................................................. 216 Appendix F. Supplementary tables for chapter 6. ................................................................................. 222  vii  List of tables Table 2.1. Host animal, date and specific location, isolation and culturing details for animals and ichthyosporeans collected from Barkley Sound, Bamfield B.C. Canada. .................................... 28 Table 3.1. Estimates of effective population size times the frequency of sex (Ne) ŧŧ  based on several estimates of population recombination parameter ρ*. ...................................................... 64 Table 3.2. Effective population sizes (Ne) based on intergenic and four fold-degenerate sites given an upper and lower bound for mutation rate μ. ................................................................... 65 Table 3.3. Index of association between HSP70 and EFL alleles. ............................................... 66 Table 4.1. Hosts, collection locations, allele and haplotype designations for isolates from three Sphaeroforma species collected from British Columbia. ........................................................... 105 Table 4.2. Region of overlap and percent similarity of our 3 SSU-rDNA phylotypes and previously identified S. arctica and S. tapetis (formerly P. tapetis). .......................................... 108 Table 4.3. Genetic differentiation between subpopulations of S. tapetis (B.C.)......................... 109 Table 5.1. Morphological variation between P. gemmata and A. whisleri. ................................ 151 Table 6.1 Features of clades within the Ichthyophonida. ........................................................... 186 Table B.1. GenBank classification, accession numbers and collection details for taxa included in phylogenetic analysis. ................................................................................................................. 194 Table C.1. Collection details and haplotypes of Pseudoperkinsis sp. ........................................ 201 Table C.2. Primer and intron positions ....................................................................................... 205 Table C.3. Tajima‘s D, Fu and Li‘s D, and Fu and Li‘s F calculated using all 24 individuals1. 206 Table D.1. Host animal, collection date, number and specific location of host animals, and numbers of isolates from each SSU-rDNA phylotype. ................................................................ 208 Table E.1. Accession number and GenBank classification of sequences used for this study. ... 216 Table F.1. ATCC accession numbers of ichthyosporean strains from newly described species. 222 Table F.2. Comparison of morphological features and life cycles of ichthyophonida observed in this thesis (bold) to a selection of previously described species. ................................................ 223  viii  List of figures Figure 2.1.  Maximum likelihood PhyML tree comparing phylogenetic relationship of C. fragrantissima to other opisthokont SSU-rDNA sequences obtained from GenBank. ................. 30 Figure 2.2. One of six maximum likelihood trees (-ln 8281) comparing SSU-rDNA sequence of C. fragrantissima to all non-redundant sequences of ichthyosporeans belonging to the ichthyophonida. ............................................................................................................................. 31 Figure 2.3. General morphology of vegetative cells and of asexual reproduction. ...................... 32 Figure 2.4. Time series photography showing separation and differential maturation of a multiply lobed cell. ...................................................................................................................................... 33 Figure 2.5. DIC time-series of an actively moving amoeba. ........................................................ 34 Figure 2.6. Electron micrographs of C. fragrantissima cells during assimilative growth. ........... 35 Figure 2.7. Spindle pole body next to nucleus, directly under Golgi body. ................................. 36 Figure 2.8. Detail of cell wall showing membrane bound tubular extensions penetrating the inner cell wall. ........................................................................................................................................ 37 Figure 2.9. Spindle pole bodies between Golgi body and nucleus during interphase. ................. 38 Figure 3.1. Polymorphic sites from the amplified regions of the EFL and HSP70 loci shown for each of the 24 individuals. ............................................................................................................ 67 Figure 3.2. Genealogical networks of the HSP70 and EFL loci. .................................................. 68 Figure 3.3. Inter- and intra- locus recombination reconstructed using Splits Tree4. .................... 70 Figure 3.4. Range of possible frequencies of sexual reproduction in Pseudoperkinsus sp. ......... 71 Figure 3.5. Numbers of isolates belonging to each haplotype that were isolated from four host animal collections on the east side of Vancouver Island. ............................................................. 72 Figure 3.6. Asexual reproduction in Pseudoperkinsus sp. ............................................................ 73 Figure 3.7. Some combinations of alleles from the HSP70 locus (H) and the EFL locus (E) occurred at higher than expected frequencies, suggesting a certain level of clonality in the population. .................................................................................................................................... 74 Figure 4.1. Animal collection sites from Vancouver Island, British Columbia, Canada. .......... 110 Figure 4.2. Numbers of Sphaeroforma isolates cultured from host animals varied by year and by season and were not directly related to sampling effort. ............................................................ 111 Figure 4.3. Maximum parsimony tree comparing the phylogenetic relationship of our isolates to other spherical marine ichthyosporeans based on the SSU-rDNA sequences. ............................ 112 ix  Figure 4.4. Maximum parsimony trees showing the phylogenetic relationships of ITS and EFL sequences from new and previously isolated spherical ichthyosporeans. .................................. 113 Figure 4.5. Uncorrected pairwise genetic distances from the ITS region were greater between species pairs than within species................................................................................................. 114 Figure 4.6. Maximum parsimony tree with midpoint rooting from the ITS, EFL and HSP70 sequences, showing concordant support for the monophyly of four species of Sphaeroforma and contrasting lack of support for within species branching order. ................................................. 115 Figure 4.7. Morphology and spore release in Sphaeroforma species. ........................................ 117 Figure 4.8. Colony morphologies in Sphaeroforma. .................................................................. 118 Figure 4.9.  Whole cells of S. tapetis undergoing vegetative growth and maturation. ............... 119 Figure 4.10. Whole cells of S. nootkatensis undergoing vegetative growth and maturation. ..... 120 Figure 4.11. Closer view of internally coated fibrous vesicles and tubular extensions (MTBs) into the cell wall of S. tapetis and S. nootkatensis. ............................................................................ 121 Figure 4.12. Cell wall development and structure in S. tapetis. ................................................. 122 Figure 4.13. Details of S. nootkatensis ultrastructure. ................................................................ 124 Figure 4.14. Details of S. tapetis ultrastructure. ......................................................................... 125 Figure 4.15. Frequency and prevalence of isolation of each British Columbian Sphaeroforma species from four types of hosts.................................................................................................. 127 Figure 5.1. Abeoforma whisleri and Pirum gemmata, the two new species described in this paper, are the only living organisms known from a clade that consists otherwise of environmental clone sequences. ................................................................................................................................... 152 Figure 5.2. Vegetative cells of Pirum gemmata. ........................................................................ 154 Figure 5.3. Timelapse series showing three Pirum gemmata cells (A-C) undergoing sporogenesis in slide chambers......................................................................................................................... 155 Figure 5.4. Endospore release and morphology in Pirum gemmata. .......................................... 156 Figure 5.5. DAPI stained whole mounts of Pirum gemmata cells showing variation in nuclear size. ............................................................................................................................................. 157 Figure 5.6. Transmission electron micrographs showing sections through vegetative and reproductive cells of Pirum gemmata. ........................................................................................ 159 Figure 5.7. Spindle pole bodies and networks of smooth tubules were associated with the nucleus in Pirum gemmata. ...................................................................................................................... 161 x  Figure 5.8. Cell walls of Pirum gemmata showing membrane bound tubules arranged in discrete patches between the plasma membrane and the inner surface of the cell wall. .......................... 162 Figure 5.9. Examples of various cell shapes and modes of reproduction of Abeoforma whisleri. ..................................................................................................................................................... 164 Figure 5.10. Various shapes and forms of amoeboid stages of Abeoforma whisleri. ................. 165 Figure 5.11. Time-lapse series showing a plasmodial form of Abeoforma whisleri transitioning into a sphere. ............................................................................................................................... 166 Figure 5.12. Time series showing the formation of a colony of Abeoforma whisleri via plasmodial growth, budding and endospore release. .................................................................. 167 Figure 5.13. Examples of various cell shapes of Abeoforma whisleri. ....................................... 168 Figure 5.14. Detail of nuclei and other cytoplasmic contents of Abeoforma whisleri. ............... 170 Figure 5.15. Detail of the multilayered walls of Abeoforma whisleri. ....................................... 171 Figure 5.16. High magnification views of the inner wall layer with membrane bound tubules (MBTs) of Abeoforma whisleri. .................................................................................................. 173 Figure 6.1. Shared characteristics between ichthyophonida from different clades (I-IV). ......... 187 Figure C.1. Pseudoperkinsus sp. isolation sites from Vancouver Island, British Columbia, Canada......................................................................................................................................... 207 Figure D.1. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the EFL sequences from our 29 of our isolates, S. arctica and S. tapetis (Japan). ........................................................................................................................................ 212 Figure D.2. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the HSP70 sequences from our 29 of our isolates, S. arctica and S. tapetis (Japan). ........................................................................................................................................ 213 Figure D.3. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the ITS sequences from our 29 of our isolates, S. arctica and S. tapetis (Japan). ........................................................................................................................................ 214 Figure D.4. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the ITS sequences from our 29 of our isolates, S. arctica and S. tapetis (Spain and Japan). ....................................................................................................................... 215 xi  List of terms and their usage Commensual -an organism that derives benefit from but does not appear to cause harm towards its host. Distinguished from a mutualist (both organisms benefit) or a parasite (below), but included within the definition of a symbiont (below). Endospore – daughter cell with no apparent form of motility such as flagella or pseudopodia, usually, but not always with a cell wall. Distinct from zoospore (flagellated) or amoebae (move using prominent pseudopodia). Hypha –one or more intransient, tubular cells with cell walls and apical growth. Hypha-like –used to describe cells that resemble hyphae in shape when the presence and permanence of the cell wall is unknown, and when the permanence of the cell shape is unknown. For example some plasmodia may temporarily form structures that superficially look like hyphae. Multinucleate stage –refers to any cell with multiple nuclei. Osmotroph -A heterotrophic organism that obtains its nutrients by absorbing organic matter in solution from its surroundings. Parasite –an organism that derives benefit from and causes harm to its host. Included within the definition of a symbiont (below). Phylotype – the set of organisms with identical DNA sequences at a specified locus or specified loci. Plasmodium – a type of multinucleate cell with either flexible or non-existent cell walls, capable of changing shape and not typically spherical. Recombination – used here to describe the process by which genetic material is rearranged during meiosis. Recombination of loci on different chromosomes occurs during segregation and random reassortment of chromosomes during meiosis. Crossing over during meiosis can also occur resulting in recombination between loci on the same chromosome or interlocus recombination when the breakpoints occur within the genetic locus. This document distinguishes interlocus recombination from other forms of recombination. Sporogenesis –describes the processes involved in the conversion of a multinucleate parent cell into mature daughter cells. Symbiont – an organism that lives in close association with a host, includes commensual, parasitic or mutualistic relationships. xii  Acknowledgements I am grateful for the contributions of others during the progress of this thesis. For training and assistance with electron microscopy I thank D.J. McLaughlin, G. Celio and M. Padamsee and staff at bioimaging centers at the University of Minnesota and at the University of British Columbia. Thank-you to Berbee lab members for reading drafts and especially J. Dee and S. Lim for technical support. To students and staff and BMSC for company in the field collecting animals. Intellectual support from M. Berbee, and committee members R. DeWreede, B. Leander and L. Rieseberg . Also from K. Needham, S. Otto and W. Maddison. Funding was provided by NSERC, BMSC, and AFTOL.  xiii  Dedication A more personal note to those who kept me human: M. Marshall, S. McKim, G. Choe, M. Berbee, M. Kirby and B. Bravi, without you I would weigh less and have no clothes. To L. Marshall and C. Ware for transportation and my first laptop. Thank-you to Ike and Chako for chasing bears from the Chimo dump so I could get home. L. Eickhoff, thank-you for keeping the fire burning.  xiv  Co-authorship statement The identification, design and implementation of the research in this thesis were mostly completed by myself. M. Berbee contributed to the experimental design, analysis and final stages of manuscript preparation for published, submitted and to be submitted manuscripts (chapters 2- 5). Co-authors G. Celio and D. McLaughlin provided assistance with preparation and interpretations of electron microscopy for the description of Creolimax (chapter 2). All manuscripts were primarily written by me, with editorial advice from M. Berbee. 1  Chapter 1. Introduction to the Ichthyosporea: a unicellular lineage within the opisthokonta.  Introduction The unicellular opisthokonta The opisthokonts, defined by a motile stage with a posterior flagellum and plate-like mitochondrial cristae (Cavalier-Smith 1987), are the most studied of the eukaryote supergroups, largely due to the success and diversity of multicellular animals and fungi and the inclusion of our own species. Less attention has been directed towards the more anciently derived unicellular relatives of the animals and fungi. Initially opisthokonts included only two basal unicellular groups; the choanoflagellates and the chytrid fungi (Cavalier-Smith 1987; Baldauf and Palmer 1993; Wainright et al. 1993; Baldauf 1999). Adding to the choanoflagellates and chytrid fungi, six more lineages of unicellular opisthokonts have been established within the last 13 years due to the combined efforts of molecular phylogenetic based reclassification and new discoveries (Adl et al. 2005). The first indication of these undiscovered lineages came from a molecular study of an unclassified salmonid parasite, the ‗rosette‘ agent, which shared sequence similarity with the choanoflagellates (Kerk et al. 1995). Sequences from more taxa soon followed, leading to the establishment of the monophyletic group consisting of Dermocystidium, the rosette agent, Ichthyophonus,and Psorospermium, provisionally referred to by the acronym ‗DRIPs‘ (Ragan et al. 1996; Spanggaard et al. 1996).  The DRIPs clade was renamed the Ichthyosporea, the Mesomycetozoa or the Mesomycetozoea (e.g. Cavelier-Smith 1998; Herr et al. 1999; Mendoza et al. 2001, 2002). We use the name Ichthyosporea according to Adl et al. (2005). Other recently discovered unicellular opisthokonts included Corallochytrium, in its own monotypic class (Cavalier-Smith and Allsopp 1996), Ministeria (Tong 1998; Cavalier-Smith and Chao 2003), Capsaspora (Owczarzak et al. 1980; Hertel et al. 2002) and the nucleariid amoebae (Amaral Zettler et al. 2001). A sixth group, the Alphelidea (Gromov 2000), is also included amongst the basal opisthokonts but has no available sequences and an unknown phylogenetic placement (Adl et al. 2005). 2  Multi-gene phylogenies consistently place the nucleariida as the unicellular sister group to the fungi and Capsaspora, Corallochytrium, Ministeria, ichthyosporea and choanoflagellates as the unicellular relatives of animals (Lang et al. 2002; Ruiz-Trillo et al. 2006; Steenkamp et al. 2006). The exact branching order of these lineages is still inconclusive but recent multi-gene phylogenies place the ichthyosporea as or within the basal-most branch along the animal side after the animal fungus divide (Ruiz-Trillo et al. 2006; Steenkamp et al. 2006; Ruiz-Trillo et al 2008; Shalchian-Tabrizi et al. 2008). Studies of these free-living and parasitic unicellular lineages provide means for interpreting the history of character evolution leading to the origins of the animals and fungi. Determining events that occurred during the course of opisthokont evolution is still in its beginning stages. Genomic studies of representatives from several of the unicellular lineages have provided evidence for ancient origins of animal characteristic signalling pathways and cell- adhesion components (King et al. 2003; Segawa et al. 2006; King et al. 2008; Ruiz-Trillo et al. 2008; Shalchian-Tabrizi et al. 2008).  However, these lineages are under-represented in ecological, morphological and physiological studies, especially considering their phylogenetic divergences. This is mostly due to the relatively recent recognition of the lineages themselves and also the low number of representative isolates in culture collections from each lineage. Five of the lineages, Ministeria spp., Capsaspora owczarzaki, Corallochytrium limacisporum, Nuclearia spp. (Nucleariida), and the Alphelidea, are represented by one to few species (Raghu- kumar 1987; Patterson et al. 1993; Tong 1998; Gromov 2000; Amaral-Zettler et al. 2001; Hertel et al. 2002; Dykova et al. 2003).  The ichthyosporea are second to the choanoflagellates in terms of known levels of species diversity and are currently comprised of at least 30 genera.  The Ichthyosporea Although the first ichthyosporean small subunit rDNA sequence was not determined until 1995 (Kerk et al. 1995), they have been documented as parasites since the 19 th  century (reviewed in Mendoza et al. (2002). Most species were termed ‗enigmatic‘ and were reclassified several times. The earliest description I could find was of Enterobryus elegans, a millipede commensal, described by Leidy in 1849. The discovery of two other distantly related arthropod associated taxa, Psorospermium haeckeli and Amoebidium parasiticum, soon followed (Haeckel 1857; 3  Cienkowski 1861). Descriptions of non-arthropod associated species began with the discovery of fish parasite Ichthyophonus hoferi (Hofer 1893), followed later by the human parasite Rhinosporidium seeberi (Seeber 1900). In 1996 the lineage was described based on four sequenced taxa (Ragan et al. 1996) but has since grown to encompass over 30 genera. The greatest increase in the number of taxa within the ichthyosporea resulted from the transfer of Enterobryus and 16 other genera from the former fungus order Eccrinales (Cafaro 2005). This recent addition included many of the first ichthyosporeans to be discovered and accounts for more than half of described genera. Four genera, Sphaeroforma, Creolimax, Amphibiocystidium, and Amphibiothecum have been described since the establishment of the clade (Jostensen et al. 2002; Pascolini et al. 2003; Feldman et al. 2005; Pereira et al. 2005; Marshall et al 2008). As far as is known, the ichthyosporea are osmotrophic symbionts of animals. Typically the cells are spherical or ovoid, often with large central vacuoles or inclusions, have thick cell walls but few additional unifying characteristics. Reproduction has been described as asexual production and release of large numbers of motile zoospores, amoebae or non-motile endospores (Mendoza et al. 2002). Phylogenetically, the ichthyosporea can be divided into two lineages: the ichthyophonida and the dermocystida (Cavalier-Smith 1998) (synonymous to ichthyophonae and rhinosporideacae (Adl etal. 2005)). The ichthyophonida are symbionts of a wide range of animals including fish (Sindermann and Scattergood 1954), amphibians (Baker et al. 1999), and invertebrates (Whisler and Fuller 1968; Nylund et al. 1983; Lichtwardt 1997; Figueras et al. 2000). Most species spend at least part of their life-cycle within the host digestive tract (Mendoza et al 2002; Cafaro 2005). They range in shape from spherical to the long and thread-like eccrinids and reproduce via non-motile spores or amoeboid spores (Mendoza et al 2002; Cafaro 2005). Four species (Amoebidium parasiticum, Ichthyophonus hoferi, Pseudoperkinsus tapetis, and Sphaeroforma arctica) had been cultured prior to my work (Whisler 1960; Okamoto et al. 1985; Figueras et al. 2000; Jostensen et al. 2002). The ichthyophonida are comprised of at least 25 genera; all but seven genera belong within the uncultured arthropod inhabiting eccrinid lineage.  The dermocystida are composed of five genera, all of which parasitize vertebrates. One species, Rhinosporidium seeberi, has been found in mammals, including humans (Herr et al. 1999). The dermocystida often have intracellular stages and form spherical or complex plasmodia and cysts in the host tissue (usually epithelial) (Mendoza et al. 2002). Members of the 4  dermocystida are typically unculturable, except for a tiny intracellular parasite of salmon named Sphaerothecum destruens, which can be cultured within host cell lines. Depending on the species, spores can be simple and non-motile or can be motile zoospores with a single, naked, posterior flagellum. Amoebae have never been observed in the dermocystida.  Thesis theme and objectives PCR based surveys of small subunit ribosomal DNA (SSU-rDNA) sequences show phylotypes of uncultured unicellular opisthokonts, primarily from the choanoflagellates, unclassified lineages related to chytrid and zygomycete fungi, and the ichthyosporean lineage ichthyophonida. I was interested in establishing cultures representing undescribed species and lineages within the osmotrophic, unicellular opisthokonts. Living cultures allow for a greater range of physical observations, such as life cycles and development, large extractions of DNA, and can be archived in culture collections for future studies. Over a three year period I isolated and maintained 177 cultures of ichthyosporeans using marine invertebrate digestive tract contents as a source of inoculum. The partially digested contents of animal stomachs and intestines seemed like they would provide a good food source for osmotrophic organisms. I mainly sampled filter feeders and detritivores such as bivalves, and peanut worms due to their potential for concentrating microbial biota from their environment. I also expected that these animals may harbour symbiotic taxa. My cultures represented six SSU-rDNA phylotypes belonging to two lineages within the icthyosporean division ichthyophonida. Individual phylotypes were represented by one up to 126 cultured isolates. Four phylotypes belonged to a clade from which only two species were previously known; the other two were the first living representatives of a divergent lineage formerly populated by environmental sequences. Because my isolates grew easily in vitro, I had an unprecedented opportunity to study the asexual reproduction and ultrastructure of healthy vigorous cells. I identified and described my isolates as three new genera and six species using a combination of light and transmission electron microscopy and molecular phylogenetic analyses. In addition to making new discoveries about ichthyosporean morphology, I used unique attributes of each species to learn more about life cycles and host relationships of ichthyosporeans. These descriptions and corresponding observations are described in the 5  following four chapters. Abstracts for published or submitted chapters are provided in appendix A. For those that grew well in culture, I followed and described their asexual reproduction. For those that were found in abundance, I looked for evidence of host specificity. Three SSU-rDNA phylotypes –representing 148 of my isolates—were very similar to each other and to isolates collected abroad by others. In order to identify or describe them I first had to delimit species. From these 148 isolates I identified and described three cryptic species with overlapping ranges, two of which were new to science, using genealogical concordance phylogenetic species recognition criteria. Once delimited, I was able to further examine their local range, local pattern of speciation and host specificity. For the most abundant species, I used population-level genetic techniques to analyze the footprints left in the DNA sequence variation to further elucidate their life cycle, population structure and host relationships. By analyzing pattern of variation between sequences from loci with multiple alleles I was able to determine the ploidy of my asexually reproductive cultures and show a history of sexual recombination within an otherwise clonal population. Ultimately I was able to combine these findings with observations from cultures and isolation data to propose a life cycle that included a host/symbiont relationship.  6  References Adl S Simpson AG Farmer MA Anderson RA Anderson OR Barta JR Bowser SS Brugerolle G Fensome RA Fredericq S James TY Karpov SA Kugrens P Krug J Lane CE Lewis LA Lodge J Lynn DH Mann DG McCourt RM Mendoza L Moestrup Ø Mozley-Standridge SE Nerad TA Shearer CA Smirnov AV Spiegel FW Taylor M (2005) The new higher level classification of eukaryotes with emphasis on the taxonomy of protists. J Eukaryot Microbiol 52:399-451 Amaral-Zettler LA Nerad TA O'Kelly CJ Sogin M (2001) The nucleariid amoebae: more protists at the animal-fungal boundary. J Eukaryot Microbiol 48:293-297 Baker GC Beebee TJC Ragan MA (1999) Prototheca richardsi, a pathogen of anuran larvae, is related to a clade of protistan parasites near the animal-fungal divergence. Microbiology 145:1777-1784 Baldauf SL (1999) A search for the origins of animals and fungi: comparing and  combining molecular data. Am Nat 154: S178-S188 Baldauf SL Palmer J (1993) Animals and fungi are each other's closest relatives:  congruent evidence from multiple proteins. Proc Natl Acad Sci USA 90: 11558-  11562 Cafaro MJ (2005) Eccrinales (Trichomycetes) are not fungi, but a clade of protists at the early divergence of animals and fungi. Mol Phyl Evol 35:21-34 Cavalier-Smith T (1987) The origin of fungi and pseudofungi. Pages 339-353 in A. Rayner, C. Brasier, and D. Moore, editors. Evolutionary biology of fungi. Cambridge University Press, Cambridge Cavalier-Smith T (1998) Neomonada and the origin of animal and fungi. Pages 375-407 in G. Coombs, K. Vickerman, M. Sleigh, and A. Warren, editors. Evolutionary relationships among protozoa. Chapman and Hall, London Cavalier-Smith T Allsopp M (1996) Corallochytrium, an enigmatic non-flagellate  protozoan related to choanoflagellates. Eur J Protistol 32: 306-310 Cavalier-Smith T Chao EE (2003) Phylogeny of choanozoa, apusozoa, and other protozoa and early eukaryote megaevolution. J Mol Evol 56:540-563 Cienkowski L (1861) Ueber parasitische Schlauche auf Crustaceen und einigen Insektenlarven (Amoebidium parasiticum). Bot Ztg 19:169-174 7  Dykova I Veverkova M Fiala I Machackova B Peckova H (2003) Nuclearia pattersoni sp. n. (Filosea), a new species of amphizoic amoeba isolated from the gills of roach (Rutilus rutilus), and its rickettsial endosymbiont. Folia Parasitol 51:161-170 Feldman SH Wimsatt JH Green DE (2005) Phylogenetic classification of the frog pathogen Amphibiothecum (Dermocystidium) penneri based on small ribosomal subunit sequencing. J Wild Dis 41:701-706 Figueras A Lorenzo G Ordas MC Gouy M Novoa B (2000) Sequence of the small subunit ribosomal DNA gene of Pseudoperkinsus tapetis isolated carpet shell clam in Galicia (NW of Spain). Mar Biotechnol 2:419-428 Gromov BV (2000) Algal parasites of the genera Aphelidium, Amoeboaphelidium and Pseudophelidium from the Cienkovski's "Monadinea" group as representatives of a new class. Zool Zhurn 79:517-525 Haeckel E (1857) Ueber die gewebe des Flusskrebses. Arch Anat Physiol Med (Muller Archiv) 24:469-568 Herr RA Ajello L Taylor JW Arseculeratne SN Mendoza L (1999) Phylogenetic analysis of Rhinosporidium seeberi's 18S small subunit ribosomal DNA groups this pathogen among members of the protoctistan Mesomycetozoa clade. J Clin Microbiol 37:2750-2754 Hertel LA Bayne CJ Loker ES (2002) The symbiont Capsaspora owczarzaki, nov. gen. nov. sp., isolated from three strains of the pulmonate snail Biomphalaria glabrata is related to members of the Mesomycetozoa. Int J Parasitol 32:1183-1191 Hofer B (1893) Eine Salmonidenerkrankung. Allg Fisch Ztg 18:168-171 Jostensen JP Sperstad S Johansen S Landfald B (2002) Molecular-phylogenetic, structural and biochemical features of a cold adapted, marine ichthysporean near the animal-fungal divergence, described from in vitro cultures. Europ J Protistol 38:93-104 Kerk D Gee A Standish M Wainwrite P Drum A Elston R Sogin M (1995) The  rosette agent of chinook salmon (Oncorhynchus tshawytscha) is closely related to  choanoflagellates, as determined by the phylogenetic analysis of its small subunit  RNA. Mar Biol 122: 187-192 King N Hittinger CT Carroll SB (2003) Evolution of key cell signalling and adhesion protein families predates animal origins. Science 301:361-363 8  King N Westbrook MJ Young SL Kuo A Abedin M Chapman J Fairclough S Hellsten U Isogai Y Letunic I Marr M Pincus D Putnam N Rokas A Wright KJ Zuzow R Dirks W Good M Goodstein D Lemons D Li W Lyons JB Morris A Nichols S Richter DJ Salamov A Sequencing JGI Bork P Lim WA Manning G Miller WT McGinnis W Shapiro H Tjian R Grigoriev IV Rokhsar D (2008) The genome of the choanoflagellate Monosiga brevicollis and the origin of metazoans. Nature 451:783-788 Lang BF O'Kelly CO Nerad TA Gray MW Burger G (2002) The closest unicellular relatives of animals. Curr Biol 12:1773-1778 Leidy J (1849) Enterobryus, a new genus of confervaceae. Proc Natl Acad Sci Phil 4:225-233 Lichtwardt RW (1997) Costa Rican gut fungi (Trichomycetes) infecting lotic insect larvae. Rev Biol Tropical 45:1339-1383 Marshall WL Celio G McLaughlin DJ Berbee ML (2008) Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. Protist 159:415-433 Mendoza L Ajello L Taylor JW (2001) The taxonomic status of Lacazia loboi and Rhinosporidium seeberi has finally been resolved with the use of molecular tools. Rev Iberoam Micol 18:95-98 Mendoza L Taylor JW Ajello L (2002) The class Mesomycetozoea: a heterogeneous group of microorganisms at the animal-fungal boundary. Ann Rev Microbiol 56:315-344 Nylund V Westman K Laintmaa K (1983) Ultrastructure and taxonomic position of the  crayfish parasite Psorospermium haeckeli Heilgendorf. Freshw Crayfish 5:307-314 Okamoto N Nakase K Suzuki H Nakai Y Fujii K Sano T (1985) Life history and  morphology of Ichthyophonus hoferi in vitro. Fish Pathol 20: 273-286 Owczarzak H Stibbs H Bayne CJ (1980) The destruction of Schistosoma mansoni mother sporocysts in vitro by amoebae isolated from Biomphalaria glabrata: an ultrastructural study. J Invert Pathol 35:26-33 Pascolini R Dasak P Cunningham AA Tei S Vagnetti D Bucci S Fagotti A Di Rosa I (2003) Parasitism by Dermocystidium ranae in a population of Rana esculenta complex in Central Italy and description of Amphibiocystidium n. gen. Dis Aquat Org 56:65-74 9  Patterson D Nygaard K Steinberg G Turley C (1993) Heterotrophic flagellates and other protists associated with organic detritus throughout the water column in the mid North Atlantic. J Mar Biol Stat UK 73:67-95 Pereira CN Rosa IDI Fagotti A Simoncelli F Pascolini R Mendoza L (2005) The pathogen of frogs Amphibiocystidium ranae ia a member of the Order Dermocystida in the class mesomycetozoa. J Clin Microbiol 43:192-198 Ragan MA Goggin CL Cawthorn RJ Cerenuis L Jamieson AVC Plourde SM Rand TG Soderhall K Gutell RR (1996) A novel clade of protistan parasites near the animal- fungal divergence. Proc Natl Acad Sci USA 93:11907-11912 Raghu-kumar S (1987) Occurrence of the Thraustochytrid, Corallochytrium limacisporum gen. et sp. nov. in the Coral Reef Lagoons of the Lakshadweep Islands in the Arabian Sea. Bot Mar 30:83-89 Ruiz-Trillo I Lane CE Archibald JM Roger AJ (2006) Insights into the evolutionary origin and genome architecture of the unicellular opisthokonts Capsaspora owczarzaki and Sphaeroforma arctica. J Eukaryot Microbiol 53:1-6 Ruiz-Trillo I Roger AJ Burger G Gray MW Lang BF (2008) A phylogenomic investigation into the origin on metazoa. Mol Biol Evol 25:664-672 Seeber GR (1900) Un nuevo esporozoario parasito del hombre, dos casos encontrados en polipos nasales. Universidad Nacional de Buenos Aires. Segawa Y Suga H Iwabe N Oneyama C Akagi T Miyata T Okada M (2006) Functional development of Src tyrosine kinases during evolution from a unicellular ancestor to multicellular animals. Proc Natl Acad Sci USA 103:12021-12026 Shalchian-Tabrizi K Minge MA Espelund M Orr R Ruden T Jakobsen KS Cavalier-Smith T (2008) Multigene phylogeny of choanozoa and the origin of animals. PLoS ONE 3:e2098 Sindermann C Scattergood L (1954) Disease of fishes of the Western North Atlantic.  II. Ichthyosporidium disease of the sea herring (Clupea harengus). Research  Bulletin No. 19: Department of Sea and Shore Fisheries, Augusta, Maine pp. 1-40 Spanggaard B Skouboe P Rossen L Taylor JW (1996) Phylogenetic relationship of the intracellular fish pathogen Ichthyophonus hoferi, and fungi, choanoflagellates  and the rosette agent. Mar Biol 126: 109-115 10  Steenkamp ET Wright J Baldauf SL (2006) The protistan origins of animals and fungi. Mol Biol Evol 23:93-106 Tong S (1998) Heterotrophic flagellates from Southampton water UK. Ophelia 47:71-131 Wainright P Hinkle G Sogin M Stickel S (1993) Monophyletic origins of the Metazoa: an evolutionary link with fungi. Science 260: 340-342 Whisler HC (1960) Pure culture of the Trichomycete Amoebidium parasiticum. Nature 186: 732-733 Whisler HC Fuller M (1968) Preliminary observations on the holdfast of Amoebidium parasiticum. Dev Biol 17:562-570 11  Chapter 2. Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. 1   Introduction Contributions to our understanding of evolutionary protistology have come from the use of ultrastructural characters, molecular gene trees and the seemingly unstoppable discovery of unique and unusual taxa. These factors have been integral to the conception and development of the Opisthokonta, defined by a posteriorly uniflagellate motile stage and mitochondria with flat, plate-like cristae (Cavalier-Smith 1987). Initially opisthokonts were comprised of the multicellular animals and fungi and included only two unicellular groups: choanoflagellates and chytrid fungi (Baldauf 1999; Baldauf and Palmer 1993; Cavalier-Smith 1987; Wainright et al. 1993). However, a combination of reclassification of enigmatic taxa and some remarkable, recent discoveries of a few very divergent organisms has resulted in the addition of six more unicellular lineages (Adl et al. 2005). The first indication of these undiscovered lineages came from a molecular study of an unclassified salmonid parasite, named the rosette agent, which shared sequence similarity with the choanoflagellates (Kerk et al. 1995). Sequences from more taxa were determined soon after (Cavalier-Smith and Allsopp 1996; Ragan et al. 1996; Spanggaard et al. 1996), leading to the establishment of two new lineages: the provisional DRIPs clade (an acronym for the monophyletic grouping of Dermocystidium, the rosette agent, Ichthyophonus, and Psorospermium) (Ragan et al. 1996) and the monotypic Class Corallochytrea (Cavalier-Smith 1995). Cavalier-Smith (1998) renamed the DRIPs clade as Class Ichthyosporea within the Choanozoa. This was amended to the Class Mesomycetozoea by Mendoza et al. (2002), based on the clade name ―mesomycetozoa‖ proposed by Herr et al. (1999), to reflect the growing variety of known hosts. All three names are used synonymously in the literature but ―mesomycetozoa‖ is also used to describe all six of the recently added opisthokont protists in some classification schemes (Adl et al. 2005; Mendoza et al. 2001, Mendoza et al. 2002). We will use the original and less confusing name ―ichthyosporea‖ without rank designation (Adl et  1  A version of this chapter has been published. Marshall WL Celio G McLaughlin DJ Berbee ML (2008). Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. Protist 159:415-433. 12  al. 2005). Since the addition of the Eccrinales, formerly thought to be fungi (Cafero 2005), the ichthyosporea now comprise 29 genera. Each of the other four recently described lineages of unicellular opisthokonts, Ministeria spp., Capsaspora owczarzaki, Alphelidea, and the nucleariid amoebae, is represented by only one or very few species (Amaral-Zettler et al. 2001; Dykova et al. 2003; Gromov 2000; Hertel et al. 2002; Patterson et al. 1993; Raghukumar 1987; Tong 1998). As far as is known, ichthyosporeans are osmotrophs in parasitic or commensal relationships with animals. Typically the cells are spherical or ovoid, often with a large central vacuole or inclusion and thick cell walls but few additional characteristics. Asexual reproduction, where known, begins with a coenocytic sporangium or plasmodium, which gives rise to numerous uni- or multi-nucleate propagules. Phylogenetically, the ichthyosporea is divided into two lineages: the dermocystida and the ichthyophonida (Cavalier-Smith 1998) (synonymous to rhinosporidaeae and ichthyophonae (Adl et al. 2005)). In the dermocystida, asexual reproduction involves flagellated zoospores while in the ichthyophonida, amoebae have been the only motile stage observed. Most taxa from both orders appear to have lost all forms of motility. The dermocystida have yet to be cultured outside of their vertebrate host cells, and are typically associated with skin, gills, mucus membranes and visceral organs. The ichthyophonida are usually associated with host gut, epicuticle or visceral organs and are found in both vertebrate and invertebrate animals. Isolation and in vitro observations of novel ichthyosporean species are fundamental to the appreciation of the evolutionary history of all opisthokonts, including the multicellular animals and fungi. Environmental PCR surveys of water and sediment samples have detected unique sequences belonging to the ichthyophonida (Coolen et al. 2006; Savin et al. 2004; Stougaard et al. 2002; Takishita et al. 2005; Takishita et al. 2007ab) and hint at the existence of undetermined, deeply branching fungal and protistan opisthokont lineages as well as unknown biodiversity within established clades (Amaral-Zettler et al. 2002; Berney et al. 2004; Takishita et al. 2005; Takishita et al. 2007ab). We are isolating fungus-like protists from marine invertebrate digestive tract contents. Once organisms are in pure culture, we are studying their morphology, life cycle and phylogeny. Digestive tracts are good source material because they are nutrient rich, enclosed environments, presumably favourable for osmotrophs and possibly also housing parasitic or commensal species. In addition to symbionts, animals may concentrate protists from the surrounding microbial flora as a result of detrivorous or filter feeding. 13   In the course of this survey, we repeatedly isolated a spherical, unicellular, vigorously growing organism, which we describe in this paper as the new genus and species ―Creolimax fragrantissima.‖ Its small subunit ribosomal DNA (SSU-rDNA) sequence showed that it belonged to the ichthyophonida within the ichthyosporea. From the ichthyophonida, the only other species in culture are Sphaeroforma arctica, Pseudoperkinsus tapetis, Amoebidium parasiticum and a few strains of Ichthyophonus (Jostensen et al. 2002; Novoa et al. 2002; Okamoto et al. 1985; Rand et al. 2000; Sindermann and Scattergood 1954; Whisler 1960; Whisler 1962). The lack of available cultures has thwarted research into ichthyophonid biology. Basic information on asexual reproduction, ultrastructure and physiology are missing for most taxa. In addition to being difficult to culture, few ichthyophonida could be collected from their hosts at high enough frequencies to permit research into their ecology, host specificities and ranges, or population genetics.  S. arctica and P. tapetis, for example, were the only non-eccrinid taxa isolated from marine invertebrates (Jostensen et al. 2002; Figueras et al. 2000) and S. arctica was isolated only once. In this paper, we describe the phylogenetic relationships, morphology and cycle of asexual reproduction for a readily isolated and easily culturable, new species that offers new opportunities for research on the biology and ecology of a predominantly uncultured group of obligate parasites and commensals.  Methods Animal collection and isolation of strains Animals were collected during low tides from intertidal zones near the Bamfield Marine Sciences Centre in Barkley Sound, British Columbia, Canada between July 2004 and December 2005. They were held for a maximum of 48 hours in running seawater, euthanized in isotonic MgSO4, and dissected using sterilized tools. Portions from different parts of the digestive tract from three to five individuals (approximately 400 l) were mixed in a sterile eppendorf tube with 1 mL of sterile seawater. Coelomic or mantle cavity fluid was removed from some animals (Table 2.1) with a sterile syringe. Ten and 30 l of diluted slurry from digestive tracts and 30 l or 60 l of undiluted coelomic fluid were spread onto 50 mm petri dishes with different media containing antibiotics. Media used were as follows: YPD (1 g yeast extract, 1 g peptone, 5 g dextrose, 12 g agar 1 L seawater); PmTG (1.6 g peptonized milk agar (Sigma-Aldrich), 1 g 14  tryptone, 5 g dextrose, 9.8 g agar, 1 L seawater); MEM-10 (100 mL 10X MEM medium (Gibco), 900 mL seawater, 12 g agar, 110 mL bovine serum (Gibco) added after medium cooled); TGV- 10 (20 g tryptone, 5 g dextrose, 0.28 g KH2PO4, 0.35 g K2HPO4, 0.26 g (NH4)2SO4, 50 g biotin, 15 g agar, 110 mL bovine serum, 1 L seawater); YPSS (20 g soluble starch, 1 g yeast extract, 1 g K2HPO4, 0.05 g MgSO4, 1 L seawater); and TGhG (16 g tryptone, 4 g gelatin hydrolysate, 2 g dextrose, 12 g agar, 1 L seawater). All media were supplemented with 200 g of thiamine and one of the following combinations of antibiotics: 125 mg streptomycin and 105 mg doxycyclin, or 250 mg ampicillin and 125 g cloxacillin. Plates were incubated at both 10 and 15 C and were checked and weeded every four to seven days for up to two months. All colonies producing sporangia-like cells were picked and streaked onto fresh media until they were free of contaminating bacteria. Cultures were observed and maintained on YPD media without antibiotics or thiamine supplements.  DNA extraction, PCR amplification and sequencing Cells were scraped from the surface of petri plates and DNA was collected using a CTAB extraction protocol (Doyle and Doyle 1987). Universal primers (White et al. 1990) were used to amplify the SSU-rDNA and the ribosomal gene internal transcribed spacers ITS 1, partial ITS 2 regions and the 5.8S ribosomal DNA subunit (ITS-rDNA). All amplifications were performed in a GeneAmp PCR system 9700 (Applied Biosystems, Foster City, CA) using puReTaqTM Ready-To-Go PCR beads (Amersham Biosciences Corp., Piscataway, NY). Annealing temperatures were set to 52 and 56 ºC for the SSU-rDNA and ITS-rDNA, respectively. The elongation factor-like (EFL) gene, a paralogous copy of the EF1- (Keeling and Inagaki 2004) was amplified using primers EFL-F1 (5‘-ATCTGCGGMCACGTMGATG) and EFL-R1 (5‘- GGGTGGTTCATRAYCTGKACCTG) with an annealing temperature of 52 ºC. These primers were designed using a nucleotide alignment of the EFL region from Monosiga brevicollis (AY026073), two chytrids, Spizellomyces punctatus and Rhizophydium sp. (http://aftol.org/data.php), and S. arctica kindly provided by I. Ruiz-Trillo. Sequencing reactions were performed with an ABI PRISM BigDye Terminator v3.1 Cycle Sequencing Kit (PE Applied Biosystems, Foster City, CA) and were determined by the Nucleic Acid Protein Service Unit at the University of British Columbia. The fragments were compared and/or assembled 15  using AutoAssembler TM v.1.4 (Applied Biosystems, Perkin Elmer Corp., Norwalk, Conn). The ITS-rDNA sequence was compared for 26 isolates and the EFL sequence for one isolate from each host collection (n=7). All C. fragrantissima sequences were deposited in GenBank under accession numbers EU124887- EU124916 and EU169924-EU169930.   Phylogenetic analysis Sequences from the ichthyophonida and several outgroup taxa were selected from GenBank (Supplementary Table 2.1, Appendix B), then aligned using Clustal X v1.8 (Thompson et al. 1997) and edited by eye using MacClade v. 4.08 (Maddison and Maddison 2000). The SSU-rDNA sequence from ―C. fragrantissima‖ and several other mesomycetozoan taxa were used in BLAST searches (Altschul et al. 1990) in order to locate related environmental isolates as of July 2007. Two alignments were analyzed (TreeBASE ID#SN3798). To show the broad-scale relationships among ―C.  fragrantissima‖, animals, fungi, choanoflagellates, related protists, and species selected to represent diversity within the ichthyosporea, we used an "opisthokont" alignment, with 1649 sites. The second ― ichthyophonid‖ alignment, with 1671 sites, included all non-redundant sequences (as of July 2007) from the ichthyophonida, except for two closely related phylotypes of I. hoferi.  Regions of ambiguous alignment were excluded from all analyses. Parsimony bootstrapping of the ―opisthokont‖ and ―ichthyophonid‖ alignments involved 500 heuristic search replicates with random sequence addition and tree-bissection reconnection branch swapping performed by PAUP v4.0b10 (Swofford 1998). For bootstrapping of the ―ichthyophonid‖ alignment, rearrangements were limited to 1 X 108 to speed analysis. Bayesian analysis with MrBayes v3.0.2 (Huelsenbeck and Ronquist 2001) consisted of two runs with three heated and one cold chain. 1,000,000 and 2,000,000 generations were used for the ―opisthokont‖ and ―ichthyophonid‖ data matrices respectively with a 1/1000 sample frequency. The first 250 and 500 trees were discarded as burn in. The mean likelihoods of both runs were – ln 11415 (+/- 0.7) and –ln 8322 (+/- 0.8) for each data matrix and the convergence diagnostics were within 1 (+/- 0.005) for both analyses. A GTR+I+G model of nucleotide substitution was used in both cases based on parameters estimated by ModelTest v3.7 (Posada and Crandall 1998). Maximum likelihood analyses were calculated using PhyML v 2.6 (Guindon and Gascuel 16  2003) with 500 bootstrap replicates using a Tamura and Nei (1993) model of nucleotide substitution. An additional maximum likelihood analysis of the ―ichthyophonid‖ alignment was generated by PAUP using a GTR+I+G substitution model and 10 replicates.  Light and electron microscopy All microscopy was done on isolate CH-2 from a sub-tidal tunicate (Corella sp.) (Table 2.1). Growth cycles of individual cells were observed and photographed daily for 7-10 days in both seawater and YPD broth in Brown‘s Fungus Slide chambers (Brown 1942). Time-lapse photography was used to record more frequent images during time-spans of less than 24 hours; here cells were grown in chambers of nutrient broth between a slide and coverslip ringed by Carolina Observation Gel (Carolina Biological Supply Company, Burlington North Carolina). Nuclei were stained using DAPI for whole mounts and feulgen staining of sections. For staining, cells were fixed for one hour in 2% gluteraldehyde in 0.1 M cacodylate buffer pH 7.2, rinsed with the same buffer and then incubated in 0.001 g DAPI in 1 ml filtered seawater for 1 hour. Feulgen stained cells were fixed as above, dehydrated in ethanol and embedded in glycol methacrylate (Electron Microscopy Sciences). Two micron thick sections were stained according to O'Brien and McCully (1981). A culture was grown in YPD broth at room temperature for one week for high pressure- freeze substitution. Before freezing, the cells were centrifuged at low speed and transferred by pipet to petri plates containing solid medium to absorb excess moisture. Moist clumps of cells were loaded into 100 and 200 l brass planchettes and frozen in a BAL-TEC HPM-010 (Technotrade International, Manchester, NH), and transferred to 0.1% uranyl acetate and 2% osmium tetroxide in 100% anhydrous acetone at -80 C. Cells were left at -80 C for one week, slowly brought to room temperature, microwaved in Quetol II solution (Ted Pella, Inc) using a Pelco 3450 Lab Micro Processor equipped with a vacuum chamber, and embedded flat between Teflon coated microscope slides. Sections, 90-95 nm thick, were cut with a diamond knife, sections were collected on slot grids according to Rowley and Moran (1975) and stained with 3% uranyl acetate for 20 mins followed by lead citrate for three mins. Sections were observed and photographed on a Phillips CM12 transmission electron microscope.  17  Results  ―Creolimax fragrantissima‖ was isolated 27 times from four species of marine invertebrates. It was most regularly associated with the peanut worm, Phascolosoma agassizii, and over two years, it was cultured from four of five collections of these animals (Table 2.1). Gut contents were sampled from 32 additional animal collections (not shown); ―C. fragrantissima‖ was cultured from three of these (Table 2.1). No cultures developed from the coelomic or mantle cavity fluid (for bivalves) sampled from 14 of the 37 animal collections. All cultures have been maintained easily for 2-3 years through many transfers since isolation. Lyophilized material from each sequenced isolate was deposited at the Spencer Entomological Museum under accession numbers SEM-UBC MES-001-026 and one live culture from each animal collection (Table 2.1) was deposited at the Canadian Centre for the Culture of Microorganisms under accession numbers CCCM 101-107, both at the University of British Columbia, Vancouver, B.C., Canada.  Colonies of ―C. fragrantissima‖ were white with irregular margins that appeared to spread out in waves composed of piled spherical cells. Smaller, round cells were observed at a distance from the edges of the colonies. All cultures released an unfamiliar, sweet, aromatic scent and media in tubes or slants became suffused with a sulphur yellow pigment (Kornerup and Wanscher 1963).  Diagnoses Genus Creolimax gen. nov. (ICZN): Smooth, walled, multinucleate, spherical cells with a prominent central vacuole. Generic epithet denotes the prolific release of limax shaped amoebae upon maturity under culture conditions. Cultured colonies white, rough, irregular and swarming. Osmotrophic nutrition, with no evidence of phagocytosis during amoeboid phases. Dispersal via endospores or amoebae released through numerous small pores or tears in cell wall. Uninucleate amoebae actively crawl in jerky motions using a single broad anterior pseudopodium. Multiply lobed or aggregated cell clusters are occasionally present in cultured material; no hypha-like, budding or flagellated stages were observed. Ultrastructure: cell wall with an ~80 nm thick homogenous electron opaque outer layer and a translucent inner layer with embedded tubular outgrowths from the plasma membrane, numerous mitochondria with flat cristae, spindle pole 18  body associated with nucleus and Golgi apparatus. Placed within the ichthyosporean lineage ichthyophonida with P. tapetis and S. arctica as nearest relatives based on phylogenetic inference. Distinguished from Pseudoperkinsus and Sphaeroforma by swarming colonies and production of amoebae under culture conditions. In Creolimax endospores are released through numerous small perforations or pores while in S. arctica endospores escape when the parent cell wall bursts. Type species: C. fragrantissima.  C. fragrantissima sp. nov. : Cells 30-60 µm in diameter at maturity, release numerous motile amoebae 12 x 5 µm, which encyst into 6-8 µm spheres. Specific epithet refers to sweet aromatic scent released by cultured material. Isolated from marine invertebrate digestive tracts from the Pacific Coast of Canada but may not be limited to this environment. Holotype: C. fragrantissima live strain CH-2 deposited at the Canadian Centre for the Culture of Microorganisms (accession number CCCM 101) and lyophilized material at the Spencer Entomological Museum (SEM-UBC MES-001) both at the University of British Columbia, Vancouver, British Columbia, Canada. Type locality: From the digestive tract of a sub-tidal tunicate (Corella sp.) collected off the South Dock at the Bamfield Marine Sciences Centre, Bamfield Inlet, Barkley Sound, British Columbia, Canada.  Phylogenetic inference The SSU-rDNA sequence of C. fragrantissima was 1747 bp in length and shared 97 % identity to P. tapetis and S. arctica within a 1729 bp region of overlap. Identities between ITS- rDNA of C. fragrantissima and P. tapetis were 56 %, 65 % and 95 % respectively. Sequences of S. arctica ITS-rDNA regions were not available for comparisons. All 26 ITS-rDNA (includes the ITS 1, 5.8S and ITS 2 regions) sequences from different isolates of C. fragrantissima were identical based on comparisons of at least 405 (mean= 772) overlapping nucleotides. Also identical were all seven sequences from the EFL gene of different isolates, based on at least 599 (mean=854) nucleotides. The seven isolates used in the EFL sequencing included one isolate from each host collection. 19  C. fragrantissima nested deeply within the ichthyosporean  group ichthyophonida, in a clade that included P. tapetis and S. arctica -- two other species associated with marine invertebrates (Fig. 2.1). Several environmental clones isolated from marine environments were also within this clade (Figs. 2.1 and 2.2). C. fragrantissima‘s position at the base of a clade including S. arctica, P. tapetis, and four uncultured clones was consistently supported in all analyses. Uncultured clone BB01 diverged early in this putative marine clade (Fig. 2.1 and 2). All analyses of both data sets showed support for a sister group relationship between this marine clade and a clade of freshwater species including Anurofeca richardsi. Taxa within the ichthyophonida were consistently divided into three main clades (Figs. 2.1 and 2.2). The amoebidiales, eccrinales and Ichthyophonus formed one grouping, the putative marine clade containing the C. fragrantissima and the freshwater sister clade containing Anurofeca formed another, and finally there was a distinct cluster of uncultured aquatic clones. Psorospermium haeckeli had a highly derived sequence and moved between a basal position (Fig. 2.1), or closer to Ichthyophonus and the arthropod commensals (Fig. 2.2).  The eccrinales with the exception of Eccrinidus were monophyletic and supported by all methods (Fig. 2.2). The ichthyophonida and dermocystida were both monophyletic and supported by all methods; the ichthyosporeans formed a monophyletic group but without bootstrap support (Fig. 2.1). Substitution rate variation among the outgroup taxa, as evident in the long branch to Ministeria vibrans, may have weakened support for this node.  Light and electron microscopy Most cells were spherical and solitary, ranging from 6 to 65 µm in size, with a large central vacuole that broke up into smaller vacuoles and disappeared as the cell matured (Fig. 2.3A-C). Occasionally multiply lobed cells or aggregations of cells were present, either without (Fig. 2. 3D) or with (Fig. 2.3E-G) visible dividing cell walls or septa. The development of several of these cell types was observed daily in Brown‘s slide chambers and the dividing wall became visible as the cells grew (not shown). Budding was never observed, but the clusters could have originated from adjacent endospores or amoebae sticking together. The typical asexual life cycle as observed under Brown‘s slide conditions was as follows: New cysts or small endospores developed a central vacuole almost immediately, which increased 20  in size in proportion to cell growth, until maturation. At approximately 25-60 µm the sporangium became compartmentalized (Fig. 2.3C), and then numerous motile amoebae exited through one or several points through the parental cell wall (Fig. 2.3H). Amoebae were motile for varying periods of time and encysted at various distances from the parent cell. A second round of reproduction and release of amoebae was observed for many cells in both media and seawater. In many instances non-motile endospores were observed within or near the parental cell. Their release was not observed. Endospores may have had a brief amoeboid stage and their arrangement suggested that they exited through several points in the parental cell wall (Fig. 2.3I). Parental cell walls appeared intact, without tears or visible pores. In other instances the endospores did not exit the parental cell wall at all but developed a vacuole and continued to grow (Fig. 2.3J). Maturation times of lobes or aggregated cells were not synchronous (Fig. 2.3K).  Figure 2.4 shows a time series of the maturation of a three-celled cluster. All three cells initially appeared fused but dividing walls became visible and each cell matured independently, based on the lack of synchrony in the production of amoebae. DAPI staining resulted in some autofluorescence; however, it was clear that amoebae were uninucleate and that endospores and small cysts were either uni or binucleate (not illustrated). Endospores and small cysts in feulgen stained sections were also uni- or bi-nucleate and larger cells were multinucleate with prominent unstained nucleoli. Endospores growing within parental cell walls often achieved multi-nucleate stages (not illustrated). Amoebae were 12 µm long by 4.5-5.0 µm wide, with a rounded anterior, a single central inclusion or vacuole and posterior constriction or uroid (Fig. 2.5). They moved by abrupt protoplasmic streaming into a blunt anterior or slightly lateral pseudopodium -- depending on direction of travel (Fig. 2.5). Amoebae changed direction frequently and travelled various distances before encysting, often colliding with other amoebae or cysts (Fig. 2.4C-H). Often contact with another amoeba or cyst was sustained; occasionally, an amoeba would encyst in contact with an older cyst. Neither fusion nor fission of amoebae was observed. All cells sectioned for transmission electron microscopy were at a non-reproductive stage and no amoeboid cells or cells undergoing cleavage were observed. The cells were multinucleate and bound by a wavy plasmalemma and cell wall (Fig. 2.6A-C). A large central vacuole was surrounded by cytoplasm containing nuclei, areas with numerous mitochondria and membrane bound vesicles containing flocculent material, and patches of densely packed glycogen rosettes 21  (Fig. 2.6A-C). Glycogen patches were distributed throughout the endoplasm but were usually most abundant around the vacuole (Fig. 2.and B). The vesicles containing the flocculent contents were often seen fused to the central vacuole (Fig. 2.6A) and occasionally budded off or coalesced together (Fig. 2.6B). Other membrane bound inclusions appeared to be derived from, or fusing to the Golgi bodies (Fig. 2.7). Multivesicular bodies were also present, usually near golgi bodies (Fig. 2.7). Exceptionally electron opaque inclusions were scattered throughout the cell and were sometimes concentrated around the Golgi body (Fig. 2.6A) or within flocculent vesicles (Fig. 2.8B). Golgi bodies with approximately five flattened cisternae were centered over spindle pole bodies associated with each nucleus (Fig. 2.7) and typically had the maturing (trans) face towards the outside of the cell (Figs. 2.8A and 2.9A). The double membrane bound mitochondria had plate-like cristae (Figs. 2.6A-B, 2.8B). Nuclei were 2.5 µm in diameter with distinct nucleoli and patches of chromatin-like material, and they were bound by a double membrane (Figs. 2.7A and 2.8A) with round pores 85 nm in diameter (Figs. 2.7A, 2.9). Layers of rough endoplasmic reticulum (RER) were associated closely with the nucleus, often encircling it on many sides (Figs. 2.7A and 2.8A). Adjacent nuclei often shared the same network of endoplasmic reticulum (Fig. 2.8A). Layers of RER were also situated near the outer regions of the cell (Fig. 2.8A). Cell walls were composed of an outer homogenous electron opaque layer ranging from 40 to 100 nm in thickness and an inner electron translucent layer of similarly variable thickness (Figs. 2.6, 2.8 and 2.9A). The electron opaque layer of parent cells persisted beyond endospore release and cell death (Fig. 2.8A) and cells that were joined shared this outer electron opaque layer (Figs. 2.6A and 2.9A). The plasmalemma was typically wavy (as in lower part of Figures 2.8C-F and 2.9A) but sometimes formed a membranous tubular network (Fig. 2.8C-F). These tubules appeared to be joined to the plasmalemma; they sometimes lay parallel to the plasmalemma but more were more often perpendicular, and occasionally contacted the electron opaque layer of the cell wall (Fig. 2.8C-F). Serial sections showed that the tubules were cylindrical, not plate-like and were concentrated in specific areas or patches (Fig. 2.8). Vesicles with an electron opaque coating were at times visible near the plasmalemma but were not correlated with the presence or absence of the tubules (Fig. 2.8C-F). Spindle pole bodies resembling round, concave, crescent shaped, electron opaque plaques were present between the nucleus and Golgi apparatus (Figs. 2.7A-B and 2.9A-C). The bodies 22  were 285-300 nm in diameter and 80-95 nm deep. Extranuclear microtubules were evident in longitudinal and transverse sections (Fig. 2.9).  Discussion Phylogenetic relationships Molecular as well as morphological and ecological features of C. fragrantissima were consistent with the ichthyosporea.  Although the monophyly of the ichthyosporea lacked bootstrap support, the ichthyophonida and dermocystida were both supported.  Using the first phylogenetic analysis to include all known ichthyophonida taxa, that included members of the eccrinales and environmental sequences, we showed that C. fragrantissima was positioned deeply within the ichthyophonida, in a clade including two other species with spherical thalli from marine invertebrates, P. tapetis and S. arctica (Figueras et al. 2000; Jostensen et al. 2002; Novoa et al. 2002). This clade also included several unidentified environmental clones from marine sources, suggesting that C. fragrantissima may be one of many related species, most as yet undescribed, but also associated with marine invertebrates. The phylogenetic relationships of the three described taxa and the environmental clones is not clearly supported. This is likely due to the short sequence lengths of several of the clones, most notably Ikaite-53 and BB01. The sister group to the clade with C. fragrantissima and the other marine invertebrate associates was the freshwater Anurofeca and allies. This arrangement was evident in all prior phylogenetic analyses that included members of both groups (Hertel et al. 2004; Pekkarinen et al. 2003; Jostensen et al. 2002; Mendoza et al. 2002) and received support here in all analyses. The clade containing both C. fragrantissima and Anurofeca excluded all members of the ichthyophonida that have thick, multiply layered (three or more layers) cell walls and non- spherical morphologies. A. richardsi is spherical with a prominent vacuole or inclusion and has a thin cell wall although it differs from both C. fragrantissima and S. arctica by being considerably smaller, intractable to culture, and dividing by binary fission (Baker et al. 1999). Anurofeca and allies have been associated with anurans, snails, and perhaps other invertebrates, although the nature of the interaction remains unclear (Hertel et al. 2004; Beebee and Wong 1993; Wong et al. 1994).  Suggesting some form of host dependence, Wong et al. (1994) were successful in using fluorescent antibodies to locate cells of A. richardsi only in ponds where 23  anurans were present.  Perhaps both sister groups of spherical ichthyosporeans had an ancestor that lived in invertebrate digestive systems and has since diverged into marine and freshwater specialists.  Host-symbiont relationships  We assume that all of our 27 C. fragrantissima isolates from invertebrate hosts in four different phyla were closely related and likely conspecific, based on the identical ITS-rDNA and EFL sequences from all isolates sequenced and on the general lack of variation in growth or colony morphology among all isolates. Other ichthyophonid species, A. parasiticum, A. richardsi, and Ichthyophonus hoferi, also inhabit multiple hosts (Whisler 1960; Wong et al. 1994; Jones and Dawe 2002). Whether or not the ichthyosporea are usually host specific or distinguishable based on morphology is unclear; however, in Rhinosporidium ITS-rDNA variation is correlated with geography and host (Silva et al. 2005), while dissimilarities of ITS- rDNA sequences were consistent with morphological differences in Psorospermium (Bangyeekhun et al. 2001).  All ichthyosporeans have either been isolated from or observed within living animals and nearly all described ichthyophonida taxa have some association with the host digestive tract. Ichthyophonus and Psorospermium are the only two genera known to infect internal tissues but can enter the host orally (Gydemo 1996; Henttonen et al. 1997; Sindermann and Scattergood 1954; Vogt and Rug 1999). Given the frequency of isolation from digestive tract tissue and the absence of isolation from any of the coelomic fluid samples, it seemed likely that C. fragrantissima also inhabited the digestive tract as an osmotroph through at least part of its life cycle. All host animals appeared healthy and no lesions were noted on digestive tracts; when isolated, C. fragrantissima was not causing evident disease. The nature of the relationship between C. fragrantissima and its hosts could, however, range anywhere from parasitism to casual ingestion and determining the degree of harm or benefit would require in depth studies using appropriate host species.  Little is known about the host relationships of S. arctica and P. tapetis, the readily culturable taxa closest to C. fragrantissima.  P. tapetis was isolated on several occasions from the gills and blood of the carpet shell clam, Ruditapes decussatus, from the Atlantic coast of 24  Spain (Figueras et al. 2000; Ordas and Figueras 1998). S. arctica was isolated only once from a homogenate of the entire amphipod, Gammarus setosus, that had been collected in the Arctic Ocean, near Greenland (Jostensen et al., 2002). Attempts to re-inoculate the animal were unsuccessful (Jostensen et al., 2002); therefore, nothing is known of its location within the host. The eccrinales and Paramoebidium live as commensals in the hindguts of arthropods (Cafaro 2005) whereas A. richardsi, Ichthyophonus, and P. haeckeli tend towards inhibiting host growth or causing disease (Beebee 1991; Kocan et al. 1999; Thörnqvist and Söderhall 1993). Environmental DNA sequences of ichthyophonida detected from water or sediment samples in GenBank total 19 (July 2007), and these probably represent ichthyophonida cells and perhaps propagules persisting outside of a host. Cultured cells of C. fragrantissima continued growing and reproducing in sterile seawater for a week and cells left in seawater were still viable after six months. The cells may persist passively until being ingested by a new host and recommencing growth in a nutritionally rich digestive tract. Unlike many other osmotrophic saprophytes, C. fragrantissima produced no penetrating or attachment structures such as hyphae, rhizoids or holdfasts, at least under our cultural conditions. However, cells were sticky at certain stages and this may allow them to adhere to a host. Colonies adhered to solid media and individual cells sometimes clumped together and were occasionally attached to glassware.  Comparative morphology of C. fragrantissima vegetative cells Light microscopy  The multinucleate, spherical cells with a prominent central vacuole and cell wall in C. fragrantissima were typical for ichthyosporeans. S. arctica, A. richardsi, and most dermocystida are also spherical and typically have a large central vacuole or inclusion at some stage in their life cycle. The sporangia of the amoebidiales, eccrinales, and Psorospermium range from ovoid to thread-like in shape. Even when pH of media was adjusted to three or 9, C. fragrantissima did not produce the hyphal or budding stages found in Ichthyophonus (Franco-Sierra and Alvarez- Pellitero 1999). However, aggregate cells with two or more lobes or compartments were frequent in C. fragrantissima, and these cells resembled the lobed forms described for Psorospermium (Henttonen et al. 1997). 25   The spontaneous release of amoebae of C. fragrantissima under culture conditions in both media and seawater was unique among ichthyosporeans. Although amoebae are known in other ichthyophonida, in the amoebidiales, in Psorospermium, and in Ichthyophonus  (Henttonen et al. 1997; Lichtwardt and Williams 1992; Okamoto et al. 1985), they are released under specific conditions, such as moulting of host (Whisler 1965) or pH changes (Okamoto et al. 1985; Spanggaard et al. 1995). Additional ichthyophonida species could perhaps be induced to produce amoebae under the appropriate conditions.   The amoebae described here were morphologically similar to those from the amoebidiales (Dang and Lichtwardt 1979) and I. hoferi (Okamoto et al. 1985). Psorospermium amoebae were variously described as having a single pseudopodium (Henttonen et al. 1997) or being filose in form (Vogt and Rug 1999). Movement in C. fragrantissima was abrupt and in jerks (eruptive) as in Psorospermium (Henttonen et al. 1997) and was much more active than movement described from I. hoferi (Okamoto et al. 1985). P. haeckeli amoebae have been reported to divide by fission (Vogt and Rug 1999); however, this was not observed in C. fragrantissima, nor has it been reported for Ichthyophonus, or either genus of amoebidiales. Fusion was also not seen although cysts often formed in clumps or pairs; this clumping was also observed in studies of Amoebidium (Whisler 1962; Whisler 1963). Phagocytosis has not been observed in any ichthyophonida amoebae and the amoebae most likely function for dispersal. In C. fragrantissima, amoebae and endospores were most likely escaping through many small tears or pores, leaving an otherwise intact parent cell wall. This mechanism of dispersal was different from the sudden bursting of the parent cell wall exhibited by S. arctica (Jostensen et al. 2002).  Amoeboid phases are found in some, but not all, other opisthokonts. Capsaspora and Nuclearia are both filose in shape and phagocytic in nature, with the amoeboid phase as their main assimilative form (Patterson 1984; Stibbs et al. 1979). Even though Corallochytrium limacisporum (Corallochytrea) is a distantly-related opisthokont and probably not a ichthyosporean, its method of sporangial opening followed by the dispersal of its non- phagocytosing amoebae (Raghukumar 1987) were much like C. fragrantissima.  In contrast to the ichthyophonida, the dermocystida disperse via posteriorly flagellate zoospores or non-motile endospores and amoebae are completely unknown. Motile free-ranging amoebae are also absent from choanoflagellates and fungi, although they are present as phagocytic cell types in many multicellular animals. 26  Ultrastructure In addition to illustrating the ultrastructural features typical of opisthokonts, such as nuclei with prominent nucleoli and glycogen storage, this study captured one of the only spindle pole bodies known from an ichthyosporean. No dividing nuclei were encountered during this study and the spindle pole bodies were present during interphase. The spindle pole bodies in this study differed in size and consistency from the thin plaque-like organelle that was pressed against the nuclear envelope in I. hoferi (Franco-Sierra and Alvarez-Pellitero 1999). Centrioles were observed in Dermocystidium percae (Pekkarinen et al. 2003). Intra and extranuclear microtubules have been observed in the nuclearids (Dykova et al. 2003; Patterson 1983) but no organizing center.  Centrioles were never encountered in serial sections of C. fragrantissima and are probably not present. Mitochondria had plate-like cristae, archetypal of the Opisthokonta with the exception of I. hoferi, which has tubular cristae (Franco-Sierra and Alvarez-Pellitero 1999; Spanggaard et al. 1996). A further unique feature of C. fragrantissima was the organization of regions of the plasma membrane into branching tubules. The purpose of these tubules is unknown but they may increase the surface area of the plasma membrane, function in secretion, or affect the composition of the cell wall. Vesicular structures between the plasmalemma and outer cell wall have occasionally been observed in other ichthyosporeans; however, it was not clear whether these were cross-sections of tubules, fixation artefacts, or non-homologous structures (Franco- Sierra and Alvarez-Pellitero 1999; Moss 1999). Cell walls are present in all ichthyosporeans except during amoeboid or flagellated stages; however, cell wall structure and composition vary among the few species where they have been characterized.  Among the opisthokonts, persistent cell walls and osmotrophy are features of the ichthyosporeans, C. limacisporum and the Fungi. Cell walls from C. fragrantissima are relatively thinner and less complex than in most other ichthyosporeans. Many ichthyophonida, including I. hoferi, and most members of the dermocystida have thick, multi- layered walls. C. limacisporum has thin homogeneous cell walls but has projections or scales covering the external surface (Raghukumar 1987).  27  Conclusion Creolimax fragrantissima is one of few ichthyosporeans that can be isolated regularly and repeatedly from nature and then grown easily through cycles of asexual reproduction. Its quick luxuriant growth facilitates further studies into the morphology and physiology of the ichthyosporea, such as ultrastructure and composition of secondary metabolites. This is the first study documenting what is likely a single species found in a broad range of animal phyla and the first example of an ichthyosporean within a sipunculid and echinoderm. Further isolations may allow studies on prevalence, range, population biology and whether or not sexual recombination occurs within C. fragrantissima.  Acknowledgements We are thankful for the assistance of staff from the University of Minnesota College of Biological Sciences Imaging Center and the University of British Columbia Bio-Imaging Facility; students and staff of the Bamfield Marine Sciences Centre for assistance collecting animals; M. Padamsee, S. Lim, and J. Dee for technical assistance; I. Ruiz-Trillo and A. Rogers for the S. arctica EFL sequence; and K. Needham and S. Millen for helpful advice on nomenclature. This work was supported by the Natural Sciences and Engineering Research Council of Canada, Assembling the Fungal Tree of Life Project NSF grant EF-022867 to D. J. McLaughlin, and the Bamfield Marine Sciences Centre.   28  Table 2.1. Host animal, date and specific location, isolation and culturing details for animals and ichthyosporeans collected from Barkley Sound, Bamfield B.C. Canada. Host Animal (n= number sampled) Location and Date of Sampling Number of Colonies Isolated Isolation Media Coelomic Fluid Tested? Peanut worm (Phascolosoma agassizii) n=3 Scott‘s Bay Aug, 2004 1 YPSS Yes, 30 l Peanut worm (Phascolosoma agassizii) n=5 Scott‘s Bay Nov, 2004 10 YPD, PmTG, MEM-10, TGhG, TGV-10 No Peanut worm (Phascolosoma agassizii) n=4 BMSC Foreshore Oct, 2005 3 YPD Yes, 60 l Peanut worm (Phascolosoma agassizii) n=5 Dixon Island Oct, 2005 8 MEM-10, YPD, TGhG Yes, 60 l Tunicate (Corella sp.) n=5 BMSC South Dock Dec, 2004 1 MEM-10 No Burrowing sea cucumber (Leptosynapta clarki) n=4 Scott‘s Bay Nov, 2004 1 YPD No  29  Host Animal (n= number sampled) Location and Date of Sampling Number of Colonies Isolated Isolation Media Coelomic Fluid Tested? Chiton (Katharina tunicata) n=3 Scott‘s Bay Nov, 2005 3 YPD, TGhG, PmTG No   30  0.1 Astreptonema sp. Palavascia patagonica Astreptonema gammari Enteropogon sexuale Palavascia sphaeromae Enterobryus oxidi Enterobryus sp. SPA2C10 Enterobryus sp. MA11C1 Enterobryus sp.  SPA10C6 Amoebidium parasiticum Paramoebidium sp. Eccrinidus flexil is SPA11C45 Eccrinidus flexil is SPA10C2 Ichthyophonus hoferi U25637 Ichthyophonus irregularis Pseudoperkinsus tapetis Uncultured eukaryote clone TAGIRI-25 Sphaeroforma arctica Uncultured mesomycete clone Ikaite un-c53 Uncultured eukaryote clone BB01 Uncultured Anurofeca clone LAH2003 Anurofeca richardsi Uncultured Ichthyophonida clone LKM51 Uncultured eukaryote clone s25 Uncultured eukaryote clone TAGIRI-27 Uncultured eukaryote clone s16 Psorospermium haekeli Rhinosporidium seeberi Dermocystidium sp. Dermocystidium salmonis Dermocystidium percae Sphaerothecum destruens Ministeria vibrans Capsaspora owczarzaki Corallochytrium limacisporum Monosiga ovata Diaphanoeca grandis Nuclearia simplex Spizellomycete sp. Olpidium brassicae Saccharomyces cerevisiae Mytilus trossulus Xenopus laevisAcanthamoeba polyphaga Creolimax fragrantissima 99/100/99 -/-/73 -/-/65 -/-/73 88/100/72 100/100/74 100/100/98 -/81/61 100/100/100 86/100/67 99/100/100 77/100/69 68/65/79 -/-/89 60/79/73 74/84/74 87/100/68 96/100/97 99/100/100 100/100/100 100/100/100 97/100/80 84/-/84 -/61/- 95/100/99 -/67/- -/100/97 -/63/- 79/-/- 100/100/100 99/100/97 -/99/79 C H T H Y O S P O R E A ANIMALS FUNGI Dermocystida Ichthyophonida 82/100/- -/85/- I CHOANO- FLAGELLATES a a d a a a a d d d d d d d d d d ? a/d a/d d d a/d a/d L L L L L L L L L L L L L L L SSU-rDNA (MP/Ba/PhyML) a a  Figure 2.1.  Maximum likelihood PhyML tree comparing phylogenetic relationship of C. fragrantissima to other opisthokont SSU-rDNA sequences obtained from GenBank. Numbers at nodes represent 500 parsimony or PhyML likelihood pseudoreplications, or Bayesian posterior probabilities (MP/Ba/PhML) over 60%. Icons of amoebae and flagellated stages are shown for taxa where these structures are known. Lower case ‗a‘ or ‗d‘ denotes whether these stages are found during assimilative or dispersal stages. An ‗L‘ represents the loss of motile stages. 31  AMOEBIDIALES E C C R I N A L E S  Invertebrates (?)/ Marine Fish/Anadromous and Marine Freshwater A r t h r o p o d s U ltrastru cture C u ltu rab le M u ltiple  iso late s YY Y Y Y Y Y+/- Y Y+/- Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y 0.1 Palavascia sphaeromae Enterobryus oxidi Astreptonema sp. Palavascia patagonica Alacrinella limnoriae Astreptonema gammari Taeniellopsis sp. Enteromyces callianassae Taeniella carcini Enteropogon sexuale Enterobryus sp. MA11C1 Enterobryus sp. SPA2C10 Enterobryus sp. SPA10C6 Amoebidium parasiticum Paramoebidium sp. Eccrinidus flexilis SPA11C45 Eccrinidus flexilis SPA10C2  Ichthyophonus hoferi U43712 Ichthyophonus hoferi U25637 Ichthyophonus irregularis Psorospermium haeckeli Uncultured clone D4P08E02 Uncultured  clone D3P06H09 Sphaeroforma arctica Pseudoperkinsus  tapetis Uncultured clone TAGIRI-25 Uncultured clone Ikaite unc-53 Creolimax fragrantissima Uncultured clone BB01 Anurofeca richardsi Uncultured Anurofeca  clone LAH2003 Uncultured  clone LKM-51 Uncultured clone Qinghai 5-2 Uncultured clone s16 Uncultured clone s25 Uncultured clone s712 Uncultured clone s416 Uncultured clone NAMAKO-33 Uncultured Amoebidium  clone 1DB4 Uncultured clone TAGIRI-27 Uncultured clone ms313 Uncultured clone TAGIRI-26 Uncultured clone  BSDGGE Uncultured clone DSGM65 Uncultured clone NAMAKO-34 Rhinosporidium seeberi E N V  R O N M E N T A L I C L O N E S 83/98/80 -/64/- -/90/- 70/93/72 -/90/- -/90/- -/72/- -/66/- -/85/- 99/100/96 100/100/99 89/100/90 100/100/100 75/98/71 84/100/81 90/97/93 100/100/100 -/74/62 76/99/88 76/100/89 77/100/91 66/90/86 64/77/84 83/87/- 78/73/81 100/100/100 98/100/99 -/71/- -/98/- 75/97/- 67/100/83 99/100/85 65/100/95 100/100/100 100/100/100 Y Y SSU-rDNA (MP/Ba/PhyML)  Figure 2.2. One of six maximum likelihood trees (-ln 8281) comparing SSU-rDNA sequence of C. fragrantissima to all non-redundant sequences of ichthyosporeans belonging to the ichthyophonida. Numbers at nodes represent 500 parsimony or PhyML likelihood pseudoreplications, or Bayesian posterior probabilities (MP/Ba/PhML) over 60%. Grey bars highlight taxa meet two or more criteria from the sidebar. A. parasiticum, some strains of Ichthyophonus hoferi, and C. fragrantissima are the only ichthyophonida species with published electron micrographs that have been repeatedly isolated into culture. 32   Figure 2.3. General morphology of vegetative cells and of asexual reproduction. A and B. Spherical cells with a single large or multiple vacuoles. C. A more mature cell, which has subdivided prior to release of amoebae or endospores. D. An elongate or bilobed cell with two prominent vacuoles, no cell wall is present between the halves. E, F and G. Multiply lobed or aggregated cells with walls or septa (arrowheads) separating the cells or compartments. H. Amoebae (A) escaping through pores or tears in parental cell wall (arrow) and continuing to crawl along slide. I. Endospores escaping through parental cell wall (arrow).  J. Walled, vacuolated endospores growing within and escaping through a tear in parental cell wall. K. Differential maturation times of cells separated by septa. Note that one cell of each pair is releasing endospores while its neighbour remains vegetative. Scale bar equals 20 µm. 33   Figure 2.4. Time series photography showing separation and differential maturation of a multiply lobed cell. A. No differentiation or visible septa are seen between lobes. B. Lobe or cellular partition in posterior most position is releasing amoeboid (a) cells. C-E. Septum dividing the top and bottom lobes into individual cells is well developed. The top cell has matured and has subdivided and begun to release amoebae. The central vacuole still persists in the lower cell. F. Top cell is empty but the original cell wall persists (arrow). The bottom cell is now subdivided. G. Amoebae are escaping from the bottom cell. H. All cells are empty but the cell wall remains. Some amoebae have encysted into small spheres. I. There are no longer any active amoebae. The original cell walls are still present (arrow) and many of the encysted amoebae have begun to increase in size. T= number of minutes since frame A. Scale bar is equal to 20 µm. 34   Figure 2.5. DIC time-series of an actively moving amoeba. The circled dot is an immobile particle in the medium. The pseudopodium (P) is anterior and smooth compared to the granular mid and posterior regions. A central contractile vacuole (V) and posterior uroid (U) are evident. Time interval equals 3.5 seconds; scale bar is equal to 5 µm. 35   Figure 2.6. Electron micrographs of C. fragrantissima cells during assimilative growth. Multiple nuclei, mitochondria, several vesicle types and patches of glycogen rosettes are visible in endoplasm surrounding the central vacuole. A. Section through cells with a shared cell wall or septum (S). B and C. Sections through solitary cells.  (N) nuclei; (M) mitochondria; (S) shared cell wall or septum; (CV) central vacuole; (FV) flocculent vesicle; (D) endoplasm surrounded by double membrane; (G) glycogen; circle around several electron opaque inclusions or vesicles. Arrowhead shows flocculent vesicle fused to central vacuole; asterisk budding or coalescing flocculent vesicles. Scale bar equals 5 µm. 36   Figure 2.7. Spindle pole body next to nucleus, directly under Golgi body. A. Detail of nucleus, golgi body and vesicles, and rough endoplasmic reticulum from cell in figure 2.6C. B. Golgi cisternae, vesicles and spindle pole body from nucleus in A. C. Flattened cisternae and associated vesicles. (N) nucleus; (Go) Golgi body; (R) rough endoplasmic reticulum; (SPB) spindle pole body; (MVB) multivesicular body; arrowhead points to a nuclear pore. Scale bar equals 500 nm. 37   Figure 2.8. Detail of cell wall showing membrane bound tubular extensions penetrating the inner cell wall. A. Adjacent cells with cell wall remnant from previous generation (arrowhead). Asterisk marks continuation of endoplasmic reticulum between nuclei. B. Detail of endoplasm from cell shown in Figure 2.6A. Electron opaque inclusions (P) were sometimes found within flocculent vesicles (FV). Finger-like extensions of the plasmalemma can be seen at top right. C-F. Serial sections through cell wall showing the wavy plasmalemma with network of tubular extensions. (N) nucleus; (Go) golgi body; (CV) central vacuole; (M) mitochondria; (R) rough endoplasmic reticulum; (E) extensions of plasmalemma into tubules; (C) areas where tubules contact electron opaque layer of cell wall; (CdV) coated vesicles. Scale bar equals 500 nm. 38   Figure 2.9. Spindle pole bodies between Golgi body and nucleus during interphase. A. Microtubules are visible in space between nucleus and Golgi. B. Adjacent section of nucleus in A showing an extra-nuclear microtubule from spindle pole body. C. Oblique section through a spindle pole body and attached microtubules. (N) Nucleus; (SPB) spindle pole body; (S) shared wall or septum; (NP) nuclear pore; (Mt) microtubule; (Go) Golgi body. Scale bar equals 500 nm. 39  References Adl SM Simpson AG Farmer MA Anderson RA Anderson OR Barta JR Bowser SS  Brugerolle G Fensome RA Fredericq S James TY Karpov S Kugrens P Krug  J Lane CE Lewis LA Lodge J Lynn DH Mann DG McCourt RM Mendoza L  Moestrup Ø Mozley-Standridge SE Nerad TA Shearer CA Smirnov AV  Spiegel FW Taylor MFJR (2005) The new higher level classification of  eukaryotes with emphasis on the taxonomy of protists. J Eukaryot Microbiol 52:  399-451 Altschul SF Gish W Miller W Meyers EN Lipman DJ (1990) Basic local alignment  search tool. J Mol Biol 215:403-410 Amaral-Zettler LA Gómez F Zettler E Keenan BG Amils R Sogin ML (2002)  Microbiology: Eukaryotic diversity in Spain‘s River of Fire. Nature 417:137 Amaral-Zettler LA Nerad TA O'Kelly CJ Sogin M (2001) The nucleariid amoebae:  more protists at the animal-fungal boundary. J Eukaryot Microbiol 48:293-297 Baker GC Beebee TJC Ragan MA (1999) Prototheca richardsi, a pathogen of anuran  larvae, is related to a clade of protistan parasites near the animal-fungal  divergence. Microbiology 145:1777-1784 Baldauf SL (1999) A search for the origins of animals and fungi: comparing and  combining molecular data. Am Nat 154:S178-S188 Baldauf SL Palmer J (1993) Animals and fungi are each other's closest relatives:  congruent evidence from multiple proteins. Proc Natl Acad Sci USA 90:11558-  11562 Bangyeekhun E Ryynanen HJ Cerenius L Söderhall K (2001) Sequence analysis of  the ribosomal internal transcribed spacer DNA of the parasite Psorospermium  haeckeli. Dis Aquat Org 46:217-222 Beebee TJC (1991) Purification of an agent causing growth inhibition in anuran larvae  and its identification as a unicellular unpigmented alga. Can J Zool 69:2146-2153 Beebee T Wong ALC (1993) Stimulation of cell division and DNA replication in  Prototheca richardsi by passage through larval amphibian guts. Parasitology 107:  119-124 Berney C Fahrni J Pawlowski J (2004) How many novel eukaryotic ‗kingdoms‘? 40   Pitfalls and limitations of environmental DNA surveys. BMC Biol 2:1-13 Brown H (1942) A micro culture slide for fungi. J Bacteriol 43:16 Cafaro MJ (2005) Eccrinales (Trichomycetes) are not fungi, but a clade of protists at the  early divergence of animals and fungi. Mol Phyl Evol 35:21-34 Cavalier-Smith T (1987) The origin of fungi and pseudofungi. In Rayner A Brasier C,  Moore D (eds) Evolutionary biology of fungi. Cambridge University Press,  Cambridge, pp 339-353 Cavalier-Smith T (1995) Zooflagellate phylogeny and classification. Cytology 37:1010-  1029 Cavalier-Smith T (1998) Neomonada and the origin of animal and fungi. In Coombs G Vickerman K Sleigh M Warren A (eds) Evolutionary relationships among protozoa. Chapman and Hall, London, pp 375-407 Cavalier-Smith T Allsopp M (1996) Corallochytrium, an enigmatic non-flagellate  protozoan related to choanoflagellates. Eur J Protistol 32:306-310 Coolen MJL Boere A Abbas B Baas M Wakeman SG Damasté JSS (2006) Ancient  DNA derived from alkenene biosynthesizing haptophytes and other algae in  Holocene sediments from the Black Sea. Paleoceanography 21:PA1005  doi:101029/2005PA001188 Dang S-N Lichtwardt RW (1979) Fine structure of Paramoebidium (Trichomycetes)  and a new species with virus-like particles. Am J Bot 66:1093-1104 Doyle JJ Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh  leaf tissue. Phytochem Bull 19:11-15 Dykova I Veverkova M Fiala I Machackova B Peckova H (2003) Nuclearia pattersoni  sp. n. (Filosea), a new species of amphizoic amoeba isolated from the gills of  roach (Rutilus rutilus), and its rickettsial endosymbiont. Folia Parasitol 51:161-  170 Figueras A Lorenzo G Ordas MC Gouy M Novoa B (2000) Sequence of the small  subunit ribosomal DNA gene of Pseudoperkinsus tapetis isolated carpet shell  clam in Galicia (NW of Spain). Mar Biotechnol 2:419-428 Franco-Sierra A Alvarez-Pellitero P (1999) The morphology of Ichthyophonus sp. in  their mugilid hosts (Pisces: Teleostei) and following cultivation in vitro. A light 41   and electron microscopy study. Parasitol Res 85:562-575 Gromov BV (2000) Algal parasites of the genera Aphelidium, Amoeboaphelidium and  Pseudophelidium from the Cienkovski‘s ―Monadinea‖ group as representatives  of a new class. Zool Zhurn 79:517-525 Guindon S Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large  phylogenies by maximum likelihood. Sys Biol 52:696-704 Gydemo R (1996) Signal crayfish, Pacifastacus leniusculus, as a vector for  Psorospermium haeckeli to noble crayfish, Astacus astacus. Aquaculture 148:1-9 Henttonen P Huner JV Rata P Lindqvist OV (1997) A Comparison of the known life  forms of Psorospermium spp. in freshwater crayfish (Arthropoda, decapoda) with  emphasis on Astacus astacus L. (Astacidae) and Procambarus clarkii (Girard)  (Cambaridae). Aquaculture 149:15-30 Herr RA Ajello L Taylor JW Arseculeratne SN Mendoza L (1999) Phylogenetic  analysis of Rhinosporidium seeberi's 18S small subunit ribosomal DNA groups  this pathogen among members of the protoctistan Mesomycetozoa clade. J Clin  Microbiol 37:2750-2754 Hertel LA Barbosa CS Santos RAAL Loker ES (2004) Molecular identification of  symbionts from the pulmonate snail Biomphalaria glabrata in Brazil. J Parasitol  90:759-763 Hertel LA Bayne CJ Loker ES (2002) The symbiont Capsaspora owczarzaki, nov. gen.  nov. sp., isolated from three strains of the pulmonate snail Biomphalaria glabrata  is related to members of the Mesomycetozoa. Int J Parasitol 32:1183-1191 Huelsenbeck JP Ronquist F (2001) MrBayes: Bayesian inference of phylogenetic trees.  Bioinformatics 17:754-755 Jones SRM Dawe SC (2002) Ichthyophonus hoferi (Plehn & Mulsow) in British  Columbia stocks of Pacific herring, Clupea pallasi (Valenciennes), and its  infectivity to chinook salmon, Oncorhynchus tshawytscha (Walbaum). J Fish Dis  25:415-421 Jostensen JP Sperstad S Johansen S Landfald B (2002) Molecular-phylogenetic,  structural and biochemical features of a cold adapted, marine ichthysporean near  the animal-fungal divergence, described from in vitro cultures. Eur J Protistol 42   38:93-104 Keeling PJ Inagaki Y (2004) A class of eukaryotic GTPase with punctuate distribution  suggesting multiple functional replacements of translation elongation factor 1.  Proc  Natl Acad Sci USA 101:15380-15384 Kerk D Gee A Standish M Wainwrite P Drum A Elston R Sogin M (1995) The  rosette agent of chinook salmon (Oncorhynchus tshawytscha) is closely related to  choanoflagellates, as determined by the phylogenetic analysis of its small subunit  RNA. Mar Biol 122:187-192 Kocan RM Hershberger PK Mehl T Elder N Bradley M Wildermuth D Stick K  (1999) Pathogenicity of Ichthyophonus hoferi for laboratory-reared Pacific  herring Clupea pallasi and its early appearance in wild Puget Sound herring. Dis  Aquat Org 35:23-29 Kornerup A Wanscer JH (1963) Methuen handbook of colour. Methuen and Co. Ltd.,  London, UK Lichtwardt RW Williams MC (1992) Two new Australian species of Amoebidiales  associated with aquatic insect larvae, and comments on their biogeography.  Mycologia 84:376-383 Maddison WP Maddison DR (2000) MacClade 4: Interactive analysis of phylogeny and  character evolution. Sinauer, Sunderland, MA, USA Mendoza L Taylor JW Ajello L (2001) The taxonomic status of Lacazia loboi and  Rhinosporidium seeberi has finally been resolved with the use of molecular tools.  Rev Iberoam Micol 18:95-98 Mendoza L Taylor JW Ajello L (2002) The class Mesomycetozoa: A heterogeneous  group of microorganisms at the animal-fungal boundary. Ann Rev Microbiol 56:  315-344 Moss ST (1999) Astreponema gammari: an eccrinid with appendaged spores. Kew Bull  54:637-650. Novoa B Ordas MC Figueras A (2002) Hypnospores detected by RFTM in clam  (Ruditapes decussatus) tissues belong to two different protozoan organisms,  Perkinsus atlanticus and a Perkinsus-like organism. Aquaculture 209:11-18 O'Brien TP McCully ME  (1981) The study of plant structure and principles and 43   selected mothods. Termarcarphi Pty Ltd., Victoria, Australia Okamoto N Nakase K Suzuki H Nakai Y Fujii K Sano T (1985) Life history and  morphology of Ichthyophonus hoferi in vitro. Fish Pathol 20:273-286 Ordas MC Figueras A (1998) In vitro culture of Perkinsus atlanticus. Dis Aquat Org  33:129-136. Patterson D (1984) The genus Nuclearia (Sarcodina, Filosea): Species composition and  characteristics of the taxa. Arch Protistenk 128:127-139 Patterson D Nygaard K Steinberg G Turley C (1993) Heterotrophic flagellates and  other protists associated with organic detritus throughout the water column in the  mid North Atlantic. J Mar Biol Stat UK 73:67-95 Patterson DJ (1983) On the organization of the naked filose amoeba, Nuclearia  moebiusi Frenzel, 1897 (Sarcodina, Filosea) and its implications. J Protozool 30:  301-307 Pekkarinen M Lom J Murphy CA Ragan MA Dykova I (2003) Phylogenetic position  and ultrastructure of two Dermocystidium species (Ichthyosporea) from the  common perch (Perca fluviatilus). Acta Protozool 42:287-307 Posada D, Crandall KA (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818 Ragan MA Goggin CL Cawthorn RJ Cerenuis L Jamieson AVC Plourde SM Rand TG Söderhall K Gutell RR (1996) A novel clade of protistan parasites near the animal-fungal divergence. Proc Natl Acad Sci USA 93:11907-11912 Raghukumar S (1987) Occurrence of the Thraustochytrid, Corallochytrium limacisporum gen. et sp. nov. in the Coral Reef Lagoons of the Lakshadweep Islands in the Arabian Sea. Bot Mar 30:83-89 Rand TG White K Cannone JJ Gutell RR Murphy CA Ragan MA (2000)  Ichthyophonus irregularis sp. nov. from the yellowtail flounder Limanda  ferruginea from the Nova Scotia shelf. Dis Aquat Org 41:31-36 Rowley CR Moran DT (1975) A simple technique for mounting wrinkle-free sections  on formvar-coated slot grids. Ultramicrotomy 1:151-155 Savin MC Martin JL LeGresley M Giewat M Rooney-Varga J (2004) Plankton  diversity in the Bay of Fundy as measured by morphological and molecular 44  methods. Microb Ecol 48:51-65 Silva V Pereira CN Ajello L Mendoza L (2005) Molecular evidence for multiple host-  specific strains in the genus Rhinosporidium. J Clin Microbiol 43:1865-1868 Sindermann C Scattergood L (1954) Disease of fishes of the Western North Atlantic.  II. Ichthyosporidium disease of the sea herring (Clupea harengus). Research  Bulletin No. 19: Department of Sea and Shore Fisheries, Augusta, Maine pp. 1-40 Spanggaard B Huss HH Bresciani J (1995) Morphology of Ichthyophonus hoferi  assessed by light and scanning electron microscopy. J Fish Dis 18:567-577 Spanggaard B Skouboe P Rossen L Taylor JW (1996) Phylogenetic relationship of the intracellular fish pathogen Ichthyophonus hoferi, and fungi, choanoflagellates  and the rosette agent. Mar Biol 126:109-115 Stibbs HH Owczarzak A Bayne CJ DeWan P (1979) Shistosome sporocyst-killing amoebae isolated from Biomphalaria glabrata. J Inv Pathol 33:159-170 Stougaard P, Jorgensen F, Johnson MG, Hansen OC (2002) Microbial diversity in ikaite tufa columns: an alkaline, cold ecological niche in Greenland. Environ Microbiol 4:487-493 Swofford DL (1998) PAUP*: phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer Associates. Sunderland, Mass Takishita K Miyake H Kawato M Maruyama T (2005) Genetic diversity of microbial eukaryotes in anoxic sediment around fumaroles on a submarine caldera floor based on the small-subunit rDNA phylogeny. Extremophiles 9:185-196 Takishita K Tsuchiya M Kawato M Oguri K Kitazato H Maruyama T (2007a) Genetic diversity of microbial eukaryotes in anoxic sediment of the saline meromictic Lake Namako-ike (Japan): On the detection of anaerobic or anoxic- tolerant lineages of eukaryotes. Protist 158:51-64 Takishita K Yubuki N Kakizoe N Inagaki Y Maruyama T (2007b) Diversity of  microbial eukaryotes in sediment at a deep-sea methane cold seep: surveys of ribosomal DNA libraries from raw sediment samples and two enrichment cultures. Extremophiles 11:563-576 Tamura K Nei M (1993) Estimation of the number of nucleotide substitutions in the  control region of mitochondrial DNA in humans and chimpanzees. Mol Biol Evol 45  10:512-526 Thompson JD Gibson TJ Plewniak F Jeanmougin F Higgins DG (1997) The Clustal X windows interface flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl Acid Res 24:4876-4882 Thörnqvist P-O Söderhall K (1993) Psorospermium haeckeli and its interaction with the crayfish defence system. Aquaculture 117:205-213 Tong S (1998) Heterotrophic flagellates from Southampton water UK. Ophelia 47:71-  131 Vogt G Rug M (1999) Life stages and tentative life cycle of Psorospermium haeckeli, a species of the novel DRIPs clade from the animal-fungal dichotomy. J Exp Zool 283:31-42 Wainright P Hinkle G Sogin M Stickel S (1993) Monophyletic origins of the Metazoa: an evolutionary link with fungi. Science 260:340-342 Whisler HC (1960) Pure culture of the Trichomycete Amoebidium parasiticum. Nature 186:732-733 Whisler HC (1962) Culture and nutrition of Amoebidium parasiticum. Am J Bot 49: 193-199 Whisler HC (1963) Observations on some new and unusual enterophilous phycomycetes. Can J Bot 41:887-900 Whisler HC (1965) Host-integrated development in Amoebidiales. J Protozool 13: 183-188 White TJ Bruns T Lee S Taylor JW (1990) Amplification and direct sequencing of   fungal ribosomal RNA genes for phylogenetics. In Innis MA Gelfand DH Sninsky JJ White TJ (eds) PCR Protocols: A Guide to Methods and Applications. Academic Press, Inc., San Diego, pp 315-322 Wong ALC Beebee TJC Griffiths RA (1994) Factors affecting the distribution and  abundance of an unpigmented heterotrophic alga Prototheca richardsi. Freshw  Biol 32:33-38 46  Chapter 3. Population level analyses indirectly reveal cryptic sex and life-history traits of Pseudoperkinsus sp., an ecological model for basal Opishthokont lineage Ichthyosporea. 1   Introduction Traditionally determination of microbial life-cycles has relied on tenacity, microscopy and real- time observation. Using population genetic techniques and samples of conspecific isolates we have worked in ‗reverse‘ describing aspects of an organisms' life-cycle and ecology based on the footprints left in DNA sequence variation. The opisthokonts (Cavalier-Smith 1987), which include animals and fungi, also encompass seven unicellular lineages (Adl et al. 2005). For these unicellular groups, ploidy and sexual cycles have yet to be elucidated. Reasons for this include a lack of cultures in culture collections and difficulty in identification of con-specifics. Ichthyosporeans are unicellular opisthokonts and may represent the basal-most branch on the animal lineage after the animal/fungus divide (Ruiz- Trillo et al. 2006; Steenkamp et al. 2006; Ruiz-Trillo et al. 2008; Shalchian-Tabrizi et al. 2008). As parasites or commensals, most are dependent on animal hosts (Mendoza et al. 2002). During a survey of marine invertebrate digestive tracts for novel and culturable unicellular opisthokonts, we noticed that ichthyosporeans were often present. We cultured over 100 isolates of an unknown ichthyosporean from the digestive tracts of a variety of marine invertebrates. These isolates were similar in ribosomal small subunit (SSU-rDNA) sequence similarity to the ichthyosporean Pseudoperkinsus tapetis (Figueras et al. 2000; Takishita et al. 2008) and we provisionally call the species 'Pseudoperkinsus sp.' We screened all cultures for internal transcribed ribosomal spacers (ITS) similarity, using 99.6% or greater as our cut-off for putative conspecificity. Rather than adapt cytological methods to a divergent and possibly host adapted organism, we applied population level genetic analyses to our collection of isolates. Using a primer based approach, we found two loci with intraspecific variation, which we used to indirectly determine the ploidy, or chromosome copy number, of our isolates and detect genetic evidence of recombination. Using measures of nucleotide diversity, we estimated the effective  1  A version of this chapter has been submitted. Marshall WL Berbee ML. Population level analyses reveal cryptic sex and life-history traits of Pseudoperkinsus sp., an ecological model for the basal Opisthokont lineage Ichthyosporea. 47  population size and frequency of recombination. We also describe an enquiry into host specificity and transmission.  Genetic evidence of chromosome copy number  Ploidy is integral to a lifecycle but the traditional approach of characterizing ploidy by cytological observation can be difficult to adapt to a newly discovered organism. Flow cytometry can be used to sort haploid from diploid cells (e.g. Bernander et al. 2001), but we had no means to establish values  for 1N and 2N chromatin amounts in the first place. One of the most widely adaptable approaches to determining ploidy has been analysis of the frequency of heterozygotes in microsatellite data sampled from populations (Brondani et al. 2000; Rynearson and Armbrust 2000; van der Verde et al. 2001;Weeks et al. 2001). For example, Oliveira et al. (1998) showed diploidy in Trypanosoma cruzi after cytology gave inconclusive results and Santos and Coffroth (2003) used a lack of heterozygotes from microsatellite data to show that a population of Symbiodinium dinoflagellates was haploid. A similar principle can be applied using sequence data from a population with polymorphisms. Direct sequencing of PCR products from heterozygous diploids results in sequence chromatograms with double peaks at the segregating sites (Clark 1990). Conversely, a lack of heterozygotes at expected frequencies could imply haploidy. The multicellular animals are mainly diploid and the higher fungi are mostly haploid or dikaryotic. To support or reject diploidy in Pseudoperkinsus sp. we carefully compared the observed versus expected frequency of double peaks among the segregating sites in our population.  Genetic evidence of recombination in microeukaryotes  As in most other microorganisms, asexuality is easily observed in the few ichthyosporeans that grow in culture (Whisler 1968; Hibbits 1978; Jostensen et al. 2002; Marshall et al. 2008). Meiosis has yet to be reported in any ichthyosporean. Scaliform conjugation and nuclear fusion have been observed in the uncultured ichthyosporean, Enteropogon sexuale, but the fused cells did not grow or divide (Hibbits 1978). E. sexuale is host obligate, and unfortunately Hibbits‘ observations were limited to short term observations on dissected animals. Sexual recombination, while widespread, can be difficult to observe and is unknown in many unicellular 48  protozoa and fungi (e.g. Reynolds, 1993); however, the inability to find it does not rule out its occurrence.  On the other hand, even when mating can be induced in the lab, it may not necessarily occur in nature. Molecular analysis of variable loci in populations has the advantage of being able to show that sexual recombination has occurred in nature and has revealed evidence for recombination amongst natural populations of pathogenic and free-living bacteria, fungi, and protozoa (Maynard Smith et al. 1993; Burt et al. 1996; Geiser et al. 1998; LoBuglio and Taylor 2002; Xu 2004; Campbell and Carter 2006; Cooper et al. 2007) where sexual cycles were previously unknown. To detect recombination, appropriate delimitation of species and populations is critical (Geiser et al. 1998; Campbell et al. 2005; Saul et al. 2008). If populations are defined too broadly, the barriers to gene flow and linkage disequilibrium may be misinterpreted as clonality. Historical recombination may also be difficult to detect if populations are defined so narrowly that they are dominated by asexual siblings due to proximity (Halliday and Carter 2003), by progeny from recent clonal expansions (Su et al. 2003), by highly inbred strains (Razakandrainibe et al. 2005) or by genetically homogeneous survivors of population bottlenecks (Goodwin et al. 1994). Even with careful consideration of species and population boundaries, finding either exclusively clonal or sexual populations does not rule out the existence of populations of the opposite type (Urdaneta et al. 2001; Cooper et al. 2007). Provided genetic markers capable of resolving different genotypes are available from suitable representatives of a population or species, the occurrence of sex should be evident. Genomes of asexual organisms are inherited as effectively identical copies of the parent. Sexual genomes are shuffled and reassembled by independent assortment of chromosomes and crossing over between paired chromosomes. We tested whether Pseudoperkinsus sp. showed characteristics of recombination including linkage equilibrium and incongruent phylogenetic histories between loci.  Life-cycle and relationship with host Population level analyses have been rare for the protozoa and are lacking altogether for the unicellular opisthokont lineages. Of the ichthyosporea, no genetic studies have been conducted to look further at population structure or host affinities. In order to use population level genetic 49  variation to indirectly characterize the life-cycle of Pseudoperkinsus sp., we sequenced two protein-coding loci with intraspecific variation. By screening sequences for heterozygosity we were able to determine the chromosome copy number. With ploidy established we were able to apply common population genetic tests for evidence of a sexual cycle, population size and structure, and host specificity. Using limited evidence of recombination we were able to infer the frequency of sex by comparing the relative contribution of homologous recombination to nucleotide diversity. In order to determine whether sex was obligate or facultative we compared the average time between recombination events to host age. Using measures of nucleotide diversity we estimated the effective population size and looked for evidence of barriers to gene flow due to host fidelity. Ultimately we were able to combine these findings with culture and isolation observations to propose a life cycle that included the host/symbiont relationship.  Methods Animal collection, isolation of strains and culture conditions. We cultured Pseudoperkinsus sp. from eight of 42 marine invertebrate hosts plus one sample of water surrounding captive oysters from the east and west sides of central Vancouver Island, British Columbia, Canada between August 2004 and April 2007 (Fig. C.1, Table C.1 (Appendix C)). Because geographic population sizes of ichthyosporeans are unknown, we sampled animals from overlapping (cohabiting), near and distant sites (Fig. C.1, Table C.1(Appendix C)). From each collection of 2- 5 conspecific host animals, we dissected and cultured the contents from pooled digestive tracts according to Marshall et al. (2008). For most animals we used pieces from all gut regions; for bivalves we used a portion of the digestive gland and syringed stomach material. We diluted the gut contents in 400 μl of sterile seawater before plating. As a control, we sampled coelomic fluid from 50 % of non-bivalve animals and mantle cavity fluid from 57 % of the bivalves. From water surrounding captive oysters we cultured isolates from the pellet formed after centrifugation of 150 ml of water collected from a 0.5 L container containing three oysters. Ichthyosporean colonies were white and raised and consisted of large round cells. We repeatedly streaked colonies from each original isolate until they were axenic. For isolates CG2 and LE7 we observed asexual reproduction and DAPI stained cells according to Marshall et al. (2008). These 50  strains are available at the American type culture collection (ATCC) under accession numbers PRA-281/282.  DNA extraction, PCR amplification, and sequencing. We extracted DNA from cultures of each individual isolate using a DNeasy® Plant Mini Kit (QIAGEN Inc., Mississauga, Ontario, Canada). We amplified and sequenced the SSU-rDNA, internal transcribed spacers ITS1 and ITS2 and the 5.8S (ITS), and a region of the elongation factor α-like gene (EFL) according to Marshall et al. (2008) except we used internal primers ELF3 and ELR3 (Table C.2, Appendix C) to sequence the EFL.  Initially we sequenced a region of the heat shock protein 70 gene (HSP70) using primers H7150F and H7965R (Table C.2, Appendix C) with an annealing temperature of 48 °C, and cloned it into the vector pCR2.1 using the TOPO TA cloning kit (Invitrogen). For subsequent isolates we amplified and sequenced the HSP70 using H7F2 and H7R2 (Table C.2, Appendix C) with an annealing temperature of 52 °C. Sequences were determined by Macrogen at Seoul, Korea or the Nucleic Acid Protein Service Unit at the University of British Columbia, Canada. To verify point mutations, we sequenced each allele from at least two PCR products and in both directions (Table C.1, Appendix C). We also re-amplified and sequenced between 36 % of HSP70 and 50 % of EFL genes to check for cross contamination (Table C.1, Appendix C). A subset of eight of our isolates (Table S1) were 100% identical in SSU sequence to each other over a maximum overlap of 1718 bp and were 99.9% identical the only other known P. tapetis isolates which were previously collected from Spain and Japan (Figueras et al. 2000; Takishita et al. 2008). All 126 of our isolates shared 99 – 100% ITS identity to each other and to the single isolate of P. tapetis from Spain over a maximum overlap of 667 bp region including ambiguous sites with secondary peaks. Compared to the P. tapetis isolate from Japan our isolates shared 96.9 – 98.8% ITS similarity over a maximum overlap of 675 bp. In contrast, ITS sequences from other ichthyosporeans that we isolated (not shown) were 88.4% or less identical to P. tapetis over a 725 bp region of overlap and were not considered to be conspecific or included in this study. These levels of identity were comparable to the 5.1% interspecific versus 0.51% intraspecific ITS divergence within a data set of Penicillium species (Seifert et al. 2007). We therefore worked under the assumption that our isolates were most likely conspecific and could be identified as P. tapetis. Of our 126 putatively conspecific isolates, we chose 34 isolates for sequencing of the EFL and HSP70 (Table 1). These isolates were selected from all eight 51  animal hosts (Table 1, S1). Because inbreeding can result in elevated levels of homozygosity we increased the number of sequences to 52 for the EFL locus. Genetic analysis. We aligned sequences in Clustal X v1.8 (Thompson et al. 1997) and edited them in MacClade 4.08 (Maddison and Maddison 2000). We calculated pairwise nucleotide diversity, π, and theta per site, θw, (Watterson 1975), Tajima‘s D, and Fu and Li‘s F *  and D *  in DnaSP v.4.50.3 (Rozas et al. 2003) and calculated the ratio of non-synonymous to synonymous substitutions (DN:DS) by eye. To construct parsimony trees we used TCS v.1.13 (Clement et al. 2000) and PAUP v.4.0b10 (Swofford 1998). Parsimony bootstrapping in PAUP involved 500 heuristic search replicates with 10 random sequence additions and tree-bisection reconnection (TBR) branch swapping limited to 10 6  rearrangements. We used the most parsimonious tree from each locus to calculate the consistency index (C. I.), retention index (R. I.), and tree length using PAUP. Chromosome copy number. We aligned all 52 EFL sequences and checked all polymorphic sites for the double peaks indicative of diploid heterozygotes. We calculated the heterozygosity at two adjacent variable sites (Fig. 3.1) that would be expected if Pseudoperkinsus sp. were diploid based on the nucleotide frequencies found for the set of 24 ‗individuals‘ (see below). Definition of data sets. Asexual reproduction was obvious in culture. For analyses where a history of sexual recombination could be obscured by including asexual siblings, we defined three data sets with different degrees of clone correction. In order to eliminate clone siblings that may have been isolated during dissection and culturing, we included only unique haplotypes from each animal collection (hereafter referred to as ‗individuals‘) in the first data set (n=24). Some isolates came from animals collected from the same beach. Assuming clones dispersed locally, we used unique haplotypes from the each location for a ‗location corrected‘ data set (n=21). In order to assess the relationship of the loci independent of population structure a third data set consisted of unique haplotypes (n=18). In addition to allowing for alternative degrees of clone correction, we also applied three alternative definitions of alleles for analyses where results depend on how stringently the alleles are defined (see Fig. 3.2). Animals collected from the east side of Vancouver Island gave rise to a high number of isolates. When isolates with identical haplotypes were cultured from the same animal collection, it was impossible to know whether they originated from one or more initial infections of the host. 52  However, assuming that haplotypes infected randomly and that infections were Poisson distributed based on average haplotype frequency, we used an arithmetic approach to estimate the actual, or ‗expected‘ number of infective individuals from our sequenced isolates. We calculated the expected proportion of each haplotype in each animal collection according to (–ln (1 - f/4)) 4, where f was equal to the number of host collections the haplotype was found in out of a total of four host species.  Recombination, effective population size, and frequency of sex Tests for recombination. Sexual recombination in eukaryotes requires segregation and independent assortment of chromosomes during meiosis resulting in different geneologies for loci on separate chromosomes. We tested for incongruent histories between the HSP70 and the EFL using the incongruence length difference (ILD) test (Farris et al. 1994). The ILD tests a null hypothesis of clonality by comparing the summed tree length of the observed data to the sums after random partitioning. We generated 1000 trees in PAUP using heuristic searches with random sequence addition and TBR branch swapping limited to 10 6 rearrangements.  Our cut-off for significance was p<0.01 and we excluded invariant characters (Cunningham 1997).  To map the history of recombination between the two loci, we made recombination networks (Huson and Kloepper 2005) for the EFL, HSP70, and both regions combined in Splits Tree4 v4.10 (Huson 1998; Huson and Bryant 2006). To minimize homoplasies and keep the number of segregating sites within each locus comparable we only used sequence from the second exon of HSP70 (Fig. 3.1) when the loci were combined. We tested for recombination within each locus and within the combined loci using the ‗Pairwise Homoplasy Index‘ (PHI) (Bruen et al. 2006) implemented in Splits Tree4.  Homologous recombination is a second consequence of meiosis and measurements of covariation between nucleotides provides a basis for estimating the frequency of sexual recombination. We tested both loci using the PHI test (as above). We also compared alternate methods, LARD (Holmes et al. 1999), and Bootscan (Martin et al. 2005a), implemented by RDP3 software (Martin et al. 2005b) after excluding alleles with single unique mutations. We calculated the population recombination parameter (ρ) for HSP70 using the moment method of (Hudson 1987) and R minimum of (Hudson and Kaplan 1985) in DnaSP. We used LDhat  v.2.1 53  (McVean et al. 2002) to calculate ρ using Wakeley‘s moment method (Wakeley 1997) and the pairwise approximate likelihood method (McVean et al. 2002). Effective population size. We calculated the effective population size (Ne) according to Ne = θw/2μ. The upper and lower bounds of the mutation rate per base pair per mitosis (μ) were 0.22 – 2.5 x 10-9, based on ranges from unicellular eukaryotes (Lynch 2006). We calculated overall per site nucleotide variation (θw) from the HSP70 intron and the four-fold degenerate sites from the coding regions of HSP70 and EFL. For comparison, we calculated θw for the EFL using the east and west side isolates separately. Frequency of sex. We estimated the frequency of sex by comparing the relative contributions of mutation and recombination to nucleotide diversity (Tsai et al. 2008). In the standard neutral model Ne can be calculated from the population mutation parameters (as above), or recombination parameters according to Ne  =ρ/2(rσ). The population recombination parameter (ρ) is a measure of overall covariation of nucleotides at different sites and (rσ) is a compound value based on the inferred number of chiasmata in Morgans per kb (M/kb) during meiotic division or per meiosis recombination rate (r) multiplied by the frequency of sex (σ). We used the HSP70 intron to calculate θw to avoid the effects of selection and codon use bias. The per meiosis recombination rate (r) was bound between 0.000017 – 0.0048 M/kb, based on observations from unicellular eukaryotes (Awadella 2003; Lynch 2006). An r value of 10 -3  M/kb corresponds to roughly one chiasmata for a Mb chromosome. We did not account for inbreeding nor consider the series of mitotic divisions between endospore growth and release of new daughter cells.  Clonality and population structure Genetic tests for clonality. Linkage disequilibrium or non-random association among alleles between loci can arise from genetic isolation, inbreeding or lack of recombination, and clonal expansion. We tested the null hypothesis of clonality by calculating the index of association (IA) (Brown et al. 1980; Maynard Smith et al. 1993) using Multilocus v.1.3 (Agapow and Burt 2001). Values range between 0 for randomly associated alleles between loci (due to panmictic mating) or one for complete linkage. To assess significance (p < 0.05) we compared the observed data set to 10, 000 replications that mimicked the effects of random mating. We 54  used Arlequin (Schneider et al. 2000), to test the expected versus observed association for allele pairs between loci. Where tests for linkage are based on allele frequencies phylogenetic methods, such as the parsimony tree length permutation test (PTLPT) (Archie 1989; Burt et al. 1996), utilize all the sequence data.  For the PTLPT, in the absence of recombination the combined tree length for both loci should be significantly shorter (p < 0.05) than trees generated after simulating random mating with random mixing of the alleles. We generated 1000 randomizations using Multilocus with alleles coded as character states for each locus. Following randomization, we replaced the allele codes with the original sequence data and calculated the tree lengths in PAUP according to the specifications provided by Multilocus. Population structure. To test for possible correlations of genetic differentiation with hosts, we calculated dθ using Multilocus. dθ uses allele frequencies and is a variation of Wier‘s (1996) formulation of Wright‘s FST (Agapow and Burt 2001). Subpopulations with dθ close to 0 have similar allele frequencies and those close to one have dissimilar frequencies, which could be due to a reduction in gene flow. We assessed significance using 10, 000 randomizations. We used our sequence data to calculate Snn (Hudson 2000) using DnaSP compared to 10, 000 permutations.  Results The unicellular organism in this study probably represents an undescribed species that we isolated regularly from invertebrate guts. A BLAST search (Altschul et al. 1990) of both SSU and ITS regions showed that our 126 isolates were most closely related to two Pseudoperkinsus tapetis isolates, one from Japan (Takishita et al. 2008) and the other from Spain (Figueras et al. 2000). A 1622 bp SSU-rDNA sequence from five isolates (Table C.1, Appendix C) was identical to P. tapetis (Japan) AB300505 and had one deletion compared to P. tapetis (Spain) AF192386. The ITS sequences were more similar to the isolate from Spain. Out of a total of 975 sequenced sites within the ITS, 29 were frequently ambiguous, with secondary peaks approximately 25 % of the dominant peak within a cleanly sequenced region. However, because the ITS region is present in multiple copies per nucleus, these variants were not analysed further. 55  Ignoring ambiguous sites, 124 isolates (including CG2) were identical and three others, including LE7 and OI22B, differed at three sites. Using a working definition of 100 % SSU identity and 99.6 % or more identity at the ITS, our 126 isolates were considered part of the same operational taxonomic unit.  Use of a single gene, such as the ITS, to define conspecificity can be controversial. Since other ichthyosporeans in our culture collection had a maximum of 88.4 % similarity to our Pseudopoerkinsus sp., we considered it unlikely that we were excluding con- specifics from our data set. However the possibility that we were including more than one species, or hybrids could not be entirely dismissed. Isolates LE7 and OI22B were the only isolates from our analysis with unambiguous differences in ITS sequence and were most divergent in their EFL and HSP70 sequences. We repeated our tests for index of association, PTLPT and ILD without LE7 and OI22B, but the significance of our results was unchanged.  Population variation and lack of heterozygosity Both of the single-locus data matrices had segregating sites and none showed evidence of selection as measured by Tajima‘s D, and Fu and Li‘s F* and D* (Table C.3, Appendix C).  The EFL data matrix, based on the 24 individual isolates, had 922 nucleotides with 19 segregating sites and 10 alleles (Fig. 3.1, 3.2). The amino acid translation was 307 codons in length with a DN:DS ratio of 7:12. All but one non-synonymous substitution was in the first or second position. All synonymous substitutions were in the third position. The HSP70 matrix, based on 24 individuals, consisted of 1006 nucleotides with 64 segregating sites and 11 alleles (Fig. 3.1, 3.2). Two coding regions with a total of 13 synonymous substitutions flanked a 524 bp intron (Fig. 3.1). If our isolates were diploid, assuming Hardy-Weinberg equilibrium, the frequency of double peaks evident as conflicting signals on DNA sequence chromatograms at a chosen site within the EFL (Fig. 1) should have been 56.9 %, but the observed frequency was zero. Neither locus had double peaks at any polymorphic site and therefore we conducted all further analyses under the premise that Pseudoperkinsus sp. was haploid.  56  Recombination, effective population size, and frequency of sex  Genetic evidence for sex. Interlocus phylogenetic conflict pointed to a history of recombination in the population. The parsimony trees from each locus (Fig. 3.2) were incongruent and the tree lengths of the combined data sets were not additive. The C.I., R.I. and lengths of individual matrices were 0.94, 0.99 and 68 for the HSP70 and 1.0, 1.0 and 20 for the EFL. The values for the combined data sets were C.I.= 0.8077, R.I. = 0.9373, length = 104. The difference in the number of steps for each locus was one (EFL) and four (HSP70) steps longer than the minimum possible; however, it was 21 steps longer for the combined data set and the ILD test was significant (p=0.001). Similarly, the combined network for the coding region of the HSP70 and the EFL had several reticulations (Fig. 3.3a), suggestive of inter or intra- chromosomal recombination. A reticulation was detected when a fragment of the second exon of the HSP70 was analyzed alone (Fig. 3.3a), although the evidence for it was not significant according to the PHI test (P=0.2), and the EFL had no reticulations. The PHI test for the combined data was significant (p<0.01). Reticulations and intra-locus recombination were evident in the entire sequenced region of the HSP70 (PHI: p< 0.01) (Fig. 3.3b). Both Bootscan and LARD detected three significant recombination events ( p ≤ 0.05) but neither could confidently distinguish daughter sequences from the minor parent or identify breakpoints. Both methods suggested OI22B as a major parent, BM6 as a minor parent and NO6 and CRG9 were two potential daughters. A third scenario was BM6 as a major parent, OI22B as a minor parent and LE7 a daughter. Estimates of ρ for the HSP70 ranged between 0.285 – 1.6 (Table 3.1). Effective population size. Estimates of nucleotide diversity (π) and θw were higher for the HSP70 than the EFL, although this difference was minimized when only 4-fold degenerate sites were considered (not shown). θw ranged between 0.0126 and 0.0227 using degenerate or intergenic sites (Table 3.2). For the EFL, nucleotide diversity was higher on the west side than the east side and the intergenic region of the HSP70 had higher diversity than the degenerate sites from the coding regions (Table 3.2). Using intergenic and degenerate sites, the Ne was on the order of 10 -6 to 10 -7  depending on the mutation rate (Table 3.2). Frequency of sex. Ratios of sexual to asexual generations or frequency of sex, σ,  ranged between one in 154 to one in 495,000 with a median value of approximately one in 22,700 for ρ = 1 (Fig. 3.4). Median values were based on the two less extreme estimates. Our estimate of ρ 57  was based on a relatively short fragment of DNA and ρ can vary dramatically between loci (Awadella 2003) for reasons including chromosomal position. When we repeated our calculations using a hypothetical high ρ of 50 the frequency of sex ranged between one in three to one in 9, 910 generations with a median value of one in 418 generations. Calculations based on the expected data set including seven additional infective isolates (Fig. 3.5) had a lower frequency of sex with a median value near one in 23, 900 cycles for ρ = 1.  Clonality and population structure Morphological and genetic evidence of clonality. DAPI stained cells undergoing asexual reproduction in culture typically had 32, 64 or more nuclei by the time endospores were visible and that uninucleate endospores grew into multi-nucleate plasmodia before maturation (Fig. 3.6a). Numbers higher than 64 nuclei per cell could not be accurately counted.  The rare cells that had nuclear counts on the order of 250 were often irregular in shape and did not appear to have endospores. Nuclear division was synchronous and nuclei in mature cells were the product of five to nine mitotic divisions. After 3-5 days endospores were usually released through pores or splitting of the parent cell wall (Fig. 3.6b). The population level footprint of asexual reproduction was evident as a significant IA value when all individuals were included (Table 3.3). The IA decreased as allele definitions were adjusted to encompass minor variants but still indicated significant clonality (Table 3.3). The null hypothesis of recombination could not be rejected when the sample was reduced to either the largest subgroup (G1) within the sample (Fig. 3.7a, Table 3.3), to the ‗location corrected‘ data set (Table 3), or the third data set that consisted of unique haplotypes (Table 3.3). Subgroup G2 was collected from different hosts at a single location and varied by a single point mutation in the HSP70 locus. When sequences were used in place of allele codes (PTLPT) the observed tree length was significantly shorter (p<0.023) compared to a permutated data set including all individuals; however, the null hypothesis of recombination could not be rejected using the clone corrected data set (p>0.078). Population structure. We found six unique haplotypes one time each from the west side of Vancouver Island and all the other isolates came from four host populations from three beaches within a 2 km range on the east side of the island (Table C.1, Fig. C.1 (Appendix C)). 58  The 12 different haplotypes found on the east side represented a minimum of 18 independent host colonizations. The number is a minimum because it was calculated by ignoring repeated cultures with the same haplotype from the same host collection under the assumption that genetically identical cells originated and spread from a single initial colonization. However, in some instances, cells of the same haplotype may have colonized independently. Factoring in the probability of multiple independent infections by the same haplotype, increased the possible number of independent colonizations to 25 (Fig. 3.5). Haplotypes were not correlated with host species and dɵ and Snn did not differ significantly from random (not shown). Instead, the same haplotypes were found in more than one host species (Fig. 3.7b). An excess of Pseudoperkinsus sp. haplotypes relative to number of host animals pooled for culture indicated that some host animals contained more than one haplotype (Table C.1, Appendix C). One host sample of five animals had eight different Pseudoperkinsus sp. haplotypes (Table C.1, Appendix C). One animal collection from the west side and one from the east had a number of haplotypes equal to the number of animals sampled (Table C.1, Appendix C). According to the expected data set for infections from three of four host species collected from the east side of Vancouver Island as many as 24 of the 25 expected infections would have occurred in 15 animals (Fig. 3.5, Table C.1, Appendix C).  Discussion The combined results from microscopy and genetic analysis from our collection of con-specifics enabled us to indirectly unveil aspects of the lifecycle and ecology of Pseudoperkinsus sp. that could not be observed in the field or in culture. Our results are the first to introduce ploidy to the life cycle of a basal opisthokont. The complete absence of heterozygotes at polymorphic sites indicated that our Pseudoperkinsus sp. population lives in its invertebrate host primarily as a haploid and that it has a haploid asexual cycle.  If life cycles in other ichthyosporea are predominantly haploid, Hibbits‘ (Hibbits 1978) observations of cells of Enteropogon sexuale undergoing scaliform conjugation and nuclear fusion could be interpreted as fusing haploid cells that form a zygote, likely followed by zygotic meiosis. No diploid asexual cycle was evident in Pseudoperkinsus sp., which may indicate the species has a short lived zygote, or a patchy 59  distribution of haploid and diploid populations, or that diploids for some reason do not colonize animals or grow in culture. An alternate explanation for the lack of observed heterozygotes could be a diploid population that is highly inbred. Johnson et al. (2003) found only one heterozygote amongst 28 isolates from a recombining population of the diploid terrestrial yeast Saccharomyces paradoxus. For Pseudoperkinsus sp. haploidy seems more likely than absolute inbreeding. Considering the large number of alleles, outcrossing would have to be extremely rare to produce 100 % homozygosity. No forms of cellular fusion such as seen in yeasts (Johannsen and van der Walt 1980; Johnson et al. 2003) were observed under our culture conditions. Compared to non-motile terrestrial microorganisms, Pseudoperkinsus sp. may have more opportunities to disperse since cells are carried in currents and tides, and to outcross since host animals ingest cells and concentrate them. Based either on the minimum number of colonizations or on the expected number of colonizations, the number of infections was greater than the number of animals sampled.  Since Pseudoperkinsus sp. was not genetically isolated within its host, it should have potential partners for outcrossing. Evidence for both recombination and clonality The lines of evidence for sexuality rely on the premise that under a clonal model all regions of the genome have the same history. The EFL and HSP70 have significantly incongruent histories, an expected consequence of a meiotic stage with independent chromosome assortment (or recombination between loci on the same chromosome). Tests for clonality, such as the index of association, use panmictic mating as a null hypothesis, and show that panmixis alone is not an adequate explanation for genetic structure.  Pseudoperkinsus sp. has sex frequently enough to obscure the genetic signals of clonality and we were unable to reject recombination once the data were clone corrected. Evidence of intra-locus recombination in the HSP70 provided a second level of support. Intra-locus recombination is detected less frequently than inter-locus recombination and in a haploid organism meiosis is the only opportunity for crossover between single copy, homologous genes. Although difficult to observe directly, evidence for recombination is not unexpected. Theoretical and empirical evidence of the selective benefits of sex is plentiful (e.g. Müller 1964; Chao et al. 1997; Zeyl and Bell 1997; Burt 2000) and meiosis may be present in all extant eukaryotes except for some lineages with recent 60  losses (Ramesh et al. 2005). Candidate ‗ancient asexual‘ lineages are difficult to prove (Judson and Normark 1996) and the only ‗decisively corroborated ancient asexual‘ (Normark et al. 2003) might be the bdelloid rotifers ( Mark Welch and Meselson 2000). The ρ/θw ratio was lower in Pseudoperkinsus sp. (Table 3.1) than in S. paradoxis (0.31- 0.81) (Tsai et al. 2008) or Plasmodium falciparum (0-6.2) (Awadella 2003) but greater than in the predominantly clonal  (Llewellyn et al. 2009) Trypanosoma cruzi (0) (Awadella 2003). In the animals Drosophila melanogaster and Caenorhabditis remanii, the ratios range from 0.6-7.6 (Tsai et al. 2008). Fraser et al. (2007) estimated that ρ/θw ratios greater than 0.25 prevented clonal divergence in bacteria, which rely on homologous recombination for sex. Uniformly low values, below 0.07 in three different methods of analysis, suggest that clonality has been an important force in shaping the genetic structure in our Pseudoperkinsus sp. population. This superimposition of clonality on sexuality is often observed in populations of unicells, e.g. (Ngouanesavanh et al. 2006; Bui et al. 2008). Subgroup G2, which included isolates NO6, MMB5 and MB16, may have been an example of a clonal expansion within a single location incorporating a single point mutation (Figs. 3.2, 3.7). Tibayrenc et al. (1990) proposed four criteria (d-g) to recognize clonality in haploid populations.  We found little evidence for a widespread unique genotype (criterion d) as expected under clonality. Evidence in favour of clonality came from correlated independent genetic markers (criterion g) and significant linkage between alleles at the two loci (criterion f) when all Pseudoperkinsus sp. individuals were considered.  Also suggesting clonality, all four gametic types were never found for any pair of alleles (criterion e), but allele diversity was high and the absence of some of the expected recombinant combinations may have been due to insufficient sampling.  Frequency of sex The ratio of sexual to asexual cycles is an estimate based on a series of assumptions. Inbreeding coefficients and the number of mitotic divisions before endospore release within the host environment are unknown for Pseudoperkinsus sp. Adjustment for these factors would increase estimates of the relative frequency of mating. For example, several mitotic divisions occur before endospore release in culture. If a single asexual generation is defined according to each spore release rather than each mitotic division the relative frequency of mating could be up 61  to 10 times higher. As inbreeding frequency increases, its effects become more pronounced. Using Ne = ρ/2r (1-F) and Ne = θ (1+F)/2μ where F is the inbreeding coefficient (Tsai et al. 2008), a coefficient of 0.5 would increase the ρ/θw ratios and mating frequency by ~1.33 times whereas a coefficient of 0.9 would increase them by ~ five times. Regardless of the margins of error, Pseudoperkinsus sp. is most likely facultatively sexual and can probably undergo unlimited rounds of asexual reproduction, much like a yeast. Outcrossing rates in yeasts have been estimated at once per 3000 – 9000 asexual cycles for S. paradoxus (Tsai et al. 2008) and once per 50 000 asexual cycles for Saccharomyces cerevisiae (Ruderfer et al. 2006). Some protists have an obligate sexual stage because of constraints such as morphology (e.g. diatom frustule size), development (alternation of generations in foramniferans), or host relationships (e.g. some apicomplexans). Even given our wide range of parameters, if Pseudoperkinsus sp. had a similar obligate sexual cycle we would have expected to see a lower bound for the number of asexual cycles per sexual cycle.  Population size and host relationship According to our estimates of the frequency of sex, infection of a host animal does not depend on a sexual stage. If Pseudoperkinsus sp. undergoes about two mitotic divisions per day in the host as it does in culture, it can undergo approximately 700 mitotic divisions per year. Food can pass through a bivalve digestive tract within a day (Ren et al. 2006) which is less than maturation time for Pseudoperkinsus sp. in culture. Yet Pseudoperkinsus sp. was never isolated from undiluted mantle cavity fluid, suggesting that the gut is its main habitat in the animal. Anurofeca richardsi, the closest relative to Pseudoperkinsus sp. with any in situ observations, grows and divides in tadpole digestive tracts (Beebee and Wong 1993). Oysters on the east side of Vancouver Island had no isolates, in contrast to other co-habiting bivalves and to oysters on the west side. Given this variation it seems likely that Pseudoperkinsus sp. is neither retained indefinitely nor cumulatively accrued. Assuming an average retention time of a year and our median estimates of frequency of sex, Pseudoperkinsus sp. may pass through several hosts before mating. If sex was obligate and infection limited to sexually derived resting spores or zygotes, the retention time necessary before sexual reproduction could be longer than the host lifespan (Fig. 4). 62  The average level of silent site pairwise nucleotide diversity (θw) was within the range (0.0103 – 0.2522) but lower than the average value (0.057), of other unicellular eukaryotes (Lynch 2006). The effective population size of Pseudoperkinsus sp. was on the order of 10 6 to 10 7 , which is typical for unicellular eukaryotes (Lynch 2006).  Effective population sizes are based on numbers of individuals expected to contribute progeny to the next generation. Because many cells die and do not contribute to the gene pool the effective size is lower than the censused population. Pseudoperkinsus sp. can grow and divide for a few days in un-supplemented seawater but cells arrest at uni- and bi-nucleate stages. Re-infection is probably a limiting factor in Pseudoperkinsus sp.‘s demographics.In 2003, Lynch and Conery speculated that Ne values for parasitic or pathogenic taxa may be lower than those of freeliving taxa due to dependence on a multicellular host. Initially this was supported by observations of lower estimates of nucleotide diversity from pathogens (Lynch 2006) compared to higher estimates from free-living ciliates (Snoke et al. 2006; Barth et al. 2008). However, Catania et al. (2009) found no difference between parasitic species and free living ciliates after applying stricter species delimitations. Overall there seems to be no obvious difference between Pseudoperkinsus sp. and other microeukaryotes whether they be symbionts, phototrophs (Smith and Lee 2008), or heterotrophs. The challenge of colonization and re-infection faced by symbionts such as Pseudoperkinsus sp. may be offset by factors such as reduced competition, ample supply of nutrients or the ability to proliferate and mate once past the re-infection hurdle. While we were unable to definitively distinguish whether identical haplotypes from a single animal collection originated from multiple infections, from endospore release during dissection, or from asexual reproduction during host colonization, our large total number of isolates implied that Pseudoperkinsus sp. grew actively within the guts. Cases where we found multiple isolates from several genotypes in a single host suggested that reproduction occurred before dissection. Whether Pseudoperkinsus sp. has host related restrictions or not, its large effective population size implies that it was sampled from its natural habitat. Nucleotide diversity and effective population size decrease during population bottlenecks such as recent colonization (Smith 1991; Rich and Ayala 2000). The presence of a recombining populations also suggests that Pseudoperkinsus sp was in its natural ecological niche (Campbell et al. 2005). Recombination in Giardia was shown once isolates were collected from an endemic area with high transmission rates and mixed infections (Cooper et al. 2007). Although most frequently isolated from bivalves, (Figueras et al. 2000; Takishita et al. 2008), 63  Pseudoperkinsus showed no obvious detriment to its hosts. No evidence of host specificity was evident based on either the collection data or the genetic structure. Our finding of a large effective population size with evidence of sexual recombination suggests that Pseudoperkinsus sp. is an overlooked endemic of a wide host range of invertebrates compared to the clonal and genetically homogenous population structure expected of a recent invader.  Conclusions We have demonstrated, for the first time amongst basal unicellular opisthokonts, that sexual recombination occurs and that sex is most likely facultative.  Given a low frequency of sex, Pseudoperkinsus sp. most likely enters the host as an asexually derived endospore. Asexually derived endospores have been described in most ichthyosporea (Mendoza et al. 2002) and formation of a multi-nucleate parent cell, followed by endospore release is common amongst ichthyosporeans (Chatton 1920, Lichtwardt and Chen 1964, Marshall et al. 2008). Ichthyosporeans are one of few host dependent lineages among the opisthokonts. Phylogenetically, multicellular animals originated after the ichthyosporean lineage diverged and so the earliest ichthyosporeans could not have had metazoan hosts. The lack of host specificity in Pseudoperkinsus may reflect the relatively recent evolution of ichthyosporean dependence on metazoan hosts. A potentially haploid lineage at the base of the metazoan tree contrasts with the animals diploid nature and brings the ploidy of the metazoan ancestor to question.  Acknowledgements We thank S. Otto for helpful discussions on coalescent theory, W. Maddison for Poisson probability calculations and L. Rieseberg for comments on an earlier version of this manuscript. This work was supported by the Natural Sciences and Research Council of Canada and the Bamfield Marine Sciences Centre. 64  Table 3.1. Estimates of effective population size times the frequency of sex (N e) ŧŧ  based on several estimates of population recombination parameter ρ *. The ρ/θw **  ratio was calculated using θw from the HSP70 intron and Neσ based the upper and lower bounds of per meiosis recombination rate, r * .  Effective population size x frequency of sex (Neσ)  ŧŧ Method ρ * ρ/θw r *  = 0.000017  r *  =  0.0048 Wakeley 1997 0.285 0.0136 8, 380 30 Rm / Pairwise 1 0.044 29, 400 104 Hudson 1987 1.6 0.070 47, 100 167 ŧŧ (Ne) is the effective population size, Ne, times the frequency of sex, . (Ne) is calculated based on Ne = ρ/2r. *ρ and r are expressed in Morgans per kb (M/kb). **  θw  is Watterson‘s (1975) estimator of nucleotide diversity expressed per kb. 65  Table 3.2. Effective population sizes (Ne) based on intergenic and 4 fold-degenerate sites given an upper and lower bound for mutation rate μ. Nucleotide diversity (θw) was calculated from all 24 individuals.   Locus and sites used (number of nucleotides) θw Ne given μ = 0.22 x 10 -9  Ne  given μ = 2.5 x 10 -9 HSP70 coding (74) 0.0179 40, 680, 000 3, 580, 000 HSP70 intron (524) 0.0227 51, 590, 000 4, 540, 000 EFL coding, All (155) 0.0126 29, 090, 000 2, 560, 000 EFL coding, east side Vancouver Island (155) 0.0110  25, 000, 000  2, 200, 000  EFL coding, west side Vancouver Island (155) 0.0176 31, 360, 000 2, 760, 000 66  Table 3.3. Index of association between HSP70 and EFL alleles. Significant association between HSP70 and EFL alleles (with P < 0.05 or lower) was evident when all individual isolates were included, but not when datasets were clone corrected or location corrected. Allele definitions Isolates included # of isolates IA P All sites considered All individuals (Fig. 7a-i) 24 0.1814 0.015 Alleles reduced by 1 variable site (Fig. 3) All individuals 24 0.1482 0.025 Alleles defined by balanced sites (Fig. 3) All individuals 24 0.1328 0.036 All sites considered Fewer individuals (Fig. 7a-ii) 22 0.1726 0.023 All sites considered Fewer individuals (Fig. 7a- iii) 19 0.1210 0.095 All sites considered Location corrected 21 0.0717 0.23 All sites considered Unique haplotypes 18 -0.0736 1.0  67    Figure 3.1. Polymorphic sites from the amplified regions of the EFL and HSP70 loci shown for each of the 24 individuals.  A small ‗g‘ denotes a gap, each gap was coded as one character regardless of its length. Each individual was assigned one of 18 unique haplotypes defined after the EFL and HSP70 sequences were concatenated. The non-coding region of the HSP70 is shaded in grey. The polymorphic region used to calculate hypothetical heterozygosity of the EFL is outlined by a box. Balanced sites were defined as any site where polymorphisms were present in at least three individuals and are marked by an asterisk above the consensus line. Allele and haplotype numbers correspond with Figures 3.2, 3.5 and 3.7. Specific collection locations are listed in table C.1, Appendix C.   EFL HSP70 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Consensus Allele #C C C C C T A A A C G C C C G C C C C Allele #G G T A C G T C T T T C T A T C T A A C T T T C C A T G C T T T C G T C G C T T G A T G G C A C C C T G G C G T C C C T G T T T Haplotype # Location Host CG5 1 T A -- -- -- -- G -- -- -- -- -- -- A A ? ? ? ? 4 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 8 West Oyster KS1 1 T A -- -- -- -- G -- -- -- -- -- -- A A -- -- -- -- 2 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 9 West Chiton CG2 2 T A -- -- -- -- G -- -- -- -- -- -- A -- -- -- -- -- 3 ? -- -- -- -- -- -- -- -- -- -- -- -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- ? 7 West Oyster GM8 2 T A -- -- -- -- G -- -- -- -- -- -- A -- -- -- -- -- 6 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- C 1 East-1 Mussel MB4 2 T A -- -- -- -- G -- -- -- -- -- -- A -- -- ? ? ? 6 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- ? 1 East-2 Mussel BM33 3 -- -- -- -- -- -- G -- -- -- -- -- -- -- -- -- ? ? ? 6 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- C 4 East-2 Mussel BM19 3 -- -- -- -- -- -- G -- -- -- -- -- -- -- -- -- ? ? ? 4 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 5 East-2 Mussel NO32 3 -- -- -- -- -- -- G -- -- -- -- -- -- -- -- -- -- -- -- 4 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 5 East-2 Varnish Clam PV23 3 -- -- -- -- -- -- G -- -- -- -- -- -- -- -- -- -- -- -- 4 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 5 East-3 Manilla Clam VP19 3 -- -- -- -- -- -- G -- -- -- -- -- -- -- -- -- -- -- -- 5 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- T -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- A T T -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 11 East-3 Manilla Clam BM6 4 -- -- -- -- -- -- -- T T -- -- -- -- -- -- -- -- -- -- 6 A -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- C 3 East-2 Mussel PV25 4 -- -- -- -- -- -- -- T T -- -- -- -- -- -- -- -- -- -- 6 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- C 3 East-3 Manilla Clam NO6 5 -- -- -- -- -- -- -- C -- -- -- -- -- -- -- -- -- -- -- 8 ? T g -- -- C C -- G -- C -- C -- C T -- T G T C -- G T -- G C -- -- -- -- -- -- -- G T A T -- A -- T G -- T -- -- -- T T C -- A -- -- -- -- -- -- -- -- -- -- -- 12 East-2 Varnish Clam MMB5 5 -- -- -- -- -- -- -- C -- -- -- -- -- -- -- -- -- -- -- 8 ? T g -- -- C C -- G -- C -- C -- C T -- T G T C -- G T -- G C -- -- -- -- -- -- -- G T A T -- A -- T G -- T -- -- -- T T C -- A -- -- -- -- -- -- -- -- -- -- -- 12 East-2 Mussel ON1M 6 -- -- -- -- -- -- -- C -- -- -- -- G -- -- -- -- -- -- 1 ? -- -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- C -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 10 East-2 Varnish Clam PV3 6 -- -- -- -- -- -- -- C -- -- -- -- G -- -- -- -- -- -- 6 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- C 2 East-3 Manilla Clam MB16 7 -- -- T -- -- -- -- C -- -- -- -- -- -- -- -- -- -- -- 8 -- T g -- -- C C -- G -- C -- C -- C T -- T G T C -- G T -- G C -- -- -- -- -- -- -- G T A T -- A -- T G -- T -- -- -- T T C -- A -- -- -- -- -- -- -- -- -- -- -- 14 East-2 Mussel BM17 8 -- -- -- -- -- -- -- C -- -- -- -- -- -- -- -- -- T T 9 ? T g -- T C C -- G -- C -- C -- C T -- T G T C -- G T -- G C -- -- -- -- -- -- -- G T A T -- A -- T G -- T -- -- -- T T C -- A -- -- -- -- -- -- -- -- -- -- -- 13 East-2 Mussel ON12 8 -- -- -- -- -- -- -- C -- -- -- -- -- -- -- -- -- T ? 9 -- T g -- T C C -- G -- C -- C -- C T -- T G T C -- G T -- G C -- -- -- -- -- -- -- G T A T -- A -- T G -- T -- -- -- T T C -- A -- -- -- -- -- -- -- -- -- -- -- 13 East-2 Varnish Clam MB29 8 -- -- -- -- -- -- -- C -- -- -- -- -- -- -- -- -- T T 4 ? -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- G C -- T -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- A -- -- -- -- -- T T -- -- -- -- -- 6 East-2 Mussel NO16 8 -- -- -- -- -- -- -- C -- -- -- -- -- -- -- -- -- T T 7 -- T g -- -- C C -- G -- C -- C -- C T C T G T -- -- G T -- G C -- -- -- -- -- -- A G T A T -- A -- T G -- T -- -- -- T T C -- A -- A -- -- T -- -- -- -- -- -- 16 East-2 Varnish Clam CRG9 9 -- -- -- A T -- -- C -- -- A A -- -- -- -- -- -- -- 7 ? T g -- -- C C -- G -- C -- C -- C T C T G T -- -- G T -- G C -- -- -- -- -- -- A G T A T -- A -- T G -- T -- -- -- T T C -- A -- A -- -- T -- -- -- -- -- -- 15 West Oyster LE7 10 -- -- -- -- -- C -- -- -- A -- -- -- -- -- A T -- -- 10 ? T g -- -- C C -- G C C T C -- C T -- T G T C -- G T -- G C -- -- -- -- -- -- -- G T A T -- A -- T G -- T -- -- -- T T C -- A -- -- -- T -- -- -- A C C -- 17 West Sea Cucumber OI22B 10 -- -- -- -- -- C -- -- -- A -- -- -- -- -- A T ? ? 11 ? T C -- -- C -- T G -- C -- -- -- -- T -- T G T -- -- G T -- G C A T -- -- G -- -- -- T A T C A A -- -- -- T -- -- -- T T C -- A T -- C T -- -- A -- C C -- 18 West Oyster Water HSP70 KS1 2 ON1M 1 CG2 3 4 VP19 5 BM6 BM33 GM8 MB4 PV3 PV25 6 BM17 ON12 9 CRG9 NO16 7 MB16 MMB5 NO6 8 LE7 10 OI22B 11 A B C D E F G H H1 H2 H3 H4 H5 2 1 BM19 CG5 MB29 NO32 PV23 1 4 5 1 35 24 4 1 3 1 3 13 4 9 12 4 (100) (100) (100) (98) (92) (92)C. I. = 0.94 R. I. = 0.99 S = 64 L = 68 (59) (62) E1 E4 BM17 MB29 NO16 ON12 8 NO6 MMB5 5 BM6 PV25 4 ON1M PV3 6 MB16 7LE7 OI22B 10 1 1 1 1 1 2 4 4 DC F E3 E5 CG5 KS1 9 1 3 B A E2 CG2 GM8 MB4 BM19 BM33 NO32 PV23 VP19 1 2 3 EFL (65) (94) (63) (97) (80) C. I. = 1.0 R. I. = 1.0 S =19 L = 20 (63) (73) CRG9 9 E  Figure 3.2. Genealogical networks of the HSP70 and EFL loci. The genealogical network from the HSP70 locus (top) conflicts with the network from the EFL locus (bottom), suggesting a history of recombination between the two loci. The HSP70 haplotypes fell into three networks, designated in blue, black, and red. Each alphanumeric code designates an isolate. Isolate codes in the EFL network retain their colour from the HSP70 network to illustrate incongruence.  Circles around isolates with identical sequences designate 11 HSP70 and 10 EFL alleles. To minimize the effects of 69  recent somatic mutations, two alternative approaches to designating alleles were used: 1) sequences with only one unique difference were lumped together as the same allele (dashed boxes A-H), or 2) alleles were defined using only balanced sites (Fig. 1) (solid boxes E/H1-5). Numbers above branches are branch lengths; numbers below are parsimony bootstrap percentages.  Consistency index (C.I.), Retention Index (R.I.), segregating sites (S), and tree length (L) are shown for both trees.  Networks were generated by TCS (Clements 2001). Dashed lines on HSP70 network show connectivity after the connection limit was fixed at 60 steps.  70  BM17 NO16 CRG9 LE7 OI22B MB16 NO6 BM17 BM6 BM33 GM8 CG2 KS1 CG5 MB29 BM6 BM33 KS1 LE7 OI22B MB16 NO6 GM8 CG2 CG5 MB29 PV3 PV23 VP19 ON1M NO16 CRG9 PV3 PV23 VP19 ON1M HSP70 HSP70 + EFL a) Partial HSP70 b) Complete HSP70 OI22B LE7 NO6 BM17 MB16 CRG9 NO16 BM6 BM33 GM8 PV3 ON1M VP19 CG5 MB29 KS1CG2 PV23  Figure 3.3. Inter- and intra- locus recombination reconstructed using Splits Tree4. Edges representing potential recombination events are shown in bold grey and networks are not drawn to scale. a) Networks from the combined EFL plus the HSP70 exon 2 region, compared to the exon 2 region of HSP70 region alone (inset) . The dashed lines represents edges shared by both trees. b) Recombination network using the all 1006 base pairs from our HSP70 sequence.  71  1 10-2 10-3 10-4 10-5 10-6 0.285 1 F re q u e n c y  o f s e x 10-1 542 6150 153 000 1 740 000 * 43 600 495 000 154 1750 * 10-7 1.6 1 100 27 200 310 000 * 96 Population Recombination Factor ( ) 79 600 1 year 5 years 10 years 22 700 14 100 Retention time  Figure 3.4. Range of possible frequencies of sexual reproduction in Pseudoperkinsus sp. Range of possible frequencies of sexual reproduction in Pseudoperkinsus sp., given three estimates for the population recombination factor (ρ) calculated for the heat shock protein 70 locus (x axis) where a y axis value of 1 is equivalent to no asexual reproduction. The vertical ranges represent alternative assumptions about the per replication recombination rate (r) and mutation rate (μ). Black symbols represent μ = 2.5 x 10-9, white symbols μ = 2.2 x 10 -10  and rectangles represent r = 4.8 x 10 -3 , circles r = 1.7 x 10 -5 . Median values (*) were calculated between the two least extreme mutation and recombination rate combinations. If sexual reproduction was less frequent than once in 700 nuclear divisions (shaded region), and assuming that Pseudoperkinsus is retained in its host for a year, then initial host infection is probably via asexual spores. The dotted lines are cut-offs, below which asexual spores would be constitute the infective stage given retention times or host life-spans of 1, 5 and 10 years. 90 % inbreeding would increase the range of frequency of sex by a factor of ~ 5.  72  0 1 2 3 4 5 6 7 Haplotype # N u m b er  o f Is o la te s # cultured individuals expected 1 2 3 4 5 6 10 11 12 13 14 16  Figure 3.5. Numbers of isolates belonging to each haplotype that were isolated from four host animal collections on the east side of Vancouver Island. The total number of cultured isolates of each haplotype (n=28) is compared to the number of isolates defined as ‗individuals‘ (n=18), and the ‗expected‘ number of individuals (n=25) based on the probability of multiple infections. Expected values were rounded to the nearest whole number. Haplotype # was based on the combined EFL and HSP70 data and corresponds to Figure 3.1 and table C.1.  73   Figure 3.6. Asexual reproduction in Pseudoperkinsus sp. a. (i-vi) DAPI stain of nuclear proliferation in Pseudoperkinsus sp. Cells start as uni- nucleate endospores, nuclei divide synchronously producing 2, 4, 8, 16, etc. multinucleate stages. Scale bar equals 20 microns. b and c. Uni-nucleate non-motile endospores are released (b) by squeezing through pores in the parent cell wall or (c) when parent cell wall splits into two valves. Scale bars equal 20 and 10 microns.    74    IA=0.18   p=0.02 H IA= 0.12 p= 0.10   IA= 0.17   p= 0.02i) ii) iii) 1 3 6 4 2 9 7 5 6 2 4 3 1 8 9 8 5 7 10 11 10 E G1 G2 G3 a) E1 E2E3E4E5E6 E7E8 E9E10 H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 1 1 1 1 1 2 2 1 1 1 1 1 3 1 12 1 2 b) Figure 3.7. Some combinations of alleles from the HSP70 locus (H) and the EFL locus (E) occurred at higher than expected frequencies, suggesting a certain level of clonality in the population. a. Numbers in circles are allele codes and their colours are from the HSP70 genealogy (see Fig. 3.2). Alleles are shared within groups G1, G2 and G3 but not among the groups. G2 includes isolates NO6, MMB5 and MB16; G3 includes isolates LE7 and OI22B. Solid lines specify allele combinations that appeared at frequencies that were significantly higher than expected.  The index of association (IA) is given for (i) all isolates, (ii) for G1/G2, and (iii) for G1 considered separately. b. For the 24 ‗individual‘ isolates, the combinations of alleles found together at significantly higher frequencies than would be expected under random mating are indicated by black boxes. Numbers in boxes indicate number of host animal species in which the allele combination was found. Finding the same allele combinations in different hosts supports a lack of host specificity.  75  References Adl S Simpson AG Farmer MA Anderson RA Anderson OR Barta JR Bowser SS Brugerolle G Fensome RA Fredericq S James TY Karpov SA Kugrens P Krug J Lane CE Lewis LA Lodge J Lynn DH Mann DG McCourt RM Mendoza L Moestrup Ø Mozley-Standridge SE Nerad TA Shearer CA Smirnov AV Spiegel FW Taylor M (2005) The new higher level classification of eukaryotes with emphasis on the taxonomy of protists. J Eukaryot Microbiol 52:399-451 Agapow PM Burt A (2001) Indices of multilocus linkage disequilibrium. Mol Ecol Notes 1:101-102 Altschul SF Gish W Miller W Meyers EN Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215:403-410 Archie JW (1989) A randomization test for phylogenetic information in systematic data. Syst Zool 38:239-252 Awadella P (2003) The evolutionary genomics of pathogen recombination. Nat Rev Genet 4:50-60 Barth D Tischer K Berger H Schlegel M Berendonk TU (2008) High mitochondrial haplotype diversity of Coleps sp. (Ciliophora: Prostomatida). Environ Microbiol 10:626-634 Beebee T Wong ALC (1993) Stimulation of cell division and DNA replication in Prototheca richardsi by passage through larval amphibian guts. Parasitology 107:119-124 Bernander R Palm JED Svärd SG (2001) Genome ploidy in different stages of the Giardia lamblia life cycle. Cell Microbiol 3:55-62 Brondani C Brondani RPV Garrido LR Ferreira ME (2000) Development of microsatellite markers for the genetic analysis of Magnaporthe grisea. Genet Mol Biol 23:753-762 Brown AHD Feldman MW Nevo E (1980) Multilocus structure of natural populations of Hordeum spontaneum. Genetics 96:523-536 Bruen T Philippe H Bryant D (2006) A simple and robust statistical test for detecting the presence of recombination. Genetics 172:2665-2681 76  Bui T Lin X Malik R Heitman J Carter D (2008) Isolates of Cryptococcus neoformans from infected animals reveal genetic exchange in unisexual, a mating type populations. Eukaryot Cell 7:1771-1780 Burt A (2000) Perspective: sex, recombination, and the efficacy of selection -- was Weismann right? Evolution 54:337-351 Burt A Carter DA Koenig GL White TJ Taylor JW (1996) Molecular markers reveal cryptic sex in human pathogen Coccidiodes immitis. Proc Natl Acad Sci USA 93:770-773 Campbell LT Carter D (2006) Looking for sex in the fungal pathogens Cryptococcus neoformans and Cryptococcus gattii. FEMS Yeast Res 6:588-598 Campbell LT Currie BJ Krockenberger M Malik R Meyer W Heitman J Carter D (2005) Clonality and recombination in genetically differentiated subgroups of Cryptococcus gattii. Eukaryot Cell 4:1403-1409 Catania F Wurmser F Potekhin AA Przyboś E Lynch M (2009) Genetic diversity in the Paramecium aurelia species complex. Mol Biol Evol 26:421-431 Cavalier-Smith T (1987) The origin of fungi and pseudofungi. Pages 339-353 in A. Rayner, C. Brasier, and D. Moore, editors. Evolutionary biology of fungi. Cambridge University Press, Cambridge Chao L Tran TT Tran TT (1997) The advantage of sex in the RNA virus phi6. Genetics 147:953-959 Chatton E (1920) Les peridiniens parasites, morphologie, reproduction, ethologie. Arch Zool Exp Genet 59:1-475 Clark A (1990) Inference of haplotypes from PCR-amplified samples of diploid populations. Mol Biol Evol 7:111-122 Clement M Posada D Crandall KA (2000) TCS: a computer program to estimate gene geneologies. Mol Evol 9:1657-1659 Cooper MA Adam RD Worobey M Sterling CR (2007) Population genetics provides evidence for recombination in Giardia. Curr Biol 17:1984-1988 Cunningham CW (1997) Can three incongruence tests predict when data should be combined? Molecular Biology and Evolution 14:733-740. 77  Farris JS Källersjö M Kuge AG Bult C (1994) Testing significance of incongruence. Cladistics 10:315-319 Figueras A Lorenzo G Ordas MC Gouy M Novoa B (2000) Sequence of the small subunit ribosomal DNA gene of Pseudoperkinsus tapetis isolated carpet shell clam in Galicia (NW of Spain). Mar Biotechnol 2:419-428 Fraser C Hanage WP Spratt BG (2007) Recombination and the nature of bacterial speciation. Science 315:476-480 Geiser DM Pitt JI Taylor JW (1998) Cryptic speciation and recombination in the aflatoxin-producing fungus Aspergillus flavus. Proc Natl Acad Sci USA 95:388- 393 Goodwin SB Cohen BA Fry WE (1994) Panglobal distribution of a single clonal lineage of the Irish potato famine fungus. Proc Natl Acad Sci USA 91:11591- 11595 Halliday C Carter D (2003) Clonal reproduction and limited dispersal in an environmental population of Cryptococcus neoformans var. gattii strains from Australia. J Clin Microbiol 41:703-711 Hibbits J (1978) Marine Eccrinales (Trichomycetes) found in crustaceans of the San Juan Archipelago, Washington. Syesis 11:213-261 Holmes EC Worobey M Rambaut A (1999) Phylogenetic evidence for recombination in dengue virus. Mol Biol Evol 16:405-409 Hudson RR (1987) Estimating the recombination parameter of a finite population model without selection. Genet Res 50:245-250 Hudson RR (2000) A new statistic for detecting genetic differentiation. Genetics 155:2011-2014 Hudson RR Kaplan N (1985) Statistical properties of the number of recombination events in the history of a sample of DNA sequences. Genet 111:147-164 Huson DH (1998) SplitsTree: a program for analyzing and visualizing evolutionary data. Bioinformatics 14:68-73 Huson DH Bryant D (2006) Application of phylogenetic networks in evolutionary studies. Mol Biol Evol 23:254-267 78  Huson DH Kloepper TH (2005) Computing recombination networks from binary sequences. Bioinformatics 21:ii159-ii165 Johannsen E van der Walt JP (1980) Hybridization studies within the genus Saccharomyces Klocker. Can J Microbiol 26:1199-1203 Johnson LJ Koufopanou V Goddard MR Hetherington R Schäfer SM Burt A (2003) Population genetics of the wild yeast Saccharomyces paradoxus. Genetics 66:43- 52 Jostensen JP Sperstad S Johansen S Landfald B (2002) Molecular-phylogenetic, structural and biochemical features of a cold adapted, marine ichthysporean near the animal-fungal divergence, described from in vitro cultures. Europ J Protistol 38:93-104 Judson OP Normark BB (1996) Ancient asexual scandals. Trends Ecol Evol 11:41-46 Lichtwardt RW Chen AW (1964) A Parataeniella (Trichomycetes, Eccrinales) in an Isopod. Mycologia 56:163-169 Llewellyn MS Miles MA Carrasco HJ Lewis MD Yeo M Vargas J Torrico F Diosque P Valente V Valente SA Gaunt MW (2009) Genome-scale multilocus microsatellite typing of Trypanosoma cruzi discrete typing unit 1 reveals phylogeographic structure and specific genotypes linked to human infection. PLoS Pathog 5:e1000410 LoBuglio KF Taylor JW (2002) Recombination and genetic differentiation in the mycorrhizal fungus Cenococcum geophilum Fr. Mycologia 94:772-780 Lynch M (2006) The origins of eukaryotic gene structure. Mol Biol Evol 23:450-468 Lynch M Conery JS (2003) The origins of genome complexity. Science 302:1401-1404 Maddison WP Maddison DR (2000) MacClade 4: Interactive analysis of phylogeny and character evolution. Sinauer, Sunderland, MA, USA Mark Welch DB Meselson MS (2000) Rates of nucleotide substitution in sexual and anciently asexual rotifers. Science 288:1211-1215 Marshall WL Celio G McLaughlin DJ Berbee ML (2008) Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. Protist 159:415-433 79  Martin D Posada D Crandall KA Williamson C (2005a) A modified bootscan algorithm for automated identification of recombinant sequences and recombination breakpoints. AIDS Res Hum Retroviruses 21:98-102 Martin D Williamson C Posada D (2005b) RDP2: recombination detection and analysis from sequence alignments. Bioinformatics 21:260-262 Maynard Smith J Smith NH O'Rourke M Spratt BG (1993) How clonal are bacteria? Proc Natl Acad Sci USA 90:4384-4388 McVean G Awadella P Fearnhead P (2002) A coalescent-based method for detecting and estimating recombination from gene sequences. Genetics 160:1231-1241 Mendoza L Taylor JW Ajello L (2002) The class Mesomycetozoea: a heterogeneous group of microorganisms at the animal-fungal boundary. Ann Rev Microbiol 56:315-344 Müller HJ (1964) The relation of recombination to mutational advance. Mutat Res 1:2-9 Ngouanesavanh T Guyot K Certad G Le Fichoux Y Chartier C Verdier R-I Cailliez J-C Camus D Dei-Cas E Bañuls A-L (2006) Cryptosporidium population genetics: evidence of clonality in isolates from France and Haiti. J Eukaryot Microbiol 53:S33-S36 Normark BB Judson OP Moran NA (2003) Genomic signatures of ancient asexual lineages. Biol J Lin Soc 79:69-84 Oliveira RP Broude NE Macedo AM Cantor CR Smith CL Pena SDJ (1998) Probing the population genetic structure of Trypanosoma cruzi with polymorphic microsatellites. Proc Natl Acad Sci USA 95:3776-3780 Ramesh MA Malik SB Logsdon JMJ (2005) A phylogenomic inventory of meiotic genes: evidence for sex in Giardia and an early eukaryotic origin of meiosis. Curr Biol 15:185-191 Razakandrainibe FG Durand P Koella JC De MT Rousset F Ayala FJ Renaud F (2005) "Clonal" population structure of the malaria agent Plasmodium falciparum in high-infection regions. Proc Natl Acad Sci USA 102:19388-17393 Ren JS Ross AH Hayden BJ (2006) Comparison of assimilation efficiency on diets of nine phytoplankton species of the greenshell mussel  Perna canaliculus. J Shellfish Res 25:887-892. 80  Reynolds DR (1993) The fungal holomorph: an overview. Pages 15-26 in D. R. Reynolds and J. W. Taylor, editors. The fungal holomorph: a consideration of mitotic and pleomorphic speciation. CAB International., Wallingford, U. K. Rich SM Ayala FJ (2000) Population structure and recent evolution of Plasmodium falciparum. Proc Natl Acad Sci USA 97:6994-7001 Rozas J Sanchez-DelBarrio JC Messeguer X Rozas R (2003) DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19:2496-2497 Ruderfer DM Pratt SC Seidel HS Kruglyak L (2006) Population genomic analysis of outcrossing and recombination in yeast. Nat genet 38:1077-1081 Ruiz-Trillo I Lane CE Archibald JM Roger AJ (2006) Insights into the evolutionary origin and genome architecture of the unicellular opisthokonts Capsaspora owczarzaki and Sphaeroforma arctica. J Eukaryot Microbiol 53:1-6 Ruiz-Trillo I Roger AJ Burger G Gray MW Lang BF (2008) A phylogenomic investigation into the origin on metazoa. Mol Biol Evol 25:664-672 Rynearson TA Armbrust EV (2000) DNA fingerprinting reveals extensive genetic diversity in a field population of the centric diatom Ditylum brightwelli. Limnol Oceanogr 45:1329-1340 Santos SR Coffroth MA (2003) Molecular genetic evidence that dinoflagellates belonging to the genus Symbiodinium Freudenthal are haploid. Biol Bull 204:10- 20 Saul N Krockenberger M Carter D (2008) Evidence of recombination in mixed- mating-type and a-only populations of Cryptococcus gattii sourced from single Eucalyptus tree hollows. Eukaryot Cell 7:727-734 Schneider S Roessli D Excoffier L (2000) Arlequin: a software for population genetics data analysis., Genetics and Biometry Lab, Dept. of Anthropology, University of Geneva, Geneva Seifert KA Samson RA de Waard JR Houbraken J Levesque CA Moncalvo J-M Louis-Seize G Hebert PDN (2007) Prospects for fungus identification using C01 DNA barcodes, with Penicillium as test case. Proc Natl Acad Sci USA 104:3901- 3906 81  Shalchian-Tabrizi K Minge MA Espelund M Orr R Ruden T Jakobsen KS Cavalier-Smith T (2008) Multigene phylogeny of choanozoa and the origin of animals. PLoS ONE 3:e2098 Smith D Lee R (2008) Nucleotide diversity in the mitochondrial and nuclear compartments of Chlamydomonas reinhardtii: investigating the origins of genome architecture. BMC Evol Biol 8:156. Smith JM (1991) The Population Genetics of Bacteria. Proc Biol Sci 245:37-41 Snoke MS Berendonk TU Barth D Lynch M (2006) Large global effective population sizes in Paramecium. Mol Biol Evol 23:2474-2479 Steenkamp ET Wright J Baldauf SL (2006) The protistan origins of animals and fungi. Mol Biol Evol 23:93-106 Su C Evans D Cole RH Kissinger JC Ajioka JW Sibley LD (2003) Recent expansion of Toxoplasma through enhanced oral transmission. Science 299:414-416 Swofford DL (1998) PAUP*: phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer Associates. Sunderland, Mass Takishita K Fujiwara Y Kawato M Kakizoe N Miyazaki M Maruyama T (2008) Molecular identification of the ichthyosporean protist "Pseudoperkinsus tapetis" from mytilid mussel Adipicola pacifica associated with submerged whale carcasses in Japan. Mar Biotechnol 10:13-18 Thompson JD Gibson TJ Plewniak F Jeanmougin F Higgins DG (1997) The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl Acids Res 24:4876-4882 Tibayrenc M Kjellberg F Ayala FJ (1990) A clonal theory of parasitic protozoa: the population structures of Entamoeba, Giardia, Leishmania, Naegleria, Plasmodium, Trichomonas, and Trypanosoma and their medical and taxonomical consequences. Proc Natl Acad Sci USA 87:2414-2418 Tsai IJ Bensasson D Burt A Koufopanou V (2008) Population genomics of the wild yeast Saccharomyces paradoxus: quantifying the life cycle. Proc Natl Acad Sci USA 105:4957-4962 82  Urdaneta L Lal A Barnabe C Oury B Golman I Ayala FJ Tibayrenc M (2001) Evidence for clonal propagation in natural isolates of Plasmodium falciparum from Venezuela. Proc Natl Acad Sci USA 98:6725-6729 van der Verde M During HJ van de Zande L Bijlsma R (2001) The reproductive biology of Polytrichum formosum: clonal structure and paternity revealed by microsatellites. Mol Ecol 10:2423-2434. Wakeley J (1997) Using the variance of pairwise differences to estimate the recombination rate. Genet Res 69:45-48. Watterson GA (1975) On the number of segregating sites in genetical models without recombination. Theor Pop Biol 7:256-276 Weeks AR Marec F Breeuwer JAJ (2001) A mite species that consists entirely of haploid females. Science 292:2479-2482 Weir BS (1996) Genetic data analysis II. Sinauer., Sunderland, MA. Whisler HC (1968) Developmental control of Amoebidium parasiticum. Dev Biol 17:562-570 Whisler HC Zebold SL Shemanchuk JA (1974) Alternate host for mosquito parasite Coelomomyces. Nature 251:715-716 Xu J (2004) The prevalence and evolution of sex in microorganisms. Genome 47:775- 780 Zeyl C Bell G (1997) The advantage of sex in evolving yeast populations. Nature 388:465-468 83  Chapter 4. Genealogical concordance species recognition criteria show cryptic species among morphologically similar sympatric isolates of Sphaeroforma (Ichthyosporea, Opisthokonta). 1   Introduction During a culture based survey for undescribed marine osmotrophs inhabiting marine invertebrate digestive tracts we isolated 148 strains of spherical unicells with 98.8 % or greater SSU-rDNA sequence similarity to spherical ichthyosporeans Sphaeroforma arctica and Pseudoperkinsus tapetis. Ichthyosporeans are osmotrophic protists classified in one of six known unicellular lineages related to the multicellular animals and fungi, which together make up the eukaryote supergroup Opisthokonta (Cavalier-Smith 1998; Adl et al. 2005). Our isolates, plus three previously described taxa, S. arctica, P. tapetis and Creolimax fragrantissima all belonged to a monophyletic clade of ichthyosporeans that are spherical and found in marine invertebrate hosts. Unlike most other ichthyosporeans, these grow well in culture (Figueras et al. 2000; Jostensen et al. 2002; Marshall et al. 2008; Takishita et al. 2008).  Species recognition in microorganisms The discovery of defining characters that separate a species from its close relatives is a critical element of taxonomy. However, finding reliable measures of discontinuous character variation that may signal this divergence can be challenging, especially in cryptic species with few morphological characters – a condition that applies to most microbes including our ichthyosporea with spherical cells. It was unclear whether our 148 isolates represented one or several new species, or were conspecific with S. arctica or P. tapetis. Our isolates were collected from a wide variety of hosts and from two geographically separated sites. We were aware that too broad a species definition  1  A version of this chapter will be submitted. Marshall WL Berbee ML. Genealogical concordance species recognition criteria show cryptic species among morphologically similar sympatric isolates of Sphaeroforma (Ichthyosporea, Opisthokonta). 84  would make impossible the recognition of correlated biologically important characters, such as cryptic morphological variation, host or substrate specificities, geographical ranges and temperature and nutrient requirements (e.g. de Vargas et al. 2001; Amato et al. 2007). Too narrow a definition would underestimate variation and fail to predict new character combinations that may appear in the next generations. Accurate species delimitation would allow testing for geographic isolation, niche differentiation or other possible causes for divergence among species (Reece et al. 2001).  Most working species definitions are practical derivations of the theoretical evolutionary species concept (ESC). The ESC defines a species as a lineage of descendent populations from a single ancestor, which maintains its identity from other such lineages and has its own evolutionary tendencies and historical fate (Wiley 1978). Approaches to recognition of species of microorganisms vary and some criteria for species recognition are more applicable to some groups than to others (Tibayrenc 2006). Amongst microbes, mating crosses (biological species concept), discontinuity of morphological traits (morphological species concept) and various molecular techniques have all been used to delimit species (e.g. Taylor et al. 2000; Schlegel and Meisterfeld 2003; Tibayrenc 2006). Biological species concepts are impractical for protozoans with uncharacterized lifecycles and morphological concepts do not resolve cryptic species. Molecular phylogenetic concepts, depending on the genes analyzed, can potentially recognize species before the accumulation of measurable changes in morphology (Taylor et al. 2000). Examples where both biological and molecular phylogenetic species recognition criteria had better resolving power than morphological methods are prevalent for both fungi and protozoa (e.g. Kasuga et al. 1999; Amato et al. 2007). In some cases (e.g. Amato et al. 2007) morphological characters for recognizing species have been discovered after close study of species delimited by other methods.  Different criteria for recognition often converge on the same answer but not always and for example, biological versus multilocus phylogenetic species delimitations gave discordant results when applied to ciliates of the genus Paramecium (Catania et al. 2009). Molecular phylogenetic approaches and morphological concepts share a drawback in that the cut off between intra- and interspecific variation is arbitrary. Hebert et al. (2004) proposed a ‗10 fold rule,‘ basically that divergence between species is at least 10 times the divergence 85  within species. Divergence thresholds do not, however, have widespread biological support (Frézal and Leblois 2008). Both morphological similarities and SSU-rDNA similarities pointed to a close relationship among our 148 isolates and the previously identified spherical ichthyosporeans. We soon noticed discontinuous variation among the ribosomal internal transcribed spacer (ITS) DNA sequences of our isolates. However, a single measure of sequence divergence from one locus in the absence of any other data was not likely to provide convincing delimitation of species. Application of a morphological species concept would have united all isolates in a single species, given their similarity in form and their overlapping morphological characters. Given the discontinuous variation in the ITS regions, it seemed likely that the morphological species concept would have been too broad for application in studies of host specificity or geographical distribution. Ichthyosporeans reproduce prolifically via asexual reproduction (Mendoza et al. 2002) and mating between cultures has never been observed, therefore the application of a biological species recognition system was not feasible.  An alternate solution to species recognition, the genealogical concordance phylogenetic species recognition (GCPSR) (proposed by Baum and Shaw 1995, and reviewed by Taylor et al. 2000) seemed applicable to our situation. The GCPSR uses multi-locus phylogenies and takes advantage of the genetic recombination that occurs within but not among species. Conflict among phylogenies from different loci is considered to be the result of recombination within a single species. Alternatively, genealogical concordance among phylogenies indicates genetic isolation among different species. Assumptions underlying application of the GCPSR are that hypothetical species are recombining, are not completely clonal and are not hybridizing (Taylor et al. 2000). Fortunately, although asexual reproduction predominated, we were able to show that recombination occurred among a subset of the 148 isolates of spherical ichthyosporeans (Marshall and Berbee 2009, submitted). Also fortunate, an EST libary for S. arctica (Ruiz-Trillo et al. 2004) was available, providing the DNA sequences we needed to design primers to several protein coding genes.  Having both the necessary primers and living cultures we were able to easily sequence several loci from our isolates. Therefore, 86  we expected to see a transition from complete concordance between phylogenies from different loci between species to phylogenetic conflict generated by the recombination associated with normal sexual exchange within species according to a GCPSR approach. Our first goal with this paper was to apply the GCPSR using data from three loci to delimit species of Sphaeroforma, comparing our148 sympatric isolates with one another and with the only three related strains isolated previously, S. arctica  (Jostensen et al. 2002), P. tapetis from Spain (Figueras et al. 2000) and P. tapetis collected from Japan (Takishita et al. 2008). Our second goal was to describe for the first time the light microscopic and ultrastructural features of Sphaeroforma species from our newly isolated species. Once we had delimited species, our third goal was apply population level approaches to look for evidence of geographical or host range separation.  Methods Collection of isolates and genetic analysis Animal collection, isolation of strains and culture conditions. Our 148 ichthyosporeans were isolated from 12 of 42 collections of marine invertebrate hosts and one sample of water surrounding captive oysters (Fig. 4.1). We collected the host animals from overlapping (cohabiting), near and distant sites in the intertidal or high subtidal regions at sites on the east and west sides of central Vancouver Island, British Columbia, Canada between March 2003 and April 2007 (Fig. 4.2, Table D.1). From each collection consisting of 2-5 conspecific host animals (Fig. 4.2, Table D.1), we dissected and cultured the contents from digestive tracts pooled from each host collection according to Marshall et al. (2008). For most animals we used pieces from all gut regions; for bivalves we used a portion of the digestive gland and syringed stomach material. We ground arthropods, Idotea sp. and Pollicipes polymerus, and a limpet whole. We diluted the gut contents or animal homogenates in 400 μl of sterile seawater before plating. As a control, we sampled coelomic fluid from 50 % of non-bivalve animals and mantle cavity fluid from 57 % of the bivalves. From water surrounding captive oysters we cultured isolates from the pellet formed after centrifugation of 150 ml of water collected from a 0.5 L 87  container containing three oysters. Ichthyosporean colonies were white, raised and consisted of large round cells. We repeatedly streaked colonies from each original isolate until they were axenic. Additional cultures of S. arctica, P. tapetis (Spain) and P. tapetis (Japan) were kindly provided by J. De Mestral and A. Rogers, A. Figuerus, and K. Takishita. P. tapetis (Spain) was not viable but we were able to confirm its unique insertion in the SSU-rDNA.  DNA extraction, PCR amplification, and sequencing. We extracted DNA using a DNeasy® Plant Mini Kit (QIAGEN Inc., Mississauga, Ontario, Canada). We amplified and sequenced the SSU-rDNA, internal transcribed spacers ITS1, 5.8S, and ITS2 (ITS) and a region of the elongation factor α-like gene (EFL) according to Marshall et al. (2008) except we used internal primers ELF3 and ELR3 (Appendix C, Table C.2) to sequence the EFL.  Initially we sequenced a region of the heat shock protein 70 gene (HSP70) using primers H7150F and H7965R (Appendix C, Table C.2) with an annealing temperature of 48 °C, and cloned it into the vector pCR2.1 using the TOPO TA cloning kit (Invitrogen). For subsequent isolates we amplified and sequenced the HSP70 using H7F2 and H7R2 (Appendix C, Table C.2) with an annealing temperature of 52 °C. Sequences were determined by Macrogen at Seoul, Korea or the Nucleic Acid Protein Service Unit at the University of British Columbia, Canada. Genetic analysis. To evaluate the phylogenetic relationship between our new isolates and other members of the spherical ichthyosporea we compared sequences from the SSU-rDNA, ITS and EFL using maximum parsimony, maximum likelihood and Bayesian methods. We aligned sequences in Clustal X v1.8 (Thompson et al. 1997), edited them in MacClade 4.08 (Maddison and Maddison 2000) and excluded ambiguously aligned regions. Our SSU-rDNA matix was 1738 bp in length and was comprised of 10 phylotypes. In addition to three phylotypes representing our isolates, we included S. arctica, P. tapetis, C. fragrantissima and several uncultured clone sequences from environmental samples. Both ITS and EFL matrices included each phylotype representing our isolates, S. arctica, at least one previously isolated P.tapetis, and C. fragrantissima as an outgroup. Our ITS alignment was 487 bp and comprised of 10 phylotypes. C. fragrantissima was difficult to align, therefore our ITS alignment was mostly made up of the 5.8 S with small unambiguously aligned sections of the ITS-1 and 88  ITS-2 spacer regions. Our ITS sequences had many polymorphic sites. Whenever a site had a secondary peak of at least 25 % of the main peak in one or both sequencing directions we coded it as polymorphic. However, because the forward and reverse sequences did not always show the same strength for the second peak we did not use polymorphic sites to define phylotypes. Our EFL matrix was 932 bp and included 21 phylotypes. We were unable to sequence the EFL from P.tapetis Spain. For the ITS, we compared the uncorrected p distances between the nonambiguous sites from 151 ITS sequences including S. arctica and both P. tapetis isolates using PAUP v.4.0b10 (Swofford 1998). Preliminary phylogenetic species assignments. Ichthyosporeans reproduce prolifically via asexual reproduction and our 148 isolates were likely to contain a high proportion of genetically identical asexual clones. Rather than sequencing several loci from all of our isolates we used sequences from the ITS and SSU-rDNA to assign our isolates into three subgroups. We first sequenced the shorter and more variable ITS region from all isolates. We sequenced the larger and less variable SSU-rDNA for two to six isolates from each of the four recognized ITS phylotypes, based on non-ambiguous sites. We then assigned our isolates to one of three hypothetical species defined by having 100 % SSU-rDNA similarity and at least 99.6 % ITS similarity. Our only other alternate possibility for defining phylogenetic species based on our ITS and  SSU-rDNA sequence data was to assign four categories based on 100 % ITS similarity. We refer to these phylogentic species as S. tapetis British Columbia (B.C.), ―S. gastric‖ and ―S. nootkatensis‖. In order to test whether our three hypothetical phylogenetic species were supported according to the GCPSR criteria we chose between six and 17 isolates from each for a sequencing of the EFL and HSP70 loci.  When possible we included one representative of each unique haplotype (based on the combined EFL and HSP70 sequences) and representatives from both sides of Vancouver Island (Table 4.1). We compared the results of a combined ITS, EFL and HSP70 analysis and each locus individually and looked for transitions from concordance to conflict. To test whether or not our isolates would be supported as phylogenetically distinct taxa from previously isolated taxa we also included S. arctica and P. tapetis (Japan) in our analyses. We were 89  unable to amplify the EFL and HSP70 sequences from P. tapetis (Spain).  To construct parsimony trees we used heuristic searches with 10 random sequence additions and TBR branch swapping limited to 10 6 rearrangements in PAUP. Parsimony bootstrapping in PAUP involved 100 replicates using heuristic search parameters as above. Maximum likelihood analyses were calculated by RAxML (Stamatakis 2006) or PAUP using model parameters estimated by Modeltest v. 3.7 (Posada and Crandall 1998). RAxML searches involved 200 heuristic search replicates and bootstrapping using 500 replications. For maximum likelihood analyses using PAUP we used a heuristic search strategy, limited to 10 random sequence additions, tree-bisection branch swapping limited to 2500 rearrangements and 100 bootstrap replications. Population structure.  To test for possible correlation of genetic differentiation with collection locations (east versus west side of Vancouver Island) or hosts, we calculated dθ using Multilocus v.1.3 (Agapow and Burt 2001) and EFL and HSP70 alleles for ―S. nootkatensis‖, ―S. gastric‖ and S. tapetis. dθ uses allele frequencies and is a variation of Wier‘s (1996) formulation of Wright‘s FST (Agapow and Burt 2001). Subpopulations with dθ close to 0 have similar allele frequencies and those close to 1 have dissimilar frequencies, which could be due to a reduction in gene flow. We assessed significance using 10, 000 randomizations. For S. tapetis, which had the largest data set, we used sequence data to calculate Snn (Hudson 2000) using DnaSP v4.50.3  (Rozas et al. 2003) and assessed significance using 10, 000 permutations. Isolates with the haplotypes that were cultured from the same host animal collection may have been asexual clone- mates. Therefore only unique haplotypes, based on the combined EFL and HSP70 alleles, from each host collection were used to represent  individuals (Table 1). Because S. tapetis was isolated in such high numbers in 2007 (Fig. 4.1) we tested for evidence of population growth using Fu‘s Fs statistic (Fu 1997) and R^2 (Ramos-Onsins and Rozas 2002), which are based on changes in shape in the expected genealogy, using  DnaSP and compared our result to 1000 simulations.  90  Light and electron microscopy We observed asexual reproduction in the S. tapetis and ―S. nootkatensis‖ because they grew most reliably in culture. ―S. gastrica‖ grew relatively slowly. We chose one representative Sphaeroforma-like  (isolate RU1) and followed single cells through at least one round of asexual reproduction in Brown‘s fungus slide chambers (Brown 1942) according to (Marshall et al. 2008). Individual cells were followed daily for up to 10 days and grown in both unsupplemented sterile seawater and in seawater broth supplemented with 1 % yeast extract, 1 % peptone and 10 % dextrose. S. tapetis isolates varied in colony morphology and size. We chose one isolate (CG2) that had small cells and non- slimy colonies and another (isolate LE7) that had larger cells and slimy colonies. In order to evaluate the consistency of four morphological characters we grew two cultures per isolate from several isolates of each new species on YPD agar plates. After one week we compared seven S. tapetis (B.C.), one P. tapetis (Japan), three ―S. gastrica‖, five ―S. nootkatensis” and one S. arctica cultures. We repeated these observations after two weeks.  For each culture, we screened plates for the presence or absence of slimy colonies and edges of colonies for the presence or absence of plasmodial cells. We also made slides of each and measured the sizes of mature cells with visible endospores and recorded the morphology of the empty parental wall cast-offs. To evaluate whether our organisms produced holdfasts or had any mechanisms for adhesion we also grew several ―S. tapetis” and ―S. nootkatensis‖ isolates in broth with sterile cotton fibres, plant roots (Zostera marina) and blond puppy fur.  We DAPI stained ―S. nootkatensis‖ (isolate RU1) and S. tapetis (isolates CG2 and LE7) according to (Marshall et al. 2008). For transmission electron microscopy we chose one ―S. nootkatensis‖ isolate (RU1) and two S. tapetis isolates (CG2 and LE7). We used two “S. tapetis” isolates because of their morphological differences using light microscopy. We high pressure freeze substituted S. tapetis  isolate LE7 and ―S. nootkatensis‖ isolate RU1 according to (Marshall et al. 2008). We also high pressure froze S. tapetis isolate CG2, but we used 0.1% uranyl acetate and 1% OsO4 in acetone as a substitution media and a Leica AFS chamber. We microwave infiltrated and embedded the cells in a 50/50 Spurrs/Epon resin (Spurr 1969).  We embedded all isolates flat between slides and cut 60 nm sections. We 91  stained the sections with 3 % uranyl acetate for 20 mins and lead citrate for 5 and photographed them using a Hitachi H7600 transmission electron microscope.  Results Phylogenetic species recognition and genealogical concordance Our SSU-rDNA sequences ranged from 98.8 % similarity between ―S. gastrica‖ and S. arctica to 99.9 % similarity between S. tapetis (B.C.) and previously identified S. tapetis (Spain) (formely Pseudoperkinsus tapetis) (Table 4.2). Three of our S. tapetis (B.C.) isolates, including isolate LE7, had a slightly different ITS phylotype than our other “S. tapetis‖ (B.C.) isolates that but all SSU-rDNA sequences were identical from the six isolates that were sequenced.  Figure 4.3 shows the relationship of our Sphaeroforma isolates to previously isolated S. arctica and S. tapetis (Spain and Japan). Ours and previously isolated S. arctica and S. tapetis sequences were more closely related to each other than they were to C. fragrantissima. Our SSU-rDNA tree shows a sister group relationship between ―S. gastrica‖ and S. tapetis (Fig. 4.3). In contrast our ITS and EFL trees (Fig. 4.4) show a sister group relationship between ―S. gastrica‖ and ―S, nootkatensis”. Figure 4.5 shows a histogram of pairwise distances between ITS sequences. At least two clusters of higher frequencies were present and the distances between ITS sequences from isolates with the same SSU-rDNA sequence were low (0-0.013), whereas the distances between sequences from different SSU-rDNA phylotypes ranged from 0 to 117. S. tapetis (B.C.) and S. tapetis (Spain and Japan) ranged in difference from 0 to 0.025. The difference between S. arctica and ―S. nootkatensis‖ was 0.028 to 0.034; the differences between all other SSU-rDNA phylotypes was 0.064 or greater.  Figure 4.6 shows the phylogenetic relationship of concatenated ITS, EFL and HSP70 sequences from unique combined haplotypes from our isolates, S. arctica and P. tapetis  (Japan). Our 29 isolates belonged to three clusters, each separated by long branches with 99 % or greater bootstrap support according to maximum parsimony and RAxML analyses (nodes 1-3).  Nodes 1-3 did not show any evidence of conflict when 92  each locus was compared individually and each locus had 97 % or greater bootstrap support for all three nodes (Figs. 4.6, D.1-4.3). Thus there was strong evidence of concordance between all three loci in distinguishing S. tapetis (B.C.), “S. nootkatensis‖ and ―S. gastrica‖. Nodes i –vi represented potential subdivisions or further taxonomic splits within our hypothetical taxa from British Columbia. Within our ―S. nootkatensis‖ - isolates our combined analysis supported an east/west division (nodes i and ii) (Fig. 4.6). However when each locus was analyzed separately, this division was only supported by the EFL (Fig. 4.6, D.2). The HSP70 locus contradicted this arrangement (Fig. S3) and a closer look at the alleles (Table 4.1) showed that an allele typically associated with the west side of Vancouver Island was found in one of the isolates collected on the east side. Likewise nodes iii-iv and v-vi were not supported as sister taxa when all loci were examined separately and therefore did not support further taxonomic divisions in our ―S. gastrica‖ or S. tapetis isolates.  The relationships between ―S. nootkatensis‖ and S. arctica and between S. tapetis B.C., Spain and Japan were close (Figs. 4.3-4.4) but the limits between species were not well defined. When S. tapetis (Spain) was included in our ITS analysis it grouped within our S. tapetis isolates (Figs. 4.3, D.4).  S. tapetis (Japan) was excluded from our  S. tapetis  (B.C.) isolates according to our combined analysis and supported by maximum parsimony methods for the ITS and HSP70 (node B) (Figs. 4.6, D.1, D.3-4.4). However, our EFL analysis showed conflict and placed S. tapetis (Japan) within our S. tapetis isolates. The separation of S. arctica from our ―S. nootkatensis‖ isolates had 97 % or greater support (node A) in our combined analysis. All loci were concordant in supporting the division of S. arctica from ―S. nootkatensis” but the support from the maximum likelihood analysis of the HSP70 locus was not strong at 65 % (Figs. 4.6, D.3). Additional isolates of S. arctica and S. tapetis (Japan) and wider geographical sampling is needed to resolve these nodes further.  Light microscopy and asexual reproduction “S. nootkatensis‖ isolate RU1, matured and released endospores after 2-3 days in Brown‘s slide chambers. These endospores grew and matured a second time by day 8. 93  For the first round of asexual reproduction mature cells measured between 40 and 45 µm and the endospores were 5-6 µm in diameter (Fig. 4.7A). The number of endospores released from the first round of division was approximately 120 but was this variable and the spores were difficult to count due to clumping. Spores were released when the parental cell wall split suddenly into two valves. Some spores were projected a short distance but most remained clumped. Often the parental cell wall valves rolled back and inverted. Empty cell walls, were distinctively split, persisted in cultures and provided a strong indicator for this type of release mechanism (Fig. 4.7 A, H). Rounds of asexual reproduction in sterile unsupplemented seawater were faster and the first release occurred at day two and most endospores matured and underwent a second round of release at day 3-4. Cells grown in seawater were smaller at maturation than those in media and released 16-32 spores after the first maturation and 8-16 at the second round of spore release. Endospores in seawater also did not divide a third time (Fig. 4.7H).  DAPI staining of cultures left in seawater suspensions for long periods of time showed that endospores arrested at a uni- and bi- nucleate states. These endospores were viable for at least six months.  Of the four additional ―S. nootkatensis‖ isolates (MC5, CH1, A9, PDD9), plus S. arctica, that we screened for colony morphology, plasmodia and empty parental cell wall morphology, all cultures, including S. arctica, (Fig. 4.7B) showed strong evidence of splitting during spore release with persistent bi-valved wall cast-offs. Slimy colonies and plasmodial cells were never observed. Between cultures, the size of mature ―S. nootkatensis‖ cells with visible endospores was variable (23-29 µm) and smaller than measurements from Brown‘s slide cultures. According to daily observations of S. tapetis isolate CG2 in Brown‘s fungus slide chambers, cells reached maturity after three days and were 14-17 µm in diameter. Cells from isolate LE7 were larger at maturity and measured 30-35 µm in diameter. New endospores measured between 5-7 µm in diameter for both isolates. Endospores in isolate CG2 had a short-lived amoeboid stage. They escaped the parental cell wall by squeezing through narrow pores in the cell wall (Fig. 4.7 D-F).  Empty parental walls were intact, circular and often had one or two small remaining refractive bodies. The numbers of 94  spores released was usually 32. Isolate LE7 reproduced by splitting (Fig. 4.7G) and released 64 or more endospores. Colony morphology amongst S. tapetis isolates was variable. Isolate CG2 was similar to ―S. nookatensis‖ and had white raised colonies of refractive spherical cells. Isolate LE7 and most other S. tapetis isolates had variably slimy or non slimy colonies. Slimy colonies usually had large plasmodial cells along the periphery (Fig. 4.8B) or extracellular material (Fig. 4.8D). Plasmodial cells were present in all additional S. tapetis cultures examined as well as in S. tapetis (Japan). In some isolates, such as CG2, plasmodia were rare but were still present (Fig. 4.8C), whereas in other isolates, for example LE7, they were common (Fig. 4.8B). Movement and budding of plasmodial cells was observed on agar plates several times and for more than one S. tapetis isolate. Figure 4.8C shows extracellular material near one end of the plasmodial cell, presumably left in its wake after moving.  Cell sizes at maturity on agar plates were variable between isolates as they were in Brown‘s fungus slides. Both split and round, intact empty parental wall cast-offs were present depending on the isolate. Nuclei were not visible without DAPI staining and numbered 1, 2, 4, 8, 16, 32 or higher in both ―S. nootkatensis‖ and S. tapetis. Cells with three or six were seen rarely and nuclear division was for the most part synchronous and seemingly instantaneous. Individual chromosomes were never visible in DAPI stains. Cells from both taxa adhered to all types of fibres tested but did not produce any visible holdfast (Fig. 4.7I). For cultures of ―S. gastrica‖ (Fig. 4.7C), parental cast-offs were of the circular type suggesting release via pores rather than splitting in the three isolates examined. Plasmodial cells and slimy colonies were evident in ―S. gastrica‖ isolate CGE, but isolates BM11 and ON36 had non slimy colonies that were similar to ―S. nootkatensis‖ and S. tapetis isolate CG2.  Sizes of mature spherical cells of ―S. gastrica‖ on agar plates ranged from 29 to 39 µm.  Ultrastructure Our ―S. nootkatensis‖ and S. tapetis isolates showed no strong indication of contrasting features at the ultrastructural level. We did not observe all structures and 95  stages in both; however, this could be due to variables such as culture condition at time of fixation and probably does not represent true differences. Both taxa were multinucleate, coenocytic, and had cell walls during vegetative growth (Figs. 4.9, 4.10). Cytoplasmic contents included large amounts of glycogen, mitochondria with flat cristae, and vesicles and inclusions with varied content (Figs. 4.9, 4.10). Cell walls were composed of two layers and were similar for both taxa. Some cells had an external electron lucent cell coat or calyx that appeared to be composed from tightly pressed fibres oriented perpendicular to the cell surface (e.g. Figs. 4.9C, 4.10A, 4.11C).  The thickness of the calyx varied from not visible to 500 nm; this variation may have been due to cell age, cell condition or treatment during fixation and processing. It was present on unreleased endospores (Fig. 4.12G), young cells (Fig. 4.10A) and parental cells with differentiated endospores (Fig. 4.10C). The outer layer of the cell wall appeared homogenous but was composed of very densely packed electron opaque fibres. The inner layer, when present, was made of a homogenous electron lucent material (e.g. 4.11C). When the inner layer was present, the plasma membrane was usually wavy. Often there were membrane bound tubules (MTBs) that extended from the cell cytoplasm and penetrated into the inner wall layer (e.g. Figs. 4.11C, 4.12J); the membrane bounding the tubules appeared contiguous with the cell membrane and glycogen rosettes very sometime visible in the cytoplasm within the MTBs (Fig. 4.11). Of the 11 multinucleate cells from S. tapetis that we sectioned, two cells had thickened cell walls at polar ends of the cell (Fig. 4.9 B, D). The opposing arrangement of the thickenings suggested that these sections may have bisected the meridian of wall destined for the formation of the split during endospore release. Electron lucent vesicles embedded in the thickenings (12 H) could perhaps function to weaken that area of the wall. Inconveniently for this hypothesis, during light microscopic observation, isolate CG2 (Fig. 4.9D) usually released amoeboid spores via pores rather than by bivalve splitting. However TEM sections (Figs. 4.9D, 4.12D) showed that some endospores had cell walls that would prevent their escape as amoebae. This may mean that not all spores are released as amoebae and that splitting of cell walls may also occur in isolate CG2. Serial sections or sections from cells at later stages of development are needed to confirm this hypothesis. 96  Nuclear structure was similar in ―S. nootkatensis‖ and S. tapetis. Nuclei had peripheral nucleoli (Figs. 4.9, 4.13 A, E, 4.14A) and sometimes two nucleoli within a single nucleus were visible (Fig. 9B). Golgi bodies, best preserved in ―S. nootkatensis,‖ had up to seven stacked cisternae with electron dense contents (Figs. 4.10C, 4.13 A-E). They were always associated with the nucleus and often there were two Golgi per nucleus (e.g. Fig. 4.13B). When there were two Golgi bodies, they were centered at opposite ends of the nucleus. In S. tapetis, the Golgi bodies were also associated with the nucleus (Fig. 4.14). In S. tapetis isolate CG2 the stacked cisternae were evident but were lightly stained (Fig. 4.14B-D). Isolate Le7 had up to three Golgi bodies associated with some nuclei (not shown) but the cisternae were not organized in neat stacks (Fig. 4.14H). Spindle pole bodies (SPB) with extranuclear microtubules were positioned between the Golgi body and nucleus for both ―S. nootkatensis‖ and S. tapetis (Figs. 4.13 B,D, 4.14 B-D) . In some nuclei there were two SPBs at opposite ends of the cell, presumably in preparation for mitotic division (Fig. 4.13B). The microtubule organizing center was crescent shaped in cross section, round in tangential section and composed of electron opaque material. Some cells from S. tapetis isolate LE7 appeared to be undergoing a different form of development. These cells had irregularly shaped developing spores, patches of discontinuous plasma membrane and unusually thick cell walls, usually greater than 1 µm. Cell wall material appeared in variously thick patches between developing spores and along the outer cells wall (Fig. 4.14G). Membrane bound extensions of the cell cytoplasm was often associated with the patches of wall material (Figs. 4.11 A-B, 4.14I). The correspondence of the MTBs and the wall material suggests that the MTBs may have a function in cell wall deposition as these cells did not appear to have the same method of parentally derived wall deposition as the other cells (see below). Intranuclear microtubules were observed from several nuclei from two of the four sectioned cells exhibiting this morphology(Fig. 4.14H). Vesicles with electron opaque contents were abundant in vegetative cells and developing endospores (Figs. 4.9, 4.11-4.13). These vesicles often appeared to encircle or be in patches near the nucleus (e.g. Figs 4.9A, 4.14A), and may have been associated with the Golgi body (Fig. 4.13E). The contents were attached to the circumference of the vesicle and probably bound to the membrane and the lumen of these vesicles usually 97  appeared transparent. Tangential sections through the opaque material showed that it was either arranged in striations or was fibrous in composition (Fig. 4.11A). Breaks or areas where no material was attached to the vesicle circumference were evident in cross and tangential sections (e.g. Fig. 4.11A). Occasionally this or similar material was arranged in a linear form (Fig. 4.12A). We were unable to determine the source or destinations of the material in these vesicles. Maturation of walled endospores was similar in both taxa. Several of our sectioned cells were no longer coenocytic. Instead the cellular contents consisted of membrane bound or sometimes walled endospores (Figs. 4.9 C-E, 4.10 B-C, 4.12 A-H). We did not section any cells that were in clear transition between these states, although some cells had endopores that had more than one nucleus and may have been undergoing cleavage (Fig. 4.9D). The electron opaque material composing the outer cell wall layer of each endospore originated from what remained of the parental or investing cell‘s cytoplasm and adhered to the outside of the developing endospores (Fig. 4.12 C-F). Pores were visible in the endospore wall, even during the earlier stages of wall deposition (Fig. 4.12 D, F-G). Figure 4.12G shows mature unreleased endospores with an intact calyx. These mature spores, and others with advanced cell wall development had few or none of the vesicles bound with fibrous contents.  The contents of the vesicles may have somehow contributed to the cell wall.  Circular and linear segments of plasma membrane were present within many developing spores (Fig. 4.12 B-C). These were occasionally contiguous with the outer cell membrane and may have contributed to cell growth.  Host specificity and gene flow between east and west Vancouver Island  All three species from British Columbia were found on both East and West Vancouver Island (Figs. 4.1, 4.15, Table D.1). S. tapetis was found 126 times from nine different collections of eight different host species, ―S. gastrica‖ was found eight times in three host species and ―S. nootkatensis‖ was found 14 times from five host species (Figs. 4.1, 4.15, Table D.1). Both S. tapetis and ―S. gastrica‖ were found more often in bivalve collections than in other hosts, even after correcting for the higher frequency of collection 98  of bivalve hosts (Fig. 4.15). This contrasted with the broader host distribution of ―S. nootkatensis‖ (Fig. 4.15). No haplotypes, defined after concatenating the EFL and HSP70, were found on both east and west Vancouver Island for any of our three species. Each species did, however, have at least one allele that was present in isolates collected on each side (Table 4.1). Genetic differentiation in S. tapetis was significant as measured by dθ (Table 4.3) between the east and west sides of Vancouver Island even though one EFL and two HSP70 alleles were found on both sides of the Island. Using combined sequence data, geographical subdivision as measured by Hudson‘s Kst (Hudson et al. 1992) was not significant (not shown) but the SNN statistic of (Hudson 2000), which is more sensitive for smaller sample sizes, showed significant population subdivision for the east and west side (Table 4.3). When analysing each locus alone both Kst and SNN methods showed east/west division for the EFL, and neither method showed subdivision for the HSP70 (Table 4.3). Populations of ―S. nootkatensis‖ and ―S. gastrica‖ showed no genetic differentiation between either side of the Vancouver Island as measured by dθ (not shown). Populations of S. tapetis at two sites with high prevalence 1 km apart showed no evidence of population differentiation (Table 4.3). There was also no evidence of differentiation between the two most populous host animals (Table 4.3). Fu‘s Fs and R^2 showed no support for population expansion despite the unusually large number of isolates collected in the spring of 2007 (Fig. 4.1).  Diagnosis.  We saw no argument for maintaining Sphaeroforma and Pseudoperkinsus as separate genera. ‗Pseudoperkinsus tapetis‘ (Spain)‘ (=our ‗S. tapetis’) had not been formally described. Our studies revealed overlapping morphological variation between S. tapetis isolates and isolates of the species that were phylogenetically closer to S. arctica. We therefore provide a formal description for ‗Pseudoperkinsus tapetis (Spain)‘ as ―Sphaeroforma tapetis‖ S. tapetis (B.C.) was phylogenetically indistinguishable from previously identified S. tapetis (Spain).  Former P. tapetis (Japan) was genetically divergent but we could not rule out that the single isolate was nested within S. tapetis so 99  we conservatively included it in that species. We describe ―S. gastrica‖ and ―S. nootkatensis‖ as new species.  Sphearoforma (Jostensen et al. 2002). Typically smooth, coenocytic, spherical, walled cells. Nuclear proliferation concurrent with cellular growth, division of nuclei synchronous and ‗instantaneous‘. Asexual reproduction mainly via complete transformation of parent cell into numerous endospores. No flagellated stages but a plasmodial stage and short-lived amoeboid stages present in some species. Cells capable of adhesion but no holdfasts produced. Ultrastructure: mitochondria with plate-like cristae, Golgi bodies with stacked cisternae, nuclei with multiple peripheral nucleoli and a spindle pole body between the nucleus and Golgi body. Cell wall composed of an outer densely fibrous electron opaque layer, often perforated with pores and surrounded by an electron lucent calyx. Inner layer, when present, is comparatively lucent and often impregnated with MTBs. Most closely related to genus Creolimax based on phylogenetic inference. Distinguished from Creolimax based on having non-swarming colonies, no long lived amoeboid stages, lacking a central vacuole and having one or more peripheral nucleoli. Habitat: Marine, found in marine invertebrate digestive tracts, but may not be restricted to that habitat. Includes four species: S. arctica (Jostensen et al. 2002), S. nootkatensis, S. tapetis and S. gastrica.  S. tapetis (Figuerus et al. 2000) Usually spherical but irregularly shaped plasmodial forms may be present. Endospores between 5-7 µm released either actively by amoeboid ‗squeezing‘ through pores, or passively after the parental cell splits into two valves. Budding of plasmodia also occurs. Endospores squeezing through pores become non- motile almost instantaneously. Extracellular material associated with plasmodial forms contributes to the slimy appearance of the colonies. Distinguished from S. arctica and S. nootkatensis based on the variable presence of plasmodia, endospore release through pores and phylogenetic inference. Range: Northern hemisphere in marine environments. Spain, Japan and Western Canada. 100  Hosts: Primarily bivalves, but also found on single occasions in Leptosynapta clarki (sea cucumber), Katharina tunicata (chiton), Phascolosoma agassizii (peanut worm). Type: isolates CG2 accessioned at the ATCC under accession numbers PRA-281 check. Type locality: Isolated from pacific oyster Crassostrea gigas collected from Dixon Island, Barkley Sound, British Columbia, Canada. November, 2004. S. gastrica nov. sp. Distinguished from others based on phylogenetic inference. Endospores released through pores but may not be restricted to this form of reproduction. Colony morphology variably slimy and plasmodia sometimes present. Range: Western Canada, marine. Hosts: Various bivalve molluscs Type: Isolate BM11 accessioned at the ATCC under accession number PRA-278. Type locality: Isolated from bivalve mussel Mytilus sp. collected from Brickyard beach, Gabriola Island, British Columbia, Canada. April, 2007.  S. nootkatensis nov. sp. Distinguished from S. tapetis and S. gastrica based on a lack of plasmodial stages, colonies without slimy exudates and spore release exclusively via splitting of the parental cell into two equal valves, and based on phylogenetic inference. Range: Western Canada. Hosts:  Broad range of marine invertebrate hosts. Type: Isolate RU1, available at ATCC under accession number look up. Type locality: isolated from sea urchin, Strongylocentrotus franciscanus, collected from Scott‘s Bay, Bamfield, British Columbia, Canada.  Discussion Applying GCPSR supported the distribution of our isolates into three distinct species. Our sympatric isolates belonged to three different SSU-rDNA lineages. Combined data from three variable loci gave each of three distinct lineages 97% or more bootstrap support. Further subdivisions of our taxa were not supported by all loci, signifying a lack of support for further species division. Two of our three species S. gastrica and S. 101  nootkatensis were new to science and the third, which had been informally named Pseudoperkinsus tapetis (Figueras et al. 2000) is now described as Sphaeroforma tapetis. Because our isolates shared habitats within overlapping ranges, the lack of gene flow and relatively long branch lengths suggested reproductive isolation across species. Two of three of our isolated species were more closely related to isolates collected from different oceans than they were to other taxa collected from British Columbia. Sphaeroforma tapetis is widely distributed at least across the Northern hemisphere rather than narrowly endemic. This pattern shows that our three taxa did not undergo recent, local speciation but instead that speciation probably occurred before range expansion. Because our isolates were collected within overlapping geographic range the interpretation of our results was straightforward according to the GCPSR criteria. However, distinguishing our species from S. arctica and former collections of S. tapetis proved challenging because the latter were represented by one or two isolates and genetic differences could be due to geographical separation and reduced gene flow rather than to speciation. Encouragingly, ITS variation between S. tapetis isolated in Spain and BC was small, suggesting that ITS similarity may be a good marker for conspecificity, even across wide geographical distances. ITS divergence may indicate that the genetically distant isolate from Japan should be a new species separate from ‗S. tapetis‘ (B.C.). This could be testedby sampling additional isolates from populations in Japan and the geographically intermediate Aleutian Islands. Of course, the ITS similarity of S. tapetis from BC and Spain may reflect other factors such as recent host translocations. Many marine invertebrates, especially bivalves, have been introduced to British Columbia for commercial farming or as larvae in ballast water (e.g. Bourne 1982). Although the east and west Vancouver Island populations of S. tapetis showed some genetic differentiation, some alleles were common to both sides of the island. The presence of shared alleles on both coasts could be due to incomplete lineage sorting or to slow but significant gene flow, either by population connectivity, long-range dispersal, or by anthropogenic host translocations. Current velocities and directions could account for population connectivity and dispersal (Thomson 1982). Anthropogenic translocations could be linked to the rapid spread of some introduced host species such as clams (Dudas and Dower 2006). Given the evidence for geographic separation between east and west 102  Vancouver Island populations of S. tapetis, we assume that isolates collected as far away as BC and Japan are reproductively isolated. Although our isolates all came from marine invertebrate digestive tracts and showed no evidence for species specific host relationships, the three species were not necessarily occupying the same niche. We isolated S. tapetis and S. gastrica most frequently from bivalves and previously isolated S. tapetis came from bivalves (Figueras et al. 2000; Takishita et al. 2008). Our collection may have been biased because only bivalves were sampled in the spring of 2007 when for some reason, S. tapetis was in its highest abundance. Among other spherical ichthyosporeans, S. nootkatensis seemed to have no apparent host preference and C. fragrantissima was found most often in sipunculid peanut worms (Marshall et al. 2008). Even without host preference, it is possible that each species occupies a different region of the gut, or has different physiological or nutritional requirement. S. gastrica was the most difficult to culture and may have different physiological constraints. However all species with multiple isolates, except for S. gastrica, were found in several host phyla and speciation due to host preference is unlikely. Although Sphaeroforma species have all been isolated from marine invertebrate animals they did not co-evolve with specific hosts. Ichthyosporean population density was highly variable. The number of S. tapetis isolates collected from one east side site was greater than the total number collected from all other sites, yet there was no genetic evidence of long-term population expansion. If S. tapetis was undergoing some sort of population expansion or bloom, it was not dominated by a single clone. Phytoplankton blooms can be represented by a high diversity of lineages within a single species (Rynearson and Armbrust 2005).  Comparative morphology Species could not be distinguished by their morphology. The number of species as measured by the GCPSR was higher than what would be have been delimited using morphological criteria, and Sphaeroforma, while being morphologically nondescript, showed evidence for cryptic diversity. Characteristics of S. tapetis varied from isolate to isolate and generally overlapped with characteristics of S. nootkatensis. S. tapetis and S. 103  gastrica had the more complex life cycles which included a plasmodial stage that was not seen in either S. nootkatensis or S. arctica. Even using ultrastructure, S. tapetis and S. nootkatensis could not be distinguished reliably from one another. S. tapetis and S. nootkatensis exhibited many subcellular features shared throughout opisthokonta including glycogen storage and mitochondria with plate-like cristae. Characteristics common to most ichthyosporea, included a cell wall (Mendoza et al. 2002) a multi-nucleate stage, spindle pole bodies (Marshall et al. 2008), and MTBs (Marshall et al. 2008) The cell wall material and association of the Golgi bodies with the spindle pole bodies was similar to C. fragrantissima (Marshall et al. 2008) but the pores within the cell wall, calyx, peripheral nucleoli, polar thickenings of the cell wall, lack of amoebae and splitting of the parental cell wall distinguish Sphaeroforma from Creolimax. Central vacuoles, characteristic of most ichthyosporeans including C. fragrantissima, were not observed. Capsule-like material that may be homologous to the calyx has been described on the endospores of ichthyosporeans Rhinosporidium (Pereira et al. 2005) and Amphibiocystidium (Pereira et al. 2005) and holdfast material is present on the external surface of the amoebidiales and eccrinales (Dang and Lichtwardt 1979, Cafaro 2000). The eccrinids and amoebidiales use their holdfasts to attach to the exoskeleton of their insect host (Cafaro 2005). Pores have been described in the cell wall of S. arctica (Jostensen et al. 2002) and near the holdfasts of eccrinids and amoebidiales (Dang and Lichtwardt 1979, Mayfield and Lichtwardt 1980, Jostensen et al. 2002). Mayfield and Lichtwardt (1980) described the secretion of holdfast material through pores in eccrinid Enterobryus attenuatus and it is possible that the pores found in Sphaeroforma are homologous, and function for secretion of the calyx material. The function of the calyx was undetermined but may, like the holdfasts in the eccrinales and amoebidiales, be used for adhesion. Multiple nucleoli within a nucleus is not uncommon amongst eukaryotes, but has not been reported for any ichthyosporea. A peripheral nucleolus was reported in the freshwater spherical ichthyosporean Anurofeca (Beebee 1991). Spindle pole bodies nested between Golgi bodies and nuclei were also described in C. fragrantissima, and a plaque appressed to the nucleus, which was probably a spindle pole body, was described 104  for Ichthyophonus hoferi (Franco-Sierra and Alvarez-Pellitero 1999). More distantly related and typically flagellated icthyosporeans have centrioles (Ashworth 1923, Pekkarinen et al. 2003). Our sections showed that a multinucleate coenocytic stage preceded endospore development. It was clear that the increase in plasma membrane on developing endospores came from the endospore itself and the outer layer of the endospore was derived from the parental cell. The unusual electron opaque fibrous vesicles prevalent in multinucleate coenocytic cells and developing endospores had an undetermined function. Similar fibrous vesicles were described in developing endospores by (Mayfield and Lichtwardt 1980). They speculated that the vesicles may extrude through the pores to form the holdfast. Our section through a linear vesicle supports the theory of exportation, however if the vesicles were to form the calyx, they material would have to undergo some form of chemical transformation to account for the change in electron absorption. The presence of these vesicles near developing walls within the unusual cells types observed in S. tapetis supports an alternate hypothesis that they function in cell wall production.  Conclusions The application of the genealogical concordance phylogenetic species recognition allowed us to distinguish three cryptic sympatric organisms and elucidate differences in asexual lifecycles. It also allowed us to recognize that individual species showed wide host preferences and provided preliminary indications of gene flow and species ranges. 105  Table 4.1. Hosts, collection locations, allele and haplotype designations for isolates from three Sphaeroforma species collected from British Columbia. Species Host animal Location (Date) Isolate # 1 EFL allele  HSP70 allele 2 Haplotype 1 S. nootkatensis Dendraster excentricus East A22 S1 V1*  1  D. excentricus East A9 S1 V2 2  D. excentricus East A36 S1 V2 2  Phascolosoma agassizii West Pdd9 S3 V1* 3  Strongylocentrotus franciscanus West RU1 S2 -- 4  Corella sp. West CH1 S3 V1* 3  Mytilus californianus West MC2 S3 V1* 3  M. californianus West MC7 S2 -- 4  M. californianus West MC3 S3 V1* 3  M. californianus West MC10 -- V1* -- S. gastrica Crassostrea gigas West CRG1 R1 PR1 5  C. gigas West CRG3 R2* PR2 6  C. gigas West CGE R3 PR3 7  Nuttalia obscurata East ON36 R4 PR4 8  Mytilus sp. 1 East MB6 R5 PR5 9  Mytilus sp. 1 East BM11 R2* PR6 10  106  Species Host animal Location (Date) Isolate # 1  EFL allele HSP70 allele 2  Haplotype 1  S. tapetis C. gigas West CG5 P1 C4* 18  Katharina tunicata West KS1 P1 C2 19  C. gigas West CG2 P2* C3 17  C. gigas West CRG9 P9 C7* 25  Leptosynapta clarki West LE7 P10 C10 27  C. gigas water West OI22B P10 C11 28  Mytilus sp. 2 East GM8 P2* C6 11  Mytilus sp. 1 East MB4 P2* C6 11  Mytilus sp. 1 East BM33 P3 C6 14  Mytilus sp. 1 East BM19 P3 C4* 15  N. obscurata East NO32 P3 C4* 15  Venerupis phillipinnarium East PV23 P3 C4* 15  V. phillipinnarium East VP19 P3 C5 21  Mytilus sp. 1 East BM6 P4 C6 13  V. phillipinnarium East PV25 P4 C6 13  N. obscurata East NO6 P5 C8 22  Mytilus sp. 1 East MMB5 P5 C8 22  N. obscurata East ON1M P6 C1 20  V. phillipinnarium East PV3 P6 C6 12  Mytilus sp. 1 East MB16 P7 C8 24 107   Species Host animal Location (Date) Isolate # 1  EFL allele HSP70 allele 2  Haplotype 1  S. tapetis cont. Mytilus sp. 1 East BM17 P8 C9 23  N. obscurata East ON12 P8 C9 23  Mytilus sp. 1 East MB29 P8 C4* 16  N. obscurata East NO16 P8 C7* 26 1 Isolates and haplotypes shown in boldface represent unique sequence types (or individuals) for a particular source animal collection thereby excluding potential clone- mates. 2 HSP70 alleles are defined using the entire sequenced region, including the intron(s). *Alleles shared by isolates collected from both east and west Vancouver Island. 108  Table 4.2. Region of overlap and percent similarity of our 3 SSU-rDNA phylotypes and previously identified S. arctica and S. tapetis (formerly P. tapetis). Species or taxonomic subgroup (isolate ID) Number of substitutions including indels Length of region of overlap (nt) Percent similarity S. tapetis (B.C. isolate LE7) **  vs S. tapetis (B.C. isolate PV27) 0 1718 100 S. tapetis  vs S. tapetis (Japan) 2 1732 99.9 S. tapetis vs S. tapetis (Spain) 2 1731 99.9 S. tapetis (Japan) vs S. tapetis (Spain) 2 1733 99.9 S. arctica vs ―S. nootkatensis‖ (MC10) 4 1724 99.8 S. arctica vs S. tapetis (Japan) 10 1729 99.4 S. arctica vs S. tapetis (Spain) 11 1730 99.4 ―S. gastrica‖ vs S. tapetis (Spain) 15 1717 99.1 ―S. gastrica‖ vs S. arctica 20 1710 98.8 ** Isolate LE7 was one of three S. tapetis isolates that varied in ITS sequence compared to all other S. tapetis isolates. 109  Table 4.3. Genetic differentiation between subpopulations of S. tapetis (B.C.) Genetic differentiation among S. tapetis subpopulations was evident between West and East coasts of Vancouver Island, but not between neighbouring beaches on the East side (Brickyard and Maples). No genetic differentiation was evident between cohabiting hosts clam Nuttalia obscurata (n=5) and Mytilus sp. (n=8) collected from Brickyard beach. Subpopulations Number of individuals dθ (p value) Combined Snn (P value) Combined Snn (P value) EFL locus Snn (P value) H70 locus West/East Vancouver Island 24 0.0833 (0.02) 0.875 (0.005) 0.875 (0.004) 0.677 (0.1) Brickyard/Maples beach 17 -0.0232 (0.6) 0.346 (0.8) 0.565 (0.7) 0.647 (0.4) Clam/Mussel (Brickyard beach) 13 -0.0232 (0.6) 0.637 (0.4) 0.428 (0.9) 0.436 (0.8)  110  N Pacific Ocean Vancouver Victoria Vancouver Island West: Barkley Sound Mainland Georgia Strait East: Parksville East: Gabriola Island  Figure 4.1. Animal collection sites from Vancouver Island, British Columbia, Canada. Western sites were distributed within Barkley Sound, near the Bamfield Marine Sciences Centre. East side sites were at Parksville Beach and three beaches along a 3 km SSW shoreline of Gabriola Island. Specific locations are listed in table D.1. 111  10 5 100 120 Winter 2003 East: Parksville Dendraster excentricus (3)  1 sp. Summer 2004 West Phascolosoma agassizii (1) 1 sp. Leptosynapta clarki Brisaster latifrons Thelepus sp. Idotea sp. Cucumaria miniata Strongylocentrotus     droebachiensis Limpet 48 plates 240 plates156 plates 20 30 Fall 2004 West  Mytilus californianus (8) 1 sp. Crassostrea gigas (8) 2 sp. Corella  sp. (1) 1 sp. Leptosynapta  clarki (1) 1 sp. Mytilus californianus -2 Protothacea staminea Thelepus sp. Cucumaria miniata Patiria miniata Phascolosoma agassizii Fall 2005 West Mytilus sp.-1 (2) 1 sp. Mytilus sp.-2 (45) 2 sp. Venerupis philipinnarium (26) 1 sp. Nuttalia obscurata (49) 2 sp.   Strongylocentrotus franciscanus  (1) 1 sp. Phascolosoma agassizii -1  (1) 1 sp. Katharina  tunicata -1 (1) 1 sp. Oyster water (1) 1 sp. Phascolosoma agassizii -2 Mytilus californianus -1 Styela montereyensis Katharina tunicata -2 Dendraster excentricus Protothacea staminea Katharina tunicata -3 Strongylocentrotus     franciscanus -2 Leptosynapta clarki Phascolosoma agassizii -3 Leptosynapta clarki -2 Crassostrea gigas-1 Crassostrea gigas-2 Mytilus californianus -2 Pollicipes polymerus 312 plates Spring 2007 East: Gabriola Island 120 plates Crassostrea gigas 3 1 12 2 2 56 117 # # # # # ## # # # # # # #* * * ** * * * * "S. tapetis" "S. nootkatensis" "S. gastrica"  Figure 4.2. Numbers of Sphaeroforma isolates cultured from host animals varied by year and by season and were not directly related to sampling effort. Bars show number of colonies of S. tapetis, S. gastrica, and S. nootkatensis. The number of culture plates reflects sampling effort.  Hosts where S. tapetis was found are marked by # ; and bivalve molluscs by *. Invertebrate host collections that produced one or more isolates are listed above the x-axis with the total number of colonies from each host in brackets followed by the number of species. Invertebrate collections that were sampled but did not provide colonies are listed below the x-axis. 112  Creolimax fragrantissima Sphaeroforma arctica Sphaeroforma nootkatesis nov. sp. Uncultured clone D3P06H09 Uncultured clone D4P08E02 Sphaeroforma tapetis (British Columbia) Sphaeroforma tapetis  (Japan) (formerly Pseudoperkinsus) Sphaeroforma gastrica nov. sp.Uncultured clone ikaite 5 changes Sphaeroforma tapetis (Spain) (formerly Pseudoperkinsus) SSU-rDNA 74/68 80/76 73/65 77/64 90/86  Figure 4.3. Maximum parsimony tree comparing the phylogenetic relationship of our isolates to other spherical marine ichthyosporeans based on the SSU-rDNA sequences. Numbers below nodes represent maximum parsimony and maximum likelihood bootstrap support (MP/ML). New Sphaeroforma isolates from British Columbia are in bold. 113  5 changes Sphaeroforma arctica S. nootkatensis (A22) S.  nootkatensis (PDD9) S. gastrica (CRG1) S. gastrica (MB6) S. tapetis (LE7) S. tapetis  (CG2) S. tapetis  (Spain) S. tapetis  (Japan) Creolimax fragrantissima 94/- 65/76 98/97 78/83 92/72 ITS  Sphaeroforma arctica S. tapetis Japan Creolimax fragrantissima5 changes 100/82 63/72 98/100 100/100 S. gastrica S. tapetis British Columbia 99/92 S. nootkatensis EFL  Figure 4.4. Maximum parsimony trees showing the phylogenetic relationships of ITS and EFL sequences from new and previously isolated spherical ichthyosporeans. Numbers at nodes represent 100 maximum parsimony and maximum likelihood bootstrap replications (MP/ML). The ITS tree was based on 10 phylotypes including S. tapetis (Spain). The EFL tree was based on 21 phylotypes. New Sphaeroforma isolates from British Columbia are in bold. 114  Between identical SSU phylotypes Between  S.tapetis Spain, Japan and British Columbia Between S. arctica and S. nootkatensis 0 50 100 150 200 250 0 .0 0 1 0 .0 0 5 0 .0 0 9 0 .0 1 3 0 .0 1 7 0 .0 2 1 0 .0 2 5 0 .0 2 9 0 .0 3 3 0 .0 3 7 0 .0 4 1 0 .0 4 5 0 .0 4 9 0 .0 5 3 0 .0 5 7 0 .0 6 1 0 .0 6 5 0 .0 6 9 0 .0 7 3 0 .0 7 7 0 .0 8 1 0 .0 8 5 0 .0 8 9 0 .0 9 3 0 .0 9 7 0 .1 0 1 0 .1 0 5 0 .1 0 9 0 .1 1 3 0 .1 1 7 Uncorrected pairwise distance F re q u e n c y Between all other SSU phylotypes  Figure 4.5. Uncorrected pairwise genetic distances from the ITS region were greater between species pairs than within species. Horizontal bars above graph show the relative ranges of ITS distances between versus within species. 7601 zero values are not included. 148 of our isolates and the isolates of S. arctica and of S. tapetis (Spain and Japan) were included for a total of 11026 comparisons. 115  5 changes Sphaeroforma arctica S. nootkatesis A22 S. nootkatesis CH1 S. nootkatesis A9 S. nootkatesis A36 S. nootkatesis MC2 S. nootkatesis PDD9 S. gastrica CRG1 S. gastrica BM11 S. gastrica CRG3 S. gastrica ON36 S. gastrica CGE S. gastrica MB6 S. tapetis OI22B S. tapetis LE7 S. tapetis CRG9 S. tapetis NO16 S. tapetis CG2 S. tapetis VP19 S. tapetis PV23 S. tapetis KS1 S. tapetis BM33 S. tapetis GM8 S. tapetis BM6 S. tapetis PV3 S. tapetis ON1M S. tapetis MB29 S. tapetis ON12 S. tapetis MMB5 S. tapetis MB16 S. tapetis Japan 100/97 100/100 85/86 100/100 90/95 92 /95 100/100 100/99 91/- 98/98 81/98 97/99 1 2 3 A B i ii iii iv v vi I  E  H P L P L P L 100 100 100 100 100 99 I  E  H P L P L P L 100 97 100 100 100 100 I  E  H P L P L P L 100 99 100 99 100 100 I  E  H P L P L P L 100 99 98 96 87 65 I  E  H P L P L P L 76 74 X -/89 i ii I E H nr nr X X # # iii iv I E H nr nr nr vi v I E H nr nr X X * nr nr *nr ITS + EFL + HSP70  Figure 4.6. Maximum parsimony tree with midpoint rooting from the ITS, EFL and HSP70 sequences, showing concordant support for the monophyly of four species of Sphaeroforma and contrasting lack of support for within species branching order. 116  Numbers below nodes represent 100 maximum parsimony and 500 RAxML bootstrap replicates (MP/RAxML). Nodes numbered 1-3, A-B, i-vi range from complete concordance (red background) to conflict (grey backgrounds) when each locus was analyzed separately. The results from independent analyses of each of the loci, ITS (I) EFL (E) and HSP70 (H) are given in boxes near each of the nodes 1-3, A-B; maximum parsimony bootstrap support is given over ‗P‘ and maximum likelihood or RAxML bootstrap support over the ‗L‘. Letters shaded in grey indicate lack of resolution or bootstrap support below 60 %; letters are crossed out if an alternate topology received 60 % or more likelihood or parsimony bootstrap support. Boxes on right show support patterns for within species nodes i-vi. Check marks signify support, X‘s conflict and nr signifies a lack of resolution. Nodes that were unresolved phylogenetically but clearly not supported by distribution of alleles (Table 4.1) are marked by a X # . Nodes that did have support but not as a sister group to the other node are marked by *. New Sphaeroforma isolates from British Columbia are in bold. 117   Figure 4.7. Morphology and spore release in Sphaeroforma species. A.  S. nootkatensis isolate RU1 showing a variety of cell sizes including mature cells (M) that were subdivided into visible endospores and persistent cast-off parental cell walls (W). Parental walls split into two valves during endospore release and each cast off wall was visibly torn B. S. arctica wall cast-offs were also visibly torn into two valves. Some endospores (e) were still visible within parental wall remnants. C.  S. gastrica at varied developmental stages. Unreleased endospores are still visible within some cells. D-F. Endospore release in S. tapetis isolate CG2. Amoeboid endospores (a) squeezed through several pores in the parental cell wall. Once outside these amoebae quickly became round endospores. The parental cell wall did not split. G. S. tapetis isolate LE7 with growing endospores next to a parental cell wall that had split into two valves. H. Endospores of S. nootkatensis isolate RU1 in arrested development after two rounds of reproduction in unsupplemented seawater. The original cell wall from the first round of reproduction was still persistent (O). I. Cells of S. nootkatensis isolate MC5 are shown here attached to cotton fibres. A-C scale bars equal 50 microns; D-F scale bars equal 10 microns; G-I scale bars equal 20 microns. 118   Figure 4.8. Colony morphologies in Sphaeroforma. A. S. nootkatensis isolate RU1 colony with no evidence of slimy extracellular material or plasmodial cells. B. S. tapetis isolate LE7 colony with plasmodial cells (P). C. S. tapetis isolate CG2 colony with one rare plasmodial cell and a trail of extracellular material (E). D. S. tapetis isolate LE7 colony surrounded by slimy extracellular material. Scale bar equals 100 microns. 119   Figure 4.9.  Whole cells of S. tapetis undergoing vegetative growth and maturation. A. Vegetatively growing cell with multiple nuclei (N) positioned around the periphery of the cell. The cell was bound by an opaque cell wall and the cytoplasm contained several types of vesicles, many of which appeared to ring the nucleus. (Isolate CG2). B. Multinucleate, vegetative cell with large deposits of glycogen (G). Vesicles (V) with opaque and lucent contents were near the nuclei. At least two peripheral nucleoli (Nu) were visible in some nuclei. The cell wall was thicker at polar ends of the cell (white arrowheads). (Isolate Le7). C. Parental cell with wall-less endospores seperated by plasma membranes (PM). A thin cell coat or calyx (Ca) was attached to the outside of the parental cell wall. (Isolate CG2). D. Parental cell with developing endospores; some endospores shifted during sectioning producing white spaces as artifacts. Most of the endospores were wall-less but CW* shows a tangential section through a spore with a developing cell wall. Thickened regions of the parental cell wall were visible at polar ends of the cell. The endospore indicated by the black arrow had more than one nucleus and may have been a precursor to an endospore.  (Isolate CG2). E. Parental cell. The central endospore had a cell wall. (Isolate Le7). Scale bars equal 2 microns. 120   Figure 4.10. Whole cells of S. nootkatensis undergoing vegetative growth and maturation. A. Young cell, enclosed in a cell wall (CW) with a peripheral coating or calyx (Ca), showing a nucleus (N) with an adjacent Golgi body (go), and with mitochondria (M) with flat cristae, vesicles (V), and glycogen rosettes (G). B. A mature cell with unreleased, walled endospores. Some vesicles‘ contents may have changed during processing.  C. Endospores developing within a parent cell. Cell walls were at various stages and one irregularly shaped spore (lower left), bound only by a plasma membrane (PM), contained internally-coated vesicles (fv).  Wall material (*) in the mother cell was presumably on route to the developing spore(see below Figure 4.12 E-F). Some vesicles had opaque contents and two Golgi bodies were associated with one of the nuclei.  Scale bars equal 2 microns. 121   Figure 4.11. Closer view of internally coated fibrous vesicles and tubular extensions (MTBs) into the cell wall of S. tapetis and S. nootkatensis. A. Cross sections taken next to a patch of cell wall between ‗spores‘ within a cell similar to the one shown in Figure 4.14G. The contents of internally coated fibrous vesicles (fv) appear striated and fibrous in the tangential section. The cross section clearly shows that the contents adhered to the inner circumference of the vesicle. Most vesicles had two, sometimes three gaps where there was no adhering material (arrows). Internal cell wall material (ICW) between developing irregular shaped endospores was accompanied by MTBs (T) with a visible plasma membrane (PM). (Isolate LE7). B. MTBs penetrating the electron lucent inner layer of the external cell wall. The outer layer of the cell wall (CW) was comparatively more opaque. (Isolate Le7). C. MTBs in S. nootkatensis that were clearly contiguous with the cellular cytoplasm (E) and bound by a plasma membrane. Pores (P) in the outer wall layer and an external calyx (Ca) were visible. Glycogen, G. Scale bars equal 100 nm. 122   Figure 4.12. Cell wall development and structure in S. tapetis. A. Partial view of three unreleased endospores bound by plasma membranes but no cell wall. Vesicles lined with fibrous content (fv) were abundant, some of the fibrous material was in linear form (lfv) rather than enclosed in a vesicle. (Isolate CG2).  B. Developing spore with a cell wall (CW) and lengths of plasma membrane (PM*) that appeared to be joining with the outer membrane of the spore. Electron opaque inclusion bodies (I) with no visible bounding membrane were present. (Isolate Le7). C. Detail of Figure B, showing circular, and linear pieces of plasma membrane (PM*), including a piece that 123  was contiguous with the outer cell membrane of the endospore. D. Cell wall deposition (*) of electron opaque material originating from the parental or investing cell and binding to the outside of the developing endospore. Pores (P) were visible in the endospore wall. (Isolate CG2). E. Spore wall deposition (*) of developing endospore next to parental cell wall (PCW). Mitochondria (M) with plate-like cristae, fibrous vesicles, other vesicles (V) and nuclei (N) were visible within most endospores. (Isolate CG2). F.  Material that originated from the parental cell was deposited on the developing endospore wall. Pores were visible in the endospore wall. (Isolate Le7). G. Endospores with developed cell wall and calyx (Ca) that were not yet released from the parent cell. Pores were present in the endospore walls but not in the thinner parental cell wall. (Isolate CG2). H. Detail of one polar thickened region of the parental cell wall from the cell shown in Fig. 4.9D. A thin calyx was visible, as were lucent vesicles (Va) within the thickened region of the wall. (Isolate CG2). I. Detail of the polar thickening of the cell wall from vegetative cell shown in Fig. 4.9B. Smooth endoplasmic reticulum (SER) was present in the cytoplasm near the wall thickening and small vesicles were present within the wall material. At least two distinct layers were present in the cell wall (L1 and L2). (Isolate Le7). J. Cell wall and MTBs (T) extending into the inner electron lucent wall layer. (Isolate Le7). Glycogen, G. Scale bars equal 500 nm. 124   Figure 4.13. Details of S. nootkatensis ultrastructure. A. Nucleus (N) with peripheral nucleolus (nu), encircled by vesicles lined with fibrous material (fv). One Golgi body (go) with four stacked cisternae was visible in this section. B. Two visible Golgi bodies, each with a spindle pole body (SPB), located at the polar ends of the nucleus that was probably preparing for mitotic division. Fibre containing vesicles were visible along one side of the nucleus. MTBs (T) were visible penetrating the inner electron lucent layer of the cell wall (CW). C. Golgi body and associated multi- vesicular body (MVB) and membrane (PM) bound vesicles. D. Nucleus with Golgi body and spindle pole body with extranuclear microtubules (Mt). E. Nucleus with peripheral nucleolus, Golgi body with stacked cisternae and possibly associated fibrous vesicle. F. Cell walls of developing endospores. The loops of plasma membrane were joined to the outer cell membrane (PM). Pores (P) were visible in the outer layer of the cell wall. G-H. Cell walls of two adjacent cells. The calyx (Ca) and both layers of the cell wall were visible. The pores in the electron opaque outer layer were visible in tangential section and the MTBs were visibly embedded in the lucent inner layer. Mitochondria, M; glycogen, G, vesicle, V. Scale bars equal 500 nm. 125   Figure 4.14. Details of S. tapetis ultrastructure. A-F. Nuclei and associated structures of S. tapetis. G-H. Images from alternate forms of S. tapetis isolate Le7. A. Nucleus (N) with peripheral nucleolus (Nu) encircled with fibrous vesicles (fv). Cell wall (CW) was thick and composed of an outer opaque layer and irregular, less opaque inner layer. The plasma membrane and cell boundary was wavy and convoluted.  (Isolate Le7). B. Nucleus, spindle pole body (SPB) with extranuclear microtubules (Mt), and surrounding fibrous vesicles. (Isolate CG2). C. 126  Nucleus with spindle pole body and associated Golgi body (Go). (Isolate CG2). D. Tangential section through spindle pole body. Golgi stacks and nuclear pores (np) were visible. (Isolate CG2). E. Golgi body and vesicles next to nucleus. (Isolate Le7). F. Fibrous vesicles and unusual membrane bound vesicles (MBV) near the nucleus of a developing endospore. (Isolate Le7). G. Alternate cellular arrangement found in isolate Le7 with irregularly shaped cells with patches of cell wall material (ICW) between what appeared to be irregularly shaped spores. H.  Nuclei in these irregular cells often had spindle pole bodies with intranuclear microtubules (IMt) as well as extranuclear mictrotubules (EMt). (Isolate Le7). I. Patch of cell wall material between three irregularly shaped spores from a cell similar to G. MTBs (T) were present near these patches of wall material. Vesicles, V; glycogen, G. Scale bars equal 500 nm for A-F, H; 2 microns for G, and 200 nm for I. 127  0 0.1 0.2 0.3 0.4 0.5 0.6 S . t ap et is S . g as tr ic a S . n oo tk at en si s P re v a le n c e Peanut worm (n=5) tunicate (n=2) echinoderms (n=13) other molluscs (n=4) biv alv es (n=14) 0 1 2 3 4 5 6 S . t ap et is S . g as tr ic a S . n oo tk at en si sN u m b e r o f h o s t c o ll e c ti o n s  Figure 4.15. Frequency and prevalence of isolation of each British Columbian Sphaeroforma species from four types of hosts. The left graph shows the total number of collections of each type of animal host harbouring each Sphaeroforma species. The graph to the right shows prevalence, calculated by dividing the number of host animal collections infected by the total number of host animal collections for each host animal type. 128  References Adl S Simpson AG Farmer MA Anderson RA Anderson OR Barta JR Bowser SS Brugerolle G Fensome RA Fredericq S James TY Karpov SA Kugrens P Krug J Lane CE Lewis LA Lodge J Lynn DH Mann DG McCourt RM Mendoza L Moestrup Ø Mozley-Standridge SE Nerad TA Shearer CA Smirnov AV Spiegel FW Taylor M (2005) The new higher level classification of eukaryotes with emphasis on the taxonomy of protists. J Eukaryot Microbiol 52:399-451 Agapow PM Burt A (2001) Indices of multilocus linkage disequilibrium. Mol Ecol Notes 1:101-102 Amato A Kooistra WHCF Ghiron JHL Mann DG Pröschold T Montresor M (2007) Reproductive isolation among sympatric cryptic species in marine diatoms. Protist 158:193-207 Ashworth JH (1923) On Rhinosporidium seeberi (Wernike 1903), with special reference to its sporulation and affinities. Trans R Soc Edinb 53:301-342 Baum DA Shaw KL (1995) Genealogical perspectives on the species problem. Pages 289-303 in P. C. Hoch and A. G. Stephenson, editors. Experimental and Molecular Approaches to Plant Biosystematics. Missouri Botanical Garden, St. Louis Beebee TJC (1991) Purification of an agent causing growth inhibition in anuran larvae and its identification as a unicellular unpigmented alga. Canadian Journal of Zoology 69:2146-2153. Bourne N (1982) Distribution, reproduction and growth of Manila clam, Tapes philippinarum (Adams and Reeve), in British Columbia. J Shellfish Res 2:47-54 Brown H (1942) A micro culture slide for fungi. J Bacteriol 43:16 Cafaro MJ (2000) Gut fungi of isopods: the genus Palavascia. Mycologia 92:361-369 Cafaro MJ (2005) Eccrinales (Trichomycetes) are not fungi, but a clade of protists at the early divergence of animals and fungi. Mol Phyl Evol 35:21-34 Catania F Wurmser F Potekhin AA Przyboś E Lynch M (2009) Genetic diversity in the Paramecium aurelia species complex. Mol Biol Evol 26:421-431 129  Cavalier-Smith T (1998) Neomonada and the origin of animal and fungi. Pages 375-407 in G. Coombs, K. Vickerman, M. Sleigh, and A. Warren, editors. Evolutionary relationships among protozoa. Chapman and Hall, London Dang S-N Lichtwardt RW (1979) Fine structure of Paramoebidium (Trichomycetes) and a new species with virus-like particles. Am J Bot 66:1093-1104 de Vargas C Renaud S Hilbrecht H J P (2001) Pleistocene adaptive radiation in Globorotalia truncatulinoides: genetic, morphologic, and environmental evidence. Paleobiology 27:104-125 Dudas SE Dower JF (2006) Reproductive ecology and dispersal potential of varnish clam Nuttallia obscurata, a recent invader in the Northeast Pacific Ocean. Mar Ecol Prog Ser 320:195-205 Figueras A Lorenzo G Ordas MC Gouy M Novoa B (2000) Sequence of the small subunit ribosomal DNA gene of Pseudoperkinsus tapetis isolated carpet shell clam in Galicia (NW of Spain). Mar Biotechnol 2:419-428 Franco-Sierra A Alvarez-Pellitero P (1999) The morphology of Ichthyophonus sp. in their mugilid hosts (Pisces: Teleostei) and following cultivation in vitro. A light and electron microscopy study. Parasitol Res 85:562-575 Frézal L Leblois R (2008) Four years of DNA barcoding: current advances and prospects. Infect Genet Evol 8:727-736 Fu YX (1997) Statistical tests of neutrality of mutations against population growth, hitchiking and background selection. Genetics 147:915-925 Hebert PDN Stoeckle MY Zemlak TS Francis CM (2004) Identification of birds through DNA barcodes. Plos Biol 2:1657-1663 Hudson RR (2000) A new statistic for detecting genetic differentiation. Genetics 155:2011-2014 Hudson RR Boos DD Kaplan N (1992) A statistical test for detecting population subdivision. Mol Biol Evol 9:138-151 Jostensen JP Sperstad S Johansen S Landfald B (2002) Molecular-phylogenetic, structural and biochemical features of a cold adapted, marine ichthysporean near the animal-fungal divergence, described from in vitro cultures. Europ J Protistol 38:93-104 130  Kasuga T Taylor JW White TJ (1999) Phylogenetic relationships of varieties and geographical groups of the human pathogenic fungus, Histoplasma capsulatum. J Clin Microbiol 37:653-663 Maddison WP Maddison DR (2000) MacClade 4: Interactive analysis of phylogeny and character evolution. Sinauer, Sunderland, MA, USA Marshall WL Celio G McLaughlin DJ Berbee ML (2008) Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. Protist 159:415-433 Mayfield S-ND Lichtwardt RW (1980) Comparative study of the holdfast structure in four Trichomycetes. Can J Bot 58:1074-1087 Mendoza L Taylor JW Ajello L (2002) The class Mesomycetozoea: a heterogeneous group of microorganisms at the animal-fungal boundary. Ann Rev Microbiol 56:315-344 Pekkarinen M Lom J Murphy CA Ragan MA Dykova I (2003) Phylogenetic position and ultrastructure of two Dermocystidium species (Ichthyosporea) from the common perch (Perca fluviatilus). Acta Protozool 42:287-307 Pereira CN Rosa IDI Fagotti A Simoncelli F Pascolini R Mendoza L (2005) The pathogen of frogs Amphibiocystidium ranae ia a member of the Order Dermocystida in the class mesomycetozoa. J Clin Microbiol 43:192-198 Posada D Crandall KA (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818 Ramos-Onsins SE Rozas J (2002) Statistical properties of new neutrality tests against population growth. Mol Biol Evol 19:2092-2100 Reece KS Bushek D Hudson KL Graves JE (2001) Geographic distribution of Perkinsus marinus genetic strains along the Atlantic and Gulf coasts of the USA. Mar Biol 139:1047-1055 Rozas J Sanchez-DelBarrio JC Messeguer X Rozas R (2003) DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19:2496-2497 131  Ruiz-Trillo I. Burger G Holland PWH King N Lang BF Roger AJ Gray MW (2007  T he origins of multicellularity: a multi-taxon genome initiative. Trends in Genet  23:113-118 Rynearson TA Armbrust EV (2005) Maintenance of clonal diversity during a spring bloom of the centric diatom Ditylum brightwellii. Mol Ecol 14:1631-1640 Schlegel M Meisterfeld R (2003) The species problem in protozoa revisited. Europ J Protistol 39:349-355 Spurr AR (1969) A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26:31-43 Stamatakis A (2006) RAxML-VI-HPC: Maximum likelihood-based analyses with thousands of taxa and mixed models. Bioinformatics 22:2688-2690 Swofford DL (1998) PAUP*: phylogenetic analysis using parsimony (*and other methods). Version 4. Sinauer Associates. Sunderland, Mass Takishita K Fujiwara Y Kawato M Kakizoe N Miyazaki M Maruyama T (2008) Molecular identification of the ichthyosporean protist "Pseudoperkinsus tapetis" from mytilid mussel Adipicola pacifica associated with submerged whale carcasses in Japan. Marine Biotechnology 10:13-18 Taylor JW Jacobson DJ Kroken S Kasuga T Geiser DM Hibbit DS Fisher MC (2000) Phylogenetic species recognition and species concepts in fungi. Fungal Genet Biol 31:21-32 Thompson JD Gibson TJ Plewniak F Jeanmougin F Higgins DG (1997) The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl Acids Res 24:4876-4882 Thomson RE (1982) Oceanography of the British Columbia coast. Can Spec Publ Fish Aquat Sci 56:291 Tibayrenc M (2006) The species concept in parasites and other pathogens: a pragmatic approach. Trends Parasitol 22:66-70 Weir BS (1996) Genetic data analysis II. Sinauer., Sunderland, MA Wiley EO (1978) The evolutionary species concept reconsidered. Syst Zool 27:17-26 132  Chapter 5. Facing unknowns: Living cultures (Pirum gemmata gen. nov., sp. nov., and Abeoforma whisleri, gen. nov., sp. nov.) from invertebrate digestive tracts represent an undescribed clade within the unicellular opisthokont lineage ichthyosporea (Mesomycetozoa). 1   Introduction Culture independent molecular surveys of eukaryotes have shown unexpectedly high diversity and several novel clades (e.g. Dawson and Pace 2002; Massana et al. 2004). These molecular surveys rely on PCR amplification and sequences from a single gene, usually the small subunit ribosomal DNA (SSU-rDNA), to represent the organisms present in a sample. There are several advantages to these surveys compared to time consuming isolation and culture of individual species. Unculturable and morphologically indistinguishable organisms can be identified by a phylotype and microbial communities can be compared relatively quickly using easily standardized methodologies. The largest drawback to this form of sampling is that the only information obtained is the phylotype and any environmental and community variables related to the sample source. The organisms from which these sequences are derived are most often otherwise unknown. Several PCR based surveys have detected lineages with no sequences from closely related taxa (e.g. Dawson and Pace 2002; Amaral Zettler et al. 2003; Berney et al. 2004). Without cultures or fixed preparations of these divergent organisms we now know that we are missing some of the range of the diversity of form and function of microscopic eukaryotes. Isolation and detailed descriptions of novel taxa corresponding to these divergent phylotypes promises to unveil new and unique life forms and increase our appreciation of the diversity of adaptations found in living organisms. During a culture based survey for novel species of osmotrophic marine opisthokonts, we isolated two unicellular organisms with SSU-rDNA sequence similarity to undescribed phylotypes belonging to the protistan group ichthyosporea.  1  A version of this chapter will be submitted.. Marshall WL Berbee ML. Facing unknowns: Living cultures (Pirum gemmata gen. nov., sp. nov., and Abeoforma whisleri, gen. nov., sp. nov.) from invertebrate digestive tracts represent an undescribed clade within the unicellular opisthokont lineage ichthyosporea (Mesomycetozoa). 133  Ichthyosporeans are one of six unicellular lineages belonging to the eukaryote supergroup opisthokonta which also includes the multicellular animals and fungi (Adl et al. 2005). The unicellular relatives of the animals and fungi are not well studied due to a lack of available cultures in culture collections and relatively low levels of taxonomic diversity. The SSU-rDNA sequences from our isolates had no closely related sequenced relatives and were instead most similar to a divergent clade of ichthyosporeans that was represented solely by phylotypes collected during culture independent PCR surveys. The ichthyosporea are comprised of at least 30 genera, five of which can be grown in culture. SSU-rDNA phylogenies support the division of the ichthyosporea into two lineages, the dermocystida and the ichthyophonida (Cavalier-Smith 1998) (synonymous to rhinosporideacae and ichthyophonae) (Adl et al. 2005). All ichthyosporeans have been isolated from animals but as far as is known the dermocystida are restricted to vertebrate hosts and do not culture outside host cells. The ichthyophonida utilize a wide range of host phyla; whilst many are host obligate, all cultured species belong to this group. When motile, the dermocystida have a posterior flagellum whereas motile ichthyophonida have amoeboid stages. All described ichthyosporeans from both lineages have cell walls and obtain nutrients from their hosts and in culture osmotrophically (Mendoza et al. 2002). Further defining characteristics of the ichthyophonida, dermocystida, and the ichthyosporea as a whole, awaits description and comparative studies of more taxa, especially those from previously undescribed lineages. In this paper we use time lapse photography, light microscopy and transmission electron microscopy to describe the morphology, growth and asexual reproduction of two new genera where only environmental sequences had been known before.  We will describe our two new genera and species as ―Pirum gemmata” and ―Abeoforma whisleri‖.  Methods Isolation, DNA extraction and phylogenetic analyses We discovered “P. gemmata” and ―A. whisleri” in cultures started from marine invertebrate digestive tract contents. Host animals were collected from Barkley Sound 134  and Gabriola Island in British Columbia, Canada on November 2004 and April 2007.  We isolated “P. gemmata” and “A. whisleri” n. gen. et sp. one time each on YPSS media at 15 °C and maintained axenic cultures on YPD seawater media according to Marshall et al. (2008). Isolate “P. gemmata” came from the gut contents of the detritivorous peanut worm Phascolosoma agassizii and was not present in cultures of coelomic fluid. “A. whisleri” came from material from the digestive tract of the filter feeding mussel Mytilus sp.; we did not culture the mantle cavity fluid. Both cultures remained viable through successive rounds of transfer for over two years and were deposited at the American Type Culture Collection (ATCC) where they are accessioned as PRA-279 and PRA-280. We extracted the DNA, and amplified and sequenced the SSU-rDNA from both isolates according to Marshall et al. (2008). We aligned our sequences with representative sequences from most ichthyosporean genera, other opisthokont lineages and outgroup taxa selected from GenBank (Table E.1) using Clustal X v1.8 (Thompson et al. 1997). Our isolates appeared most closely related to uncultured clone sequences, therefore we included sequences from each known opisthokont lineage and a selection of environmental sequences in an alignment of 76 sequences. We located our environmental sequences by using the results from BLAST searches (Altschul et al. 1997) from representatives of each unicellular opisthokont lineage. We screened the environmental sequences using the distance trees generated in the BLAST program and fast parsimony analyses. Our preliminary parsimony analyses involved 500 bootstrap replications using simple sequence addition in PAUP v4.0b10 (Swofford 1998). We did not include environmental sequences that did not appear to be opisthokonts, nor ones that grouped within animals, higher fungi, or choanoflagellates. Due to the high number of environmental sequences belonging to the choanoflagellates and  fungi we restricted the sequences from these groups to a few representatives of previously identified ‗novel‘ lineages related to the chytrid and zygomycete fungi (e.g. Luo et al. 2005). We excluded any sequence grouping within the choanoflagellates or established fungal lineages with greater than 60 % support. We analyzed the non-ambiguous regions of our data matrix using maximum parsimony, maximum likelihood and Bayesian methods. For parsimony bootstrapping, we used PAUP v.4.0b10 (Swofford 1998) to generate 500 heuristic search replicates with 135  random sequence addition (n=10) and tree-bisection branch swapping limited to 1 x 10 8  rearrangements. For Bayesian and maximum likelihood analyses we used a GTR+G model of nucleotide substitution based on parameters estimated by ModelTest v3.7 (Posada and Crandall 1998).We used MrBayes v3.0.2 (Huelsenbeck and Ronquist 2001) for our Bayesian analysis, which consisted of 2, 000, 000 generations with a 1/1000 sample frequency and the first 500 trees discarded as a burnin. The mean likelihood of the run was -17576.4 (± 0.31) and the convergence diagnostic was within 1.0 (± 0.003). For maximum likelihood we used RAxML (Stamatakis 2006) to find the most likely tree using 200 heuristic search replicates and bootstrapping using 500 replications.  Light and electron microscopy  Both new species grew slowly under our culture conditions so we could not follow single cells through complete cycles of asexual reproduction. Our interpretations were therefore synthesized from combined observations of many cells and life stages during a variety of time periods. We followed single cells on a daily basis for up to one week using Brown‘s fungus slides (Brown 1942). For shorter term observations we either observed cells directly, used time lapse photography within slide chambers (Marshall et al. 2008), or VHS video recording. We DAPI stained according to (Marshall et al. 2008). For transmission electron microscopy, we cultured, fixed and prepared “P. gemmata” according to (Marshall et al. 2008).  For “A. whisleri” we cultured cells on 1% yeast extract, 1% peptone, 10% dextrose, seawater agar with a broth overlay. We harvested “A. whisleri” from cultures ranging from four days to two weeks old by adding more broth and using a widened pipette tip to loosen and remove cells. We concentrated the cells using gravity and low speed centrifugation and removed the excess moisture with filter paper. We froze the cells in 100 and 200 µm brass planchettes under high pressure using a BAL-TEC HPM-010 and freeze substituted them in a Leica AFS chamber using 0.1% uranyl acetate and 1% OsO4 in acetone as a substitution medium. We microwave infiltrated and embedded “A. whisleri” in a 50/50 Spurrs/Epon resin. “P. gemmata” was also high pressure frozen but we used a 0.1% uranyl acetate and 2% OsO4 in acetone as a substitution medium and Quetol II for resin (Marshall et al 2008). We 136  embedded both isolates flat between slides and cut 60 nm sections. We stained the sections with 3 % uranyl acetate for 20 minutes and lead citrate for 5 minutes. We photographed our sections using a Hitachi H7600 transmission electron microscope.  Results Phylogenetic inference The SSU-rDNA sequences of “P. gemmata” and “A. whisleri” were 1678 and 1674 nucleotides in length. Phylogenetic inference based on the SSU-rDNA placed both isolates within the ichthyosporean lineage ichthyophonida with each other as their closest sequenced relative. Including gaps they shared a 95.7 % similarity to each other over a 1675 bp region of overlap and they both shared 95% similarity to uncultured clone NAMAKO-34 over the entire sequenced region. According to a BLAST search, the most closely related sequences from described taxa were Amoebidium parasiticum, Sphaeroforma arctica and Creolimax fragrantissima which each shared 90 % similarity over 96 % coverage relative to our sequences.  Our analysis showed that both “P. gemmata” and “A. whisleri” belong to a clade within the ichthyophonida composed of 13 undescribed sequences amplified from culture independent PCR surveys. This clade was distinguished from other ichthyophonida by > 90 % support in all of our analyses. Our species were most closely related to environmental clones s16 and NAMAKO-34 (Fig. 5.1). NAMAKO-34 was sequenced during a PCR-based survey of DNA found in anoxic sediment from Lake Namako, Japan (Takishita et al. 2007a). Clone s16 was one of several clone sequences belonging to this clade that were isolated during a survey of open water from the Atlantic Sea of Fundy (Savin et al. 2004) (Table E.1). Other uncultured clones were in a clade of spherical ichthyosporeans related to freshwater Anurofeca and marine genera Pseudoperkinsus, Sphaeroforma and Creolimax. We also found one or two potentially undescribed opisthokont lineages based on the environmental clone sequences included in our data set (Fig. 5.1). Uncultured clone M1_18C07 (uncultured opisthokont lineage 1) did not group within any of the molecularly described lineages and was supported as a sister sequence to 137  Corallochytrium according to our Bayesian analysis. Parsimony topology placed clone M1_18C07 as a sister to the choanoflagellates, but with low statistical support. The second group of uncultured opisthokonts (uncultured opisthokont lineage 2) included cloned sequences RT5iin16 and RT5iin16. These sequences appeared as a sister group to the nuclearia in each method but this placement was only supported by moderate Bayesian posterior probability (Fig. 5.1). We found no environmental sequences that grouped with Capsaspora, Ministeria, within the nuclearia, Corallochytrium (other than clone M1_18C07) or the ichthyosporean dermocystida. This contrasted strongly with the numbers of environmental sequences available for the fungi and choanoflagellates and to a lesser degree the ichthyophonida.  Light and electron microscopy “P. gemmata:” Our observations of asexual growth and reproduction in “P. gemmata” were based on direct observations of spore release and time-lapse photography showing sporogenesis. “P. gemmata” did not grow well in Brown‘s fungus slide chambers and we were unable to follow individual cells through complete rounds of asexual reproduction. We divided the life cycle of “P. gemmata” into three stages: vegetative growth (Fig. 5.2), sporogenesis (Figs. 5.2D, 5.3) and endospore release and dispersal (Fig. 5.4). Vegetative cells were round in cross section and pear shaped in longitudinal section (Fig. 5.2). Cells undergoing vegetative growth had a prominent central or system of vacuoles that occupied most of the cell volume. The cytoplasm was typically pressed along the cell periphery and in thin strands or bridges between vacuoles (Fig. 5.2). DAPI staining showed that vegetative and mature cells were multinucleate (Fig. 5.5). Younger cells had 1, 2, 4, 8, 16, 32, and 64 nuclei, demonstrating that nuclei reproduced by synchronous division; older cells had nuclei in the hundreds. Cells commonly reached between 45-65 microns in width and 70 to 100 microns in length before the onset of sporogenesis (Fig. 5.3). Sporogenesis involved serial cleavage of the entire cytoplasm within a common parental cell wall. Each cleavage produced smaller divisions of membrane bound 138  cytoplasm (Fig. 5.3). We could not tell whether each cleavage was bilaterally symmetrical but each newly formed subdivision appeared to undergo further subdivision. As the overall number of subdivisions increased, the endospore precursors became too small to see and the cytoplasm became granular in appearance (Fig. 5.3). Endospores of 4-6 m in diameter exited in pulses through one or more tears in the cell wall. Usually the spores were solitary and unattached but occasionally they formed chains as in Figure 5.4B. Often some endospores remained within the parent cell wall and began vegetative growth (Fig. 5.2E). Endospores were non-motile but did exhibit some amoeboid properties. They made subtle changes in shape and had pseudopod-like extensions. The pseudopod-like extensions appeared to be stationary and were not used for locomotion and in liquid medium the endospores drifted passively away from the parent cell (Fig. 5.4).  In developing cells the size of the nuclei increased until the onset of nuclear division, after which they decreased in size (Fig. 5.5). The nuclear size of emerging endospores was approximately 1.8 m in diameter (Fig 5.5I) and grew to approximately 4 m as the uninucleate endospores grew vegetatively (Fig. 5.5 B-C). Following the onset of nuclear division the nuclear size gradually decreased as the number of nuclei increased and the cell approached maturity (Fig. 5.5 D-K).  Ultrathin sections of “P. gemmata” showed cells undergoing vegetative growth (Fig. 5.6A). These cells were multinucleate, coenocytic and bound by a cell wall. A large central vacuole and what appeared to be either budding or fusing smaller vesicles occupied much of the cell volume. The cytoplasm was densely packed with glycogen rosettes and additional vesicles with contents of varying consistency (Fig. 5.7). Mitochondria were not well preserved but had plate-like cristae (Fig. 5.6D). Figure 5.6 (B-E) show mature reproductive cells with membrane bound endospores, some of which were multinucleate and may have been endospore precursors. The parent cell wall was still present (Fig. 5.6 B, D-E) but endospores and potential precursors did not have cell walls. Young endospores or their precursors had small lipid bodies and internal concentric rings of plasma membrane. The cell shown in Figure 6E had an opening in the parent cell wall and at least one multinucleate daughter cell and could be interpreted in two ways. This cell could have been immature, the opening in the wall due to a fixation 139  artifact, and the multinucleate cell could have been an incompletely cleaved endospore precursor. Alternately it could have been past maturity with at least one growing multinucleate daughter cell (similar to the cell shown in Figure 5.2E). If it was past maturity, the multinucleate daughter cell had undergone nuclear division prior to production of a cell wall. Inward extensions of the outer plasma membrane (Fig. 5.6 C, E) were present on at least one endospore. It was not clear whether these invaginations were due to increase in cell circumference, or part of the cleavage process during endospore formation, or both. Figure 5.6D shows a segment of plasma membrane that was close to the outer plasma membrane and may have been about to join in order to increase the cell circumference. Nuclei had prominent nucleoli and regular nuclear pores (Fig. 5.7). A spindle pole body with extranuclear microtubules was often next to the nucleus (Fig. 5.7). The spindle pole bodies were electron opaque plaques, were crescent shaped in cross section, round in tangential section and approximately 200 nm in diameter. Microtubules appeared to originate from a layer of electron lucent material surrounding each plaque. Networks of small tubules were usually found associated with the spindle pole bodies and at each pole of the nucleus. The tubules resembled smooth endoplasmic reticulum (SER) and stacked cisternae characteristic of Golgi bodies were not observed.  Cell walls were composed of an outer, continuous, fibrous, electron opaque layer and a discontinuous electron lucent layer (Fig. 5.8).  In cross section the outer fibrous layer was at least 100 nm thick and the fibres were arranged in layers that ran parallel to the cell surface (Fig. 5.8F). In tangential section the fibres of the outer layer were interwoven (Fig. 5.8G). The inner wall layer was variable in distribution and thickness. In some areas it was not visible and the plasma membrane appeared to be adjacent to the outer fibrous layer. In most vegetative cells the inner layer formed periodic thickenings, extending inwards and creating indentations in the cell‘s cytoplasm. Membrane bound tubules (MBTs) extended from the cell and penetrated the lucent wall layer in each of the thickenings. Occasionally coated vesicles and networks of tubules resembling SER were present in the cytoplasm near the thickenings (Fig. 5.8C,-D).  “A. whisleri:” 140   Our culture of “A. whisleri” grew relatively slowly compared to “P. gemmata”. Cells had a more diverse variety of states, morphologies and methods of reproduction. We were unable to determine the complete cycle of asexual reproduction nor determine whether there was a fixed order for change between the various states. Our observations were based on daily observations of single cells and colonies up to seven days, direct observation and VHS video of cells in slide chambers or on agar plates. “A. whisleri” was most frequently spherical in shape with prominent vacuoles (Fig. 5.9) but plasmodia and various amoeboid shapes were also common (Figs. 5.9- 5.12). Spherical cells typically ranged from 20-50 µm in diameter. Plasmodia were usually large, irregularly shaped and often up to 120 µm in length (Fig. 5.9 B-C, I-K). Cells also took hyphal forms (Fig. 5.9L) and were sometimes as narrow as 5µm. Amoebae came in three forms (Fig. 5.10). The most common amoebae appeared to function in post reproductive dispersal (Fig. 5.10A). These ‗dispersal‘ amoebae actively crawled away from parent cells and had one to several dynamic and often bifurcating pseudopodia. Filose amoebae (Fig. 5.10C) with short pseudopodia and branched amoebae (Fig. 5.10 D-F) were less common but were each observed on several occasions. Filose amoebae had short, fine, active pseudopodia covering their surface and were usually between 20-30 µm in diameter. Branched amoebae had the majority of their cytoplasm confined to the termini of two or more branches. Each terminus had actively moving pseudopodia and was separated from the other termini by long narrow cytoplasmic strands as narrow as 1.5 µm (Fig. 5.10 D-F). One terminus always appeared enlarged and ovoid relative to the others and probably contained the nucleus (Fig. 5.10 D- F). The amoeba in Figure 5.10 (D-E) was attached to debris in the media by one of the terminal ends for two days before it released and moved away (Fig. 5.10F). In this instance the attached terminus appeared to be adapted as a holdfast. Both filose and branched amoebae had dynamic pseudopodia but did not actively change location and often stayed in the same spot for up to a day. In contrast, the dispersal amoebae were actively crawling except when they were undergoing division. DAPI staining showed that dispersal amoebae were uninucleate and the single enlarged terminus of the branched amoebae suggested that these amoebae were also uninucleate. The number of nuclei in the larger diameter filose amoebae was not determined. 141  Reproduction was also variable. We occasionally saw spherical cells releasing endospores in spurts, as in “P. gemmata” (Fig. 5.9D). In other instances the spherical cells formed tetrads, octets, or greater numbers of subdivisions (Fig. 5.9 E-H) ultimately releasing weakly amoeboid endospores (Fig. 5.9G) or actively dispersing amoebae. The amoeboid endospores were similar in volume to the dispersal amoebae and may have been the same cell type but in a less active state. Both underwent binary fission (Fig. 5.9G). When dispersal amoebae underwent binary fission they became inactive for at least a day and strongly resembled the amoeboid endospores. Plasmodia reproduced both by budding (Fig. 5.9 K-L), or changing into a spherical state and releasing amoebae (Figs. 5.9J, 5.10A). In addition to budding, plasmodia also grew hypha-like shapes (Fig. 5.9 K-L). Empty trails composed of discarded extracellular cell coat or cell wall material were frequently observed in the cultured material. These trails were evidently left after plasmodial growth. This was typically observed as plasmodia in extended and hypha-like forms retracted their cellular content into a single spherical form (Fig. 5.11). Therefore both the transition between a sphere to a plasmodial or hypha-like state and the reverse both occurred; however, we did not determine whether a single cell could switch between each state multiple times. Figure 12 shows the formation of a colony from a single plasmodial cell. For this colony reproduction was primarily due to plasmodial budding, although after one day the original cell produced amoeboid endospores (Fig. 12 G-I). Video footage from other colonies showed that smaller peripheral buds underwent fission and produced dispersal amoebae.  Ultrathin sections of “A. whisleri” were embedded in an organic matrix, presumed to be remnants of slimy extracellular material that contributed to the slippery appearance of the older cultures. The outer plasma membrane was frequently torn during processing resulting in an artifact that looked like a homogenous gap between the cytoplasm and the plasmalemma (e.g. Fig. 5.13). Several cells showed distinct tearing, convincing us that this gap was indeed an artifact. Spherical, plasmodial and amoeboid stages were present in our sections but were not undergoing reproduction (Fig. 5.13).  Like “P. gemmata”, vegetative cells of “A. whisleri” were also multinucleate and bound by a layered cell wall. Cytoplasmic content included glycogen rosettes, 142  mitochondria with flat cristae, SER and vesicles with variable contents (Figs. 5.13-5.14). SER was often found along the cell periphery (Fig. 5.13E) and around the nuclei (Fig. 5.14 A-B). Internal membranes were present in many cells (Figs. 5.13 E-I, 5.14F). Some vesicles had further internalized and sometimes concentric rings of membranes. Their contents ranged from electron lucent (Figs. 5.13H, 5.14 D-E) to electron opaque (Figs. 5.13 E-F, 5.14F). Nuclei had a prominent nucleolus and often had a spindle pole body with extranuclear microtubules (Fig. 5.14C). The thickness and layering of the cell wall was variable between and within different cell states. Sections of what were most likely spherical cells had layered cell walls ranging from 500 nm to two µm (Fig. 5.13 A-C). Plasmodial walls ranged from thick and multilayered in the hypha-like cell shown in Figure 5.13F to a single layer with embedded membrane bound tubules (MBTs) (Fig. 5.13 H-I). Figure 5.13D was most likely a spherical cell that had initiated budding plasmodial growth, similar to Figure 5.12.  The cell wall was usually composed of two distinct layers although sometimes each layer was present more than once. The inner layer had a moderately electron lucent, homogenous matrix embedded with MBTs and electron lucent channels (Figs. 5.15-5.16). This layer was present for all cell types except amoebae, which lacked walls altogether (Fig. 5.13G). Many cells also had an outer layer composed of electron opaque fibres (Fig. 5.13 A-D, F), similar to the outer layer of “P. gemmata”. Occasionally a calyx or cell coat was present (Fig. 5.15A). Like “P. gemmata”, the fibres of the outer layer appeared to be arranged parallel to the cell circumference in cross section (Fig. 5.16 D, G), and formed a loosely interwoven network in tangential section. Vesicles, coated on the outside with what resembled fibres, were visible in the wall layers (Fig. 5.15 E-F). Within the inner layer, the MBTs and channels were often distributed in patches (Fig. 5.15 A-C). The inner and outer layers were usually distinct (Fig. 5.15 A, G), but sometimes patches of inner wall-like material with MBTs were sandwiched between layers of fibrous outer wall material (Fig. 5.15D). The contents of the MBTs varied from electron lucent to electron dense (Fig. 5.16). Figure 5.16 (B-D) shows examples where the membrane from the MBTs connected to the cell‘s plasma membrane. MBTs were twisted, convoluted, 143  sometimes branched and ranged between 10 – 100 nm in diameter. It was not clear how long the MBTs were, nor whether they remained contiguous for their entire length.  Diagnosis Genus gen. nov. (ICZN): Pirum Smooth, walled, multinucleate, cells with prominent, large, often convoluted vacuole(s) during vegetative growth. Cells round in cross section and ovoid to pear shaped in longitudinal view. Cultured colonies white, raised and rough. Bilayered cell wall with an electron opaque outer layer and a variably thick electron lucent inner layer. Outer layer fibrous, with fibres parallel to plasma membrane in cross section and forming an interwoven network in tangential section. Membrane bound tubules (MBTs) extending from and the cell penetrating the thickened regions of the inner wall layer. Mitochondria with flat cristae, nuclei with a prominent nucleolus and spindle pole bodies with extranuclear microtubules. Nutrition osmotrophic with no phagotrophic stages. Asexual reproduction and dispersal primarily through weakly amoeboid wall-less endospores released in spurts through tears in parent cell wall. Less frequently endospores formed a chain. Sporogenesis via serial cleavage of cytoplasm after reaching a multinucleate state. No multi-cell aggregations, hyphae, flagellated or motile amoeboid stages observed. Distinguished from Abeoforma by lack of amoebae, plasmodia and differences in structure of the inner cell wall layer and the arrangement of MBTs in patches associated with thickened regions of the inner wall. Generic epithet describes the pear-like shape in longitudinal section derived from the Latin pirum = pear.  Type species:  P. gemmata P. gemmata sp. nov.: Cells 70-100 m long by 45-65 m wide at maturity. At maturity releases hundreds of weakly amoeboid but non-motile endospores 4-6 m in diameter. Isolated on a single occasion from peanut worm,  P. agassizii. Specific epithet refers to the large system of vacuoles that refract light reminiscent of facets on a gemstone, derived from the Latin gemma = gem. 144  Holotype: P. gemmata live strain bPW deposited at the ATCC under accession number PRA-279. Type locality: Isolated from material collected from the digestive tract of a peanut worm, P. agassizii. collected from the intertidal zone of Scott‘s Bay, Bamfield, British Columbia, Canada, November 2004.  Genus gen. nov. (ICZN): Abeoforma Multinucleate, walled, typically spherical cells with one to several vacuoles occupying a majority of the cell volume. Cell wall, when present, of one or two layers. Single and/or innermost layer composed of a moderately electron lucent matrix embedded with MBTs and patches of electron lucent channels.  Second layer, if present, external and composed of a network of electron opaque fibres. Mitochondria with flat cristae, nuclei with prominent nucleolus and an associated spindle pole body with extranuclear microtubules.Complex and undetermined lifecycle involving several stages including plasmodia and amoebae. Multinucleate plasmodia irregularly shaped, dynamic, variable in size, typically with a wall or cell coat. Amoebae uninucleate with no cell wall and numerous pseudopodia. Cultured colonies white, slimy with irregular edges. Osmotrophic nutrition with no evidence of phagocytosis during amoeboid phases. Reproduction variable: 1) Release of amoeboid endospores by complete and partial conversion of parent cell, common.  2) Frequent and repeated binary fission of amoebae and endospores, common. 3) Budding of plasmodial stages, common. 4) Release of numerous non-motile endospores through tear in parent cell wall, occurs but less common. Distinguished from Pirum by arrangement of MBTs in a contiguous inner cell wall layer rather than in patches, also by presence of amoebae, plasmodia and slimy colonies. Generic epithet refers to the dynamic range of shapes exhibited as the organism transitions between amoeboid and walled stages derived from the Latin, abeo = change and forma = shape.  Type species: Abeoforma whisleri. A. whisleri sp. nov.: Spherical cells up to 50 µm in diameter. Plasmodial stages frequently exceeded 120 µm in length, often leaving a transparent wall or cell coat after changing shape. Plasmodia occasionally forming hypha-like forms with diameters as 145  narrow as 5 µm. Dispersal via amoebae with numerous narrow pseudopodia ranging from approximately 5 µm wide up to 25 µm long. Some amoebae filose with short, narrow pseudopodia originating from most of the surface. Other amoebae branched with cytoplasm concentrated on branch termini with each terminus separated by narrow (1.5 µm) cytoplasmic connection. Specific epithet was selected in honour of Dr. Howard Whisler in recognition of his outstanding contributions to ichthyosporean biology through his culture and experimental studies of Amoebidium. Holotype:  A. whisleri live strain MB14 deposited at the ATCC under accession number PRA-280. Type locality: Isolated from material collected from the stomach, intestine and digestive gland of a mussel, Mytilus sp., collected from the intertidal zone of Brickyard beach, Gabriola Island, British Columbia, Canada. April 2007.  Discussion We described our new species in two different genera, Pirum and Abeoforma, because they were phylogenetically and morphologically distinct and they shared few morphological characters to distinguish them from other ichthyophonids. On an ultrastructural level, their synapomorphic similarities were their fibrous outer cell walls and their systems of smooth tubules near the spindle pole bodies. Otherwise, the morphology and life cycles of P. gemmata and A. whisleri were different (Table 5.1). P. gemmata was relatively simple with no evident motile or plasmodial stages. A. whisleri had several forms of spore release and a complex series of alternative vegetative forms. Characters such as the arrangement of the tubular extensions of the plasma membrane into the inner cell wall layer (MTBs), and the structure of the inner wall layers of Pirium were no more similar to those of Abeoforma than to other genera such as Creolimax. The SSU-rDNA similarity between Pirum and Abeoforma was 95 %, equivalent to the difference between two of the best-studied genera within the ichthyosporea, Amoebidium and Ichthyophonus. As a result of the divergence between the species and the lack of synapomorphies uniting them, describing independent genera for Pirum and Abeoforma was less awkward than erecting a genus broad enough for both of them. 146  We isolated P. gemmata and A. whisleri one time each from marine invertebrate digestive tracts and the only closely related sequences were from uncultured environmental clones. A. whisleri was the sister taxon to a clade including P. gemmata plus two environmental DNA sequences. At the next node down in the tree, the sister clade to the group with our new species consisted only of clones from uncultured environmental samples. These related environmental sequences, each represented by only by one to two cloned DNA sequences, were collected during seven independent surveys of SSU-rDNA diversity within water and sediment samples (Savin et al. 2004; Takishita et al. 2005, 2007a, 2007b). All of these environmental sequences were from marine or saline environments except for ‗uncultured clone Amoebidium’ which came from the sulphurous Lake Vilar, Spain.  All ichthyosporean species described to date were isolated from animal hosts and most are known to be symbionts (Mendoza et al. 2002). Most ichthyophonida colonize or enter their host through the digestive tract (Jones and Dawe 2002; Mendoza et al. 2002; Cafaro 2005). Suggesting that it needed a nutrient-rich environment, P. gemmata neither grew nor made resting spores in unsupplemented seawater. To test whether P. gemmata might also colonize areas of the body cavity other than the gut, we had attempted, without success, to culture ichthyophonids from coelomic cavity fluid from its host at the same time as we isolated P. gemmata successfully from the guts. Indicating more generally that marine ichthyophonids did not colonize internal body tissues, in our studies, we isolated177 ichthyophonid colonies from 13 of 42 hosts but never isolated any ichthyophonids from 26 of 42 host coelomic or mantle cavity fluid samples. (Marshall et al. 2008, Chapter 4). We cannot rule out that spores of A. whisleri and P. gemmata were ingested accidentally or as food. Likewise we cannot know whether the environmental sequences were from free-living organisms or were resting stages of host obligate species. On the balance, we consider the digestive tract to be the most likely habitat for both new isolates. However, because these isolates were only found once, we can not be certain that these were their definitive hosts.  147  Comparative morphology  In addition to their many unique features (Table 5.1), both P. gemmata and A. whisleri shared characteristics with other ichthyosporeans and opisthokonts. Typical of opisthokonts, their ultrastructure showed glycogen rosettes, nuclei with prominent nucleoli, and mitochondria with plate-like cristae. Like all other described ichthyosporeans, both P. gemmata and A. whisleri had cell walls and osmotrophic nutrition (Mendoza et al. 2002).  Generally, dermocystida (e.g. Ashworth 1923; Pekkarinen et al. 2003) and ichthyophonida (e.g. Whisler 1968; Cafaro 2005; Marshall et al. 2008) become multinucleate before dividing. Near simultaneous nuclear division as observed in P. gemmata is also a transitional phase leading to multinucleate stages in other ichthyophonida and dermocystida (Ashworth 1923, Marshall and Berbee 2009, submitted). Division in plasmodia, although reported less commonly, has been described for Dermocystidium percae (Pekkarinen et al. 2003) and Ichthyophonus hoferi (Franco- Sierra and Alvarez-Pellitero 1999). The production of amoebae (seen in A. whisleri), and the spindle pole bodies and MBTs seen in both P. gemmata and A. whisleri have also been described in other ichthyophonida (Whisler 1965; Okamoto et al. 1985; Vogt and Rug 1995; Marshall et al. 2008). Amoebae are known in Psorospermium, Ichthyophonus, Amoebidium, Paramoebidium and Creolimax fragrantissima (Whisler 1965; Dang and Lichtwardt 1979; Okamoto et al. 1985; Vogt and Rug 1995; Marshall et al. 2008).  Spindle pole bodies may be a general feature of the ichthyophonids although they were first described in the group only recently, in C. fragrantissima (Marshall et al. 2008). The replacement of the centriole with a spindle pole body was probably congruent with the loss of a flagellated stage and both flagella and centrioles may be absent from the ichthyophonida. Dermocystida in contrast to the ichthyosporea lack known amoeboid stages and they produce a flagellated stage if motile. Centrioles have been described in Dermocystidium percae (Pekkarinen et al. 2003) and the non-motile Rhinosporidium (dermocystida) is reported to have an internal centrosome (Ashworth 1923). Spindle pole bodies have not been reported in any of the dermocystida. The MBTs may provide ultrastructural characters distinguishing the different ichthyophonid genera, although fixation by freeze substitution may be required to 148  preserve their distinctive forms.  In freeze-substituted C. fragrantissima cells, clearly defined MBTs extended into an undulating inner wall layer and were organized in loosely defined patches and (Marshall et al. 2008).  In P. gemmata the MBTs were confined to discrete thickened patches in the cell circumference while in A. whisleri MBTs formed an extensive, thick and widespread layer in the inner wall. Electron micrographs of chemically-fixed cells of other ichthyophonida show vesicular structures which may also be MBTs, which are evident between the plasmalemma and cell wall (Franco-Sierra and Alvarez-Pellitero 1999; Moss 1999). The function of the MBTs is unknown although they would effectively increase the surface area of the cell, possibly increasing nutrient assimilation efficiency. Given that A. whisleri assumes such a diversity of shapes and is often motile, the MBTs may function to mediate changes in cell wall composition during movement or phase shift. The presence of the inner wall layer and MBTs in irregularly shaped and seemingly plasmodial cells suggests that the inner wall layer may be flexible and allow for movement. MBTs have never been reported from dermocystida.  The serial equilateral cleavage of cytoplasm leading to the formation of endospores in P. gemmata has not described in other ichthyosporeans. Several other types of maturation and spore production have however been reported, including serial binary fission in Anurofeca richardsi (Beebee 1991; Vogt and Rug 1999) and basipetal maturation from a persistent mother cell in the elongate eccrinid taxa (Cafaro 2000). Although distantly related, spore production in the dermocystid R. seeberi shared features with P. gemmata including a decrease in nuclear size with maturity (Ashworth 1923). Release of large numbers of endospores in pulses was observed in R. seeberi as well as in P. gemmata and occasionally in A. whisleri. In R. seeberi, however, endospores were released in small spurts through a pore in the cell wall as new endospores mature in a region on the opposite side of the cell (Mendoza et al. 1999). In P. gemmata and A. whisleri, no such pore was evident and the endospores probably exited through tears in the wall. The chains of spores occasionally seen in P. gemmata were unusual and have not been reported in other ichthyosporea. With their numerous variably shaped pseudopodia the amoebae of A. whisleri resembled the filose amoebae produced by P. haeckeli (Vogt and Rug 1995). Among the ichthyosporea, the highly branched amoebae that sometimes attach to the substrate and 149  the larger filose amoebae with short filopodia of A. whisleri are so far unique. Their function within the lifecycle of A. whisleri is unknown. The amoebae of A. parasiticum, C. fragrantissima, I. hoferi and some P. haeckeli differ in having a single, broad, anterior pseudopod (Dang and Lichtwardt 1979; Okamoto et al. 1985; Henttonen et al. 1997; Marshall et al. 2008). Growth and asexual reproduction may vary within single ichthyosporean species. Reproduction in P. gemmata was not particularly complex. However, A. whisleri, like Ichthyophonus hoferi produced spherical, plasmodial, hypha-like and amoeboid forms and the  asexual life cycles of both are complex and not completely resolved (e.g. Okamoto et al. 1985). A. whisleri and I. hoferi produce strikingly similar empty wall remnants.  In I. hoferi these empty hyphae were produced by the continuous evacuation of the cytoplasm as it advanced as a bulbous tip, eventually rounding up into a spherical multinucleate terminal body (Spanggaard et al. 1994; Spanggaard et al. 1995). In A. whisleri the empty walls remained behind after irregularly shaped cells changed shape and position. As in I. hoferi, the remaining cytoplasm did eventually ball up into a spherical, walled cell. Despite the similarities in the deposition of empty wall material, the cell wall of A. whisleri was not similar to I. hoferi or any other ichthyosporean. The thick inner wall layer with long MBTs is, so far, unique to A. whisleri. We have found two species corresponding to a previously undescribed lineage of clone sequences but the search for reciprocal matches between phylotypes and captive organisms is far from over. As Fig. 5.1 shows, several opisthokonts represented by DNA sequences were only distantly related to described species. Some of the sequences probably represent as yet undiscovered lineages of unicellular opisthokonts. Others may represent described but unsequenced taxa such as the Alphelidea (Gromov 2000), which were classified as the opisthokonts (Adl et al. 2005). The environmental sequences did not appear to be scattered uniformly through the phylogeny. Environmental PCR based surveys often produce new sequences from the choanoflagellates, ichthyophonids (e.g. Stoeck and Epstein 2003; Takishita et al. 2005). In contrast, unicellular lineages including the symbiotic dermocystids, eccrinales, and Capsaspora had no closely related environmental sequences. Environmental sequences may be biased against obligate symbionts and towards free-living taxa.  However, no 150  environmental clone sequences were closely related to members of the free-living nuclearids, Corallochytrium, or Ministeria either.  It is unclear whether these patterns result from PCR bias, from differences in abundance of different organisms in different habitats, or from biases in isolation of living organisms in culture-based surveys.  Conclusions P. gemmata and A. whisleri represent an ichthyosporean clade that was previously known only from DNA sequences from seven unrelated environmental surveys. Each species has unique traits, such as its MBTs organization, and for A. whisleri, the trail of extracellular material remaining after plasmodial movement.  Like other ichthyosporeans, both new species are walled osmotrophs, with a multinucleate, coenocytic stage preceding spore production, and with spindle pole bodies rather than centrioles . Many of these characters shared between P. gemmata and A. whisleri were likely also ancestral characters for the ichthyosporeans. Discovery and characterization of the biological diversity of living species such as these is completing the picture of morphological features that unite the Ichthyosporea, a group that shared a common ancestor with the progenitors of choanoflagellates and animals. 151  Table 5.1. Morphological variation between P. gemmata and A. whisleri. Character P. gemmata A. whisleri Forms of reproduction Pulsed endospore release Pulsed endospore release Binary fission Budding Dispersal amoebae Shape Pear to spherical Spherical Plasmodial (large and small) Various amoeboid shapes Inner cell wall layer Completely lucent Forms intermittent thickenings causing the cytoplasm to appear indented MBTs penetrate only at thickenings Moderately lucent More consistent thickness within cells Impregnated with long MBTs and lucent chanels Inner layer may be the only layer present Extracellular secretions Absent Present   152   Uncultured Opisthokont II 0.1 Uncultured clone NAMAKO-33 (Japan 1) Uncultured clone s416 (Atlantic Canada) Uncultured clone s712 (Atlantic Canada) Uncultured clone Qinghai-52 (China) Uncultured clone s25 (Atlantic Canada) Uncultured clone Amoebidium (Spain) Uncultured clone TAGIRI-27 (Japan 2) Uncultured clone s16 (Atlantic Canada) Uncultured clone NAMAKO-34 (Japan 1) Pirum gemmata Abeoforma whisleri Astreptonema sp. Palavascia patagonica Enteropogon sexuale Enterobryus oxidi Eccrinidus flexilis Amoebidium parasiticum Ichthyophonus hoferi Psorospermium haeckeli Uncultured clone ikaite-53 Uncultured clone TAGIRI-25 Sphaeroforma arctica Uncultured clone D3P06H09 Uncultured clone D4P08E02 Creolimax fragrantissima Uncultured clone BB01 Uncultured clone Anurofeca LAH-2003 Anurofeca richardsi Uncultured clone LKM51 Capsaspora owczarzaki Savillea micropora Taphrina johansonii Pneumocystis wakefieldiae Agaricus bisporus Uncultured clone WIM-48 Nuclearia simplex A Nuclearia moebius Nuclearia delicatula Nuclearia thermophila Nuclearia simplex B Uncultured clone RT5iin16 Uncultured clone RT5iin14 Apusomonadidae  sp. Planomonas micra 79/63/100 90/74/100 60/61/- 85/91/99 62/-/83 91/-/100 87/-/100 Uncultured clone BS-DGGE (Black Sea) Uncultured clone DSGM-65 (Japan 3) Uncultured clone TAGIRI-26 (Japan 2) Uncultured clone ms313 (Atlantic Canada) 98/87/100 73/69/91 92/89/100 81/71/100 -/-/69 83/67/98 92/72/100 66/64/67 92/-/67 62/ 66/ 85 95/86/100 Amphibiothecum penneri Dermocystidium sp. Amphibiocystidium ranae Rhinosporidium seeberi Sphaerothecum destruens -/-/67 68/-/99 Salpingoeca amphoridium Monosiga ovata Salpingoeca pyxidium Diaphanoeca grandis -/-/75 76/-/100 -/-/98 -/-/95 Uncultured clone M1_18C07 Corallochytrium limacisporum -/-/65 -/-/87 Pisaster Corella Aurelia aurita Ministeria vibrans inflata ochraceus96/87/100 86/69/100 96/84/100 97/82/100 -/-/86 -/ -/ 100 Uncultured clone DSGM-63 Uncultured clone TAGIRI-23 Uncultured clone RT5iin3 Rozella allomycis73/64/100 -/-/100 93/77/100 -/-/86 -/-/96 Uncultured clone CCW-24 Smittium commune Uncultured clone TAGIRI-24 Physoderma maydis 91/75/100 -/-/98-/ -/ 99 Rhizophidium elyensis Catenomyces sp. Spizellomycete sp. Mortierella wolfi Rhizophydium sphaerotheca-/-/100 -/-/77 75/-/98 C H I H T Y O P O N I A H D Dermocystida Uncultured Opisthokont I Choano- flagellates Animals Fungi Nuclearia SSU-rDNA C H I H T Y O S O R E P A  Figure 5.1. Abeoforma whisleri and Pirum gemmata, the two new species described in this paper, are the only living organisms known from a clade that consists otherwise of environmental clone sequences. 153  This maximum likelihood RAxML tree compares the phylogenetic relationships of A. whisleri and P. gemmata to other GenBank SSU-rDNA sequences from opisthokonts and includes identified species as well as uncultured environmental clones.  Numbers at nodes represent 500 RaxML or parsimony bootstrap replications or Bayesian posterior probabilities (RAxML/MP/Ba) over 60 %. Thickened branches show nodes that have 90 % or greater support by all three methods. 154   Figure 5.2. Vegetative cells of Pirum gemmata. A-C. P. gemmata cells were round in cross section and pear shaped in longitudinal view with large, often convoluted, central vacuoles (CV). Arrowhead in B points to a narrow bridge of cytoplasm between two large vacuoles. D. Mature cells with developing endospores had granular cytoplasm. E.  Endospores (e) that were not released grew within and protruded through the opening in the parent cell wall. All scale bars equal 20 microns. 155   Figure 5.3. Timelapse series showing three Pirum gemmata cells (A-C) undergoing sporogenesis in slide chambers. Pictures were taken at eight minute intervals and minutes since starting frame are shown on the upper right. Immediately before sporogenesis the cells had a single large central vacuole (CV), then furrows (F) appeared as the cytoplasm cleaved (e.g. cell B at 8 minutes). Cell A shows how each new endospore precursor was bound by a plasma membrane (PM). As cleavage continued, the developing endospore precursors became smaller. The outer wall of the parent cell remained intact throughout sporogenesis (arrowheads). All scale bars equal 20 microns. 156   Figure 5.4. Endospore release and morphology in Pirum gemmata. A. Endospores (e) were usually released in pulses through oneor more openings in the parent cell wall. With each pulse endospores moved towards the opening from deep within the parental cell (arrow). B. Occasionally the endospores were joined to form a chain. C-F. Endospores were weakly amoeboid and some had pseudopod-like extensions (p). A-B. Scale bar equals 10 microns. C-F. Scale bar equals 5 microns. 157   Figure 5.5. DAPI stained whole mounts of Pirum gemmata cells showing variation in nuclear size. A. Three DAPI stained cells showing a range of nuclear sizes from smallest (1.8 – 2.0 m diameter) on the left to the largest (4.8 m) in the middle cell. The nucleolus was visible in some nuclei as the dark unstained centre. Images B-K show light micrographs and corresponding fluorescent images of DAPI stained nuclei. B-C. Comparison of two uninucleate endospores and the corresponding fluorescent image. The cell on the top was young and still had a small nucleus (2.0 m); the bottom cell had undergone some 158  vegetative growth and had a vacuole (CV) and a larger nucleus (3.5 x 4.0 m). D-K shows cells at equal scale. Nuclear size in microns is shown in the top right. D-E. Vegetative, vacuolate cells at eight and 16 nucleus stages. F-G. Mature cell that was undergoing sporogenesis with barely visible endospores. H-I. Mature cell that was releasing endospores (e). Nuclei were at their smallest size between 1.8 and 2.0 m in diameter. J-K. These endospores were growing vegetatively within the parent cell wall and some nuclei were increasing in size, for example the nucleus in endospore (e) was 2.5 m. All scale bars are equivalent to 10 microns. 159   Figure 5.6. Transmission electron micrographs showing sections through vegetative and reproductive cells of Pirum gemmata. A. Section through a vegetative cell with a cell wall (CW), multiple nuclei (N), a large central vacuole (CV) and numerous smaller vacuoles (V). Some of the smaller vesicles may have been fusing with or budding from the central vacuole (arrowhead). B-C. Endospores within mature cells. Parent cells had a cell wall and the endospores were bound by a plasma membrane (PM) but no cell wall. In addition to membrane bound vacuoles there were also lipid drops (*) and concentric rings of plasma membrane (R) within the cytoplasm. A section of internal plasma membrane appeared to be joining the 160  outer plasma membrane (jm1). D. Detail of a developing or unreleased endospore from Figure 6C. A section of plasma membrane (jm2) was in close proximity to the outer plasma membrane.  E. Cleaved cell with wall-less endospores and an opening in the parent cell wall (arrow O). An invagination of the plasma membrane was present (jm3) and could have been either been a cleavage furrow in an endospore precursor or new section plasma membrane being added to a new daughter cell that had begun vegetative growth.  A. Scale bar equals 10 microns. B-E. Scale bar equals 2 microns. 161   Figure 5.7. Spindle pole bodies and networks of smooth tubules were associated with the nucleus in Pirum gemmata. A. Spindle pole bodies (SPB) were often at one pole of the nucleus and were associated with a network of smooth tubules resembling smooth endoplasmic reticulum (SER) or part of a Golgi body. B. Detail of spindle pole body with extranuclear microtubules (Mt) and a mitochondrion (poorly preserved) (M). C. Tangential section through the spindle pole body. Nuclear pores (Np) are visible. Scale bar equals 500 nm. CW, cell wall; N, nucleus; V, vesicle; G, glycogen. 162   Figure 5.8. Cell walls of Pirum gemmata showing membrane bound tubules arranged in discrete patches between the plasma membrane and the inner surface of the cell wall. The MBTs (T) may be embedded in the inner cell wall and the electron lucent material that surrounds them may be  an internal protuberance of the  inner cell wall layer. A. Cell walls (CW) of two adjacent cells (boundary at arrowheads) showing patches (P) of thickened sections of the electron lucent inner wall layer which was penetrated by MBTs (T). B. Tangential section near cell surface. The cell wall was composed of loosely networked electron opaque fibres that often extended into the intracellular space (f). Plasma membrane (PM) was visible on the MBTs and appeared contiguous with the 163  plasma membrane of the cell C. Cross section through patches of MBTs showing a coated vesicle (Co) nearby. D. System of tubules or smooth endoplasmic reticulum (SER) was visible in the cytoplasm near this patch of MBTs. E. Section that shows continuity with external plasma membrane and MBTs. F. Cross section through cell wall showing a layered arrangement of fibrils. G. Tangential section through cell wall showing an interwoven network of cell wall fibrils. All scale bars equal 500 nm. V, vesicle; CV, central vacuole; G, glycogen. 164   Figure 5.9. Examples of various cell shapes and modes of reproduction of Abeoforma whisleri. A. A commonly observed spherical form with prominent vacuoles (V) that occupied a large proportion of the cell volume. B. Example of an irregularly shaped cell that may have been in transition between a plasmodial or spherical stage. C. Large plasmodial cell. D. Reproduction via release of large numbers of endospores through an opening in the parent cell wall. E-G. Reproduction via production of amoeboid endospores from tetrads and octets. The endospores from the mature octet in Figure 9G were undergoing binary fission (bf). H. Mature cell with at least 64 endospores. I-K. Smaller cells undergoing plasmodial type growth. L. The same plasmodium as in Figure 9K after one day now showing hypha-like extensions. This plasmodium separated or budded into at least two physically seperate. Scale bars equal 20 microns for all images. 165   Figure 5.10. Various shapes and forms of amoeboid stages of Abeoforma whisleri. A. ‗Dispersal‘ amoebae (a) crawled away from parent cell (same cell as shown in Figure 9J one day later). Material from parent cell or plasmodium was visible (w). B. Dispersal amoebae had multiple pseudopodia (p) of variable number, size and shape. C. Rare, filose amoeba had short narrow pseudopodia. D-F. Rare, branched amoeba. Most of the cytoplasm was at the termini which were separated by long narrow connections of cytoplasm (arrowheads). Figures D-E show the amoeba attached to a piece of debris by one of the terminal ends of the cytoplasm (H). Pseudopodia on the ends were visibly moving and the length of each connecting branch was also dynamic. F. Same amoeba as in figures D-E one day later. This time it was not attached and only had two terminal ends. One of the terminal ends of the branched amoebae was always relatively larger and ovoid in form (e). All scale bars equal 10 microns. 166   Figure 5.11. Time-lapse series showing a plasmodial form of Abeoforma whisleri transitioning into a sphere. As the cytoplasm withdrew the extracellular material or the remnant cell wall was left (arrowheads) (especially D-F) and the sphere increased in volume. Images were taken at 4 minutes intervals. Scale bar equals 20 microns for all images. 167   Figure 5.12. Time series showing the formation of a colony of Abeoforma whisleri via plasmodial growth, budding and endospore release. Plasmodial growth (A-F), budding (G-H) and endospore (e) release (H) culminating in a maze-like plasmodial network (I). Arrows in A and I mark debris on media as a reference point; (pb) shows peripheral buds formed from the plasmodium. Scale bar equals 100 microns. 168   Figure 5.13. Examples of various cell shapes of Abeoforma whisleri. Ultrathin sections of whole cells of A. whisleri showing walled and wall-less stages. An artifact from processing often produced a gap between the cytoplasm and the outer plasma membrane of the cell; examples of these gaps are marked with an ‗xx‘. Cells were typically embedded in a matrix of extracellular material (em). Arrowheads designate the external boundaries of each cell. Electron lucent vesicles (V) may have lost their contents during processing. A-C. Cells with multiple nuclei (N), multilayered cell walls (CW), glycogen (G) and several types of vesicles (V), possibly corresponding to spherical phase in light micrograph Fig. 5.9A. D. Cross section of a growing colony or cell cluster, possibly corresponding to plasmodial stage in Fig. 5.12. A thick cell wall surrounds the 169  densely vacuolated inner-most cell or compartment and a thin outer cell wall encircled the entire colony or cell. E. Multinucleate cell with a network of internal membranes (im). The cell wall detached during processing and is not in view. Cytoplasm at the cell periphery was permeated with membranous tubules or smooth endoplasmic reticulum (SER). Cellular contents included glycogen patches, mitochondria (M), vesicles (V), pigmented vesicles or inclusions (PV) and membranous vesicles with opaque contents (omv). F. Tip of a hypha-like cell with multiple nuclei, vesicles and a cell wall. G. Wall- less amoebae nested between wall-less or minimally walled neighbouring amoebae. Neighbouring cells are marked by ‗NC‘.  The nucleus had clear nuclear pores (np) and occupied much of the cell volume. Membrane bound vesicles with varying content and glycogen rosettes were present in the cytoplasm.  H-I. Probable plasmodial cells. These cells were irregular in shape and had a single layer of cell wall material composed of membrane bound tubules (T) embedded in an electron lucent matrix. Cytoplasm had mitochondria, various vesicles including membranous vesicles (mv), some with electron lucent contents (lmv), and a network of internal membranes. Scale bar equals 2 microns for all images. 170   Figure 5.14. Detail of nuclei and other cytoplasmic contents of Abeoforma whisleri. A-B Nuclei (N) had prominent nucleoli (Nu), nuclear pores (np) and associated patches of tubules resembling smooth endoplasmic reticulum (SER).  Mitochondria (M) had plate-like cristae. C. Spindle pole body (SPB) with extranuclear microtubules (Mt) positioned next to the nucleus. D-F. Detail of several types of membranous vesicles (mv) containing internal membrane bound compartments, sometimes in several concentric layers (*). Electron lucent vesicles (lmv) and electron opaque membrane bound lipid bodies (omv) were present. E. Detail of electron lucent membranous vesicles showing internal membrane bound compartments. F. Detail of cell from Figure 13E showing membrane bound internal compartments of the opaque membranous vesicles. Scale bar equals 500 nm for all images. Fixation artifact, xx; vesicle, V; glycogen, G; plasma membrane pm; internal membrane, im. 171   Figure 5.15. Detail of the multilayered walls of Abeoforma whisleri. A. Edge of a cell showing the transition between a thick inner layer with MBTs (TL), fibrous layer (FL) and an outer calyx or extracellular matrix (Ca). The cell edge is defined by the arrowheads. The tubule containing layer consisted of an amorphous moderately electron lucent matrix (Ma) impregnated with tubular structures (T) and electron lucent channels (Ch). Extracellular membrane bound vesicles were visible in the tubular layer (Ev). Electron dense globules (*) were visible near and along the cell 172  periphery. B. Higher magnification clearly showing the fibres in the outer layer and the arrangement of tubules and channels within the matrix of the inner layer. C. Section through a thick cell wall with very distinct layering and long tubules. D. Tubules and channels were present throughout the matrix of the inner layer and smooth endoplasmic reticulum (SER) was prevalent around the edge of the cell. E. This cell had some evidence of tubules near the plasma membrane but the wall was mostly composed of the outer fibrous layer.  F-G. Tangential sections through the cell wall layers showing fibres (f), membrane bound vesicles coated in fibrous material (co) and MBTs with clear plasma membranes. Scale bar equals 2 microns for A-B, D-E and 500 nm for C. F-G. G, glycogen; em, extracellular material; N, nucleus; omv, electron opaque membranous vesicle; M, mitochondria. 173   Figure 5.16. High magnification views of the inner wall layer with membrane bound tubules (MBTs) of Abeoforma whisleri. A. Detail of cell wall from Figure 15B showing membrane bound tubules (T) within the inner layer and the cross-linked network of fibres of the outer layer (FL). Figures B-D are of cells without an outer wall layer. Arrowheads in Figure B define the cell boundary. Connectivity of the tubules with the cell‘s plasma membrane is shown by ‗ext‘. The plasma membrane in Figure D detached from the cytoplasm leaving an artificial gap (xx). Longitudinal and cross sections of the tubules show that they varied in width, twist and branch (b). Scale bar equals 500 nm for A-B and 100nm for C-D. Glycogen, G; Ma, matrix; plasma membrane, pm; smooth endoplasmic reticulum, SER; mitochondria, M. 174  References Adl S Simpson AG Farmer MA Anderson RA Anderson OR Barta JR Bowser SS Brugerolle G Fensome RA Fredericq S James TY Karpov SA Kugrens P Krug J Lane CE Lewis LA Lodge J Lynn DH Mann DG McCourt RM Mendoza L Moestrup Ø Mozley-Standridge SE Nerad TA Shearer CA Smirnov AV Spiegel FW Taylor M (2005) The new higher level classification of eukaryotes with emphasis on the taxonomy of protists. J Eukaryot Microbiol 52:399-451 Altschul SF Madden TL Schaffer AA Zhang J Zhang Z Miller W Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25:3389-3402 Amaral Zettler LA Messerli MA Laatsch AD Smith PJS Sogin ML (2003) From Genes to Genomes: Beyond Biodiversity in Spain's Rio Tinto. Biol Bull 204:205- 209 Ashworth JH (1923) On Rhinosporidium seeberi (Wernike 1903), with special reference to its sporulation and affinities. Trans R Soc Edinb 53:301-342 Beebee TJC (1991) Purification of an agent causing growth inhibition in anuran larvae and its identification as a unicellular unpigmented alga. C J Zool 69:2146-2153 Berney C Fahrni J Pawlowski J (2004) How many novel eukaryotic ‗kingdoms‘? Pitfalls and limitations of environmental DNA surveys. BMC Biol 2:13 Brown H (1942) A micro culture slide for fungi. J Bacteriol 43:16 Cafaro MJ (2000) Gut fungi of isopods: the genus Palavascia. Mycologia 92:361-369 Cafaro MJ (2005) Eccrinales (Trichomycetes) are not fungi, but a clade of protists at the early divergence of animals and fungi. Mol Phyl Evol 35:21-34 Cavalier-Smith T (1998) Neomonada and the origin of animal and fungi. Pages 375-407 in G. Coombs, K. Vickerman, M. Sleigh, and A. Warren, editors. Evolutionary relationships among protozoa. Chapman and Hall, London Dang S-N Lichtwardt RW (1979) Fine structure of Paramoebidium (Trichomycetes) and a new species with virus-like particles. Am J Bot 66:1093-1104 Dawson SC Pace NR (2002) Novel kingdom-level eukaryotic diversity in anoxic environments. Proc Natl Acad Sci USA 99:8324-8329 175  Franco-Sierra A Alvarez-Pellitero P (1999) The morphology of Ichthyophonus sp. in their mugilid hosts (Pisces: Teleostei) and following cultivation in vitro. A light and electron microscopy study. Parasitol Res 85:562-575 Gromov BV (2000) Algal parasites of the genera Aphelidium, Amoeboaphelidium and Pseudophelidium from the Cienkovski's "Monadinea" group as representatives of a new class. Zool Zhurn 79:517-525 Henttonen P Huner JV Rata P Lindqvist OV (1997) A Comparison of the known life forms of Psorospermium spp. in freshwater crayfish (Arthropoda, decapoda) with emphasis on Astacus astacus L. (Astacidae) and Procambarus clarkii (Girard) (Cambaridae). Aquaculture 149:15-30 Huelsenbeck JP Ronquist F (2001) MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17:754-755 Jones SRM Dawe SC (2002) Ichthyophonus hoferi Plehn & Mulsow in British Columbia stocks of Pacific herring, Clupea pallasi Valenciennes, and its infectivity to chinook salmon, Oncorhynchus tshawytscha (Walbaum). J Fish Dis 25:415-421 Luo Q Krumholz LR Najar FZ Peacock AD Roe BA White DC Elshahed MS (2005) Diversity of the Microeukaryotic Community in Sulfide-Rich Zodletone Spring (Oklahoma) Appl Environ Microbiol 71:6175-6184 Marshall WL Celio G McLaughlin DJ Berbee ML (2008) Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. Protist 159:415-433 Massana R Castresana J Balague V Guillou L Romari K Groisillier A Valentin K Pedros-Alio C (2004) Phylogenetic and Ecological Analysis of Novel Marine Stramenopiles. Appl Environ Microbiol 70:3528-3534 Mendoza L Herr RA Arseculeratne SN Ajello L (1999) In vitro studies on the mechanisms ofendospore release by Rhinosporidium seeberi. Mycopathologia 148:9-15 Mendoza L Taylor JW Ajello L (2002) The class Mesomycetozoea: a heterogeneous group of microorganisms at the animal-fungal boundary. Ann Rev Microbiol 56:315-344 176  Moss ST (1999) Astreponema gammari: an eccrinid with appendaged spores. Kew Bull 54:637-650 Okamoto N Nakase K Suzuki H Nakai Y Fujii K Sano T (1985) Life history and morphology of Ichthyophonus hoferi in vitro. Fish Pathol 20:273-286 Pekkarinen M Lom J Murphy CA Ragan MA Dykova I (2003) Phylogenetic position and ultrastructure of two Dermocystidium species (Ichthyosporea) from the common perch (Perca fluviatilus). Acta Protozool 42:287-307 Posada D Crandall KA (1998) MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818 Savin MC Martin JL LeGresley M Giewat M Rooney-Varga J (2004) Plankton diversity in the Bay of Fundy as measured by morphological and molecular methods. Microb Ecol 48:51-65 Spanggaard B Gram L Okamoto N Huss HH (1994) Growth of the fish-pathogenic fungus, Ichthyophonus hoferi, as measured by conductimetry and microscopy. J Fish Dis 17:145-153 Spanggaard B Huss HH Bresciani J (1995) Morphology of Ichthyophonus hoferi assessed by light and scanning electron microscopy. J Fish Dis 18:567-577 Stamatakis A (2006) RAxML-VI-HPC: Maximum likelihood-based analyses with thousands of taxa and mixed models. Bioinformatics 22:2688-2690 Stoeck T Epstein S (2003) Novel Eukaryotic lineages inferred from small-subunit rRNA analysis of oxygen-depleted marine environments. Appl Environ Microbiol 69:2657-2663 Swofford DL (1998) PAUP*: phylogenetic analysis using parsimony (*and other  methods). Version 4. Sinauer Associates. Sunderland, Mass Takishita K Miyake H Kawato M Maruyama T (2005) Genetic diversity of microbial eukaryotes in anoxic sediment around fumaroles on a submarine caldera floor based on the small-subunit rDNA phylogeny. Extremophiles 9:185-196 Takishita K Tsuchiya M Kawato M Oguri K Kitazato H Maruyama T (2007a) Genetic diversity of microbial eukaryotes in anoxic sediment of the saline meromictic Lake Namako-ike (Japan): On the detection of anaerobic or anoxic- tolerant lineages of eukaryotes. Protist 158:51-64 177  Takishita K Yubuki N Kakizoe N Inagaki Y Maruyama T (2007) Diversity of  microbial eukaryotes in sediment at a deep-sea methane cold seep: surveys of  ribosomal DNA libraries from raw sediment samples and two enrichment  cultures. Extremophiles 11:563- 576 Thompson JD Gibson TJ Plewniak F Jeanmougin F Higgins DG (1997) The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucl Acids Res 25:4876-4882 Vogt G Rug M (1995) Microscopic anatomy and histochemistry of the crayfish parasite Psorospermium haeckeli. Dis Aquat Org 21:79-90 Vogt G Rug M (1999) Life stages and tentative life cycle of Psorospermium haeckeli, a species of the novel DRIPs clade from the animal-fungal dichotomy. J Exp Zool 283:31-42 Whisler HC (1965) Host-integrated development in Amoebidiales. J Protozool 13:183- 188 Whisler HC (1968) Developmental control of Amoebidium parasiticum. Dev Biol 17:562-570 178  Chapter 6. Concluding chapter Summary I have described six species of ichthyosporeans, five of which were new to science, using microscopy and molecular phylogenetic techniques. Living representatives of these species are all accessioned, by request, at the American Type Culture Collection (Table F.1). My cultures belong to two the four clades (defined below) within the Ichthyophonida. Two species belong to a divergent clade populated by unknown organisms represented by DNA sequences from environmental samples. The others belong to a clade that included two taxa previously isolated from marine invertebrates. Each of these formerly isolated taxa had only been found once or twice whereas my four species from British Columbia were represented by 8-126 conspecific isolates. Prior to my work, four genera had been grown in culture and of these only two species, Amoebidium parasiticum and Ichthyophonus hoferi, had been isolated more than twice. Multiple isolations allowed for preliminary ecological observations, such as host specificity, and an unprecedented opportunity for population level genetic analyses member of a unicellular opisthokont lineage. Healthy living cultures of ichthyosporeans from two distantly related clades provided new opportunities for morphological and life cycle studies. Most described ichthyosporea are host dependent and electron micrographs are often of poorly preserved, perhaps compromised, cells. I followed single cells through complete cycles of asexual reproduction for three species, and I described spore release in all six. I documented several morphological features new to the ichthyosporea, including an extracellular secretion visible as a slime trail after plasmodial movement, and membrane-bound tubular extensions of the cell (MBTs) (see below). To my knowledge these characters have not been described in other eukaryotes and may be unique to the ichthyosporea. I also found characters that were shared by other ichthyosporeans in different clades (Fig. 6.1), some of which may have been present in ancestral ichthyosporeans. Identification of defining and possibly ancestral traits of ichthyosporeans and other unicellular lineages is an essential stage for mapping the character evolution of early opisthokonts. Documentation of ichthyosporean life cycles and morphology is also an important step 179  for defining what an ichthyosporean is, and may provide insight into how they interact with their environment. Of the four species with multiple isolates, three were closely related. I generated sequence data for multiple loci from these closely related isolates. These data were the basis for a genealogical concordance phylogenetic species concept. Combined with morphological and geographical data, my analysis revealed three cryptic species living in sympatry within marine invertebrate guts. One of these three species had been found previously, suggesting that at least some ichthyosporeans may be cosmopolitan. By applying population level genetic techniques to the most common of these defined species I have provided evidence of occasional sexual reproduction, lack of host specificity and of a haploid asexual cycle. This is the first time ploidy or recombination has been revealed in any of the protistan opisthokonts. For this chapter I aim to extrapolate from the results included in this thesis and relate them to our current understanding of ichthyosporean biology. I emphasize and speculate regarding ichthyosporean and opisthokont evolution, relationships within the ichthyosporea, life cycles, and life history of ichthyosporeans. In each section I suggest further avenues of research and provide context for how they would be beneficial.  Comparative morphology and speculation on ancestral traits Relationship of ichthyosporeans to other unicellular opisthokonts The unicellular relatives of the multicellular animals and fungi are diverse in form and are likely to be as different from each other as the animals are from the fungi. A striking example of this is their variety of shapes and modes of nutrition. Forms range from stalked Ministeria, flagellated choanoflagellates, walled symbiotic ichthyosporeans, wall free living Corallochytrium, to the filose free living amoeboid nucleariids and symbiotic filose amoeboid Capsaspora. Most of these protists ingest food via phagocytosis but the walled lineages, ichthyosporea and Corallochytrium, are osmotrophic and do not appear to ingest food during their amoeboid or plasmodial phases. These differences in form and nutrition follow an interesting phylogenetic pattern with both osmotrophic and phagotrophic forms in the metazoan and in fungal lineages. 180  The ancestral opisthokont either had both types of form and nutrition with subsequent losses, or convergent gains have occurred. Within opisthokonts walled resting stages and cysts are formed by choanoflagellates and nucleariids. Outside the opisthokonts the amoebozoan slime molds also have amoeboid stages and produce walled spores for dispersal or resting stages.  Thus the switch between walled and wall-less assimilative and dispersal stages has occurred several times. The ichthyosporea, while osmotrophic, are unique amongst the opisthokonts and have both walled and amoeboid vegetative states. Abeoforma whisleri provided a clear example of variety in form and was one of several species that I described (Chapter 5) that switch between walled and plasmodial stages. The presence of microvilli or the ability to form amoeboid pseudopods are common and possibly homologous features of the unicellular optisthokonts. Shalchian- Tabrizi et al. (2008) classified the choanoflagellates, Capsaspora, Ministria, and animals as class Filasterea based on their phylogeny and their synapomorphic projecting microvilli or pseudopodia with internal microfilaments. Both Ministeria and choanoflagellates have villus-like tentacles. Like choanoflagellates, Ministeria has a basal body, structures resembling microtubules in its stalk and indistinct microfilaments in its tentacles (Cavalier-Smith and Chao 2003). Capsaspora can also produce long pseudopodia, but I was unable to find electron micrographs to substantiate the report of internal microfilaments by Shalchian-Tabrizi et al. (2008) (Owczarzak et al. 1980; Hertel et al. 2002).  Like Capsaspora, the nucleariids can also produce long, narrow branching pseudopodia and do not appear to have microfilaments (Patterson 1983, Dykova et al. 2003). Whether the cytoplasmic extensions between the termini of the branched amoebae in A. whisleri have a cytoskeleton with microfilaments or not remains to be determined. Broad pseudopodia are found in amoebozoans (Shalchian-Tabrizi et al. 2008) and in Corallochytrium (Raghu-kumar 1987) as well as several ichthyosporeans (Mendoza et al. 2002).  Overall pseudopod shape varies greatly amongst the unicellular lineages and even, as was the case for A. whisleri (Chapter 5), within a single species in the ichthyosporea.  Few attempts have been made to map morphological characters onto the phylogeny of unicellular opisthokonts, partly because of lack of information about 181  characters, partly because of the diversity of opisthokont forms, and partly because of phylogenetic uncertainty.  Molecular studies were helpful in confirming that well known morphological and physiological similarities between animals and fungi (Cavalier-Smith 1998) reflected shared ancestry (e.g.Baldauf and Palmer 1993).  However, most multigene analyses of the unicellular lineages were missing representatives from at least one of the unicellular lineages. Ruiz-Trillo‘s (2006) study had the most comprehensive sampling of the unicellular opisthokonts to date but was limited to three genes. A large data set that includes representatives from all lineages and perhaps even the dermocystid Sphaerothecum destruens (ATCC 50615) (Kerk et al. 1995),  might help resolve these relationships.  Relationships among clades within the ichthyosporea Prior to my studies the ichthyosporea encompassed a collection of taxa belonging to two clades, ichthyophonida and dermocystida (Mendoza et al. 2002) without further phylogenetic subdivision. The order Eccrinales, originally considered fungal, was transferred to the ichthyosporea in 2005 (Cafaro). Our phylogeny included all sequenced taxa within the ichthyophonida, dividing the group into four distinct clades (Marshall et al. 2008). These were: Clade I, with A. whisleri, P. gemmata and previously uncultured DNA sequences (Chapter 5); Clade II, the former fungal families Eccrinales and Amoebidiales and the fish parasite Ichthyophonus;  Clade III, Psorospermium haeckeli; and Clade IV, the spherical ichthyophonida. Characteristics of each clade are listed in Table 6.1. Within the ichthyosporea we found several characters that were shared among taxa from different clades (Fig. 6.1). Characters unique to the ichthyosporea (compared to other unicellular opisthokont lineages) and possibly ancestral to the group  include: MBTs, pores in the cell wall, an external calyx which might be homologous to a holdfast, spindle pole bodies, plasmodia, extracellular secretions, pulsed endospore release and prominent central vacuoles (Table F.2). Shared characters also found in other opisthokont lineages and likely ancestral to all animals and fungi included: an amoeboid stage, mitochondria with plate-like cristae, glycogen food storage, and prominent nucleoli. 182  The origin of the ichthyosporean lineage  predated the radiation of modern animals and fungi, therefore the  ancestor of the modern ichthyosporean was not a parasite of multicellular animals. Holdfast-like material or sticky calyxes may have been used by ancestral osmotrophic ichthyosporeans for attachment in nutritious environments. In this form they may have lived in biofilms like modern-day yeasts (Wahl 1989). However, both amoeboid and plasmodial stages may have been present in the ancestral form, and although phagocytosis has not been observed in any extant species (e.g. Marshall et al 2008; Chapter 5), it could have been present in the ichthyosporean ancestor. Studies on nutrient uptake in free-living osmotrophic Corallochytrium may provide insight into how spherical, walled osmotrophs assimilate nutrients outside hosts. Mode of motility appears to be decisively divided between ichthyophonida and dermocystida. Flagellated forms are only found amongst dermocystida, likewise centrioles have only been observed in the dermocystida (Pekkarinen et al. 2003), even in the non-motile Rhinosporidium (Ashworth 1923). The ichthyophonida, so far, only have amoeboid motility and spindle pole bodies are present in at least two lineages (Fig. 6.1) (Marshall et al. 2008; Chapter 5). Centrioles are probably absent. Flagellated and amoeboid stages are present in animals and amoebozoans; both stages may have also been present in the ancestral ichthyosporean. P. gemmata and A. whisleri do share a feature with the dermocystid Rhinosporidium in the form of pulsed spore release. This trait may have been present in the common ancestor to the ichthyophonida and dermocystida and since been lost from many taxa. Membrane bound tubules that penetrate the inner cell wall layers are unique to the ichthyosporea. These MBTs are contiguous with the cell‘s cytoplasm and plasma membrane and extend outward, penetrating the inner most layer of the cell wall (Marshall et al. 2008, Chapter 4, Chapter 5). Ichthyosporeans grow as relatively large unicells and the MBTs may function to increase surface area for nutrient assimilation. Alternately, these tubules may play a role in cell wall dynamism as ichthyosporeans shift between amoeboid, plasmodial and walled stages. The MBTs were most extensive in A. whisleri, the species that showed the most dramatic changes in shape. In C. fragrantissima MBTs were longer and more dense next to a newly formed and possibly incomplete shared cell wall (not shown). MBTs were also longer and more numerous near what seemed to be 183  incompletely formed patches of wall material in Sphaeroforma tapetis (Chapter 4). However, MBTs were still present, albeit in lower numbers, in Sphaeroforma nootkatensis,  a species without known plasmodial or amoeboid forms (Chapter 4). I. hoferi also has a complex life cycle and transitions between plasmodial, amoeboid and walled osmotrophic stages (Okamoto et al. 1985) but MBTs are not evident in any micrographs of this species. The lack of MBTs in previous studies may be an artifact of chemical fixation procedures and further studies using high pressure freezing of healthy cultured cells may be necessary to survey the prevalence of MBTs throughout the ichthyosporea. Establishing prevalence and correlation of MBTs with life cycle stages may help determine their function. Amongst the ichthyosporeen lineage ichthyophonida, plasmodial stages with budding and micrographs with ‗empty‘ wall material had only been shown in Ichthyophonus (Spanggard et al. 1996). We show that this character is present in at least two additional clades (Figure 6.1) (Chapter 4, Chapter 5).  Life history characteristics of the ichthyosporea Life cycles  Asexual life cycles among the ichthyosporea are notably complex and few have been documented in completion. For example, I showed that motile plasmodial stages are present in several genera (e.g. Chapters 4 and 5), however, the function and placement of plasmodial stages within the life cycles is still unknown. Complete asexual lifecycles from host integrated studies of ichthyosporean life cycles have only conducted in the Amoebidiales and Eccrinales (e.g. Whisler 1965; Hibbits 1978). In these species at least two types of asexual reproduction each culminate in a different type of spore. In the Amoebidiales, either amoebae or non-motile spores are produced and each type functions either to recolonize the same animal or to disseminate to a new host (Whisler 1965). The eccrinids are similar but produce non-motile spores that differ in the number of nuclei depending on their function (Hibbits 1978). Unfortunately, eccrinids have not been cultured. In my studies, having living cultures permitted me to make repeated observations on asexual reproduction in large numbers of single cells. However relating 184  structures and stages, such as plasmodia, extracellular secretions and calyxes to the ichthyosporean life history will require further studies, likely involving host inoculations.  My studies of asexual reproduction and development corroborate other studies in showing that the formation of a multinucleate stage usually precedes the onset of sporogenesis (Mendoza et al. 2002). However the process of sporogenesis that I described for Pirum gemmata, was different from the development described for Amoebidium (Whisler 1965) or Palavascia sp. (Cafaro 2000) (Table F.2). Prior to this thesis, ploidy had not been determined for any member of the unicellular opisthokonts. Marshall and Berbee (2009, submitted (Chapter 3) provide evidence for haploid asexual reproduction in a Sphaeroforma. tapetis (= Pseudoperkinsus tapetis). Lack of heterozygosity in other Sphaeroforma species having intralocus variation (not shown) suggests that haploidy may predominate within Sphaeroforma, however further studies would be required from more distant relatives to demonstrate whether this can be generalized within the ichthyosporea.  In Chapter 3, I demonstrated that sexual recombination occurred in the ichthyosporean S. tapetis and that sex was most likely facultative. My work provided the first genetic evidence for a sexual cycle for any member of the unicellular basal opisthokont lineages, although finding evidence for recombination was perhaps consistent with the general rarity of ancient asexual lineages. I detected different haplotypes of S. tapetis from the same host animal collection, and the frequent cohabitation of conspecifics with different genotypes may have provided frequent enough opportunities for mating to maintain a signal of recombination despite the overall clonal population structure (Chapter 3). The mechanism for genetic exchange is still unknown. It may be that ichthyosporeans begin sexual cycles with some form of cell fusion. Scaliform conjugation and nuclear fusion was observed in Enteropogon sexuale (Hibbits 1978) but meiosis has not been observed directly in any species. C. fragrantissima makes multicellular aggregates that temporarily lose their shared cell wall and then build a new dividing wall (Marshall et al. 2008). Video shows that the aggregates are not formed by budding but occurs when two or more amoebae encyst next to each other (not shown). However this type of temporary fusion may be a form of colonial behaviour rather than a stage in genetic exchange. 185  Ecology Although some genera and clades of ichthyophonida tend to be preferentially associated with specific host animal phyla, for most ichthophonids, species specific host relationships and host/symbiont co-evolution have not been evident. Amoebidium parasiticum (Whisler 1963) showed little host specificity, although a morphological criterion was used for recognition of this species and morphologically-identical host- specific crypic species would not have been detected. Lack of genetic variation in the ITS and EFL loci among C. fragrantissima isolates from four hosts also suggests a lack of host specificity amongst several invertebrate phyla (Marshall et al 2008). However once again, cryptic but host-specific species, this time with identical ITS and EFL sequences could not be ruled out.  More compellingly, Jones and Dawe (2002) demonstrated that Ichthyophonus hoferi was able to infect new hosts as it passed through the food chain via predation. Genetic evidence from my studies, from loci with intraspecific variation showed that Sphaeroforma species from British Columbia were clearly not host specific (Chapters 3 and 4). All of these taxa with a probable lack of host specificity also grew well in culture, suggesting that they may not be host obligate, and may have evolved from a free living ancestor. In contrast, the dermocystida are host obligate and may have undergone co-evolution with their vertebrate hosts. Silva et al. (2005) showed how strains of a single SSU-rDNA phylotype of Rhinosporidium seeberi varied in their ITS sequence depending on both host species and geographic location.  Suggesting that many ichthyosporeans remain to be described, I discovered five new species and many sequences detected during DNA based surveys still do not correspond to any described species. Some of our new species may be endemic to British Columbia. But one of three Sphaeroforma species from British Columbia was more likely conspecific with isolates from different oceans, suggesting that endemism is not the rule for all species (Chapter 4). So far only eccrinids, amoebidiales and Rhinosporidium have been collected in the Southern hemisphere. No phylogeographical studies have been conducted on ichthyosporeans other than for R. seeberi which had a small sample size (Silva et al. 2005). Phylogeographic and population based studies could show species ranges, evidence for endemics, or co-evolution amongst symbiotic species. 186  Table 6.1 Features of clades within the Ichthyophonida.  List of genera or family Hosts Clade Characteristics Clade I Abeoforma Pirum Marine invertebrates 1.  Electron dense, fibrous   outer cell wall layer 2. Spindle pole bodies 3. Membrane bound tubular extensions (MBTs) 4. Pulsed endospore release through tears Clade II Ichthyophonus 1 Amoebidiales 2 Eccrinales 3 Fish 1 Arthropods 2,3 1. Elongate, thread-like and hyphal 1  shapes 2. Multi-layered walls 3. Amoebae 1,2 4. Walled spores 5. Holdfasts 2,3 Clade III Psorospermium Crustaceans 1. Thick multi-layered wall with plates 2. Amoebae 3. Double nucleus in developing cells Clade IV Sphaeroforma 4 Creolimax 4 Anurofeca 5 Various marine invertebrates 4 , tadpoles 5 1. Spherical 2. Homogenous electron dense outer wall layer 3. MBTs  and lucent inner wall layer 4 4. Peripheral nucleolus (except Creolimax) 1-5 Clarification provided when characteristics are not shared between all taxa. See table F.2 for sources. 187  SSU-rDNA IV. Spherical Ichthyophonida I. Previously uncultured clade II. Eccrinales,    Amoebidiales    & Ichthyophonus 0.1 Capsaspora owczarzaki 76/79/100 77/-/99 86/-/100 -/-/100 85/100/100 Dermocystida III.Psorospermium haeckeli Ichthyophonus Amoebidiales and Eccrinales Marine Freshwater A. Spindle pole body B. MBTs C. Plasmodia D. Extracellular      "wall remnants" E. Amoebae F. Binary fission C. Plasmodia D. Extracellular      "wall remnants" E.  Amoebae G. Pores in wall H. Holdfast/calyx E. Amoebae A. Spindle pole body B. MBTs C. Plasmodia D. Extracellular      "wall remnants" E. Amoebae F. Binary fission G. Pores in wall H. Holdfast/calyx *Abeoforma *Pirum *Creolimax *Sphaeroforma   Figure 6.1. Shared characteristics between ichthyophonida from different clades (I- IV). Maximum likelihood RAxML. Numbers at nodes represent 500 RAxML or parsimony bootstrap replications or Bayesian posterior probabilities (RAxML/MP/Ba) over 60 % (according to parameters used in chapter 5). Thickened branches show nodes that have 90 % or greater support by all three methods. 188   References Ashworth JH (1923) On Rhinosporidium seeberi (Wernike 1903), with special reference to its sporulation and affinities. Trans R Soc Edinb 53:301-342 Baker GC Beebee TJC Ragan MA (1999) Prototheca richardsi, a pathogen of anuran larvae, is related to a clade of protistan parasites near the animal-fungal divergence. Microbiology 145:1777-1784 Baldauf SL Palmer J (1993) Animals and fungi are each other's closest relatives: congruent evidence from multiple proteins. Proc Natl Acad Sci USA 90:11558- 11562 Beebee TJC (1991) Purification of an agent causing growth inhibition in anuran larvae and its identification as a unicellular unpigmented alga. Can J Zool 69:2146-2153 Cafaro MJ (2000) Gut fungi of isopods: the genus Palavascia. Mycologia 92:361-369 Cafaro MJ (2005) Eccrinales (Trichomycetes) are not fungi, but a clade of protists at the early divergence of animals and fungi. Mol Phyl Evol 35:21-34 Cavalier-Smith T (1998) Neomonada and the origin of animal and fungi. Pages 375-407 in G. Coombs, K. Vickerman, M. Sleigh, and A. Warren, editors. Evolutionary relationships among protozoa. Chapman and Hall, London Cavalier-Smith T Chao E (2003) Phylogeny of choanozoa, apusozoa, and other protozoa and early eukaryote megaevolution. J Mol Evol 56:540-563 Dang S-N Lichtwardt RW (1979) Fine structure of Paramoebidium (Trichomycetes) and a new species with virus-like particles. Am J Bot 66:1093-1104 Dykova I Veverkova M Fiala I Machackova B Peckova H (2003) Nuclearia pattersoni sp. n. (Filosea), a new species of amphizoic amoeba isolated from the gills of roach (Rutilus rutilus), and its rickettsial endosymbiont. Folia Parasitol 51:161- 170 Franco-Sierra A Alvarez-Pellitero P (1999) The morphology of Ichthyophonus sp. in their mugilid hosts (Pisces: Teleostei) and following cultivation in vitro. A light and electron microscopy study. Parasitol Res 85:562-575 Hertel LA Bayne CJ Loker ES (2002) The symbiont Capsaspora owczarzaki, nov. gen. nov. sp., isolated from three strains of the pulmonate snail Biomphalaria glabrata is related to members of the Mesomycetozoa. Int J Parasitol 32:1183-1191 189  Hibbits J (1978) Marine Eccrinales (Trichomycetes) found in crustaceans of the San Juan Archipelago, Washington. Syesis 11:213-261 Jostensen JP Sperstad S Johansen S Landfald B (2002) Molecular-phylogenetic, structural and biochemical features of a cold adapted, marine ichthysporean near the animal-fungal divergence, described from in vitro cultures. Europ J Protistol 38:93-104 Kerk D Gee A Standish M Wainwrite P Drum A Elston R Sogin M (1995) The rosette agent of chinook salmon (Oncorhynchus tshawytscha) is closely related to choanoflagellates, as determined by the phylogenetic analysis of its small subunit RNA. Mar Biol 122:187-192 Marshall WL Celio G McLaughlin DJ Berbee ML (2008) Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. Protist 159:415-433 Mayfield S-ND Lichtwardt RW (1980) Comparative study of the holdfast structure in four Trichomycetes. Can J Bot 58:1074-1087 Mendoza L Taylor JW Ajello L (2002) The class Mesomycetozoea: a heterogeneous group of microorganisms at the animal-fungal boundary. Ann Rev Microbiol 56:315-344 Moss ST (1999) Astreponema gammari: an eccrinid with appendaged spores. Kew Bull 54:637-650 Okamoto N Nakase K Suzuki H Nakai Y Fujii K Sano T (1985) Life history and morphology of Ichthyophonus hoferi in vitro. Fish Pathol 20:273-286 Ordas MC Figueras A (1998) In vitro culture of Perkinsus atlanticus. Dis Aquat Org 33:129-136 Owczarzak A Stibbs HH Bayne CJ (1980)The destruction of Schistosoma mansoni mother sporocysts in vitro by amoebae isolated from Biomphalaria glabrata: an ultrastructural study. J Invert Pathol 35:26-33 Patterson DJ (1983) On the organization of the naked filose amoeba, Nuclearia moebiusi Frenzel, 1897 (Sarcodina, Filosea) and its implications. J Protozool 30:301-307 190  Pekkarinen M Lom J Murphy CA Ragan MA Dykova I (2003) Phylogenetic position and ultrastructure of two Dermocystidium species (Ichthyosporea) from the common perch (Perca fluviatilus). Acta Protozool 42:287-307 Raghu-kumar S (1987) Occurrence of the Thraustochytrid, Corallochytrium limacisporum gen. et sp. nov. in the Coral Reef Lagoons of the Lakshadweep Islands in the Arabian Sea. Bot Mar 30:83-89 Richards CM (1958) The inhibition of growth in crowded Rana pipiens tadpoles. Physiol Zool 31:138-151 Rug M Vogt G (1995) Histology and histochemistry of developing and mature spores of the two morphotypes of "Psorospermium haeckeli". Freshw Crayfish 10:374-384 Ruiz-Trillo I Lane CE Archibald JM Roger AJ (2006) Insights into the evolutionary origin and genome architecture of the unicellular opisthokonts Capsaspora owczarzaki and Sphaeroforma arctica. J Eukaryot Microbiol 53:1-6 Shalchian-Tabrizi K Minge MA Espelund M Orr R Ruden T Jakobsen KS Cavalier-Smith T (2008) Multigene phylogeny of choanozoa and the origin of animals. PLoS ONE 3:e2098 Silva V Pereira CN Ajello L Mendoza L (2005) Molecular evidence for multiple host-  specific strains in the genus Rhinosporidium. J Clin Microbiol 43:1865-1868 Spanggaard B Huss HH Bresciani J (1995) Morphology of Ichthyophonus hoferi assessed by light and scanning electron microscopy. J Fish Dis 18:567-577. Spanggaard B Skouboe P Rossen L Taylor JW (1996) Phylogenetic relationship of the intracellular fish pathogen Ichthyophonus hoferi, and fungi, choanoflagellates and the rosette agent. Mar Biol 126:109-115 Takishita K Fujiwara Y Kawato M Kakizoe N Miyazaki M Maruyama T (2008) Molecular identification of the ichthyosporean protist "Pseudoperkinsus tapetis" from mytilid mussel Adipicola pacifica associated with submerged whale carcasses in Japan. Marine Biotechnology 10:13-18 Vogt G Rug M (1995) Microscopic anatomy and histochemistry of the crayfish parasite Psorospermium haeckeli. Dis Aquat Org 21:79-90 191  Vogt G Rug M (1999) Life stages and tentative life cycle of Psorospermium haeckeli, a species of the novel DRIPs clade from the animal-fungal dichotomy. J Exp Zool 283:31-42 Whisler HC (1962) Culture and nutrition of Amoebidium parasiticum. Am J Bot 49:193- 199 Whisler HC (1963) Observations on some new and unusual enterophilous phycomycetes. Can J  Bot 41:887-900 Whisler HC (1965) Host-integrated development in Amoebidiales. J Protozool 13:183- 188 Whisler HC Fuller MS (1968) Preliminary observations on the holdfast of Amoebidium parasiticum. Mycologia 60:1068-1079 Wahl M (1989) Marine epibiosis. I. Fouling and antifouling: some basic aspects. Mar  Ecol Prog Ser 58:175-189 192  Appendices Appendix A. Abstracts for published or submitted manuscripts Abstract: Chapter 2. Multiple isolations of a culturable, motile Ichthyosporean (Mesomycetozoa, Opisthokonta), Creolimax fragrantissima n. gen., n. sp., from marine invertebrate digestive tracts. A fragrant, spherical, osmotrophic eukaryote was isolated 27 times from the digestive tracts of marine invertebrates collected from the Northeast Pacific. The isolates were cultured from seven animal collections over a two-year period, most from the peanut worm, Phascolosoma agassizii. A small subunit ribosomal DNA phylogeny placed the spherical organism within the ichthyosporea, closest to Sphaeroforma arctica and Pseudoperkinsus tapetis. Supporting the close relationship of isolates, the sequences of ribosomal gene internal transcribed spacers determined for 26 isolates were identical, as were the elongation factor 1-alpha-like gene fragments from seven isolates. Dispersal via amoeboid cells distinguished this species from its closest relatives and led to the erection of a new genus and species, “Creolimax fragrantissima.”   Vegetative cells reproduced asexually in vitro after they reached 30 to 60 microns in diameter by producing amoebae or endospores, which escaped through openings in the parent cell wall.  Ultrathin sections of vegetative cells prepared by high-pressure freeze substitution provided some of the first images of ichthyosporean spindle pole bodies and document, for the first time, tubular extensions of the plasma membrane into an electron-translucent inner layer of the cell wall. Ichthyosporeans are parasites and commensals of animals and culturable species are few. Because ―C. fragrantissima” can be isolated regularly and repeatedly from nature and then grown easily through cycles of asexual reproduction, it has the potential to serve as a model organism for further research into marine ichthyosporeans. 193  Abstract: Chapter 3. Population level analyses indirectly reveal cryptic sex and life-history traits of Pseudoperkinsus sp., an ecological model for basal Opishthokont lineage Ichthyosporea. In this paper, we use population genetics to detect the molecular footprint of a sexual cycle, of a haploid vegetative state and of lack of host specificity in Pseudoperkinsus sp., a marine unicellular relative of the animals. Prior to this study complete lifecycles were not known for any of the unicellular lineages sharing common ancestry with multicellular animals and fungi. We established the first collection of conspecific cultures of any member from the unicellular opisthokont lineages, isolating 126 cultures of Pseudoperkinsus  sp. from guts of marine invertebrates ranging from clams to sea cucumbers. We considered isolates with 99.6% or more identity in the ribosomal internal transcribed spacer regions to be conspecific. We sequenced fragments of the elongation factor alpha-like for 52 isolates and heat shock protein 70 genes for 34 isolates.  Absence of heterozygotes from 52 isolates provided evidence for haploidy. Phylogenetic incongruence and a lack of support for linkage between two loci signified a history of recombination consistent with a sexual cycle.  Shared haplotypes in different invertebrate phyla indicated the species was not host specific. Based on estimates of the frequency of sex and on observations in cultures, we propose that Pseudoperkinsus sp. is transmitted between hosts via asexual endospores. New protists are continually being discovered and as this study illustrates, analysis of culturable collections from natural habitats can transform a new species from a complete unknown to a model system for better understanding of evolution of life histories in whole phyla and kingdoms. 194  Appendix B. Supplementary table for chapter 2. Table B.1. GenBank classification, accession numbers and collection details for taxa included in phylogenetic analysis. Taxa    GenBank classification Host and collection location Accession number Anurofeca richardsi  Ichthyosporea; Ichthyophonida Rana temporaria tadpoles AF070445 Uncultured Anurofeca sp. LAH-2003 Ichthyosporea; Ichthyophonida Biomphalaria glabrata; Brazil AY363958 Sphaeroforma arctica   Ichthyosporea; Ichthyophonida Gammarus setosus; Offshore from Spitsbergen, Svalbard archipelago Y16260 Ichthyophonus hoferi  Ichthyosporea; Ichthyophonida Limada ferruginea; Northwest Atlantic U25637 Ichthyophonus hoferi  Ichthyosporea; Ichthyophonida Herring; North Sea U43712 Ichthyophonus irregularis  Ichthyosporea; Ichthyophonida Limada ferruginea; Nova Scotia, Canada AF232303 Pseudoperkinsus tapetis  Ichthyosporea; Ichthyophonida Ruditapes decussatus; Spain AF192386 Psorospermium haeckeli  Ichthyosporea; Ichthyophonida Astacus astacus, Decapoda; Sweden U33180  195  Taxa    GenBank classification Host and collection location Accession number Amoebidium parasiticum FRA-1-14 Ichthyosporea; Ichthyophonida; Amoebidiacea Daphnia sp., Cladocera; Herault, France AF274051 Paramoebidium sp. KS-61-w6 Ichthyosporea; Ichthyophonida; Amoebidiacea Ephemeroptera; Uniramia; Kansas, USA AY336708 Eccrinidus flexilis SPA-10-C2 Eccrinales; Eccrinaceae Glomeris sp., Diplopoda; Barcelona, Spain AY336698 Eccrinidus flexilis SPA-11-C45 Eccrinales; Eccrinaceae Glomeris sp., Diplopoda; Tarragona, Spain AY336700 Enterobryus sp. SPA-10-C6 Eccrinales; Eccrinaceae Iulidae, Diplopoda; Barcelona, Spain AY336711 Enterobryus sp. MA-11-C1 Eccrinales; Eccrinaceae Diplopoda; Massachusetts USA AY336701 Enterobryus sp. SPA-2C-10 Eccrinales; Eccrinaceae Brachydesmus sp., Diplopoda; Barcelona, Spain AY336712 Enterobryus oxidi KS-79-W2 Eccrinales; Eccrinaceae Diplopoda; Kansas, USA AY336710 Astreptonema gammari MN-3-W6 Eccrinales; Eccrinaceae Gammarus sp., Amphipoda; Minnesota, USA AY336709  196  Taxa    GenBank classification Host and collection location Accession number Alacrinella limnoriae MA-8-W4 Eccrinales; Eccrinaceae Limnoria sp., Isopoda; Massachusetts, USA AY336703 Asteptonema sp. WA-3-C3 Eccrinales; Eccrinaceae Sphaeromatidae, Isopoda; Washington, USA AY336706 Enteromyces callianassae CA-12-C4b Eccrinales; Eccrinaceae Callianassa californiensis, Decapoda; California, USA AY336702 Palavascia sphaeromae  Eccrinales; Eccrinaceae Sphaeroma sp., Isopoda; USA AY336713 Palavascia patagonica ARG-D1-C15 Eccrinales; Eccrinaceae Exosphaeroma studeri, Isopoda; Argentina AY682845 Enteropogon sexuale WA-1-C5 Eccrinales; Eccrinaceae Upogebia pugettensis, Decapoda; Washington, USA AY336705 Taeniella carcini WA-1-C37 Eccrinales; Eccrinaceae Hemigrapsus nuda, Decapoda; Washington, USA AY336707 Taeniellopsis sp. MA-5-C17 Eccrinales; Eccrinaceae Orchestia sp., Amphipoda; Massachusetts, USA AY336704 Uncultured Ichthyophonida sp. Clone LKM-51 Ichthyosporea; environmental sample Water culture; Lake Ketelmeer, Netherlands AJ130859  197  Taxa    GenBank classification Host and collection location Accession number Uncultured eukaryote clone BB01 Eukaryota; environmental sample Seawater; Barneget Bay, Atlantic, USA AY884988 Uncultured eukaryote clone NAMAKO-34 Eukaryota; environmental sample Anoxic sediment, saline lake; Lake Namako, Japan AB252774 Uncultured eukaryote clone NAMAKO-33 Eukaryota; environmental sample Anoxic sediment, saline lake; Lake Namako, Japan AB252773 Uncultured eukaryote clone BS-DGGE Euk-1 Eukaryota; environmental sample Marine sediment; Black Sea DQ234281 Uncultured eukaryote clone DSGM-65 Eukaryota; environmental sample Marine methane cold seep sediment; Sagumi Bay, Japan AB275065 Uncultured Amoebidium clone Eukaryota; environmental sample Sulphurous lake water; Lake Vilar, Spain AJ862378 Uncultured eukaryote clone D3P06H09 Eukaryota; environmental sample Anoxic intertidal sediment; Greenland EF100301 Uncultured eukaryote clone D4P08E02 Eukaryota; environmental sample Anoxic intertidal sediment; Greenland EF100330  198  Taxa    GenBank classification Host and collection location Accession number Uncultured eukaryote clone Quinhai 5-2 (Q-52) Eukaryota; environmental sample Saline lake water; Lake Quinhai, China AM709508 Uncultured eukaryote clone TAGIRI-25 Eukaryota; environmental sample Marine anoxic mud; Japan AB191433 Uncultured eukaryote clone TAGIRI-26 Eukaryota; environmental sample Marine anoxic mud; Japan AB191434 Uncultured eukaryote clone TAGIRI-27 Eukaryota; environmental sample Marine anoxic mud; Japan AB191435 Uncultured mesomycete Ikaite un-c53 Ichthyosporea; environmental sample Ikaite tufa columns; Ikka Fjord, Greenland AJ431360 Uncultured marine eukaryote Clone s16 a  Eukaryota; environmental sample Seawater; Bay of Fundy, Atlantic Ocean, USA AY331727 Uncultured marine eukaryote Clone ms313 a  Eukaryota; environmental sample Seawater; Bay of Fundy, Atlantic Ocean, USA AY331725 Uncultured marine eukaryote Clone s25 a  Eukaryota; environmental sample Seawater; Bay of Fundy, Atlantic Ocean, USA AY331726  199  Taxa    GenBank classification Host and collection location Accession number Uncultured marine eukaryote Clone s712 a  Eukaryota; environmental sample Seawater; Bay of Fundy, Atlantic Ocean, USA AY331729 Uncultured marine eukaryote Clone s416 a  Eukaryota; environmental sample Seawater; Bay of Fundy, Atlantic Ocean, USA AY331730 Sphaerothecum destruens Isolate RA-3 Ichthyosporea; Dermocystida Onchorhynchus tschawytscha; California, USA AY267345 Dermocystidium sp.  Ichthyosporea; Dermocystida Salvelinus fontinalis; Nova Scotia, Canada U21336 Rhinosporidium seeberi  Ichthyosporea; Dermocystida Human; Sri Lanka  AF118851 Dermocystidium salmonis  Ichthyosporea; Dermocystida  Onchorhynchus tschawytscha; Oregon, USA U21337 Dermocystidium percae  Ichthyosporea; Dermocystida  Perca fluviatilis; Lake Hirvijärvi, Finland AF533949 Capsaspora owczarzaki Strain 1316R1 Ichthyosporea Biomphalaria glabrata; Mollusca AF436889 Corallochytrium limacisporum  Eukaryota; Choanozoa Free living marine; coral reef lagoons Arabian Sea India L42528 200  Taxa    GenBank classification Host and collection location Accession number Nuclearia simplex  Nucleariidae; Nuclearia Free living AF484687 Ministeria vibrans ATCC 50519 Eukaryota; Choanozoa Free living AF271998 Monosiga ovata  Choanoflagellida; Codonosigidae Free living AF084230 Diaphanoeca grandis  Choanoflagellida; Acanthoecidae Free living AF084234 Xenopus laevis  Metazoa; Chordata  X04025 Mytilus trossulus  Metazoa; Mollusca  L33455 Olpidium brassicae  Fungi; Chytridiomycota  DQ322624 Spizellomycete sp. JEL335 AFTOL ID-637 Fungi; Chytridiomycota  DQ536477 Saccaromyces cerevisiae  Fungi; Ascomycota  J01353 Acanthamoeba polyphaga  Eukaryota; Acanthamoebidae  AF019056 a These sequences were removed from GenBank during preparation of this manuscript. 201  Appendix C. Supplementary tables and figures for chapter 3. Table C.1. Collection details and haplotypes of Pseudoperkinsis sp. Sources of Pseudoperkinsus sp. isolates, by location of collection and by host animal species, along with the number of replicate PCR products sequenced and allele/haplotypes designations of each isolate according to HSP70 and EFL sequences. Animal†† (number collected), location and date Total number of isolates per host Total number of unique haplotypes sequenced Isolate ID** EFL allele # EFL PCR products sequenced HSP70 allele #  HSP70 PCR products sequenced Unique Haplotype # Mytilus sp. (n=5) East-2, Gabriola Is. Brickyard beach April 2007 42 8 BM6 BM17 BM19 BM20 BM33 4 8 3 5 3 1 2 1 2 1 6 9 4  6 1 2 1  1 3 13 5  4    BM31 BM37 5 5 1 1     BM28 + 1    MB4 MB5 MB18 2 4 4 2 2 1 6 6 6 1 1 2 1 3 3    MMB5 5 2 8 2 12    MB10 5 1    MB13 MB15 MB16 MB28 MMB3 5 5 7 + 5 1 2 1 1 1 8  8  8 1  1  2 12  14  12    MB20 5 2 + 1    MB22 MB29 5 8 2 2  4  2  6  202  Animal†† (number collected), location and date Total number of isolates per host Total number of unique haplotypes sequenced Isolate ID** EFL allele # EFL PCR products sequenced HSP70 allele #  HSP70 PCR products sequenced Unique Haplotype # Venerupis philippinarium (n=5) East-3 Gabriola Is. Maples beach 26 5 PV23 PV25 PV47 PV3 PV8 3 4 5 6 6 1 1 2 2 1 4 6  6 1 1  1 5 3 ++ 2 April 2007   PV38 6 2 6 1 2    VP2 4 2 6 1 2    VP7 VP19 4 3 2 2 6 5 2 1 3 11    VP17 3 2 5 2 11 Nuttallia obscurata (n=5) East-2 Gabriola 47 5 ON1M ON50M ONSL4 6 5 5 2 2 2 1 8 2 1 10 12 Is. Brickyard    ON5 8 1 beach April 2007   ON8 ON9 ON11 ON12 ON14 5 8 5 8 5 2 1 1 1 1 8 9 1 2 12 13    ON35 5 2    NO6 5 1 8 1 12    NO32 3 1 4 1 5    NO16 8 1 7 1 16 Mytilus sp. (n=5) East-1 Gabriola Is. Green Wharf April 2007 2 1 GM8  2 2 6 2 1  203  Animal†† (number collected), location and date Total number of isolates per host Total number of unique haplotypes sequenced Isolate ID** EFL allele # EFL PCR products sequenced HSP70 allele #  HSP70 PCR products sequenced Unique Haplotype # Leptosynapta clarki (n=5) West, Barkley Sound, Scott‘s Bay November 2004 1 1 LE7 10 2 10 2 15 Phascolosoma agassizii (n=5) West, Barkley Sound, Scott‘s Bay August 2004 1 0 Crassostrea gigas (n=3) West, Barkley Sound, Dixon Island, November 2004 5 3 CRG9 CG2 CG5  9 2 1 2 1 1 7 3 4  1 2 1 15 7 8 Katharina tunicata (n=3) West, Barkley Sound, Scott‘s Bay November 2005 1 1 KS1 1 1 2 2 1 Oyster water West, Barkley Sound December 2005 1 1 OI22B 10 2 11 2 18 †† We did not find Pseudoperkinsus sp. in 34 additional animal collections including C. gigas (3 additional collections including one cohabiting population at East-2), Dendraster excentricus (2 collections), P. agassizii (3 additional collections), L. clarki (2 additional collections), Briaster latifrons, Thelepus sp. (2 collections), Idotea sp., Cucumaria miniata (2 collections), Strongylocentrotus droebachiensis, Limpit, Mytilus californianus 204  (3 collections), Protothacea staminea (2 collections), Corella sp., Patiria miniata, Strongylocentrotus franciscanus (2 collections), K. tunicana (2 additional collections), Styela montereyensis, and Pollicipes polymerus. ** Only sequences with EFL sequences are listed in this table. + Sequence too short to designate allele. ++ Haplotype unique to host based on EFL alone. 205  Table C.2. Primer and intron positions Primer/Intron Primer sequence Position on corresponding S. arctica DQ403166,  DQ403164 H7150F 5‘-GAGATYATYGCCAAYGAYCARGG-3‘ 49-71 H7965R 5‘-GAANCGNGCACGGGTGAT-3‘ 862-879 H7F2 5‘-TAACCGCACAACCCCTTC-3‘ 73-90 H7R2 5‘-GAAATCCTCACCACCCAAG-3‘ 654-672 HSP70 intron 1 N/A Between 148-151 ELF3 5‘-AAGTCCACGACTACCGGT-3‘ 58-75 ELR3 5‘-TGATCTGTACCTGAGCGG-3‘ 1076-1093  206  Table C.3. Tajima’s D, Fu and Li’s D, and Fu and Li’s F calculated using all 24 individuals 1 . Locus Tajima’s D (P value) Fu and Li’s D (P value) Fu and Li’s F (P value) HSP70 0.684 (>0.1) -0.310 (>0.1) 0.00224 (>0.1) EFL 1.41 (>0.1 5.66 (>0.1) 0.943 (>0.1) 1 There was also no evidence of selection on clone corrected data sets (not shown).  207  N Pacific Ocean Vancouver Victoria Vancouver Island West: Barkley Sound Mainland Georgia Strait East: Gabriola Island  Figure C.1. Pseudoperkinsus sp. isolation sites from Vancouver Island, British Columbia, Canada. Western sites were distributed within Barkley Sound, near the Bamfield Marine Sciences Centre. East side sites were from three beaches along a 3 km SSW shoreline of Gabriola Island. Specific locations are listed in table C.3. 208  Appendix D. Supplementary tables and figures for chapter 4. Table D.1. Host animal, collection date, number and specific location of host animals, and numbers of isolates from each SSU-rDNA phylotype. Animal ††  (number collected), location and date Number of S. tapetis isolates Number of S. gastrica isolates Number of S. nootkatensis isolates Dendraster excentricus (n=4) East, Parksville beach March 2003 0 0 3 Phascolosoma agassizii (n=5) West, Barkley Sound, Scott‘s Bay August 2004 1 0 0 Leptosynapta clarki (n=5) West, Barkley Sound, Scott‘s Bay November 2004 1 0 0 Mytilus californianus West, Barkley Sound, Scott‘s Bay November 2004 0 0 8 209  Animal ††  (number collected), location and date Number of S. tapetis isolates Number of S. gastrica isolates Number of S. nootkatensis isolates Crassostrea gigas (n=3) West, Barkley Sound, Dixon Island November 2004 5 3 0 Katharina tunicata (n=3) West, Barkley Sound, Scott‘s Bay November 2005 1 0 0 Corella sp. (n=4) West, Barkley Sound, BMSC South Dock November 2005 0 0 1 Phascolosoma agassizii (n=5) West, Barkley Sound, Dixon Island November 2005 0 0 1   210  Animal ††  (number collected), location and date Number of S. tapetis isolates Number of S. gastrica isolates Number of S. nootkatensis isolates Strongylocentrotus franciscanus (n=3) West, Barkley Sound, Dixon Island November 2005 0 0 1 Oyster water West, Barkley Sound December 2005 1 0 0 Mytilus sp. (n=5) East-2, Gabriola Is. Brickyard beach April 2007 42 3 0 Mytilus sp. (n=5) East-1 Gabriola Is. Green Wharf April 2007 2 0 0       211  Animal ††  (number collected), location and date Number of S. tapetis isolates Number of S. gastrica isolates Number of S. nootkatensis isolates Venerupis philippinarium (n=5) East-3 Gabriola Is. Maples beach April 2007 26 0 0 Nuttallia obscurata (n=5) East-2 Gabriola Is. Brickyard beach April 2007 47 2 0  212  5 changes Sphaeroforma arctica S. nootkatensis  A22 S. nootkatensis  CH1 S. nootkatensis  A9 S. nootkatensis  A36 S. nootkatensis  MC2 S. nootkatensis  PDD9 S. gastrica  CRG1 S. gastrica  BM11 S. gastrica  CRG3 S. gastrica ON36 S. gastrica  CGE S. gastrica  MB6 S. tapetis  ON12 S. tapetis  OI22B S. tapetis  LE7 S. tapetis  MMB5 S. tapetis  CRG9 S. tapetis  CG2 S. tapetis  VP19 S. tapetis  PV23 S. tapetis  BM33 S. tapetis  KS1 S. tapetis  GM8 S. tapetis  BM6 S. tapetis  NO16 S. tapetis  MB29 S. tapetis  MB16 S. tapetis  ON1M S. tapetis  PV3 S. tapetis  (Japan) 100/100 98/96 96/95 100/100 100/100 72/64 100/99 -/64 97/98 68/65 93/96 -/62 EFL 1 A i ii 2 iii 3 v  Figure D.1. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the EFL sequences from our 29 of our isolates, S. arctica and S. tapetis (Japan). Numbers below nodes represent 100 maximum parsimony and 100 maximum likelihood bootstrap replications (MP/ML). Nodes labelled 1-3, A-B, i-vi, correspond to Figures 4.6 and supplementary Figures D.1-D.3. New Sphaeroforma isolates from British Columbia are in bold. 213  1 change Sphaeroforma arctica S. nootkatensis A22 S. nootkatensis CH1 S. nootkatensis A9 S. nootkatensis A36 S. nootkatensis MC2 S. nootkatensis PDD9 S. gastrica BM11 S. gastrica CRG1 S. gastrica CRG3 S. gastrica ON36 S. gastrica CGE S. gastrica MB6 S. tapetis ON12 S. tapetis LE7 S. tapetis MMB5 S. tapetis VP19 S. tapetis BM33 S. tapetis KS1 S. tapetis GM8 S. tapetis BM6 S. tapetis MB29 S. tapetis MB16 S. tapetis OI22B S. tapetis CRG9 S. tapetis  CG2 S. tapetis PV23 S. tapetis NO16 S. tapetis ON1M S. tapetis PV3 S. tapetis (Japan) 100/99 86/64 100/100 100/100 74/- 87/61 62/81 97/94 3 B 2 vi 1 B HSP70  Figure D.2. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the HSP70 sequences from our 29 of our isolates, S. arctica and S. tapetis (Japan). Numbers below nodes represent 100 maximum parsimony and 100 maximum likelihood bootstrap replications (MP/ML). Nodes labelled 1-3, A-B, i-vi, correspond to Figures 4.6 and supplementary Figures D.1, D.3-D.4. New Sphaeroforma isolates from British Columbia are in bold. 214  5 changes Sphaeroforma arctica S. nootkatensis A22 S. nootkatensis CH1 S. nootkatensis A9 S. nootkatensis A36 S. nootkatensis MC2 S. nootkatensis PDD9 S. gastrica CRG1 S. gastrica BM11 S. gastrica CRG3 S. gastrica ON36 S. gastrica CGE S. gastrica MB6 S. tapetis ON12 S. tapetis OI22B S. tapetis LE7  S. tapetis MMB5 S. tapetis CRG9 S. tapetis CG2 S. tapetis VP19 S. tapetis PV23 S. tapetis BM33 S. tapetis KS1 S. tapetis GM8 S. tapetis BM6 S. tapetis NO16 S. tapetis  MB29 S. tapetis MB16 S. tapetis ON1M S. tapetis PV3 S. tapetis (Japan) ITS 100/100 100/99 100/97 87/97 74/89 81/82 100/99 76/- 1 2 3 A B iii iv v .  Figure D.3. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the ITS sequences from our 29 of our isolates, S. arctica and S. tapetis (Japan). Numbers below nodes represent 100 maximum parsimony and 100 maximum likelihood bootstrap replications (MP/ML). Nodes labelled 1-3, A-B, i-vi, correspond to Figures 4.6 and supplementary Figures D.1-D.2, D.4. New Sphaeroforma isolates from British Columbia are in bold. 215  Sphaeroforma arctica S. nootkatesis   A22 5 changes S. nootkatesis  CH1 S. nootkatesis  A9 S. nootkatesis  A36 S. nootkatesis  MC2 S. nootkatesis  PDD9 S. gastrica  CRG1 S. gastrica  BM11 S. gastrica  CRG3 S. gastrica  ON36 S. gastrica  CGE S. gastrica  MB6 S. tapetis  ON12 S. tapetis  OI22B S. tapetis  LE7 S. tapetis  MMB5 S. tapetis  CRG9 S. tapetis  CG2 S. tapetis  VP19 S. tapetis  PV23 S. tapetis  BM33 S. tapetis  KS1 S. tapetis  GM8 S. tapetis  BM6 S. tapetis  NO16 S. tapetis  MB29 S. tapetis  MB16 S. tapetis  ON1M S. tapetis  PV3 S. tapetis  (Japan ) S. tapetis  (Spain) 100/100 100/99 100/97 88/97 75/87 100/99 77/- 86/83 ITS -/65 1 2 3 A i ii B  Figure D.4. Maximum parsimony tree with mid-point rooting comparing the phylogenetic relationship based on the ITS sequences from our 29 of our isolates, S. arctica and S. tapetis (Spain and Japan). Numbers below nodes represent 100 maximum parsimony and 500 RAxML bootstrap replications (MP/RAxML). Nodes labelled 1-3, A-B, i-vi, correspond to Figures 4.6 and supplementary Figures D.1-D.3. New Sphaeroforma isolates from British Columbia are in bold. 216  Appendix E. Supplementary table for chapter 5. Table E.1. Accession number and GenBank classification of sequences used for this study. Taxa Accession number GenBank Classification Uncultured clone s16 AY331727 Eukaryota; environmental sample Uncultured clone NAMAKO-34 AB252774 Eukaryota; environmental sample Pirum gemmata  TBA Abeoforma whisleri  TBA Uncultured clone s416 AY331730 Eukaryota; environmental sample Uncultured clone s712 AY331729 Eukaryota; environmental sample Uncultured clone NAMAKO-33 AB252773 Eukaryota; environmental sample Uncultured clone Qinghai-52 AM709508 Eukaryota; environmental sample Uncultured clone s25 AY331726 Eukaryota; environmental sample Uncultured clone Amoebidium AJ862378 Eukaryota; environmental sample Uncultured clone TAGIRI-27 AB191435 Eukaryota; environmental sample Uncultured clone BSDGGE DQ234281 Eukaryota; environmental sample Uncultured clone DSGM-65 AB275065 Eukaryota; environmental sample 217  Taxa Accession number GenBank Classification Uncultured clone ms313 AY331725 Eukaryota; environmental sample Astreptonema sp. AY336706 Eccrinales; Eccrinaceae Astreptonema gammari AY336704 Eccrinales; Eccrinaceae Palavascia patagonica AY682845 Eccrinales; Eccrinaceae Enteropogon sexuale AY336705 Eccrinales; Eccrinaceae Enterobryus oxidi AY336709 Eccrinales; Eccrinaceae Taeniella carcini AY336707 Eccrinales; Eccrinaceae Taeniallopsis sp. AY336704 Eccrinales; Eccrinaceae Enterobryus sp. isolate CMJ-2003 AY336711 Eccrinales; Eccrinaceae Eccrinidus flexilis isolate SPA10C2 AY336698 Eccrinales; Eccrinaceae Paramoebidium sp. AY336708 Ichthyosporea; Ichthyophonida; Amoebidiacea Amoebidium parasiticum strain ATCC 32708 Y19155 Ichthyosporea; Ichthyophonida; Amoebidiacea Ichthyophonus hoferi U25637 Ichthyosporea; Ichthyophonida Ichthyophonus irregularis AF232303 Ichthyosporea; Ichthyophonida Psorospermium haeckeli U33180 Ichthyosporea; Ichthyophonida Uncultured clone Ikaite 53 AJ431360 Ichthyosporea; environmental sample Uncultured clone TAGIRI-25 AB191433 Eukaryota; environmental sample Uncultured clone TAGIRI-26 AB191434 Eukaryota; environmental sample Pseudoperkinsus tapetis AF122386 Ichthyosporea; Ichthyophonida 218  Taxa Accession number GenBank Classification Sphaeroforma arctica Y16260 Ichthyosporea; Ichthyophonida Uncultured clone D3P06H09 EF100301 Eukaryota; environmental sample Uncultured clone D4P08E02 EF100330 Eukaryota; environmental sample Creolimax fragrantissima EU124914 Ichthyosporea; Ichthyophonida Uncultured clone BB01 AY884988 Eukaryota; environmental sample Uncultured clone Anurofeca sp. LAH-2003 AY363958 Eukaryota; environmental sample Anurofeca richardsi AF070445 Ichthyosporea; Ichthyophonida Uncultured clone LKM-51 AJ130859 Ichthyosporea; environmental sample Amphibiothecum penneri  AY772000 Ichthyosporea; Dermocystida Dermocystidium sp. U21336 Ichthyosporea; Dermocystida Amphibiocystidium ranae  AY692319 Ichthyosporea; Dermocystida Rhinosporidium seeberi AF118851 Ichthyosporea; Dermocystida Sphaerothecum destruens AY267345 Ichthyosporea; Dermocystida Capsaspora owczarzaki  AF349564 Ichthyosporea Salpingoeca amphoridium EU011925 Choanoflagellida; Salpingoecidae Monosiga ovata AF084230 Choanoflagellida; Codonosigidae  219  Taxa Accession number GenBank Classification Salpingoeca pyxidium strain ATCC 50929 EU011930 Choanoflagellida; Salpingoecidae Diaphanoeca grandis AF084234 Choanoflagellida; Acanthoecidae  Savillea micropora EU011928 Choanoflagellida; Acanthoecidae Uncultured clone M1_18C07 DQ103822 Eukaryota; environmental sample Corallochytrium limacisporum L42528 Eukaryota; Choanozoa Pisaster ochraceus DQ060813 Metazoa; Echinodermata Corella inflata AY903930 Metazoa; Chordata Aurelia aurita AY428815 Metazoa; Cnidaria Ministeria vibrans AF271998 Eukaryota; Choanozoa Rozella allomycis isolate AFTOL- ID 297 AY635838 Fungi; Chytridiomycota Uncultured clone RT5iin3 AY082996 Eukaryota; environmental sample Uncultured clone DSGM-63 AB275063 Eukaryota; environmental sample Uncultured clone TAGIRI-23 AB191431 Eukaryota; environmental sample Taphrina johansonii strain HA1569 AJ495835 Fungi; Ascomycota Pneumocystis wakefieldiae L27658 Fungi; Ascomycota Agaricus bisporus strain Ab-12 FJ869172 Fungi; Basidiomycota Uncultured clone WIM-48 AM114806 Eukaryota; environmental sample 220  Taxa Accession number GenBank Classification Uncultured clone CCW-24 AY2302 Eukaryota; environmental sample Smittium commune AF277035 Uncultured clone TAGIRI-24 AB191432 Eukaryota; environmental sample Physoderma maydis isolate AFTOL-ID 19 AY601708 Fungi; Blastocladiomycota Rhizophydium elyensis isolate AFTOL-ID 693 DQ536479 Fungi; Chytridiomycota Catenomyces sp. JEL342 isolate AFTOL-ID 47 AY635830 Fungi; Blastocladiomycota Spizellomycete sp. JEL355 DQ536477 Fungi; Chytridiomycota Mortierella wolfii AF113425 Fungi; incertae sedis Rhizophydium sphaerotheca isolate AFTOL-ID 37 AY635823 Fungi; Chytridiomycota Nuclearia simplex A AF349566 Eukaryota; Nucleariidae Nuclearia moebiusi AF349565 Eukaryota; Nucleariidae Nuclearia delicatula AF349563 Eukaryota; Nucleariidae Nuclearia thermophila AB433328 Eukaryota; Nucleariidae Nuclearia simplex B AF484687 Eukaryota; Nucleariidae Uncultured clone RT5iin16 AY082991 Eukaryota; environmental sample Uncultured clone RT5iin14 AY082985 Eukaryota; environmental sample Apusomonadidae sp. EU349230 Eukaryota; Apusozoa 221  Taxa Accession number GenBank Classification Planomonas micra AF053088 Eukaryota; Apusozoa  222  Appendix F. Supplementary tables for chapter 6. Table F.1. ATCC accession numbers of ichthyosporean strains from newly described species. ATCC Organism Strain Designation PRA-278 Sphaeroforma gastrica BM11 PRA-279 Pirum gemmata bPW PRA-280 Abeoforma whisleri MB14 PRA-281 Sphaeroforma tapetis Cg2 PRA-282 Sphaeroforma tapetis Le7 PRA-283 Sphaeroforma nootkatesis RU1 PRA-284 Creolimax fragrantissima CH2 223  Table F.2. Comparison of morphological features and life cycles of ichthyophonida observed in this thesis (bold) to a selection of previously described species. Species Host animal  General Morphology Cell Wall Organelles observed (TEM) Sporogenesis Types of asexual (and sexual) reproduction Motile stage Sphaeroforma arctica19 Gammarid amphipod  Spherical No central vacuole 100 nm thick Homogeneous Electron dense Pores Mitochondria: flat cristae Prominent nucleolus Electron dense bodies Nuclear division preceeds spore formation  Complete transformation of parental cell into spores No known sexual stage None observed Sphaeroforma nootkatesis Marine invertebrates, various Spherical No central vacuole Outer electron dense layer Lucent inner layer Pores Calyx Flat cristae Glycogen Golgi SPB Peripheral nucleoli MBTs Fibrous opaque vesicles Nuclear division preceeds spore formation Cell walls of spores from parental cytoplasm Complete transformati on of parental cell into spores No known sexual stage None observed 224   Species Host animal  General Morphology Cell Wall Organelles observed (TEM) Sporogenesis Types of asexual (and sexual) reproduction Motile stage Sphaerforma tapetis18 Various marine invertebrates especially bivalves18,7  Spherical, often with large central vacoule18** No central vacuole Outer electron dense layer and lucent inner layer Pores Calyx Flat cristae Glycogen Golgi SPB Peripheral nucleoli Some MBT’s Fibrous opaque vesicles Occasional bipartitions of mature sporangia 18** Nuclear division preceeds spore formation Cell walls of spores from parental cytoplasm Complete transformati on of parental cell into spores Budding plasmodia Genetic evidence for a sexual stage Flagellated stage18**(disputed) Plasmodia Briefly amoeboid stage Creolimax fragrantissima Various marine invertebrates, especially peanut worm Spherical to ovoid Central vacuole Outer electron dense layer Lucent inner layer   Flat cristae Glycogen Golgi SPB Centric nucleolus MBTs in patches Nuclear division preceeds spore formation  Complete transformati on of parental cell into spores No known sexual stage Amoeboid stage Plasmodia Anurofeca richardsi Tadpoles1,2 Spherical Central vacuole Peripheral nucleous -0.5-1μm thick -thicker with maturity1,2 Electron dense vacuoles2 Fission producing 2-8 cell conglome- rates3  Binary fission2 No known sexual stage None observed 225   Species Host animal  General Morphology Cell Wall Organelles observed (TEM) Sporogenesis Types of asexual (and sexual) reproduction Motile stage Ichthyophonus hoferi Fish Spherical Produces hyphae in culture and in host (post mortem) Vacuolate Thick with three layers20 Fibrillar 21  Mitochondria with tubular cristae21,22 Golgi21 Nucleus associated organelle21 Not described Undescribed mechanism of spore and amoebae production20 Budding20 No known sexual stage Amoeboid stage 20,16 Plasmodia20 Psorospermium haeckeli Crayfish  Egg shaped sporangia4,5,6 Vacoulate4,5,6  Immature: thin Mature: three distinct layers4  Double and single nuclei4 Mitochondria with plate-like cristae4 Granule studded vesicles4 Develop from  spore precursors visible in developing sporangium4 Complete transformation of parental cell No known sexual stage Amoeboid stage4 Paramoebidium Hind guts of specific aquatic insect larvae17 Elongate, curved17 Holdfast17 Thick electron dense layer17 Membrane bound vacuoles17 Oil droplets17 Electron light pores near holdfast17 Not described Not described Amoeboid cells17 226   Species Host animal  General Morphology Cell Wall Organelles observed (TEM) Sporogenesis Types of asexual (and sexual) reproduction Motile stage Amoebidium parasiticum Daphnia and other aquatic arthropods Moderately elongate thallus Holdfast13 Thin but distinct14  Electron transparent pits near holdfast14 Golgi14 Mitochondria with plate-like cristae 14  Simultaneous diagonal cleavage for spores13 Horizontal cleavage for amoebae13 Complete transformation of parental cell 13,17 No known sexual stage Amoeboid cells13,15 Pirum gemmata Peanut worm Pear shaped in length and round in cross section Prominent system of vacuoles Outer electron dense fibrous layer Intermittent lucent inner layer Flat cristae Glycogen SPB Centric nucleolus MBTs in patches Nuclear division preceeds spore formation Serial cleavage of cytoplasm to produce spores Complete transformati on into endospores No known sexual stage None observed 227   Species Host animal  General Morphology Cell Wall Organelles observed (TEM) Sporogenesis Types of asexual (and sexual) reproduction Motile stage Abeoforma whisleri Mussel Mostly spherical Vacuolate Various plasmodial forms Outer electron dense fibrous layer sometimes present Complex lucent inner layer Flat cristae Glycogen SPB Centric nucleolus Membranous vesicles with concentric rings MBTs Nuclear division preceeds spore formation  Complete transformati on into endospores Binary fission Budding No known sexual stage Amoeboid and plasmodial stages Enteropogon sexual8 Hindgut of mudshrimp Unbranched elongate thalli Holdfast Vacuolate Not described Not described Not described Possible sexual reproduction by scaliform conjugation  None observed Enterobryus sp. Arthropoda, Diplopoda, Nematoda Narrow and elongate11 Holdfast11 Some species with bifurcated thallus11 Multiple layers11 Electron transparent pores in wall near holdfast11  Reticulate ring structure in wall near holdfast11 Prominent nucleolus11  Not described Not described None observed 228   Species Host animal  General Morphology Cell Wall Organelles observed (TEM) Sporogenesis Types of asexual (and sexual) reproduction Motile stage Astreptonema gammari Gammarid amphipod9 Elongate9 Holdfast9 Some stages with large vacuole9 Absent on spores9 Thickens with maturity with two distinct layers9 Appendages on spores9 Golgi9 ―Septal‖ pores between developing sporangiospores9   Not described Not described None observed Palavacia sp. Isopoda12 Elongate12 Holdfast12 Not described Not described Basipetal septation12 Partial conversion of parental cell12 None observed12 ** S. tapetis ( formerly = Pseudoperkinsus tapetis) was described based on a mixed culture with Perkinsus which has a central vacuole and flagellated spores. 1. (Richards, 1958), 2. (Beebee, 1991), 3. (Baker, Beebee, and Ragan, 1999), 4. (Vogt and Rug, 1999), 5. (Rug and Vogt, 1995), 6. (Vogt and Rug, 1995), 7. Takishita et al. 2008 , 8. (Hibbits, 1978), 9. (Moss, 1999), 10. (Whisler, 1963), 11. (Mayfield and Lichtwardt, 1980), 12. (Cafaro 2000), 13. (Whisler, 1965), 14. (Whisler and Fuller, 1968), 15. (Whisler, 1962), 16. (Spanggaard, Huss, and Bresciani, 1995), 17. (Dang and Lichtwardt, 1979), 18. (Ordas and Figueras, 1998), 19. (Jostensen et al., 2002), 20. (Okamoto et al., 1985), 21. (Franco-Sierra and Alvarez-Pellitero, 1999), 22. (Spanggaard et al., 1996).    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0070884/manifest

Comment

Related Items