Genomic, Transcriptomic and Functional Studies of Diplomonads

Home , Diplomonad, Hydrogenosome, Mitosome, Naegleria, Octomitus, Spironucleus

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 990

Hidden Diversity Revealed

JON JERLSTRÖM-HULTQVIST

ACTA UNIVERSITATIS UPSALIENSIS ISSN 1651-6214 ISBN 978-91-554-8520-7 UPPSALA urn:nbn:se:uu:diva-182831 2012 Dissertation presented at Uppsala University to be publicly examined in B22, Biomedicinskt centrum (BMC), Husargatan 3, Uppsala, Friday, December 14, 2012 at 09:00 for the degree of Doctor of Philosophy. The examination will be conducted in English.

Abstract Jerlström-Hultqvist, J. 2012. Hidden Diversity Revealed: Genomic, Transcriptomic and Functional Studies of Diplomonads. Acta Universitatis Upsaliensis. Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 990. 104 pp. Uppsala. ISBN 978-91-554-8520-7.

The diplomonads are a diverse group of eukaryotic microbes found in oxygen limited environments such as the intestine of animals were they may cause severe disease. Among them, the prominent human parasite Giardia intestinalis non-invasively colonizes the small intestine of humans and animals where it induces the gastrointestinal disease giardiasis. Two of the eight genetic groups of G. intestinalis, assemblage A and B, are known to infect humans and have zoonotic potential. At the start of project, genome scale data from assemblage B-H was either sparse or entirely missing. In this thesis, genome sequencing was performed on the assemblage B isolate GS (Paper I) and the P15 isolate (Paper III) of the hoofed-animals specific assemblage E to investigate the underlying components of phenotypic diversity in Giardia. Comparisons to assemblage A isolate WB revealed large genomic differences; entirely different repertoires of surface antigens, genome rearrangements and isolate specific coding sequences of potential bacterial origin. We established that genomic differences are also manifested at the transcriptome level (Paper VIII). In a follow up analysis (Paper IV) we concluded that the Giardia assemblages are largely reproductively isolated. The large genomic differences observed between Giardia isolates can explain the phenotypic diversity of giardiasis. The adaptation of diplomonads was further studied in Spironucleus barkhanus (Paper II), a fish commensal of grayling, that is closely related to the fish pathogen Spironucleus salmonicida, causative agent of systemic spironucleosis in salmonid fish. We identified substantial genomic differences in the form of divergent genome size, primary sequence divergence and evidence of allelic sequence heterozygosity, a feature not seen in S. salmonicida. We devised a transfection system for S. salmonicida (Paper VI) and applied it to the study of the mitochondrial remnant organelle (Paper VII). Our analyses showed that S. salmonicida harbor a hydrogenosome, an organelle with more metabolic capabilities than the mitosome of Giardia. Phylogenetic reconstructions of key hydrogenosomal enzymes showed an ancient origin, indicating a common origin to the hydrogenosome in parabasilids and diplomonads. In conclusion, the thesis has provided important insights into the adaptation of diplomonads in the present and the distant past, revealing hidden diversity.

Keywords: Giardia intestinalis, Spironucleus salmonicida, Spironucleus barkhanus, intestinal parasite, hydrogenosome, mitosome, lateral gene transfer, horizontal gene transfer, diplomonad, metamonad, sexual recombination, transfection, protein complex purification

Jon Jerlström-Hultqvist, Uppsala University, Department of Cell and Molecular Biology, Microbiology, Box 596, SE-751 24 Uppsala, Sweden.

ISSN 1651-6214 ISBN 978-91-554-8520-7 urn:nbn:se:uu:diva-182831 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-182831)

”The anus was a prerequisite for intelligence; without it, heads and brains would not have evolved.”

- T. Cavalier-Smith (2006), evolutionary biologist

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Franzén O, Jerlström-Hultqvist J, Castro E, Sherwood E, Ankarklev J, Reiner DS, Palm D, Andersson JO, Andersson B, Svärd SG. (2009) Draft genome sequencing of Giardia intestinalis assemblage B isolate GS: is human giardiasis caused by two different species? PLoS Pathogens 5(8) II Roxström-Lindquist K, Jerlström-Hultqvist J, Jørgensen A, Troell K, Svärd SG, Andersson JO. (2010) Large genomic differences between the morphologically indistinguishable diplomonads Spironucleus barkhanus and Spironucleus salmonicida. BMC Genomics. 11:258. III Jerlström-Hultqvist J, Franzén O, Ankarklev J, Xu F, Nohýnko- vá E, Andersson JO, Svärd SG, Andersson B. (2010) Genome analysis and comparative genomics of a Giardia intestinalis assemblage E isolate. BMC Genomics. 11:543. IV Xu F, Jerlström-Hultqvist J, Andersson JO. (2012) Genome- Wide Analyses of Recombination Suggest That Giardia intestinalis Assemblages Represent Different Species. Molecular Bi- ology and Evolution. 29(10): 2895-98. V Jerlström-Hultqvist J, Stadelmann B, Birkestedt S, Hellman U, Svärd SG. (2012) Plasmid vectors for proteomic analyses in Giardia: purification of virulence factors and analysis of the proteasome. Eukaryotic Cell. 11(7):864-73. VI Jerlström-Hultqvist J, Einarsson E, Svärd SG. (2012) Stable transfection of the diplomonad parasite Spironucleus salmonicida. Eukaryotic Cell. 11(11):1353-61 VII Jerlström-Hultqvist J , Einarsson E, Xu F, Hjort K, Ek B, Stein- hauf D, Bergqvist J, Andersson JO, Svärd SG. Spironucleus mitochondrial remnants suggest that hydrogenosomes are ancient organelles. Submitted. VIII Franzén O*, Jerlström-Hultqvist J*, Einarsson E, Ankarklev J, Ferella M, Andersson B, Svärd SG. Transcriptome Profiling of Giardia intestinalis Using Strand-specific RNAseq. Submitted.

*These authors contributed equally

Reprints were made with permission from the respective publishers.

Review articles or other publications, not included in the thesis.

I Ankarklev J, Jerlström-Hultqvist J, Ringqvist E, Troell K, Svärd SG. (2010) Behind the smile: cell biology and disease mechanisms of Giardia species. Nature Reviews Microbiolo- gy. 8(6):413-22. (Review article) II Jerlström-Hultqvist J, Ankarklev J, Svärd SG. (2010) Is human giardiasis caused by two different Giardia species? Gut Mi- crobes. 1(6):379-82. (Review article) III Hertz HM, von Hofsten O, Bertilson M, Vogt U, Holmberg A, Reinspach J, Martz D, Selin M, Christakou AE, Jerlström- Hultqvist J, Svärd S. (2012) Laboratory cryo soft X-ray microscopy. Journal of Structural Biology. 177(2):267-72.

Contents

Sammanfattning på svenska (Summary in Swedish) ...... 11 Genetiska studier av Giardia (Paper I, III & IV) ...... 12 Transkriptomstudier av Giardia (Paper VIII) ...... 12 Miljöanpassning hos diplomonader (Paper II & VIII) ...... 13 Funktionella studier av diplomonader (Paper V, VI, VII) ...... 13 1. Introduction ...... 15 1.1 The eukaryotic cell ...... 15 1.1.1 Origins ...... 15 1.1.2 The demise of the Archezoan hypothesis ...... 16 1.1.3 Hydrogenosomes: discovery and origins ...... 17 1.2 Excavata ...... 19 1.3 Fornicata ...... 20 1.4 The Giardia cell ...... 21 1.4.1 Giardia classification and assemblages ...... 22 1.5 The Giardia life cycle ...... 23 1.5.1 The trophozoite ...... 23 1.5.2 Differentiation...... 28 1.6 Cell biology and metabolism of Giardia ...... 30 1.6.1 Cellular metabolism ...... 31 1.6.2 Sex in Giardia? ...... 33 1.7 Giardia pathogenesis ...... 34 1.7.1 Giardiasis ...... 35 1.7.2 Host-parasite interactions ...... 35 1.7.3 Antigenic variation ...... 37 1.7.4 Treatment ...... 41 1.8 The G. intestinalis genome and transcriptome ...... 41 1.8.1 Transcriptional landscape ...... 42 1.8.2 Gene families ...... 45 1.9 Spironucleus ...... 47 1.9.1 Taxonomy and host association ...... 47 1.9.2 Genomics, cell biology and metabolism ...... 48 1.9.3 Infection and transmission ...... 50 2. Present investigation ...... 52

2.1 Comparative genomics and transcriptomics in Diplomonads (Paper I, II, III, IV and VIII) ...... 53 2.1.1 Genome sequencing of G. intestinalis ...... 53 2.1.2 The GS genome (Paper I) ...... 53 2.1.3 The P15 genome (Paper III) ...... 55 2.1.4 Transcriptomics of G. intestinalis (Paper VIII) ...... 57 2.1.5 Genomic characterization of S. salmonicida and S. barkhanus (Paper II) ...... 59 2.2 Applications of genome and transcriptome data (Paper IV, V, VI and VII) ...... 61 2.2.1 Expression and purification of proteins in Giardia (Paper V) .... 61 2.2.2 Establishing a transfection system in S. salmonicida ...... 62 2.2.3 Genome-wide analyses of recombination suggest that Giardia intestinalis assemblages represent different species (Paper IV) ...... 64 2.2.4 Hydrogenosomes in the ancestor of diplomonads and parabasalids (Paper VIII) ...... 65 3. Discussion ...... 69 3.1 Are G. intestinalis assemblages different species? ...... 69 3.1.1 The plastic genome of Giardia ...... 69 3.1.2 Gene families present in the plastic genome ...... 70 3.1.3 The mysteries of G. intestinalis gene expression ...... 70 3.1.4 Presence of ASH in Giardia and its implications for the cell ..... 71 3.2 Evolutionary trends in diplomonads ...... 72 3.2.1 Gene acquisitions: a common theme in diplomonads ...... 72 3.2.2 Alternative genetic code and selenocysteine ...... 72 3.2.3 Sex in diplomonads ...... 73 3.3 Finding the essence of diplom(on)adness? ...... 74 3.3.1 S. salmonicida as a hexamitid model system ...... 74 3.3.2 Where are the S. salmonicida cysts? ...... 74 3.3.3 Cellular organization in diplomonads ...... 75 3.4 The MRO in diplomonads ...... 75 3.4.1 Hydrogenosomes in the ancestor of diplomonads and trichomonads ...... 76 3.4.2 Hydrogenosome function in S. salmonicida ...... 76 3.4.3 MROs of other diplomonads ...... 77 3.4.4 A hypothesis of hydrogenosome origin in Eukaryotes ...... 78 4. Conclusions and future perspectives ...... 80 5. Acknowledgements ...... 82 6. References ...... 85

Abbreviations

[FeFe] hydrogenase Iron-only hydrogenase ADI Arginine deiminase ASCT Acetate:succinate CoA-transferase ASE Allele specific expression ASH Allelic sequence heterozygosity ATP Adenosine triphosphate BI Bayesian inference CLO Carpediemonas-like organism ER Endoplasmatic reticulum EST Expressed sequence tag ESV Encystation-specific vesicle FPKM Fragments / kilobase /million reads HCMP High-cysteine membrane protein Hyd[EFG] Hydrogenase maturase [EFG] IMF Immunofluorescence LGT Lateral gene transfer MALDI Matrix-assisted laser desorption ionization ML Maximum likelihood MRO Mitochondrion-related organelle MS Mass spectrometry NEK Never in mitosis related kinase NO Nitric oxide OCT Ornithine carbamoyltransferase ROS Reactive oxygen species SDS-PAGE Sodium dodecyl sulphate- polyacrylamide gel electrophoresis SHMT Serine hydroxy methyltransferase PAS Polyadenylation signal PFL Pyruvate formate lyase PFOR Pyruvate:ferredoxin oxidoreductase VSP Variant-surface protein

Sammanfattning på svenska (Summary in Swedish)

Diplomonader är en grupp av organismer som återfinns i syrefattiga miljöer såsom i mag-tarmkanalen hos djur eller i bottenslam. Merparten av dessa organismer är parasiter som behöver en värdorganism för att överleva och sprida sig vidare till nästa värd. I denna avhandling har studier av arvsmassan och biologin hos framförallt två diplomonader, Giardia intestinalis och Spironucleus salmonicida, undersökts med en rad olika bioinformatiska och molekylärbiologiska tekniker. Giardia intestinalis orsakar giardiasis, en sjukdom hos människor och djur. Människor blir vanligtvis smittade genom att de dricker vatten som innehåller den sjukdomsframkallande cyst-formen. Att en människa (eller ett djur) får i sig så lite som 10-50 cystor kan leda till att en infektion inleds. Giardia koloniserar ytan av tunntarmen och utlöser en rad av symptom där diarré, magkramper och förlust av aptiten är vanliga inslag. Sjukdomen uppvisar dock väldigt stora skillnader i symptombilden och många människor kan bära parasiten utan att märka av det. I dagsläget är det inte känt om det är parasiten, värden eller en kombination av båda som avgör hur grava symptom som uppstår vid infektion. Längre ner i matsmältningskanalen leder en rad fysiologiska förändringar till att Giardia genomgår en utveckl- ingsprocess och bildar den infektiösa cyst-formen som sedan sprids i miljön via avföringen. I Sverige dokumenteras runt 1500 fall av giardiasis varje år. Merparten av dessa patienter blir smittade under en utomlandsvistelse. Tre hundra miljoner fall av giardiasis tros inträffa varje år runt om i världen, framförallt i länder där tillgången på rent vatten är begränsad. Spironucleus salmonicida orsakar en allvarlig sjukdom hos laxfiskar, i synnerhet hos Atlantlax. I motsats till Giardia kan Spironucleus också or- saka en utspridd infektion där vävnader och organ angrips. Detta leder vanligtvis till en hög dödlighet hos ovan nämnda fiskar. Väldigt lite forskning har bedrivits på Spironucleus och dess spridningsvägar fiskar emellan är inte kartlagda.

11 Genetiska studier av Giardia (Paper I, III & IV) Avhandlingen bygger till stora delar på jämförande analyser där egenskaper som är gemensamma eller unika mellan parasiter identifieras och analyseras. Giardia kan delas in i åtta olika grupper av organismer som kan infektera olika typer av värddjur, varav två grupper kan etablera infektion hos männi- ska. Även om det var känt sedan tidigare att dessa organismer är olika rent biologiskt så fanns det endast begränsat med information om vad som föran- leder dessa skillnader rent genetiskt. I den jämförande studien mellan två parasiter från dessa två människoinfekterande grupper (Paper I) samt ett djurisolat (Paper III) påvisades stora genetiska skillnader. Människoisolaten var 80% identiska på proteinnivå medan ett av människoisolaten var närmare släkt med djurisolatet med en identitet på 90%. Uppemot 10% av parasiter- nas proteiner var helt olika, där merparten utgjordes av proteiner som an- vänds för att undvika immunförsvaret. En samling med unika proteiner kunde påvisas som förmodligen överförts nyligen från bakterier som lever i magtarm-kanalen (Paper I & III). De stora, tidigare okända skillnaderna som observerats mellan de tre organismerna, speciellt i stora genfamiljer, skulle kunna vara en förklaring till hur Giardia-infektioner ger så olika symptom- bilder. Sexuell reproduktion är en vanlig strategi för att förbättra oddsen för överlevnad genom skapandet av nya varianter av organismer med blandade eller nya egenskaper. Tidigare observationer på genetisk nivå har indikerat att en sexuell cykel finns i Giardia, inget är dock känt om hur en sådan process skulle fungera. Genetiska spår av en förmodad sexuell process kunde påvisas i vårt studerade människoisolat (Paper I). Mer ingående studier av detta fenomen med ytterligare tekniker kunde senare påvisa att en sådan sexuell process inte verkar påverka den genetiska strukturen mellan tre av de åtta grupperna av Giardia (Paper IV). I detta hänseende är det därför troligt att de åtta grupperna, eller åtminstone delar av dessa, utgör separata arter och inte bara underarter.

Transkriptomstudier av Giardia (Paper VIII) Skillnader på gennivå förmedlas till proteinnivå via två processer i cellen som kallas transkription (DNA→RNA) och translation (RNA→Protein). Reglering av genaktivitet är viktigt för att Giardia ska kunna anpassa sig till sin omgivning under infektionen eller vid utvecklingen till cyst-formen. Vi studerade hur transkriptionen fungerar i fyra olika isolat; tre från människor och ett djurisolat. Transkriptionen från alla gener (som är aktiverade) i dessa isolat kan potentiellt studeras med hjälp av nya kraftfulla tekniker. Skillnader som tidigare observerats på gennivå stod också att finna i analysen av gener som aktiveras i transkriptionsprocessen. Hur gener regleras generellt i Giar-

12 dia är inte känt, men resultaten i denna studie påvisade en potentiellt viktig roll för regionen mellan stopkodonet och poly(A)-svansen i reglering av genaktivitet. De genetiska spåren av den sexuella processen som vi funnit i det tidigare analyserade människoisolatet gav i denna analys upphov till närmare 2000 proteiner med något förändrade egenskaper. En sådan upp- sättning med olika proteiner skulle kunna vara en fördel vid situationer som kräver stor flexibilitet hos parasiten som t ex vid anpassning till en ny värd- organism.

Miljöanpassning hos diplomonader (Paper II & VIII) Diplomonadgruppens organismer har egenskaper som är gemensamma, men varje enskild grupp har också utmärkande egenskaper. Alltsedan diplomona- dernas anfader levde för kanske en halv miljard år sedan har livsförutsätt- ningar och miljöer sannolikt förändrats åtskilliga gånger. Detta har resulterat i arter med vitt skilda egenskaper och anpassningar där gener har förlorats, vunnits och fått utrymme att förändras i funktion och antal. Skillnader mellan organismer kan uppstå snabbt som en konsekvens av att de återfinns i olika miljöer. S. barkhanus, en nära släkting till S. salmonicida, har studerats för att förstå vilka skillnader som kan finnas mellan organismer som till utseendet är helt oskiljbara från varandra. S. barkhanus är inte en parasit utan lever tillsammans med sin värdorganism (fisken harr) utan att framkalla några sjukdomssymptom. Trots att de två arterna av Spironucleus inte går att skilja åt till utseendet så uppvisar de lika stora genetiska skillnader (~80% identitet) som också tidigare dokumenterades mellan de två mänskliga Giardia isolaten. Mikrobiella arter som lever i samma miljö har upprepade gånger funnits ha närbesläktade gener utan att organismerna ifråga är nära släkt. Dessa gener tros ha spridits från sitt ursprung via så kallad horisontell genöverföring, en välkänd och väldokumenterad process hos bakterier. Detta arbete har vid åtskilliga tillfällen visat att diplomonader har anpassat sig genom att dra nytta av denna process. Gener inblandade i metabolism verkar vara den van- ligaste kategorin av gener som överförs, även om många av de överförda generna än så länge saknar en känd funktion.

Funktionella studier av diplomonader (Paper V, VI, VII) Många intressanta hypoteser kan formuleras med hjälp av genomik och tran- skriptionstudier. Det är dock först genom funktionella studier som modeller kan testas och utvärderas. Mycket av min tid och energi har ägnats åt att utveckla nya molekylära verktyg för Giardia (Paper V) samt ett helt nytt

13 transfektionssystem för S. salmonicida (Paper VI) där sådana verktyg tidigare saknades. I Giardia användes de nya verktygen för att rena fram proteiner som tidigare visats vara inblandade i parasitens strategi för att ta udden av immun- försvaret i tarmen. Komponenterna i en proteinmaskin i Giardia, prote- asomen, kunde bestämmas genom upparbetning och identifiering via masspektrometri. Det nyutvecklade transfektionssystemet i S. salmonicida (Paper VII) an- vändes för att studera lokaliseringen av ett antal proteiner i cellen. Genom att den subcellulära positionen hos ett protein är känt, ger det möjlighet att dra slutsatser om dess funktion. S. salmonicida tillhör nu, genom vårt arbete, en liten grupp av mikrobiella eukaryoter med ett sekvenserat genom samt ett robust transfektionssystem. Vi använde transfektionssystemet för att studera den mitokondriella res- ten i S. salmonicida. Alla eukaryota organismer, celler med en kärna för sitt genomiska DNA, har en mitokondrie av något slag. Mitokondrier är dome- sticerade bakterier och de flesta använder syre för att bryta ned socker till energi för cellulära aktiviteter. I avsaknad av syre, som i den miljö där Giar- dia och Spironucleus finns, har mitokondrien i många fall omvandlats till att producera vätgas vid energiproduktion (i hydrogenosomer) eller förlorat sin roll som cellens kraftverk (i mitosomer). Giardia är känd för att ha en mito- som medan ingen har undersökt denna fråga i Spironucleus. Analyser av arvsmassan i S. salmonicida visade att den skulle kunna ha en hydrogeno- som, något som kunde bekräftas med de molekylära verktygen vi utvecklat (Paper VII). Vår slutsats är att anfadern till den grupp av organismer som Giardia och Spironucleus tillhör kan ha varit den första organism som an- passade sig till syrefattiga miljöer via upptag av gener från bakterier, något som sannolikt hände för över en miljard år sedan.

Sammanfattningsvis har mina studier tillsammans med mina kollegors expertis bidragit till ökad förståelse kring diversiteten hos en evolutionärt och medicinskt intressant grupp av organismer, diplomonaderna. Informat- ionen om arvsmassan i Giardia har gjorts tillgänglig för andra forskare och har redan stimulerat vidare studier. Onekligen återstår mycket arbete innan vi kan förstå hur dessa organismer orsakar sjukdom hos sin värd. Mitt arbete har visat att vi kan öka förståelsen för diplomonaderna genom att använda en kombination av tekniker som både är datorbaserade och experimentella.

14 1. Introduction

Our perception of eukaryote diversity is often limited to the magnificent range of forms and features present in the few eukaryotic lineages (animals, plants, fungi, brown algae) that have evolved organismal complexity and are visible to the naked eye. In fact, there are around fifty additional groups of poorly studied single-celled eukaryotes that each shows similar degree of diversity as the above mentioned macroscopic groups. The majority of eukaryotic microbial groups remain virtually uncharacterized beyond ultrastructural descriptions and in some cases expressed sequence tag (EST) studies.

1.1 The eukaryotic cell Eukaryota is one of the three lineages of current life together with the archaea and bacteria (Fig. 1). The former two are mostly single-celled microscopic organisms, although some bacteria, like Streptomycetes and Myxo- bacteria, display colonial behavior reminiscent of multicellularity. Eukaryotes are characterized by the presence of a true nucleus, where the genomic DNA of the cell is housed, maintained and copied. The decoupling of transcription from translation offered by the eukaryotic nucleus allows additional control and regulation of gene expression. Eukaryotic cells also contain a complex endomembrane system that allows existence of subcom- partments with distinct chemistry and function within a single cell. The dynamic actin and tubulin cytoskeletons of many eukaryotes allow amoeboid movements and phagotrophic behavior. In fact, an attempted reconstruction of the common ancestor of eukaryotes recovers a highly complex heterotrophic cell (Fritz-Laylin et al. 2010).

1.1.1 Origins The origin of the eukaryotic cell has been a hot topic debated among re- searchers throughout the years. A multitude of different theories have been imposed to explain the fact that eukaryotic genes involved in information processing most closely resemble those found in archaea and that eubacterial origin can be assigned for many genes involved in cellular metabolism (Embley and Martin 2006). The most common theory holds that a protoeukaryote, which might have been an archaeon (Cox et al. 2008), engulfed

15 an α-proteobacterial cell which over time was established as an endosymbiont, the mitochondrion (Embley and Martin 2006). Transfer of genes from the endosymbiont to the host nuclear genome ensued and this resulted in the small size of the current mitochondrial genome. The mitochondrion endowed eukaryotic cells with increased energy-generating capabilities which allowed an increase in organism complexity and size compared to prokaryotes. As we soon will see, not all extant eukaryotes conform to the standard blueprint described above.

Figure 1. Eukaryote phylogeny of the five recognized supergroups with major subgroups according to (Adl et al. 2012). Colored circles show the presence of mitochondria (red), hydrogenosomes (blue) and mitosomes (green) within the respective subgroups, adapted from (Shiflett and Johnson 2010).

1.1.2 The demise of the Archezoan hypothesis The last 10-15 years have profoundly altered the view of how mitochondrial acquisition happened in eukaryotes. Electron microscopy studies had failed to reveal mitochondria in several groups of eukaryotic organisms. These so called Archezoans, microbial eukaryote lineages found in oxygen-free environments, clustered at the base of the eukaryotic tree in molecular phylogenies and were suspected to have diverged before the assimilation of the mitochondrion into ‘higher’ eukaryotes. The first crack in the Archezoan hypothesis was the discovery that genes of mitochondrial origin existed in En- tamoeba (Clark and Roger 1995), one of the Archezoan lineages. This discovery was soon followed by the recognition of secondary-loss of mitochondria in all Archezoan groups (microsporidia, diplomonads, parabasalids) (Germot et al. 1996; Germot et al. 1997; Roger et al. 1998). Subcellular lo-

16 calization soon established the presence of highly derived organelles of mitochondrial descent in Archezoan organisms (Mai et al. 1999; Tovar et al. 1999; Embley et al. 2003; Tovar et al. 2003). The stubborn retention of mitochondrion-related organelles (MROs) in eukaryotes is attributed to the synthesis of Fe-S clusters, a pathway universally conserved in investigated MROs. The loss of the ancestral Isc-type and gain of a bacterial Nif-type Fe- S cluster biogenesis system that also operates in the MROs of the Archamoebae Entamoeba and Mastigamoeba underscores the functional importance of this process for the organelle (van der Giezen et al. 2004; Maralikova et al. 2010). Interestingly, studies of the mitochondrion of Sac- charomyces cerevisae indicated that Fe-S cluster synthesis is the sole pathway indispensable for cell survival (Lill and Kispal 2000).

1.1.3 Hydrogenosomes: discovery and origins Hydrogenosomes are organelles that take part in energy-generation through substrate level phosphorylation, with a characteristic production of molecular hydrogen as a byproduct. Hydrogenosomes were discovered in the parabasalids, Tritrichomonas foetus and T. vaginalis in the 1970’s (Lindmark and Muller 1973; Lindmark et al. 1975). The organelles are surrounded with double membranes just as mitochondria, but lacks cristae. Hydrogenosomes usually lack an organellar genome, but exceptions to this are known (Perez- Brocal and Clark 2008; de Graaf et al. 2011).

Origins The origin of hydrogenosomes was for a long time enigmatic and early theories suggested a separate origin from the mitochondrial acquisition, possibly from Clostridia (Whatley et al. 1979). Not long thereafter, hydrogenosomes were described from several other organisms living in oxygen-depleted environments. These organisms were phylogenetically well separated from Pa- rabasalia and included organisms such as chytrid fungi and certain lineages of ciliates. The patchy occurrence of hydrogenosomes (Fig. 1) was suggested to be evidence that the protoeukaryote was a facultative anaerobe having both aerobic and anaerobic capabilities within its mitochondrion, and that differential loss of either machinery occurred during specialization to specific niches. The anaerobic mitochondria of some insects and algae that respire with alternative electron acceptors under anoxic conditions were thought to be intermediates of this process. The hypothesis generated from these discoveries, the differential loss model, obviated the need for additional endosymbiotic events to explain the current phylogenetic patchiness of hydrogenosomes. Surprisingly, studies of diversity in ciliates revealed closely related lineages that were either aerobic or anaerobic, from within lineages that were otherwise aerobic. In addition, chytrid fungi were found to utilize pyruvate formate lyase (PFL) in pyruvate catabolism rather than py-

17 ruvate:ferredoxin oxidoreductase (PFOR) that is used in T. vaginalis and Giardia (Akhmanova et al. 1999). These studies suggested polyphyletic origins of hydrogenosomes in ciliates and chytrid fungi and questioned the differential loss model. A new model was needed to explain these unexpected results. At this time, lateral gene transfer (LGT) was increasingly being recognized as a major force in the adaptation of an organism to new environments. Molecu- lar studies of several hallmark enzymes of anaerobic pyruvate metabolism, PFOR and [FeFe]-hydrogenase, characteristic of most hydrogenosomes had failed to recapitulate accepted eukaryotic relationships, showing that genes had been inherited horizontally between eukaryotes and perhaps multiple times from bacterial donors (Hug et al. 2010). Along the same lines, evolutionary history reconstruction of the previously mentioned PFL enzyme has revealed that the gene was probably acquired by an eukaryote from a firmicute bacterium (Hampl et al. 2009). The patchy eukaryotic distribution was suggested to be the result of eukaryote-to-eukaryote lateral gene transfer events (Stairs et al. 2011). Interestingly, the hydrogenase maturases HydE, HydF and HydG, associated with high level hydrogen production, appear to display eukaryotic monophyly (Hug et al. 2010). There is no evidence of an α-proteobacterial origin of these genes as no extant α-proteobacterial genome encodes all five hallmark enzymes of hydrogenosomes (Hug et al. 2010). To summarize, the current picture holds that hydrogenosomes are modified mitochondria and that genes for anaerobic pyruvate metabolism in eukaryotes have been acquired from bacteria by lateral gene transfer allowing adaptation to anaerobic conditions. The sequence of events and which donors and acceptor lineages involved mostly remains undetermined.

Metabolism Hydrogenosomes partake in the anaerobic oxidation of pyruvate to acetate with the concomitant production of ATP and release of molecular hydrogen. The description of anaerobic pyruvate metabolism given here is mainly based on the well characterized system of T. vaginalis, but several avenues or solutions to this metabolism exist in other eukaryotes, as previously mentioned. This pathway is restricted to anaerobic or microaerophilic conditions due to the exquisite oxygen sensitivity of the hallmark enzymes PFOR and [FeFe]-hydrogenase. Pyruvate is decarboxylated by PFOR to carbon dioxide and acetyl-CoA. The electrons abstracted by PFOR are transferred to the iron-sulfur protein Ferredoxin. The electrons shuttled by Ferredoxin are utilized by [FeFe]-hydrogenase to produce molecular hydrogen by reduction of protons, yielding reoxidized Ferredoxin ready for another run of the catalytic cycle. The acetyl-CoA produced by PFOR can then be utilized for energy generation by one of several pathways ending with acetate as an end- product. T. vaginalis utilizes a hydrogenosomal pathway comprising Ace-

18 tate:succinate CoA-transferase (ASCT) coupled with Succinate thiokinase and Adenylate kinase to generate ATP with acetate as the end product. Ace- tyl-CoA synthetase (ADP-forming) is thought to be utilized in the hydrogenosomes of some ciliates as well as in the cytoplasm of Giardia and Entamoeba to generate ATP from ADP (van Grinsven et al. 2008). Organisms that have or are suspected to have hydrogenosomes are often difficult to study due to difficulties of culturing (Blastocystis, chytrid fungi) or genetic intractability (Mastigameoba). To the exception of T. vaginalis, there are no organisms equipped with hydrogenosomes and sequenced genomes that are amenable for transfection. As such, biochemical studies or the presence of N-terminal pre-sequences alone on proteins have mostly been utilized to interrogate the composition and capabilities of these organelles.

1.2 Excavata The excavates (Fig. 1) is a proposed group of eukaryotes that include several taxa (diplomonads, parabasalids, retortamonads, oxymonads, Malawimonas heterolobosea and euglenozoa) some of which contain medically important organisms. The initial proposal for excavates as a major eukaryotic group was based on morphological data. Excavate organisms are heterotrophs that utilize a longitudinal groove to collect food particles stirred up by beating from a posterior flagellum (Simpson 2003). As it happens, this feature has been frequently lost or remodeled throughout evolution and is not a common trait of all extant excavates. Many excavate taxa were previously thought to be “deep-branching” basal eukaryotes diverging before the emergence of the “crown-group” eukaryotes (Simpson 2003). This view has now been largely discredited with the emergence of more sophisticated phylogenetic tools, awareness of methodo- logical shortcomings and greater taxon sampling. Initial phylogenetic analyses by single genes placed several members of Excavata in different places of the eukaryotic tree. To put this in perspective, many characterized members of excavate taxa display extremely long branches in molecular phylogenies (Giardia, Trichomonas) that might lead to artefactual attraction to the long branches of bacterial outgroups or to other eukaryotic lineages (Micro- sporidia, Entamoeba) that have experienced increased evolutionary rates (Brinkmann et al. 2005). Excavata is generally recovered as several separately monophyletic groups. Phylogenetic analyses robustly recovers the two superphyla Discoba (Jakobida, Euglenozoa, Heterolobosea) and Metamonada (Preaxostyla, Pa- rabasalia, Fornicata; Fig. 1) but the presence of suspected long-branch arti- facts have frustrated attempts to recover monophyly for the entirety of Ex- cavata (Parfrey et al. 2006; Hampl et al. 2009). If fast-evolving sites or taxa

19 are removed in large phylogenomic analyses rich in excavates, monophyly can be obtained with moderate support (Hampl et al. 2009). With the removal of all the long-branched metamonads this support increases. Curiously, all metamonads live in anaerobic or microaerophilic environments and organelles reminiscent of aerobic mitochondria are missing (Adl et al. 2012). The parabasalids are known to harbor hydrogenosomes. Forni- cata includes several important groups of pathogens, including retortamonads and diplomonads, as well as free-living organisms such as Car- pediemonas (Fig. 2).

Figure 2. Phylogeny of Fornicata. Phylogenetic reconstruction of diplomonads, Carpediemonas and Retortamonas employing HSP90. Numbers show bootstrap values. The bodyplan of different diplomonads, showing single or double karyomastigont are indicated to the right. Picture reprinted from Fig. 2.1 of the book Anaerobic Parasitic Protozoa (Horizon Press).

1.3 Fornicata Members of the group Fornicata lack classical mitochondria and are found in oxygen-deprived milieus. Fornicata consists of “Carpediemonas-like” organisms (CLOs) and Eopharyngia, with the former being free-living and the latter mostly being endobiotic or parasitic. Fornicata has been expanded recently to encompass a series of newly discovered CLOs (Takishita et al. 2007; Yubuki et al. 2007; Park et al. 2009; Kolisko et al. 2010; Park et al. 2010) that make up a series of increasingly basal branches towards the near- est outgroups of Metamonada (Takishita et al. 2012). The CLOs all appear to possess MROs that superficially resemble hydrogenosomes of parabasalids, indicating that not all fornicatan organisms might have undergone the drastic reduction of mitochondria seen in Giardia.

20 Eopharyngia emerges at the tip of the CLO lineages and are mostly found in the oxygen limited gut environment of animals. The free-living members within Eopharyngia have likely readapted to this niche from an host- associated organism (Kolisko et al. 2008). Eopharyngia contains the group Diplomonadida as well as the probably polyphyletic retortamonads (Cepicka et al. 2008). Most diplomonads have a highly unusual cell organization with two karyomastigonts. Diplomonads usually have a recurrent flagellum that passes through a cytopharyngeal tube. Within the diplomonads, there are two subdivisions, the Giardiinae and Hex- amitinae. The Giardiinae (Giardia, Octomitus) have lost the cytopharynx with the recurrent flagella passing through the cellular cytoplasm. The Hex- amitinae retain a functional feeding apparatus and employ an alternative genetic code, using UAA and UAG to encode glutamine and leaving UGA as the only stop codon (Keeling and Doolittle 1996; Keeling and Doolittle 1997). The members of Hexamitinae include the parasites or commensals Hexamita and Spironucleus as well as the free-living Trepomonas. Spironu- cleus will be introduced in section (1.9). Curiously, there are also diplomonads that only have one karyomastigont and resemble one half of a typical diplomonad cell. These cells, collectively termed ‘enteromonads’ were previously grouped together and were seen as a separate fornicatan branch representing the type of primitive state that diplomonads evolved from. However, recent phylogenetic analyses reject the monophyly of this group with ‘enteromonads’ emerging within Hexamitinae in separate events of cellular reduction (Kolisko et al. 2008).

1.4 The Giardia cell The discovery of Giardia intestinalis, also known as Giardia lamblia and Giardia duodenalis, is attributed to the “father of microbiology”, Antony van Leeuwenhoek (Dobell 1920). In a 1681 letter to Robert Hooke of the Royal Society van Leeuwenhoek described the observation “animalcules” in his own diarrheic stool. He wrote :-

“..I have at times seen very prettily moving animalcules, some rather large, others somewhat smaller than a blood corpuscle, and all of one and the same structure. Their bodies were somewhat longer than broad, and their belly which was flattened, provided with several feet, with which they made such a movement through the clear medium and the globules that we might fancy we saw a pissabed running up against the wall. But although they made a rapid movement with their feet, yet they made but slow progress.” (Dobell 1920).

This description along with the circumstances of his observations is consistent with the “animalcules” being representatives of the Giardia vegeta-

21 tive stage, the trophozoite (Fig. 3). Indeed, modern work with single-lens microscopes of the same magnifying power lend credence the discovery of Giardia more than 330 years ago (Ford 2007). The rediscovery of the organism by Lambl in 1859 introduced the name Cercomonas intestinalis (Adam 2001). The current genus name Giardia was introduced by Kunstler in the early 1880’s. The next 100 years saw the introduction of a large number of species of Giardia, owing to the belief that each species of animal infected with Giardia harbours its own parasite species (Adam 2001).

1.4.1 Giardia classification and assemblages Classification of Giardia entered a new era with the work of Filice in 1952. Morphological differences in the median body structures of Giardia isolates allowed him to differentiate between three species, G. muris (from rodents), G. duodenalis (from humans and other mammals) and G. agilis (from amphibians). Additional work by electron microscopy have proposed three more species; G. psittaci (from psittacine birds), G. ardeae (from herons) and G. microti (from voles), based on morphological differences of the ventral disc, ventrolateral flange and the cyst (Table 1) (Adam 2001). Newly introduced molecular biology techniques were applied to the classification of G. intestinalis, the Giardia species that infects humans, from the middle of the 1980’s and onwards. Molecular analyses demonstrated the presence of three groups of G. intestinalis that infects humans, with two of the groups showing large genetic differences that the authors suggested to be worthy of species designation. These two genetic groupings, assemblage A and B, have further been recognized as displaying biological differences. The pathogenicity of assemblage B (GS/M isolate) was greater than that of assemblage A (Isr isolate) in experimental human infections (Nash et al. 1987) and assemblage B isolates typically grow slower than assemblage A isolates in in vitro cultures (Karanis and Ey 1998). In total, genetic typing of diverse Giardia intestinalis isolates has resulted in six additional assemblages (C-H) being recognized (Table 1). In contrast to assemblage A and B parasites, both of which have been regarded as having zoonotic potential, these assemblages display a higher degree of species specificity. Assemblage C and D infects dogs; E infects hoofed-animals; F infects cats; G infects rats; H infects marine mammals. The synonymic use of Giardia intestinalis (mainly in Europe), Giardia lamblia (the Americas) and Giardia duodenalis (Aus- tralia) has together with the newly discovered genetic substructure ignited a debate for revision of the G. intestinalis nomenclature, with a recent call for revision of the genus (Monis et al. 2009). The authors suggest the name G. duodenalis for assemblage A, G. enterica for assemblage B along with species names for 4 additional assemblages (Table 1). Throughout this thesis I will use the assemblage designations to indicate G. intestinalis genetic

22 groups in keeping with current practice in the literature. Giardia will also be used to denote G. intestinalis unless otherwise indicated.

Table 1. Recognized Giardia species, G. intestinalis assemblages, proposed species names and their respective host-ranges. Adapted from (Monis et al. 2009). Proposed species Species Genotype Host name G. intestinalis Assemblage A G. duodenalis many mammals, including humans Assemblage B G. enterica many mammals, including humans Assemblage C, D G. canis canids (dogs) Assemblage E G. bovis hoofed animals Assemblage F G. cati cats Assemblage G G. simondi rats Assemblage H marine mammals G. muris rodents G. agilis amphibians G. microti rodents G. ardeae herons

1.5 The Giardia life cycle The ability to leave the cell-cycle and enter into a differentiation program that results in the formation of a refractory stage, a cyst, is a widely practiced strategy to survive unfavorable environmental conditions in both free-living and parasitic organisms. In Giardia, it is the vegetative trophozoite which transforms into a dormant cyst by a simple two-stage life cycle without the need for an intermediate host (Fig. 3). The Giardia differentiation program involves a tightly orchestrated disassembly of the cytoskeleton and the coordinated deposition of a fluid-phase outer casing that hardens within seconds to form the fibrillar cyst wall. The protection offered by the cyst wall enables Giardia to survive for long periods of time in the environment or to withstand chemical treatment that would inactivate most organisms.

1.5.1 The trophozoite The Giardia trophozoite has the shape of a half-pear during proliferation in the small intestine of an infected individual. In keeping with being a diplomonad cell, the trophozoite displays two equally shaped nuclei with a bilat- erally symmetric body plan (Fig. 3). Each cell measures around 10-15 µm x 5 µm and carries four distinct pairs of flagella for different modes of motility (Fig. 3). The most prominent feature of the trophozoite is the presence of an

23 attachment organelle, the ventral disc, which allows the parasite to adhere to various surfaces.

Figure 3. Giardia trophozoite and cyst. A Giardia trophozoite is schematically drawn to the left. Prominent features of the trophozoite include two nuclei each with two copies of the genome, the ventral disk for attachment, eight basal bodies and their corresponding flagellar pairs (CF-caudal flagella, PF-Posteriolateral flagella, VF- ventral flagella, AF- anterior flagella) for motility, peripheral vesicles close to the membrane, a median body and mitosomes that are found in a linear arrangement between the nuclei as well as scattered in the cytoplasm. As the trophozoite undergo encystation the ventral disc fragments, flagella are disassembled and the cell is en- cased in an ovoid cyst wall. The mature cyst contains four nuclei each containing four copies of the genome as a result of two rounds of genome replication, where the last one occurs without division of the nuclei. The cyst transforms back into the vegetative stage through the excystation process, giving rise to four trophozoites.

The nuclei The individual nuclei of the trophozoite contain two copies (2N) of the haploid genome (Fig. 3) (Bernander et al. 2001). Both nuclear copies in each

24 nucleus are replicated (4N) and then segregated at cytokinesis during each round of the cell cycle (Yu et al. 2002). Despite the apparent similarity of the two nuclei, a number of differences between them have been noticed. Asyn- chronous initiation and completion of replication has been described (Wiesehahn et al. 1984) along with morphological differences of the nuclei and variation in the number of nuclear pores (Benchimol 2004). The number of nuclear pores appears to vary throughout the cell cycle with a lower abundance during mitosis. In addition, the snoRNA precursor (GLsR17) to the miR2 microRNA in Giardia has been shown to primarily localize to a subnuclear region in one of the nuclei (Saraiya and Wang 2008). The subnuclear organization of the Giardia nuclei is not well known, but a nucleolus-like subcompartment in the anterior of the nuclei can be dis- cerned by electron microscopy (Jimenez-Garcia et al. 2008). Several markers characteristic of the nucleolus (16S rRNA, Fibrillarin and CBF5) have been demonstrated to co-localize to the anterior part of nuclei (Li and Wang 2005; Jimenez-Garcia et al. 2008).

The cytoskeleton The elaborate cage-like cytoskeleton of Giardia can be seen in detergent extracted cells (Dawson 2010). The cytoskeleton is mainly microtubule based but also incorporates Giardia expanded gene families such as the α- giardins, Nek kinases, ankyrin repeat proteins, striated-fiber assemblins (SALP-1, β-, δ-giardin) (Palm et al. 2005) as well as novel proteins such as the ventral disc protein γ-giardin (Nohria et al. 1992). The highly divergent Giardia actin functions without many of the canonical actin-binding proteins and in the absence of myosin (Morrison et al. 2007). The actin cytoskeleton in Giardia has recently been characterized, providing evidence of discrete filamentous structures associated with the cortex, nuclei and the flagellar axonemes (Paredez et al. 2011). The caudal flagellar axonemes are bundled by a helix of actin and C-shaped fibers have been observed to tether the anterior flagellar axonemes. Actin appears to play a role in protein trafficking during encystation as well as in cytokinesis and endocytosis (Paredez et al. 2011). The ventral disc of Giardia (Fig. 3) is regarded as a virulence factor since the trophozoite needs to attach to the epithelium to avoid being eliminated from the small intestine via peristalsis. The ventral disc is a concave array of spiraling microtubules that are perpendicularly joined by microribbon structures on the cytoplasmatic side of the disc (Dawson 2010). The spirals are joined at the tips and create a circular area devoid of cytoskeleton proteins. The lip-like ventrolateral flange, lateral shields, lateral crest and bare area sequentially make contact with the substrate during attachment (House et al. 2011). The knock-down of Median body protein, a disc protein, was shown to impair attachment by interrupting the formation of a domed configuration of the disc (Dawson 2010).

25 There are additional microtubule based cytoskeletal structures in trophozoites which remain enigmatic. The median body is a microtubule array of symmetrically organized stacks which give rise to the crooked smile of some Giardia trophozoites (Fig. 3). The median body is a dynamic structure whose appearance is coordinated with the cell cycle (Sagolla et al. 2006). Regulatory kinases such as phosphorylated aurora kinase and ERK1 kinase are dynamically associated components of the median body (Ellis et al. 2003; Davids et al. 2008). The structure has been suggested to be the source of microtubules for assembly of the daughter disc before mitosis or to be an organizer region for components destined for the disc (Crossley et al. 1986). The rib-cage like funis has currently no known function but has been proposed to participate in a movement called dorsal-tail flexation, whereby cell detachment is achieved (Ghosh et al. 2001).

Flagella and basal bodies The flagella are used by Giardia to find a site suitable for colonization. In addition, flagellar motility is needed for completion of cytokinesis and perhaps entry into or exit out of dormancy. The eight flagella of Giardia (“9+2” microtubule arrangement) are organized into four pairs (ventral flagella - VF, anterior flagella - AF, posteriolateral flagella - PF, caudal flagella - CF) that differ in their beating pattern, length of the axoneme and ancilliary structures (Fig. 3) (Dawson 2010; Dawson and House 2010). The flagella emanate from eight basal bodies positioned between and slightly anterior to the nuclei (Fig. 3) (Benchimol 2005).The basal bodies appear to be signaling centers that coordinate signaling during mitosis (Aurora kinase) (Davids et al. 2008) and excystation (Calmodulin, Protein kinase A, Protein phosphatase 2A) (Manning et al. 2011). An intricate scheme for trans-generational maturation and migration of the giardial flagella has been proposed (Nohynkova et al. 2006), where each newly divided trophozoite will carry four newly duplicated and four parental flagella. The vane-bearing ventral flagella (VF) beat in an undulating wave-form that has been proposed to provide a current for feeding and perhaps provide a negative pressure below the disc for attachment (Holberton 1973). However, cells that express a dominant negative form of α2-giardin, a protein of the ventral disc and flagella, display a decreased flagellar waveform and only miniscule defects in attachment (House et al. 2011). Altogether the four flagellar pairs contribute to directional and tumbling movement dependent on the attachment status of the cell. The caudal flagellar complex and the funis might be responsible for dorsal-tail flexation required to break the seal of the disc and accomplish detachment.

Membrane trafficking The Giardia membrane trafficking system is very different from a typical eukaryote by being reduced and adapted to the parasitic life style. The troph-

26 ozoite possesses an endoplasmatic reticulum (ER) that is contiguous with the nuclear membrane as in other eukaryotes. The ER permeates the cell and is employed for sorting and maturation of proteins as no conventional Golgi compartment is present in trophozoites. There are few Rab GTPases (Morrison et al. 2007) and SNAREs (soluble N-ethylmalemide-sensitive factor attachment protein receptors) (Elias et al. 2008) indicating a restricted number of distinct organelle or membrane domains in the parasite. Giardia uses a very rudimentary system for glycosylation of proteins, adding N-acetyl glucosamine(s) (GlcNAc1-2) via Asn- (Robbins and Samuelson 2005) and potentially via O-linkage (Banerjee et al. 2009), without encoding an ER quality control system. The N-glycome of Giardia has been investigated by lectin-affinity chromatography followed by proteomics and exhibits differentiation induced changes (Ratner et al. 2008). There are no peroxisomes or lysosomes described in Giardia. Instead, the dorsal cell periphery and the bare area of the ventral disc are lined with peripheral vesicles (PVs; Fig. 3) which engage both in endocytosis from the extracellular milieu and in exocytosis of secretory cargo and cellular debris. PVs have the capacity to become acidified and acquire lysosomal characteristics with hydrolase and protease activities. PVs are in communication with the ER and endocytosed material is selectively transported into the cell inte- rior. Proteins are targeted for the secretory pathway by the presence of N- terminal signal peptides, or as in the case of VSPs, additionally via signals in the semi-conserved C-terminal region, and transit to the plasmamembrane via ER. Secretion of metabolic enzymes via non-conventional export has also been demonstrated (Ringqvist et al. 2008).

The mitosome Giardia was previously seen as a primitively amitochondriate organism; a vestigial cell that diverged from other eukaryotes before the acquisition of the α-proteobacterial endosymbiont (described further in section 1.1.2). This belief was questioned by the discovery of nuclear encoded genes of mitochondrial ancestry (Chaperonin 60 and IscS) in the Giardia genome at the turn of the 20th century (Roger et al. 1998; Tachezy et al. 2001). The discovery in Giardia of mitosomes, tiny (140 ×70 nm) double-membraned organelles which contained enzymes (IscU and IscS) and functioned in the synthesis of Fe-S clusters, proved that Giardia harbored mitochondrial remnant organelles and was not primitively amitochondriate (Tovar et al. 2003). There are between 25-100 mitosomes in each Giardia cell, with a prominent linear arrangement of central mitosomes along the caudal basal bodies between the nuclei as well as peripheral mitosomes distributed randomly in the cytoplasm (Regoes et al. 2005) (Fig. 3). The import of proteins to mitosomes is in some cases dependent on N-terminal pre-sequences (IscU, Ferredoxin, IscA, Glutaredoxin) while other proteins are received with internal signals.

27 N-terminal pre-sequences are removed by the single-subunit Giardia processing peptidase that has evolved to recognize the endogenous Giardia pre- sequence substrates (Smid et al. 2008). The mitosomal import machinery is extremely reduced compared to other eukaryotes and very few components have been identified. The recognized components are the outer membrane porin Tom40 and inner membrane translocases Pam16 and Pam18 (Dolezal et al. 2005; Jedelsky et al. 2011). Protein import and homeostasis proteins Chaperonin 60 and 10 as well as mtHSP70 with co-chaperons GrpE and Jac1 have been demonstrated to localize to mitosomes (Regoes et al. 2005; Smid et al. 2008; Jedelsky et al. 2011). The Fe-S machinery consists of Ferredox- in, IscU, IscS, IscA and Glutaredoxin together with the alternative Fe-S scaffold Nfu (Smid et al. 2008; Jedelsky et al. 2011). No energy metabolism or membrane potential have been detected in giardial mitosomes and PFOR and [FeFe]-hydrogenase which would enable ATP synthesis through substrate phosphorylation are not identified in mitosome enriched fractions (Emelyanov and Goldberg 2011; Jedelsky et al. 2011). A NADPH- dependent mitosomal oxidoreductase, GiOR, has been reported but the natural electron acceptor remains unknown (Jedelsky et al. 2011). Mitosomal replication and intercommunication with other compartments are poorly understood. However, Elias and colleagues demonstrated a mitosomal-like localization of the SNARE protein Sec20 (Elias et al. 2008) and Jedelsky and coworkers identified a mitosomal C-tail anchored protein, VAMP- associated protein, in their proteomics study of mitosomes (Jedelsky et al. 2011).

1.5.2 Differentiation Giardia differentiation entails the transition from a vegetative trophozoite via the process of encystation into the dormant cyst, and upon re-entry into a host the awakening from dormancy and emergence of the vegetative stage via a process called excystation. The completion of the life cycle in vitro by Gillin and co-workers (Gillin et al. 1989) makes Giardia one of the few parasite systems where the life cycle can be entirely completed outside the host.

Encystation Encystation is the process that enables Giardia to differentiate from an ac- tively growing organism into a dormant stage. The process entails a drastic restructuring of the morphology and metabolism of the parasite to enable transmission to the next host. Encystation naturally occurs when trophozoites experience the conditions in the distal small intestine or jejunum (Gillin et al. 1987). Cholesterol starvation alone is sufficient to induce encystation (Lujan et al. 1996). The in- clusion of a higher pH to simulate physiological conditions of the distal small intestine is used in protocols to accomplish in vitro encystation. Low

28 cholesterol is usually achieved by high amounts of bile in the encystation medium, but the use of lipid-depleted serum has also been employed and is equally effective for this purpose (Lujan et al. 1996; Morf et al. 2010). It is not known how Giardia senses cholesterol starvation and how signaling works to induce the downstream events that lead to encystation. However, the downstream effects that result from these events are much more characterized. Cells preferentially enter encystation in the G2 phase of the cell cycle, where the cell has replicated cellular structures and the nuclei (Reiner et al. 2008). The transcription factor Myb2 (Sun et al. 2002), which appears to be the encystation master regulator, induces transcription of cyst wall proteins 1-3 (CWPs) and genes in the pathway to synthesize carbohydrate monomers (UDP-GalNAc) for the Giardia cyst wall. Comparative transcriptome analyses of encystation employing two different ways of inducing encystation have revealed a surprisingly small number of 18 core genes that are upregulated during early encystation (up to 7 hrs) (Morf et al. 2010), including CWPs, Myb2 and genes for carbohydrate synthesis. The Giardia cyst wall is composed of ~60% carbohydrate and ~40% protein. The inducible and allosterically synthesized carbohydrate monomers are assembled into giardan (β(1-3)-N-Acetyl-D-galactosamine polymer), the unique carbohydrate of the Giardia cyst wall, by the activity of “cyst wall synthase” (Karr and Jarroll 2004). Chatterjee et al. demonstrated that CWP1 display lectin-activity towards the unique GalNAc homopolymer of the Gi- ardia cyst wall (Chatterjee et al. 2010). Elegantly, they utilized CWP1 as a probe to show that the GalNAc homopolymer is manufactured in small foci distinct from ESVs and deposited early on the encysting cell surface. Cyst wall protein transcripts are massively upregulated and accumulation of the corresponding proteinaceous cyst wall material (CWM) at ER exit sites induces the de novo formation of encystation-specific vesicles (ESVs) (Faso et al. 2012). The ESVs are Golgi-like cisternae that undergo maturation from individual vesicles in early encystation 8 hours post induction (p.i) to become a trans-Golgi-like tubular vesicular network during late encystation (Stefanic et al. 2009). A number of post-translational modifications, disulfide bonds, isopeptide-linkages and addition of phospho-groups, within and between CWP proteins have been demonstrated to be important for proper cyst wall assembly (Reiner et al. 2001; Slavin et al. 2002; Davids et al. 2004). Recently, Konrad and coworkers showed that CWM segregates into two phases with different mobilities within ESVs between 12 and 16 hours p.i concurrent with the proteolytical processing of the a short piece of the CWP2 C-terminus (Konrad et al. 2010). The inner condensed core, with low mobility, consists of CWP3 together with most of the CWP2 C- terminus. CWP1 and the N-terminus of CWP2 remain in a fluid state that dynamically exchanges between the trans-Golgi vesicles close to the parasite surface (Konrad et al. 2010). Secretion of the fluid CWM leads to rapid polymerization (within minutes) to create the first layer of the cyst wall. At

29 this stage the cyst attains it final shape as the ventral disk and flagella are internalized and disassembled. The condensed core is secreted over the space of several hours to create the final water resistant cyst form (Konrad et al. 2010). Apart from assembly of the cyst wall and disassembly of the cell cytoskeleton, Giardia undergoes another round of genome replication without nuclear division to give a ploidy of 16N in the mature cyst (Bernander et al. 2001).

The cyst The cyst represents the dormant and transmissive stage of the Giardia life cycle. Giardia cysts are refractile, nonmotile and measure 8-12 µm by 7-10 µm. The cyst is ovoid with a 0.3-0.5 µm thick cyst wall outside dual plasma membranes, has four prominent nuclei with highly condensed chromatin, ribbon-like flagellar axonemes and disassembled pieces of the ventral disc (Adam 2001) (Fig. 3). The cyst keeps large stores of glycogen to sustain basal energy levels during dormancy. Cysts are shed with stool and can remain infectious for many months in cold water.

Excystation Upon entry into the digestive tract of a host a series of consecutive stimuli triggers the awakening from dormancy. This incompletely understood process called excystation allows reentry into the replicative stage of the life cycle. The first cue for excystation is contact with acidic digestive juices of the stomach followed by the slightly alkaline and protease-laden environment of the small intestine. These events trigger the exocytosis of pre- manufactured parasite components (hydrolases, acid phosphatases and proteases) that act on the cyst wall. The rupture of the cyst at one pole is rapidly followed by the emergence of a flagellated short lived pleiomorphic parasite stage (the excyzoite). The excyzoite rapidly goes through a first round of cytokinesis, followed by a second round to give rise to four colonization competent trophozoites. The emergence of four trophozoites from one cyst partly explain the very low infectious dose of Giardia (10-50 cysts) (Rendtorff 1954).

1.6 Cell biology and metabolism of Giardia At one time, cells were viewed as bags of enzymes, where random reactions of freely diffusing protein species and their substrates governed the inner workings of life. The realization that macromolecular machines, or protein complexes, are involved in just about every major process of the cell has replaced this archaic view of how our cells are organized (Alberts 1998). The list of sophisticated assembly-line style protein complexes is long and

30 includes examples such as the splicing-machinery, the proteasome and the ribosome. The study of protein complexes was for a long time a time- consuming and complicated task employing different types of chromato- graphic separation techniques necessitating monumental amounts of starting material. Affinity chromatography has since simplified many purification endeavors, as many affinity tags enable near native conditions to be maintained throughout the procedure. Proteome scale protein complex purification was pioneered in the yeast S. cerevisiae (Gavin et al. 2002) by the development of the tandem affinity purification (TAP) system (Rigaut et al. 1999; Puig et al. 2001). This system and other approaches have since been applied to many more organisms, including multicellular systems such as the fruit-fly Drosophila melanogaster (Veraksa et al. 2005) and the nematode Caenorhabditis elegans (Zanin et al. 2011). One of the major conclusions of the Giardia genome project was the doc- umentation of a general reduction in the number of components of many molecular machines (Morrison et al. 2007). In principle, this could be caused by several factors; stream-lining of processes resulting from a parasitic life- style, primitive absence, replacement of components by novel lineage- specific proteins, our inability to recognize highly diverged proteins or a combination of all the above.

1.6.1 Cellular metabolism One of the tenets for any living organism is the need to generate or obtain energy. The energy is then utilized to maintain cellular homeostasis and to synthesize cellular building blocks. The range of chemical building blocks includes diverse chemicals such as nucleotides for nucleic acid synthesis and fatty acids for membrane lipids. To this end, a significant portion of the cellular machinery is devoted to maintaining the steady supply of cellular building blocks for cell survival and proliferation. Being a parasite, Giardia has a severely restricted assortment of metabolic pathways and is reliant on the host for lipids and on salvage pathways for pyrimidine and purine synthesis. Influx of genes encoding metabolic proteins from bacterial and archaeal donors has had a major impact on giardial metabolism and cellular detoxification systems (Morrison et al. 2007).

Energy metabolism Most eukaryotes rely on aerobic respiration and oxidative phosphorylation via their mitochondria to generate ATP. In contrast, Giardia has adapted to microaerophilic conditions where oxygen, the electron acceptor of oxidative phosphorylation, is absent or severely restricted. As such, Giardia relies on glycolysis and substrate level phosphorylation to generate ATP (Adam 2001). The main activities of these processes as well as hallmark enzymes of intermediate metabolism, PFOR and [FeFe]-hydrogenase, are non-

31 sedimentable (Lindmark 1980) or membrane associated, and are not found in the MRO as in Trichomonas (Emelyanov and Goldberg 2011). Glucose is the preferred sugar for energy generation yielding predominantly alanine, ethanol, acetate and CO2 as metabolites depending on oxygen availability (Adam 2001). The arginine dihydrolase (ADH) pathway, found in many prokaryotes, but rare in eukaryotes, an exception being Trichomonas, is a major source of energy in Giardia (Schofield et al. 1990). The decomposition of arginine via the concerted action of four enzymes; arginine deiminase (ADI), ornithine carbamoyltransferase (OCT), carbamate kinase and ornithine decarboxylase, results in the generation of four times more ATP from equimolar amounts of glucose (Schofield et al. 1990; Schofield et al. 1992). Arginine is imported into the cell by an arginine-ornithine antiporter (Knodler et al. 1995). The enzymes ADI and OCT are released by trophozoites into the intestinal lumen upon host-parasite interaction (Ringqvist et al. 2008), a process thought to deprive epithelial cells of arginine (Stadelmann et al. 2012), the substrate for production of the giardicidal compound nitric oxide (NO).

Amino acid, lipid and nucleotide metabolism Giardia has a very limited ability to synthesize amino acids de novo, and instead recovers them from the extracellular milieu either by active transport or diffusion. There exists a high demand for cysteine, due to the abundance of this amino acid in VSPs. Cysteine doubles as a potent reducing agent, and gives partial protection from oxygen induced damage. Early experiments using radiolabeled precursors to phospholipids and fatty acids revealed a lack, or a very limited capability of, de novo synthesis in Giardia (Jarroll et al. 1981). Recent analyses of the Giardia complement of phospholipids (Yichoy et al. 2009) and the completion of the genomes has shifted this view, indicating that re-modeling of or even de novo synthesis of some lipids occurs (Yichoy et al. 2011). Sphingolipid metabolism is differentially regulated in cell differentiation and critical for the encystation process (Sonda et al. 2008). Nucleotide metabolism in Giardia is limited to salvage pathways both for obtaining pyrimidine and purine nucleosides (Adam 2001). The lack of ribo- nucleoside reductase necessitates the salvage of deoxynucleotides via deox- ynucleosidases (Laoworawit et al. 1993).

Oxygen detoxification

Giardia is a microaerophilic organism that tolerates low levels of O2 and reactive oxygen species (ROS). It is also highly susceptible to nitrosative stress (Lloyd et al. 2003). Recently, it was proposed that the small intestine and its high redox buffering capacity might be one of the reasons why Giar- dia colonizes this particular niche (Mastronicola et al. 2011), despite the relatively high levels of O2 that is present at the muscosal interface. One

32 reason for the O2 sensitivity is the activity of the giardial DT-diaphorase that has been found to generate hydrogen peroxide, a damaging ROS, under aerobic conditions. Another reason is that prominent enzymes of intermediary pyruvate metabolism in Giardia (PFOR and [FeFe]-hydrogenase) are both very sensitive to O2, becoming inactivated at the levels encountered in the intestine. In addition, Giardia lacks catalase, superoxide dismutase and the glutathione-system, conventional ROS-scavenging activities in most cells. To counter this, Giardia detoxification of ROS and O2 relies upon a system of thioredoxin-like proteins together with several other enzymes, NADH oxidase (Brown et al. 1998), flavodiiron protein (Di Matteo et al. 2008; Vicente et al. 2009; Mastronicola et al. 2011), flavohemoglobin (Mastronicola et al. 2010; Rafferty et al. 2010) and superoxide reductase (Testa et al. 2011), all likely acquired from a prokaryotic donors by LGT.

1.6.2 Sex in Giardia? Giardia has traditionally been seen as a clonally propagating organism, as diplomonads have been regarded as being asexual (Adam 2001). The eukaryotic common ancestor (LECA) is believed to have been endowed with a sexual cycle, although not all extant eukaryotes are thought to have retained the trait (Cavalier-Smith 2002). The arrival of the genome data from Giardia revealed the presence of homologs to genes typically involved in meiosis in other eukaryotes (Ramesh et al. 2005; Malik et al. 2008). The Giardia WB genome displays very low levels of allelic sequence heterozygosity (ASH) (Morrison et al. 2007), contrary to what would be expected for an asexual organism without any way of controlling the build-up of differences between the nuclei. Evidence for recombination within assemblage AII isolates from an endemic Peruvian setting (Cooper et al. 2007), has provided evidence for sexual outcrossing in Giardia. Around the same time, evidence was presented that nuclei can fuse and exchange episomal plasmids during encystation (Poxleitner et al. 2008). This process, termed ‘diplomixis’ by the authors, was recently confirmed to allow exchange of chromosomally integrated markers with high frequency (Carpenter et al. 2012). In one experiment, the starting clonal population consisted of 76% of cells that displayed a FISH labeling spot in one individual nucleus. One round of encystation, excystation and growth to confluency resulted in trophozoites with a single spot (49%), no spots (31%) and 20 % with two spots (one in each nucleus) (Carpenter et al. 2012). Further, the former study demonstrated that several homologs of meiosis-related genes are expressed and localize to the nucleus at this stage of the life-cycle. Diplomixis can potentially by recombination and/or gene conversion reduce the ASH between the nuclei. Genetic variation within individual Giardia trophozoites and cysts has been demonstrated (Ankarklev et al. 2012), most often signified by double peaks in sequencing chromatograms. The presence of ASH appears to be

33 common in assemblages B, C, D and E whereas A, F and G appear to have much lower amounts (Lebbad et al. 2010). Interassemblage recombination has been suggested to be low or absent entirely based on population-based studies, suggesting that assemblages are reproductively isolated (Takumi et al. 2012). However, some studies have shown signs of interassemblage recombination (Teodorovic et al. 2007; Lasek-Nesselquist et al. 2009). The single-cell cloned assemblage B isolate 12c14B was reported to show six different copies of gdh and five copies of mlh (Lasek-Nesselquist et al. 2009). It is not known if this represents erroneous gene determinations, gene duplications or the result of aneuploidic events. These observations are ex- plainable in a recently published model whereby the fusion of trophozoites is followed by genome reduction through loss of chromosomes until tetra- ploidy is reached (Andersson 2012). During this time, the genome might undergo shuffling through the action of diplomixis and intranuclear gene conversion and/or recombination. The 12c14B genotyping results, evidence of aneuploidy in Giardia (Tumova et al. 2007) and the results from other diplomonads with complicated genomes (Joint Genome Institute 2012) might all be different facets of the same phenomenon of diplomonad sex (Andersson 2012). The fact that ASH is more prevalent in certain assemblages could also indicate that this process is not operational to the same extent across G. intestinalis. The observation that transfected episomal DNA is integrated into the genome of the GS isolate, but not in the WB isolate, might be relevant to this issue (Singer et al. 1998).

1.7 Giardia pathogenesis The burden of diarrheal disease around the globe is immense; 2 billion cases are estimated to occur annually (WHO 2009). Even though bacteria and viruses are responsible for a majority of these episodes, Giardia is regarded as the premier contributor from the pathogenic protozoa. The annual numbers of Giardia infections have been estimated to be 280 million (Lane and Lloyd 2002). The vast majority of these cases are found in low-income countries, where the parasite is endemic. In recognition of this, Giardia has been included in the ‘WHO Neglected Diseases Initiative’, a collection of diseases clearly linked to poverty and hindrance to development in the localities where they occur (Savioli et al. 2006). Giardia is however a cosmopolitan pathogen, being prevalent around the world. Giardia is most common in the age group 0-4 years and exhibits a seasonal pattern in incidence, with a higher number of cases in late summer and early fall (Yoder et al. 2012). In 2009, Sweden reported 1210 cases of giardiasis, equivalent to a incidence of 13 infections per 100 000 inhabitants (ECDC 2011).

34 1.7.1 Giardiasis In classical experiments conducted on volunteers (prison inmates), Rendtorff fulfilled Koch’s third postulate and established Giardia as a causative agent of diarrheal disease in humans (Rendtorff 1954). A more stringent fulfill- ment of all Koch’s postulates for human infectious agents was attained by Nash and colleagues (Nash et al. 1987). These experiments also showed that only 10-50 cysts are enough to establish infection. Giardia infection of a host is triggered when cysts are ingested directly via the fecal-oral route, through contaminated food or drinks. The cysts are triggered to excyst by the stimuli imparted during the passage through the digestive tract, and colonization of the small intestine is initiated by the trophozoite. Symptoms of acute disease appear 1-2 weeks after trophozoite colonization is initiated, and coincide with the appearance of cysts in fecal samples. Typical symptoms of giardiasis include diarrhea, flatulence, abdominal dis- tension, intestinal cramps, nausea and malabsorption. The stool is often described as greasy and frothy but might be watery as well, in which case trophozoites can be seen. In immune competent individuals, the acute phase gives way to a chronic infectious phase in 30-50% of cases. This is often signified by lingering diarrhea, malaise and malabsorption. Cyst excretion while carrying an asymptomatic Giardia infection appears to be common. In children, Giardia infection is often the cause of “failure to thrive” diagnoses and it may be the cause of stunting and dietary deficiencies. The induction of chronic gastrointestinal disorders such as irritable bowel syndrome (IBS) as well as chronic fatigue has been documented following giardiasis (Wensaas et al. 2012). Human giardiasis is only caused by parasites of assemblage A and B, both of which are potentially zoonotic. Attempts of attributing symptoms with these or subtypes thereof have often given conflicting or inconclusive results. Recently, an association between physical symptoms (flatulence in infected children) and infection with assemblage B was established (Lebbad et al. 2011).

1.7.2 Host-parasite interactions The ability to withstand and control infections is a fundamental process in any organism. Being a non-invasive parasite that operates at the mucosa- epithelial interface, a niche rife with noxious chemicals, specialized shock- troops in the form of immune cells and antibody guided missiles, Giardia surely employs strategies to dodge this onslaught and establish infection. The response of the host involves several distinct phases and immune associated processes.

35 Pathophysiology and disease mechanism Infection by Giardia normally results in low levels of inflammation with few infiltrating immune cells. Diffuse shortening of microvilli has often been observed in the small intestine of infected individuals. Increased rates of apoptosis or repression of enterocyte proliferation could both be responsible for this observation. Increased apoptosis induction has indeed been detected in duodenal biopsies from infected individuals (Troeger et al. 2007) as well as during in vitro interactions with human cell lines (Panaro et al. 2007). Increased leak flux and chloride hyper-secretion have been suggested to be factors of chronic diarrhea in giardiasis (Troeger et al. 2007). The reduced surface area induced by shortened microvilli might be responsible for malabsorption or nutrient deficiencies. Impairment of disaccharidase activity in the small intestine following infection with Giardia has been observed in a mouse model of giardiasis (Solaymani-Mohammadi and Singer 2011) as well as in carriers on chronic giardiasis (Troeger et al. 2007). The decline in activity in the former study was dependent on the Giardia strain and upon induction of CD4+ and CD8+ T cells (Solaymani-Mohammadi and Singer 2011). Curiously, this effect was only seen with the GS isolate and no change was observed with the WB isolate. The effect in this case was not attributed to direct damage to the epithelium but rather through an indirect effect that could involve transcriptional or post-translational modifications. A direct effect on the host F-actin cytoskeleton from G. intestinalis attachment via parasite lipid-raft micro domains has recently been demonstrated (Humen et al. 2011). The authors suggested that brush-border enzymes become delocalized, trans-epithelial resistance is decreased and tight junction proteins are rearranged as a result of attachment. Inhibition of enterocyte growth via host-cell deprivation of arginine has also been suggested to affect tissue homeostasis to the benefit of parasite colonization (Stadelmann et al. 2012). The multifactorial causes of giardiasis has been recently been highlighted (Cotton et al. 2011).

Giardia responses during infection The response of the parasite whilst interacting with human cells has mainly been studied in vitro; no expression data has been obtained from naturally infected humans. That said, interaction with human epithelial cells using an in vitro system induced either up- or down-regulation of over 200 genes (Ringqvist et al. 2011; Ma'ayeh and Brook-Carter 2012). These genes included cysteine proteases, HCMPs, VSPs and proteins implicated in attachment. Giardia has been demonstrated to release toxic compounds either con- stitutively or upon contact with human cells. The Giardia secretome includes a menagerie of proteins that include VSPs (Papanastasiou et al. 1996), proteases (Jimenez et al. 2000; Rodriguez-Fuentes et al. 2006; de Carvalho et al. 2008) as well as probable moon-lighting proteins (Ringqvist et al. 2008;

36 Skarin et al. 2011) that already have an established role in the parasite but may be additionally involved in pathogenesis. The latter category includes ornithine carbamoyltransferase (OCT) and arginine deiminase (ADI) of the ADH pathway as well as enolase and EF-1α, all four of which are immunodominant during giardiasis in humans and mice (Palm et al. 2003; Davids et al. 2006).

Gut microflora Studies of interactions between Giardia and the host intestinal flora is in its infancy, but an intriguing observation of co-occurrence between Giardia infection and Helicobacter pylori has been noticed (Ankarklev et al. 2012). Singer and Nash noticed that pathogenesis in mice upon infection with GS/M-H7 isolate depended on the microflora of the host (Singer and Nash 2000). Mice with identical genetic backgrounds were either susceptible or refractory to Giardia challenge, an effect found to be transferable between animals. The exact strain responsible for this effect was not determined, but Lactobacilli, a component of the synthetic microflora of refractory animals, have been suggested to antagonize H. pylori colonization (Kabir et al. 1997; Singer and Nash 2000).

1.7.3 Antigenic variation The ability to persist at the infection site or have the ability to reinfect a host is a common theme among pathogens. One of the most successful strategies displayed by diverse infectious agents is to fool or circumvent the immune system of the host. This can range from the inherent ability to produce a distinct variant organism by antigenic drift as in the influenza virus, by coat- ing oneself with host-derived proteins as in helminths or the deployment of discrete surface molecules of different antigenicity. The last strategy, antigenic variation, is employed by some bacteria (Neisseria and Borrelia) and eukaryotic parasites such as Plasmodium falciparum and trypanosomes (Deitsch et al. 2009). Not only limited to pathogenic organisms, antigenic switching of cysteine-rich proteins in ciliates in response to environmental stimuli has been studied extensively (Simon and Schmidt 2007). The presence of antigenic variation in G. intestinalis was described in 1988 by the Nash lab both in vitro (Adam et al. 1988) and in vivo (Aggarwal and Nash 1988; Nash et al. 1990). The first Variant-specific surface protein (VSP; TSA417) was cloned and sequenced by the legendary parasitologist P. Hagblom two years later (Gillin et al. 1990). The Variant-specific surface proteins (VSPs) that undergo antigenic variation were found to be cysteine- rich and to be part of a multigene family. The VSPs are type I integral membrane proteins that coat the entire surface of the trophozoite, including the flagella and ventral disc (Pimenta et al. 1991). VSPs are continuously recy- cled from the cell surface by shedding into the extracellular milieu, being

37 potentially cleaved at the C-terminus to release the soluble ectodomain (Papanastasiou et al. 1996).

Features of VSPs VSPs have several distinct features that allow their assignment to this class of proteins. All are cysteine-rich with a high frequency of Cys-X-X-Cys motifs. An N-terminal signal peptide directs VSPs to the trophozoite surface. The central parts of VSPs are variable, with a lot of genes having tandem repeated sequence motifs (Fig. 4) (Adam et al. 2010). The sequence GGCY, as well as zinc finger motifs are found in many VSPs, but the significance of these characters has not been elucidated. The C-terminal part of VSPs has a characteristic transmembrane domain and terminates with the almost invariant pentapeptide CRGKA exposed to the cytosol (Fig. 4). The CRGKA motif has been implicated in proper delivery to the plasma membrane in one study (Marti et al. 2003), but found to be dispensable in another (Touz et al. 2005). Differences in the ectodomains used in these studies are likely to be the cause of these differing results. Post-translational modifications of the pentapeptide, palmitoylation of the cysteine (Papanastasiou et al. 1997) and citrullination of the arginine residue (Touz et al. 2008), have been described and been implicated in segregation into lipid-rafts or as a switching signal, respectively (Fig. 4).

Genomic context of VSPs All five Giardia chromosomes contain VSP genes, with rare instances of telomeric positioning. The Giardia WB genome encodes at least 303 vsp genes, of which 228 are complete (Adam et al. 2010). The sizes of vsp genes are highly variable, ranging from 222 bp to almost 6.8 kbp. VSPs are often found in linear arrays, which in some examples can consist of at least 10 vsp genes. The vsp genes are often found as inverted or tandem pairs of identical or very similar genes, which have made genome assembly challenging (Adam et al. 2010).

Figure 4. VSP structure and important regions. The variable and semi-conserved domain contain numerous C-X-X-C motifs and some also contain a GGCY motif. A conserved transmembrane region terminates in the invariant C-terminal pentapeptide sequence CRGKA. The cysteine and the arginine residues can be post-translationally modified by palmitoylation (Papanastasiou et al. 1997) and citrullination (Touz et al. 2008) respectively. The VSP transmembrane region is derived from TSA417.

Switching and regulation of VSP expression Switching of VSPs takes place on average every 6-13 generations in vitro even in the absence of any immunological pressure (Nash et al. 1990). This

39 finding questioned the relevance of antigenic variation for the induction of persistent and recurring infection and suggested additional functions in Gi- ardia biology. However, the decoupling of antigenic variation in experimental infections has demonstrated its importance as an adaptive mechanism for dodging the immune system of the host (Rivero et al. 2010). A pathway for initiating VSP switching and avoidance of antibody mediated cytotoxicity by relaying signals from the surface has been suggested (Touz et al. 2008). These events are hypothesized to lead to citrullination of the arginine in the CRGKA motif by the metabolic protein ADI (Fig. 4). How this modification would induce switching remains to be shown. The VSP switching mechanism does not appear to be reliant on alteration of the coding sequence or DNA rearrangement as in trypanosomes (Kulakova et al. 2006), even though evidence of recombination amongst individual members appears common. During the process of switching, old and new VSPs are detected at the surface of the trophozoite simultaneously over an extended period of time (Nash et al. 2001). The switching/regulation of VSP expression has been suggested to be effected by epigenetic alterations of the chromatin structure (Kulakova et al. 2006) and/or via a mechanism involving components of the RNA interference pathway (Prucca et al. 2008). Knock- down of Giardia Dicer (GiDicer) or RNA dependent RNA polymerase (GiRdRP) has been shown to decouple antigenic variation with the entire VSP repertoire being exposed on the surface of the parasite (Prucca et al. 2008). A post-transcriptional mechanism responsible for this observation has been proposed (Prucca et al. 2008), but a lack of GiDicer induced RNAi of long double-stranded RNAs and no slicing activity by the Giardia Argo- naute (GiAgo) protein questions this model (Li et al. 2012). Moreover, a regulatory network comprised of microRNAs (miRNAs) has been proposed to contribute to the selective translational inhibition of VSP mRNAs (Saraiya and Wang 2008; Li et al. 2011; Saraiya et al. 2011; Li et al. 2012). The miRNA candidates discovered so far are designated snoRNA-derived small RNAs (sdRNAs), owing to their biogenesis from snoRNAs by the action of GiDicer (Li et al. 2012). The mature miRNAs are immunoprecipi- tated with GiAgo, the potential effector of translational repression, which has affinity towards the cap of mature mRNAs (Saraiya and Wang 2008). The five recognized miRNAs have the combined capability to potentially regulate the expression of 178 VSPs, sometimes by cooperative or non- cooperative mechanisms (Li et al. 2012). Interestingly, the miRNA target sites are located in the 3’ part of the coding sequences and in the 3’UTRs where VSPs show the highest amount of conservation (Adam et al. 2010). The exact process that allows switching by this translational repression mechanism remains to be defined.

40 1.7.4 Treatment Giardia infection is normally treated by the administration of a 5- nitroimidazole, typically metronidazole or tinidazole, which shows broad and specific effects towards the parasite’s anaerobic metabolism. The 5- nitroimidazoles are in fact prodrugs which are activated when they encounter proteins of the parasite’s metabolism that have low enough reduction potential to activate them. The activated form of metronidazole induces cytotoxic effects in the parasite by release of radicals that damage cellular components. Classically, PFOR and Ferredoxin have been implicated in this activation due to their low redox potential, but evidence is mounting that there are other pathways capable of this activation as well (Leitsch et al. 2011). The nitrothiazolide nitazoxanide and the benzimidazole albendazole are other potent giardicidal compounds employed for treatment of giardiasis. Nita- zoxanide is activated in the parasite by the nitroreductase GlNR1, a protein down-regulated in nitazoxanide resistant parasites (Nillius et al. 2011). This protein has also been implicated in metronidazole activation in one study (Pal et al. 2009). Albendazole is a potent antihelminithic agent that binds to tubulin and prevents the polymerization of microtubules. Resistance to albendazole in vitro is associated with complex alterations in cell physiology and metabolism (Paz-Maldonado et al. 2012). Increasing numbers of Giardia isolates have been reported to display resistance to the commonly adminis- tered drugs and have prompted the search for new treatment options with the help of compound library screening (Bonilla-Santiago et al. 2008; Chen et al. 2011; Faghiri et al. 2011), discovering several highly potent giardicial compounds.

1.8 The G. intestinalis genome and transcriptome The haploid genome size of G. intestinalis is close to 12 Mb and is organized into five chromosomal linkage groups. The chromosomes are flanked by telomeric repeats similar to those of other eukaryotes. The first sequenced Giardia genes gave the impression of a GC-rich genome, but the WB genome sequence instead revealed a GC content of 49% (Morrison et al. 2007). Giardia possesses the canonical histones found in other eukaryotes, except histone H1 which appear to be missing (Yee et al. 2007). The study of the genome structure and architecture in Giardia was pioneered using pulsed-field gel electrophoresis (PFGE) (Adam et al. 1988) and revealed differences in size of individual chromosomes within and between G. intestinalis isolates. Such size differences have been attributed to the frequently recombining telomeric regions and differences in copy number of rDNA arrays (Le Blancq and Adam 1998). Evidence of aneuploidy have been suggested in individual Giardia cells based on cytogenetic analyses

41 (Tumova et al. 2007), with the most common karyotype differing between members of assemblages A and B. Chromosomal maps of two chromosomes have been determined for the assemblage A1 isolate BRIS/83/HEP/106 by probe mapping (Upcroft et al. 2009; Krauer et al. 2010). Efforts to improve genome continuity to chromosome scale were performed during the WB genome project by mapping of BAC clones. These chromosomal charts excluded the highly variable sub- telomeric repeat regions, which have been analyzed separately. Recently, chromosome-wide maps were established by optical mapping of the genome project strain (Perry et al. 2011). The results resolved some missassemblies from the genome project and indicated that the actual genome size is 12.1 Mb, in close agreement with sequencing data and PFGE analyses. The major discrepancy was an underestimation of the size of chromosome 5, the largest of the Giardia chromosomes. Chromosome five contains a 819 kb gap in the optical map, much of which is likely to consist of rDNA repeats (Perry et al. 2011).

1.8.1 Transcriptional landscape The regulation of transcription in Giardia has to contend with the simultane- ous and synchronous control of two nuclei, both which appear to be transcriptionally active (Kabnick and Peattie 1990). Most genes are found in very gene-rich regions of the genome, leaving small intergenic regions for transcriptional control.

Transcription Giardia relies upon short A/T-rich promoters without an identifiable TATA- box to initiate transcription (Adam 2001). The 5’UTRs of Giardia are short, usually between 0 and 14 nt, and ribosome scanning is absent. The homolog of TATA-binding protein (TBP) found in the Giardia genome is extremely divergent and there seems to be no homolog of transcription factor IIB (Best et al. 2004). Twenty-one out of 28 proteins of the eukaryal RNAPI, RNAPII and RNAPIII complexes have been found in Giardia, with six of the seven missing components being polymerase specific (Best et al. 2004). Moreover, only four of the 12 general transcriptions factors were identified (Best et al. 2004). The number of proteins having transcription factor-associated domains as well as factor diversity is low in Giardia; the genome encodes only 9 out of 29 classes of transcription factor motifs found across eukarya (Iyer et al. 2008). This pattern of reduced transcription factor diversity is typical of parasitic protists, as evidenced by the reduction of core particle and specific transcription factors in kinetoplastids and microsporidia (Iyer et al. 2008). The few transcription factors that have been described have mainly been investigated for their role in parasite differentiation (Sun et al. 2006;

42 Wang et al. 2007; Huang et al. 2008; Pan et al. 2009; Wang et al. 2010; Su et al. 2011; Chuang et al. 2012). Giardia mRNAs have blocked 5’-ends suggestive of a cap structure, and the presence of a 5′-m7G cap stimulated expression of a transgenic construct in vivo (Hausmann et al. 2005). However, the presence of a cap does not appear to be universal across Giardia mRNAs (Yu et al. 1998; Knodler et al. 1999). Giardia also encodes two paralogous Tgs enzymes that are involved in producing a 2,2,7-trimethylguanosine (TMG) cap, likely by sequential addition of methyl groups by the two enzymes (Hausmann and Shuman 2005; Hausmann et al. 2007; Benarroch et al. 2010). The TMG cap can be found on a subset of small nuclear RNAs in Giardia (RNAs A-H) (Niu et al. 1994). RNA D, also known as GLsR17, is an orphan nucleolar RNA (Yang et al. 2005) that is the source of the miR2 microRNA (Saraiya and Wang 2008). Two distinct Eukaryotic translation initiation factor 4E have been identified in Giardia, one binds to the TMG cap whereas the other one binds the m7G cap (Li and Wang 2005). Polyadenylation is accomplished by a rudimentary system that is thought to recognize the Giardia polyadenylation-sequence (PAS), AGTPuAAPy, followed by addition of the A-tail ap- proximately 10 bp downstream of the PAS (Adam 2001).

Introns Splicing was for a long time though to be absent from Giardia, but the presence of homologs of the splicing-machinery and the eventual discovery of an intron in a ferredoxin gene changed this dogma (Nixon et al. 2002). This discovery was followed by additional descriptions of cis-introns in a total of five genes (Russell et al. 2005; Morrison et al. 2007; Roy et al. 2012). The identified 5’ and 3’ intron borders show extended characteristic motifs, with the 3’ motif containing the branch-point nucleotide. Interestingly, the 5’ and 3’ intron motifs are also well conserved in T. vaginalis (Vanacova et al. 2005). Recently, in independent efforts, two groups described the presence of trans-splicing in Giardia (Kamikawa et al. 2011; Nageshan et al. 2011). HSP90 and a dynein motor protein were found to be expressed as full-length gene products despite the presence of two separate coding regions for the N- terminal and C-terminal of the proteins. Full-length mRNAs were shown to be produced by trans-splicing from two mature polyadenylated pre-mRNAs produced from the separate loci. Interestingly, the motifs governing the splicing were identical to those employed in cis-splicing and extended regions of complementarity upstream and downstream of the splice-site allowed intermolecular positioning to complete the splicing-event (Kamikawa et al. 2011; Nageshan et al. 2011). Extensive base-complementarity is also evident in some of the longer cis-spliced introns in Giardia, indicating a possible route to the evolution of trans-spliced genes (Roy et al. 2012). An- other trans-splicing event that involves the very same dynein motor protein

43 mentioned previously has since been described, again the pre-mRNAs showed extensive complementary base pairing, but unusually for Giardia employing AT-AC splice-sites (Roy et al. 2012). Spliceosomal RNAs have recently been identified in Giardia and were found to exhibit an unusual mixture of motifs of the major and minor spliceosomal small nuclear RNAs. A processing motif shared between trans-spliced and non-coding RNAs was additionally defined (Hudson et al. 2012). The low density of introns in Giardia is unlikely to reflect an ancient state as the related organisms Carpdiemonas membranifera (Simpson et al. 2002), T. vaginalis (Carlton et al. 2007) as well as the oxymonad excavate Streblo- mastix strix (Slamovits and Keeling 2006) seem to carry more introns in their genomes.

RNA homeostasis The presence of pervasive transcription in eukaryotic genomes has been noticed in several organisms, including Giardia (Elmendorf et al. 2001; Teodorovic et al. 2007). A large proportion of sterile transcripts (up to 20%) were detected in strand-specific, polyA+ cDNA libraries prepared from both WB and GS cells (Elmendorf et al. 2001). The authors proceeded to validate three antisense transcript by 5’RACE, strand-specific northern blot and RNase protection assays, implicating cryptic promoters and polyadenylation sites coupled with loose transcriptional control as contributing factors (Elmendorf et al. 2001). It should however be said that the authors defined sterile transcripts as any RNAs having >300 bp UTRs, as lengths above this was unprecedented in Giardia at the time (Elmendorf et al. 2001). In a later study, based upon the data generated in the Giardia SAGE project, the global abundance of antisense transcripts was studied in the WB isolate (Teodorovic et al. 2007). Antisense transcripts were detected at ~50% of transcribed loci, a consequence of inherently bidirectional promoters according to the authors. Another possible source of the abundance of antisense RNAs in Giardia is inefficient degradation and increased half-life of transcripts. While Giardia appears to harbor all canonical pathways for RNA degradation seen in other eukaryotes, some complexes like the exosome (3’- 5’ degradation) show a simplified configuration (Williams and Elmendorf 2011).

Transcriptome studies in Giardia Gene expression studies in Giardia have been performed to some extent. Microarray analyses were conducted on early encysting cells (Morf et al. 2010), during host-parasite interaction (Ringqvist et al. 2011) and in the analysis of drug resistant isolates (Muller et al. 2007). In addition, a serial analysis of gene expression (SAGE) project has analyzed gene expression changes during 10 stages of Giardia differentiation in vitro (Birkeland et al. 2010). These analyses have revealed the differential regulation of hundreds

44 of genes at some part of the life-cycle, involving canonical eukaryotic pathways or proteins as well novel diplomonad or Giardia specific proteins.

1.8.2 Gene families The Giardia genome consists of several uniquely expanded gene families, three of which are described in this section, whereas the VSPs are described in section 1.7.3 Antigenic variation.

Ankyrin repeat proteins – Protein 21.1 The ankyrin repeat proteins (often annotated as Protein 21.1) is the largest gene family in Giardia. The Giardia WB genome contains 436 genes with at least one ankyrin repeat, with 320 of these lacking a protein kinase domain. Ankyrin repeats are common in eukaryotes and bacteria and typically partake in forming protein-protein interactions (Li et al. 2006). The ankyrin repeat proteins in Giardia are diverse in size, from 7 to 313 kDa, and in number of repeats. The vast majority of ankyrin repeat proteins in Giardia have not been characterized. Recently, 21 ankyrin repeat proteins were identified in a proteomic study of the Giardia ventral disc, with 10 of the proteins showing subcellular localization to the disc and/or associated structures using GFP-tagging (Hagen et al. 2011).

Nek kinases The Giardia WB genome was found to contain 276 protein kinases, representing 41 out of 63 of the ancient kinase families (Morrison et al. 2007). The 80 proteins in the Giardia core kinome is the most compact known for any eukaryote (Manning et al. 2011). The Never in Mitosis Gene A-Related Kinase (Nek) family has been massively expanded in the Giardia genome, accounting for 197 of the protein kinases. Curiously, a large proportion (137) of these lacks catalytic residues and appear to be pseudokinases (Morrison et al. 2007). The significance of this finding is unsure as pseudokinases sometimes can function in signaling as a scaffold or kinase target, a prominent feature seen with some Toxoplasma rhoptry kinases (Reese and Boyle 2012). Nek kinases function in control of cell cycle entry and in control of flagellar length in cilliated organisms, having been differentially expanded in ciliates and excavates. The Nek kinases in Giardia are commonly paired with ankyrin repeat domains, 67% of Nek kinases have an N-terminal kinase domain followed by varying numbers (1-26) of ankyrin repeats (Manning et al. 2011). Four catalytically active Nek kinases have been localized by epitope- tagging to various cytoskeletal structures, whereas one that was predicted to be inert, was found in the cytoplasm (Manning et al. 2011). An additional two have been localized to the lateral crest of the Giardia ventral disc (Hagen et al. 2011).

45 High cysteine membrane proteins (HCMPs) The high cysteine membrane proteins (HCMPs) is a diverse family of cysteine-rich proteins sharing certain characteristics with the VSP gene family. Both VSPs and HCMPs have abundant Cys-X-X-Cys motifs, whereas Cys- X-Cys is almost exclusively found in HCMPs (Davids et al. 2006). VSPs and HCMPs are both predicted to be type I transmembrane proteins, but HCMPs lack the CRGKA cytoplasmic pentapeptide. The transmembrane regions of HCMPs bear some resemblance to VSPs but the putative tails, have varying size and composition (Davids et al. 2006). Domain analysis was used to classify the 61 HCMPs on the WB genome into nine classes. One HCMP, the high cysteine non-variant cyst protein (HCNCp), is the only one that has been studied (Davids et al. 2006). The expression of the protein is increased during encystation, it is rerouted from the perinuclear region via ESVs and finally localizes to the membrane beneath the cyst wall and to the cell body in cysts (Davids et al. 2006). Ringqvist et al. noticed prominent differential regulation of HCMPs during host-cell interaction experiments, highlighting important functions of this class of proteins (Ringqvist et al. 2011).

Giardins The giardins are a heterogenous gathering of cytoskeletal proteins. There are four types of giardins; α- , β- , δ- and γ-giardin. The β- and δ-giardin proteins are members of the striated fiber (SF)-assemblin family, are predominantly found in the ventral disc (Palm et al. 2005; Jenkins et al. 2009). The γ- giardin protein is a novel Giardia protein, without any known homologies, found in the ventral disc (Nohria et al. 1992). The α-giardins are highly diverged annexin homologs, which have been expanded in Giardia. The gene family contains 21 members, some of which are clustered on the chromosome, likely as a result of gene duplication events (Weiland et al. 2005). Annexins normally interact with phospholipids in a Ca2+ dependent manner and act as a link between the cytoskeleton and lipid membranes. The α-giardins where initially recognized as prominent ~30 kDa proteins in cytoskeletal preparations of Giardia (Crossley and Holberton 1983). The first identified α-giardin was α1-giardin, also known as trypsin-activated Giardia lectin (taglin), a surface exposed protein that possesses Ca2+ dependent lectin activity upon trypsin activation (Farthing et al. 1986; Lev et al. 1986). Unusually among annexins, α1-giardin disengages from phospholipids at high Ca2+ concentration, implicating a role of this protein in emergence from dormancy where calcium-signaling is critical (Weeratunga et al. 2012). The immunodominant character of α1-giardin (Palm et al. 2003) has heralded its use as a basis for a Giardia vaccine (Jenikova et al. 2011). Many α-giardins localize to the flagella, plasma membrane or other parts of the cytoskeleton (Szkodowska et al. 2002;

46 Weiland et al. 2005; Vahrmann et al. 2008; Saric et al. 2009) and display differential binding affinities to phospholipids, indicative of the basis of their functional diversification.

1.9 Spironucleus Spironucleus are hexamitid flagellates found in the digestive tracts of animal hosts, either as parasites or commensals. Spironucleus infections have been recorded in mice and rodents (Spironucleus muris), turkeys and partridges (Spironucleus meleagridis) and amphibians (Spironucleus elegans) where they cause the disease spironucleosis. Several species of fish are also known to be infected by organisms of the genus Spironucleus. Notably, the “hole- in-the-head” disease affecting some species of cichlid ornamental fish has been suggested to be the result of infection by Spironucleus vortens (Paull and Matthews 2001).

1.9.1 Taxonomy and host association The species designation among some of the hexamitid flagellates has been complicated by the necessity for at least electron microscopy (Poynton et al. 2004) or preferentially molecular methods for species determination (Jorgensen and Sterud 2007). The slight morphological difference between Spironucleus and Hexamita, the former having kidney-shaped nuclei and the latter round ones, have resulted in redescriptions when going beyond light microscopy for classification (Poynton et al. 2004). The hexamitid group additionally contains interspersed members which are free-living (Hexamita and Trepomonas), as well as the monokaryotic ‘enteromonads’. Phylogenet- ic studies have suggested that there might be three groups of Spironucleus having separate terrestrial, fresh-water and marine origins (Jorgensen and Sterud 2007). However, increased sampling of organisms has questioned this division (Kolisko et al. 2008). In the early 1990’s, the aquaculture industry in Norway and Canada experienced outbreaks signified by mass mortality and large losses of consuma- ble fish (Kent et al. 1992; Poppe et al. 1992). The causative agent was found to be a hexamitid flagellate, at that time described as Hexamita salmonis. The disease was signified by systemic infiltration of most tissues including the blood in younger fish, with larger puss and flagellate filled boils being formed in infected adult fish. Interestingly, experimental infection with the causative agent of the Canadian outbreak showed two stages of infection with initial infection in the blood and later dissemination into tissues (Guo and Woo 2004). Electron microscopy studies by Sterud (Sterud et al. 1997) prompted the reassignment of this organism to the genus Spironucleus. The new species, Spironucleus barkhanus, named after the crescent-shaped

47 ridges at the exit of the recurrent flagella, was detected in grayling (Thy- mallus thymallus) and Atlantic salmon (Salmo salar). Infection in grayling proved highly prevalent and showed no evidence of pathology, whereas the presence of S. barkhanus in Atlantic salmon was associated with disease (Sterud et al. 1997). Subsequent experiments using axenized strains demonstrated differences in temperature preference and growth speed (Sterud 1998).

Figure 5. Scanning electron micrograph of S. salmonicida. Image by Prof. An- drew Hemphill, University of Berne.

These differences prompted genetic investigations of S. barkhanus which in 2006 established S. salmonicida, “salmonid killer” (Fig. 5), as the causative agent of systemic spironucleosis in farmed salmonid fish, being differentiat- ed from S. barkhanus of grayling which was regarded as a commensal (Jorgensen and Sterud 2006). The natural host of S. salmonicida has remained unknown until recently when wild Arctic char (Salvelinus alpinus) and brown trout (Salmo trutta) were identified to carry the parasite without any signs of disease (Jorgensen et al. 2011).

1.9.2 Genomics, cell biology and metabolism Knowledge about the cell biology of Spironucleus species remains rudimentary. Most of the information regarding their biology has been inferred from a single combined genome sequence survey (GSS) and expressed sequence tag (EST) study in S. salmonicida (Andersson et al. 2007). A genome and EST project aimed at S. vortens has been initiated by the Joint Genome Insti- tute. Recently, the genome project was put on hold citing large problems

48 with assembly (Joint Genome Institute 2012). After sequencing 130 Mb, estimated to give 8x coverage of the initial suggested genome size (~15 Mb), assembled contigs were only at twofold coverage. An analysis of ESTs showed that only 18% of gene space was assembled. The problems with assembly might be caused by extensive polymorphism between haploid genome copies, high amounts of repetitive sequence, potentially uncloneable sequences of the genome or that the genome size was severely underestimated (Joint Genome Institute 2012). The Spironucleus salmonicida GSS revealed an AT-rich genome (64%), with high coding density and short intergenic regions (Andersson et al. 2007). Like Giardia, gene promoters lack typical eukaryotic motifs and A/T- rich stretches are commonly found within 50 bp of the translational start codon. The S. salmonicida genome appears to be intron-poor and the 3’UTRs appear to be very short, often not more than 9-14 nt. The S. salmonicida genome encodes a large complement (>300) of cysteine-rich proteins. The cysteines are mainly present in Cys-X-X-Cys motifs such as in the VSPs of Giardia, but bear no evidence of the hallmark of VSPs such as the CRGKA or the GGCY motif (Andersson et al. 2007). However, a pentapeptide motif that have some shared characteristics to that of VSPs ([KR][KR][GSA][KR][KR]) has recently been recovered from a set of about 100 cysteine-rich type I membrane proteins in S. salmonicida (F. Xu, unpublished data). Oddly enough, many cysteine-rich proteins show the highest similarity to proteins in the ciliate Tetrahymena thermophila.

Molecular biology Molecular studies have been performed in S. vortens, where a transfection system has been created (Dawson et al. 2008). Transfection was successfully accomplished by employing the puromycin acetyltransferase (pac) selection marker on an episomal plasmid. The episomal plasmids were observed by FISH to reside in a single nucleus of the parasite following selection. Ex- pression and nuclear localization of GFP-tagged Histone H3 was demonstrated. Unequal expression levels of the tagged protein in individual cells of the transfected population were noticed, a phenomenon previously attributed to copy number variation of the episome in transfected Giardia cells (Singer et al. 1998).

Metabolism and MROs The metabolic capacity of Spironucleus spp. has not been fully elucidated, although recent progress has been made in the study of S. vortens metabolism (Millet et al. 2010; Millet et al. 2011; Williams et al. 2012). An unusual biphasic growth curve has been seen in S. vortens, where the rapid phase of growth shows an unusually short doubling time that approaches the shortest recorded for any eukaryote (Millet et al. 2011). When growing on glucose the endpoint products of the S. vortens metabolism are acetate, alanine, CO2,

49 and lactate (Millet et al. 2011). In the case of S. salmonicida only GSS or EST data are available to reconstruct the metabolism, both of which are limited in scope and depth. However, the data indicates a metabolic profile similar to that of Entamoeba histolytica, a fermentative phagotroph (Andersson et al. 2007). Interestingly, many of the metabolic proteins in S. salmonicida are more closely related to prokaryotic homologs, suggesting adaptation to microaerophilic conditions by LGT (Andersson et al. 2003; Andersson et al. 2006; Andersson et al. 2007). S. salmonicida was found to encode sec tRNA for incorporation of selenocysteine, a feature missing in Giardia. The presence of hydrogen metabolism in hexamitid flagellates was suggested 12 years ago, when a [FeFe]-hydrogenase gene was described in S. salmonicida (Horner et al. 2000). Hydrogen evolution at a level higher than in T. vaginalis was recently reported in S. vortens (Millet et al. 2010), being close to forty times higher than in Giardia (Lloyd et al. 2002). The presence of a MRO in S. salmonicida is implied by the discovery of two proteins, Chaperonin 60 (Cpn60) and Cysteine desulfurase (IscS), both typically associated with this subcompartment (Andersson et al. 2007). The genes coding for these proteins are also found in the unpublished S. vortens genome sequence dataset. Despite the presence of double membraned electron dense organelles in S. vortens (Sterud and Poynton 2002) and S. salmonis (Poynton et al. 2004), there are no published studies that allow assignment of them as mitochondria-derived organelles, either in the form of hydrogenosomes or as mitosomes.

1.9.3 Infection and transmission Spironucleus infection is often limited to the intestine of an infected animal, although there are several examples where spironucleosis is compounded by tissue invasive or systemic infections (Kent et al. 1992; Poppe et al. 1992; Paull and Matthews 2001). S. salmonicida has been well demonstrated to invade the blood-stream and to later disseminate into organs and tissues. “Hole-in-the-head” disease caused by S. vortens is most often accompanied by infiltration into the liver, spleen and kidneys (Paull and Matthews 2001). In moderately infected fish, S. vortens was only recovered from the intestine suggesting systemic progression of the disease from an initial infection of the intestine (Paull and Matthews 2001). Intracellular parasites have been observed in some instances of infection with S. salmonicida in Arctic char and S. meleagridis in pheasants (Sterud et al. 2003; Cooper et al. 2004). In- tracellular stages of S. salmonicida appear ‘wrapped up’ with flagella surrounding the cell body (Sterud et al. 2003). S. barkhanus colonizes the fish intestine as well as the gall bladder of its commensal host (Sterud et al. 1997).

50 Virulence factors Spironucleus have to withstand the onslaught of host defensive molecules such as NO and ROS as well as tolerate the natural presence of O2 within an infected host. In contrast to Giardia, where infection remains non-invasive, at least some Spironucleus species needs to contend with higher oxygen levels to invade the bloodstream and to disseminate into host tissues. S. vortens has been found to have extremely high O2-scavening capabilities, permitting the organism to stay active in highly oxygenated environments for extended amounts of time. Although no such experiments have been performed in S. salmonicida, the genome encodes seven distinct flavodiiron proteins, two rubrerythrins, two hybrid-cluster proteins as well as arginine deiminase, all presumably involved directly or indirectly in cellular detoxification (Andersson et al. 2006). These proteins display a patchy distribution among eukaryotes and have probably been acquired from prokaryotic organisms by LGT because they confer an advantage in surviving in oxygen-poor environments. Interestingly, the acquisitions of some of these proteins pre- date the split of Giardiinae and Hexamitinae (Andersson et al. 2003). S. salmonicida encodes 50 cysteine-proteases in its draft-genome, a number more similar to the 75 proteases found in E. histolytica than to the 37 proteases of the WB isolate of G. intestinalis (F. Xu, unpublished data).

Transmission Very little is known about the transmission of Spironucleus species in general. One exception is the transmission of S. muris, which appears to proceed by cysts of similar morphological and immunological appearance as in Gi- ardia (Januschka et al. 1988). The S. muris cyst has a thinner cyst wall and is 40-50% smaller than a G. microti cyst (Januschka et al. 1988). The flagella are not disassembled as in Giardia and can be seen wrapped around the cell body. S. meleagridis has also been shown to produce cysts that often become imbedded in the mucus layer and are difficult to detect in droppings (Wood and Smith 2005). Reports of cysts from piscine Spironucleus species are sporadic and cyst-like objects are not observed during standard in vitro cultures. S. vortens has been reported to show viability for 10 hours or more in fresh-water, opening the possibility of a direct infection route. The S. salmonicida genome is known to encode homologs of Giardia cyst wall proteins (Andersson et al. 2007). The draft genome sequence of S. salmonicida encodes 11 genes with significant similarity to Giardia cyst wall proteins. One of these potential cyst wall proteins, which belong to a group of eight proteins with extensive similarity, has been expressed in G. intestinalis during encystation. The protein is sorted into ESVs and localizes in a pattern similar to CWP1 in maturing ESVs, before being deposited in the cyst wall (J. Jerl- ström-Hultqvist, unpublished data).

51 2. Present investigation

The diplomonads have attracted considerable interest throughout the years, owing to their pathogenic potential and hypothesized early divergence from other eukaryotes. The current view holds that diplomonads are not vestigial cells, but rather display remarkable examples of niche adaptation to a parasitic life-style. That said; our knowledge of diplomonads is mainly based upon a few easily cultivated isolates of G. intestinalis, a prominent pathogen of animals. The knowledge of the cell biology of other diplomonad lineages such as Spironucleus is rudimentary but some interesting observations of biology not present G. intestinalis has been noticed (Andersson et al. 2007). The present work has sought to utilize the power of comparative genomics and transcriptomics approaches to study diplomonads, with an emphasis on the G. intestinalis species complex and on Spironucleus. Upon commencing this thesis essentially all genomic knowledge about G. intestinalis was based on a single genome completed in 2007 (Morrison et al. 2007). Significant genome diversity was believed to exist in the G. intestinalis species complex, although genome scale studies were lacking (Thompson and Monis 2004). This work describes the generation of genome and transcriptome data from both human- and non-human infecting G. intestinalis isolates, aimed at understanding the extent of diversity within the species. The thesis also describes the first efforts to understand diversity in diplomonads other than Giardia, with a comparative analysis of S. salmonicida and S. barkhanus, two closely related organisms with differences in host tropism (Jorgensen and Sterud 2006). The second part of the thesis describes efforts directed at providing methods and reagents to investigate functional consequences of the differences observed in our comparative studies. We have developed and applied a novel transfection system in S. salmonicida to obtain the first information about protein localization in this organism. The availability of a completely sequenced genome together with the transfection system allowed us to identify the S. salmonicida MRO and classify it as a hydrogenosome. This effort established the S. salmonicida hydrogenosome as one best characterized organelles of its class, and contributed information about its evolutionary trajectory in diplomonads. Hopefully my work will increase the appreciation and significance of the hidden diversity present in the different lineages of eukaryotes, in particular that of the diplomonads.

52 2.1 Comparative genomics and transcriptomics in Diplomonads (Paper I, II, III, IV and VIII) 2.1.1 Genome sequencing of G. intestinalis The two different lineages of G. intestinalis that infect humans, assemblages A and B, were known to be genetically quite different at the initiation of this study. Large divergences had been noticed in individual genes when comparing the assemblage A isolate WB and assemblage B isolate GS (Lu et al. 1998). Differences had also been noticed in growth rate, disease symptoms, ability to differentiate and in maintenance of transfected episomal DNA (Nash et al. 1987; Singer et al. 1998). The basis for these biological differences was unknown. No genomic data, except data obtained through genotyping, was available for Giardia assemblages other than A and B.

2.1.2 The GS genome (Paper I) The genome of assemblage B isolate GS\M clone H7 was sequenced using the 454 Genome Sequencer FLX system (Roche). The 182 Mb data obtained was assembled de novo using the MIRA assembler into 2931 contigs that were on average covered 16x. Automated prediction of coding sequences was performed and subsequent manual inspection and refinement allowed assignment of 754 novel open reading frames not annotated in the WB genome project. The total number of called full-length protein coding genes was 4470. We also documented 221 genes that were artificially interrupted in the fragmented genome assembly. In addition, 64 genes were found to be fragmented, while being firmly found inside a contig. PCR amplification and sequencing of these genomic regions showed that 58 out of 64 events repre- sented artificial fragmentation owing to the difficulty of base-calling homopolymeric stretches using pyrosequencing technology. The remaining 6 events were regarded as potential pseudogenes.

Unique genes The average identities between the GS and WB genomes were 77% and 78% at the nucleotide and amino acid levels, respectively. In a search that excluded VSPs, HCMPs, Protein 21.1 and Nek kinases only three genes above 200 aa were found to be missing from the GS genome compared to the WB genome. The reversed search identified 28 coding regions putatively unique to the GS isolate, among them several potential lateral transfer candidates. Four of these appeared to be of bacterial origin whereas an additional four genes had their closest match in viruses replicating by rolling-circle replication. Influxes of genes have been documented previously in Giardia and other diplomonads, but the genes involved in each of these events appear to be novel acquisitions. The GS genome revealed one instance of a recent intro-

53 duction of a genomic fragment from a bacterium. A genomic fragment containing three genes had been inserted into a gene encoding Copine-1. Two of these genes had been pseduogenized, one of them likely already at the stage of integration. Phylogenetic analyses revealed that the two pseudogenes had been transferred from a relative of Porphyromonas gingivalis, a member of Bacteroidetes, commonly found in the upper gastrointestinal tract. The genomic context coupled with quantitative PCR analyses indicated that this gene segment might represent a case of aneuploidy, not being present in all chromosomal copies.

Allelic sequence heterozygosity (ASH) The WB genome displayed extremely low levels of ASH, consistent with Giardia exercising ways of eliminating sequence divergence arising between the nuclei. Assemblage B isolates have been noticed to harbor frequent double-peaks in sequencing chromatograms, often attributed to co-infection with multiple clones. The GS genome displayed significantly higher degrees of ASH, amounting to an average of 0.53% (1.25% intergenic and 0.3% in coding regions). A large fraction of the alleles were found in a 1:3 ratio, consistent with differences existing within a single nucleus, rather than solely between them. Additionally the ASH in GS is distributed differentially across the contigs, showing alternating high and low ASH regions. While most ASH is non-synonymous, 38% invoke an amino acid substitution that would extend the Giardia proteome by almost 2000 proteins if they were expressed and translated.

Synteny breaks The de novo assembly approach allowed identification of multiple instances of genomic rearrangements, insertions or deletions when comparing the WB and GS genomes. PCR amplification and sequencing allowed verification of 24 of these events, often involving members of large Giardia gene families. Synteny breaks most often involved shorter inter-chromosomal rearrangements, but some instances of breaks across chromosomal linkage groups were identified.

Potential biological differences The synthesis of the cyst wall in Giardia differentiation is transcriptionally regulated by the transcription factor Myb2 (Sun et al. 2002). The GS genome contains binding sites for Myb2 in its own upstream region and in that of the three cyst wall proteins as in the WB isolate. While the promoters of two proteins involved in sugar synthesis for the cyst wall, glucosamine-6 phos- phate isomerase and UDP-N-acteylglucosamine 4’ epimerase, lack recognizable Myb2 boxes. This implies that encystation might be regulated differ- ently in GS, an isolate known to encyst poorly in vitro.

54 The application of 454 sequencing only allowed us to partially character- ize the VSP complement of GS. The number of VSPs in GS was estimated to be around 200, of which only 16 genes were recovered as full-length. How- ever, we were able to confirm that the VSP complement of GS is highly divergent compared to WB with only 55% average similarity, in line with previous analyses indicating a unique VSP repertoire in GS (Nash and Mowatt 1992).

2.1.3 The P15 genome (Paper III) The P15 isolate belongs to assemblage E, a group of G. intestinalis mainly infecting hoofed animals (Monis et al. 2009). Assemblage E has been found to be more closely related to assemblage A than to assemblage B (Ey et al. 1997). A non-human infecting clade nested between assemblage A and B would allow assignment of adaptations associated with human infections in these two groups. We sequenced the genome of assemblage E isolate P15 using reads generated by the 454 FLX and Titanium platforms, obtaining far more sequence data and longer read-lengths than for GS. De novo assembly of the data resulted in an average coverage of 47x across the 820 contigs, for a total assembly size of 11.52 Mb. Gene calling and subsequent manual annotation resulted in 5012 protein coding sequences being annotated. Even though the coverage for each position in the assembly was considerably higher, a similar number of genes (61) were frame-shifted in the assembly. Manual inspection of the assemblies revealed 48 cases likely attributed to homopolymeric insertions or deletions. The identity of protein coding sequences was 90% between P15 and WB and 81% between P15 and GS, consistent with earlier studies (Ey et al. 1997).

The Giardia core proteome The genomes of three phylogenetically distinct Giardia groups allowed us to assign lineage specificity to the genes identified in the three genomes. We found 91% of genes to be present in all three Giardia genomes (three-way orthologs) and 9% of genes to be variable, most of which are members of the four large gene families. The selective forces acting on different processes in Giardia were estimated using the ratio of non-synonymous (dN) to synonymous (dS) mutations in core orthologs between P15 and WB. The dS values of GS-WB and P15-GS comparisons showed signs of saturation and were not analyzed. Most genes were found to be under purifying selection, but certain gene categories appeared to evolve faster/slower in comparison to all other genes in the genome. Notably, genes involved in nucleic acid and primary metabolism were found to evolve faster whereas translation associated genes accumulated differences slower.

Figure 6. Venn-diagram of unique and shared proteins in the WB, GS and P15 Gi- ardia genomes. Figure reproduced from (Jerlstrom-Hultqvist et al. 2010).

Isolate specific genes We assigned isolate specific genes by identification of open-reading frames present in a single or in two out of the three isolates. Any gene showing homology by BLAST searches (E-value > e-20) to genes in the other isolates were removed from consideration, including members of large gene families. The highest number of isolate specific genes (38) was found in the P15 isolate followed by GS (31) and WB (5) (Fig. 6). The P15 and GS isolates shared 20 proteins to the exclusion of WB, with 13 of these found in a cluster of 20 kbp in the P15 genome. We identified one isolate specific gene in P15, a putative acetyl transferase, which showed high sequence similarity to bacterial homologs. The bacterial donor lineage cannot be confidently determined but the gene is found in some Firmicutes (Clostridia, Lactobaccilli, Anaerotruncus and Enterococcus), all prominent members of the mammali- an gut. We identified a potential pseudogene in P15 that involved another potentially transferred gene. The gene in question is present in a limited number of bacteria from different groups (Ruminococcus, Collinsella, Bacil- lus).

56 Structural and gene family variation The genes in the non-core complement of P15 are most often present in non- syntenic regions of the three sequenced Giardia genomes, with 451 genes lacking clear positional orthologs. Among these genes we recovered 112 complete VSPs in the P15 assembly, likely owing to the longer read-length and higher coverage offered by technological advances in sequencing technology. Still there are additional VSPs present in smaller contigs which could not be assembled in their entirety. The number of VSPs in the P15 genome was estimated to be around 100 using a read-mapping strategy employing the conserved CRGKA motif. We recorded the number of genomic reads matching the pentapeptide sequence and using the known average coverage of the genome we could estimate an approximate number of VSPs in the genome, a method that in P15 slightly underestimated the number recovered in the assembly. Deviation from synteny was often observed between the three Giardia genomes. Of the 141 cases of insertions/deletions or rearrangements recorded across the genomes, 113 involved loss of synteny between for P15 and WB, while only 28 were detected for WB and GS. This discrepancy between phylogenetic distance and deviation from synteny is likely dependent on the higher quality of the P15 assembly that provides improved ability to recognize genomic variants. One reflection of this is the higher fraction of events that could be verified experimentally in P15 (70%) compared to in GS (59%). The chromosomal architecture in Giardia show gene-rich stable regions with maintained gene order interspersed with non-syntenic regions harboring VSPs and other non-core genes. These regions often have a higher GC% and show nucleotide signatures that deviate from that of surrounding regions, in part due to the common occurrence of VSP and HCMP genes that are more GC-rich than the genome on average.

2.1.4 Transcriptomics of G. intestinalis (Paper VIII) We performed a comparative transcriptome study using RNA sequencing (RNA-Seq) in four Giardia isolates to investigate the impact that genomic differences have on the transcriptional output. We included the assemblage B isolate GS, two isolates from different groups of assemblage A (AI-WB, AII-AS175) as well as the assemblage E representative P15. We constructed strand-specific libraries from polydenylated RNA isolated from exponentially growing WB, GS and P15 trophozoites. AS175 libraries were constructed from exponentially grown trophozoites of in vitro passage 4 and 33. The libraries were sequenced using an Illumina HiSeq2000 instrument employing 100 bp paired-end reads. We recovered 33 to 41 million paired-reads in each of the sequenced libraries, of which over 90% mapped back to the respective genomes.

57 Genome-wide transcriptional profiling The analysis of transcription across the four genomes was initially conducted on 4175 four-way orthologs. We did not analyze truncated genes and most VSPs since they rarely are present in orthologous positions across the genomes. RNA-Seq allowed refinement of the annotated gene models by adding close to 200 genes in the genomes of WB and P15 and almost 400 in the GS genome. We calculated normalized gene expression values for all genes expressed as fragments per kilobase per million reads (FPKM). Between 94% and 98% of the annotated genes were found to be transcribed employing a cut-off of 0.5 FPKM. The expression values between orthologs of individual strains correlated in a manner reflecting their respective divergence. We found that our RNA-Seq data correlated well with that of real-time quantitative polymerase chain reaction (RT-qPCR) analyses, whereas published microarray and SAGE datasets from WB trophozoites correlated moderately and poorly respectively. Many genes were differentially expressed between the isolates (between 8 and 14%), but the magnitude of variation was often small. Of these, between 52 and 176 genes were regarded as showing lineage specific differential expression. We studied expression of VSPs in WB, finding that in an uncloned population most genes show expression, although at widely different levels. VSPs containing an initiator sequence (Inr) in WB but not in P15 or AS175 were found on average to display 3x higher expression values than those without.

A silent gene cluster in Giardia Most but not all of the assemblage specific genes detected in the WB, P15 and GS genomes showed weak expression. There were a number of ORFs that showed no evidence of transcription, notably 27 ORFs in a 41 kb stretch on chromosome 5 in WB. The vast majority of these genes cannot be assigned any specific function but one is annotated as a ‘DNA polymerase’. The syntenic region in P15 is silent, including the adjoining cluster of 13 isolate-specific genes previously described in P15. In GS and AS175, this region is split in many small contigs that precludes a similar analysis.

Global analysis of polyadenylation sites No previous analysis have systematically mapped and analyzed transcript 3’ ends across the Giardia transcriptome. Reads containing a polyA-tag were extracted and mapped towards the respective genomes. In WB, although 49027 individual polyA sites were found 7,617 polyA site clusters (PACs) remained when polyA site micro-heterogeneity and low frequency sites were excluded. Around 80% of PACs contained a prominent polyA signal (PAS), with the remaining sites lacking a PAS being associated with longer 3’UTRs and genes with low expression. Ignoring the option of multiple PACs for any given gene and employing mutual proximity, we could assign 73% of the

58 PACs in WB to a protein-coding gene. The remaining non-assigned sites were termed orphan PACs and may stem from normal 3’ ends, transcripts that are leaky or antisense to genes. If orphan PACs in intergenic regions and in the sense directions of the upstream gene are assumed to represent utilization of alternative 3’ UTRs, then one in nine transcripts would be alternatively polyadenylated in Giardia. Extensive use of alternative polyadenylation sites might mean that the estimation of 19% antisense transcription in Giar- dia needs to be revised (Teodorovic et al. 2007).

Alleles in GS are expressed The presence of ASH in the GS isolate allowed us to investigate allele specific expression (ASE) in Giardia. We identified 15,913 positions in the GS genome fulfilling the criteria for being an ASH within an ORF. By employing realistic error-profiles to eliminate mapping bias we could confirm the allelic expression at 13,096 of these sites. If applied to the gene level, 1,858 genes showed evidence of biallelic expression. Allelic imbalance, where one allele shows higher expression than the other, was noticed for 1,557 genes. This can either be due to allele dose or transcriptional regulation. Allele dosage was estimated from the GS 454 genomic reads and showed that a higher dosage was positively correlated with an increased allelic imbalance. An analysis of loci where haplotype-phasing was possible confirmed the presence ASE at the majority of the investigated loci.

Splice-junction mapping The Giardia genome has been found to encode five cis-introns and four trans-spliced gene segments (Nixon et al. 2002; Russell et al. 2005; Morrison et al. 2007; Kamikawa et al. 2011; Nageshan et al. 2011; Roy et al. 2012). We confirmed the precise splicing of these introns in most isolates, the only exception being splicing of the intron in the GL50803_35332 gene that was below detection in WB and GS and barely detected in AS175 and P15. We searched for additional cis-introns by applying criteria of co- occurrence of the splicing event in non-repetitive genes from at least two isolates with a minimum of five supporting splice-junction reads. Out of the fourteen candidates matching these criteria (taking away known introns) we could confirm one novel intron upstream of the GL50803_86945 gene by RT-PCR. This intron of 36 bp conforms to the recognized splice signals in Giardia and has a canonical splice site (GT-AG).

2.1.5 Genomic characterization of S. salmonicida and S. barkhanus (Paper II) Morphologically indistinguishable closely related organisms are prevalent among eukaryotic microbes in general and diplomonads in particular. S.

59 salmonicida and S. barkhanus were recently separated into two different species based on ecological and phylogenetic grounds. We investigated the level and type of evolutionary change responsible for these two newly diverged species. Close to 5000 ESTs were generated from S. barkhanus isolated from grayling using Sanger sequencing. We obtained 1270 unique gene clusters with 871 of these having matches to previously encountered protein- coding genes. A comparison to the genes identified in the S. salmonicida genome survey recovered 233 putative orthologs showing an average identity of 84% on the protein level. This level of divergence is similar to that of assemblage A and B in G. intestinalis, which are also morphologically indistinguishable diplomonads.

Gene classes in S. barkhanus The evolutionary origin of S. barkhanus proteins were used for classification into three categories according to their closest hit in the public databases. The largest class with 538 proteins contains proteins with their closest homolog in Giardia, followed by 235 proteins that are present in Giardia but exhibit the higher matches to proteins in other organisms. Proteins of this class were most probably present in the common ancestor of the Giardiinae and Hexamitinae but the class has experienced increased evolutionary pressure in the former. Alternatively, LGT events have reshaped the historic signal of these proteins. A third class of proteins with no discernible homologs in G. intestinalis is the smallest with 58 proteins. Genes with a metabolic function is the dominant category for which an annotation is possible. A handful of these genes were also identified in the S. salmonicida genome survey where they were shown to be the result of LGT (Andersson et al. 2007).

The UGA-triplett serves three functions in S. barkhanus The UGA codon provides the only means of terminating translation of Spi- ronucleus mRNAs (Keeling and Doolittle 1996; Keeling and Doolittle 1997). Previously, the capability of using the selenocysteine (Sec) amino acid was noticed in S. salmonicida (Andersson et al. 2007). This amino acid is incorporated using the UGA codon at coding sequence internal sites in an as yet undetermined number of proteins. The full complement of proteins required for Sec utilization can be reconstructed using data from S. salmonicida and S. barkhanus. Curiously, none of these proteins is detected in genomic or EST data from S. vortens or the distant relative T. vaginalis. Selenocysteine synthetase of S. barkhanus indeed clusters with bacterial homologs consistent with a fresh recruitment of this gene. An analysis of the ends of cDNAs also implicates the UGA codon to serve as a determinant for the polyadenylation machinery, being the only conserved sequence around the site of polyA-tail addition.

60 Sequence and genome size variation in Spironucleus Allelic variants of S. barkhanus genes were detected in the EST library and could be demonstrated to be present also at the genomic level. This is in contrast with S. salmonicida where such variation is absent. The genome sizes of both Spironucleus species were examined by flow- cytometry and were found to be clearly different. The estimated haploid genome size of S. barkhanus was 18 Mb, while for S. salmonicida it was close to 12 Mb, assuming a ploidy cycling between four and eight haploid copies of the genome in the cell cycle. PFGE was unsuccessful in resolving the individual sizes of Spironucleus chromosomes, although the sizes of S. barkhanus chromosomes appeared larger.

2.2 Applications of genome and transcriptome data (Paper IV, V, VI and VII) 2.2.1 Expression and purification of proteins in Giardia (Paper V) The data available from the Giardia genomes and transcriptomes have established many interesting processes or individual proteins for study. We sought to provide episomal vectors for protein tagging followed by endogenous expression and subsequent purification of these proteins in Giardia. The development of a system for efficient recovery of tagged protein complexes in Giardia was also pursued.

Vector construction We designed cassette based N- and C-terminal vectors employing the Strep- tavidin binding peptide (SBP)-Glutathione S-transferase (GST) tag combination. A recognition site for the PreScission protease was included to allow removal of the tags from the expressed protein. We also constructed N- and C-terminal vectors that employed the StrepII- FLAG tandem affinity purification tag (SF-TAP), a significantly smaller combination less likely to inter- fere with the function of a tagged protein. These tags can also be rapidly eluted employing mild conditions, again beneficial for recovery of intact protein complexes. All vectors carry the puromycin acetyltransferase (pac) selectable marker for giardial selection.

Purification of OCT and ADI from transfected Giardia We tagged OCT C-terminally and ADI N-terminally with the SBP-GST tag and successfully established Giardia transfectants. Purification was performed on a glutathione matrix with elution being accomplished by protease cleavage using GST-tagged PreScission protease, removing the tag and pro-

61 tease in a single step. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) demonstrated protein species purified to apparent homogeneity with the expected sizes for OCT and ADI. Both elution fractions were found to possess the respective enzymatic activities. The specific activity of ADI was found to be higher than for a bacterially expressed protein purified by a similar protocol.

Tagging and purification of the Giardia 26S proteasome We generated a tagged strain of giardial homologs to the proteins Pre1, Rpn11 and Rpt1, previously shown to be amenable to tagging in S. cerevisiae. Rpn11 and Rpt1 cells grew normally and were found to express tagged proteins of the correct sizes by Western Blotting. Pre1 transfectants were not analysed further due to a severe growth defect. Parallel purifica- tions using WB wild-type, AN-Rpt1 and AC-Rpn11 were performed using the StrepTactin matrix followed by elution with desthiobiotin. The eluted fractions were analyzed by Western Blotting using specific antibodies towards Giardia Rpt1 and α4 subunit (Pre6), showing enrichment of proteins in the regulatory particle (RP) as well as in the core particle (CP) upon AN- Rpt1 pull-down. No co-purification was seen for the wild-type and the AC- Rpn11 strains. The proteins of the AN-Rpt1 elution fraction were concen- trated, separated by SDS-PAGE and silver-stained; revealing around 30 protein species with prominent clusters around 20-30 kDa and 40-70 kDa with the addition of two bands of larger molecular weights. Bands were excised and analyzed by matrix assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry (MS).

Composition of a giardial protein complex The theoretical composition of the proteasome in Giardia was studied by bioinformatics employing a reciprocal BLAST approach towards S. cerevisiae where the proteasome has been extensively characterized. Thirty shared proteins were recovered in this analysis, some of which displayed very large divergences with uncertain assignments of homologs. MS identified 37 unique proteins in the purified fraction, 28 of these were part of the bioinformatics data set. The only proteins not detected in the MS analysis were two β subunits of the CP, both of which produce few peptides suitable for analysis. The additionally co-purified proteins include Ubiquitin carbox- yl-terminal hydrolase 14, Dynamin, BiP and ribosomal proteins.

2.2.2 Establishing a transfection system in S. salmonicida The development of a second diplomonad model system in S. salmonicida would be useful both for the study of unique aspects of this pathogen and for the analyses of the evolutionary path and history of Giardia and Spironucle- us. In vitro growth of S. salmonicida is robust and comparatively rapid in

62 non-specialty media. Further, the availability of a high-quality draft genome with RNA-seq validated open-reading frames allows analysis and precision cloning of essentially the complete gene set (F. Xu, unpublished data).

Vector construction and selection markers We evaluated five different antibiotics (puromycin, blasticidin S, G418, paromomycin, hygromycin B) for their ability to inhibit the growth of S. salmonicida. All five drugs were found to be effective in inhibiting cellular proliferation, but hygromycin B and paromomycin needed to be used in amounts deemed unpractical. Modeled on the previously described Giardia vectors, we devised a cassette-based vector system where we provided three separate selectable markers (pac, blasticidin-S deaminase (bsr) and neomy- cin phosphotransferase II (nptII)), five C-terminal tagging options (3xHA, 3xMYC, 2xOLLAS, SF-TAP and SBP-GST) and the same N-terminal tagging options except for the 3xMYC tag. N-terminally tagged constructs are driven by the α-tubulin promoter.

Establishment of S. salmonicida transfectants Transfection and selection of transfectants were successfully demonstrated using all three markers using a protocol similar to that used for Giardia and S. vortens (Singer et al. 1998; Dawson et al. 2008). Transfection using episomal plasmid DNA proved robust with confluent transfectants generally obtained within three weeks from start of selection. We demonstrated locus- specific integration transfection by single-crossover recombination using the bsr selection marker.

Generation S. salmonicida transfectants expressing epitope-tagged proteins We prepared constructs of six S. salmonicida genes, three of which were tagged by a C-terminal 3xHA tag in a pac selectable background (Caltractin, IFT46 and Fibrillarin). The remaining three genes were marked by a C- terminal 2xOLLAS tag in a bsr selectable vector (BiP, PDI-2 and Sec61α). The pac vectors were transfected in episomal format whereas the bsr constructs were transfected linearized at a unique restriction site within the coding sequence of the tagged gene. Transfectants were generated and Western Blotting revealed protein of the expected sizes for all six constructs. Howev- er, only the BiP construct revealed successful integration in the native locus, whereas the other two constructs failed to integrate site-specifically and presumably remained episomal or integrated elsewhere in the genome. Discrete protein bands of smaller sizes than expected were also seen for some of the constructs. We studied the subcellular localization of these constructs by confocal microscopy to infer commonalities and differences to cellular organization in Giardia. Fibrillarin localized to the nuclei of the cell with a partial but clearly different subcellular distribution compared to nuclear

63 DNA. Caltractin was found to localize to pyramidal shaped structures anteri- or to the nuclei. These structures likely represent the basal bodies, seen by the co-localization of tagged Caltractin and the basal body marker Centrin. IFT46 was detected along the flagella in foci sometimes enriched at the tip, the basal bodies and along two C-shaped sheets running the length of the cell. These sheets transition into S-shaped structures at the tapering end of the cell. All three 2xOLLAS tagged proteins exhibited similar localizations around the nuclei and along the length of the recurrent flagella, sometimes with intense staining in the region posterior to the nuclei.

Double transfected cells for co-localization experiments We investigated the option of generating cells carrying two selectable markers simultaneously. In initial experiments we had noticed an interaction between the antibiotics during double-selection. Concomitant selection with blasticidin S and puromycin at standard selective concentrations were not effective in inhibiting cellular proliferation in a pac-harboring S. salmonicida cell line. With increased amounts of blasticidin S (from 15 to 80 µg/ml) this interaction could be overcome and double transfectants carrying PDI-2-2xOLLAS (bsr marker) and IFT46-3xHA constructs (pac marker) were created by super transfection. Confocal microscopy of double-stained cells showed indistinguishable localization to that from single-marker experiments.

2.2.3 Genome-wide analyses of recombination suggest that Giardia intestinalis assemblages represent different species (Paper IV) Criteria for recognition of species often invoke the presence of reproductive isolation. There is evidence that suggest some mode of genetic interchange is operating in and between Giardia assemblages (Cooper et al. 2007; Teodorovic et al. 2007; Lasek-Nesselquist et al. 2009). However, the preva- lence and impact of recombination in Giardia has not been studied at the genome scale. We made use of the three sequenced Giardia genomes (WB, GS and P15) that belong to three of the eight Giardia assemblages (A, B and E) to investigate if they maintain reproductive isolation.

Approaches for testing for recombination between Giardia assemblages A three-way positional ortholog dataset consisting of 4,009 genes was com- piled using stringent standards. Special care was taken to avoid size- heterogeneity in the form of truncated or frame-shifted genes which might produce false negatives in recombination analyses. Recombination can potentially be detected by anomalous dN and dS values. As previously noted, dS values of P15-GS and WB-GS comparisons showed signs of saturation,

64 which would preclude their use for recombination detection. Instead, we computed normalized pair-wise dN values, which show no sign of saturation, and visualized them in triangle-plots. We detected only small deviations from the mean for the vast majority of genes, indicative of limited amounts of recombination between assemblages. Three additional recombination tests (Sawyer’s, maximum χ2 and RDP) were applied for sensitive detection of recombination on the small-scale. A consensus approach of these tests left the vast majority of genes as unaffected by recombination.

Recombination events are rare between assemblages The 112 strongest candidates for recombination from the substitution rate analysis were plotted on the WB genome to scan for clustered regions showing strong deviations from the mean. This analysis recovered one region of 6 kbp containing five genes in P15 that is present as two slightly divergent copies (P15-1 and P15-2) in the genome. Three of the genes in the P15-2 copy display higher identity to WB than to the P15-1 copy suggestive of a foreign source of this region. Micro homology assessment further indicated a mosaic pattern of ancestry in the P15-2 copy consistent with successive gene conversion events.

2.2.4 Hydrogenosomes in the ancestor of diplomonads and parabasalids (Paper VIII) The ability to exclusively inhabit anaerobic environments is a trait found in diverse organisms across the eukaryotic phylogeny (Shiflett and Johnson 2010). Persistence in such environments decouples the possibility of using oxygen as a terminal acceptor of electrons generated via cellular respiration. These organisms have adopted other strategies for generation of ATP, chief- ly via substrate phosphorylation and concomitant production of molecular hydrogen as a consequence of using protons as electron sinks. This process is still maintained within a double-membraned compartment of mitochondrial origin, the hydrogenosome. The anaerobic group Metamonada is a diverse and distantly related collection of organisms including the diplomonads as well as the trichomonads, where hydrogenosomes where first described. Recently, the fish parasite S. vortens was reported to evolve hydrogen at levels exceeding those of T. vaginalis (Millet et al. 2010), well known for its prominent and highly active hydrogenosomes (Lindmark et al. 1975). High level hydrogen production in eukaryotes is thought to necessitate the presence of three hydrogenase maturases that together synthesize the complicated organometallic iron cluster affording high activity to [FeFe] hydrogenase. Concordantly, hydrogenase activity of Giardia, that lacks these maturases but contains [FeFe] hydrogenase, is very low. However, the underpinnings of hydrogen production in Spironucleus remains unknown.

65 S. salmonicida contains a MRO A gene for Cpn60, a well-established marker for MROs, has previously been identified in S. salmonicida (Andersson et al. 2007). We constructed an integrated S. salmonicida strain expressing 3xHA epitope tagged Cpn60. Cpn60 localized to numerous foci posterior to the nuclei. Immunoelectron microscopy analysis of these cells demonstrated labeling of electron-dense double membrane organelles measuring 300-400 × 200 nm.

Protein import machinery and Fe-S cluster biogenesis in the S. salmonicida MRO. We searched the S. salmonicida draft genome for genes known to localize to the mitosome of Giardia (or the hydrogenosome of T. vaginalis). Proteins involved in protein import (mtHSP70, Pam18, Tom40, GrpE, Jac1) and Fe-S cluster biogenesis (IscU, IscS, two Ferredoxins, Nfu) were identified on the S. salmonicida genome. A potential Frataxin homolog, previously noted to be present in S. vortens (Tachezy and Dolezal 2011), was detected in S. salmonicida. This gene, which is potentially involved in Fe-S cluster biogenesis, is not found in Giardia. The eleven genes were epitope-tagged and shown to co-localize with Cpn60 in the MRO by immunofluorescence (IMF) analyses.

Hydrogen production is common in Spironucleus The ability to produce hydrogen was surveyed in S. barkhanus, S. salmonicida, S. vortens and G. intestinalis. By gas chromatography, we detected ample hydrogen production from all three Spironucleus species, whereas hydrogen production in G. intestinalis was below the detection level. S. vortens was found to evolve around ten times more hydrogen than both S. salmonicida and S. barkhanus in a cell number and time dependent manner.

Proteins for hydrogen metabolism localize to the S. salmonicida MRO The S. salmonicida genome was found to encode homologs of proteins implicated in hydrogen production. We detected seven genes for [FeFe] hydrogenase, five genes encoding PFOR as well as genes for the hydrogenase maturases HydE, F and G. The subcellular localizations of these genes were investigated by IMF and mutual interactions were queried by a proximity- ligation assay, establishing that two hydrogenases (FeHyd5 and FeHyd6), one PFOR (PFOR 5) and the hydrogenase maturases localized to the MRO (see Fig. 7 for HydG localization). Further, direct interactions of HydF with FeHyd5, FeHyd6 and HydG were demonstrated via proximity-ligation. The presence of these gene products in a mitochondria-derived compartment is consistent with it being classified as a hydrogenosome.

66 The hydrogenosomes of S. salmonicida have an ancient origin The origins of the PFOR, [FeFe] hydrogenase and HydEFG were studied by bayesian inference (BI) and maximum likelihood (ML) phylogenetic reconstructions. The assembled datasets encompassed sequences from S. salmonicida, S. vortens and Trepomonas sp. PC1, representing both parasitic and free-living diplomonads.

Figure 7. S. salmonicida cells with labeled hydrogenosomes. Maximum intensity projection of HydG-2xOLLAS transfectants stained with anti-OLLAS (red), TAT1 anti-tubulin (green) and DAPI for the nuclei (blue), imaged by laser scanning confocal microscopy.

The phylogenetic trees for HydEFG suggest that they were present in the organism that gave rise to the diplomonads. The HydF and HydG phylogenies show eukaryotic monophyly whereas HydE branches with low support together with two bacterial groups in the ML tree. The BI tree for HydE on the other hand recovers eukaryotic monophyly. The branching orders support a common ancestry of the hydrogenase maturases in trichomonads and diplomonads. PFOR also recovers eukaryotic phylogeny with four paralogous genes being present in the diplomonad ancestor. Two of these groups, including the one with the hydrogenosomal PFOR in S. salmonicida, have since been lost in Giardia. [FeFe] hydrogenase ancestry is difficult to interpret, but indicates the presence of multiple genes in the founding diplomonad. The combined results promote ancient shared histories of the hydrogen production machinery in trichomonad and diplomonad hydrogenosomes, with intriguing signs that

67 this branch of eukarya might have been the original recipients of these genes via LGT from bacteria.

Additional functions for the S. salmonicida hydrogenosomes By comparative genomics, many homologs of the T. vaginalis hydrogenosomal components were revealed in S. salmonicida. The subcellular localizations of twenty of these were determined by epitope-tagging and IMF analyses. We concluded that two additional proteins, serine hydroxyl methyltransferase (SHMT) and a potential component of the glycine decarboxylase complex (H-protein), demonstrated hydrogenosomal localization by virtue of co-localization with Cpn60. The H-protein is too small and divergent to permit phylogenetic analyses but SHMT branches earliest among sequences in the eukaryotic cluster together with T. vaginalis. The presence of these two proteins argues for an ancient state of amino acid metabolism in the hydrogenosomes of S. salmonicida.

Organelle enrichment gives clues to hydrogenosomal ATP synthesis Cellular fractionation was employed in an effort to identify novel components of the S. salmonicida hydrogenosome. We adapted an organelle enrichment approach devised for Giardia mitosomes (Emelyanov and Goldberg 2011) to the application of S. salmonicida hydrogenosomes. Cells carrying a Cpn60-3xHA construct were lysed by sonication and cellular fractions were prepared by successive rounds of differential centrifugation, identifying hydrogenosome enrichment by Western Blotting. The hydrogenosome-enriched 50,000 × g pellet was subjected to centrifugation in a discontinous Nycodenz gradient. The layered materials of this centrifugation were analyzed by Western Blotting and the hydrogenosome-rich fraction was submitted for proteomic analysis by liquid chromatography tandem mass spectrometry (LC-MS2). In this fraction we detected 2907 peptides derived from 804 proteins, identifying the vast majority of previously recognized hydrogenosomal proteins. Additionally, we identified a divergent homolog of an acetyl-CoA synthetase (ADP-forming) and showed its hydrogenosomal localization by epitope-tagging and co-localization with Cpn60. An acetyl-CoA synthetase activity allows reconstruction of an hypo- thetical pathway for ATP generation within the S. salmonicida hydrogenosome (Fig. 8).

68 3. Discussion

3.1 Are G. intestinalis assemblages different species? Sequencing of Giardia isolates representing assemblages B and E has together with the previous assemblage A genome highlighted the presence of large genetic differences in these morphologically very similar organisms. Even though previous analyses have reported evidence of inter-assemblage recombination events (Teodorovic et al. 2007; Lasek-Nesselquist et al. 2009), it appears to have had no major impact across the three sequenced Giardia genomes. This suggests that these are rare and that assemblages are essentially reproductively isolated. Only a single event of inter-assemblage recombination, between P15 and an assemblage A representative, involving a duplicated gene segment, was identified across the three genomes. The extent of primary sequence divergence and the presence of plastic gene regions could presumably act as barriers to efficient recombination based genetic exchange. In accordance with this, recombination has been confidently identified in the AII subassemblage where primary sequence divergence among strains is low (~98% identical), but high enough to permit confident detection of recombinants (Cooper et al. 2007). The mechanism whereby Giardia recombination occurs remains obscure; fusion of Giardia trophozoites has not been observed in in vitro experiments employing strains carrying either of two selectable markers (Carpenter et al. 2012). In Trypanosoma brucei, a sexual stage was known to occur in the tse-tse fly host, but only recently did the use of fluorescent fusion proteins allow a demonstration that epimastigotes in the salivary glands are the sexual stage of this parasite (Gibson et al. 2008). A similar approach applied to Giardia might be able to capture rare occurrences of recombination or a sexual cycle (discussed in section 3.2.3) only induced in an appropriate host environment.

3.1.1 The plastic genome of Giardia The Giardia genomes consist of gene-rich syntenic regions interspersed by islands with low gene density and a marked enrichment of proteins that belong to non-core genes of the major gene families. Most cases of synteny loss in the Giardia genomes are attributed to events occurring in these plastic areas of the genome. These regions could serve as areas where genes are free to evolve by recombination, duplication and gene conversion without

69 any constraints. A similar but more extreme version of this genome arrangement is seen in the oomycete plant pathogen Phytophthora infestans, which has a rapid ability to overcome plant defenses through rapid evolution of its effector repertoire. In the case of P. infestans, a massive increase in genome size has occurred via accumulation of repetitive elements in regions of the genome that are plastic. Core genes are found in coding rich regions with extended synteny to related Phytophthora species (Haas et al. 2009). The Giardia genome is different in this respect, since only a few telomere- positioned transposons have been described (Arkhipova and Morrison 2001). Genes potentially encoding reverse transcriptases are found in non-syntenic positions across the three Giardia genomes, indicating that repetitive elements have influenced the gene organization of Giardia since the divergence of the assemblages. Additional genomes might reveal whether repetitive elements contribute to genome evolution in Giardia.

3.1.2 Gene families present in the plastic genome It is intriguing that subsets of Nek kinases, termed classes GL1-4, exhibit lineage specific expansion across the three Giardia genomes (Manning et al. 2011). The GL4 family is especially interesting due to its high rate of protein turnover. Likewise, ankyrin repeat proteins (Protein 21.1) are also differentially distributed across the three genomes. About two thirds of the HCMPs were found as three-way orthologs, with the remaining third being present in the variable part of the genomes. The function of this large group of genes has not been elucidated, but HCMPs are known to be up-regulated during various kinds of stress conditions (Muller et al. 2007; Ringqvist et al. 2011). The VSPs are clearly the most variable group of proteins in the Giardia genomes, with most genes lacking clear orthologs. Distinct repertoires of VSPs might endow Giardia assemblages with differing abilities to colonize hosts and to avoid elimination by the hands of immune defenses. One theory is that zoonotic assemblages A and B might have larger diversity of VSPs than Giardia assemblages with a narrow host-range, a feature that might aid in establishing infections across animal groups.

3.1.3 The mysteries of G. intestinalis gene expression The presence of a large silenced genomic region in WB, P15 and maybe GS is intriguing and could be a good tool used to investigate the conditional expression and silencing of individual VSPs, given that the mechanism has commonalities. In any case, it might be possible to harness the silencing machinery and apply it to endogenous genes as has been realized in E. histolytica (Morf and Singh 2012). The low number of transcription factors and paucity of motifs in the upstream regions of many Giardia genes is puzzling. Our discovery of the po-

70 tential for alternative utilization and differences in 3’UTR length between genes with high and low expression could be an indication that regulation in the 3’ UTR is more common than previously envisaged. Regulation of VSP expression in Giardia involves components of the RNAi machinery (Prucca et al. 2011). Other classes of genes might also be regulated by similar mechanisms. Chromatin alterations are involved in silencing of VSP genes (Kulakova et al. 2006), but the influence of positional effects in the genome remains unknown. It might be possible that the GC-rich genomic islands and the subnuclear position of VSP genes are important for silencing similarly to that of var genes in Plasmodium (Frank and Deitsch 2006).

3.1.4 Presence of ASH in Giardia and its implications for the cell The presence of two nuclei and four copies of the genome afford more op- portunities to recombine and accumulate gene differences with each cellular division. Several of the proteins implicated in DNA repair and meiosis show large divergences as well as insertions or deletions when comparing the WB- P15 and GS clades. Gene variants or mutations arising in one nucleus that offer a selective advantage in the population might then be reshuffled and propagated to offspring by diplomixis during encystation. Diplomixis was shown to occur in the WB isolate, but there has been no investigation of this process in isolates of assemblage B. There is a possibility that there are differences in the efficiencies of DNA repair, recombination or diplomixis between assemblages. Such differences might manifest themselves as higher levels of ASH in certain Giardia lineages. An intriguing hint of differences in the process of recombination comes from the observation of integration of episomal or linearized DNA in transfection experiments involving the GS isolate, as opposed to the WB isolate where only linearized DNA was integrated (Singer et al. 1998). Evidence of ASH at the single cell level in Giardia has been described (Ankarklev et al. 2012). The ASH detected on the genomic level in GS is also manifested at the transcriptomic level, with evidence of allele dosage being propagated to the transcriptomic level. This observation would suggest that there is no fundamental difference in expression level between the nuclei. Morphological differences between nuclei noticed by some authors (Benchimol 2004) are most likely a reflection of asynchronous start of replication in individual trophozoites (Tumova et al. 2007). Any such asynchronous initiation would be averaged out in a non-synchronized trophozoite population as in this study. Synchronized GS trophozoites might be an ave- nue to understand if there are fundamental and systematic differences in the regulation between different nuclear copies. Aside from being technically

71 demanding, this would also require a presently unavailable haplotype-phased GS assembly.

3.2 Evolutionary trends in diplomonads Several evolutionary trends or specific events have influenced the evolution of diplomonads. The gain of genes from other organisms and evidence of an enigmatic sexual cycle have been unearthed as a part of this project. These processes together with larger genetic upheavals such as the modification of the genetic code in Hexamitinae have shaped the current diplomonad lineages.

3.2.1 Gene acquisitions: a common theme in diplomonads One of the most profound findings of the Giardia genome was the ample evidence of genes with a phylogenetic history indicating acquisition by LGT. In an influential work published in 1998, the evolutionary biologist Ford Doolittle proposed a ratchet-like mechanism to explain how bacterial genes (i.e. not mitochondrion derived) might end up being encoded in the nuclear DNA of eukaryotic organisms (Doolittle 1998). The ratchet as envisaged by Doolittle, states that eukaryotes will acquire beneficial genes from bacterial prey inhabiting the same environment as the recipient eukaryote. Basically, “You are what you eat”. Influx of genes, especially metabolic genes has greatly influenced the capabilities of extant diplomonad organisms. We might assume that this process could be more frequent in hexamitid flagellates than in Giardiinae given that they retain a functional feeding apparatus. Other processes of obtaining genetic material are likely also operational since gene acquisition from bacteria inhabiting the same niche as Giardia is still an ongoing phenomenon. Viruses and plasmids might contribute novel genetic material in Giardia. Rep-genes of potential viral origin were identified in the GS isolate. Such genes have been found in the BRIS/92/HEPU/1541 isolate previously where they were found to be associated with a plasmid-like integrative element (Upcroft and Upcroft 1998).

3.2.2 Alternative genetic code and selenocysteine The use of an alternative genetic code is employed in various organisms throughout the three domains of life. The most familiar example of an alternative genetic code is perhaps the modified code of our own mitochondria (Lobanov et al. 2010). How an alternative code came to be established in Hexamitinae is not known, but upcoming genome information from diverse diplomonad organisms might be able to establish some characteristics of the ancestral genome that facilitated such a transition. Extreme genome events

72 such as high or low GC contents or severe reductive processes have been documented to promote recoding events. The alternative code utilized by hexamitid diplomonads and some ciliates (code 6) has been suggested to be more robust towards error than the standard code (Kurnaz et al. 2010). The selenocysteine encoding trait is found in S. salmonicida, S. barkhanus and Trepomonas sp. PC1, but appears to absent from S. vortens. The phylogenetic placement of these four lineages indicates that the Sec-trait was ancestral to hexamitid flagellates and subsequently lost in S. vortens (Takishita et al. 2012). Sec is not utilized by Giardia or Trichomonas and recruitment of the Sec-trait from a bacterial source into the hexamitid branch is suggested by the phylogeny of selenocysteine synthetase. Selenocysteine incorporation has been promoted as a way to increase the efficiency of redox-active proteins, having especially beneficial effects on the catalytic rate (Arner 2010). At present only one protein, selenoprotein W, has been noticed to encode selenocysteine in the S. salmonicida proteome and nothing is presently known about the physiological impact of this trait.

3.2.3 Sex in diplomonads Sex is a powerful strategy to generate new variant organisms and was probably a feature of LECA (Ramesh et al. 2005). In addition to meiotic sex, cells can entertain parasexual life-cycles or cellular hybridization by cell fusion. The low levels of and/or uneven distribution of ASH in some Giardia isolates might be explained by rounds of gene conversion and diplomixis (Andersson 2012). As previously discussed, no direct evidence of cell fu- sions has been reported in Giardia although clear evidence indicative of such events have been published (Cooper et al. 2007; Lasek-Nesselquist et al. 2009). Many other organisms require matching of mating types for completion of the sexual cycle. Such a system in Giardia could explain the failure to observe cell fusion with two differentially marked clonal Giardia WB strains, since these are of the same mating-type. It would be interesting to perform mating experiments between different isolates of the AI and AII assemblages to investigate this hypothesis. If sex in Giardia involves cell fusion, hybrids could by random-loss of excess genetic material lead to the generation of large numbers of genetic variants as a result of a single rare event. “Hybrid vigor” offered by such a mechanism could increase the chances for adaptation to adverse situations (Schurko et al. 2009). Our finding of potential ASH in S. barkhanus might be an indication that a similar sexual cycle could be operational also in other diplomonads.

73 3.3 Finding the essence of diplom(on)adness? The establishment of a transfection system combined with an essentially complete genome sequence makes S. salmonicida a rare organism amongst microbial eukaryotes. There are only about ten sequenced excavate genomes available, with most being from human pathogens. S. salmonicida is only the second organism harboring a hydrogenosome where a complete genome and genetic tools are available. Thus there are great prospects for exploring the different lineage specific adaptations in Spironucleus and Giardia and the basic biology of hydrogenosomes using the developed tools.

3.3.1 S. salmonicida as a hexamitid model system The development of a hexamitid model system with similar capabilities as in Giardia has been initiated by work described in this thesis. There are only a limited number of eukaryotic microbial systems with transfection capabilities and a completely sequenced genome. The transfection system was modeled on existing systems in Giardia (Singer et al. 1998) and S. vortens (Dawson et al. 2008). The considerable ease of cloning and transfection in S. salmonicida have allowed the generation of above one hundred transfectant strains, many of them double-transfectants, since the conception of the system. The completeness of the S. salmonicida genome makes it a more attractive model for hexamitid flagellates than S. vortens, where the incompleteness of its genome prevents whole genome analyses. Site-specific integrative transfection using single-crossover recombination, a method utilized in Giardia (Gourguechon and Cande 2011), was shown to be possible also in S. salmonicida. Methods for double-crossover integration (Singer et al. 1998; Jimenez-Garcia et al. 2008) and inducible expression (Sun and Tai 2000) exists in Giardia but have yet to be realized for S. salmonicida. Live cell microscopy using fluorescent proteins or chemical probes have in Giardia permitted study of dynamic processes, evidenced by analyses of ESVs (Konrad et al. 2010) and attachment via the ventral disc (House et al. 2011). The application of fluorescent tools have not been extensively pursued in S. salmonicida but might help in elucidating homolo- gous structures between Giardia and Spironucleus, and give clues to how hydrogenosomes are replicated and distributed during cell division.

3.3.2 Where are the S. salmonicida cysts? The transmission of S. salmonicida is not well-characterized and no cyst- stage has been confidently described in the literature for any piscine associated Spironucleus species. Cysts with similar appearance and immunogenici- ty as Giardia cysts have been described in S. muris (Januschka et al. 1988) and cysts are also known from S. meleagridis (Cooper et al. 2004). S. muris

74 and S. meleagridis are closely related to the exclusion of piscine Spironucle- us species (Kolisko et al. 2008) and might as parasites of terrestrial animals employ a similar transmission mechanism as Giardia. The presence of bona- fide homologs of CWPs in the S. salmonicida genome would suggest that an ability to differentiate into cysts is still or was until recently operational in the parasite. Cyst-like objects are not observed during regular in vitro culture of S. salmonicida, nor have we been able to induce cyst-formation by expos- ing the cells to conditions known to induce encystation in Giardia (high amounts of bile and increased pH). By employing transfectant cells with a CWP promoter fused to a reporter gene it might be possible to pinpoint stimuli that lead to activation of differentiation, and reverse engineer encystation stimuli. A similar approach where trophozoites expressing a reporter gene from a stably integrated construct is utilized in in vivo infections might also divulge if piscine diplomonads might be transmitted by other mechanisms, either as excreted trophozoites or perhaps as tissue cysts as in Toxo- plasma (Dubey et al. 1998).

3.3.3 Cellular organization in diplomonads The cellular organization in diplomonads is intriguing and subject to many speculations. The independent lineages of ‘enteromonads’ with their single karyomastigont might be ideal systems to study diplomonad specific solutions for the maintenance of the dual body plan. The expansion of certain gene families that appear associated with the cytoskeleton could also give clues to how the complicated cytoskeleton is replicated and faithfully divided during each cell-cycle. The annexins (α-giardins) appear to have been independently expanded in Giardia and Spironucleus and show large diver- sities in their localization and presumably function in both species (J. Jerl- ström-Hultqvist, E. Einarsson, unpublished data).

3.4 The MRO in diplomonads The discovery of genes encoding hydrogenase maturases of ancient origin in S. salmonicida, Trepomonas sp. PC1 and S. vortens show that Hexamitinae have inherited these genes from a common diplomonad ancestor. This suggests at the same time that these genes have been lost in the lineage leading to G. intestinalis. Whether this applies to all members of Giardiinae is unknown.

75 3.4.1 Hydrogenosomes in the ancestor of diplomonads and trichomonads There is evidence of an organelle of mitochondrial origin in all recognized major eukaryotic groups (Embley and Martin 2006). The emergence of the lineages within the anaerobic group Metamonada has been estimated to be of equal age to the split between fungi and animals (Parfrey et al. 2011). With this in mind, the MROs of metamonadan organisms have independently evolved and adapted in the different lineages from a common ancient organellar blueprint. Evidence indicative of hydrogen metabolism across all three branches of Metamonada have now been collected, even though hydrogenosomes are yet to be defined in Preaxostyla (Hampl et al. 2008). Phy- logenetic analyses of HydEFG from Trimastix, a preaxostylid flagellate, and from CLOs will soon be able answer whether all the metamonadan MROs stem from a single introductory event into Metamonada as hypothesized later.

3.4.2 Hydrogenosome function in S. salmonicida The protein import and iron-sulfur cluster machineries of the hydrogenosome in S. salmonicida vary slightly in the composition of identified components when compared to Giardia. No processing peptidase has been discovered in S. salmonicida in keeping with an apparent absence of targeting peptides on hydrogenosomal proteins. Two of the proteins with signal peptides in Giardia (IscA and Glutaredoxin) are missing from Spironucleus. It is tempting to speculate that these proteins might absolutely require a targeting signal for their delivery to the organelle, and that a loss of these proteins in S. salmonicida triggered a total reliance of internal targeting signals. The presence of Frataxin in Spironucleus might serve as a functional replacement of IscA and Glutaredoxin. Two distinct 2Fe-2S ferredoxin homologs are present in Spironucleus whereas one is present in Giardia, consistent with one copy being utilized in Fe-S cluster biogenesis and the other one in transport of electrons generated during hydrogenosomal pyruvate decarboxylation by PFOR. Two hydrogenases were found to exhibit hydrogenosomal localization, but their contribution to the organisms’ total hydrogen production is unknown. ATP might be generated by the activity of an Acetyl-CoA synthetase, a pathway not present in the hydrogenosomes of T. vaginalis. Acetyl-CoA synthetase is used for energy generation in the cytoplasm of Giardia and the S. salmonicida ortholog of this enzyme also displays a cytosolic localization. Amino acid metabolism appears to have been retained in the S. salmonicida hydrogenosome, whereas there is no evidence of such genes in Giardia.

76 3.4.3 MROs of other diplomonads The MROs of diplomonads have not been studied in other species than Gi- ardia (and now in S. salmonicida). The absence of recognizable signal peptides on hydrogenosomal proteins in Spironucleus proteins makes in silico predictions of organellar localization problematic and suggestive at best. Preliminary localization of epitope-tagged SvHydE and SvHydG in S. vortens indicates that hydrogenase maturases are distributed in multiple foci, with linear arrays of signals close to the nuclei. A similar localization is obtained with epitope-tagged SvIscU indicative of the foci representing MROs. Whether these foci corresponds to the previously visualized electron-dense double-membraned organelles close to the nuclei in S. vortens (Sterud and Poynton 2002) remains to be determined.

Figure 8. Schematic view of the verified hydrogenosomal proteome in S. salmonicida. Proteins of the G. intestinalis mitosome also identified in the hydrogenosome of S. salmonicida are shown in red. Proteins unique to the S. salmonicida hydrogenosome when compared to the G. intestinalis mitosome are shown in yellow. Suggested end products via hydrogenosomal metabolism are shown in boxes. Abbreviations: FdOX , FdRED , oxidized and reduced ferredoxin; AcCoA, Acteyl CoA, PFOR, Pyruvate ferredoxin:oxidoreductase; SHMT, serine hydroxyl methyltransferase; GDC, glycine decarboxylase complex; PYR, pyruvate.

77 3.4.4 A hypothesis of hydrogenosome origin in Eukaryotes Commonalities of hydrogenosomes in Parabasalia and Fornicata might also reflect back to the adaptation to an anoxic life-style in the original MRO of Metamonada. Phylogenetic reconstruction of the histories of the hydrogenase maturases recover diplomonads and parabasalids as diverging closest to the bacterial outgroup from monophyletic eukaryotes. Taken literally, this would indicate that the primary transfer of these proteins occurred into Met- amonada and that eukaryote-to-eukaryote transfer then ensued, giving rise to the patchy distribution we see today (LGT model, Fig. 9). It should be noted that more complex models with a secondary endosymbiotic event into Met- amonada followed by spread of key genes via intra-domain transfer events is also consistent with the available data (secondary endosymbiosis model B, Fig. 9). The available data are not in support of the differential loss model where the proteins of hydrogen metabolism were present in the α- proteobacterial endosymbiont of LECA and then lost in anaerobic or facultative anaerobic lineages. In fact, no known extant α-proteobacteria are known to encode [FeFe] hydrogenase, PFOR and HydEFG (Hug et al. 2010). More- over, the predictions of the differential loss model or the secondary endosymbiosis model A (Fig. 9), where the HydEFG phylogeny should mirror the established eukaryotic groupings (Adl et al. 2012), are not consistent with our data. The phylogenetic history of PFL in eukaryotes shows a similar evolutionary pattern with a single acquisition into eukaryotes and subsequent intra-domain spread (Stairs et al. 2011). Why these genes have not been re- peatedly transferred from bacteria and instead been acquired from eukaryotes remains puzzling. It might be possible that adaptation to eukaryotic expression conditions, whether it be reliance on specific cofactors or evolutionary constraints on auxiliary machineries were only possible in Metamon- ada at one time. Adaptation of this system to eukaryotes once it happened might have made transfer to other eukaryotes from Metamonada far more likely than a repeat of the process via inter-domain transfer. The validation of the LGT model will require further sampling of eukaryotes with inferred hydrogen metabolism, especially within Metamonada and other lineages were HydEFG have not been demonstrated so far.

Figure 9. Three models proposed to explain the presence of HydEFG of bacterial origin in eukaryotes. The three respective models (LGT, differential loss and secondary endosymbiosis) are shown applied to a eukaryotic phylogeny. Key features and their chronologies are shown to the right of each model.

79 4. Conclusions and future perspectives

The results obtained within the scope of this thesis reveal diplomonads as a diverse group of organisms that have been shaped by strong evolutionary forces in the distant as well as in the recent past.

The sequencing of Giardia genomes (Paper I & III) and transcriptomic analyses (Paper VIII) have revealed that each Giardia assemblage is divergent in primary sequence and that about one tenth of the genomes can be entirely different. The definition what constitutes the core and plastic parts of the genomes might help to explain differences in host-range and symptom induction. The Giardia genome recombination analyses (Paper VI) indicated limited gene-flow between assemblages and add to the mounting data that Giardia assemblages represent genetically divergent and distinct lineages worthy of species level designation. However, some things would still be need to be clarified before the debate can be settled. There are also pressing questions that will need answers before we can understand generation of Giardia diversity at the single-cell level.

• The genomes of representative isolates from each Giardia assemblage should be generated to define the level inter-assemblage recombination. • Hundreds of genomes from assemblages A and B with accompanying phenotypic or patient symptom data, including putative “hybrid” isolates, should be generated. This will help to understand the level of intra- assemblage recombination and if there are correlations between parasite genotype and patient symptoms. • Single-cell Giardia transcriptomes could reveal evidence about VSP expression dynamics that are not readily observable at the population level. • How are differences between the nuclei contributing to Giardia virulence? How is sexual exchange of genes accomplished?

In Paper III, VI and VII we studied Spironucleus to understand how diplomonad genomes and their biology have changed during adaptation to different life-styles or states of parasitism. In Paper III we characterized parts of the expressed genome of S. barkhanus and found large differences in genome size and in individual genes compared to the morphologically very similar S. salmonicida. We devised a transfection system for S. salmonicida

80 in Paper VI and employed it to the study of the hydrogenosome in Paper VII. Future studies of Spironucleus and the diplomonad group as a whole could answer many interesting questions that remain unanswered.

• Sequencing the genomes of additional species of Spironucleus and other diplomonads could reveal information of what adaptations are necessary for parasitism vs. commensalism (S. barkhanus), to free-living conditions (Trepomonas or Hexamita), to the maintenance of a double karyomastigont (‘Enteromonads’) and development of the ventral disc and elaborate cytoskeleton of Giardia (Octomitus). • Sequencing the genomes of CLOs would give us an excellent starting point figuring the capabilities of the diplomonad ancestor. • Metabolomic and proteomic studies of hydrogenosome-enriched fractions would allow assignment of organelle functions. What are the functions of the Spironucleus hydrogenosome and are there differences in among the individual lineages?

The development of high-throughput technologies is contributing to a flood of information and data regarding diplomonad biology. However, we are only at the initial stages of understanding the how the genomic data can be translated into function. Our transfection system for S. salmonicida has established a hexamitid model system for the study of evolution and the basis for pathogenicity in diplomonads. Some interesting aspects that could be investigated by virtue of this system includes:

• Are Spironucleus also utilizing antigenic variation of cysteine-rich proteins to avoid elimination by the immune system. • What is the function of conserved diplomonad-unique proteins?

In conclusion, the hope is that the work delineated in this thesis will lead to many new and exciting future discoveries concerning the hidden diversity that constitutes the diplomonads.

81 5. Acknowledgements

There is a great deal of people who made the completion of this work possible, and at that a most enjoyable undertaking. I’m grateful to FORMAS for funding and supporting this project.

I would like to thank: My supervisors: Staffan Svärd: Your enthusiasm for research and the “everything is possible” attitude is one of the reasons I wanted to join the group ever since I took the course “Makromolekylära maskiner”. It’s been a great and stimulating experience, with countless “we should do that” discussions! Thank you for always keeping an open mind to my random musings and for not quenching the spirit of research!! Jan Andersson: Thanks for all the help, support and great ideas throughout the duration of this project. I owe much of my fascination with eukaryotic microbes and their evolution to you! Great conferences in Orvieto and Oslo, “Rom och frukost” at Anker is legendary.

Giardia group members, past and present: Elin: Fellow Spiro pioneer. I admire your work ethic and never ending enthusiasm over “weird” projects. Do we still keep pieces of salmon in the freezer or maybe some fish bile? You have great things ahead of you (and keep up the Spiro dancing)! Britta: Current room-mate and the greatest manager of time I ever encountered! Thank you for all nice conversations and great parties. You could no doubt make a living as a party planner/chef. Your influence on the group will be long-lasting!!! Good luck back in Switzerland! Mattias: Thanks for all coffee corner discussions and your great sense of humor. Marcela: Thanks for being such a kind and caring person! Johan A: Thank you for all the good times in the lab, during teaching and at conferences. Your inventive and business savvy mind will take you far. You will rock Boston no doubt! Karin H: What great and challenging days we spent agonizing over centrifugation techniques! Who knew that pelleting Spironucleus was such an art form! Thank you for sharing all your knowledge in the lab and being a great person in general!! Emma: My former room-mate. I always enjoyed discussing science with you. Your spirit and infectious laughter would brighten the gloomiest days. Thanks a lot for the invitation to stay with you and Oscar in Palmerston North, I had a great time!

82 Karin T: Will always remember the exhilarating days of collaboration on the synchronized cells story, that’s what science is all about if you ask me!! Sorry for not always being able to answer your daily pop-quiz: ”Berätta något roligt!” Kattis: I admire your great attitude and aptitude towards science and life. Thank you for providing the first peek at Spiro and initiating that chapter of my research. Sandra B: My first student! You did such great work. That after party post Baren Baren was awesome! =) Daniel S: Thank you for all your work in exploring the Spironucleus MROs and always being up for a challenge. Good luck at IMBIM! Sara: Thank you for interesting conversations and good work with the ADI story. Franzi: Thank you for inviting me to the quiz-night at Smålands, really fun! Per and Gabriel: Super contig engineers are hopefully on your CVs. Hanna S: Thank you for the all the help when I started in the lab. Cedrique: Keep up the good work!

Collaborators: Oscar Franzén: Bioinformaticist extraordinaire. Thank you for great collaborations on the Giardia projects, nothing short of amazing times. I will never forget the fren- zied days of GS annotation. Neither will my swollen annotation elbow….a small price to pay! Feifei Xu: Bioinformatics wizard. Annotating the Spiro genome was great! Ulf Hellman: Thank you for learning me everything about MALDI-TOF. I’m sorry for (almost) always delivering pitifully weak silver-bands =). Björn A: Ugan- dan beer-sessions with Nile special will be remembered. Anders A: Always helpful and knowledgeable. Dirk “Pacho” Pacholsky: Thanks for helping me with the LSC work. Thorsten H and Petra K: Thank you for opening up your insanely warm GC room for hydrogen measurements.

Micro, past and present: Gerhart W: You passion for science is remarkable and a true inspiration. RNA club and other discussions have been great! Erik H: You are such a nice person and great scientist. The teaching at EBC was really fun! Good fiddling in Germany! Johan R: You always had time to discuss both big and small things. The spex movies were just hilarious. Cia: Thank you for fun conversations and spex movie craziness! Klas U: Thank you for making me feel welcome when starting at ICM. Nadja: Really enjoyed discussing the intricacies of science with you. Maaike and Mirthe; Good luck with your work! Bhupi, Sonchita and Henrik T are all C10:2 compatriots, thank you for all help when I was new in the lab, and for fun times later on. I’ve always enjoyed interacting with the Dicty-people, now freshly adapted growth under ICM conditions =) Fredrik S: You have always taken the time to listen to any problems, whether if it relates to science or not. Thank you! Lotta: I have enjoyed discussing peculiar eukaryotes and other things during RNA clubs and at other times. I know you will do well in Sydney. Åsa F: You have such a lovely outlook at life! Shiying: Keep up the spirit! Ulrika: The micro backbone throughout my first years. Hope IMBIM is treating you well. Magnus L, Lina, Devashish: Good-luck with your work. Jessica, Marie, Disa: The plate-people! Many spexmovies and party- moments to remember.

83 People at ICM I would like to thank all people of ICM that I have interacted and grown with during the years! Some people needs special mention. Åsa, Ewa, Anders, Sofia, Ali, Lena, Sigrid, Solan, Erling and Akiko for assistance, advice and help. Mats W, Petter H, David F, Prune for fun discussions.

Family The work described here would not have been possible without the support of my family and friends. I love you all! I would especially thank Anna for so many things related to completion of this jour- ney and beyond. No words are enough to express my gratitude. I love you so much! Thank you Mamma and Pappa for never-ending care and support whenever things have been tough, and of course at other times as well =). This work is also a reflection of you! Samuel; you may never understand my fascination with microbiology, but anyone who will almost go to a fist-fight over what constitutes vacuum have the scientific spirit for sure. You were right I must admit =). For my grandparents! I miss you! Thank you Yvonne for always being there, your kindness and always offering a possible solution to any problem. I would like to thank the Åsman family and respective partners, Lena, Ida, Matthias, Sofia, Pentti and Sven-Arne, for all the good times! Friends The “old” friends from Arboga; Bobba, Klas, Giran, Micke, Perra, Irfan Meco, Sunkan, Kajan, Rana, Jakob, Calle, Johan, Åsa. Great to have you as friends! From Uppsala; Daniel, Mats L, Rickard, Dalla, Ubbe, Linda, Lina. Many fun times!

84 6. References

Adam, R. D. (2001). "Biology of Giardia lamblia." Clin Microbiol Rev 14(3): 447-475. Adam, R. D., A. Aggarwal, et al. (1988). "Antigenic variation of a cysteine- rich protein in Giardia lamblia." J Exp Med 167(1): 109-118. Adam, R. D., T. E. Nash, et al. (1988). "The Giardia lamblia trophozoite contains sets of closely related chromosomes." Nucleic Acids Res 16(10): 4555-4567. Adam, R. D., A. Nigam, et al. (2010). "The Giardia lamblia vsp gene repertoire: characteristics, genomic organization, and evolution." BMC Genomics 11: 424. Adl, S. M., A. G. Simpson, et al. (2012). "The revised classification of eukaryotes." J Eukaryot Microbiol 59(5): 429-514. Aggarwal, A. and T. E. Nash (1988). "Antigenic Variation of Giardia- Lamblia Invivo." Infect Immun 56(6): 1420-1423. Akhmanova, A., F. G. Voncken, et al. (1999). "A hydrogenosome with pyruvate formate-lyase: anaerobic chytrid fungi use an alternative route for pyruvate catabolism." Molecular Microbiology 32(5): 1103-1114. Alberts, B. (1998). "The cell as a collection of protein machines: preparing the next generation of molecular biologists." Cell 92(3): 291-294. Andersson, J. O. (2012). "Double peaks reveal rare diplomonad sex." Trends Parasitol 28(2): 46-52. Andersson, J. O., R. P. Hirt, et al. (2006). "Evolution of four gene families with patchy phylogenetic distributions: influx of genes into protist genomes." BMC Evol Biol 6: 27. Andersson, J. O., A. M. Sjogren, et al. (2003). "Phylogenetic analyses of diplomonad genes reveal frequent lateral gene transfers affecting eukaryotes." Curr Biol 13(2): 94-104. Andersson, J. O., A. M. Sjogren, et al. (2007). "A genomic survey of the fish parasite Spironucleus salmonicida indicates genomic plasticity among diplomonads and significant lateral gene transfer in eukaryote genome evolution." BMC Genomics 8: 51. Ankarklev, J., E. Hestvik, et al. (2012). "Common Coinfections of Giardia intestinalis and Helicobacter pylori in Non-Symptomatic Ugandan Children." PLoS Negl Trop Dis 6(8): e1780. Ankarklev, J., S. G. Svard, et al. (2012). "Allelic sequence heterozygosity in single Giardia parasites." BMC Microbiol 12: 65.

85 Arkhipova, I. R. and H. G. Morrison (2001). "Three retrotransposon families in the genome of Giardia lamblia: two telomeric, one dead." Proc Natl Acad Sci U S A 98(25): 14497-14502. Arner, E. S. J. (2010). "Selenoproteins-What unique properties can arise with selenocysteine in place of cysteine?" Exp Cell Res 316(8): 1296-1303. Banerjee, S., P. W. Robbins, et al. (2009). "Molecular characterization of nucleocytosolic O-GlcNAc transferases of Giardia lamblia and Cryptosporidium parvum." Glycobiology 19(4): 331-336. Benarroch, D., M. Jankowska-Anyszka, et al. (2010). "Cap analog substrates reveal three clades of cap guanine-N2 methyltransferases with distinct methyl acceptor specificities." Rna-a Publication of the Rna Society 16(1): 211-220. Benchimol, M. (2004). "Giardia lamblia: behavior of the nuclear envelope." Parasitology Research 94(4): 254-264. Benchimol, M. (2005). "The nuclei of Giardia lamblia - new ultrastructural observations." Archives of Microbiology 183(3): 160-168. Bernander, R., J. E. D. Palm, et al. (2001). "Genome ploidy in different stages of the Giardia lamblia life cycle." Cellular Microbiology 3(1): 55-62. Best, A. A., H. G. Morrison, et al. (2004). "Evolution of eukaryotic transcription: Insights from the genome of Giardia lamblia." Genome Research 14(8): 1537-1547. Birkeland, S. R., S. P. Preheim, et al. (2010). "Transcriptome analyses of the Giardia lamblia life cycle." Mol Biochem Parasitol 174(1): 62-65. Bonilla-Santiago, R., Z. Wu, et al. (2008). "Identification of growth inhibiting compounds in a Giardia lamblia high-throughput screen." Mol Biochem Parasitol 162(2): 149-154. Brinkmann, H., M. van der Giezen, et al. (2005). "An empirical assessment of long-branch attraction artefacts in deep eukaryotic phylogenomics." Syst Biol 54(5): 743-757. Brown, D. M., J. A. Upcroft, et al. (1998). "Anaerobic bacterial metabolism in the ancient eukaryote Giardia duodenalis." International Journal for Parasitology 28(1): 149-164. Carlton, J. M., R. P. Hirt, et al. (2007). "Draft genome sequence of the sexually transmitted pathogen Trichomonas vaginalis." Science 315(5809): 207-212. Carpenter, M. L., Z. J. Assaf, et al. (2012). "Nuclear inheritance and genetic exchange without meiosis in the binucleate parasite Giardia intestinalis." Journal of Cell Science 125(10): 2523-2532. Cavalier-Smith, T. (2002). "The phagotrophic origin of eukaryotes and phylogenetic classification of Protozoa." Int J Syst Evol Microbiol 52(Pt 2): 297-354. Cepicka, I., M. Kostka, et al. (2008). "Non-monophyly of Retortamonadida and high genetic diversity of the genus Chilomastix suggested by analysis of SSU rDNA." Mol Phylogenet Evol 48(2): 770-775.

86 Chatterjee, A., A. Carpentieri, et al. (2010). "Giardia Cyst Wall Protein 1 Is a Lectin That Binds to Curled Fibrils of the GalNAc Homopolymer." Plos Pathogens 6(8). Chen, C. Z., L. Kulakova, et al. (2011). "High-throughput Giardia lamblia viability assay using bioluminescent ATP content measurements." Antimicrob Agents Chemother 55(2): 667-675. Chuang, S. F., L. H. Su, et al. (2012). "Functional redundancy of two Pax- like proteins in transcriptional activation of cyst wall protein genes in Giardia lamblia." Plos One 7(2): e30614. Clark, C. G. and A. J. Roger (1995). "Direct evidence for secondary loss of mitochondria in Entamoeba histolytica." Proc Natl Acad Sci U S A 92(14): 6518-6521. Cooper, G. L., B. R. Charlton, et al. (2004). "Hexamita meleagridis (Spironucleus meleagridis) infection in chukar partridges associated with high mortality and intracellular trophozoites." Avian Dis 48(3): 706-710. Cooper, M. A., R. D. Adam, et al. (2007). "Population genetics provides evidence for recombination in Giardia." Current Biology 17(22): 1984-1988. Cotton, J. A., J. K. Beatty, et al. (2011). "Host parasite interactions and pathophysiology in Giardia infections." International Journal for Parasitology 41(9): 925-933. Cox, C. J., P. G. Foster, et al. (2008). "The archaebacterial origin of eukaryotes." Proc Natl Acad Sci U S A 105(51): 20356-20361. Crossley, R. and D. V. Holberton (1983). "Characterization of proteins from the cytoskeleton of Giardia lamblia." Journal of Cell Science 59: 81- 103. Crossley, R., J. Marshall, et al. (1986). "Immunocytochemical Differentiation of Microtubules in the Cytoskeleton of Giardia- Lamblia Using Monoclonal-Antibodies to Alpha-Tubulin and Polyclonal Antibodies to Associated Low-Molecular-Weight Proteins." Journal of Cell Science 80: 233-252. Davids, B. J., K. Mehta, et al. (2004). "Dependence of Giardia lamblia encystation on novel transglutaminase activity." Mol Biochem Parasitol 136(2): 173-180. Davids, B. J., J. E. Palm, et al. (2006). "Polymeric immunoglobulin receptor in intestinal immune defense against the lumen-dwelling protozoan parasite Giardia." J Immunol 177(9): 6281-6290. Davids, B. J., D. S. Reiner, et al. (2006). "A new family of giardial cysteine- rich non-VSP protein genes and a novel cyst protein." Plos One 1: e44. Davids, B. J., S. Williams, et al. (2008). "Giardia lamblia aurora kinase: A regulator of mitosis in a binucleate parasite." International Journal for Parasitology 38(3-4): 353-369. Dawson, S. C. (2010). "An insider's guide to the microtubule cytoskeleton of Giardia." Cellular Microbiology 12(5): 588-598.

87 Dawson, S. C. and S. A. House (2010). "Life with eight flagella: flagellar assembly and division in Giardia." Current Opinion in Microbiology 13(4): 480-490. Dawson, S. C., J. K. Pham, et al. (2008). "Stable transformation of an episomal protein-tagging shuttle vector in the piscine diplomonad Spironucleus vortens." BMC Microbiol 8: 71. de Carvalho, T. B., E. B. David, et al. (2008). "Protease activity in extracellular products secreted in vitro by trophozoites of Giardia duodenalis." Parasitology Research 104(1): 185-190. de Graaf, R. M., G. Ricard, et al. (2011). "The organellar genome and metabolic potential of the hydrogen-producing mitochondrion of Nyctotherus ovalis." Molecular Biology and Evolution 28(8): 2379- 2391. Deitsch, K. W., S. A. Lukehart, et al. (2009). "Common strategies for antigenic variation by bacterial, fungal and protozoan pathogens." Nat Rev Microbiol 7(7): 493-503. Di Matteo, A., F. M. Scandurra, et al. (2008). "The O2-scavenging flavodiiron protein in the human parasite Giardia intestinalis." Journal of Biological Chemistry 283(7): 4061-4068. Dobell, C. (1920). "The Discovery of the Intestinal Protozoa of Man." Proc R Soc Med 13(Sect Hist Med): 1-15. Dolezal, P., O. Smid, et al. (2005). "Giardia mitosomes and trichomonad hydrogenosomes share a common mode of protein targeting." Proceedings of the National Academy of Sciences of the United States of America 102(31): 10924-10929. Doolittle, W. F. (1998). "You are what you eat: a gene transfer ratchet could account for bacterial genes in eukaryotic nuclear genomes." Trends Genet 14(8): 307-311. Dubey, J. P., D. S. Lindsay, et al. (1998). "Structures of Toxoplasma gondii tachyzoites, bradyzoites, and sporozoites and biology and development of tissue cysts." Clin Microbiol Rev 11(2): 267-+. ECDC (2011). Annual epidemiological report 2011 - Reporting on 2009 surveillance data and 2010 epidemic intelligence data. Elias, E. V., R. Quiroga, et al. (2008). "Characterization of SNAREs Determines the Absence of a Typical Golgi Apparatus in the Ancient Eukaryote Giardia lamblia." Journal of Biological Chemistry 283(51): 35996-36010. Ellis, J. G., M. Davila, et al. (2003). "Potential involvement of extracellular signal-regulated kinase 1 and 2 in encystation of a primitive eukaryote, Giardia lamblia - Stage-specific activation and intracellular localization." Journal of Biological Chemistry 278(3): 1936-1945. Elmendorf, H. G., S. M. Singer, et al. (2001). "The abundance of sterile transcripts in Giardia lamblia." Nucleic Acids Res 29(22): 4674- 4683. Embley, T. M. and W. Martin (2006). "Eukaryotic evolution, changes and challenges." Nature 440(7084): 623-630.

88 Embley, T. M., M. van der Giezen, et al. (2003). "Mitochondria and hydrogenosomes are two forms of the same fundamental organelle." Philosophical Transactions of the Royal Society of London Series B- Biological Sciences 358(1429): 191-202. Emelyanov, V. V. and A. V. Goldberg (2011). "Fermentation enzymes of Giardia intestinalis, pyruvate:ferredoxin oxidoreductase and hydrogenase, do not localize to its mitosomes." Microbiology-Sgm 157: 1602-1611. Ey, P. L., M. Mansouri, et al. (1997). "Genetic analysis of Giardia from hoofed farm animals reveals artiodactyl-specific and potentially zoonotic genotypes." J Eukaryot Microbiol 44(6): 626-635. Faghiri, Z., R. B. Santiago, et al. (2011). "High-throughput screening in suboptimal growth conditions identifies agonists of Giardia lamblia proliferation." Parasitology 138(2): 194-200. Farthing, M. J., M. E. Pereira, et al. (1986). "Description and characterization of a surface lectin from Giardia lamblia." Infect Immun 51(2): 661-667. Faso, C., C. Konrad, et al. (2012). "Export of cyst wall material and Golgi organelle neogenesis in Giardia lamblia depends on ER exit sites." Cellular Microbiology. Ford, B. (2007). "Antony van Leeuwenhoek's microscope and the discovery of Giardia." Microscopy and Analysis 21(4): 5-7. Frank, M. and K. Deitsch (2006). "Activation, silencing and mutually exclusive expression within the var gene family of Plasmodium falciparum." International Journal for Parasitology 36(9): 975-985. Fritz-Laylin, L. K., S. E. Prochnik, et al. (2010). "The Genome of Naegleria gruberi Illuminates Early Eukaryotic Versatility." Cell 140(5): 631- 642. Gavin, A. C., M. Bosche, et al. (2002). "Functional organization of the yeast proteome by systematic analysis of protein complexes." Nature 415(6868): 141-147. Germot, A., H. Philippe, et al. (1996). "Presence of a mitochondrial-type 70- kDa heat shock protein in Trichomonas vaginalis suggests a very early mitochondrial endosymbiosis in eukaryotes." Proceedings of the National Academy of Sciences of the United States of America 93(25): 14614-14617. Germot, A., H. Philippe, et al. (1997). "Evidence for loss of mitochondria in Microsporidia from a mitochondrial-type HSP70 in Nosema locustae." Mol Biochem Parasitol 87(2): 159-168. Ghosh, S., M. Frisardi, et al. (2001). "How Giardia swim and divide." Infect Immun 69(12): 7866-7872. Gibson, W., L. Peacock, et al. (2008). "The use of yellow fluorescent hybrids to indicate mating in Trypanosoma brucei." Parasit Vectors 1. Gillin, F. D., S. E. Boucher, et al. (1989). "Giardia-Lamblia - the Roles of Bile, Lactic-Acid, and Ph in the Completion of the Life-Cycle Invitro." Experimental Parasitology 69(2): 164-174.

89 Gillin, F. D., P. Hagblom, et al. (1990). "Isolation and expression of the gene for a major surface protein of Giardia lamblia." Proc Natl Acad Sci U S A 87(12): 4463-4467. Gillin, F. D., D. S. Reiner, et al. (1987). "Encystation and Expression of Cyst Antigens by Giardia-Lamblia Invitro." Science 235(4792): 1040- 1043. Gourguechon, S. and W. Z. Cande (2011). "Rapid tagging and integration of genes in Giardia intestinalis." Eukaryotic Cell 10(1): 142-145. Guo, F. C. and P. T. K. Woo (2004). "Experimental infections of Atlantic salmon Salmo salar with Spironucleus barkhanus." Diseases of Aquatic Organisms 61(1-2): 59-66. Haas, B. J., S. Kamoun, et al. (2009). "Genome sequence and analysis of the Irish potato famine pathogen Phytophthora infestans." Nature 461(7262): 393-398. Hagen, K. D., M. P. Hirakawa, et al. (2011). "Novel structural components of the ventral disc and lateral crest in Giardia intestinalis." PLoS Negl Trop Dis 5(12): e1442. Hampl, V., L. Hug, et al. (2009). "Phylogenomic analyses support the monophyly of Excavata and resolve relationships among eukaryotic "supergroups"." Proc Natl Acad Sci U S A 106(10): 3859-3864. Hampl, V., J. D. Silberman, et al. (2008). "Genetic Evidence for a Mitochondriate Ancestry in the 'Amitochondriate' Flagellate Trimastix pyriformis." Plos One 3(1). Hausmann, S., M. A. Altura, et al. (2005). "Yeast-like mRNA capping apparatus in Giardia lamblia." Journal of Biological Chemistry 280(13): 12077-12086. Hausmann, S., A. Ramirez, et al. (2007). "Biochemical and genetic analysis of RNA cap guanine-N2 methyltransferases from Giardia lamblia and Schizosaccharomyces pombe." Nucleic Acids Res 35(5): 1411- 1420. Hausmann, S. and S. Shuman (2005). "Giardia lamblia RNA cap guanine-N2 methyltransferase (Tgs2)." Journal of Biological Chemistry 280(37): 32101-32106. Holberton, D. V. (1973). "Mechanism of attachment of Giardia to the wall of the small intestine." Trans R Soc Trop Med Hyg 67(1): 29-30. Horner, D. S., P. G. Foster, et al. (2000). "Iron hydrogenases and the evolution of anaerobic eukaryotes." Molecular Biology and Evolution 17(11): 1695-1709. House, S. A., D. J. Richter, et al. (2011). "Giardia flagellar motility is not directly required to maintain attachment to surfaces." Plos Pathogens 7(8): e1002167. Huang, Y. C., L. H. Su, et al. (2008). "Regulation of cyst wall protein promoters by Myb2 in Giardia lamblia." Journal of Biological Chemistry 283(45): 31021-31029. Hudson, A. J., A. N. Moore, et al. (2012). "Evolutionarily divergent spliceosomal snRNAs and a conserved non-coding RNA processing motif in Giardia lamblia." Nucleic Acids Res.

90 Hug, L. A., A. Stechmann, et al. (2010). "Phylogenetic distributions and histories of proteins involved in anaerobic pyruvate metabolism in eukaryotes." Molecular Biology and Evolution 27(2): 311-324. Humen, M. A., P. F. Perez, et al. (2011). "Lipid raft-dependent adhesion of Giardia intestinalis trophozoites to a cultured human enterocyte-like Caco-2/TC7 cell monolayer leads to cytoskeleton-dependent functional injuries." Cellular Microbiology 13(11): 1683-1702. Iyer, L. M., V. Anantharaman, et al. (2008). "Comparative genomics of transcription factors and chromatin proteins in parasitic protists and other eukaryotes." International Journal for Parasitology 38(1): 1-31. Januschka, M. M., S. L. Erlandsen, et al. (1988). "A comparison of Giardia microti and Spironucleus muris cysts in the vole: an immunocytochemical, light, and electron microscopic study." J Parasitol 74(3): 452-458. Jarroll, E. L., P. J. Muller, et al. (1981). "Lipid and carbohydrate metabolism of Giardia lamblia." Mol Biochem Parasitol 2(3-4): 187-196. Jedelsky, P. L., P. Dolezal, et al. (2011). "The Minimal Proteome in the Reduced Mitochondrion of the Parasitic Protist Giardia intestinalis." Plos One 6(2). Jenikova, G., P. Hruz, et al. (2011). "Alpha1-giardin based live heterologous vaccine protects against Giardia lamblia infection in a murine model." Vaccine 29(51): 9529-9537. Jenkins, M. C., C. N. O'Brien, et al. (2009). "Antibodies to the ventral disc protein delta-giardin prevent in vitro binding of Giardia lamblia trophozoites." J Parasitol 95(4): 895-899. Jerlstrom-Hultqvist, J., O. Franzen, et al. (2010). "Genome analysis and comparative genomics of a Giardia intestinalis assemblage E isolate." BMC Genomics 11: 543. Jimenez-Garcia, L. F., G. Zavala, et al. (2008). "Identification of nucleoli in the early branching protist Giardia duodenalis." International Journal for Parasitology 38(11): 1297-1304. Jimenez, J. C., G. Uzcanga, et al. (2000). "Identification and partial characterization of excretory/secretory products with proteolytic activity in Giardia intestinalis." J Parasitol 86(4): 859-862. Joint Genome Institute. (2012). "Spironucleus vortens assembly and EST sequences." Retrieved 2012-08-22, 2012, from http://genome.jgi- psf.org/Spivo0/Spivo0.download.ftp.html. Jorgensen, A. and E. Sterud (2006). "The marine pathogenic genotype of Spironucleus barkhanus from farmed salmonids redescribed as Spironucleus salmonicida n. sp." J Eukaryot Microbiol 53(6): 531- 541. Jorgensen, A. and E. Sterud (2007). "Phylogeny of Spironucleus (Eopharyngia: Diplomonadida: Hexamitinae)." Protist 158(2): 247- 254.

91 Jorgensen, A., K. Torp, et al. (2011). "Wild arctic char Salvelinus alpinus and trout Salmo trutta: hosts and reservoir of the salmonid pathogen Spironucleus salmonicida (Diplomonadida; Hexamitidae)." Dis Aquat Organ 97(1): 57-63. Kabir, A. M., Y. Aiba, et al. (1997). "Prevention of Helicobacter pylori infection by lactobacilli in a gnotobiotic murine model." Gut 41(1): 49-55. Kabnick, K. S. and D. A. Peattie (1990). "In situ analyses reveal that the two nuclei of Giardia lamblia are equivalent." Journal of Cell Science 95 ( Pt 3): 353-360. Kamikawa, R., Y. Inagaki, et al. (2011). "Split introns in the genome of Giardia intestinalis are excised by spliceosome-mediated trans- splicing." Curr Biol 21(4): 311-315. Karanis, P. and P. L. Ey (1998). "Characterization of axenic isolates of Giardia intestinalis established from humans and animals in Germany." Parasitology Research 84(6): 442-449. Karr, C. D. and E. L. Jarroll (2004). "Cyst wall synthase: N- acetylgalactosaminyltransferase activity is induced to form the novel N-acetylgalactosamine polysaccharide in the Giardia cyst wall." Microbiology 150(Pt 5): 1237-1243. Keeling, P. J. and W. F. Doolittle (1996). "A non-canonical genetic code in an early diverging eukaryotic lineage." EMBO Journal 15(9): 2285- 2290. Keeling, P. J. and W. F. Doolittle (1997). "Widespread and ancient distribution of a noncanonical genetic code in diplomonads." Molecular Biology and Evolution 14(9): 895-901. Kent, M. L., J. Ellis, et al. (1992). "Systemic hexamitid (Protozoa: Diplomonadida) infection in seawater pen-reared chinook salmon Oncorhynchus tshawytscha." Diseases of Aquatic Organisms 14(2): 81-89. Knodler, L. A., P. J. Schofield, et al. (1995). "L-arginine transport and metabolism in Giardia intestinalis support its position as a transition between the prokaryotic and eukaryotic kingdoms." Microbiology 141 ( Pt 9): 2063-2070. Knodler, L. A., S. G. Svard, et al. (1999). "Developmental gene regulation in Giardia lamblia: first evidence for an encystation-specific promoter and differential 5' mRNA processing." Molecular Microbiology 34(2): 327-340. Kolisko, M., I. Cepicka, et al. (2008). "Molecular phylogeny of diplomonads and enteromonads based on SSU rRNA, alpha-tubulin and HSP90 genes: implications for the evolutionary history of the double karyomastigont of diplomonads." BMC Evolutionary Biology 8(205): 1-14. Kolisko, M., J. D. Silberman, et al. (2010). "A wide diversity of previously undetected free-living relatives of diplomonads isolated from marine/saline habitats." Environ Microbiol 12(10): 2700-2710.

92 Konrad, C., C. Spycher, et al. (2010). "Selective Condensation Drives Partitioning and Sequential Secretion of Cyst Wall Proteins in Differentiating Giardia lamblia." Plos Pathogens 6(4). Krauer, K. G., A. G. Burgess, et al. (2010). "Sequence map of the 2 Mb Giardia lamblia assemblage A chromosome." J Parasitol 96(3): 660- 662. Kulakova, L., S. M. Singer, et al. (2006). "Epigenetic mechanisms are involved in the control of Giardia lamblia antigenic variation." Molecular Microbiology 61(6): 1533-1542. Kurnaz, M. L., T. Bilgin, et al. (2010). "Certain Non-Standard Coding Tables Appear to be More Robust to Error Than the Standard Genetic Code." Journal of Molecular Evolution 70(1): 13-28. Lane, S. and D. Lloyd (2002). "Current trends in research into the waterborne parasite Giardia." Critical Reviews in Microbiology 28(2): 123-147. Laoworawit, P., C. S. Lee, et al. (1993). "Deoxynucleoside kinases of Giardia intestinalis." Mol Biochem Parasitol 60(1): 37-44. Lasek-Nesselquist, E., D. M. Welch, et al. (2009). "Genetic Exchange Within and Between Assemblages of Giardia duodenalis." Journal of Eukaryotic Microbiology 56(6): 504-518. Le Blancq, S. M. and R. D. Adam (1998). "Structural basis of karyotype heterogeneity in Giardia lamblia." Mol Biochem Parasitol 97(1-2): 199-208. Lebbad, M., J. G. Mattsson, et al. (2010). "From mouse to moose: multilocus genotyping of Giardia isolates from various animal species." Vet Parasitol 168(3-4): 231-239. Lebbad, M., I. Petersson, et al. (2011). "Multilocus genotyping of human Giardia isolates suggests limited zoonotic transmission and association between assemblage B and flatulence in children." PLoS Negl Trop Dis 5(8): e1262. Leitsch, D., A. G. Burgess, et al. (2011). "Pyruvate:ferredoxin oxidoreductase and thioredoxin reductase are involved in 5- nitroimidazole activation while flavin metabolism is linked to 5- nitroimidazole resistance in Giardia lamblia." Journal of Antimicrobial Chemotherapy 66(8): 1756-1765. Lev, B., H. Ward, et al. (1986). "Lectin activation in Giardia lamblia by host protease: a novel host-parasite interaction." Science 232(4746): 71- 73. Li, J., A. Mahajan, et al. (2006). "Ankyrin repeat: a unique motif mediating protein-protein interactions." Biochemistry 45(51): 15168-15178. Li, L. and C. C. Wang (2005). "Identification in the ancient protist Giardia lamblia of two eukaryotic translation initiation factor 4E homologues with distinctive functions." Eukaryotic Cell 4(5): 948- 959. Li, W., A. A. Saraiya, et al. (2011). "Gene Regulation in Giardia lambia Involves a Putative MicroRNA Derived from a Small Nucleolar RNA." Plos Neglected Tropical Diseases 5(10).

93 Li, W., A. A. Saraiya, et al. (2012). "The profile of snoRNA-derived microRNAs that regulate expression of variant surface proteins in Giardia lamblia." Cellular Microbiology 14(9): 1455-1473. Lill, R. and G. Kispal (2000). "Maturation of cellular Fe-S proteins: an essential function of mitochondria." Trends in Biochemical Sciences 25(8): 352-356. Lindmark, D. G. (1980). "Energy metabolism of the anaerobic protozoon Giardia lamblia." Mol Biochem Parasitol 1(1): 1-12. Lindmark, D. G. and M. Muller (1973). "Hydrogenosome, a cytoplasmic organelle of the anaerobic flagellate Tritrichomonas foetus, and its role in pyruvate metabolism." Journal of Biological Chemistry 248(22): 7724-7728. Lindmark, D. G., M. Müller, et al. (1975). "Hydrogenosomes in Trichomonas vaginalis." The Journal of Parasitology 61(3): 552-554. Lloyd, D., J. C. Harris, et al. (2003). "Nitrosative stress induced cytotoxicity in Giardia intestinalis." J Appl Microbiol 95(3): 576-583. Lloyd, D., J. R. Ralphs, et al. (2002). "Hydrogen production in Giardia intestinalis, a eukaryote with no hydrogenosomes." Trends Parasitol 18(4): 155-156. Lobanov, A. V., A. A. Turanov, et al. (2010). "Dual functions of codons in the genetic code." Crit Rev Biochem Mol Biol 45(4): 257-265. Lu, S. Q., A. C. Baruch, et al. (1998). "Molecular comparison of Giardia lamblia isolates." International Journal for Parasitology 28(9): 1341- 1345. Lujan, H. D., M. R. Mowatt, et al. (1996). "Cholesterol starvation induces differentiation of the intestinal parasite Giardia lamblia." Proceedings of the National Academy of Sciences of the United States of America 93(15): 7628-7633. Ma'ayeh, S. Y. and P. T. Brook-Carter (2012). "Representational difference analysis identifies specific genes in the interaction of Giardia duodenalis with the murine intestinal epithelial cell line, IEC-6." International Journal for Parasitology 42(5): 501-509. Mai, Z. M., S. Ghosh, et al. (1999). "Hsp60 is targeted to a cryptic mitochondrion-derived organelle ("crypton") in the microaerophilic protozoan parasite Entamoeba histolytica." Molecular and Cellular Biology 19(3): 2198-2205. Malik, S. B., A. W. Pightling, et al. (2008). "An Expanded Inventory of Conserved Meiotic Genes Provides Evidence for Sex in Trichomonas vaginalis." Plos One 3(8). Manning, G., D. S. Reiner, et al. (2011). "The minimal kinome of Giardia lamblia illuminates early kinase evolution and unique parasite biology." Genome Biology 12(7). Maralikova, B., V. Ali, et al. (2010). "Bacterial-type oxygen detoxification and iron-sulfur cluster assembly in amoebal relict mitochondria." Cellular Microbiology 12(3): 331-342.

94 Marti, M., A. Regos, et al. (2003). "An ancestral secretory apparatus in the protozoan parasite Giardia intestinalis." Journal of Biological Chemistry 278(27): 24837-24848. Mastronicola, D., A. Giuffre, et al. (2011). "Giardia intestinalis escapes oxidative stress by colonizing the small intestine: A molecular hypothesis." IUBMB Life. Mastronicola, D., F. Testa, et al. (2010). "Flavohemoglobin and nitric oxide detoxification in the human protozoan parasite Giardia intestinalis." Biochem Biophys Res Commun 399(4): 654-658. Millet, C. O., J. Cable, et al. (2010). "The diplomonad fish parasite Spironucleus vortens produces hydrogen." J Eukaryot Microbiol 57(5): 400-404. Millet, C. O., D. Lloyd, et al. (2011). "Carbohydrate and amino acid metabolism of Spironucleus vortens." Experimental Parasitology 129(1): 17-26. Millet, C. O., D. Lloyd, et al. (2011). "In vitro culture of the diplomonad fish parasite Spironucleus vortens reveals unusually fast doubling time and atypical biphasic growth." J Fish Dis 34(1): 71-73. Monis, P. T., S. M. Caccio, et al. (2009). "Variation in Giardia: towards a taxonomic revision of the genus." Trends Parasitol 25(2): 93-100. Morf, L. and U. Singh (2012). "Entamoeba histolytica: a snapshot of current research and methods for genetic analysis." Current Opinion in Microbiology 15(4): 469-475. Morf, L., C. Spycher, et al. (2010). "The Transcriptional Response to Encystation Stimuli in Giardia lamblia Is Restricted to a Small Set of Genes." Eukaryotic Cell 9(10): 1566-1576. Morrison, H. G., A. G. McArthur, et al. (2007). "Genomic minimalism in the early diverging intestinal parasite Giardia lamblia." Science 317(5846): 1921-1926. Muller, J., M. Sterk, et al. (2007). "Characterization of Giardia lamblia WB C6 clones resistant to nitazoxanide and to metronidazole." J Antimicrob Chemother 60(2): 280-287. Nageshan, R. K., N. Roy, et al. (2011). "Post-transcriptional repair of a split heat shock protein 90 gene by mRNA trans-splicing." Journal of Biological Chemistry 286(9): 7116-7122. Nash, T. E., S. M. Banks, et al. (1990). "Frequency of Variant Antigens in Giardia-Lamblia." Experimental Parasitology 71(4): 415-421. Nash, T. E., D. A. Herrington, et al. (1990). "Antigenic Variation of Giardia- Lamblia in Experimental Human Infections." J Immunol 144(11): 4362-4369. Nash, T. E., D. A. Herrington, et al. (1987). "Experimental Human Infections with Giardia-Lamblia." Journal of Infectious Diseases 156(6): 974-984. Nash, T. E., H. T. Lujan, et al. (2001). "Variant-specific surface protein switching in Giardia lamblia." Infect Immun 69(3): 1922-1923.

95 Nash, T. E. and M. R. Mowatt (1992). "Characterization of a Giardia lamblia variant-specific surface protein (VSP) gene from isolate GS/M and estimation of the VSP gene repertoire size." Mol Biochem Parasitol 51(2): 219-227. Nillius, D., J. Muller, et al. (2011). "Nitroreductase (GlNR1) increases susceptibility of Giardia lamblia and Escherichia coli to nitro drugs." J Antimicrob Chemother 66(5): 1029-1035. Niu, X. H., T. Hartshorne, et al. (1994). "Characterization of putative small nuclear RNAs from Giardia lamblia." Mol Biochem Parasitol 66(1): 49-57. Nixon, J. E., A. Wang, et al. (2002). "A spliceosomal intron in Giardia lamblia." Proc Natl Acad Sci U S A 99(6): 3701-3705. Nohria, A., R. A. Alonso, et al. (1992). "Identification and Characterization of Gamma-Giardin and the Gamma-Giardin Gene from Giardia- Lamblia." Mol Biochem Parasitol 56(1): 27-38. Nohynkova, E., P. Tumova, et al. (2006). "Cell division of giardia intestinalis: Flagellar developmental cycle involves transformation and exchange of flagella between mastigonts of a diplomonad cell." Eukaryotic Cell 5(4): 753-761. Pal, D., S. Banerjee, et al. (2009). "Giardia, Entamoeba, and Trichomonas enzymes activate metronidazole (nitroreductases) and inactivate metronidazole (nitroimidazole reductases)." Antimicrob Agents Chemother 53(2): 458-464. Palm, D., M. Weiland, et al. (2005). "Developmental changes in the adhesive disk during Giardia differentiation." Mol Biochem Parasitol 141(2): 199-207. Palm, J. E., M. E. Weiland, et al. (2003). "Identification of immunoreactive proteins during acute human giardiasis." Journal of Infectious Diseases 187(12): 1849-1859. Pan, Y. J., C. C. Cho, et al. (2009). "A novel WRKY-like protein involved in transcriptional activation of cyst wall protein genes in Giardia lamblia." Journal of Biological Chemistry 284(27): 17975-17988. Panaro, M. A., A. Cianciulli, et al. (2007). "Caspase-dependent apoptosis of the HCT-8 epithelial cell line induced by the parasite Giardia intestinalis." FEMS Immunol Med Microbiol 51(2): 302-309. Papanastasiou, P., A. Hiltpold, et al. (1996). "The release of the variant surface protein of Giardia to its soluble isoform is mediated by the selective cleavage of the conserved carboxy-terminal domain." Biochemistry 35(31): 10143-10148. Papanastasiou, P., M. J. McConville, et al. (1997). "The variant-specific surface protein of Giardia, VSP4A1, is a glycosylated and palmitoylated protein." Biochem J 322 ( Pt 1): 49-56. Paredez, A. R., Z. J. Assaf, et al. (2011). "An actin cytoskeleton with evolutionarily conserved functions in the absence of canonical actin- binding proteins." Proceedings of the National Academy of Sciences of the United States of America 108(15): 6151-6156.

96 Parfrey, L. W., E. Barbero, et al. (2006). "Evaluating support for the current classification of eukaryotic diversity." PLoS Genet 2(12): e220. Parfrey, L. W., D. J. G. Lahr, et al. (2011). "Estimating the timing of early eukaryotic diversification with multigene molecular clocks." Proceedings of the National Academy of Sciences of the United States of America 108(33): 13624-13629. Park, J. S., M. Kolisko, et al. (2009). "Light microscopic observations, ultrastructure, and molecular phylogeny of Hicanonectes teleskopos n. g., n. sp., a deep-branching relative of diplomonads." J Eukaryot Microbiol 56(4): 373-384. Park, J. S., M. Kolisko, et al. (2010). "Cell morphology and formal description of Ergobibamus cyprinoides n. g., n. sp., another Carpediemonas-like relative of diplomonads." J Eukaryot Microbiol 57(6): 520-528. Paull, G. C. and R. A. Matthews (2001). "Spironucleus vortens, a possible cause of hole-in-the-head disease in cichlids." Dis Aquat Organ 45(3): 197-202. Paz-Maldonado, M. T., R. Arguello-Garcia, et al. (2012). "Proteomic and transcriptional analyses of genes differentially expressed in Giardia duodenalis clones resistant to albendazole." Infect Genet Evol. Perez-Brocal, V. and C. G. Clark (2008). "Analysis of two genomes from the mitochondrion-like organelle of the intestinal parasite Blastocystis: complete sequences, gene content, and genome organization." Molecular Biology and Evolution 25(11): 2475-2482. Perry, D. A., H. G. Morrison, et al. (2011). "Optical map of the genotype A1 WB C6 Giardia lamblia genome isolate." Mol Biochem Parasitol 180(2): 112-114. Pimenta, P. F. P., P. P. Dasilva, et al. (1991). "Variant Surface-Antigens of Giardia-Lamblia Are Associated with the Presence of a Thick Cell Coat - Thin-Section and Label Fracture Immunocytochemistry Survey." Infect Immun 59(11): 3989-3996. Poppe, T. T., T. A. Mo, et al. (1992). "Disseminated hexamitosis in sea- caged Atlantic salmon Salmo salar." Diseases of Aquatic Organisms 14(2): 91-97. Poxleitner, M. K., M. L. Carpenter, et al. (2008). "Evidence for karyogamy and exchange of genetic material in the binucleate intestinal parasite Giardia intestinalis." Science 319(5869): 1530-1533. Poynton, S. L., M. R. S. Fard, et al. (2004). "Ultrastructure of Spironucleus salmonis n. comb. (formerly Octomitus salmonis sensu Moore 1922, Davis 1926, and Hexamita salmonis sensu Ferguson 1979), with a guide to Spironucleus species." Diseases of Aquatic Organisms 60(1): 49-64. Prucca, C. G., F. D. Rivero, et al. (2011). "Regulation of Antigenic Variation in Giardia lamblia." Annual Review of Microbiology, Vol 65 65: 611-630. Prucca, C. G., I. Slavin, et al. (2008). "Antigenic variation in Giardia lamblia is regulated by RNA interference." Nature 456(7223): 750-754.

97 Puig, O., F. Caspary, et al. (2001). "The tandem affinity purification (TAP) method: a general procedure of protein complex purification." Methods 24(3): 218-229. Rafferty, S., B. Luu, et al. (2010). "Giardia lamblia encodes a functional flavohemoglobin." Biochem Biophys Res Commun 399(3): 347- 351. Ramesh, M. A., S. B. Malik, et al. (2005). "A phylogenomic inventory of meiotic genes; evidence for sex in Giardia and an early eukaryotic origin of meiosis." Curr Biol 15(2): 185-191. Ratner, D. M., J. Cui, et al. (2008). "Changes in the N-Glycome, Glycoproteins with Asn-Linked Glycans, of Giardia lamblia with Differentiation from Trophozoites to Cysts." Eukaryotic Cell 7(11): 1930-1940. Reese, M. L. and J. P. Boyle (2012). "Virulence without catalysis: how can a pseudokinase affect host cell signaling?" Trends Parasitol 28(2): 53- 57. Regoes, A., D. Zourmpanou, et al. (2005). "Protein import, replication, and inheritance of a vestigial mitochondrion." Journal of Biological Chemistry 280(34): 30557-30563. Reiner, D. S., J. Ankarklev, et al. (2008). "Synchronisation of Giardia lamblia: identification of cell cycle stage-specific genes and a differentiation restriction point." Int J Parasitol 38(8-9): 935-944. Reiner, D. S., J. M. McCaffery, et al. (2001). "Reversible interruption of Giardia lamblia cyst wall protein transport in a novel regulated secretory pathway." Cellular Microbiology 3(7): 459-472. Rendtorff, R. C. (1954). "The experimental transmission of human intestinal protozoan parasites. II. Giardia lamblia cysts given in capsules." Am J Hyg 59(2): 209-220. Rigaut, G., A. Shevchenko, et al. (1999). "A generic protein purification method for protein complex characterization and proteome exploration." Nat Biotechnol 17(10): 1030-1032. Ringqvist, E., L. Avesson, et al. (2011). "Transcriptional changes in Giardia during host-parasite interactions." International Journal for Parasitology 41(3-4): 277-285. Ringqvist, E., J. E. D. Palm, et al. (2008). "Release of metabolic enzymes by Giardia in response to interaction with intestinal epithelial cells." Mol Biochem Parasitol 159(2): 85-91. Rivero, F. D., A. Saura, et al. (2010). "Disruption of antigenic variation is crucial for effective parasite vaccine." Nature Medicine 16(5): 551- U583. Robbins, P. W. and J. Samuelson (2005). "Asparagine linked glycosylation in Giardia." Glycobiology 15(6): 15g-16g. Rodriguez-Fuentes, G. B., R. Cedillo-Rivera, et al. (2006). "Giardia duodenalis: analysis of secreted proteases upon trophozoite- epithelial cell interaction in vitro." Memorias Do Instituto Oswaldo Cruz 101(6): 693-696.

98 Roger, A. J., S. G. Svard, et al. (1998). "A mitochondrial-like chaperonin 60 gene in Giardia lamblia: Evidence that diplomonads once harbored an endosymbiont related to the progenitor of mitochondria." Proceedings of the National Academy of Sciences of the United States of America 95(1): 229-234. Roy, S. W., A. J. Hudson, et al. (2012). "Numerous fragmented spliceosomal introns, AT-AC splicing, and an unusual dynein gene expression pathway in Giardia lamblia." Molecular Biology and Evolution 29(1): 43-49. Russell, A. G., T. E. Shutt, et al. (2005). "An ancient spliceosomal intron in the ribosomal protein L7a gene (Rpl7a) of Giardia lamblia." BMC Evol Biol 5: 45. Sagolla, M. S., S. C. Dawson, et al. (2006). "Three-dimensional analysis of mitosis and cytokinesis in the binucleate parasite Giardia intestinalis." Journal of Cell Science 119(23): 4889-4900. Saraiya, A. A., W. Li, et al. (2011). "A microRNA derived from an apparent canonical biogenesis pathway regulates variant surface protein gene expression in Giardia lamblia." Rna-a Publication of the Rna Society 17(12): 2152-2164. Saraiya, A. A. and C. C. Wang (2008). "snoRNA, a Novel Precursor of microRNA in Giardia lamblia." Plos Pathogens 4(11). Saric, M., A. Vahrmann, et al. (2009). "Dual acylation accounts for the localization of {alpha}19-giardin in the ventral flagellum pair of Giardia lamblia." Eukaryotic Cell 8(10): 1567-1574. Savioli, L., H. Smith, et al. (2006). "Giardia and Cryptosporidium join the 'Neglected Diseases Initiative'." Trends Parasitol 22(5): 203-208. Schofield, P. J., M. Costello, et al. (1990). "The arginine dihydrolase pathway is present in Giardia intestinalis." International Journal for Parasitology 20(5): 697-699. Schofield, P. J., M. R. Edwards, et al. (1992). "The pathway of arginine catabolism in Giardia intestinalis." Mol Biochem Parasitol 51(1): 29- 36. Schurko, A. M., M. Neiman, et al. (2009). "Signs of sex: what we know and how we know it." Trends in Ecology & Evolution 24(4): 208-217. Shiflett, A. M. and P. J. Johnson (2010). "Mitochondrion-related organelles in eukaryotic protists." Annual Review of Microbiology, Vol 65 64: 409-429. Simon, M. C. and H. J. Schmidt (2007). "Antigenic variation in ciliates: antigen structure, function, expression." J Eukaryot Microbiol 54(1): 1-7. Simpson, A. G., E. K. MacQuarrie, et al. (2002). "Eukaryotic evolution: early origin of canonical introns." Nature 419(6904): 270. Simpson, A. G. B. (2003). "Cytoskeletal organization, phylogenetic affinities and systematics in the contentious taxon Excavata (Eukaryota)." International Journal of Systematic and Evolutionary Microbiology 53: 1759-1777.

99 Singer, S. M. and T. E. Nash (2000). "The role of normal flora in Giardia lamblia infections in mice." Journal of Infectious Diseases 181(4): 1510-1512. Singer, S. M., J. Yee, et al. (1998). "Episomal and integrated maintenance of foreign DNA in Giardia lamblia." Mol Biochem Parasitol 92(1): 59- 69. Skarin, H., E. Ringqvist, et al. (2011). "Elongation factor 1-alpha is released into the culture medium during growth of Giardia intestinalis trophozoites." Experimental Parasitology 127(4): 804-810. Slamovits, C. H. and P. J. Keeling (2006). "A high density of ancient spliceosomal introns in oxymonad excavates." BMC Evol Biol 6: 34. Slavin, I., A. Saura, et al. (2002). "Dephosphorylation of cyst wall proteins by a secreted lysosomal acid phosphatase is essential for excystation of Giardia lamblia." Mol Biochem Parasitol 122(1): 95-98. Smid, O., A. Matuskova, et al. (2008). "Reductive Evolution of the Mitochondrial Processing Peptidases of the Unicellular Parasites Trichomonas vaginalis and Giardia intestinalis." Plos Pathogens 4(12). Solaymani-Mohammadi, S. and S. M. Singer (2011). "Host immunity and pathogen strain contribute to intestinal disaccharidase impairment following gut infection." J Immunol 187(7): 3769-3775. Sonda, S., S. Stefanic, et al. (2008). "A sphingolipid inhibitor induces a cytokinesis arrest and blocks stage differentiation in Giardia lamblia." Antimicrob Agents Chemother 52(2): 563-569. Stadelmann, B., M. C. Merino, et al. (2012). "Arginine Consumption by the Intestinal Parasite Giardia intestinalis Reduces Proliferation of Intestinal Epithelial Cells." Plos One 7(9): e45325. Stairs, C. W., A. J. Roger, et al. (2011). "Eukaryotic pyruvate formate lyase and its activating enzyme were acquired laterally from a Firmicute." Molecular Biology and Evolution 28(7): 2087-2099. Stefanic, S., L. Morf, et al. (2009). "Neogenesis and maturation of transient Golgi-like cisternae in a simple eukaryote." Journal of Cell Science 122(Pt 16): 2846-2856. Sterud, E. (1998). "In vitro cultivation and temperature-dependent growth of two strains of Spironucleus barkhanus (Diplomonadida: Hexamitidae) from Atlantic salmon Salmo salar and grayling Thymallus thymallus." Dis Aquat Organ 33(1): 57-61. Sterud, E., T. A. Mo, et al. (1997). "Ultrastructure of Spironucleus barkhanus n. sp. (Diplomonadida: Hexamitidae) from grayling Thymallus thymallus (L.) (Salmonidae) and Atlantic salmon Salmo salar L. (Salmonidae)." Journal of Eukaryotic Microbiology 44(5): 399-407. Sterud, E., T. Poppe, et al. (2003). "Intracellular infection with Spironucleus barkhanus (Diplomonadida: Hexamitidae) in farmed Arctic char Salvelinus alpinus." Dis Aquat Organ 56(2): 155-161.

100 Sterud, E. and S. L. Poynton (2002). "Spironucleus vortens (Diplomonadida) in the Ide, Leuciscus idus (L.) (Cyprinidae): a warm water hexamitid flagellate found in northern Europe." J Eukaryot Microbiol 49(2): 137-145. Su, L. H., Y. J. Pan, et al. (2011). "A novel E2F-like protein involved in transcriptional activation of cyst wall protein genes in Giardia lamblia." Journal of Biological Chemistry 286(39): 34101-34120. Sun, C. H., D. Palm, et al. (2002). "A novel Myb-related protein involved in transcriptional activation of encystation genes in Giardia lamblia." Molecular Microbiology 46(4): 971-984. Sun, C. H., L. H. Su, et al. (2006). "Novel plant-GARP-like transcription factors in Giardia lamblia." Mol Biochem Parasitol 146(1): 45-57. Sun, C. H. and J. H. Tai (2000). "Development of a tetracycline controlled gene expression system in the parasitic protozoan Giardia lamblia." Mol Biochem Parasitol 105(1): 51-60. Szkodowska, A., M. C. Muller, et al. (2002). "Annexin XXI (ANX21) of Giardia lamblia has sequence motifs uniquely sdhared by giardial annexins and is specifically localized in the flagella." Journal of Biological Chemistry 277(28): 25703-25706. Tachezy, J. and P. Dolezal (2011). The Giardia mitosomes. Giardia: A Model Organism. H. D. Lujan and G. S. Svärd, Springer- Verlag/Wien: 185-200. Tachezy, J., L. B. Sanchez, et al. (2001). "Mitochondrial type iron-sulfur cluster assembly in the amitochondriate eukaryotes Trichomonas vaginalis and Giardia intestinalis, as indicated by the phylogeny of IscS." Molecular Biology and Evolution 18(10): 1919-1928. Takishita, K., M. Kolisko, et al. (2012). "Multigene Phylogenies of Diverse Carpediemonas-like Organisms Identify the Closest Relatives of 'Amitochondriate' Diplomonads and Retortamonads." Protist 163(3): 344-355. Takishita, K., N. Yubuki, et al. (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(4): 563-576. Takumi, K., A. Swart, et al. (2012). "Population-based analyses of Giardia duodenalis is consistent with the clonal assemblage structure." Parasit Vectors 5: 168. Teodorovic, S., J. M. Braverman, et al. (2007). "Unusually low levels of genetic variation among Giardia lamblia isolates." Eukaryotic Cell 6(8): 1421-1430. Teodorovic, S., C. D. Walls, et al. (2007). "Bidirectional transcription is an inherent feature of Giardia lamblia promoters and contributes to an abundance of sterile antisense transcripts throughout the genome." Nucleic Acids Res 35(8): 2544-2553. Testa, F., D. Mastronicola, et al. (2011). "The superoxide reductase from the early diverging eukaryote Giardia intestinalis." Free Radic Biol Med 51(8): 1567-1574.

101 Thompson, R. C. and P. T. Monis (2004). "Variation in Giardia: implications for taxonomy and epidemiology." Adv Parasitol 58: 69-137. Touz, M. C., J. T. Conrad, et al. (2005). "A novel palmitoyl acyl transferase controls surface protein palmitoylation and cytotoxicity in Giardia lamblia." Molecular Microbiology 58(4): 999-1011. Touz, M. C., A. S. Ropolo, et al. (2008). "Arginine deiminase has multiple regulatory roles in the biology of Giardia lamblia." Journal of Cell Science 121(17): 2930-2938. Tovar, J., A. Fischer, et al. (1999). "The mitosome, a novel organelle related to mitochondria in the amitochondrial parasite Entamoeba histolytica." Molecular Microbiology 32(5): 1013-1021. Tovar, J., G. Leon-Avila, et al. (2003). "Mitochondrial remnant organelles of Giardia function in iron-sulphur protein maturation." Nature 426(6963): 172-176. Troeger, H., H. J. Epple, et al. (2007). "Effect of chronic Giardia lamblia infection on epithelial transport and barrier function in human duodenum." Gut 56(3): 328-335. Tumova, P., K. Hofstetrova, et al. (2007). "Cytogenetic evidence for diversity of two nuclei within a single diplomonad cell of Giardia." Chromosoma 116(1): 65-78. Upcroft, J. and P. Upcroft (1998). "My favorite cell: Giardia." Bioessays 20(3): 256-263. Upcroft, J. A., K. G. Krauer, et al. (2009). "Sequence map of the 3-Mb Giardia duodenalis assemblage A chromosome." Chromosome Res 17(8): 1001-1014. Vahrmann, A., M. Saric, et al. (2008). "alpha14-Giardin (annexin E1) is associated with tubulin in trophozoites of Giardia lamblia and forms local slubs in the flagella." Parasitology Research 102(2): 321-326. van der Giezen, M., S. Cox, et al. (2004). "The iron-sulfur cluster assembly genes iscS and iscU of Entamoeba histolytica were acquired by horizontal gene transfer." BMC Evol Biol 4: 7. van Grinsven, K. W., S. Rosnowsky, et al. (2008). "Acetate:succinate CoA- transferase in the hydrogenosomes of Trichomonas vaginalis: identification and characterization." Journal of Biological Chemistry 283(3): 1411-1418. Vanacova, S., W. Yan, et al. (2005). "Spliceosomal introns in the deep- branching eukaryote Trichomonas vaginalis." Proc Natl Acad Sci U S A 102(12): 4430-4435. Wang, C. H., L. H. Su, et al. (2007). "A novel ARID/Bright-like protein involved in transcriptional activation of cyst wall protein 1 gene in Giardia lamblia." Journal of Biological Chemistry 282(12): 8905- 8914. Wang, Y. T., Y. J. Pan, et al. (2010). "A novel pax-like protein involved in transcriptional activation of cyst wall protein genes in Giardia lamblia." Journal of Biological Chemistry 285(42): 32213-32226.

102 Weeratunga, S. K., A. Osman, et al. (2012). "Alpha-1 Giardin is an Annexin with Highly Unusual Calcium-Regulated Mechanisms." Journal of Molecular Biology 423(2): 169-181. Weiland, M. E., A. G. McArthur, et al. (2005). "Annexin-like alpha giardins: a new cytoskeletal gene family in Giardia lamblia." International Journal for Parasitology 35(6): 617-626. Wensaas, K. A., N. Langeland, et al. (2012). "Irritable bowel syndrome and chronic fatigue 3 years after acute giardiasis: historic cohort study." Gut 61(2): 214-219. Veraksa, A., A. Bauer, et al. (2005). "Analyzing protein complexes in Drosophila with tandem affinity purification-mass spectrometry." Dev Dyn 232(3): 827-834. Whatley, J. M., P. John, et al. (1979). "From extracellular to intracellular: the establishment of mitochondria and chloroplasts." Proc R Soc Lond B Biol Sci 204(1155): 165-187. WHO. (2009). "Fact sheet N°330 : Diarrhoeal disease." Retrieved 2012-09- 17, 2012, from http://www.who.int/mediacentre/factsheets/fs330/en/index.html. Vicente, J. B., F. Testa, et al. (2009). "Redox properties of the oxygen- detoxifying flavodiiron protein from the human parasite Giardia intestinalis." Arch Biochem Biophys 488(1): 9-13. Wiesehahn, G. P., E. L. Jarroll, et al. (1984). "Giardia-Lamblia - Autoradiographic Analysis of Nuclear Replication." Experimental Parasitology 58(1): 94-100. Williams, C. F., D. Lloyd, et al. (2012). "Disrupted intracellular redox balance of the diplomonad fish parasite Spironucleus vortens by 5- nitroimidazoles and garlic-derived compounds." Vet Parasitol. Williams, C. W. and H. G. Elmendorf (2011). "Identification and analysis of the RNA degrading complexes and machinery of Giardia lamblia using an in silico approach." BMC Genomics 12: 586. Wood, A. M. and H. V. Smith (2005). "Spironucleosis (Hexamitiasis, Hexamitosis) in the ring-necked pheasant (Phasianus colchicus): detection of cysts and description of Spironucleus meleagridis in stained smears." Avian Dis 49(1): 138-143. Yang, C. Y., H. Zhou, et al. (2005). "Identification of 20 snoRNA-like RNAs from the primitive eukaryote, Giardia lamblia." Biochem Biophys Res Commun 328(4): 1224-1231. Yee, J., A. Tang, et al. (2007). "Core histone genes of Giardia intestinalis: genomic organization, promoter structure, and expression." BMC Mol Biol 8: 26. Yichoy, M., T. T. Duarte, et al. (2011). "Lipid metabolism in Giardia: a post- genomic perspective." Parasitology 138(3): 267-278. Yichoy, M., E. S. Nakayasu, et al. (2009). "Lipidomic analysis reveals that phosphatidylglycerol and phosphatidylethanolamine are newly generated phospholipids in an early-divergent protozoan, Giardia lamblia." Mol Biochem Parasitol 165(1): 67-78.

103 Yoder, J. S., J. W. Gargano, et al. (2012). "Giardiasis surveillance - United States, 2009-2010." MMWR Surveill Summ 61(5): 13-23. Yu, D. C., A. L. Wang, et al. (1998). "Protein synthesis in Giardia lamblia may involve interaction between a downstream box (DB) in mRNA and an anti-DB in the 16S-like ribosomal RNA." Mol Biochem Parasitol 96(1-2): 151-165. Yu, L. Z., C. W. Birky, et al. (2002). "The two nuclei of Giardia each have complete copies of the genome and are partitioned equationally at cytokinesis." Eukaryotic Cell 1(2): 191-199. Yubuki, N., Y. Inagaki, et al. (2007). "Ultrastructure and ribosomal RNA phylogeny of the free-living heterotrophic flagellate Dysnectes brevis n. gen., n. sp., a new member of the Fornicata." J Eukaryot Microbiol 54(2): 191-200. Zanin, E., J. Dumont, et al. (2011). "Affinity purification of protein complexes in C. elegans." Methods Cell Biol 106: 289-322.

104

Acta Universitatis Upsaliensis Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 990 Editor: The Dean of the Faculty of Science and Technology

A doctoral dissertation from the Faculty of Science and Technology, Uppsala University, is usually a summary of a number of papers. A few copies of the complete dissertation are kept at major Swedish research libraries, while the summary alone is distributed internationally through the series Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology.

ACTA UNIVERSITATIS UPSALIENSIS Distribution: publications.uu.se UPPSALA urn:nbn:se:uu:diva-182831 2012