Phylogeny and molecular evolution of green algae - Phycology ...

Phylogeny and molecular evolution 

of green algae 

Ellen Cocquyt

Universiteit Gent 

Faculteit Wetenschappen, Vakgroep Biologie 

Onderzoeksgroep Algologie 

Phylogeny and molecular evolution of green algae 

Fylogenie en moleculaire evolutie van groenwieren 

Ellen Cocquyt 

Proefschrift voorgelegd tot het behalen van de graad van 

Doctor in de Wetenschappen: Biologie 

Academiejaar 2008-2009

Promotor: Prof. Dr. O. De Clerck (Universiteit Gent) 

Co-promotor: Dr. H. Verbruggen (Universiteit Gent) 

Leden van de leescommissie: 

Prof. Dr. K. Hoef-Emden (Universität zu Köln, Germany) 

Dr. P. Rouzé (Universiteit Gent) 

Prof. Dr. A. Vanderpoorten (Université de Liège) 

Overige leden van de examencommissie: 

Prof. Dr. K. Sabbe (Universiteit Gent) 

Prof. Dr. Erik Smets (Katholieke Universiteit Leuven and National Herbarium of the Nederlands) 

Prof. Dr. W. Vyverman (Universiteit Gent) 

Photographs cover: Frederik Leliaert and Heroen Verbruggen 

Photographs upper band, from left to right and from top to bottom: 

Boodlea, Phyllodictyon, Dictyosphaeria, Ulva, Cladophora and Valonia 

Photographs lower band, from left to right: 

Boergesenia, Codium, Halimeda, Dictyosphaeria and Cladophora 

Photographs at the back, from left to right and from top to bottom: 

Blastophysa, Trentepohlia, Ignatius and Chlamydomonas 

The research reported in this thesis was funded by the Special Research Fund (Ghent University, 

DOZA-01107605) and performed in the Research Group Phycology and the Center for Molecular 

Phylogenetics and Evolution, Biology Department, Ghent University, Krijgslaan 281-S8, B-9000, 

Ghent, Belgium. www.phycology.ugent.be

Dankwoord 

Vooreerst zou ik mijn promotor Olivier willen bedanken. Bijna negen jaar geleden stapte ik op het 

vliegtuig richting Zuid-Afrika. Ik mocht gedurende 2 maanden wieren inzamelen langsheen de Zuid- 

Afrikaanse kust en zou leren ‘kijken’ naar roodwieren, samen met Olivier die daar toen gedurende 

een jaar aan de Universiteit van Kaapstad werkte. Op de luchthaven van Kaapstad aangekomen vroeg 

Olivier verwondert: “Is dat alles wat je mee hebt?”, wijzend naar mijn klein rugzakje. Tja, er zat een 

rugby team op het vliegtuig en niet alle bagage was meegeraakt. De mijne stond nog in Londen. ’t Is 

gelukkig allemaal goed gekomen en ik heb daar een fantastisch tijd gehad! 

Eenmaal ik mijn licentiaatdiploma behaalde, ging ik nog even langs het labo om goedendag te 

zeggen. Olivier stelde toen voor om met een Marie Curie beurs een tijdje bij Christine Maggs aan de 

Queen’s University of Belfast te gaan werken. Hm, ik zou eigenlijk net gaan samenwonen met Toon. 

Uiteindelijk ben ik toch vertrokken voor een half jaartje. Daar heb ik voor het eerst DNA 

geëxtraheerd en PCR’s gedaan, en eveneens een fantastisch tijd beleeft. Terug in België mocht ik 

onder voorwaarde dat ik een IWT beurs zou aanvragen, beginnen als laborante bij onze 

onderzoeksgroep. Die IWT beurs werd niets, maar na anderhalf jaar kon ik dan toch beginnen aan 

een doctoraat met een BOF beurs. Het resultaat daarvan is dit doctoraat! 

Olivier, bedankt om me te begeleiden doorheen al die jaren. 

Ten tweede, zou ik mijn co-promotor Heroen willen bedanken. Als laborante heb ik voor zijn 

doctoraat veel praktisch werk gedaan, maar het laatste anderhalf jaar heb ik ontzettend veel hulp 

van hem gekregen. Heroen, bedankt voor de hulp bij het analyseren van mijn gegevens en het snel 

en grondig nalezen en verbeteren van de teksten. 

Ten derde, zou ik Olivier, Heroen en Frederik willen bedanken om me te steunen. Zonder de talloze 

brainstormmomenten en de hulp van jullie alle drie bij het verwerken en uitschrijven van de 

resultaten, was ik nooit tot dit resultaat gekomen. 

Caroline, het was leuk om gedurende drie jaar bureau en labo met je te delen. Eveneens bedankt 

voor de hulp bij het praktisch werk. Andy en Renata, bedankt voor al het sequentiewerk. 

Kadriye thanks to help me with PCR’s and cloning of some of the nuclear genes. It was often a 

frustrating job, with a lot of trial and error. 

Koen Sabbe, Ann Willems en Paul De Vos, bedankt voor de tijd die jullie hebben vrij gemaakt om, 

vooral in het begin, te luisteren naar mijn vorderingen en me met jullie suggesties telkens een stapje 

vooruit te helpen. Ook Steven Robbens en Yves van de Peer hielpen me door me in het begin de kans 

te geven over het nog niet gepubliceerde Ostreococcus genoom te beschikken. 

Klaus Valentin, thanks for the cDNA service. The generation of this cDNA library was a big step 

forward during this PhD study. The people from VERTIS Biotechnologie AG (Freising , Germany) also 

helped a lot to solve the problems I encountered during the screening of the cDNA library.

Aan de mensen van de plantkunde in de Ledeganckstraat, het was altijd leuk en gezellig tijdens de 

middag of aan de koffietafel. Liesbeth, op het bankje aan het kleine vijvertje was het ook steeds 

gezellig vertoeven. Ik heb daar goeie herinneringen aan! 

Het laatste anderhalf jaar was het met de mensen van op de Sterre minstens even gezellig tijdens de 

middagen in de Resto. 

Katrien en Elke, bedankt voor het nalezen van een stukje Nederlandstalige tekst. 

Eric, al wist je nooit goed waar ik nu precies mee bezig was, toch zou ik je willen bedanken om me 

warm te maken voor de algologie, en om je vlucht naar Sri Lanka te verzetten zodat je aanwezig kunt 

zijn op mijn publieke verdediging. 

Ook mijn ouders en vrienden zou ik willen bedanken om me steeds te blijven steunen doorheen de 

jaren. 

En tenslotte, Toon die nu al bijna negen jaar lang mijn vriend is… nog vele fijne jaren voor ons! 

Ellen 

juni 2009

Contents 

Chapter 1 General introduction and thesis outline 1 

Chapter 2 Ancient relationships among green algae inferred from nuclear and 

chloroplast genes 

Chapter 3 Gain and loss of elongation factor genes in green algae 43 

Chapter 4 Complex phylogenetic distribution of a non-canonical genetic code in green 

algae 

Chapter 5 Codon usage bias and GC content in green algae 81 

Chapter 6 A multi-locus time-calibrated phylogeny of the siphonous green algae 91 

Chapter 7 Systematics of the marine microfilamentous green algae Uronema curvatum 

and Urospora microscopica (Chlorophyta) 

Chapter 8 General discussion 129 

References 143 

Summary 161 

Samenvatting 165 

19 

69 

115

1 

Introduction 

Algae 

Algae are a large and diverse group of eukaryotic photosynthetic organisms occurring in almost every 

habitat. They exhibit a huge morphological diversity, ranging from tiny unicells to huge kelps over 50 

m long. The first algal groups arose between 1 and 1.5 billion years ago (Douzery et al. 2004, Yoon et 

al. 2004) after the symbiogenesis of a heterotrophic eukaryotic organism with a photosynthetic 

cyanobacterium. This event gave rise to the primary plastids which are still present in the 

Glaucophyta, red algae and green lineages including land plants (Reyes-Prieto et al. 2007). These 

three lineages are collectively called Plantae or Archaeplastida (Cavalier-Smith 1981, Adl et al. 2005). 

The other photosynthetic protists arose through secondary endosymbiosis of either a green or a red 

alga. The euglenids and chlorarachniophytes are thought to have acquired their plastids from a green 

alga in two separate secondary endosymbiotic events, while molecular evidence suggests that the 

red algal plastid of cryptomonads, heterokonts, haptophytes, apicomplexans and dinoflagellates was 

acquired by a single secondary endosymbiosis in their common ancestor (Archibald 2005, Archibald 

2008). This process of serial cell capture and subsequent enslavement explains the diversity of 

photosynthetic eukaryotes. Endosymbiosis forms the landmark evolutionary event, responsible for 

the spread of photosynthesis through the Eukaryotic tree of life. Photosynthesis occurs in four of the 

six supergroups: Archaeplastida (Glaucophyta, red algae, green plants), Chromalveolata 

(cryptophytes, Stramenopila or heterokonts including diatoms and brown algae, haptophytes and 

dinoflagallates), Rhizaria (Chlorarachniophyta) and Excavata (euglenoids) (Fig. 1). 

Figure 1. Eukaryotic tree of life. The first 

algae arose after the symbiogenesis of a 

heterotrophic eukaryotic organism with a 

photosynthetic cyanobacterium, giving rise 

to the Archaeplastida. The other 

photosynthetic protists arose through 

secondary endosymbiosis of either a green 

or a red alga and occur in four of six 

supergroups (marked with respectively 

green and red circles). The monophyly of 

the Archaeplastida is well-supported and 

most recent evidence favours the 

Glaucophyta as earliest diverging lineage 

within the Archaeplastida (modified after 

Baldauf 2008, Lane and Archibald 2008).

2 CHAPTER 1 

Archaeplastida 

The monophyly of primary plastids has long been suggested by several features, such as a similar 

gene content of plastid genomes, the presence of plastid-specific gene clusters that are distinct from 

those found in Cyanobacteria, the conservation of the plastid-protein import machinery and proteintargeting 

signals, and phylogenies based on plastid and cyanobacterial gene sequences (Palmer 

2003). Nevertheless, several single-gene phylogenies and a few multigene phylogenies have 

challenged this hypothesis (e.g., Stiller et al. 2001, Nozaki et al. 2003a, Nozaki et al. 2003b, Stiller and 

Harrell 2005). Conclusive evidence for the monophyly of the Glaucophyta, red algae and green plants 

was provided only relatively recently by Rodriguez-Ezpeleta et al. (2005) based on: (1) chloroplast 

gene phylogenies showing the monophyly of primary plastid and (2) a phylogenomic dataset 

containing 143 nuclear genes, ca. 30,000 amino acid positions which show the monophyly of all 

organisms with a primary plastid (Fig. 1). The latter study, however, could not reveal the relation 

among the three major lineages. Several nuclear genes suggest that red algae are the earliest 

diverging Archaeplastida, but such results are inconsistent with many plastid gene trees that identify 

glaucophytes as the earliest divergence. Most recent evidence favours the early divergence of 

glaucophytes, as demonstrated by Reyes-Prieto et al. (2007) using a concatenated dataset of 

conserved nuclear-encoded plastid targeted proteins of cyanobacterial origin. The latter evolutionary 

scenario corroborates with two important putatively ancestral characters shared by glaucophyte 

plastids and the cyanobacterial endosymbiont that gave rise to this organelle: the presence of 

carboxysomes and a peptidoglycan deposition between the two organelle membranes. Both traits 

were apparently lost in the common ancestor of red and green algae after the divergence of 

glaucophytes. 

Figure 2. The green algae exhibit a remarkable cytological diversity ranging from unicellar organisms (coccoid 

or flagellates), over multicellular filaments and foliose blades, to coenocytic and siphonous life forms that are 

essentially composed of a single giant cell containing countless nuclei (after Coppejans 1998). Arrows indicate 

trends in morphological complexity rather than evolutionary hypotheses. For example, green algae are thought 

to have evolved from a unicellular flagellate (the Ancestral Green Flagellate, AGF) rather than a coccoid life 

form.

Green lineage or Viridiplantae 

INTRODUCTION 3 

Green algae are distributed worldwide and can be found in almost every habit ranging from polar to 

tropical marine, freshwater and terrestrial environments and as symbionts (Pröschold and Leliaert 

2007). They exhibit a remarkable cytological diversity ranging from the world’s smallest free-living 

eukaryote known to date Ostreococcus taurii (Derelle et al. 2006), over multicellular filaments and 

foliose blades, to siphonous life forms that are essentially composed of a single giant cell containing 

countless nuclei (Fig. 2). Together with land plants, green algae form the green lineage or 

Viridiplantae (also written as Virideaplantae or known as green plants, Chlorobionta, Chloroplastida 

or Chlorophycophyta). Morphological and molecular studies have identified a major split within the 

Viridiplantae giving rise to two monophyletic lineages, the Chlorophyta and the Streptophyta 

(Pickett-Heaps and Marchant 1972, Lewis and McCourt 2004) (Fig. 3). The streptophytes comprise 

several lineages of predominantly freshwater green algae (often called charophytes or charophyte 

green algae) and the land plants (Embryophyta) which evolved roughly 470 million years ago from a 

charophyte ancestor. The majority of green algae, however, belong to the Chlorophyta. 

Streptophyta 

When motile cells are present, the Streptophyta are characterized by biflagellate cells with 

asymmetrically flagellar roots including a multilayered structure or MLS (a distinct parallel 

arrangement of microtubules) and a smaller root. In all representatives the nuclear envelope breaks 

down before the chromosomes separate (open mitosis) and the mitotic spindle is persistent which 

helps to keep the daughter nuclei separate until cytokinesis has been accomplished. Biochemical 

characters such as photorespiratory enzymes are different from those found in most chlorophyte 

green algae (Figs. 4 and 5; Table 1). 

Several phylogenetic studies have tried to determine the origins of the land plants, focussing on the 

green algal progenitors of the Streptophyta. Recent studies have indicated the scaly green flagellate 

Mesostigma as the earliest diverging streptophyte. Initially, the scaly flagellate was placed within the 

prasinophytes (Mattox and Stewart 1984), later ultrastructural investigations (e.g. a flagellum which 

is anchored in the cell by means of an asymmetric root) revealed the association with the 

Streptophyta (Melkonian 1989). Molecular phylogenies also showed conflicting results regarding the 

phylogenetic relationships of this enigmatic species: Mesostigma either diverges before the 

Chlorophyta/Streptophyta split (Lemieux et al. 2000, Turmel et al. 2002a, Turmel et al. 2002b) or as 

an early diverging flagellate within the Streptophyta (Bhattacharya et al. 1998, Marin and Melkonian 

1999, Karol et al. 2001). Increasing taxon and gene sampling and the use of more realistic models of 

evolution provide evidence that Mesostigma is an early diverging lineage within the Streptophyta 

(Petersen et al. 2006, Lemieux et al. 2007, Rodriguez-Ezpeleta et al. 2007). The colonial soil alga 

Chlorokybus diverges after Mesostigma in most phylogenies (Karol et al. 2001), although more recent 

studies united Mesostigma and Chlorokybus as the earliest diverging branch of the Streptophyta 

(Lemieux et al. 2007). Unbranched filaments that form the class Klebsormidiophyceae diverge next, 

followed by the Zygnematophyceae clade, which includes unicells and unbranched filaments with 

isogamous sexual reproduction. The more complex charophytes are Coleochaetophyceae and 

Charophyceae both consisting of branched filaments with oogamous sexual reproduction. It remains

4 CHAPTER 1 

inconclusive from which group of algae Embryophytes emerged. Karol et al. (2001) resolved the 

stoneworts (Charophyceae) as the closest relatives of the Embryophyta, while evidence from plastid 

genomes point toward the conjugating algae or Zygnematophyceae (Lemieux et al. 2007). 

Figure 3. Phylogenetic relationships between green algae inferred from several phylogenetic studies: 

Streptophyta (Karol et al. 2001: SSU nrDNA, mitochondrial nad5 gene and plastid rbcL and atpB genes), 

prasinophytes (Guillou et al. 2004: SSU nrDNA), Trebouxiophyceae (Karsten et al. 2005: SSU nrDNA), 

Chlorophyceae (Turmel et al. 2008: chloroplast genes), Ulvophyceae (Lopez-Bautista and Chapman 2003: SSU 

nrDNA) and (Watanabe and Nakayama 2007: SSU nrDNA). The main interest of chapter 2 is to resolve relations 

between Ulvophyceae, Trebouxiophyceae and Chlorophyceae (1) and among Ulvophyceae (2). Representatives 

from each clade are studied (except those marked in gray).


Figure 4. Variation in flagellar apparatuses found among green algae, viewed from the top (upper figure) and 

from the side (lower figure), modified after Graham et al. (2009) and Pröshold and Leliaert (2007). The flagellar 

apparatus generally include two or four basal bodies (shown here as rectangles or cylinders), microtubular 

roots (s or d), and distal (DF) or proximal (PF) connecting fibers. A. Flagellar apparatus with cruciate roots and 

basal bodies displaced in counter-clockwise direction. B. Flagellar apparatus with cruciate roots showing 

directly opposed placement of flagellar basal bodies. C. Flagellar apparatus with clockwise displaced flagellar 

basal bodies. D. Flagellar apparatus with parallel basal bodies and asymmetrical distribution of the flagellar 

roots, showing the characteristic multilayered structure (MLS). 

Figure 5. Ultrastructural features of green algae. 1. 

Comparison of cytokinesis among green algae 

(after Graham et al. 2009). A. Phycoplasts, arrays 

of microtubules which lie parallel to the developing 

cleavage furrow, are often present in the 

Chlorophyceae and Trebouxiophyceae. B. 

Furrowing, sometimes with involvement of 

microtubules, is observed in the early charophytes 

and some Ulvophyceae. C. Phragmoplasts very 

similar, if not identical, to those of land plants are 

found in later diverging charophytes and the 

Trentepohliales (Ulvophyceae). Little furrowing is 

involved, the cell plates develop from the center 

toward the cell periphery. Microtubules arranged 

perpendicular to the developing cell plate. 2. 

Different types of mitosis among green algae 

during the metaphase indicating the fate of the 

nuclear envelope and position of the centrioles 

(after Graham et al. 2009). A. Closed mitosis. B. 

Metacentric mitosis C. Open mitosis.

6 CHAPTER 1 

Chlorophyta 

The current classification of the Chlorophyta, which relies on a combination of morphology, 

ultrastructural features of the flagellar root system, and characters relating to the mitotic spindle 

during cell division and cytokinesis, has been largely confirmed by phylogenetic analysis (Figs. 4 and 

5, Table 1) (Mattox and Stewart 1984, Pröschold and Leliaert 2007). Four major groups, commonly 

regarded as classes, are recognized by consensus: “Prasinophyceae”, Chlorophyceae, 

Trebouxiophyceae and Ulvophyceae (Fig. 3) (reviewed in Lewis and McCourt 2004). The 

prasinophytes form a paraphyletic group of unicellular flagellates or coccoid cells at the base of the 

Chlorophyta (Steinkötter et al. 1994, Fawley et al. 2000, Lopez-Bautista and Chapman 2003, Guillou 

et al. 2004). The Ulvophyceae, Trebouxiophyceae and Chlorophyceae are resolved as a wellsupported 

clade (UTC clade) in most studies (Mishler et al. 1994), but the relationships among these 

lineages form the basis of a longstanding debate. Furthermore, the monophyly of the three classes 

remains to be demonstrated unequivocally (O'Kelly and Floyd 1984a, Zechman et al. 1990, Krienitz et 

al. 2003). 

Certain ultrastructural characteristic are shared between the Ulvophyceae, Trebouxiophyceae and 

Chlorophyceae. In all representatives the nuclear envelope remains intact until the chromosomes 

finally separate (closed mitosis). The flagella are anchored in the cell by means of cruciate flagellar 

roots with mostly an X-2-X-2 configuration of the microtubules (Moestrup 1978, Lewis and McCourt 

2004). Other ultrastructural observations are useful diagnostic characters to separate the three 

classes. The orientation of the basal bodies, short cylindrical arrays of microtubules at the base of a 

flagellum, is one of those discriminative characters. The Ulvophyceae and Trebouxiophyceae have a 

counter-clockwise orientation of the basal bodies, while the Chlorophyceae have a direct opposite or 

clockwise orientation of the basal bodies (Fig. 4) (Lewis and McCourt 2004). The Ulvophyceae have a 

persistent mitotic spindle which helps to keep the daughter nuclei separate until cytokinesis has 

been accomplished. The Trebouxiophyceae and Chlorophyceae both have a non-persistent mitotic 

spindle and a phycoplast composed of a set of microtubules which lie parallel to the plane of 

cytokinesis (Fig. 5) (Friedl 1995, Lewis and McCourt 2004). Based on these ultrastructural 

observations Mattox and Stewart (1984) suggest that the Ulvophyceae diverged first, followed by the 

Trebouxiophyceae and Chlorophyceae (Fig. 6). While Mattox and Stewart (1984) based their 

classification on the orientation of the basal bodies in the flagellar apparatus and differences of the 

mitotic spindle during cell division and cytokinesis, Sluiman (1989) only used the orientation of basal 

bodies in the flagellar apparatus as diagnostic character and merged the Trebouxiophyceae with the 

Ulvophyceae based on the counter-clockwise orientation of the basal bodies in the flagellar 

apparatus (Fig. 6). 

Molecular phylogenetic studies have been highly inconclusive about the relationships between UTC 

classes (Fig. 6). The first molecular phylogenies based on small subunit nuclear ribosomal DNA (SSU 

or 18S nrDNA) sequences all observed that Ulvophyceae branch first, leaving Trebouxiophyceae and 

Chlorophyceae as sisters (Friedl 1995, Bhattacharya et al. 1996, Krienitz et al. 2001, Lopez-Bautista 

and Chapman 2003), while more recent SSU nrDNA phylogenitic studies using expanded taxon 

sampling and likelihood-based methods with more realistic models of sequence evolution revealed a 

sister relation between Chlorophyceae and Ulvophyceae (Friedl and O'Kelly 2002, Lewis and Lewis 

2005, Watanabe and Nakayama 2007). Chloroplast gene order data and genomic structural features


(shared gene losses and rearrangements within conserved gene clusters), along with a phylogenetic 

analysis of seven mitochondrial genes supported this sister relation between Ulvophyceae and 

Chlorophyceae, while phylogenetic analysis of 58 concatenated chloroplast genes supported a sister 

relation between Ulvophyceae and Trebouxiophyceae (Pombert et al. 2004, Pombert et al. 2005). In 

this way, all possible relations between the UTC algae have been proposed the latest decennia. 

Figure 6. The three alternative topologies for Ulvophyceae, Trebouxiophyceae and Chlorophyceae (UTC) have 

been supported by ultrastructural characteristics, molecular phylogenies and/or genomic structural features. 

“Prasinophyceae” – The prasinophytes are planktonic and predominantly marine unicellular 

flagellates (with one to eight flagella) or coccoid organisms exhibiting a wide variety of ultrastructural 

features and photosynthetic pigment signatures (see O'Kelly 2007 and references therein). The cells 

and flagella of many members are covered by up to seven distinct types of organic scales, which are 

formed in the Golgi apparatus. The absence of clear synapomorphies suggest that the prasinophytes 

are not monophyletic but rather form a cluster of several independent lineages (Chlorodendrales, 

Picocystis clade, Pycnococcus clade, Nephroselmis clade, Mamiellales, Prasinococcales, 

Pyramimonadales) at the base of the Chlorophyta, which have been formally described and 

confirmed by SSU nrDNA phylogenies (Fig. 3 + Table 1) (Steinkötter et al. 1994, Fawley et al. 2000, 

Guillou et al. 2004, Turmel et al. 2009). It should be noted that our knowledge of these diverse, 

mainly picoplanktonic algae is far from comprehensive. The application of new technologies such 

environmental sequencing results in the discovery of new lineages, the majority of which remains 

hitherto uncultured and undescribed. 

The Chlorodendrales (Tetraselmis and Scherffelia) is the only prasinophyte lineage which is robustly 

placed at the base of the UTC clade in molecular phylogenies and which is characterised by the 

presence of a counter-clockwise orientation of the basal bodies, a non-persistent mitotic spindle and 

the occurrence of a phycoplast. All the rest of the prasinophytes contain ancestral morphological 

treats, which makes it difficult to assign them either to the Chlorophyta or Streptophyta. In this view, 

the Pyramimonadales are crucial in our understanding of the early history of green plants. The 

combination of a well-represented fossil record, a flagellar apparatus configuration from which all 

other patterns in the green plants can be plausibly derived by reduction, and the only documented 

instances of phagotrophic mixotrophy suggests that the Pyramimonadales are the modern 

representatives of the earliest green algae (O’Kelly 2007). The prasinophytes further contain 

Ostreococcus tauri (Mamiellales), which is the smallest free-living eukaryote and which has the

8 CHAPTER 1 

smallest nuclear genome of all photosynthetic eukaryotes (Derelle et al. 2006) and Nephroselmis 

(Nephroselmidales) a small unicellular flagellate. 

Trebouxiophyceae - The Trebouxiophyceae mainly consist of freshwater (e.g. Chlorella) and 

terrestrial algae (e.g. the phycobiont Trebouxia in lichens), some members (e.g. Prasiola) occur in 

marginally marine habitats. In contrast to most trebouxiophytes, the genus Prototheca is colorless 

and obligately heterotrophic and mainly lives in soil, but there are also some disease causing species. 

The enigmatic parasitic alga Helicosporidium is closely related to Prototheca (Tartar et al. 2002). 

Trebouxiophyceae occur as non-flagellate unicells or colonies, unbranched or branched filaments, or 

small blades (e.g. Prasiola) similar to those found among Ulvophyceae. Trebouxiophyceae commonly 

produce asexual, non-motile autospores. Sexual reproduction, involving flagellate sperm and nonmotile 

eggs, is only known for some representatives (e.g. Prasiola). 

The Trebouxiophyceae are characterized by a combination of ultrastructural characteristics: counterclockwise 

orientation of the basal bodies (Fig. 4), non-persistent metacentric mitotic spindles (Fig. 

5B) and the presence of a phycoplast (Fig. 5A), none of which is unique to the class. The basal body 

orientation is shared with the Ulvophyceae, metacentric spindles with the prasinophytes and nonpersistent 

spindles and phycoplasts with the Chlorophyceae. The monophyly of the 

Trebouxiophyceae has still to be proven due to the lack of unique structural or reproductive features 

and the failure of molecular phylogenetic studies to support monophyly (Krienitz et al. 2003, Lewis 

and McCourt 2004). 

Chlorophyceae - The Chlorophyceae mainly contain freshwater and terrestrial green algae. The class 

exhibits all major body plans found among green algae: unicells (flagellates and non-flagellates), 

sarcinoid organisms which are composed of packets of non-motile cells, colonies, unbranched or 

branched filaments and some multinucleate organisms. Asexual reproduction often occurs by means 

of flagellated zoospores but can also occur with non-motile aplanospores or autospores. Sexual 

reproduction may be isogamous, involving gametes that are morphological identical; anisogamous, in 

which case flagellate gametes are structurally distinguishable; or oogamous, with a large non-motile 

egg and smaller flagellate sperm. When sexual reproduction is present, Chlorophyceae always have a 

haploid vegetative phase and a single celled, often dormant, zygote as diploid stage (zygotic meiosis). 

Cell division is fairly uniform among Chlorophyceae and includes closed mitosis and a non-persitent 

mitotic spindle which collapses before cytokinesis. Cleavage is mostly mediated by a phycoplast but 

some Chlorophyceae divide by simple centripetal furrowing of the cell wall. The production of a cell 

plate that develops centrifugally by fusion of Golgi-derived vesicles in which case plasmodesmata or 

channels through the cell wall will allow intercellular communication after division have been 

reported for Cylindrocapsa and Uronema (see chapter 7 and van den Hoek et al. 1995, Graham et al. 

2009). 

The monophyly of the Chlorophyceae is corroborated by the presence of clockwise or direct opposite 

flagellar root supports (Fig. 4). The two major clades are formed by the Chlamydomonadales 

(clockwise basal body orientation) which include genera like Chlamydomonas and Volvox and the


Sphaeropleales (direct opposite basal body orientation) which contain for instance Scenedesmus. 

Other orders are Chaetophorales, Chaetopeltidales and Oedogoniales (Turmel et al. 2008). 

Ulvophyceae - The Ulvophyceae mainly include marine green macroalgae among which some wellknown 

green algae such as the sea lettuce Ulva, the model organism Acetabularia and the weedy 

Codium and Bryopsis. Morphological diversity ranges from flagellate and non-flagellate unicells and 

colonies to branched and unbranched filaments, foliose blades and multinucleate life forms. The 

sexual life cycle of most Ulvophyceae involves an alternation between a diploid sporophyte and 

haploid, free-living gametophytes (sporic meiosis), but zygotic meiosis with a haploid vegetative 

phase and single celled zygote as diploid stage also occurs. In some representatives (e.g. some 

Ulotrichales), flagellate reproductive cells are covered by a layer of small scales similar to those 

found in some prasinophytes (Sluiman 1989). 

Since the class lacks clear synapomorphies, the monophyly of the Ulvophyceae has been doubted 

since its establishment (O'Kelly and Floyd 1984b). Therefore, the Ulvophyceae are defined by a suite 

of characters such as a counter clockwise orientation of the basal bodies (Fig. 4), a persistent mitotic 

spindle (Fig. 5B) and cytokinesis mediated by centripetal furrowing of the plasma membrane, none of 

which is exclusive for the Ulvophyceae (Mattox and Stewart 1984, O'Kelly and Floyd 1984a). The 

Trentepohliales for which molecular phylogenies revealed their relationship to ulvophycean 

seaweeds, have some enigmatic streptophyte-like characteristics such as an asymmetric flagellar 

root anchored in the cell with a multilayered structures and the formation of a phragmoplast during 

cytokinesis. 

Molecular phylogenies based on SSU nrDNA failed to provide solid support for the monophyly of the 

Ulvophyceae and revealed the presence of two distinct groups within the Ulvophyceae. One group is 

represented by the orders Ulvales and Ulotrichales and contains unicellular, filamentous and bladelike 

organisms. The other group is composed of the branched filamentous order Trentepohliales, the 

siphonocladous seaweed order Cladophorales (synonymous with Siphonocladales), and the 

siphonous seaweed orders Dasycladales and Bryopsidales (Zechman et al. 1990, Watanabe et al. 

2001, Lopez-Bautista and Chapman 2003, Watanabe and Nakayama 2007). In these trees, long 

evolutionary distances separate the green seaweed orders from the rest of the Ulvophyceae which 

consequently obscures the relationships among them. Based on these long branches and the 

apparent differences in thallus architecture, cellular organization, chloroplast morphology, cell wall 

composition, and life histories, some authors suggested to elevate the ulvophycean orders 

Cladophorales, Bryopsidales, Dasycladales and Trentepohliales to the class level (van den Hoek et al. 

1995). In addition to the uncertainty regarding the affinities between the main orders, the position of 

some enigmatic Ulvophyceae such as Ignatius, Oltmannsiellopsis and Blastophysa is not well-known.

10 CHAPTER 1 

Green algal phylogenies 

The estimated divergence time between Streptophyta and Chlorophyta varies considerably among 

studies. The split between these lineages is generally situated in the Neoproterozoic to late 

Mesoproterozoic (Fig. 7) (between 1200 and 700 million years ago: Douzery et al. 2004, Hedges et al. 

2004, Yoon et al. 2004, Berney and Pawlowski 2006, Roger and Hug 2006, Zimmer et al. 2007, Herron 

et al. 2009). The ancient age in combination with the short time span over which the major lineages 

of green algae diverged likely caused the unstable relationships among these classes. Determining 

relationships among the different orders of the Ulvophyceae poses a similar problem. Up to now, 

green algal phylogenies using a broad taxon sampling were always based on SSU nrDNA (Zechman et 

al. 1990), while studies using whole plastid genome sequences contend with a small taxon sampling 

(Pombert et al. 2004, Pombert et al. 2005). It is well-known that phylogenetic information contained 

in a single gene is often insufficient to obtain firm statistical support for deep nodes (Philippe et al. 

2005). Hence, improved character sampling by the inclusion of a larger number genes often leads to 

more accurate estimates of phylogenetic relationships. Unfortunately there is a limited availability of 

genomic data for green algae. Whole genome sequences were only available for the chlorophycean 

Chlamydomonas reinhardtii and the prasinophycean Ostreococcus tauri at the start of this project. 

Some other whole genome sequences have been recently released or are on the way for the 

chlorophyceae Volvox and the prasinophytes Micromonas and Bathycoccus. However, up till present 

no genome has been sequenced for a representative of the Ulvophyceae or Trebouxiophyceae. In 

addition to whole genome sequences, there are small to moderately sized EST libraries and complete 

organelle genomes available for some green algae (table 2). However, the ulvophycean seaweed 

orders remain underrepresented: no sequenced plastid genome and only 2 smaller EST libraries are 

available. 

This limited availability of genomic information for the Chlorophyta inevitably poses a restriction on 

taxon sampling. Most studies which have addressed deep relationships of Chlorophyta are seriously 

constrained in taxon sampling (e.g., Pombert et al. 2005, Rodriguez-Ezpeleta et al. 2007). 

Furthermore, sparse and uneven taxon sampling, with most species belonging to either the 

prasinophytes or Chlorophyceae, increases the risk of systematic error due to long branch attraction 

and other biases (Verbruggen and Theriot 2008). 

Resolving deep green algal relationships will inevitably request the combined analysis of many 

markers. Since genomic data are rare, it will be necessary to design primers for new markers. Lowcopy 

nuclear markers proved to be good candidates due to a higher evolutionary rate than organellar 

genes, the potential to analyze multiple unlinked genes and the bi-parentally inheritance (Small et al. 

2004). However, the greater difficulty of isolating and characterizing low-copy nuclear markers in 

comparison with SSU nrDNA and chloroplast genes which can be amplified with universal primers 

and the difficulty of assessing orthology are the major challenge when using low-copy nuclear 

markers (Small et al. 2004).


Figure 7. The estimated divergence time between Streptophyta and Chlorophyta, which are based on 

molecular clock studies, vary considerably among studies. The split between these lineages is generally situated 

in the Neoproterozoic to late Mesoproterozoic (between 1200 and 700 million years ago). The best 

characterized Precambrian fossils which most likely are green algae are Proterocladus and Palaeastrum at 

Svanbergfjellet, Tasmanites and Pterospermella at Thule and Spiromorpha at Ruyang. The oldest reliable fossil 

green algae are 540 million years old (Dasycladales). Other fossil Ulvophyceae are found more recently. 

Evolution of nuclear markers 

One of the major challenges when using nuclear genes for phylogenetic purposes is to assess 

orthology. Gene duplications lead to the formation of gene families (e.g. actin gene family, and 

GapA/B gene duplication in Streptophytes and the prasinophyte Ostreococcus), in which case 

extreme caution should be taken to ensure that the same orthologues gene is used in each species 

(An et al. 1999, Petersen et al. 2006, Robbens et al. 2007). Apart from difficulties in assessing 

orthology, there are indications of the occurrence of lateral gene transfer between eukayotes 

(Andersson 2005, Keeling and Palmer 2008). Two key genes of the translational apparatus, 

elongation factor-1 alpha (EF-1α) and elongation factor-like (EFL) have an almost mutually exclusive 

distribution in eukaryotes. In the green plant lineage, the Chlorophyta encode EFL except 

Acetabularia (Dasycladales, Ulvophyceae) where EF-1α is found, and the Streptophyta possess EF-1α 

except Mesostigma, which has EFL (Noble et al. 2007). The punctuated distribution of these two 

genes make it worth to further explore their distribution and evolutionary patterns of gain and loss 

within the green plant lineage. In this context it is interesting that Acetabularia not only has a 

different elongation factor gene compared to the rest of the Chlorophyta but also uses a different 

genetic code. In Acetabularia, and its close relative Batophora, stop codons TAG and TAA code for 

the amino acid glutamine (Schneider et al. 1989, Schneider and de Groot 1991). It is interesting to 

see if other green algae, especially close relatives of Acetabularia, also have the EF-1α gene or use 

the same alternative (non-canonical) code.

12 CHAPTER 1 

Designing primers for low-copy nuclear markers 

For phylogenetic analyses we will concentrate on the development of low-copy nuclear markers. 

There are several alternative approaches to isolate and characterize novel low-copy nuclear markers: 

1) the design of new primers from information in sequence databases (e.g., GenBank), 2) obtaining 

novel sequences via cDNA cloning 3) isolation of homologous DNA using a gene probe from another 

organism, and 4) characterization of sequence markers from DNA fingerprints (Schluter et al. 2005). 

In order to find suitable nuclear markers we first searched the literature for potentially useful lowcopy 

nuclear markers and then applied the first two approaches (Fig. 8). When designing new 

primers from information already available in GenBank, we used genomic sequences of two 

unicellular green algae (Chlamydomonas reinhartii and Ostreococcus tauri) and aligned them with 

Genbank sequences from land plants and other green algae if available. For the second approach we 

made an EST library for the siphonocladous ulvophyte Cladophora coelothrix. All Cladophora 

coelothrix EST sequences were blasted against the GenBank protein database (blastx) for annotating 

the genes. Only genes with a certain annotation (hits to Swiss-Prot database) were retained and 

aligned with GenBank sequences from other green algae and land plant. If those genes were long 

enough and contained conserved regions, primers were designed and tested on a variety of green 

algae. 

Figure 8. Flow chart showing three different methods to select suitable nuclear markers in green algae for 

which primers can be designed and tested on a wide range of green algae.

Tree building methods 

Maximum likelihood and Bayesian Inference 


Using the most accurate tree building methods and evolutionary models available is a basic necessity 

to obtain accurate phylogenies (Fig. 9) (Delsuc et al. 2005). Likelihood-based methods (maximum 

likelihood and Bayesian Inference) generally outperform methods based on distance or parsimony 

criteria because they allow the explicit incorporation of the processes of character evolution into 

probabilistic models to calculate the likelihood of the data given the model and tree. Maximum 

likelihood (ML) selects the tree that maximizes the probability of observing the data under a given 

model of sequence evolution. Bayesian methods derive the distribution of trees according to their 

posterior probability, using Bayes’ mathematical formula to combine the likelihood function 

(including tree and model parameters) with prior probabilities on trees. Since prior knowledge is 

mostly lacking or bias towards one or the other tree is not generally desirable, flat priors are usually 

chosen, i.e. giving the same prior probability to all trees. Consequently, posterior tree probabilities 

depend primarily on the tree likelihood. Unlike ML, which optimizes model parameters to find the 

highest peak in parameter space and where confidence is obtained by non-parametric bootstrapping, 

Bayesian approaches integrate the model parameters by measuring the volume under a posterior 

probability surface rather than finding its maximum height and simultaneously estimates trees and 

measurements of uncertainty for every branch (Holder and Lewis 2003, Delsuc et al. 2005, 

Verbruggen and Theriot 2008). During Bayesian analysis, Markov chain Monte Carlo (MCMC) 

simulation is used to approximate the posterior probability distribution because the complexity of 

the phylogenetic likelihood functions prevents its analytical calculation. During each generation, a 

parameter change is proposed (topology, branch lengths and model parameters) and accepted if it 

increases the posterior probability. If the posterior probability decreases, the parameter change is 

either accepted or rejected depending on the amount of change in posterior probability. Whereas 

small changes are often accepted, large decreases are usually rejected. Because parameters are 

usually not near their optimal values during initial generations these first generations, called burn-in, 

need to be removed before a consensus tree of all post-burn-in samples can be made. In order to 

search tree space even more thoroughly, Metropolis-coupled MCMC, in which several chains are run 

in parallel can be applied. Metropolis-coupled MCMC is implemented in the commonly used BI 

program MrBayes (Ronquist and Huelsenbeck 2003). The first chain is the called the cold chain and 

only propose small parameter changes. The other chains are incrementally heated and propose 

larger parameter changes in order to find distant regions with high posterior probabilities. After each 

generation, chains can be swapped, i.e. a heated chain in a higher posterior probability region than 

the current cold chain can become the cold chain in order to find the local optimum. Only the output 

from the cold chain is used to summarize the posterior distribution and, due to chain swapping, this 

chain will contain a more complete image of the high posterior probability regions of tree space 

compared with a BI analysis based on a single MCMC chain. The downside of Metropolis-coupled 

MCMC is a considerably higher computational cost because several chains have to be run in parallel.

14 CHAPTER 1 

Missing data 

Deep phylogenies require the simultaneous analysis of many characters and many taxa (Delsuc et al. 

2005). Individual, orthologous genes can be combined into a supermatrix which inevitably involves a 

certain amount of missing data. Many studies have studied the effects of missing data on 

phylogenetic reconstruction. A simulation study suggests that the placement of individual taxa in a 

tree is robust to large amounts of missing data in the sequences of the taxa in question (up to 50% 

under the simulated conditions) and that model-based methods can deal with even greater amounts 

of missing data (Wiens 2005). Another simulations study demonstrates that Bayesian analyses are 

even more robust to missing data, i.e. the phylogenetic position of taxa with 95% of missing data in 

their sequence is still accurate, as long as the total number of characters in the dataset is large 

(Wiens and Moen 2008). Studies of empirical datasets have shown that datasets with up to 92% of 

missing data are still able to provide insights into various parts of the tree of life (Driskell et al. 2004, 

Philippe et al. 2004, Delsuc et al. 2005). 

Models of sequence evolution 

The General Time Reversible (GTR) model and its simpler variants include one or more parameters to 

describe the substitution rate between the different bases. The GTR model uses a set of parameters 

to describe the relative substitution rate between all combinations of bases (AC, AG, AT, CG, CT, and 

GT). The simpler models only consider transitions versus transversions or attribute an equal 

substitution rate to all possible changes. A second important component of a model are the base 

frequencies. They can be calculated directly from the dataset (‘empirical’ base frequencies) or 

optimized along with the other parameters of the model. A third common element of the model 

allows for variations of evolutionary rate across site (e.g. different codon positions in protein coding 

genes, loops and stems in ribosomal DNA). Such among site rate variation is commonly accounted for 

by assuming that the site rates follow a gamma distribution and/or by incorporating a proportion of 

invariable sites. 

Partitioning strategies 

A supermatrix, a dataset composed of different genes, often demands data partitioning to account 

for across site heterogeneity in evolutionary rate (Delsuc et al. 2005). Therefore, careful attention 

has to be paid to the selection of suitable partitioning strategies (Brown and Lemmon 2007, Li et al. 

2008, Verbruggen and Theriot 2008). Protein coding genes usually benefit from partitioning into 

codon position. Empirical studies showed that codon position models perform better than models 

which do not take codon position into account (Shapiro et al. 2006). In order to accommodate 

differences in evolutionary rate among partitions rate multipliers can be used.


Figure 9. Flow chart for accurate phylogenetic reconstruction. A. Phylogenetic reconstruction. B. Removal of 

fast-evolving sites. 

Selection of the optimal partitioning strategy and model 

A great number of models of sequence evolution have been described, ranging from simple models 

to complex models incorporating a lot of parameters. To reconstruct an accurate phylogenetic tree it 

is important to select a model of sequence evolution that approximates the evolutionary history of 

genes under study. A number of criteria have been developed to evaluate the fit of the different 

models to the data. The Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) 

are two of these criteria that can be used for selection of the optimal partitioning strategy and 

evolutionary model (Figure 9A). Whereas AIC only penalizes for the number of model parameters, 

BIC also incorporates alignment length and thus penalizes a situation in which many parameters have 

to be estimated from a small dataset. In other words for the same dataset, AIC will prefer a more 

complex model than BIC. Because Bayesian analysis appears to be more sensitive to model underspecifications 

than ML, some authors have suggested that AIC scores can be used to choose a 

complex model for Bayesian analysis and BIC to choose a less complex model for ML analysis 

(Verbruggen and Theriot 2008). AIC and BIC calculation starts from a guide tree which is inferred with 

a fast distance based method (e.g. NJ) or a fast ML search under a simple model (e.g. PhyML,

16 CHAPTER 1 

Treefinder). In the second step log likelihoods of the guide tree under different partitioning strategies 

and models are calculated. Subsequently, the corresponding AIC or BIC scores are calculated and 

compared. The condition with the lowest AIC and/or BIC score is chosen for phylogenetic analysis. 

Alternatively, Bayes factors (Nylander et al. 2004) can be used to compare different partitioning 

strategies and models. For each tested condition a separate Bayesian analyses has to be run which 

implies high computational times. This makes it unrealistic to compare many partitioning strategies 

and models in a Bayesian framework. 

Complex models of sequence evolution 

The secondary structure of ribosomal RNA consists of loops and stems. The nucleotides in the stem 

regions form base pairs and are interdependent because a change on one side of the stem has to be 

compensated in the other side of stem to avoid malfunction of the molecule. Since models of 

sequence evolution have to approach real evolution as close by as possible, it is recommended to 

incorporate this site interdependence in the model. This can be done by partitioning the ribosomal 

RNA into loops and stems and using a doublet model for the stem regions (Schöniger and Von 

Haeseler 1994). However, the use of a doublet model is computational demanding. 

Instead of partitioning protein coding genes into codon positions, a codon substitution model can be 

applied. In this model, nucleotide triplets are considered as a single character and changes from one 

triplet to another one are considered taking into account that some changes are more likely than 

others (e.g. synonymous versus non-synonymous substitution). Although codon substitution models 

are a more realistic approximation of protein sequence evolution than codon position models, they 

come with a very high computational cost, hindering their use for large datasets (Shapiro et al. 2006). 

Mixture models 

Mixture models offer an attractive alternative to data partitioning and applying different models to 

the partitions. Whereas a partitioned analysis assumes that all sites within a partition arose from a 

single evolutionary process, mixture models relax this assumption by not expecting any prior 

partitioning and applying a set of different models to each site in the alignment. The log likelihood of 

each site is calculated as a weighted sum of the log likelihoods of each model for that site. The model 

weights correspond to the probability that the site has evolved under the model in question. Mixture 

models can thus apply different rate matrices to different parts of the dataset without explicitly 

partitioning it (Pagel and Meade 2004, Venditti et al. 2008). This is an elegant way to incorporate 

across site heterogeneity in the evolutionary process because it does not require prior knowledge 

about differences of evolutionary processes between different parts of the dataset and it avoids 

problems associated with differences of the evolutionary process within partitions that are defined a 

priori. Although analyses using mixture models outperform analyses based on partitioned datasets, 

they are restrictively time-consuming for large datasets.

Removal of fast-evolving sites 


It has been shown that even the most accurate phylogenetic method and evolutionary models are 

unable to account for differences in the evolutionary process between species causing the wellknown 

long-branch attraction artifact (Delsuc et al. 2005). Long-branch attraction artifacts can be 

reduced by improving taxon sampling and removing fast-evolving sites (Brinkmann et al. 2005, Delsuc 

et al. 2005). Fast-evolving sites are particularly challenging because they mask the true phylogenetic 

signal, resulting in loss of resolution and decrease of accuracy (Rodriguez-Ezpeleta et al. 2007). In 

chapter 2, we will illustrate that removal of fast-evolving sites (site stripping: Waddell et al. 1999) 

improves phylogenetic signal in the desired epoch (Fig. 9B). 

Aims and outline of this thesis 

First, we want to improve the phylogeny of the green lineage. To avoid biases caused by taxon or 

character sampling we examine deep relations of Chlorophyta by analyzing a combined dataset of 

several single copy nuclear markers, together with SSU nrDNA and plastid rbcL and atpB genes, 

including taxa from all major green algal lineages. Broad taxon and character sampling in 

combination with careful selection of models of sequence evolution and partitioning strategies, and 

thorough phylogenetic analyses, including methods to reduce data saturation will yield optimal 

resolution in the desired epoch (relation between UTC classes and relations within Ulvophyceae). 

Guided by this improved green algal phylogenetic tree, we address various topics relating to 

molecular evolution of the Chlorophyta. The patterns of gain and loss of EF-1α and EFL genes, key 

enzymes in the translational apparatus, are interpreted in this phylogenetic framework. In addition 

we will screen green algal nuclear genes for the presence of non-canonical genetic codes, with 

emphasis on the ulvophycean relatives of the Dasycladales, where a non-canonical code had been 

detected earlier. Their phylogenetic distribution pattern will be studied and we will look for biological 

factors influencing the origin of a non-canonical code. 

In Chapter 2 deep relations of Chlorophyta will be examined and interpreted with respect to 

evolution of multicellularity and multinucleate cells. 

In Chapter 3 a 74-taxon phylogeny of the green lineage based on SSU nrDNA and two plastid genes 

(rbcL and atpB) is inferred and used to study the distribution and gain-loss patterns of elongation 

factor genes in green algae. 

In Chapter 4 the presence of a non-canonical code in some green algae is evaluated and discussed. 

Codon usage of canonical and non-canonical glutamine codons are calculated and their evolution is 

reconstructed. 

Chapter 5 explores synonymous codon usage bias and GC content among the green plant lineage. 

Chapter 6 calibrates a multi-locus phylogeny of the siphonous green algae in a geological timeframe.

18 CHAPTER 1 

In Chapter 7 the phylogenetic positions of the enigmatic microfilamentous species Uronema 

curvatum and Urospora microscopica are assessed by molecular phylogenetic analysis of nuclearencoded 

small and large subunit rDNA sequences 

Finally, the main conclusions are summarized and discussed in Chapter 8 and perspectives for future 

research are provided. 

Authors’ contribution 

Chapter 2: Ellen Cocquyt and Olivier De Clerck designed the study. Ellen Cocquyt carried out lab 

work. Ellen Cocquyt and Frederik Leliaert maintained algal cultures and performed sequence 

alignment. Ellen Cocquyt and Heroen Verbruggen analyzed data. All authors wrote, revised and 

approved the final manuscript. 

Chapter 3: Ellen Cocquyt, Olivier De Clerck, Heroen Verbruggen and Koen Sabbe designed the study. 

Ellen Cocquyt carried out lab work. Ellen Cocquyt and Frederik Leliaert maintained algal cultures and 

performed sequence alignment. Ellen Cocquyt and Heroen Verbruggen analyzed data and wrote the 

manuscript. Fredirick W. Zechman provided atpB sequences. All authors revised and approved the 

final manuscript. 

Chapter 4: Ellen Cocquyt and Olivier De Clerck designed the study. Ellen Cocquyt and Gillian H. Gile 

carried out labwork. Ellen Cocquyt, Frederik Leliaert and Gillian H. Gile maintained algal cultures. 

Ellen Cocquyt and Heroen Verbruggen analyzed data. Ellen Cocquyt, Heroen Verbruggen and Olivier 

De Clerck wrote the manuscript. All authors revised and approved the final manuscript. 

Chapter 5: Ellen Cocquyt and Olivier De Clerck designed the study. Ellen Cocquyt carried out lab 

work. Ellen Cocquyt and Heroen Verbruggen analyzed data and wrote the manuscript. All authors 

revised and approved the final manuscript. 

Chapter 6: Heroen Verbruggen and Steven T. LoDuca designed and wrote the manuscript. Frederick 

W. Zechman, Diane S. Littler, Mark M. Littler, Frederik Leliaert, and Olivier De Clerck collected 

samples. Ellen Cocquyt, Matt Ashworth, Caroline Vlaeminck and Thomas Sauvage carried out lab 

work. Heroen Verbruggen analyzed data. All authors revised and approved the final manuscript. 

Chapter 7: Frederik Leliaert, Jan Rueness and Christine A. Maggs designed the study. Jan Rueness 

provided the culture strains of Uronema curvatum and Urospora microscopica. Christian Boedeker 

provided sequences of Chaetomorpha and Wittrockiella. Frederik Leliaert and Ellen Cocquyt carried 

out lab work and analyzed the data. All authors revised and approved the final manuscript.

2 

Ancient relationships among green algae inferred from nuclear and 

chloroplast genes 1 

Ellen Cocquyt, Heroen Verbruggen, Frederik Leliaert and Olivier De Clerck 

Phycology Research Group and Center for Molecular Phylogenetics and Evolution, Ghent University, 

Krijgslaan 281 S8, 9000 Ghent, Belgium 

Abstract 

The Chlorophyta is one of the two divisions of green plants and harbors a wide range of green algae. 

The ancient relationships among three classes of this division, the Ulvophyceae, Trebouxiophyceae 

and Chlorophyceae (UTC) have been at the center of a long-standing debate. Our phylogenetic 

analyses (ML and BI) of seven nuclear genes, SSU nrDNA and two plastid markers with carefully 

chosen partitioning strategies and models of sequence evolution result in high support across the 

topology of the Chlorophyta, show the monophyly of the UTC classes and resolve the branching 

order among them. Even though topology tests (AU) do not exclude an alternative branching order of 

UTC classes, we show that moderate removal of fast-evolving sites improves the phylogenetic signal 

in the desired epoch. We also infer the relationships among the orders of the Ulvophyceae, providing 

novel insights into the evolution of multicellularity and multinucleate cells in the green tree of life. 

Keywords 

single-copy nuclear genes, Chlorophyta, green algae, molecular phylogenetics, Ulvophyceae, 

Chlorophyceae, Trebouxiophyceae 

1 submitted article

20 CHAPTER 2 

Introduction 

The green plant lineage or Viridiplantae represents one of three groups of photosynthetic eukaryotes 

that diverged after enslavement of a cyanobacterium to make a primary chloroplast (Rodriguez- 

Ezpeleta et al. 2005). Ultrastructural and molecular studies have identified a major split within the 

Viridiplantae giving rise to two lineages, the Chlorophyta and the Streptophyta (Pickett-Heaps and 

Marchant 1972, Lewis and McCourt 2004). The Streptophyta consists of several lineages of 

freshwater green algae from which land plants evolved approximately 470 million years ago 2 (Karol 

et al. 2001, McCourt et al. 2004, Hall and Delwiche 2007). Whereas considerable progress has been 

made during the past decade in clarifying the relationships among the streptophyte green algae and 

land plants (Parkinson et al. 1999, Turmel et al. 2003, Turmel et al. 2006, Lemieux et al. 2007, Moore 

et al. 2007, Rodriguez-Ezpeleta et al. 2007, Saarela et al. 2007), the phylogeny and evolutionary 

history of the Chlorophyta has been more difficult to elucidate. 

Members of the Chlorophyta are common inhabitants of aquatic environments and exhibit a 

remarkable morphological and cytological diversity ranging from the world’s smallest free-living 

eukaryote Ostreococcus, over multicellular filaments and foliose blades, to highly complex siphonous 

life forms. Four classes are recognized within the Chlorophyta: Prasinophyceae, Ulvophyceae, 

Trebouxiophyceae and Chlorophyceae. The predominantly marine planktonic Prasinophyceae form a 

paraphyletic assemblage of unicellular organisms from which the Ulvophyceae, Trebouxiophyceae 

and Chlorophyceae (UTC) are derived (Steinkötter et al. 1994, Fawley et al. 2000, Lopez-Bautista and 

Chapman 2003, Guillou et al. 2004). The latter three classes form a monophyletic group termed UTC 

clade but the relationships among them form the basis of a longstanding debate (Pröschold and 

Leliaert 2007). All possible relationships between the UTC classes have been hypothesized over the 

years, depending on which ultrastructural characters were interpreted or which DNA locus or 

phylogenetic analysis method was used (Fig. 1). The unstable relationships exhibited among these 

three classes are likely due to a combination of their ancient age and the short time span over which 

they diverged from one another (O'Kelly 2007). The fossil record indicates the presence of the classes 

in the mid-Neoproterozoic (Butterfield et al. 1994) and molecular clock estimates situate the UTC 

divergence in the early Neoproterozoic (Douzery et al. 2004, Zimmer et al. 2007, Herron et al. 2009). 

The UTC lineages occupy distinct and widely divergent ecological niches. The Chlorophyceae and 

Trebouxiophyceae are ubiquitous in freshwater and soil ecosystems. Furthermore, trebouxiophycean 

algae are the favored partner in symbiotic relationships with fungi to form lichens. Members of the 

Ulvophyceae, on the other hand, are best known as marine macroalgae in coastal ecosystems, but 

some of them live in damp subaerial habitats such as humid soil, rocks, tree bark and leaves (Lopez- 

Bautista and Chapman 2003, Watanabe and Nakayama 2007). Having evolved from a marine 

prasinophyte ancestor, the UTC lineages have experienced one to several transitions to a freshwater 

2 This estimate is based on a plant phylogeny derived from 27 plastid protein-coding genes calibrated with 

plant fossils (Sanderson et al. 2003). Estimates based on phylogenies of eukaryotes derived from 75 nuclear 

protein coding genes calibrated with vertebrate fossil suggest that vascular plant already diverged from mosses 

about 700 mya, which suggest that land plants are at least 700 my old (Heckman et al. 2001; Hedges 2004). A 

possible explanations for this huge difference might be the use of other calibration points (i.e. plant versus 

vertebrate fossils).

PHYLOGENY OF GREEN ALGAE 21 

environment, which involves profound physiological adaptation (Mann 1996, Becker and Marin 

2009). Furthermore, marine versus freshwater lifestyles coincide with differentiations in life-style: 

whereas the marine Ulvophyceae have a haplodiplontic life cycle with alternating generations of 

free-living gametophyte and sporophyte phases and sporic meiosis, freshwater green algae 

predominantly have a haploid vegetative phase and a single-celled, often dormant zygote as the 

diploid stage (haplontic life cycle with zygotic meiosis) (Lewis and McCourt 2004). Resolving the 

branching order among the three classes can help to understand the diversification of ecological, 

physiological and life history features of the UTC clade. 

Whereas Prasinophyceae are unicellular marine plankton and Trebouxiophyceae and Chlorophyceae 

are unicells, colonies or in a few cases simple filaments, the predominantly marine class Ulvophyceae 

has evolved an unrivalled diversity of morphologies and cytological types. Morphologies in the class 

range from simple unicells to multicellular organisms of various levels of complexity (the green 

seaweeds). Cytological diversity is equally diverse and ranges from uni- to multinucleate cells and, in 

the extreme, giant tubular cells with millions of nuclei. Five major groups are recognized in the 

Ulvophyceae. The first has fairly typical cells and ranges in morphology from unicells to multicellular 

filaments and sheets, the multicellular types being more derived (orders Oltmannsiellopsidales, 

Ulvales and Ulotrichales) (Cocquyt et al. 2009). The second group has similar cells in a filamentous 

organization (order Trentepohliales). In contrast to these two groups, the other groups possess 

multinucleate cells. Representatives of the third group (Cladophorales) are branched filamentous 

seaweeds consisting of multinucleate cells with a few to thousands of nuclei arranged in non-motile 

cytoplasmic domains (siphonocladous organization). Members of the fourth group (Bryopsidales) are 

essentially unicellular: they consist of a single, giant tubular cell with thousands to millions of nuclei 

and complex patterns of cytoplasmic flow (siphonous organization). This siphonous cell can branch 

and coalesce to form highly complex seaweeds that dominate the flora of tropical coastal ecosystems 

(Vroom and Smith 2003). The seaweeds belonging to the fifth group (order Dasycladales) also feature 

a siphonous organization. Acetabularia and Batophora (Dasycladales) form an exception, because 

they have a single, giant nucleus situated in the rhizoid with which the alga is attached to its rocky 

substrate. Before sexual reproduction, this nucleus divides meiotically followed by several mitotic 

divisions, leading to the formation of numerous nuclei that populate the gametes. In addition to 

these five major multicellular or multinucleate groups, the Ulvophyceae also include the genus 

Ignatius, a soil alga forming clusters of a few cells (Watanabe and Nakayama 2007), and Blastophysa, 

a microscopic endophytic green alga. Several problems surround the classification of the 

Ulvophyceae. First, the monophyly of the Ulvophyceae has been questioned since its establishment 

because it lacks unique ultrastructural synapomorphies (Mattox and Stewart 1984, O'Kelly and Floyd 

1984, Lewis and McCourt 2004). Second, molecular phylogenetic studies have not fully resolved the 

relationships among the orders. Most studies did reveal the presence of two major groups: 

Oltmannsiellopsidales–Ulvales–Ulotrichales, and a clade consisting of Trentepohliales and the 

siphonocladous and siphonous seaweed orders (Zechman et al. 1990, Watanabe et al. 2001, Lopez- 

Bautista and Chapman 2003, Watanabe and Nakayama 2007). However, the relationships within the 

latter clade, and the positions of the enigmatic genera Ignatius and Blastophysa have not been 

resolved satisfactorily. 

The Chlorophyta thus offer a unique opportunity to study the proliferation of cytological types, 

morphological complexity, ecophysiological adaptations and life cycle evolution. A first step towards

22 CHAPTER 2 

understanding this diversification is to resolve the phylogenetic relationships among the UTC classes 

and the ulvophycean orders. The fact that previous molecular phylogenetic analyses have not yielded 

a satisfactory resolution of these taxa can be attributed at least in part to the difficulty in obtaining a 

good combination of taxon and gene sampling. Single-gene analyses with good taxon sampling have 

suffered from poor resolving power due to the low number of characters (Watanabe and Nakayama 

2007). Conversely, studies based on longer alignments from complete organelle genomes have 

suffered from sparse and uneven taxon sampling (Pombert et al. 2004, Pombert et al. 2005), which 

can lead to systematic error in phylogenetic analysis (Brinkmann et al. 2005, Philippe et al. 2005). 

Our goal in this study is to infer ancient relationships within the Chlorophyta, more specifically the 

divergence of UTC classes and ulvophycean orders. Our approach consists of phylogenetic analysis of 

a dataset that offers a good balance between taxon and gene sampling. Our alignment consists of 

seven single-copy nuclear markers, SSU nrDNA and two plastid genes for 43 taxa representing the 

major lineages of the Viridiplantae. Phylogenetic analyses are carried out with model-based 

techniques, paying careful attention to the selection of suitable partitioning strategies and models of 

sequence evolution. Noise-reduction techniques targeted at improving signal in the epoch of interest 

are applied. 

Figure 1. The three alternative topologies for Ulvophyceae, Trebouxiophyceae and Chlorophyceae (UTC) have 

been supported by ultrastructural characteristics, molecular phylogenies and/or genomic structural features. 

Topology hypothesis testing using an AU tests based on our dataset containing seven nuclear genes, SSU 

nrDNA and two plastid genes suggested that an alternative branching order for the UTC classses (second 

column) was not significantly worse than the ML solution (first column), which supports a sister relationship 

between the Ulvophyceae and Chlorophyceae. Topology hypothesis testing using Bayesian factors provided 

very strong support in favor of the T(UC) topology. 

Results 

We generated EST data from a siphonocladous ulvophyte, Cladophora coelothrix, to facilitate the 

development of PCR primers for single-copy nuclear genes across the green algae. We selected 43 

taxa representing all major lineages of the green algae and obtained DNA sequences for seven 

nuclear genes, the nuclear SSU rDNA and two plastid genes (Fig. S1, Table S1). The concatenated 

matrix is 10209 bp long and is 63% filled at the gene level and 61% at the nucleotide level (Fig. S1).


Our phylogenetic tree shows high resolution of the backbone of the green algal tree of life, including 

the branching order among the UTC classes and most ulvophycean orders (Figs. 2, S2 and S3). It 

reveals a sister relationship between the Ulvophyceae and Chlorophyceae, with the branch joining 

these two classes being very short, indicative of a rapid diversification. Although Bayes factors 

strongly favor the T(UC) configuration over the C(UT) and U(TC) topologies, AU tests suggested that 

an alternative branching order for the UTC classes was not significantly worse than the ML solution 

(Fig. 1). 

Because the former result may in part be due to the application of tests that rely on character 

resampling to a dataset with a suboptimal signal to noise ratio, we used a site removal approach to 

counteract the erosion of ancient phylogenetic signal caused by fast-evolving sites. Our approach 

consisted of maximizing the phylogenetic signal in the epoch corresponding to the radiation of UTC 

classes and ulvophycean orders (Fig. 3). Incremental removal of fast-evolving sites did not change the 

topology but yielded a positive trend in phylogenetic signal in the relevant epoch (Fig. 3D, GE = gain in 

relevant epoch), with an optimum at moderate amounts of site removal (25%). Fig. 3C clearly 

illustrates that statistical support for phylogenetic relationships within the relevant epoch is 

substantially improved by removing the 25% fastest sites, at the expense of signal in more recent 

epochs that are outside the scope of our study. The tree in Fig. 2 was inferred from the dataset with 

optimal signal in the relevant epoch (25% sites removed); the result of the phylogenetic analysis on 

the full data matrix is given in Fig. S2. 

Discussion 

Ancient, rapid radiations are difficult to resolve even with large datasets because there has been 

little time for the accumulation of substitutions, the underlying genes may have discordant 

phylogenetic histories, and inference methods can suffer from systematic error in some 

circumstances, e.g. the long branch attraction artifact (Delsuc et al. 2005, Wiens et al. 2008). We 

have taken some precautions to avoid systematic error: the use of multiple genes, broad taxon 

sampling, and the incremental removal of fast-evolving sites. The use of broad taxon sampling has 

been proven to be one of the most important design criteria leading to improved accuracy in 

phylogenetic studies (Zwickl and Hillis 2002, Geuten et al. 2007). Additionally, because fast-evolving 

sites can erode the overall phylogenetic signal of a dataset by masking the more reliable signal of 

slow sites, removing them can reveal a more accurate signal (Delsuc et al. 2005). 

Radiation of the UTC classes 

Our phylogenetic tree indicates an early divergence of the freshwater-terrestrial clades 

Trebouxiophyceae and Chlorophyceae from marine prasinophycean progenitors, closely followed by 

the diversification of the marine Ulvophyceae. A consistent timeframe of chlorophytan evolution is 

still lacking, mainly due to ambiguities in the interpretation of the scarce fossils, the estimated age of 

the clade being highly dependent on the taxonomic assignment of key fossils like Proterocladus and 

Palaeastrum (Butterfield et al. 1994, O'Kelly 2007). Yet, our data clearly shows that the radiation of

24 CHAPTER 2 

the UTC clades and the coinciding ecological diversification was rapid. Even with a liberal estimate of 

the Streptophyta-Chlorophyta split set at 1200 My (Herron et al. 2009), the UTC radiation would have 

occurred in ca. 20 to 30 million years (unpublished results), complicating interpretation of the 

ecological diversification. 

The relationships between the Ulvophyceae, Trebouxiophyceae and Chlorophyceae have not been 

unambiguously resolved up until now. Phylogenies based on SSU nrDNA sequences have been 

inconclusive, with the branching order of the UTC classes depending on taxon sampling, alignment 

methods and phylogenetic techniques (Fig. 1). Some early SSU phylogenies showed a sister 

relationship between Chlorophyceae and Trebouxiophyceae (Friedl 1995, Bhattacharya et al. 1996, 

Krienitz et al. 2001, Lopez-Bautista and Chapman 2003) while others revealed a sister relation 

between the Chlorophyceae and Ulvophyceae (Friedl and O'Kelly 2002, Lewis and Lewis 2005, 

Watanabe and Nakayama 2007). Although chloroplast phylogenomics has been valuable to resolve 

problematic relationships among green algae (Qiu et al. 2006, Jansen et al. 2007, Lemieux et al. 2007, 

Turmel et al. 2008, Turmel et al. 2009), it has not been able to resolve the branching order of the UTC 

classes conclusively. Whereas phylogenetic analysis of 58 concatenated chloroplast genes supported 

a sister relation between Ulvophyceae and Trebouxiophyceae, chloroplast gene order data and 

genome structure (gene losses and rearrangements within conserved gene clusters), as well as a 

phylogenetic analysis of seven mitochondrial genes supported a sister relationship between 

Ulvophyceae and Chlorophyceae (Pombert et al. 2004, Pombert et al. 2005). 

Ultrastructural evidence has been interpreted as either providing support for a sister relation 

between Chlorophyceae and Trebouxiophyceae (Mattox and Stewart 1984) or between 

Trebouxiophyceae and Ulvophyceae (Sluiman 1989) based on the shared presence of a phycoplast 

and a non-persistent mitotic spindle in the first case and a counterclockwise orientation of the 

flagellar basal bodies in the second case (Fig 1). Our phylogeny suggests an alternative branching 

order that necessitates reinterpretation of the evolution of ultrastructural features. It now appears 

more likely that the ancestor of the UTC clade had a phycoplast and a non-persistent mitotic spindle, 

and that both characters have been lost in the Ulvophyceae. The ancestral status of a phycoplast and 

a non-persistent mitotic spindle is congruent with the presence of these features in the 

Chlorodendrales (e.g. Tetraselmis), the closest prasinophyte relatives of the UTC clade (Mattox and 

Stewart 1984). Likewise the ancestor of the UTC clade probably had a counterclockwise orientation 

of the flagellar basal bodies, evolving to a direct opposite or clockwise orientation in the 

Chlorophyceae. Again this interpretation is congruent with the presence of a counterclockwise basal 

body orientation in Tetraselmis. The modern UTC clade appeared after that a new mode of cell 

division, which is mediated by a phycoplast and enables complex multicellular growth, was 

introduced in the marine ancestor of the UTC clade. The following ecophysiological adaptations lead 

to the success of the Chlorophyceae and Trebouxiophyceae in freshwater habitats (Becker and Marin 

2009).


Figure 2. Phylogeny of the green plant lineage obtained by maximum likelihood inference of the 25% site 

stripped dataset containing 7 nuclear genes, nuclear SSU nrDNA and plastid genes rbcL and atpB. Numbers at 

nodes indicate ML bootstrap values (top) and posterior probabilities (bottom); values below respectively 50 

and 0.9 are not shown. BCD clade stands for the orders Bryopsidales, Cladophorales and Dasycladales. 

Diversification and specialization of the Ulvophyceae 

For the first time, the monophyly of the Ulvophyceae is inferred with high statistical support (BV = 

98; PP = 1.00). In addition, our phylogeny provides a solid hypothesis to explain the cytological and 

ecophysiological differentiation of this predominantly marine class of green algae, by resolving the

26 CHAPTER 2 

relationships among the major clades. Our phylogeny reveals several ancient lineages which diverged 

early in the evolution of the Ulvophyceae. In contrast to previous studies, the basal divergences in 

the present tree are well-supported. Congruent with earlier studies, the clade comprising Ulvales and 

Ulotrichales branches first. Ignatius and Trentepohliales, two relatively inconspicuous and somewhat 

neglected taxa both confined to subaerial habitats, form separate, ancient lineages. They are 

associated with the clade consisting of Bryopsidales, Cladophorales and Dasycladales (BCD clade) (BV 

84; PP 0.96). Contrary to expectations based on ultrastructure (O'Kelly and Floyd 1984), the 

Dasycladales group with the Bryopsidales instead of the Cladophorales 3 . 

The position of Ignatius has never been fully resolved, being either embedded in the Ulvales— 

Ulotrichales clade or clustering with the Trentepohliales and the BCD clade (Watanabe and 

Nakayama 2007). The relationship of Ignatius with the Trentepohliales and the BCD clade revealed in 

this study is corroborated by features of the flagellar apparatus (Watanabe and Nakayama 2007) and 

the shared presence of the elongation factor-1 alpha gene instead of the elongation factor-like gene 

used in all earlier diverging chlorophytan lineages (Cocquyt et al. 2009). The enigmatic, endophytic 

green alga Blastophysa, which has often been allied with the Bryopsidales (Burrows 1991, Brodie et 

al. 2007, Kraft 2007), is here revealed to be sister to the Cladophorales (BV 91 and PP 1.00). This 

relationship is supported by morphological, ultrastructural, cytological, and biochemical features 4 

(O'Kelly and Floyd 1984, Chappell et al. 1991). 

The observed branching pattern yields insight in the morphological and cytological diversification of 

the Ulvophyceae. Multicellularity evolved independently in the Ulvales—Ulotrichales and the 

Trentepohliales + BCD clade. After multinucleate cells evolved in a common ancestor of the BCD 

clade, the siphonous thallus structure evolved from a multicellular state in the Bryopsidales and 

Dasycladales. This interpretation is further supported by the occurrence of cross walls at the base of 

reproductive structures in one of the two major bryopsidalean lineages (Bryopsidineae). The 

evolution of siphonocladalean and siphonous architectures coincides with several cytological and 

cytoskeletal specializations. These giant-celled green algae have evolved a unique mechanism of 

wounding response, in which injured cells contract their protoplasm into numerous spheres that 

regenerate into new cells and thalli (O'Neil and La Claire 1984, Menzel 1988, Kim et al. 2001). This 

mechanism has also given rise to segregative cell division, a specialized type of cell division found in 

some Cladophorales (Leliaert et al. 2007). Unlike the Cladophorales, the siphonous seaweeds 

Bryopsidales and Dasycladales are characterized by vigorous cytoplasmic streaming in which 

organelles are transported in a thin parietal layer of cytoplasm. This cytoplasmic transport has been 

shown to have certain selective advantages, such as transport of nutrients from the rhizoidal holdfast 

3 Common features of the Dasycladales and Cladophorales relate to certain aspects of the flagellar apparatus: 

its flattened aspect, structure of the striated distal fibers and absence of terminal caps. Based on these 

observations O’Kelly and Floyd O'Kelly C. J. and G. L. Floyd. 1984. Correlations among patterns of sporangial 

structure and development, life histories, and ultrastructural features in the Ulvophyceae. Pp. 121-156 in D. 

Irvine, and D. John, eds. Systematics of the green algae. The Systematics Association Special Volume 27, 

Academic Press, London and Orlando. suggested a closer relationship between Dasycladales and Cladophorales 

than to any other group within the Ulvophyceae. 

4 See Chapter 8 for a more detailed discussion


to the upright photosynthetic thallus parts (Littler et al. 1988, Chisholm et al. 1996), or chloroplast 

migration for optimal photosynthesis (Drew and Abel 1992). Despite the fact that siphonous thalli are 

essentially composed of a single giant cell, many bryopsidalean and dasycladalean representatives 

form large, differentiated thalli with a complex organization of interwoven siphons (Verbruggen et al. 

2009). 

The Ulvophyceae are an essentially marine clade, but a few representatives have adapted to 

freshwater environments (a number of Ulvales, Ulotrichales and Cladophorales) and subaerial 

habitats (Ignatius, Trentepohliales and two Cladophorales species). Our phylogeny thus implies that 

the transition to freshwater and terrestrial habitats must not only have occurred in the 

Chlorophyceae and Trebouxiophyceae (Becker and Marin 2009), but also several times 

independently within the Ulvophyceae. 

Figure 3. Illustration of the site stripping approach. 

(A) Based on a rate-smoothed tree, the epoch in which the relationships of interest are situated is selected 

(gray band across figure). 

(B) The strength of the phylogenetic signal in the data (measured as average bootstrap values [BV] in a sliding 

window across the tree) is plotted for an analysis from which the 25% fastest-evolving characters had 

been stripped and compared to the condition without site stripping. 

(C) The gain in bootstrap values, calculated as the BV of site stripped condition minus the BV of the condition 

without site stripping is plotted. This graph shows net gain in older epochs and net loss in younger epochs. 

In this example, there is net gain in the epoch of interest (dark gray). 

(D) The gain in the epoch of interest is calculated for all site stripping conditions. This table shows that 

moderate site stripping yields optimal phylogenetic signal in the relevant epoch.

28 CHAPTER 2 

Progress and perspectives 

The slow progress in green algal phylogenetics can in part be attributed to the limited availability of 

genomic data and the difficulties in consistently amplifying single-copy nuclear markers over the 

entire spectrum of a group of organisms that dates back well into the Proterozoic. Our results 

indicate that advanced phylogenetic analyses of multiple markers from different genomic 

compartments show great promise in resolving ancient divergences within the green algae. The 

present phylogeny has provided us with a framework to advance our understanding of the nature of 

the morphological, cytological and ecological diversification of the Chlorophyta. Additional insight 

would be gained from a dated phylogeny, which would allow an assessment of the timing of these 

evolutionary events and, by consequence, the global ecological circumstances under which this 

radiation took place. 

Material and methods 

Algal strains 

Algal strain information is provided in Table S2. Cultures were grown under cool white fluorescent 

lights at 18°C, with a 12/12h light/dark cycle. Dasycladales, Cladophorales and Derbesia were grown 

at 23°C. Bold's basal medium and f/2 medium was used for freshwater and marine species, 

respectively (Andersen 2005). 

RNA isolation 

Total RNA was extracted with a RNeasy Plant Mini Kit (Qiagen) or a NucleoSpin® RNA Plant kit 

(Machery-Nagel) according to the manufacturer’s instructions, including a DNAse step to eliminate 

genomic DNA contamination. RNA quality was checked on a 1% agarose gel and with 

spectrophotometric analysis (Sambrook et al. 1989). 

cDNA library construction and primer design 

A standard cDNA library was constructed from 30 µg of total RNA of Cladophora coelothrix by VERTIS 

Biotechnologie AG (Germany) and screened for nuclear genes potentially useful for green algal 

phylogeny. After the cDNA library had been plated, 200 clones were randomly picked and sequenced 

with M13 reverse primer (Table S3). Clones with a positive blastx hit to the Swiss-Prot database were 

also sequenced with the M13 forward primer (Table S3). 

For each recovered gene, an alignment including GenBank sequences from other green plants was 

made. Primers were designed for sufficiently long genes that contained conserved regions and tested 

on a variety of green algae, leading to successful amplification of 4 nuclear genes across a wide range 

of green algae: 40S ribosomal protein S9, 60S ribosomal proteins L3 and L17 and oxygen-evolving 

enhancer protein (OEE1). Actin and some glyceraldehyde-3-phosphate dehydrogenase (GapA)


primers were found in literature (Table S3). Glucose-6-phosphate isomerase (GPI) and other GapA 

primers were based upon aligned GenBank sequences from green algae (e.g., available genome 

sequences of Ostreococcus and Chlamydomonas) and land plants (Table S3). Primers of rbcL 

(Hanyuda et al. 2000), atpB (Wolf 1997, Karol et al. 2001) and SSU nrDNA (Zechman et al. 1990, Lewis 

and Lewis 2005) were taken from the literature. 

Reverse Transcriptase and Polymerase Chain Reaction 

For amplification of the nuclear genes, cDNA was constructed from total RNA with the Omniscript RT 

kit (Qiagen) according to the manufacturer’s instructions. The reaction was incubated for several 

hours at 37°C. PCR amplification was performed with the following reaction mixture: 1 µl of cDNA, 

2.5 µl of 10x Buffer (Qiagen), 0.5 µl dNTP’s (10 mM), 0.5 µl MgCl (25 mM, Qiagen), 0.5 µl of each 

primer (10 µM), 0.25 µl BSA (10 µg/µl), 18.125 µl sterilized MilliQ water and 0.125 µl Taq polymerase 

(5 U/µl, Qiagen). The amplification profile consisted of an initial denaturation of 2 min at 94 °C, 

followed by 35 cycles of 30 s at 94 °C, 30 s at 50°C or 55 °C and 45 s at 72 °C and a final extension of 

10 min at 72 °C. Products of expected size (Table S4) were either sequenced directly or cloned first. 

Cloning and sequencing 

PCR products were first sequenced with the forward primer with an Applied Biosystems 3130xl. 

Sequences were blasted against the GenBank protein database (blastx) to check for potential 

bacterial contaminants. Sequences without ambiguous base calls yielding a significant hit for 

Viridiplantae were further sequenced with the reverse primer. When ambiguous base calls were 

present in sequences, samples were cloned if the rough sequence gave a significant blastx hit for 

Viridiplantae. Cloning was performed with the pGEM®-T Vector System (Promega) according to the 

manufacturer’s instructions. After ligation, transformation and incubation, the white colonies were 

transferred to 15 µl distilled water, boiled for 10 minutes to lyse cells and centrifuged to pellet cells 

walls and harvest the DNA in the liquid phase. Three to five clones were PCR amplified and 

sequenced with the vector specific primers T7 and SP6 following the protocol described above. 

Alignment 

Sequence reads were assembled with AutoAssembler 1.4.0 (ABI Prism, Perkin Elmer). Additional 

sequences were retrieved from GenBank and aligned with our own sequences (Table S1). Nuclear 

and plastid genes were aligned by eye, taking their corresponding amino acids into account. SSU 

nrDNA sequences were aligned based on their RNA secondary structure (Cocquyt et al. 2009). The 

ten loci were concatenated, yielding an alignment of 10209 bases.

30 CHAPTER 2 

Model selection procedure 

A suitable partitioning strategy and suitable models of sequence evolution were selected with a 

three-step procedure based on the Bayesian Information Criterion (BIC). The guide tree used during 

the entire procedure was obtained by ML analysis of the unpartitioned concatenated alignment with 

PhyML, using a JC+ 8 model (Guindon and Gascuel 2003). All subsequent likelihood optimizations 

and BIC calculations were carried out with Treefinder (Jobb et al. 2004). The first step consisted of 

optimizing the likelihood for eight potential partitioning strategies, assuming a HKY+ 8 model for 

each partition. The three partitioning strategies with the best fit to the data (lowest BIC scores) were 

retained for further evaluation. The second step involved model selection for individual partitions. 

The likelihood of each partition present in the three retained partitioning strategies was optimized 

for three variants of the general time reversible model (F81, HKY and GTR), with and without among- 

site rate heterogeneity (+ 8). Because not all genes were sampled for all taxa, the guide tree was 

pruned to the taxa present in the partition in question before each optimization. The partitionspecific 

models obtaining the lowest BIC score were passed on to the third step, which consisted of a 

re-evaluation of the three partitioning strategies retained from the first step using the models 

selected for these partitions in the second step. The partitioning strategy plus model combination 

that received the lowest BIC score in the third step was used in the phylogenetic analyses. The model 

selection procedure proposed 8 partitions: SSU nrDNA was partitioned into loops and stems (2 

partitions), nuclear and plastid genes were partitioned according to codon position (2 3 partitions). 

GTR+ 8 was the optimal model for all partitions. 

Phylogenetic analysis 

Maximum likelihood analysis was performed with TreeFinder, which allows likelihood searches under 

partitioned models (Jobb et al. 2004). Due to the relatively low tree space coverage in TreeFinder 

compared to other ML programs, an analysis pipeline was created to increase tree space coverage by 

running analyses from many starting trees. A first set of starting trees was created by randomly 

modifying the PhyML guide tree by 100 and 200 nearest neighbor interchange (NNI) steps (50 

replicates each). ML searches were run from these 100 starting trees and the three trees yielding the 

highest likelihood were used as the starting point for another set of NNI modifications of 20 and 50 

steps (50 replicates each). A second set of ML searches was run from each of the resulting 300 

starting trees. The tree with the highest likelihood resulting from this set of analyses was retained as 

the global ML solution. The bootstrap resampling method was used to assess statistical support 

(1000 pseudo-replicates). 

Bayesian phylogenetic inference was carried out with MrBayes 3.1.2 (Ronquist and Huelsenbeck 

2003). Two parallel runs, each consisting of four incrementally heated chains were run for 25 million 

generations, sampling every thousand generations. Convergence of log-likelihoods and parameter 

values was assessed in Tracer v1.4 (Rambaut and Drummond 2007). A burnin sample of 2.5 million 

trees was removed before constructing the majority rule consensus tree.

Topological hypothesis testing 


Approximately unbiased tests (AU test, Shimodaira 2002) were used to test the alternative 

relationships between UTC classes. Site likelihoods were calculated by maximum likelihood 

optimization in Treefinder using the same model specifications as for phylogenetic inference. AU 

tests were performed with CONSEL V0.1i (Shimodaira and Hasegawa 2001). 

In addition Bayes Factors were used to evaluate the alternative branching order between UTC classes 

(Kass and Raftery 1995). Two additional Bayesian analyses using identical conditions as the original 

analysis that yielded the T(UC) phylogeny were run while constraining the topology to conform to 

either C(UT) or U(TC). The harmonic mean estimates of the marginal likelihoods of the competing 

hypotheses were obtained with the sump command in MrBayes and used to calculate the Bayes 

Factors (BF) as twice the difference in marginal likelihood between the competing hypotheses 

(Nylander et al. 2004). 

Removal of fast-evolving sites: site stripping 

We applied a modified site stripping approach (Waddell et al. 1999) to improve the signal to noise 

ratio in the epoch of interest. Site-specific rates were calculated in HyPhy (Pond et al. 2005). 

Progressive removal of the fastest sites, 5% each time, resulted in a set of alignments that were 

subjected to ML analysis as described above. We used a novel modus operandi to reveal the trend of 

signal change with increasing amounts of site stripping and select an optimal site stripping condition 

(Fig. 3). First, we performed rate smoothing on the ML tree obtained from the complete alignment to 

make branch lengths roughly proportional to evolutionary time (Fig. 3A), using the nonparametric 

rate smoothing (NPRS) method with the Powell optimization algorithm implemented in r8s 

(Sanderson 2003). Second, the epoch of interest (containing the UTC and ulvophycean 

diversification) was located in the tree. Third, the strength of phylogenetic signal, measured as 

average bootstrap values (BV) in a sliding window across the rate-smoothed tree, was plotted for all 

site-stripping conditions (e.g. Fig. 3B). Fourth, we calculated signal gain in the epoch of interest (GE) 

by integrating the function representing the difference between the BV of the site-stripped condition 

and the BV of the analysis on the complete alignment across the epoch of interest (Fig. 3C). Finally, 

these gain values were compared between site stripping conditions, leading to the conclusion that 

moderate site stripping (25% of characters removed) yielded maximum signal in the epoch of interest 

(Fig. 3D). The rationale of using BV as a proxy of signal strength is that these values reflect the clarity 

of the signal in the data, i.e., are inversely related to the amount of conflict in the data (Felsenstein 

1985, Soltis and Soltis 2003). It may seem as circular reasoning to use BV to decide on the most 

suitable site stripping condition and subsequently use BV to come to conclusions about the strength 

of relationships. However, this is not entirely true because GE is calculated over a wide temporal 

range and across all lineages in the relevant epoch, so it does not rely exclusively on the BV of the 

branches we wish to resolve.

32 CHAPTER 2 

Acknowledgements 

We thank Barbara Rinkel, Hervé Moreau, Tatiana Klotchkova, Wytze Stam and Jeanine Olsen for 

providing cultures, Caroline Vlaeminck for assisting with the molecular work, Klaus Valentin for cDNA 

library services, Steven Robbens and Yves Van de Peer for early access to Ostreococcus data, Rick 

Zechman for atpB sequences, Peter Dawyndt for additional processing power, and Wim Gillis for IT 

support. Funding was provided by the Special Research Fund (Ghent University, DOZA-01107605) 

and the Research Foundation Flanders (postdoctoral fellowships to HV, FL and ODC). Phylogenetic 

analyses were largely carried out on the KERMIT computing cluster (Ghent University).

Additional files 


Figure S2. Phylogeny of the green plant lineage obtained by maximum likelihood inference of the complete 

dataset containing 7 nuclear genes, nuclear SSU rDNA and plastid genes rbcL and atpB. Numbers at nodes 

indicate ML bootstrap values (top) and posterior probabilities (bottom); values below respectively 50 and 0.9 

are not shown.

ibosomal RNA plastid genes ribosomal protein-coding genes other nuclear genes 

nuclear SSU rDNA (18S) 

atpB 

rbcL 

40S S9 60S L3 

60S L17 actin 

G6PI 

GapA 

OEE1 

1974 bp 

1380 bp 

1380 bp 

537 bp 1152 bp 

468 bp 1122 bp 

753 bp 

966 bp 

477 bp 

Cyanophora paradoxa 1953 1380 1380 520 1152 468 1122 753 966 477 99% of sites 

Cyanidioschyzon merolae 1974 1380 633 1122 753 723 477 62% of sites 

Mesostigma viride 1953 1380 1380 1141 424 1122 684 966 453 92% of sites 

Chlorokybus atmophyticus 1971 1380 1380 741 966 

63% of sites 

Klebsormidium flaccidum 1953 1254 658 537 1137 424 966 428 63% of sites 

Entransia fimbriata 1936 1254 1316 1140 451 57% of sites 

Coleochaete sp. 1953 1255 1354 1122 963 


Spirogyra sp. 1254 1372 1110 966 


Closterium sp. 1974 1353 1060 753 474 54% of sites 

Chara sp. 1940 1254 1380 900 966 477 64% of sites 

Marchantia polymorpha 1953 1380 1380 537 1068 468 648 753 966 477 89% of sites 

Physcomitrella patens 1974 1380 1310 537 1152 468 1122 753 966 477 99% of sites 

Oryza sativa 1974 1380 1380 537 1152 468 1122 753 966 477 100% of sites 

Arabidopsis thaliana 1971 1380 1380 537 1152 468 1122 753 966 477 100% of sites 

Ostreococcus tauri 1973 1380 1380 1134 468 1011 753 966 453 92% of sites 

Nephroselmis olivacea 1974 1380 1380 412 657 519 441 54% of sites 

Scherffelia dubia 242 1059 465 1122 966 264 34% of sites 

Tetraselmis sp. 1953 1266 1356 424 883 738 418 64% of sites 

Chlorella sp. 1953 1380 1380 424 900 753 966 428 77% of sites 

Prototheca wickerhamii 1974 1380 373 1101 468 1118 


Helicosporidium 399 468 893 


Scenedesmus obliquus 1953 1380 1380 537 1120 424 833 507 420 77% of sites 

Chlamydomonas reinhardtii 1974 1380 1380 537 1152 468 1122 753 966 477 100% of sites 

Volvox carteri 1227 1350 537 1152 1122 477 56% of sites 

Halochlorococcum 1960 891 742 


Bolbocoleon piliferum 1922 613 687 519 


Acrochaete repens 1928 1325 537 753 520 426 48% of sites 

Ulva intestinalis 1971 1353 537 870 744 513 451 56% of sites 

Ulva sp. 1563 1354 1122 753 966 417 56% of sites 

Ignatius tetrasporus 1954 424 519 441 27% of sites 

Trentepohlia aurea 1961 1167 417 32% of sites 

Acetabularia sp. 1973 1142 1320 551 130 949 753 428 57% of sites 

Bornetella sp. 1973 1141 1318 906 435 53% of sites 

Halimeda sp. 1949 1209 1326 753 


Derbesia sp. 1323 714 966 435 32% of sites 

Bryopsis sp. 716 1197 1323 1083 900 442 39% of sites 

Codium sp. 729 1380 424 955 452 26% of sites 

Blastophysa rhizopus 741 519 428 12% of sites 

Cladophora albida 1972 424 714 906 


Valonia utricularis 1889 424 753 440 33% of sites 

Phyllodictyon sp. 1889 537 1149 424 831 753 438 55% of sites 

Boodlea composita 1889 537 424 900 753 519 453 46% of sites 

Cladophora coelothrix 1889 537 1152 468 900 753 477 58% of sites 

Boergesenia forbesii 1927 537 424 900 753 


Siphonocladus tropicus 1927 537 424 900 753 441 45% of sites 




37% of sites 36% of sites 




Fig. S1. Data availability matrix. Graphical representation of our concatenated alignment, showing the availability of sequence data. The color of column and row headers indicate the 

amount of data available for that column or row. Green indicates high data availability, red indicates low data availability and yellow/orange represents intermediate data availability. 

outgroup 

Streptophyta 

Prasinophyceae 

Trebouxiophyceae 

Chlorophyceae 

Ulvophyceae 

0% 20% 40% 60% 80% 100% 

Data availability: percent of sites present


Figure S3. Phylogeny of the green plant lineage inferred from amino-acid sequences of 7 nuclear genes and 

plastid genes rbcL and atpB and nucleotide sequences of nuclear SSU rDNA with Bayesian techniques. Numbers 

at nodes indicate posterior probabilities; values below 0.9 are not shown. 

Because TreeFinder cannot simultaneously analyze nucleotide and amino acid sequences only MrBayes was 

used to infer this tree. Amino acid sequences used a WAG model with among-site rate heterogeneity (+ 8), 

SSU nrDNA was partitioned into loops and stems and used a GTR+ 8 model. Two parallel runs, each consisting 

of four incrementally heated chains were run for 3 million generations, sampling every thousand generations. 

Convergence of log-likelihoods and parameter values was assessed in Tracer v1.4 (Rambaut and Drummond 

2007). A burnin sample of 300.000 trees was removed before constructing the majority rule consensus tree.

Table S1. Genbank accession numbers for nucleotide sequences of actin, glucose-6-phosphate isomerase, glyceraldehydes-3-phosphate dehydrogenase, histone, oxygen 

evolving protein, 40S ribosomal protein S9 and 60S ribosomal protein L3 and L17. 

actin G6PI GapA histone OEE1 40S_S9 60S_L3 60S_L17 

Chlorophyta 

Ulvophyceae 

Ulotrichales 

Halochlorococcum moorei R102 R102 R73_2 

Ulvales 

Acrochaete repens R94 R94_Gap2 R94_2 R94 R94_2 

Bolbocoleon piliferum R95 R95_GapM R60_2 

Enteromorpha sp. R109 R106 R109_GapM R109_1 R109_1_Oxy1 R109_3UTR 

Ulva sp. AB106563 (U. pertusa) R110_10 R49_2 R110_1 R110_Oxy2 

Ignatius-clade 

Ignatius tetrasporus R66_GapM R66_4 R66 R66 

R105 

Bryopsidales 

Blastophysa rhizopus R104 R104_GapM R104 R104 

Bryopsis sp. R105 R105_1 R105/ AB293980 

(B. plumosa) 

Codium sp. R63 R63 R63_1_Oxy1 R63 

Derbesia sp. R98 R98_2 R98 R98 

Halimeda spp KZN2K4_20 (H. cuneata2) R65_1 

Dasycladales 

Acetabularia acetabulum R7 AAL00001487 (TBestDB) R103 R103_1_Oxy2 EC096877 CF259099 

Bornetella sphaerica R80 R80_1 R80 

Siphonocladales 

Boergesenia forbesii R21_1 R21_2 R78 R78_1 R78 

Boodlea composita R8_1 R38_6 R79_GapM R79 R79_1 R79 

Cladophora albida R88_2 R88_19 R88 R88 

Cladophora coelothrix R35_3 R35_12 cDNA_lib117 cDNA_lib122 cDNA_lib128 cDNA_lib44 cDNA_lib80 

Phyllodictyon spp R18 R85 R85 R85 R85_2 R85 R85 

Siphonocladus tropicus R29_3 R84 R84_5 R84 R84_1 R84 

Valonia utricularis R86 R86_5 R86 R86 

Trentepohliales 

Trentepohlia aurea R107 R107_Oxy2 

Chlorophyceae 

Chlamydomonas reinhardtii D50838 R44_2 L27668 R96_5 X13826 R96_2 = RPS9 R96 = RPL3 XM_001693402 

Scenedesmus obliquus R69 SOL00005809 (TBestDB) R69 R69 EC189050 EC189501 R69 

Volvox carteri M33963 X06963 AF110780 FD918100 FD837306 


Chlorella kessleri R97 R97_5 R97_5 R97 R97 R97_1 

Helicosporidium sp. AF317896 CX128748 CX128902 CX128917 

Prototheca wickerhamii EC181529 EC181191 EC183246 EC180670

actin G6PI GapA histone OEE1 40S_S9 60S_L3 60S_L17 


Chlorodendrales 

Scherffelia dubia AF061018 DQ270259 AJ919716 AJ919712 AL132935 AJ919390 

Tetraselmis spp R76 (T. striata) R76 (T. striata) R76 (T. striata) R76_Oxy2/ AB293977 

R76 (T. striata) 

(T. cordiformis) 

Mamiellales 

Ostreococcus tauri CR954216 Steven DQ649076 R68_2 R68 R68 CR954206 

Pseudoscourfieldiales 

Nephroselmis olivacea EC732532 R112_clon R112_2_GapM R112 R112/ AB293978 

Streptophyta 

Mesostigmatales 

Mesostigma viride AF061020 R111 R111_11 R111 R111/ DN255652 R111 R111 

Chlorokybales 

Chlorokybus atmophyticus R99 DQ270263 R99 

Klebsormidiales 

Entransia fimbriata R101 R101_1_Oxy1 R101 

Klebsormidium flaccidum R46_7 R67 R67_1_Oxy2 R67_2 R67 R67 

Zygnematales 

Closterium sp. R4 R4_3 AB293981 

R113_3 (C. orbicularis) 

Spirogyra sp. AF061021 AJ246030 

Coleochaetales 

Coleochaete spp AF061019 (C. scutata) DQ270264 (C. 

scutata) 

Charales 

AB293979 (C. braunii) 

Chara spp DQ846905 (C. contraria) DQ270262 (C. 

vulgaris) 

Embryophyta (land plants) 

Marchantiophyta (Liverworts) 

Marchantia polymorpha AB100427 BJ863864 AJ246026 AB185062 BJ844290 BJ861755 BJ852563 BJ853031 

Bryophyta (Mosses) 

Physcomitrella patens XM_001782636 XM_001760154 DQ270266 BJ975811 XM_001763206 XM_001754317 XM_001782386 XM_001767807 

Spermatophyta (Seed plants) 

Arabidopsis thaliana M20016 AB044951 NM_101161 X60429 AJ145957 AB010077 M32655 AC004393 

Oryza sativa X16280 D45217 NM_001059519 NM_001056811 NM_001049669 NM_001055539 D12630 NM_001069236 

other eukaryotes 


Glaucocystophyta 

Cyanophora paradoxa CPU90325 DQ812897 DQ270258 ES235644 AJ784854 EC666063 EC661414 EC666069 

Rhodophyta 

Cyanidiophyceae 

Cyanidioschyzon merolae D32140 CMT497C_Steven AP006492 AP006496 AB159597 AP006495

38 CHAPTER 2 

Table S1 continued. Genbank accession numbers for nucleotide sequences of atpB, rbcL and SSU rDNA. 

Chlorophyta 

Ulvophyceae 

Ulotrichales 

Ulvales 

atpB rbcL SSU rDNA 

Halochlorococcum moorei AY198122 

Acrochaete repens FJ715715 FJ715684 

Bolbocoleon piliferum FJ715716 AY205330 

Enteromorpha sp. AY422552 AJ000040 

Ulva sp. AF499668 (U. fenestrate) 


Ignatius tetrasporus AB110439 

Bryopsidales 

Bryopsis sp. FJ480417 (B. plumosa) FJ715718 FJ715685 (B. plumosa) 

Codium sp. M67453 (C. fragile) FJ715686 (C. platylobium) 

Derbesia sp. AF212142 (D. marina) 

Halimeda spp FJ480416 (H. discoidea) FJ715719 (H. incrassate) AY786526 (H. gracilis) 

Dasycladales 

Acetabularia spp FJ480413 (A. dentate) FJ715714 (A. acetabulum) Z33461 (A. acetabulum) 

Bornetella spp FJ480414 (B. nitida) FJ715717 (B. sphearica) Z33464 (B. nitida) 


Boergesenia forbesii AM498746 

Boodlea composita AF510157 

Cladophora albida Z35317 

Cladophora coelothrix Z35315 

Phyllodictyon spp AF510163 (P. pulcherrimum) 

Siphonocladus tropicus AM498761 

Valonia utricularis Z35323 


Chlorophyceae 



Streptophyta 

Trentepohlia aurea FJ534608/FJ715722 AB110783 

Chlamydomonas reinhardtii NC 005353 NC 005353 M32703 

Scenedesmus obliquus NC008101 NC008101 X56103 

Volvox carteri AB013999 D63446 

Chlorella spp NC 001865 (C. vulgaris) NC 001865 (C. vulgaris) X13688 (C. vulgaris) 

Prototheca wickerhamii AJ245645 X74003 


Tetraselmis spp DQ173248 (T. chuii suecica) DQ173247 (T. chuii suecica) X70802 (T. striata) 

Mamiellales 

Ostreococcus tauri NC 008289 NC 008289 AY329635 


Nephroselmis olivacea NC 00927 NC 00927 X74754 


Mesostigma viride NC 002186 NC 002186 AJ250109 

Chlorokybales 

Chlorokybus atmophyticus DQ422812 DQ422812 M95612 


Entransia fimbriata AY823688 E88 slordig LB2793 

Klebsormidium flaccidum AF408801 E87 rbcL2 X75520 

Zygnematales 

Closterium spp AJ553936 (C. lunula) AF352237 (C. gracile)


atpB rbcL SSU rDNA 

Spirogyra spp AF408797 (S. maxima) FJ715721 (S. sp.) 


Charales 

Coleochaete scutata AY082303 AY082313 X68825 

Chara spp AF408782 (C. connivens) L13476 (C. connivens) U18493 (C. connivens) 





Marchantia polymorpha NC 001319 NC 001319 AB021684 


Physcomitrella patens AP005672 AB066207 X80986 



Rhodophyta 

Arabidopsis thaliana NC 000932 NC 000932 AC006837 

Oryza sativa AC092750 AJ746297 X00755 

Cyanophora paradoxa NC 001675 NC 001675 AY823716 


Cyanidioschyzon merolae AB002583 AB158485

40 CHAPTER 2 

Table S2. Algal strain information. 

Chlorophyta 

Ulvophyceae 

Ulvales 

number Culture collection 

Acrochaete repens E093db Barbara Rinkel (Natural History Museum, London) 

Bolbocoleon piliferum E344pc Barbara Rinkel (Natural History Museum, London) 

Ulva intestinalis EE3 Field collection at Goese Sas (Netherlands) 

Ulva sp. EE2 and EE6 Field collection at Goese Sas (Netherlands) 

Ulotrichales 

Halochlorococcum tenue 19.92 Sammlung von Algenkulturen (University of Göttingen, Germany) 


Ignatius tetrasporus 2012 UTEX culture collection of algae (University of Texas at Austin, USA) 

Bryopsidales 

Blastophysa rhizopus LB 1029 UTEX culture collection of algae (University of Texas at Austin, USA) 

Bryopsis sp. EE4 Field collection at Goese Sas (Netherlands) 

Codium sp HEC 15711 Field collection in Madeira 

Derbesia sp. 2773-1 Tatiana Klotchkova (Kongju National University, Korea) 

Ostreobium quekettii 6.99 Sammlung von Algenkulturen (University of Göttingen, Germany) 


Trentepohlia aurea 483-1 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Dasycladales 

Acetabularia acetabulum LB 2694 UTEX culture collection of algae (University of Texas at Austin, USA) 

Bornetella sphaerica LB 2690 UTEX culture collection of algae (University of Texas at Austin, USA) 


Chlorophyceae 



Streptophyta 

Boergesenia forbesii Boerg2 and 3 Phycology Research Group, Ghent University* 

Boodlea composita Bcomp2, 4 and 6 Phycology Research Group, Ghent University* 

Cladophora albida Calb3 Phycology Research Group, Ghent University* 

Cladophora coelothrix Ccoel1 and 2 Phycology Research Group, Ghent University* 

Phyllodictyon orientale Struv1, West 1631 John West (University of Melbourne, Australia) 

Siphonocladus tropicus Siph2 and 3 Phycology Research Group, Ghent University* 

Valonia utricularis Vutric2 Phycology Research Group, Ghent University* 

Chlamydomonas reinhardtii CC1690 Chlamydomonas Center (Duke University, USA) 

Scenedesmus obliquus 1450 UTEX culture collection of algae (University of Texas at Austin, USA) 

Chlorella kessleri 211-11g Sammlung von Algenkulturen (University of Göttingen, Germany) 

Tetraselmis striata 41.85 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Ostreococcus tauri OTH95 Hervé Moreau (Observatoire Océanologique de Banyuls) 

Nephroselmis olivacea 40.89 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Chlorokybus atmophyticus 48.80 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Entransia fimbriata LB 2793 UTEX culture collection of algae (University of Texas at Austin, USA) 

Klebsormidium flaccidum KL1 Hans Sluiman (Royal Bot Garden Edinburgh, Scotland); described in 

G.M. Lokhorst (1996) 

Mesostigma viride 50-1 Sammlung von Algenkulturen (University of Göttingen, Germany) 


Spirogyra sp. 169.80 Sammlung von Algenkulturen (University of Göttingen, Germany) 

*donated by Jeanine Olsen and Wytze Stam (University of Groningen, Netherlands)


Table S3. Primer sequences used for PCR amplification and sequencing. Primers were found in literature, adapted 

from existing primers or based on aligned Genbank sequence of green algae and land plants (Viridiplantae), often 

complemented with a Cladophora coelothrix cDNA sequence. 

Primer name Primer sequence (5’-3’) Primers based on 

Actin 

Actin_F GAC ATG GAR AAR ATN TGG CAC CAC AC 244F; An et al. (1999) 

Actin_R ATC CAC ATC TGY TGA AAR GTR G Ac3-R; Bhattacharya, Stickes, and Sogin (1993) 

Ac3-R GAA GCA YTT GCG RTG SAC RAT Bhattacharya, Stickes, and Sogin (1993) 

Fern5-F CTT GTY TGY GAC AAT GGA TCW GGA ATG GT An et al. (1999) 

Glucose-6-phosphate isomerase 

G6PI_F TTY GCR TTY TGG GAC TGG G Ostreococcus and land plants 

G6PI_R CCC CAC TGG TCR AAI GAR TT Ostreococcus and land plants 

Glyceraldehyde-3-phosphate dehydrogenase 

GapA_Fb GSN ATH AAY GGN TTY GG Petersen et al. (2006) 

GapA_Re CCA YTC RTT RTC RTA CCA Petersen et al. (2006) 

GapA_Fm GCI YSI TGC ACI ACY AAC TG green algae and land plants 

Oxygen-evolving enhancer protein, chloroplastic (OEE1) 

Oxy_F1 TGA CCT AYI TGC ARR TYA AGG G Viridiplantae and cDNA library Cladophora 

Oxy_F2 CIG GCI TKG CCA ACA CIT GCC C Viridiplantae and cDNA library Cladophora 

Oxy_R GAR CCA CCR CGG CCC TTG GGG TC Viridiplantae and cDNA library Cladophora 

40S ribosomal protein S9 (RPS9) 

40S_S9_F ATG CCG AAG ATC GSC YAK TAY CGC Viridiplantae and cDNA library Cladophora 

40S_S9_R CAC KCK AGC GTG GTG GAT GGA CTT Viridiplantae and cDNA library Cladophora 

40S_S9_3’UTR GTC TAT CCT GTC CAG CCT GAG CAG C Viridiplantae and cDNA library Cladophora 

60S ribosomal protein L3 (RPL3) 

60S_L3_F ATG TCI CAC AGI AAG TTT GAA CAC CC Viridiplantae and cDNA library Cladophora 

60S_L3_R GTC TTG CGI GGM AGI CGG GTG AC Viridiplantae and cDNA library Cladophora 

60S ribosomal protein L17 (RPL17) 

60S_L17_F CAA GGC GCG CGG GTC GGA YCT Viridiplantae and cDNA library Cladophora 

60S_L17_R CAT GTA CGG GTT GAT GCG RCC RTG CGC Viridiplantae and cDNA library Cladophora 

cDNA library pExCell primers 

M13_F GTA AAA CGA CGG CCA GT 

M13_R GGA AAC AGC TAT GAC CAT G

42 CHAPTER 2 

Table S4. PCR conditions, expected length of PCR products and total length of the alignment used for 

phylogenetic inferences. 

primer combination Annealing Length of Total length Comment 

temperature PCR product alignment 

Actin 1122 bp 

Actin_F + Actin_R 55°C ≈ 850 bp 

Ac3-R + Fern5-F 55°C ≈ 1000 bp Closterium 

Glucose-6-phosphate isomerase 753 bp 

G6PI_F + G6PI_R 55°C ≈ 760 bp 

Glyceraldehyde-3-phosphate dehydrogenase 966 bp 

GapA_Fb + GapA_Re 50°C ≈ 1000 bp 

GapA_Fm +GapA_Re 50°C ≈ 550 bp 

Oxygen-evolving enhancer protein, chloroplastic 762 bp 

Oxy_F1 + Oxy_R 50°C ≈ 470 bp 

Oxy_F2 + Oxy_R 50°C ≈ 500 bp 

40S ribosomal protein S9 (RPS9) 537 bp 

40S_S9_F + 40S_S9_R 55°C ≈ 400 bp 

40S_S9_F + 40S_S9_3’UTR 55°C ≈ 550 bp Ulva intestinalis 

60S ribosomal protein L3 (RPL3) 1185 bp 

60S_L3_F + 60S_L3_R 55°C ≈ 730 bp 

60S ribosomal protein L17 (RPL17) 477 bp 

60S_L17_F + 60S_L17_R 55°C ≈ 380 bp

3 

Gain and loss of elongation factor genes in green algae 1 

Abstract 

Background 

Two key genes of the translational apparatus, elongation factor-1 alpha (EF-1α) and elongation 

factor-like (EFL) have an almost mutually exclusive distribution in eukaryotes. In the green plant 

lineage, the Chlorophyta encode EFL except Acetabularia where EF-1α is found, and the Streptophyta 

possess EF-1α except Mesostigma, which has EFL. These results raise questions about evolutionary 

patterns of gain and loss of EF-1α and EFL. A previous study launched the hypothesis that EF-1α was 

the primitive state and that EFL was gained once in the ancestor of the green plants, followed by 

differential loss of EF-1α or EFL in the principal clades of the Viridiplantae. In order to gain more 

insight in the distribution of EF-1α and EFL in green plants and test this hypothesis we screened the 

presence of the genes in a large sample of green algae and analyzed their gain-loss dynamics in a 

maximum likelihood framework using continuous-time Markov models. 

Results 

Within the Chlorophyta, EF-1α is shown to be present in three ulvophycean orders (i.e., Dasycladales, 

Bryopsidales, Siphonocladales) and the genus Ignatius. Models describing gene gain-loss dynamics 

revealed that the presence of EF-1α, EFL or both genes along the backbone of the green plant 

phylogeny is highly uncertain due to sensitivity to branch lengths and lack of prior knowledge about 

ancestral states or rates of gene gain and loss. Model refinements based on insights gained from the 

EF-1α phylogeny reduce uncertainty but still imply several equally likely possibilities: a primitive EF- 

1α state with multiple independent EFL gains or coexistence of both genes in the ancestor of the 

Viridiplantae or Chlorophyta followed by differential loss of one or the other gene in the various 

lineages. 

Conclusions 

EF-1α is much more common among green algae than previously thought. The mutually exclusive 

distribution of EF-1α and EFL is confirmed in a large sample of green plants. Hypotheses about the 

gain-loss dynamics of elongation factor genes are hard to test analytically due to a relatively flat 

likelihood surface, even if prior knowledge is incorporated. Phylogenetic analysis of EFL genes 

indicates misinterpretations in the recent literature due to uncertainty regarding the root position. 

1 Published as: Cocquyt, E.*, H. Verbruggen*, F. Leliaert, F. W. Zechman, K. Sabbe, and O. De Clerck. 2009. Gain 

and loss of elongation factor genes in green algae. BMC Evolutionary Biology 9:39. 

* Equal contributors

44 CHAPTER 3 

Background 

Elongation factor-1 alpha (EF-1α) is a core element of the translation apparatus and member of the 

GTPase protein family. The gene has been widely used as a phylogenetic marker in eukaryotes; either 

to resolve their early evolution (e.g., Baldauf et al. 1996, Roger et al. 1999) or more recent 

phylogenetic patterns (e.g., Hashimoto et al. 1994, Baldauf and Doolittle 1997, Beltran et al. 2007, 

Sung et al. 2007, Zhang and Qiao 2007). The evolutionary history of genes used for such inferences 

should closely match that of the organisms and not be affected by ancient paralogy or lateral gene 

transfer (Keeling and Inagaki 2004). A gene related to but clearly distinguishable from EF-1α, called 

elongation factor-like (EFL), appears to substitute EF-1α in a scattered pattern: several unrelated 

eukaryote lineages have representatives that encode EFL and others that possess EF-1α. The EFL and 

EF-1α genes are mutually exclusive in all but two organisms: the zygomycete fungus Basidiobolus and 

the diatom Thalassiosira (James et al. 2006, Kamikawa et al. 2008). Although EFL is found in several 

eukaryotic lineages, EF-1α is thought to be the most abundant of both (Rogers et al. 2007). So far, 

EFL has been reported in chromalveolates (Perkinsus, dinoflagellates, diatoms, haptophytes, 

cryptophytes), the plant lineage (green and red algae), rhizarians (cercozoans, foraminifera), unikonts 

(some Fungi and choanozoans) and centrohelids (Keeling and Inagaki 2004, Gile et al. 2006, Noble et 

al. 2007, Kamikawa et al. 2008, Sakaguchi et al. 2008). 

The mutually exclusive distribution of EF-1α and EFL suggests similar functionality. The main function 

of EF-1α is translation initiation and termination, by delivering aminoacyl tRNAs to the ribosomes 

(Negrutskii and El'skaya 1998). Other functions include interactions with cytoskeletal proteins: 

transfer, immobilization and translation of mRNA and involvement in the ubiquitine-dependent 

proteolytic system, as such forming an intriguing link between protein synthesis and degradation 

(Negrutskii and El'skaya 1998). In contrast, the function of EFL is barely known. It is assumed to have 

a translational function because the putative EF-1β, aa-tRNA, and GTP/GDP binding sites do not differ 

between EF-1α and EFL (Keeling and Inagaki 2004). Based on a reverse transcriptase quantitative PCR 

assay in the diatom Thalassiosira, which possesses both genes, it was proposed that EFL had a 

translation function while EF-1α performed the auxiliary functions (Kamikawa et al. 2008). 

The apparently scattered distribution of EFL across eukaryotes raises questions about the gain-loss 

patterns of genes with an important role in the cell. This mutually exclusive and seemingly scattered 

distribution can be explained by two different mechanisms: ancient paralogy and lateral gene 

transfer. Ancient paralogy was considered unlikely because this would imply that both genes were 

present in ancestral eukaryotic genomes during extended periods of evolutionary history while the 

genes rarely coexist in extant species (Keeling and Inagaki 2004). Furthermore, a prolonged 

coexistence of both genes in early eukaryotes would have likely resulted in either functional 

divergence or pseudogene formation of one or the other copy (Van de Peer et al. 2001), as is 

suggested for EFL and EF-1α coexisting in the diatom Thalassiosira (Kamikawa et al. 2008). Keeling 

and Inagaki (Keeling and Inagaki 2004) proposed lateral gene transfer of the EFL gene between 

eukaryotic lineages as the most likely explanation for the scattered distribution of both genes. 

In the green plants (Viridiplantae), EF-1α and EFL seem to show a mutually exclusive distribution. Of 

the two major green plant lineages, the Chlorophyta were shown to have EFL with the exception of 

Acetabularia where EF-1α is found, and the Streptophyta were shown to possess EF-1α with the

GAIN AND LOSS OF ELONGATION FACTOR GENES 45 

exception of Mesostigma, which has EFL (Noble et al. 2007). Noble et al. (2007) proposed the 

hypothesis that EFL was introduced once in the ancestor of the green lineage, followed by 

differential loss of EF-1α or EFL in the principal clades of the Viridiplantae (i.e., Streptophyta and 

Chlorophyta). 

The goals of the present study are to extend our knowledge of the distribution pattern of EF-1α and 

EFL in the green algae and investigate patterns of gain and loss of these key genes of the translational 

apparatus. We applied a RT-PCR and sequencing-based screening approach across a broad spectrum 

of green algae, with emphasis on the ulvophycean relatives of Acetabularia. To test the hypothesis of 

Noble et al. (2007), we modeled patterns of gene gain and loss. To this goal, a reference phylogeny 

based on three commonly used loci was inferred, and gain-loss dynamics of EFL and EF-1α were 

optimized along this phylogeny using continuous-time Markov models. 

Results and Discussion 

Distribution of elongation factors in the green algae 

EF-1α sequences were retrieved from streptophytes Entransia (Klebsormidiophyceae) and 

Chlorokybus (Chlorokybophyceae), confirming previous observations that all Streptophyta except 

Mesostigma have EF-1α. We found EFL sequences in Chlorella (Trebouxiophyceae), Acrochaete and 

Bolbocoleon (Ulvophyceae), Nephroselmis and Tetraselmis striata (prasinophytes), further confirming 

the formerly established distribution pattern within the Chlorophyta. We reaffirmed the presence of 

EFL in Chlamydomonas and Scenedesmus (Chlorophyceae), Ulva intestinalis and U. fenestra 

(Ulvophyceae) and Ostreococcus (prasinophytes), previously shown by Noble et al. (2007). In addition 

to Acetabularia, EF-1α was discovered in representatives of the ulvophycean orders Dasycladales 

(Bornetella), Bryopsidales (Blastophysa, Bryopsis, Codium, Derbesia, Ostreobium), Siphonocladales 

(Boodlea, Cladophora, Dictyosphaeria, Ernodesmis, Phyllodictyon) and in Ignatius (see Figures 1 and 

2). The RT-PCR approach did not reveal the presence of both genes in any of the screened species 

despite the fact that our primers could amplify the target genes across the Viridiplantae. Our RT-PCR 

experiments on two species whose genomes have been sequenced (Chlamydomonas and 

Ostreococcus) yielded a single gene for each species, a result in compliance with the knowledge 

derived from their genome sequences (DOE Joint Genome Institute). 

The reference phylogeny, inferred from a DNA matrix consisting of 72 taxa representing all major 

plant lineages and three loci (SSU rDNA, rbcL and atpB), is in accordance with recent phylogenetic 

studies, including the position of Mesostigma within the Streptophyta (Lemieux et al. 2007, 

Rodriguez-Ezpeleta et al. 2007). Figure 1 shows the phylogenetic relationships among the taxa for 

which we have information on elongation factors; the full 72-taxon phylogeny can be found as an 

online supplement [see Additional file 1]. Even though the tree shows improved resolution from 

previous studies, large parts of the backbone remained poorly resolved. In order to obtain a solid 

hypothesis of green algal evolution, much additional sequence data may have to be gathered. The 

occurrence of EF-1α and EFL in terminal taxa was plotted on the reference phylogeny in Figure 1. 

Mesostigma is the only streptophyte which encodes EFL. Within the chlorophytan class Ulvophyceae,

46 CHAPTER 3 

the order Ulvales possesses EFL whereas the other orders encode EF-1α (Dasycladales, 

Siphonocladales, Bryopsidales and Ignatius). 

Figure 1. Distribution of EF-1α and EFL in the green plants. The type of elongation factor is indicated with black 

(EF-1α) or gray (EFL) squares. The reference phylogeny was obtained by Bayesian phylogenetic inference of 

nuclear SSU rDNA and the plastid genes rbcL and atpB. Numbers at nodes indicate posterior probabilities (top) 

and ML bootstrap values (bottom); values below respectively 0.9 and 50 are not shown.


Figure 2. Phylogenies inferred from EF-1α and EFL amino-acid sequences with Bayesian techniques. Sequences 

belonging to the green plant lineage are in gray boxes. Whereas all green plant EF-1α sequences group in a 

single clade, the green plant EFL sequences seem to form separate lineages. Sequences generated for this 

study are indicated with triangles. Numbers at nodes indicate posterior probabilities (top) and ML bootstrap 

values (bottom); values below respectively 0.9 and 50 are not shown. 

Phylogenies of EF-1α and EFL 

All green plant EF-1α sequences form a monophyletic group clearly differentiated from EF-1α 

sequences of a variety of other eukaryotes (Figure 2A). Even though the Viridiplantae form a strongly 

supported group, resolution among and within Streptophyta and Chlorophyta is generally low, which 

could in part be due to some short EF-1α sequences included in the analysis. 

In contrast, green plant EFL genes do not form a monophyletic lineage (Figure 2B). Although the 

backbone of the phylogeny is moderately resolved, monophyly of green plant EFL genes is unlikely 

because it is not observed in the MCMC output (zero posterior probability). EFL sequences of the 

Viridiplantae can be found in several clades. The chlorophytes, trebouxiophytes, ulvophytes and 

prasinophyte Tetraselmis form a single monophyletic group. The other prasinophyte EFL sequences 

form two separate groups. The last clade consists of the streptophyte Mesostigma. 

To obtain an accurate root position for our EFL tree, we included related subfamilies of the GTPase 

translation factor superfamily: EF-1α, eukaryotic release factor 3 (eRF3), heat shock protein 70 

subfamily B suppressor (HBS1) and archaebacterial EF-1α sequences in our analyses. In accordance 

with Keeling and Inagaki (Keeling and Inagaki 2004), the tree is rooted with archaebacterial EF-1α 

sequences. All analyses (Bayesian and ML) resulted in a phylogeny very similar to the one shown in

48 CHAPTER 3 

Figure 2B, the complete phylogeny with all related subfamilies can be found as an online supplement 

[see Additional file 2]. This phylogeny shows seven EFL clades, with the following branching order: 

Bigelowiella, the diatoms, Planoglabratella, the cryptophyte Goniomonas, red algae, choanozoans, 

and a large clade containing the green plant lineage, chromalveolates (dinoflagellates, haptophytes, 

cryptophytes), fungi and Rhaphidiophrys (Figures 2B). Deep branches generally received low 

statistical support, preventing strong conclusions about the relationship between the seven clades. 

Gain-loss dynamics 

The scattered distribution of EF-1α and EFL in the green plant lineage is a remarkable phenomenon 

that raises questions about evolutionary patterns of gain and loss of both genes. Noble et al. (2007) 

proposed the hypothesis that EF-1α was present in the common ancestor of the plant lineage, 

followed by a single gain of EFL early in evolution of the green lineage and subsequent differential 

loss of one or the other gene in the various lineages. Our aim was to test this hypothesis by modeling 

gain-loss dynamics and inferring ancestral presence-absence patterns of both genes in a maximum 

likelihood framework. Gene gain and loss rates were estimated by maximum likelihood (ML) 

optimization, using a dataset of presence-absence patterns of EF-1α and EFL and a reference 

phylogeny derived from the Bayesian analysis of three commonly used loci (SSU nrDNA, rbcL and 

atpB). 

A first analysis, based on the reference tree, shows uncertain character state probabilities along the 

backbone of the Viridiplantae and suggests a loss of EF-1α in early Chlorophyta evolution and regain 

in some Ulvophyceae (Figure 3A). Because branch lengths play a crucial role in model optimization, 

the analysis was repeated on an alternative version of the reference tree in which branch lengths 

were transformed using a rate smoothing approach. Since our tree deviates from the molecular 

clock, we performed rate smoothing to obtain branch lengths roughly proportional to time. Rate 

smoothing techniques relax the assumption of constant rates of evolution throughout the tree: 

differences in rates of molecular evolution are smoothed out by assuming that evolutionary rates 

change gradually throughout the phylogeny. The result is an ultrametric tree in which branch lengths 

are roughly proportional to evolutionary time instead of amounts of molecular evolution. Modeling 

gain-loss dynamics of elongation factor genes along the rate-smoothed tree yields results that 

strongly deviate from those obtained with the original reference tree: probabilities of the character 

states along the major part of backbone are now around 50 % for EFL and around 50 % for the 

presence of both genes (Figure 3B). Subsequently, an additional level of realism was introduced by 

taking phylogenetic uncertainty into account because several nodes in the reference tree are poorly 

supported. To this goal, all post-burnin MCMC trees were rate-smoothed and analyzed individually. 

The results were summarized on the rate-smoothed reference tree. Taking phylogenetic uncertainty 

into consideration had a minor influence on the probabilities of the characters states (Figure 3C). 

Although the exact numbers differ between analyses, gene gain rates were always lower than gene 

loss rates, reinforcing the notion that gene transfers are rare events in comparison to losses of 

redundant genes (Barker et al. 2007). Whereas the analysis based on the original reference tree 

returned faster gain and loss rates for EFL than for EF-1α, analyses based on rate-smoothed trees


(including MCMC trees) suggested the inverse, marking the sensitivity of Markov models to the unit 

of operational time. 

Figure 3. Gain-loss dynamics of green algal elongation factor genes and their inferred presence in ancestral 

genomes. Gain and loss rates, as well as the estimated probabilities for presence of the genes in ancestral 

genomes are given for a variety of analysis conditions. Panels A-C show the outcome of models in which EF-1α 

and EFL gain and loss rates were not constrained. In panels D-F, the gain rate of EF-1α was constrained to be 

10 -6 . Colors were used to visualize estimated probabilities for presence of genes along the tree. Red indicates a 

high probability for EF-1α, blue marks a high probability of EFL and yellow stands for a high probability of the 

presence of both genes. Intermediate colors indicate uncertainty.

50 CHAPTER 3 

From these results, it seems fair to conclude that there is major uncertainty about the ancestral 

states for a variety of reasons, including sensitivity to branch lengths and lack of prior knowledge 

about ancestral states or rates of gene gain and loss. Considering that the ancestors must have had 

either EF-1α, EFL or both genes opens perspectives for hypothesis comparison in a likelihood 

framework. Additionally, information about rates of gene gain and loss could be gleaned from the EF- 

1α and EFL phylogenies. 

Analyses constrained with various hypotheses about ancestral gene content resulted in a confidence 

set of 8 trees that differ extensively [see Additional file 3]. The fact that strongly different hypotheses 

are also present in the confidence set denotes that the likelihood surface is too flat to draw firm 

conclusions in favor of one or another hypothesis. 

The last option to reduce uncertainty is to inform the Markov models with information on gains and 

losses gleaned from the EF-1α and EFL trees (cf. Barker et al. 2007). Because green plant EF-1α 

sequences form a monophyletic and strongly supported lineage, it seems fair to assume vertical 

descent of EF-1α throughout the Viridiplantae. This knowledge can be introduced in our Markov 

model by setting a very low gain rate of EF-1α. If the analysis is constrained in this way, both EFL and 

EF-1α were inferred to be present along the backbone of the Viridiplantae in the original reference 

tree (Figure 3D) and a 50/50 probability for the presence of EF-1α or both genes was obtained in the 

rate-smoothed trees (Figures 3E-F). Comparison of hypotheses about ancestral gene content 

constrained with a very low EF-1α gain rate reduced the confidence set to 3 trees in which either EF- 

1α or both genes are present along the backbone [see Additional file 4]. The ML solution (hypothesis 

122) assumes that only EF-1α was present along the backbone of the tree and consequently shows 

independent gains of EFL in Mesostigma, prasinophytes, Chlorophyceae, Trebouxiophyceae and 

Ulvales. An alternative scenario (hypothesis 123) in the confidence set has EF-1α at the base of the 

Viridiplantae, a gain of EFL in the ancestor of the Chlorophyta, and subsequent differential loss of 

one or the other gene in the various lineages. Information from the EFL phylogeny may provide clues 

for further distinction between either multiple transfers or ancient paralogy with subsequent losses. 

The green EFL sequences form a highly supported clade together with dinoflagellates, cryptophytes, 

haptophytes, fungi and Rhaphidiophrys, suggesting lateral gene transfer of the EFL gene between 

these distant eukaryotic lineages (Andersson 2005, Keeling and Palmer 2008). Considering the ability 

of chromalveolates (i.e., dinoflagellates, cryptophytes and haptophytes) and Raphidiophrys to feed 

through phagocytosis (Sakaguchi et al. 2002) and the absence of this behavior in green algae, it 

would be tempting to assume that lateral gene transfer occurred from green algae to the 

dinoflagellates, cryptophytes, haptophytes and Raphidiophrys instead of the other way around. 

Phagotrophic eukaryotes have been shown to have elevated rates of lateral gene transfer (Gogarten 

2003, Andersson 2005) because this feeding mechanism enables them to continually recruit genes 

from engulfed prey (Nosenko et al. 2006). Lateral gene transfers to fungi, although known to exist 

(Andersson et al. 2003), would require a different explanation because neither phagotrophy nor 

endosymbiosis occur in fungi. However, in the light of this peripheral information, it would be 

tempting to conclude that both EF-1α and EFL essentially show vertical descent in green plants and 

that the observed mutually exclusive pattern of EFL and EF-1α sequences results from differential 

loss. In this scenario, lateral gene transfer must have occurred from green algal cells to other 

eukaryotic lineages.


In previous studies of functionally similar eukaryotic genes with mutually exclusive distributions, 

distinction between ancient paralogy with subsequent losses and multiple transfers was made based 

on two main criteria (Rogers et al. 2007). The first criterion states that if one gene dominates the tree 

and the other occurs in only a few lineages, multiple independent transfers should be regarded as 

the most probable explanation whereas equal representation would suggest common ancestry with 

subsequent differential loss. The second criterion is about the age of the taxa involved. If the 

mutually exclusive pattern occurs between closely related species, one can conclude common 

ancestry with subsequent losses. If the pattern is more ancient, multiple lateral transfers are a more 

probable explanation. It is obvious that such criteria are very difficult to apply in real situations. 

These difficulties can be overcome by taking a probabilistic angle on the problem and modeling gainloss 

dynamics with continuous-time Markov models. This approach brings statistical rigor to the 

analysis of gene presence-absence patterns and has the potential to discriminate between the 

alternative scenarios of ancient paralogy with differential losses and multiple independent lateral 

transfers. Application of this technique to our dataset of green algal elongation factors revealed the 

difficulty of arriving at firm conclusions about ancient gene transfer events because of a relatively flat 

likelihood surface and, consequently, ambiguous probabilities for gene content at ancestral nodes. 

When informed with external information, the analyses allow somewhat more definitive conclusions. 

The broader eukaryotic picture 

Figure 4. Visualization of the posterior 

probability of rooting of the EFL tree. The 

topology represents the unrooted topology of 

EFL genes. Branch width is proportional to the 

posterior probability that the outgroup, 

consisting of archaebacterial EF-1α, EF-1α, 

eRF3 and HBS1 sequences, attaches to the 

ingroup tree at that point. Numbers at 

branches represent the total posterior 

probability that the root is situated along the 

branch in question. 

In addition to the information gained about elongation factor evolution in green algae, our results 

also highlight misinterpretations in recent literature on EFL evolution across the eukaryotes. Previous 

studies have not been explicit about whether or how their phylogenetic trees were rooted, but have 

drawn conclusions that require directionality in the tree. Kamikawa et al. (2008) concluded that 

lateral gene transfer from a foraminifer (Planoglabratella) to the ancestor of the diatoms must have

52 CHAPTER 3 

occurred because the diatom sequences were nested within the Rhizaria (foraminifera and 

cercozoans). In case their tree was unrooted, this conclusion is flawed due to a lack of directionality 

in the tree. In their presentation of the tree, choanozoans are used as one of the basal clades, 

probably because they were the earliest-branching lineage in the tree presented by Keeling and 

Inagaki (2004). Our EFL tree, which includes EF-1α, eRF3 and HBS1 sequences and is rooted with 

archaebacterial EF-1α sequences, indicates that the directionality inferred by Kamikawa et al. (2008) 

is likely to be wrong. Our phylogram (Figures 2B) suggest that the root position of EFL lies on the 

branch leading towards the cercozoan Bigelowiella, but support is lacking for the basal relationships. 

A plot of the posterior distribution of root placements (Figure 4) illustrates the uncertainty about the 

root placement more clearly. It is evident from this plot that the choanozoans are not the oldest 

diverging lineage. This finding overturns the conclusion from Kamikawa et al. (2008) because the 

nested position of the diatom EFL genes within the Rhizaria sequences can no longer be maintained. 

Our EFL phylogeny supports the presence of lateral gene transfer between eukaryotic lineages, 

however, the direction of lateral gene transfer is difficult to evaluate. 

Conclusions 

The mutually exclusive nature of EF-1α and EFL is confirmed in a large sample of green algae. The 

Streptophyta possess EF-1α with the exception of Mesostigma, which has EFL. The Chlorophyta 

encode EFL with the exception of Dasycladales, Bryopsidales, Siphonocladales and Ignatius, where 

EF-1α is found. This result establishes EF-1α as a widespread gene among green algae. 

Gain-loss models revealed that the probabilities of the presence of EF-1α, EFL or both genes along 

the backbone of the plant phylogeny are highly uncertain, and that a previously published hypothesis 

(Noble et al. 2007) is as likely as several other hypotheses. Model refinements based on insights 

gained from the EF-1α phylogeny were unable to distinguish between three possibilities: (1) multiple, 

independent gains of EFL throughout the plant lineage, (2) a single gain of EFL early in evolution of 

the plant lineage followed by differential loss, or (3) independent gains of EFL in Mesostigma and the 

ancestor of the Chlorophyta followed by differential loss of one or the other gene in the various 

lineages (Figure 3 D-F and Additional file 4). 

Further research into the gain-loss dynamics of elongation factors of green plants and eukaryotes in 

general is needed to come to more definitive conclusions about their evolution. First, the EFL 

phylogeny should be refined by obtaining full-length sequences for a set of relevant taxa to confirm 

or reject the presence of multiple independent green lineages in this tree. The use of codon models 

may help to achieve this (Seo and Kishino 2008). An alternative approach would be to learn about the 

processes responsible for lateral transfer of elongation factors by studying their flanking regions for 

signature sequences of mobile elements (Zhang et al. 2006, Dybvig et al. 2007). Finally, studying gainloss 

dynamics across a wider spectrum of eukaryotic supergroups should lead to more stable 

conclusions. In addition to yielding more precise parameter estimates for gene gain and loss rates, a 

eukaryote-wide study would allow the use of more specific models for lateral gene transfer because 

both donor and recipient lineages would be present in the analysis (Than et al. 2006, Galtier 2007, 

Lake 2008). It remains an enigma that the evolution of elongation factors, genes crucial for cell 

functioning, is marked by such complex gain-loss patterns.

Methods 

Algal strains 


Algal strain information is provided as additional material online [see Additional file 5]. All cultures 

were grown at 18°C, except Dasycladales, Siphonocladales and Derbesia (23°C). Cool white 

fluorescent lamps were used for a 12/12 h light/dark cycle. Marine cultures were maintained in f/2 

medium and freshwater cultures in Bold’s Basal Medium (Andersen 2005). 

RNA isolation and cDNA library construction of Cladophora coelothrix 

Total RNA was extracted with a RNeasy Plant Mini Kit (Qiagen Benelux b.v., Venlo, the Netherlands) 

or a NucleoSpin® RNA Plant kit (Macherey-Nagel GmbH & Co. KG, Düren, Germany) according to the 

manufacturer’s instructions, including a DNase step to eliminate genomic DNA contamination. RNA 

quality was checked on a 1 % agarose gel (made with 1x TAE diluted in 0,1 % DEPC water). RNA 

concentration and purity were measured in a spectrophotometer at 260 and 280 nm according to 

standard methods (Sambrook et al. 1989). 

Approximately 30 µg of total RNA of Cladophora coelothrix was extracted as described above. A 

standard cDNA library was constructed by VERTIS Biotechnologie AG (Freising , Germany). An EF-1α 

sequence of 624 bp was obtained by sequencing randomly picked clones. 

Reverse Transcriptase and Polymerase Chain Reaction 

cDNA construction was performed with the Omniscript RT kit (Qiagen) and oligodT primers according 

to the manufacturer’s instructions; the reaction was incubated for several hours at 37°C. 

Primers were designed to fit the most conserved regions of EF-1α and EFL sequences across 

Viridiplantae. Primers for EF-1α were based upon aligned GenBank sequences from green algae 

(Acetabularia and Chara) and land plants, completed with our Cladophora coelothrix cDNA sequence 

(EF-1α-F: 5’-GGC CAT CTT ATC TAC AAG CTT GGC GG-3’ and EF-1α-R: 5’-CCA GGA GCA TCA ATC ACG 

GTG CAG-3’). EFL primers were adapted from Noble et al. (Noble et al. 2007) (EFL-F: 5’-TCC ATY GTS 

ATY TGC GGN CAY GTC GA-3’ and EFL-R: 5’-CTT GAT GTT CAT RCC RAC RTT GTC RCC-3’). PCR 

amplification was performed with the following reaction mixture: 1 µl of cDNA, 2.5 µl of 10x Buffer 

(Qiagen), 0.5 µl dNTP’s (10 mM), 0.5 µl MgCl (25 mM, Qiagen), 0.5 µl of each primer (10 µm), 0.25 µl 

BSA (10 µg/µl), 18.125 µl sterilized MilliQ water and 0.125 µl Taq polymerase (5 U/µl, Qiagen). The 

amplification profile consisted of an initial denaturation of 2 min at 94 °C, followed by 35 cycles of 30 

s at 94 °C, 30 s at 55 °C and 45 s at 72 °C and a final extension of 10 min at 72 °C. Products of 

expected size (300 bp for EF-1α and 900 bp for EFL) were either sequenced directly or cloned and 

sequenced.

54 CHAPTER 3 

Cloning and sequencing 

PCR products were first sequenced with the forward primer with an Applied Biosystems 3130xl. 

Sequences were blasted against the GenBank protein database (blastx), to check for potential 

bacterial contaminants. Sequences without ambiguous base calls yielding a significant hit for 

Viridiplantae were further sequenced with the reverse primer. When ambiguous base calls were 

present in sequences, samples were cloned if the rough sequence gave a significant blastx hit for 

Viridiplantae. Cloning was performed with the pGEM®-T Vector System (Promega Benelux b.v., 

Leiden, the Netherlands) according to the manufacturer’s instructions. After ligation, transformation 

and incubation, the white colonies were transferred to 15 µl double distilled water, boiled for 10 

minutes to lyse cells and subsequently centrifuged to pellet the cells walls and allow harvest of the 

DNA in the liquid phase. Between three and five clones were PCR amplified and sequenced with the 

vector specific primers T7 and SP6 following the protocol described above. Cloning showed minor 

polymorphisms that most likely represent different alleles. 

Alignments and phylogenetic analysis of EF-1α and EFL 

Sequences [see Additional file 6] were assembled with AutoAssembler 1.4.0 (ABI Prism, Perkin Elmer, 

Foster City, CA, USA) and aligned manually for both genes separately, resulting in EF-1α and EFL 

alignments of 1374 and 1653 bp, respectively [see Additional file 7]. Sequences generated with our 

primers begin in the N-terminal part the of the gene and are 900 bp for EFL and 150-300 bp for EF- 

1α. We included eukaryotic EF-1α, eRF3 and HBS1 sequences as well as archeabacterial EF-1α 

sequences to serve as outgroups for the EFL phylogeny (Keeling and Inagaki 2004). Due to the large 

divergences between EFL and the other genes, Gblocks was run to remove ambiguously aligned 

regions (Castresana 2000). We ran Gblocks v.0.91b, allowing smaller final blocks, gap positions within 

the final blocks and less strict flanking positions, resulting in an alignment of 358 amino acids [see 

Additional file 7]. The resulting EFL and EF-1α alignments were subjected to Bayesian phylogenetic 

inference with MrBayes 3.1.2 (Ronquist and Huelsenbeck 2003) using the model suggested by 

ProtTest 1.4 (Abascal et al. 2005) (WAG with among site rate heterogeneity: gamma distribution 

with 8 categories). Two parallel runs, each consisting of four incrementally heated chains were run 

for 1,000,000 generations, sampling every 1,000 generations. Convergence of log-likelihoods was 

assessed in Tracer v1.4 (Rambaut and Drummond 2007). A burnin sample of 100 trees was removed 

before constructing the majority rule consensus tree for each of the genes. Maximum likelihood 

phylogenies were inferred for EF-1α and EFL with Treefinder (Jobb 2008). The analyses were based 

on amino acid sequences and used a WAG model with among site rate heterogeneity (gamma 

distribution with 8 categories). One thousand non-parametric bootstrap trees were inferred. 

Bootstrap values were summarized with consense from the Phylip package (Felsenstein 2005) and 

plotted onto the Bayesian consensus tree. 

Phylogeny of the green plants: SSU rDNA, rbcL and atpB 

A reference phylogeny of green plants for which the presence of EF-1α or EFL is known was 

constructed using three commonly used phylogenetic markers: nuclear SSU rDNA and plastid atpB


and rbcL genes and rooted with red algae and a glaucophyte. To obtain an even species distribution 

and consequently a better phylogenetic tree (Zwickl and Hillis 2002), many additional species were 

included in the phylogenetic analysis [see Additional file 7]. Sequences were retrieved from GenBank 

and aligned with our own sequences [see Additional file 6]. DNA was extracted using a standard CTAB 

method. PCR conditions followed standard protocol. Primers were based on other publications: SSU 

rDNA (Zechman et al. 1990, Lewis and Lewis 2005), rbcL (Hanyuda et al. 2000) and atpB (Wolf 1997, 

Karol et al. 2001). The rbcL and atpB sequences were aligned by eye. The SSU rDNA sequences were 

aligned based on their RNA secondary structure with DCSE [see Additional file 8]. 

The model selection procedure [see Additional file 8] proposed eight partitions: atpB and rbcL genes 

were partitioned into codon positions (6 partitions) and the SSU rDNA was partitioned into RNA 

loops and stems (2 partitions). Bayesian phylogenetic inference was carried out using a GTR model 

with gamma distribution and 8 rate categories per partition (all parameters unlinked) and rate 

multipliers to accommodate rate differences among partitions. Two parallel runs, each consisting of 

four incrementally heated chains were run for 5,000,000 generations, sampling every 1,000 

generations. Convergence of log-likelihoods was assessed in Tracer v1.4 (Rambaut and Drummond 

2007). A burnin sample of 3,000 trees was removed before constructing the majority rule consensus 

tree. A maximum likelihood phylogeny was inferred with Treefinder (Jobb 2008). The analysis used a 

GTR model with among site rate heterogeneity (gamma distribution with 8 categories). One 

thousand non-parametric bootstrap trees were inferred. Bootstrap values were summarized with 

consense from the Phylip package (Felsenstein 2005) and plotted onto the Bayesian consensus tree. 

To obtain trees suitable for modeling gene gain and loss, the Bayesian consensus tree and the 

complete post-burnin set of trees were pruned to the set of species for which the type of elongation 

factor is known using the APE package (Paradis et al. 2004). Because our data deviate from the 

molecular clock, we performed rate smoothing to obtain branch lengths that are roughly 

proportional to time. We used the penalized likelihood method (Sanderson 2002) implemented in 

the r8s program (Sanderson 2003), with a log-additive penalty and a smoothing value of 2, which was 

the optimal value in cross-validation (Sanderson 2002). PL rate smoothing was applied to the 

Bayesian consensus tree as well as the post-burnin set of MCMC trees. 

Modeling gene gain and loss 

If the presence of EF-1α and EFL are coded as two binary characters, their gain-loss dynamics can be 

modeled along a reference phylogeny using a continuous-time Markov model. Given the likely 

dependency of gain and loss between EF-1α and EFL, a model designed to study interdependent trait 

evolution was used (Pagel 1994). The rate matrix of this model is given by: 

Q 

D 

0, 

0 

0, 

1 

1, 

0 

1, 

1 

q 

q 

0, 

0 0, 

1 1, 

0 1, 

1 

q q 0 

21 

31 

0 

q 

12 

0 

42 

q 

13 

0 

43 

q 

q 

24 

34 

(1)

56 CHAPTER 3 

where (0,0) indicates the absence of both genes from the genome, (0,1) and (1,0) denote the 

presence of EFL and EF-1α, respectively, and (1,1) is the state where both genes are present in the 

genome. Different q's indicate relative rates of the respective changes in gene content. Transitions 

that require more than one event (e.g. 1,0 → 0,1) are not allowed to occur as a single step in this 

model, the logic being that the probability of two traits changing at exactly the same time is 

negligible. This is consistent with the fact that transitions from EF-1α to EFL and vice versa should 

pass through a stage where both genes are present in the genome. The elements of the diagonal are 

determined by the requirement that each row sums to zero. Because the absence of both genes is 

likely to be lethal, the matrix was constrained by introducing a series of very low rates as follows: 

Q 

D 

10 

10 

-6 

-6 

0 

10 

l 

0 

EF1 

-6 

10 

l 

0 

6 

EFL 

g 

g 

0 

EF1α 

EFL 

(2) 

In this matrix, gEF1α and gEFL denote gain rates and lEF1α and lEFL loss rates. It must be noted that the 

model does not take gene duplications into account because our data provided no indications for the 

presence of such events. 

The rate matrix (2) was specified as a special case of the "discrete dependent" model in BayesTraits 

(Pagel and Meade 2006). The model parameters were estimated by maximum likelihood (ML) 

optimization, using a dataset of presence-absence patterns of EF-1α and EFL. One hundred 

optimization attempts were carried out to find the ML solution. Ancestral state probabilities were 

calculated using the addNode command. The reference phylogeny used for inferring patterns of gain 

and loss was derived from the Bayesian analysis of SSU nrDNA, rbcL and atpB, and was varied as 

follows. First, the majority rule consensus tree provided by MrBayes was used. Second, a ratesmoothed 

version of this consensus tree was used to have branch lengths roughly proportional to 

evolutionary time. Third, topological uncertainty was introduced in the analysis by repeating analyses 

on the entire post-burnin set of MCMC trees after they had been rate-smoothed. For the analysis on 

MCMC trees, ancestral state probabilities were calculated with the addMRCA instead of the addNode 

command. Rate estimates and ancestral state probabilities were averaged across the MCMC trees. 

We opted not to use BayesTraits' Bayesian inference because we found its output to be strongly 

influenced by prior settings. 

In addition to these analyses, several specific hypotheses about ancestral genome content (EFL, EF- 

1α or both) were compared using ML optimization on the rate-smoothed reference tree. Constraints 

on ancestral genome content were placed on 5 ancestral nodes with the fossil command in 

BayesTraits, resulting in 3 5 = 243 hypotheses for which the log-likelihoods could be compared. Only 

hypotheses within two log-likelihood units from the ML solution were retained for interpretation. 

This set of hypotheses can be seen as a confidence set because two log-likelihood units is considered 

a significance threshold for such analyses (Pagel 1999). 

The BayesTraits output was mapped onto the trees with TreeGradients v1.02 (Verbruggen 2008). 

This program plots ancestral state probabilities on a phylogenetic tree as colors along a color 

gradient.

Authors' contributions 


EC, ODC, HV and KS designed the study. EC carried out lab work. EC and FL maintained algal cultures 

and performed sequence alignment. EC and HV analyzed data and drafted the manuscript. FWZ 

provided atpB sequences. All authors revised and approved the final manuscript. 


We thank Barbara Rinkel, Hervé Moreau, Tatiana Klotchkova, Wytze Stam and Jeanine Olsen for 

providing cultures, Caroline Vlaeminck for assisting with the molecular work, Klaus Valentin for cDNA 

library services, Tom Degroote and Wim Gillis for IT support. We thank two anonymous referees for 

their constructive criticisms on a previous version of the manuscript. This research was funded by a 

BOF grant (Ghent University) to EC and FWO-Flanders funding to HV, FL and ODC. Phylogenetic 

analyses were carried out on the KERMIT computing cluster (Ghent University) and the 

Computational Biology Service Unit (Cornell University and Microsoft Corporation).

58 CHAPTER 3 

Additional files 

Additional file 1

Additional file 2 

GAIN AND LOSS OF ELONGATION FACTOR GENES 59

60 CHAPTER 3 

Additional file 3


GAIN AND LOSS OF ELONGATION FACTOR GENES 61

62 CHAPTER 3 


Table S1. Algal strain information. 

Chlorophyta 

Ulvophyceae 

Ulvales 

number Culture collection 

Acrochaete repens E093db Barbara Rinkel (Natural History Museum, London) 

Bolbocoleon piliferum E344pc Barbara Rinkel (Natural History Museum, London) 

Ulva fenestrata EE2 and EE6 Field collection at Goese Sas (Netherlands) 

Ulva intestinalis EE3 Field collection at Goese Sas (Netherlands) 


Ignatius tetrasporus B 2012 UTEX culture collection of algae (University of Texas at Austin, USA) 


Trentepohlia aurea 483-1 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Bryopsidales 

Blastophysa rhizopus LB 1029 UTEX culture collection of algae (University of Texas at Austin, USA) 

Bryopsis sp. EE4 Field collection at Goese Sas (Netherlands) 

Codium sp HEC 15711 Field collection in Madeira 

Derbesia sp. 2773-1 Tatiana Klotchkova (Kongju National University, Korea) 

Ostreobium quekettii 6.99 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Dasycladales 

Acetabularia acetabulum LB 2694 UTEX culture collection of algae (University of Texas at Austin, USA) 

Bornetella sphaerica LB 2690 UTEX culture collection of algae (University of Texas at Austin, USA) 


Chlorophyceae 



Streptophyta 

Boodlea composita Bcomp4, BoTTd75 Jeanine Olsen and Wytze Stam (University of Groningen, Netherlands)* 

Cladophora coelothrix Ccoel2, C83.14 Jeanine Olsen and Wytze Stam (University of Groningen, Netherlands)* 

Dictyosphaeria cavernosa Dcav3, D.cavSJ25b Jeanine Olsen and Wytze Stam (University of Groningen, Netherlands)* 

Ernodesmis verticillata Erno4, EvVGa88 Jeanine Olsen and Wytze Stam (University of Groningen, Netherlands)* 

Phyllodictyon orientale Struv1, West 1631 John West (University of Melbourne, Australia) 

Valonia utricularis Vutric2, VU1546 Jeanine Olsen and Wytze Stam (University of Groningen, Netherlands)* 

Chlamydomonas reinhardtii CC1690 Chlamydomonas Center (Duke University, USA) 

Scenedesmus obliquus 1450 UTEX culture collection of algae (University of Texas at Austin, USA) 

Chlorella kessleri 211-11g Sammlung von Algenkulturen (University of Göttingen, Germany) 

Tetraselmis striata 41.85 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Ostreococcus tauri OTH95 Hervé Moreau (Observatoire Océanologique de Banyuls) 


Chlorokybus atmophyticus 48.80 Sammlung von Algenkulturen (University of Göttingen, Germany) 

Entransia fimbriata LB 2793 UTEX culture collection of algae (University of Texas at Austin, USA) 

Spirogyra sp. 169.80 Sammlung von Algenkulturen (University of Göttingen, Germany) 

*now maintained in the Phycology Research Group, Ghent University


Table S2. GenBank accession numbers or DOE Joint Genome Institute (JGI) transcript identifiers for nucleotide sequences of atpB, rbcL, SSU rDNA, EF-1α and EFL. Newly 

generated sequences are in boldface. GenBank accession numbers for protein sequences of eRF3 (Candida maltosa BAB12681, Homo sapiens NP_002085, Nicotiana 

tabacum AAA79032, Saccharomyces cerevisiae EDN60510), HBS1 (Arabidopsis thaliana NP_196625, Homo sapiens NP_006611, Saccharomyces cerevisiae NP_01301) and 

archaebacterial EF-1α (Sulfolobus sulfataricus CAA50033, Termoplasma acidophilum P19486) are given here between brackets. 

atpB rbcL SSU rDNA EF-1α EFL 

Chlorophyta 

Ulvophyceae 

Ulotrichales 

Halochlorococcum moorei AY198122 

Ulothrix zonata Z47999 

Ulvales 

Acrochaete repens FJ715715 FJ715684 FJ715704 

Bolbocoleon piliferum FJ715716 AY205330 FJ715705 

Pseudendoclonium akinetum NC 008114 NC 008114 DQ011230 

Ulva fenestrata AF499668 EF551331/ FJ715706 

Ulva intestinalis AY422552 AJ000040 EF551324/ FJ715707 


Ignatius tetrasporus AB110439 FJ715687 

Oltmannsiellopsidales 

Oltmanssiellopsis viridis NC 008099 NC 008099 D86495 

Bryopsidales 

Blastophysa rhizopus FJ715688 

Bryopsis spp FJ480417 (B. plumosa) FJ715718 FJ715685 (B. plumosa) FJ715689 

Caulerpa spp FJ480415 (C. prolifera) AF479703 (C. sertularioides) 

Codium spp FJ715686 (C. platylobium) FJ715690 

Derbesia sp. FJ715691 

Halimeda spp FJ480416 (H. discoidea) FJ715719 (H. incrassate) AY786526 (H. gracilis) 

Ostreobium quekettii FJ715720 FJ715692 

Dasycladales 

Acetabularia spp FJ480413 (A. dentata) FJ715714 (A. acetabulum) Z33461 (A. acetabulum) EF551321/ FJ715693 (A. acetabulum) 

Bornetella spp FJ480414 (B. nitida) FJ715717 (B. sphearica) Z33464 (B. nitida) FJ715694 (B. sphearica) 


Boergesenia forbesii AM498746 

Boodlea composita AF510157 FJ715695 

Cladophora albida Z35317 

Cladophora coelothrix Z35315 FJ715696 

Dictyosphaeria cavernosa AM498756 FJ715697


Ernodesmis verticillata AM498757 FJ715698 

Phyllodictyon spp AF510163 (P. pulcherrimum) FJ715699 

Siphonocladus tropicus AM498761 

Struvea plumosa AF510161 

Valonia utricularis Z35323 FJ715700 


Trentepohlia aurea FJ715722 AB110783 

Chlorophyceae 

Chlamydomonas incerta DQ122889 

Chlamydomonas reinhardtii NC 005353 NC 005353 M32703 DS496185/ FJ715708 

Chlorococcum echinozygotum EF113500 EF113430 U57698 EF551323 

Cylindrocapsa geminella EF119849 EF113434 AF387159 

Oedogonium cardiacum EF113523 EF113458 U83133 

Paulschulzia pseudovolvox AB014040 D86837 U83120 

Pleodorina spp AB214424 (P. starrii) D86834( P.indica) AB095178 (P. sp.) 

Scenedesmus obliquus NC008101 NC008101 X56103 SOL00000077/ FJ715709 

Sphaeroplea robusta EF113536 EF113472 U73472 

Stigeoclonium helveticum NC 008372 NC 008372 U83131 

Tetraspora sp. EF113540 EF113477 U83121 

Uronema belkae EF113544 EF113481 AF182821 


Chlorella spp NC 001865 (C. vulgaris) NC 001865 (C. vulgaris) X13688 (C. vulgaris) FJ715710 (C. kessleri) 

Closteriopsis acicularis EF113502 EF113433 Y17470 

Helicosporidium sp. AY729488 

Oocystis spp EF113524 (O. apiculata) EF113549 (O. apiculata) AF228686 (O. solitaria) 

Prototheca wickerhamii AJ245645 PWL00000297 

Trebouxia magna EF113541 AJ969630 Z21552 



Tetraselmis spp DQ173248 (T. suecica) DQ173247 (T. suecica) X70802 (T. striata) EF551330 (T. tetrathele) 

FJ715711 (T. striata) 

Mamiellales 

Micromonas pusilla AY955031 AJ010408 EF551325 

Ostreococcus lucimarinus XM 001420948 

Ostreococcus tauri NC 008289 NC 008289 AY329635 CR954213/ FJ715713 


Nephroselmis olivacea NC 00927 NC 00927 X74754 FJ715712 

Streptophyta 


Mesostigma viride NC 002186 NC 002186 AJ250109 DQ394295


Chlorokybales 

Chlorokybus atmophyticus DQ422812 DQ422812 M95612 FJ715701 


Entransia fimbriata AY823688 E88 slordig LB2793 FJ715702 

Klebsormidium flaccidum AF408801 E87 rbcL2 X75520 

Zygnematales 

Gonatozygon spp AF408796 (G. monotaenium) U71438 (G. monotaenium) X91346 (G. aculeatum) 

Spirogyra spp AF408797 S. maxima FJ715721 (S. sp.) EF551328/ FJ715703 

Staurastrum punctulatum NC 008116 NC 008116 AF115442 

Zygnema spp NC 008117 (Z. circumcarin) NC 008117 (Z. circumcarin) AJ853450 (Z. pseudogedeanum) 


Chaetosphaeridium globosum NC 004115 NC 004115 AF113506 

Coleochaete scutata AY082303 AY082313 X68825 

Charales 

Chara spp AF408782 (C. connivens) L13476 (C. connivens) U18493 (C. connivens) EF551322 (C. australis) 

Nitella flexilis AB110837 AB076056 U05261 


Anthoceratophyta (Hornworts) 

Anthoceros spp NC 004543 (A. formosa) NC 004543 (A. formosa) X80984 (A. agrestis) 


Marchantia polymorpha NC 001319 NC 001319 AB021684 


Physcomitrella patens AP005672 AB066207 XM 001753007 


Actinidia spp AJ235382 (A. chinensis) L01882 (A. chinensis) U42495 (A. sp.) AY946009 (A. deliciosa) 

Arabidopsis thaliana NC 000932 NC 000932 AC006837 AK230352 

Cichorium intybus L13652 AY378166 

Oryza sativa AC092750 AJ746297 X00755 AF030517 

Triticum aestivum NC 002762 NC 002762 AY049040 M90077 

Vicia faba AJ222579 




Cyanophora paradoxa NC 001675 NC 001675 AY823716 AF092951 

Rhodophyta 

Bangiophyceae 

Porphyra yezoensis AP006715 D79976 U08844 

Porphyridium aerugineum AJ421145 


Cyanidioschyzon merolae AB002583 AB158485 AB095182


Cyanidium caldarium X66698 AB091232 

Florideophyceae 

Chondrus crispus CCU02984 Z14140 CO652990 - CO652099- CO652788 

Gracilaria spp AY673996 (G. tenuistipitata) AY673996 (G. tenuistipitata) L26210 (G. verrucosa) DV963090 - DV963156 - DV964877 (G. 

changii) 

Chromalveolates 

Apicomplexans 

Plasmodium falciparum X60488 

Ciliates 

Tetrahymena pyriformis D11083 

Tetrahymena thermophila XM 001032213 

Cryptophytes 

Goniomonas amphinema AB332031 

Guillardia theta AM183813 

Rhodomonas salina DQ659244 

diatoms 

Ditylum brightwellii AB368772 

Phaeodactylum tricornutum 18475 (JGI, v2.0) 

Skeletonema costatum AB368773 

Thalassionema nitzschioides AB368774 

Thalassiosira pseudonana 3858 (JGI, v3.0) 41829 (JGI, v3.0) 

Dinoflagellates 

Heterocapsa triquetra AY729485 

Karlodinium micrum EF134135 

Oxyrrhis marina DQ659243 

Haptophytes 

Emiliania huxleyi CV068986 

Isochrysis galbana AY729486 

Pavlova lutheri AY729487 

Heterokonts 

Oomycetes 

Phytophthora infestans AJ249839 

Opisthokonts 

Fungi 

Allomyces macrogynus EC637105 

Blastocladiella emersonii EF064246 

Mucor racemosus MRATEF1A 

Saccharomyces martiniae AF402021 

Metazoa (Animals) 

Homo sapiens NM 001958


choanoflagellates 

Monosiga brevicollis XM_001745603 

Sphaeroforma arctica DQ403164 

Excavates 

Diplomonads 

Giardia intestinalis D14342 

Giardia lamblia XM 76292 

Euglenids 

Euglena gracilis X16890 

Parabasalids 

Trichomonas tenax D78479 

Trichomonas vaginalis XM 001325448 

Kinoplastids 

Leishmania braziliensis XM 001563727 

Amoebozoa 

Entamoeba histolytica ENHEF1ALPH 

Rhizaria 

Cercozoa 

Bigelowiella natans AY729489 

Foraminifera 

Planoglabratrella opecularis AB334123 

Centrohelids 

Raphidiophrys contractilis AB332032

68 CHAPTER 3 


Nexus file of EF-1α, EFL, GBlocks stripped EFL and SSU-rbcL-atpB alignments 


SSU rDNA alignment and partitioning 

The SSU rDNA sequences were aligned on the basis of their rRNA secondary structure information 

using the following procedure. The SSU rDNA sequences of several green plant representatives 

incorporated in the European Ribosomal RNA Database (Wuyts et al. 2004, 

http://www.psb.ugent.be/rRNA/) were used as an initial model for building the alignment. The 

alignment editor DCSE v2.6 (De Rijk and De Wachter 1993) was used to annotate and check the 

secondary structure, and manually align the sequences. For the newly generated sequences, the 

alignment of the highly variable helices 43 and 49 (see De Rijk et al. 1999 for secondary structure 

nomenclature of the SSU gene) was refined and aided by folding the RNA sequences using mfold 

v.3.2 with default temperature and RNA parameters (Zuker 2003, http://mfold.bioinfo.rpi.edu/). 

Finally, we used the Xstem software (Telford et al. 2005) to extract the RNA secondary structure 

information from DCSE to a nexus format with stem/loop partitions. 

Model selection procedure 

Selection of a suitable partitioning strategy and suitable models for the partitions followed a threestep 

procedure and uses the Akaike Information Criterion (AIC) as a selection criterion. The guide 

tree used during the entire procedure was obtained by MP analysis of the concatenated data using 

PAUP* 4.0b10 (Swofford 2002). All subsequent likelihood optimizations and AIC calculations were 

carried out with Treefinder (Jobb et al. 2004). The first step consisted of optimizing the likelihood for 

nine potential partitioning strategies, assuming a HKY+G8 model for each partition. The three 

partitioning strategies with the lowest AIC scores (i.e., those providing the best fit to the data) were 

retained for further evaluation. The second step involved model selection for individual partitions. 

The likelihood of each partition present in the three retained partitioning strategies was optimized 

for three variants of the general time reversible model (F81, HKY and GTR), with and without 

inclusion of a discrete gamma distribution (eight categories) to model among-site rate heterogeneity. 

Because not all genes were sampled for all taxa, the guide tree was pruned to the taxa present in the 

partition in question before each optimization. The partition-specific models obtaining the lowest AIC 

score were passed on to the third step, which consisted of re-evaluation of the three partitioning 

strategies retained from the first step using the models selected for these partitions in the second 

step. The partitioning strategy + model combination that received the lowest AIC score in the third 

step was used in the phylogenetic analyses documented in the main text.

4 

Complex phylogenetic distribution of a non-canonical genetic code 

in green algae 

Ellen Cocquyt 1 , Gillian H. Gile 2 , Frederik Leliaert 1 , Heroen Verbruggen 1 , Patrick Keeling 2 and Olivier De 

Clerck 1 

1 Phycology Research Group and Center for Molecular Phylogenetics and Evolution, Ghent University, 


2 Canadian Institute for Advanced Research, Department of Botany, University of British Columbia, 

Vancouver, V6 T 1Z4 Canada 

Abstract 

A non-canonical code, in which TAG and TAA have been reassigned from stop codons to glutamine, 

has previously been reported for the green algal orders Dasycladales. This study demonstrates that 

the Dasycladales share an identical alternative genetic code with the related clade Trentepohliales 

and the genus Blastophysa, but not with its sister clade the Bryopsidales. A single transition to the 

non-canonical code followed by a reversal to the canonical code in the Bryopsidales is highly 

improbable due to the profound genetic changes that coincide with codon reassignment. Multiple 

independent gains of the non-canonical code are not very plausible because a single type of noncanonical 

code evolved in some closely related ulvophyte lineages. Instead we favor a stepwise 

acquisition model, congruent with the ambiguous intermediate theory, whereby the alternative 

codes observed in these green algal orders share a single origin. The transition from a canonical to a 

non-canonical code has been completed only in the Trentepohliales, Dasycladales, Cladophorales and 

Blastophysa. This transition process, however, has not been completed in the Bryopsidales and 

hence no codon reassignment has taken place. 

Keywords 

non-canonical genetic code, glutamine, Chlorophyta, Ulvophyceae, stop codon, genome evolution

70 CHAPTER 4 

Introduction 

The genetic code, which translates nucleotide triplets (codons) into amino acids, is one of the most 

highly conserved features in living organisms. The genetic code is universal in nearly all genetic 

systems, including viruses, bacteria, archaebacteria, eukaryotic nuclei and organelles. However, a 

small number of eubacterial, eukaryotic nuclear, plastid, and mitochondrial genomes have evolved 

slight variations on this standard or canonical genetic code (Jukes and Osawa 1993, Knight et al. 

2001). Various models have been proposed to explain the evolutionary changes of the genetic code, 

including codon capture and ambiguous intermediate models (reviewed in Knight et al. 2001, Santos 

et al. 2004). Only five lineages of eukaryotes are known to have evolved non-canonical nuclear 

genetic codes, including ciliates, hexamitid diplomonads, fungi in the genus Candida and many 

Ascomycetes, polymastigid oxymonads and dasycladalean green algae (reviewed in Knight et al. 

2001, Keeling and Leander 2003, de Koning et al. 2008). 

By far the most common change to the genetic code in nuclear genomes is the reassignment of stop 

codons TAG and TAA to glutamine. This has happened independently in at least four eukaryotic 

lineages, including hexamitid diplomonads (Keeling and Doolittle 1996, Kolisko et al. 2008), some 

ciliates (Horowitz and Gorovsky 1985, Hanyu et al. 1986), polymastigid oxymonads (Keeling and 

Leander 2003, de Koning et al. 2008) and dasycladalean green algae (Schneider et al. 1989, Schneider 

and de Groot 1991). Interestingly, this particular non-canonical code has never evolved in 

prokaryotes or organelles. This bias towards reassignment of the stop codons TAG and TAA to 

encode glutamine in eukaryotes has been attributed to differences in the translation termination 

apparatus, tRNAs and tRNA synthetases, and mutation frequencies (Knight et al. 2001, Keeling and 

Leander 2003). Knowledge about the frequency and distribution of non-canonical codes across the 

branches of the tree of life will enable a better understanding of the evolution of genetic codes 

(Keeling and Leander 2003). 

The aim of the present study is to extend our knowledge of the phylogenetic distribution of the noncanonical 

genetic code in green algae, with emphasis on the ulvophycean relatives of the 

Dasycladales. Our approach consists of screening green algal nuclear genes for the presence of noncanonical 

glutamine codons and interpreting the evolution of the genetic code and glutamine codon 

usage in a phylogenetic framework. 

Methods 

Polymerase Chain Reaction and sequencing 

Total RNA was extracted from 43 taxa representing the major lineages of the Viridiplantae as 

described previously (Cocquyt et al. submitted). Portions of seven nuclear genes (actin, GPI, GapA, 

OEE1, 40S ribosomal protein S9 and 60S ribosomal proteins L3 and L17) were amplified, cloned when 

necessary and sequenced as described in Cocquyt et al. (submitted). A histone gene was amplified 

using the same PCR conditions with an annealing temperature of 55°C. The primers were based on a 

Cladophora coelothrix cDNA sequence aligned with GenBank sequences from green algae and land 

plants (His-F: 5’-ATG GCI CGT ACI AAG CAR AC-3’ and His-R: 5’-GGC ATG ATG GTS ACS CGC TT-3’). In

NON-CANONICAL GENENTIC CODE 71 

addition, total RNA was extracted from Ignatius tetrasporus and the bryopsidalean species Caulerpa 

cf. racemosa as described previously (Gile et al. 2009). Portions of actin, -tubulin, and HSP90 genes 

including the stop codon were amplified from these taxa by 3’ RACE using the First Choice RLM-RACE 

kit (Ambion) using nested degenerate primers (actin-F: 5’-TAC GAA GGA TAC GCA CTN CCN-3’ C and 

actin-R: 5’- GAG ATC GTG CGN GAY ATH AAR GA-3’; β-tubulin-F: 5’-GAT AAC GAG GCT CTN TAY GAY 

ATH TG-3’ and β-tubulin-R: 5’-CCT TTC CGA CGG CTN CAY TTY TT-3’; HSP90-F: 5’-ATG GTC GAT CCN 

ATH GAY GAR TA-3’ and 5’-GCT AAG ATG GAG MGN ATH ATG AA-3’). 

Genetic codes 

The presence of a non-canonical code in green algal taxa was detected by screening alignments of 

nuclear genes for apparent stop codons at positions coding for glutamine in other green plant taxa 

and by the presence of only TGA as a functional stop codon at the predicted 3’ end of genes. The 

presence of the standard code is confirmed by the presence of only canonical glutamine codons and 

the use of all three stop codons at the predicted end of genes. 

Molecular phylogenetics 

We constructed a reference phylogeny of the Viridiplantae based on the analysis of multiple genes as 

described in Cocquyt et al. (submitted) to study the phylogenetic occurrence of the standard and 

non-canonical genetic code. The phylogenetic analysis was carried out on an alignment consisting of 

the nuclear genes mentioned above, together with SSU nrDNA and the plastid genes rbcL and atpB. 

Histone genes were excluded from the analysis because they are known to be duplicated across 

genomes (Nei and Rooney 2005, Wahlberg and Wheat 2008). Phylogenetic analyses were carried out 

with model-based techniques (ML and BI), paying careful attention to the selection of suitable 

partitioning strategies and models of sequence evolution. The model selection procedure proposed 8 

partitions: SSU nrDNA was partitioned into loops and stems (2 partitions) and nuclear and plastid 

genes were partitioned into codon positions (2 3 partitions). GTR+ 8 was the preferred model for 

all partitions. Noise-reduction techniques were applied to counteract the erosion of ancient 

phylogenetic signal caused by fast-evolving sites. The phylogenetic tree presented here is based on 

the 75% slowest-evolving sites (Cocquyt et al. submitted). 

Evolution of glutamine codon usage 

The evolution of glutamine codon usage was estimated using ancestral state estimation techniques. 

The frequency of the two canonical and two non-canonical glutamine codons was calculated for each 

species in the phylogenetic tree. Codon frequencies were mapped along the reference tree using the 

ace function of the APE package (Paradis et al. 2004). This function estimates ancestral character 

states by maximum likelihood optimization (Schluter et al. 1997). The branch lengths were based on 

ML estimates because we consider them to be a more relevant approximation of the amount of 

codon usage evolution that can be expected to take place than absolute time (cf. Cocquyt et al.

72 CHAPTER 4 

2009). The output from APE was mapped onto the reference tree with TreeGradients v1.03 

(Verbruggen 2009) to plot ancestral states of continuous characters on a phylogenetic tree as colors 

along a color gradient. 

Topological hypothesis testing 

Approximately unbiased tests (AU test, Shimodaira 2002) were used to test an alternative 

relationship between ulvophycean orders as suggested by the distribution of the canonical genetic 

code (see results). Site-specific likelihoods were calculated by maximum likelihood optimization in 

Treefinder using the same model specifications as for phylogenetic inference (Cocquyt et al. 

submitted). AU tests were performed with CONSEL V0.1i (Shimodaira and Hasegawa 2001). 

Results 

The dasycladaceans Acetabularia and Batophora have been shown previously to use the noncanonical 

code (Schneider et al. 1989, Schneider and de Groot 1991). Looking more broadly, we find 

that the non-canonical glutamine codons appear in coding sequences of additional members of the 

Dasycladales (genus Bornetella), as well as Cladophorales (genera Boergesenia, Boodlea, Cladophora, 

Dictyosphaeria, Phyllodictyon, Siphonocladus, Valonia), Trentepohliales (genus Trentepohlia) and the 

genus Blastophysa, which is currently not assigned to an order. Most non-canonical codons were 

found at highly conserved positions that encoded glutamine codons in the rest of the green plant 

lineage. Only TGA is used as stop codon in ribosomal protein 40S S9 and OEE1 of Cladophora. 

Moreover, other sequences from both Cladophora and Acetabularia deposited in GenBank only have 

TGA as stop codon: GapA gene (DQ270261) of Cladophora, and EF-1α (EF551321) and PsbS genes 

(BK006014) of Acetabularia. The presence of the standard code is confirmed for the genus Ignatius 

and the order Bryopsidales (genera Caulerpa, Bryopsis) by the presence of only canonical glutamine 

codons and the use of all three stop codons at the ends of various genes. The Ignatius actin gene has 

TAG as stop codon, while the β tubulin and HPS90 genes have TAA as stop codons. β tubulin and 

HPS90 genes of Caulerpa have TAA as stop codons. Additional Bryopsis sequences deposited in 

GenBank confirmed the presence of all three stop codons in the Bryopsidales: TAA in ribonuclease 

Bm2 gene (AB164318), TAG in lectin precursor and oxygen evolving protein of photosystem II genes 

(EU410470 and AB293980), and TGA in the bryohealin precursor gene (EU769118). 

The occurrence of the standard and non-canonical code are plotted onto the reference phylogeny in 

Fig. 1. The Streptophyta, prasinophytes, Trebouxiophyceae and Chlorophyceae possess the standard 

genetic code. Within the class Ulvophyceae, the standard code is found in the orders Ulvales, 

Ulotrichales, Bryopsidales and the genus Ignatius whereas the orders Dasycladales, Siphonocladales, 

Trentepohliales and the genus Blastophysa have a non-canonical code. However, the taxa with a 

non-canonical code do not form a monophyletic group (Fig. 1).

Figure 1. The occurrence of a non-canonical genetic code in green algae is indicated with gray squares. Taxa 

with a non-canonical genetic code are not monophyletic. Three alternative scenarios can explain its 

distribution on the tree: 

(1) a single origin of the non-canonical code along the branch leading to the orders Trentepohliales, 

Dasycladales, Bryopsidales, Cladophorales and the genus Blastophysa and a reversal to the universal 

code in the Bryopsidales (indicated with black arrow and cross) 

(2) three independent gains of the non-canonical code in the Trentepohliales, the Dasycladales and the 

Cladophorales + Blastophysa (indicated with gray arrows) 

(3) a single initiation of the formation of the non-canonical code along the branch leading to the orders 

Trentepohliales, Dasycladales, Bryopsidales, Cladophorales and the genus Blastophysa that has been 

completed in all lineages except the Bryopsidales (black arrow combined with gray arrows) 

The reference phylogeny of the green plant lineage was obtained by maximum likelihood inference of the 25% 

site stripped dataset containing 7 nuclear genes, SSU nrDNA and plastid genes rbcL and atpB (Cocquyt et al. 

submitted). Numbers at nodes indicate ML bootstrap values (top) and posterior probabilities (bottom); values 

below respectively 50 and 0.9 are not shown.

74 CHAPTER 4 

If both the phylogeny and the distribution of the genetic codes shown in Fig. 1 are correct, then more 

than one gain and/or loss event of the non-canonical code must be postulated. To examine the 

validity of the phylogeny, we performed topology tests to evaluate whether the data rejected a 

topology in which all taxa with the non-canonical code formed a monophyletic group. Specifically, we 

found that the topology where Bryopsidales is sister to all Ulvophyceae with a non-canonical code is 

rejected with high significance ( lnL = 27.9; p < 0.0001). 

The estimated evolution of glutamine codon usage frequencies is shown in Fig. 2. The canonical 

codon CAG is most commonly used, followed by the canonical codon CAA. Among the non-canonical 

codons, TAG is used more commonly then TAA. This bias is likely a product of the overall bias of 

these species and/or genes for GC residues, which favors a G in the third position. This observation is 

also congruent with canonical glutamine codons CAG and CAA which mutate to non-canonical 

glutamine codons TAG and TAA by a single transition (C → T) at the first codon position (Keeling and 

Leander 2003). 

Discussion 

Our results reveal the distribution of a non-canonical genetic code in the Ulvophyceae, where 

glutamine is encoded by both canonical CAG and CAA codons as well as non-canonical TAG and TAA 

codons. Unexpectedly, we find the taxa with this non-canonical code do not form a monophyletic 

group according to the seemingly robust phylogeny of the organisms in which it is found (Fig. 1). If 

the inferred phylogeny is indeed correct, three alternative scenarios can explain the distribution of 

the code on that tree: (1) a single origin of the non-canonical code along the branch leading to the 

orders Trentepohliales, Dasycladales, Bryopsidales, Cladophorales and the genus Blastophysa and a 

reversal to the universal code in the Bryopsidales (Fig. 1: indicated with black arrow and cross); (2) 

three independent gains of the non-canonical code in the Trentepohliales, the Dasycladales and the 

Cladophorales + Blastophysa (Fig. 1: indicated with gray arrows), and (3) a single initiation of a 

stepwise formation of the non-canonical code along the branch leading to the orders 

Trentepohliales, Dasycladales, Bryopsidales, Cladophorales and the genus Blastophysa, but where 

the process has gone to completion in all lineages except the Bryopsidales (Fig. 1: black arrow 

combined with gray arrows). Alternatively, because changes in the genetic code are so rare, the 

possibility that the reference phylogeny is wrong should not be passed over too easily. More 

specifically, if one assumes that the current position of the Bryopsidales is wrong and that in reality 

this group is sister to all the taxa with a non-canonical code, only a single transition from the genetic 

code would have to be invoked. In what follows, we will discuss each of these possibilities in more 

detail and report on some cytological correlates of the non-canonical genetic code.


Figure 2. The estimated ancestral frequencies of glutamine codon usage. Canonical codon CAG is most 

commonly used, followed by CAA, the second canonical codon. Among the non-canonical codons, TAG is used 

more commonly then TAA. The estimate ancestral frequencies of non-canonical codon usage along the nodes 

of interest are indicated with an arrow.

76 CHAPTER 4 

Phylogenetic uncertainty 

Our phylogenetic tree is based on the most comprehensive dataset currently available for the 

Chlorophyta. It is inferred from a concatenated dataset including seven nuclear genes, SSU nrDNA 

and two plastid genes using model-based techniques with carefully chosen partitioning strategies 

and models of sequence evolution and application of a site removal approach to optimize signal for 

the relevant level of divergence (Cocquyt et al. submitted). Our tree shows a sister relationship 

between Dasycladales and Bryopsidales with moderately high statistical support (BV 87 and PP 1.00). 

This relationship is concordant with a recently published 74-taxon phylogeny of the green lineage 

based on SSU nrDNA and two plastid genes (Cocquyt et al. 2009). Both phylogenies show major 

improvements in taxon and gene sampling within the Ulvophyceae compared to previously published 

phylogenies, which were either based on a single marker, did not include the Bryopsidales, or could 

not resolve the relationships among the Bryopsidales, Dasycladales and Cladophorales (Zechman et 

al. 1990, Lopez-Bautista and Chapman 2003, Watanabe and Nakayama 2007). Based on the dataset 

used to infer our reference tree, the alternative topology in which the Bryopsidales are sister to all 

taxa with a non-canonical code is significantly less likely than the ML tree as shown by AU tests. As a 

consequence, unless the phylogenetic signal contained in the molecular data is fundamentally wrong, 

the ulvophycean taxa with the non-canonical code form a paraphyletic group and one of the more 

complex evolutionary scenarios for the gain of the non-canonical code has to be invoked. 

Gain–reversal hypothesis 

A reversal from the non-canonical to the standard genetic code is unlikely for several reasons. First, 

following a transition to the non-canonical code, TAG and TAA codons would be present in many 

coding sequences. In order to revert to the standard or canonical code, these codons would all have 

to revert to canonical codons or they would terminate translation, with obvious detrimental effects. 

Our ancestral state estimates indicate a non-negligible usage frequency of both non-canonical 

codons along the ancestral nodes of interest (Fig. 2 C,D, indicated with arrows). These results must 

be considered with caution because of the intrinsic limitations of ancestral state estimation (Martins 

1999) and the fact that the non-independence of the evolution of the genetic code and glutamine 

codon usage cannot be captured by the Brownian motion model. Nevertheless, it seems likely that 

glutamine-encoding TAA and TAG codons in a genome using the non-canonical code would be more 

common than TAA and TAG were in the ancestral genome where the standard code was still in use, 

because the frequency of stop codons is at most once per gene whereas even disfavored glutamine 

codons can appear several times in a single gene. The major argument against a reversal from the 

non-canonical to the canonical code is that the original change would have been accompanied by 

changes in the translation apparatus, e.g. alterations in tRNAs, aminoacyl-tRNA-synthethases, 

eukaryotic release factor (eRF1) and also potentially a nonsense-mediated mRNA decay (NMD) 

mechanism, all of which would have to reverse in the Bryopsidales in order to support this scenario 

(Knight et al. 2001, Keeling and Leander 2003). Taken together, the reversal of a non-canonical code 

to the standard code appears highly unlikely.

Multiple independent gains 


Several independent acquisitions of non-canonical codes have been reported for ciliates 

(Tourancheau et al. 1995, Knight et al. 2001, Lozupone et al. 2001). Indeed, stop codon 

reassignments are surprisingly frequent in this group of organisms: the same non-canonical code 

than the one present in the Ulvophyceae has evolved several times, another non-canonical code in 

which TGA codes for tryptophan evolved twice, and a third non-canonical code that translates TGA to 

cysteine evolved once. In the present study, the distribution of the non-canonical code in the 

phylogenetic tree would require three gains: in the Trentepohliales, the Dasycladales and the 

Cladophorales + Blastophysa. Contrary to the situation in ciliates, however, Ulvophyceae only 

evolved a single type of non-canonical code and they did so in closely related lineages. The 

combination of these two arguments weakens the plausibility of multiple independent gains 

although the possibility cannot be excluded altogether. 

Stepwise acquisition model 

Several studies have revealed that stop codon reassignment is a gradual process requiring changes to 

tRNA and eRF1 genes (reviewed in Knight et al. 2001, Santos et al. 2004, Sengupta and Higgs 2005). 

In several eukaryotes, the two glutamine tRNAs, recognizing CAG and CAA, are also able to read TAG 

and TAA codons by a G U wobble base pairing at the third anticodon position (Beier and Grimm 

2001). In normal circumstances (canonical code), eRF1 out-competes the glutamine tRNAs when it 

comes to binding TAG and TAA. Mutations in glutamine tRNA copies that allow them to bind TAG and 

TAA may increase the capacity to translate TAG and TAA to glutamine. This leads to an intermediate 

step in which both eRF1 and the mutated tRNAs can easily bind to TAG and TAA. The ciliate 

Tetrahymena thermophila has three major glutamine tRNAs, one that recognizes the normal 

glutamine codons CAG and CAA, and two supplementary tRNAs that recognize the non-canonical 

codons TAG and TAA. These two supplementary tRNAs were shown to have evolved from the normal 

glutamine tRNA (Hanyu et al. 1986). A similar situation likely exists in diplomonads (Keeling and 

Doolittle 1996). In order to alter the genetic code, mutations are required not only in glutamine 

tRNAs but also in eukaryotic release factor 1 (eRF1) so it no longer recognises TAG and TAA codons. 

In ciliates it has been shown that eRF1 sequences are highly divergent in domain 1 between species 

with a canonical and non-canonical code, which could indicate that eRF1 can no longer recognize 

TAG and TAA codons in the species with a non-canonical code (Lozupone et al. 2001, Inagaki et al. 

2002). An additional mechanism that constrains the genetic code and therefore represents yet 

another potential step in the process of codon assignment is nonsense-mediated mRNA decay (NMD) 

(Baker and Parker 2004, Maquat 2004). NMD reduces errors in gene expression by eliminating 

mRNAs that encode for an incomplete polypeptide due to the presence of stop codons. In the case of 

a non-canonical code where stop codons TAG and TAA encode glutamine, as observed here, this 

NMD mechanism would have to be altered also in order to prevent degradation of mRNAs containing 

TAG and TAA codons. 

A stepwise acquisition model can reconcile the opposed and problematic hypotheses of multiple 

gains versus a single gain with subsequent loss. For example, the ambiguous intermediate theory 

(Schultz and Yarus 1994, Santos et al. 2004) would explain the distribution of the non-canonical code

78 CHAPTER 4 

in Ulvophyceae as follows: mutations in the anticodons of canonical glutamine tRNAs occurred once 

along the branch leading to the orders Trentepohliales, Dasycladales, Bryopsidales, Cladophorales 

and the genus Blastophysa (Fig. 1, black arrow). The presence of these mutated tRNAs allowed TAG 

and TAA codons to be translated to glutamine instead of terminating translation. At this step, the 

mutated tRNAs compete with eRF1 for the TAA and TAG codons. To complete the transition to the 

non-canonical code, a subsequent mutation of eRF1 that prevents binding of eRF1 with TAG and TAA 

is required. If one assumes that this step occurred three times independently in the Trentepohliales, 

Dasycladales and Cladophorales + Blastophysa (Fig. 1, gray arrows), whereas the mutated tRNAs 

decreased in importance or went extinct through selection or drift along the branch leading to the 

Bryopsidales, the distribution of the non-canonical code in the Ulvophyceae would be explained. 

However, a detailed comparison of eukaryotic release factors (eRF1) and glutamine tRNAs in the 

respective clades of the Ulvophyceae is needed to test this evolutionary scenario. 

Cytological correlates of non-canonical code 

In the ciliates, the multiple appearances of alternative codes have been attributed to their nuclear 

characteristics. Ciliates have two nuclei: a small, diploid micronucleus which is not expressed and is 

the germ line for DNA exchanges during the sexual process, and a large, polyploid macronucleus 

which is actively transcribed and ensures vegetative cell growth, but is not passed on to progeny 

after sexual conjugation and is replaced by a newly formed macronucleus after a number of rounds 

of mitotic division. There is therefore a time lag between the occurrence of mutations in the 

micronucleus and the expression of these mutations in the macronucleus, and this has been 

postulated to be a contributing factor to why ciliates have evolved alternative genetic codes more 

frequently (Cohen and Adoutte 1995). In this context it is worth mentioning that hexamitid 

diplomonads, for which a single origin of a non-canonical code together with the uninucleate 

enteromonads has been shown, have two similar nuclei per cell (Keeling and Leander 2003, Kolisko 

et al. 2008) and that several ulvophycean groups are characterized by multinucleate cells namely the 

Dasycladales, Bryopsidales, Cladophorales and Blastophysa. The Cladophorales and Blastophysa are 

branched filamentous seaweeds consisting of multinucleate cells with a few to thousands of nuclei 

arranged in non-motile cytoplasmic domains (siphonocladous organization). Members of the 

Bryopsidales and Dasycladales have a siphonous organization: they consist of a single, giant tubular 

cell with numerous nuclei and complex patterns of cytoplasmic flow. Acetabularia (Dasycladales) 

forms an exception in the sense that the number of nuclei has been reduced to one in this genus: its 

members possess a single, giant nucleus situated in the rhizoid. It thus appears that the ancestors of 

these taxa in which changes of the genetic code have occurred were very likely all multinucleate. 

Despite the fact that some eukaryotes with a non-canonical code do not feature multinucleate cells 

and that there are plenty of examples of groups with multinucleate cells that have not evolved 

alternative codes, this observation suggests that a multinucleate cytology can facilitate codon 

reassignments. To our knowledge, no explanatory mechanisms have been described.



We thank Caroline Vlaeminck for assisting with the molecular work and Wim Gillis for IT support. 

Funding was provided by the Special Research Fund (Ghent University, DOZA-01107605), a grant 

from the Natural Sciences and Engineering Research Council of Canada (227301), an NSERC 

postgraduate doctoral scholarship to GHG and Research Foundation Flanders postdoctoral 

fellowships to HV, FL and ODC. PJK is a Fellow of the Canadian Institute for Advanced Research and a 

Senior Investigator of the Michael Smith Foundation for Health Research. Phylogenetic analyses were 

largely carried out on the KERMIT computing cluster (Ghent University).

5 

Codon usage bias and GC content in green algae 

Ellen Cocquyt, Heroen Verbruggen and Olivier De Clerck 

1 Phycology Research Group and Center for Molecular Phylogenetics and Evolution, Ghent University, 


Abstract 

Large differences in synonymous codon usage bias and GC content have been observed within and 

among genomes and a plethora of hypotheses have been put forward to explain them. In this study, 

we characterize patterns of codon usage bias and GC content in eight nuclear housekeeping genes 

from 43 green algae and land plants. We analyze the evolution of these biases in a phylogenetic 

framework. We observe stronger codon usage bias in the ancestral streptophytes Mesostigma and 

Chlorokybus than in the remainder of the Streptophyta. Within the Chlorophyta, the prasinophytes, 

Trebouxiophyceae and Chlorophyceae have markedly stronger codon usage bias than the 

Ulvophyceae. One exception is Ignatius, which is a member of the Ulvophyceae yet has a markedly 

stronger codon usage bias than other members of this class. GC content patterns show congruent 

trends, species with strong codon usage bias having a high GC content. We interpret these results 

along with the biology of the organisms in the framework of two models: the mutation-selection-drift 

model and the co-evolutionary model of genome composition and resource allocation. It is 

remarkable that unicellular organisms and colony-forming species have much more pronounced GC 

and codon usage biases as compared to multicellular and macroscopic species. This may follow from 

unicells having large population sizes, which leads to more codon usage bias due to stronger 

selection as compared to species with smaller population sizes where drift can more rapidly fix 

mutation. We also interpret the correlation of the observed biases with growth rates, generation 

time and rates of molecular evolution of the organisms under study. 

Keywords 

codon usage bias, GC content, unicellular, multicellular, effective population size, green plants

82 CHAPTER 5 

Introduction 

The genetic code, which allows translating codons (nucleotide triplets) to amino acids, is redundant 

because all amino acids except methionine and tryptophan are encoded by two to six synonymous 

codons. Even though the fact that synonymous codons translate to the same amino acid would lead 

to the expectation that synonymous codons are used in equal proportions, DNA sequence data from 

a wide range of organisms have shown that this is not the case: some synonymous codons are clearly 

preferred over others, and different organisms often have different preferences. This phenomenon is 

called codon usage bias and has been associated with a wide range of biological factors (Salim and 

Cavalcanti 2008). Within a single organism, codon usage bias is more pronounced in highly expressed 

and long genes, and near the 3’end of genes. Codon usage bias is stronger when the population of 

tRNAs is less diverse and correlates with GC content. Differences in codon usage bias between 

organisms appear to correlate with their population size; species with large effective population sizes 

having more biased codon usage than species with smaller population sizes (Cutter et al. 2006, 

Ingvarsson 2008, Subramanian 2008). 

Conceptual models of various levels of complexity have been proposed to explain the presence and 

degree of codon usage bias in and between organisms. The mutation-selection-drift balance model 

assumes that codon usage ensues from the delicate balance between mutation, selection and 

genetic drift in populations (Bulmer 1991, Hershberg and Petrov 2008). Whereas natural selection 

can be expected to favor the organism's preferred (major) codons over minor codons, mutational 

pressure and genetic drift allow minor codons to persist in the population (Fig 1. A external factors 

influencing mutation bias and selection, reviewed in Hershberg and Petrov 2008). A more complex 

model, which we will refer to as the co-evolutionary model, assumes that genome biases (e.g. GC and 

codon usage bias) result from the co-evolution between genome composition and resource 

concentration, both of which act on genome bias via mutation bias and selection (Fig 1. B internal 

mechanism influencing mutation bias and selection, Vetsigian and Goldenfeld 2009). More 

specifically, for a certain genome composition the nucleotide and tRNA concentrations are altered to 

increase speed and accuracy, which in turn leads to mutation and selection pressures favoring the 

preferred nucleotides and codons. The combination of all these factors leads to differences in 

genome biases between and within species. To date, most studies have concentrated on genome 

bias within a single species and relatively few studies examined genome biases between organisms. 

As far as we know, only one study examined codon usage bias between more distantly related 

nematode species which inhabit different ecological niches being either free-living or parasitic 

(Cutter et al. 2006). 

The green algae form a promising model system to study the evolution of codon usage bias and GC 

content over longer time-scales for several reasons. First, some ulvophycean algae were shown to 

have a non-canonical genetic code in which the stop codons TAG and TAA have been reassigned to 

glutamine (Cocquyt et al. in prep.). This non-canonical code involves changes in glutamine codon 

usage, the translation termination apparatus, tRNAs and tRNA synthetases, and mutations 

converting canonical glutamine codons (CAG and CAA) to non-canonical glutamine codons (TAG and 

TAA) by a single transition at the first codon position (C � T). Second, green algae and land plants 

occupy a wide variety of ecological niches, including marine, freshwater and terrestrial habitats. 

Finally, the green algae are characterized by markedly different rates of molecular evolution as

CODON USAGE BIAS AN GC CONTENT 83 

indicated by root to tip path length differences in phylogenetic trees, indicating differences in 

molecular evolution between different groups of green algae. Finally, green algae exhibit remarkable 

morphological and cytological diversity ranging from unicells and colonies, over multicellular 

filaments and foliose blades, to highly complex multicellular and multinucleate life forms. Two types 

of multinucleate organisms are present: (1) a siphonocladous organization with cells that contain a 

few to thousands of nuclei arranged in non-motile cytoplasmic domains (e.g. order Cladophorales 

and genus Blastophysa), and (2) a siphonous organization with a single, giant tubular cell that 

contains numerous nuclei and complex patterns of cytoplasmic flow (e.g. orders Bryopsidales and 

most Dasycladales). These morphological and cytological differences may correlate with differences 

in population size and as such have an impact on genome biases. 

Figure 1. Origin of genome biases such as GC or AT bias and codon usage bias. 

A. Genome composition is influenced by a delicate balance between mutation bias and selection for fast and 

accurate transcription and translation. Selection and mutation are in turn influenced by several external and 

internal factors. 

(1) species with large population sizes have more codon usage bias due to stronger selection compared to 

species with smaller population sizes where drift can more rapidly fix mutations. 

(2) In rapidly multiplying organisms, the time spent in translation is a limiting factor in cell division, increased 

selection leads to an optimal translation. 

(3) Selection for optimal speed and accuracy of replication and translation involves a co-evolution between 

genome composition and resource concentration, which in turn influence the direction of mutation bias. 

B. Selection and mutation are influenced by the co-evolution between genome composition and resource 

concentration, effective population size and growth rate, each causing a specific genome bias.

84 CHAPTER 5 

The goal of this study is to characterize patterns of codon usage bias and GC content in green algal 

nuclear genes and analyze their evolution in an explicit phylogenetic framework. Our approach 

consists of calculating codon usage bias and GC content in 43 representatives of green algae and land 

plants based on eight nuclear housekeeping genes. The evolutionary patterns of codon usage bias 

and GC content are subsequently approximated with ancestral state estimation techniques and 

mapped along the reference tree. 

Methods 

DNA amplification and sequencing 

Seven nuclear genes (actin, GPI, GapA, OEE1, 40S ribosomal protein S9 and 60S ribosomal proteins L3 

and L17) were amplified and sequenced for 43 taxa representing the major lineages of the 

Viridiplantae as described in Cocquyt et al. (submitted). A histone gene was amplified using the same 

PCR conditions with an annealing temperature of 55°C. The primers were based on a Cladophora 

coelothrix cDNA sequence aligned with GenBank sequences from green algae and land plants (His-F: 

5’-ATG GCI CGT ACI AAG CAR AC-3’ and His-R: 5’-GGC ATG ATG GTS ACS CGC TT-3’). 

Molecular phylogenetics 

We constructed a reference phylogeny of the Viridiplantae as described in Cocquyt et al. (submitted) 

to study patterns of codon usage bias and GC content. The phylogenetic analysis was carried out on 

an alignment consisting of the nuclear genes mentioned above, SSU nrDNA and the plastid genes 

rbcL and atpB. Histone genes were excluded from the analysis because they are known to be 

duplicated across genomes (Nei and Rooney 2005, Wahlberg and Wheat 2008). The analysis was 

based on the 75% slowest-evolving sites to improve signal for the reconstruction of ancient 

relationships (Cocquyt et al. submitted). 

Synonymous codon usage bias and GC content 

Synonymous codon usage order (SCUO) was used to measure synonymous codon usage bias in each 

species and was calculated with CodonO based on concatenated sequences of the eight nuclear 

genes mentioned above (Wan et al. 2006, Angellotti et al. 2007). SCUO represents a value between 0 

to 1, larger values indicating stronger codon usage bias. GC content was calculated with TreeFinder 

(Jobb et al. 2004) based on the same eight nuclear genes. The evolution of SCUO and GC content was 

approximated using ancestral state estimation techniques and mapped along the reference tree. 

Ancestral state estimation was carried out with the ace function of the APE package (Paradis et al. 

2004), using maximum likelihood optimization for continuous characters (Schluter et al. 1997). The 

output from APE was mapped onto the reference tree with TreeGradients v1.03 (Verbruggen 2009). 

This program plots the estimated ancestral character states on a phylogenetic tree as colors along a 

color gradient.

Results and discussion 


The estimated evolution of synonymous codon usage bias (SCUO) and GC content is shown in Fig. 2. 

Mesostigma, closely followed by the genera Chlorokybus and Entransia, have stronger codon usage 

bias than the remainder of the Streptophyta. Within the Chlorophyta, the prasinophytes, 

Trebouxiophyceae, Chlorophyceae have markedly stronger codon usage bias than the Ulvophyceae. 

One exception is Ignatius, which is a member of the Ulvophyceae yet has a markedly stronger codon 

usage bias than other members of this class. GC content patterns show congruent trends: within the 

Streptophyta, GC bias is highest for Mesostigma and Chlorokybus. Within the Chlorophyta, GC bias is 

much more pronounced in the prasinophytes, Trebouxiophyceae and Chlorophyceae as compared to 

the Ulvophyceae. Within the Ulvophyceae, the orders Trentepohliales, Dasycladales and Bryopsidales 

have the weakest GC bias. 

It follows from Fig. 2 that there have been only a few transitions in GC and codon usage bias during 

the evolution of the green algae, more specifically two independent reductions of the biases in the 

Streptophyta and Ulvophyceae. This apparent conservatism of genome biases over long evolutionary 

periods stands in strong contrast with the notion that variation of codon usage bias and GC content 

has been shown between closely related species and even within individual genomes (e.g. Duret and 

Mouchiroud 1999, Derelle et al. 2006, Ingvarsson 2008). In the light of these indications for fast 

evolution of genome biases, it seems unlikely that their conservatism over long periods of time in the 

green algae is due to slow evolution but rather that it is a consequence of the association of genome 

biases with biological traits that have evolved slowly. In the following paragraphs, we will highlight 

several correlations between the observed patterns of genome bias and organismal traits that are 

associated with them and potentially responsible for them. 

Body and population size 

As illustrated in Fig. 2, the genes under study have more pronounced GC and codon usage bias in 

unicellular organisms and colony-forming species as compared to multicellular and/or multinucleate 

species. As such, the trends towards multicellularity and macroscopic algal bodies in both the 

Streptophyta and Chlorophyta are associated with decreasing amounts of GC and codon usage bias. 

It has been documented that codon usage bias is stronger for species with large populations due to 

stronger selection compared to species with smaller population sizes where drift can more rapidly fix 

mutations (Cutter et al. 2006, Ingvarsson 2008). Because the selective advantages associated with 

the usage of alternative codons are subtle, selection can only offset the stochastic effects of genetic 

drift in species with large effective population sizes (Akashi 1995, Cutter et al. 2006, dos Reis and 

Wernisch 2009). Because body size and population size appear to be inversely related for aquatic 

organisms (Finlay 2002, Fenchel and Finlay 2004), it seems likely that the strong codon usage bias 

observed in unicellular and colonial organisms may be a result of their larger effective population 

sizes.

86 CHAPTER 5

Growth rate and generation time 


In rapidly multiplying organisms, the time spent in translation is among the time-limiting factors in 

cell division (Higgs and Ran 2008). The speed of translation can be improved through the use of 

preferred or optimal codons accompanied with a higher concentration of the corresponding tRNAs 

most likely mediated by tRNA gene duplications (Higgs and Ran 2008). In other words, stronger 

codon usage bias leads to more efficient translation and as such allows for higher growth rates. 

Under optimal environmental conditions, unicellular organisms have very high growth rates and 

short generation time. Since in small, unicellular organisms every division leads to a doubling of the 

number of individuals, large population sizes are quickly reached. On the other hand, multicellular 

organisms and macroscopic unicellular algae (e.g. siphonous ulvophytes) make a larger investment in 

structural components and have lower population sizes. For example, under optimal conditions the 

unicellular flagellate Chlamydomonas divides every 6.7 h (Badour 1981), whereas the siphonous 

ulvophyte Acetabularia needs 115 days to complete its life cycle (Berger and Liddle 2003). These 

characteristic growth rates and generation times can be compared to r/K strategies in ecology. rselected 

species are opportunists characterized by high fecundity, small body sizes, early onset of 

maturity, short generation times and the ability to disperse offspring widely. K-selected species live 

in more stable environments and are characterized by large body sizes, long life expectation, and the 

production of fewer offspring. Generally speaking, unicellular algae fit the r-selection theory, 

whereas macroscopic algae can be indicated as K-selected species. The strong codon usage bias 

observed in small unicellular algae can likely be explained by ecological traits such as fast growth 

rates, short generation time, small body size and large population size all being characteristic for rselected 

species. 

Rate of molecular evolution 

Some Ulvophyceae, i.e. Trentepohliales, Dasycladales, Bryopsidales, Cladophorales and Blastophysa, 

have markedly elevated rates of molecular evolution as indicated by long root to tip path lengths in 

phylogenetic trees (branch lengths). In other words, these taxa have accumulated mutations at a 

faster pace than the rest of the green algae, potentially suggesting higher levels of genetic drift in 

those algae. The amount of codon usage bias depends on the delicate balance between selection for 

optimal translation and mutations fixed by genetic drift. Selection for speed and accuracy of 

translation is expected to favor optimal codons over minor codons, leading to codon usage bias. On 

the other hand, when selection for optimal translation is less stringent, mutation events that are 

fixed through genetic drift will result in a uniform codon usage. The observed uniform codon usage in 

some Ulvophyceae is likely due to elevated rates of molecular evolution leading to the fixation of 

many mutations and pronounced genetic drift. Studies on various other groups of organisms (both 

eukaryotes and prokaryotes) have also revealed a negative correlation between codon usage bias 

and rate of synonymous nucleotide substitution (Powell and Moriyama 1997, Kawahara and Imanishi 

2007).

88 CHAPTER 5 

Codon usage bias and GC content 

In our data, GC content and codon usage bias show congruent trends, species with strong codon 

usage bias having a high GC content (Fig. 2). Similar correlations have been recovered by other 

workers (Knight et al. 2001, Chen et al. 2004). The causal direction of this correlation has mostly been 

interpreted from GC content to codon usage bias. More specifically, it has been suggested that base 

composition drives codon usage through mutational bias in favor of the common nucleotides (Knight 

et al. 2001). Knight et al. (2001) conclude that 71-87% of the observed differences in codon usage 

within and between genomes can be explained by GC content influencing codon usage, the 

remaining 29-23% being a consequence of selective pressures for optimal translation. 

The co-evolutionary model offers a more mechanistic perspective on the correlation between GC 

content and codon usage bias. More specifically, this model predicts that base composition and 

codon usage are linked through internal feedback loops. If one assumes that translation efficiency 

improves with increased usage of G and C at synonymous codon positions, a GC-rich genome will 

result if selection for translation efficiency is sufficiently strong. Although many authors have 

reported a positive relationship between GC content and codon usage bias, we are not aware of 

studies showing empirical evidence for intrinsically more efficient translation of codons that have G 

and C at synonymous positions. It is possible that the use of G and C improves codon–anticodon 

recognition because of the extra hydrogen bond that is formed between them, with positive effects 

on the speed and efficiency of translation. However, this is in contradiction with result from the 

recently published genomes of two strains of the marine prasinophyte Micromonas (Worden et al. 

2009). This study revealed that the average GC content is about 65 %, and that both strains 

contained a region with a GC content 14 % lower than average (comprising about 7% of the total 

genome). These low GC regions have a higher transcriptional activity and a different codon usage 

compared to the rest of the genome. 

Conclusion and perspectives 

The mutation-selection-drift model (Bulmer 1991, Hershberg and Petrov 2008) and the coevolutionary 

model of genome composition and resource allocation (Vetsigian and Goldenfeld 2009) 

form theoretical frameworks that allow predicting how a variety of biological factors can be expected 

to influence genome biases (Fig. 1). In the green plants studied here, we have been able to show that 

several biological factors that have previously been associated with genome bias are effectively 

associated with genome bias. All of these factors may have some explanatory power in the observed 

evolutionary patterns of codon usage bias and GC content but at present it is difficult to estimate 

their relative contributions. Incorporation of these and other biological factors (e.g. population size, 

growth rate, life style) in more detailed models would allow for a more precise estimation about 

their contribution. 

The strong correlation observed between ecological characteristics and genome biases invites 

studying the mechanisms responsible for this apparent association with sets of more closely related


taxa that allow addressing the questions within the timeframe they act. The green plant lineage 

offers some useful cases as the members of this lineage occupy many distinct and widely divergent 

ecological niches, including freshwater, terrestrial and marine ecosystems, and exhibit a remarkable 

morphological and cytological diversity, ranging from unicells to complex macroscopic and 

multicellular organisms. Studying closely related organism exhibiting less biological variation but that 

differ in terms of r/K selection would allow testing the impact of these ecological affinities on 

genome biases. 

Comparison of genome biases among different species is usually performed on EST data (e.g. Cutter 

et al. 2006, Ingvarsson 2008). Such data are not available for the broad sample of taxa we have used 

here. Our calculations are based on eight housekeeping genes, implying that the codon usage biases 

we are observing may be higher than average due to the relatively high expression level one can 

expect of these genes. However, because we use the same eight genes for comparisons across the 

phylogeny, we expect that the observed patterns of genome bias will generally be preserved when 

EST data are compared. Obviously, studying larger datasets would allow for more definitive 

conclusions but this awaits the generation of EST or whole-genome data for a broader range of green 

algae, particularly siphonocladous and siphonous Ulvophyceae. 

Acknowledgments 

We thank P. Rouzé, K. Vandepoele and P. Vanormelingen for valuable discussions relating to this 

work, X. F. Wan for providing a standalone version of CodonO, Caroline Vlaeminck for assisting with 

the molecular work and W. Gillis for IT support. Funding was provided by the Special Research Fund 

(Ghent University, DOZA-01107605) and the Research Foundation Flanders (postdoctoral fellowships 

to HV, FL and ODC). Phylogenetic analyses were largely carried out on the KERMIT computing cluster 

(Ghent University).

6 

A multi-locus time-calibrated phylogeny of the siphonous green 

algae 1 

Abstract 

The siphonous green algae are an assemblage of seaweeds that consist of a single giant cell. They 

comprise two sister orders, the Bryopsidales and Dasycladales. We infer the phylogenetic 

relationships among the siphonous green algae based on a five-locus data matrix and analyze 

temporal aspects of their diversification using relaxed molecular clock methods calibrated with the 

fossil record. The multi-locus approach resolves much of the previous phylogenetic uncertainty, but 

the radiation of families belonging to the core Halimedineae remains unresolved. In the Bryopsidales, 

three main clades were inferred, two of which correspond to previously described suborders 

(Bryopsidineae and Halimedineae) and a third lineage that contains only the limestone-boring genus 

Ostreobium. Relaxed molecular clock models indicate a Neoproterozoic origin of the siphonous green 

algae and a Paleozoic diversification of the orders into their families. The inferred node ages are used 

to resolve conflicting hypotheses about species ages in the tropical marine alga Halimeda. 

Keywords 

molecular phylogenetics; relaxed molecular clock; fossil algae; Bryopsidales; Dasycladales; 

Ulvophyceae; Chlorophyta 

1 Published as: Verbruggen, H., M. Ashworth, S. T. LoDuca, C. Vlaeminck, E. Cocquyt, T. Sauvage, F. W. 

Zechman, D. S. Littler, M. M. Littler, F. Leliaert, and O. De Clerck. 2009. A multi-locus time-calibrated 

phylogeny of the siphonous green algae. Molecular Phylogenetics and Evolution 50:642-653.

92 CHAPTER 6 

Introduction 

The siphonous green algae form a morphologically diverse group of marine macroalgae. They are 

readily distinguished from other green algae by their ability to form large, differentiated thalli 

comprised of a single, giant tubular cell. This tubular cell branches and fuses in various patterns to 

form a broad range of forms (Fig. 1). Individual branches of the tubular cell are called siphons. The 

siphonous green algae include two sister orders (Bryopsidales and Dasycladales) and belong to the 

chlorophytan class Ulvophyceae. The Cladophorales, an order closely related to the Bryopsidales and 

Dasycladales (Zechman et al. 1990), is sometimes included in the siphonous algae but its members 

are not truly siphonous because they are generally multicellular. The siphonous green algae are 

unique among their chlorophytan relatives in having a relatively rich fossil record because many of 

them biomineralize. 

Fig. 1. Morphology and anatomy of the siphonous green algae comprising the orders Bryopsidales (A–F) and 

Dasycladales (G–J). A. Derbesia, B. Bryopsis, C. Codium, D. Udotea, E. Halimeda, F. Caulerpa, G. Batophora, H. 

Neomeris, I. Cymopolia, J. Acetabularia. 

The Bryopsidales range in morphology from simple, branched thalli (Fig. 1A,B) to more complex 

organizations with interwoven siphons, differentiated thalli and various morpho-ecological 

specializations (Fig. 1C-F). They have multiple nuclei scattered throughout the siphons. These algae 

form an important component of the marine flora, particularly in tropical marine environments, 

where they are among the major primary producers on coral reefs, in lagoons and seagrass beds. 

They comprise roughly 500 recognized species (Guiry and Guiry 2008). The calcified representatives 

are major contributors of limestone to coral reef systems and are well-represented in the fossil 

record (Hillis-Colinvaux 1986; Mu 1990). The Bryopsidales are also common in warm-temperate

A MULTI-LOCUS TIME-CALIBRATED PHYLOGENY OF THE SIPHONOUS GREEN ALGAE 93 

marine habitats where they form a significant component of the flora (e.g., Codium). Several 

bryopsidalean taxa are vigorous invasive species (e.g., Codium fragile, Caulerpa taxifolia and 

Caulerpa racemosa var. cylindracea) that are known to have profoundly affected the ecology and 

native biota in areas of introduction. Molecular phylogenetic studies have shown two principal 

bryopsidalean lineages, both comprising simple as well as complex forms (Lam and Zechman, 2006; 

Verbruggen et al., 2009; Woolcott et al., 2000). 

Extant Dasycladales are characterized by a central axis surrounded by whorls of lateral branches (Fig. 

1G-J). Members of this group contain a single giant nucleus situated in the rhizoid during the 

vegetative phase of their life cycle. Albeit relatively inconspicuous, they are common algae of shallow 

tropical and subtropical marine habitats. Both calcified and non-calcified representatives have left a 

rich fossil record dating back to the Cambrian Period (540–488 my) (Berger and Kaever 1992; LoDuca, 

Kluessendorf, and Mikulic 2003). Fossil remains suggest that non-calcified dasyclads were most 

diverse during the Ordovician and Silurian periods and declined in favor of calcified representatives 

after the Early Devonian (± 400 my), perhaps as a result of selection for resistance to herbivory 

(LoDuca, Kluessendorf, and Mikulic 2003). In all, over 700 species are known from the fossil record, 

and fossil dasyclads are important tools for both biostratigraphic and paleoenvironmental studies 

(Berger and Kaever 1992; Bucur 1999). The present dasycladalean diversity consists of 37 species 

belonging to 10 genera and the two families Dasycladaceae and Polyphysaceae (Berger 2006). 

Molecular phylogenetic studies have shown that the Polyphysaceae arose from the Dasycladaceae, 

leaving the latter paraphyletic (Berger et al., 2003; Olsen et al., 1994; Zechman, 2003). 

Currently available phylogenetic studies of the siphonous green algae suffer from a few 

shortcomings. First, the studies have been based on single loci, yielding partially resolved trees with 

some unresolved taxa. Second, several important groups within the Bryopsidales have not been 

included. Finally, the temporal aspects of siphonous green algal diversification have not been 

explored in sufficient detail. Several recent studies point to the necessity of a time-calibrated 

phylogenetic framework. For example, the large discrepancy between species ages resulting from 

interpretations of molecular data and the fossil record creates confusion (Dragastan, Littler, and 

Littler 2002; Kooistra, Coppejans, and Payri 2002). Furthermore, biogeographic interpretations are 

difficult without reference to a temporal framework (Verbruggen et al., 2005; Verbruggen et al., 

2007). So far, the fossil record has been used on one occasion to calibrate a dasycladalean molecular 

phylogeny in geological time (Olsen et al. 1994). In the years that have passed since the publication 

of this study, however, several important discoveries have been made in dasyclad paleontology (e.g. 

Kenrick and Li, 1998; LoDuca et al., 2003) and more advanced calibration methods have been 

developed (reviewed by e.g., Verbruggen and Theriot 2008). 

The present study sets out to achieve two goals. First, it aims to resolve the phylogeny of the 

siphonous green algae more fully and include previously omitted taxonomic groups. Second, it 

aspires to provide a temporal framework of siphonous green algal diversification by calibrating the 

phylogeny in geological time using information from the fossil record. Our approach consists of 

model-based phylogenetic analyses of a five-locus alignment spanning both of the orders and uses a 

composite (partitioned) model of sequence evolution. Calibration of the phylogeny in geological time 

is achieved with Bayesian implementations of relaxed molecular clock models, using a selection of 

fossil reference points.

94 CHAPTER 6 

Materials and methods 

Data generation 

In preparation for this study, the rbcL gene of a wide spectrum of taxa was sequenced as described 

below and additional rbcL sequences were downloaded from GenBank. From a neighbor joining 

guide tree produced from these sequences, 56 taxa representing all major clades were selected. For 

these taxa, we attempted to amplify four additional loci (plastid encoding atpB, tufA and 16S rDNA, 

and nuclear 18S rDNA) or downloaded this information from GenBank. 

DNA extraction followed a CTAB protocol modified from Doyle and Doyle (1987). DNA amplification 

followed previously published protocols (Famà et al., 2002; Hanyuda et al., 2000; Karol et al., 2001; 

Kooistra, 2002; Lam and Zechman, 2006; Olson et al., 2005; Wolf, 1997), with additional primers for 

amplification of partial sequences of the rbcL gene and the 16S rDNA. A complete overview of the 

primers used is given in Fig. 2. Newly generated sequences were submitted to GenBank. A complete 

list of Genbank accession numbers is provided in Table 1. Our specimens are vouchered in the Ghent 

University Herbarium or the US National Herbarium (see GenBank records for voucher information). 

The ulvophycean algae Trentepohlia aurea, Oltmannsiellopsis viridis, Ulva intestinalis, Ulothrix zonata 

and Pseudendoclonium akinetum were used as outgroup taxa (Lopez-Bautista and Chapman 2003). 

Despite the fact that the order Cladophorales is the closest relative of Bryopsidales and Dasycladales 

(Zechman et al. 1990), we did not use it as an outgroup because it has proven impossible to amplify 

chloroplast DNA in its members using standard primer combinations. 

Sequence alignment 

After removal of introns in the rbcL gene (Hanyuda, Arai, and Ueda 2000), all sequences were of 

equal length and their alignment was unambiguous. The atpB sequences were of equal length and 

their alignment was straightforward. The tufA gene was aligned by eye on the basis of the 

corresponding amino-acid sequences and ambiguous regions were removed. After introns had been 

removed, the 16S rDNA sequences were aligned using Muscle v. 3.6 using standard parameters 

(Edgar 2004). Ambiguous alignment regions were localized by eye and removed. Alignment of 18S 

rDNA sequences followed an analogous procedure. The insert near the 3' terminus of the 18S gene of 

certain Bryopsidales (Durand et al., 2002; Hillis et al., 1998; Kooistra et al., 1999) was removed 

because it was not present in all taxa and virtually impossible to align among genera. After removal 

of this region, sequences were aligned with Muscle v. 3.6 and stripped of their ambiguous regions. 

The concatenated alignment used for analysis is available through TreeBase and www.phycoweb.net. 

Partitioning and model selection 

Selection of a suitable partitioning strategy and suitable models for the various partitions used the 

Bayesian Information Criterion (BIC) as a selection criterion (e.g., Sullivan and Joyce 2005). The guide 

tree used during the entire procedure was obtained by maximum likelihood (ML) analysis of the 

concatenated data with a JC69 model with rate variation among sites following a discrete gamma


distribution with 8 categories (JC69+Γ8) inferred with PhyML (Guindon and Gascuel 2003). All 

subsequent likelihood optimizations and BIC calculations were carried out with TreeFinder (Jobb, von 

Haeseler, and Strimmer 2004). Six alternative partitioning strategies were considered (see results). 

Per partitioning strategy, six substitution models were optimized with unlinked parameters between 

partitions and a partition rate multiplier (see results). The partitioning strategy + model combination 

receiving the lowest BIC score was used in the phylogenetic analyses documented below. 

Fig. 2. Primers used for amplification and/or sequencing of the five loci used in this study.

96 CHAPTER 6 

Phylogenetic analyses 

Phylogenetic analyses consisted of Bayesian inference (BI) and maximum likelihood tree searches 

using the unrooted partitioned model selected using the BIC procedure (section 2.3). For Bayesian 

inference, three independent runs, each consisting of four incrementally heated, Metropolis-coupled 

chains were run for ten million generations using MrBayes v. 3.1.2 (Ronquist and Huelsenbeck 2003). 

The default heating factor, priors, proposal mechanisms and other settings were used. Rate 

heterogeneity among partitions was modeled by using a variable rate prior. Parameter values and 

trees were saved every thousand generations. Convergence and stationarity of the runs was assessed 

using Tracer v. 1.4 (Rambaut and Drummond 2007). An appropriate burn-in value was determined 

with the automated method proposed by Beiko et al. (2006). Their method was applied to each run 

individually, with a sliding window of 100 samples. The post-burnin trees from different runs were 

concatenated and summarized using MrBayes' sumt command. 

Maximum likelihood tree searches were carried out with TreeFinder, which allows likelihood tree 

searches under partitioned models (Jobb, von Haeseler, and Strimmer 2004). Tree space coverage in 

the TreeFinder program is low compared to other ML programs. Therefore, an analysis pipeline was 

created to increase tree space coverage by running analyses from many start trees. First, a set of 

start trees was created by randomly modifying the guide tree used for model selection by 100 and 

200 nearest neighbor interchange (NNI) steps (50 replicates each). Analyses were run from these 100 

start trees. The 3 trees yielding the highest likelihood were used as the starting point for another set 

of NNI modifications of 20 and 50 steps (50 replicates each). A second set of ML searches was run 

from each of the resulting 300 start trees. The ML tree resulting from this set of analyses was 

retained as the overall ML solution. The same partitions and models as in the BI analysis were used, 

with the second-level tree search and proportional partition rates. Branch support was calculated 

using the bootstrap resampling method (1000 pseudo-replicates). Bootstrap analyses used the same 

settings but started from the ML tree. 

Relaxed molecular clock 

Likelihood ratio tests significantly rejected a strict (uniform) molecular clock for the alignment. Node 

age estimates were therefore obtained by fitting a relaxed clock model to our molecular data. The 

assumption of the molecular clock was relaxed by allowing rates of molecular evolution to vary along 

the tree according to a Brownian motion model (Kishino et al., 2001; Thorne et al., 1998). First, an 

F84 model with rate variation across sites following a discrete gamma distribution with 20 rate 

categories was optimized in PAML v.4 (Yang 2007), using the topology obtained with Bayesian 

inference from which all but two outgroup taxa were removed (Trentepohlia and Oltmannsiellopsis). 

Second, a variance-covariance matrix was built with estbNew from the T3 package (Thorne et al. 

1998; http://abacus.gene.ucl.ac.uk/software.html). Finally, node age estimates were obtained by 

running three independent MCMC chains with the PhyloBayes program (Lartillot, Blanquart, and 

Lepage 2007) on the variance-covariance matrix, using the lognormal auto-correlated clock model. 

Sensitivity analyses were carried out to evaluate the impact of some potentially erroneous fossil 

assignments with chains of 50,000 generations that were sampled every 100th generation. The final 

analysis consisted of three independent runs of 1,000,000 generations, sampled every 100th


generation. Convergence and stationarity of the chains was evaluated by plotting trace files in Tracer 

v. 1.4 (Rambaut and Drummond 2007). The fossils used as calibration points are documented in the 

results section. 

Results 

Data exploration 

The data compiled for our analyses consisted of 155 sequences. The resulting five-locus data matrix 

was 51% filled. The rbcL gene was best represented (90% filled), followed by 18S rDNA (62%), tufA 

(49%), 16S rDNA (33%) and atpB (20%). After ambiguously aligned parts had been removed, the 

matrix measured 61 taxa by 5588 characters. The 18S rDNA provided most characters (1555 bp), 

followed by rbcL (1386 bp), 16S rDNA (1327 bp), tufA (804 bp) and atpB (516 bp). 

The BIC-based model selection procedure selected a model with four partitions and GTR+Γ8 

substitution models for each partition (Table 2). The four partitions were: (1) ribosomal loci: 18S and 

16S, (2) first codon positions of rbcL, tufA and atpB combined, (3) second codon positions of the 

three genes, and (4) third codon positions of the three genes. 

Phylogeny of the siphonous algae 

Different runs of the Bayesian phylogenetic analysis of the five-locus dataset under an unrooted 

partitioned model rapidly converged onto highly similar posterior distributions. The chains reached 

convergence after at most 384,000 generations. The analyses resulted in a well-resolved 

phylogenetic hypothesis of the siphonous green algae (Fig. 3A). The backbone of the dasycladalean 

clade was determined with high confidence except for the branching order between the Bornetella 

lineage, the Cymopolia-Neomeris clade and the Polyphysaceae. Relationships among Acetabularia 

species are not resolved with our dataset. In the Bryopsidales, three main clades were inferred, two 

of which correspond to previously described suborders (Bryopsidineae and Halimedineae). A third 

lineage contained only the limestone-boring genus Ostreobium. Relationships among the three 

families of the Bryopsidineae were inferred with high confidence (Fig. 3B) whereas within the 

Halimedineae, relationships among what we will refer to as the "core Halimedineae" (families 

Caulerpaceae, Rhipiliaceae, Halimedaceae, Pseudocodiaceae and Udoteaceae) remained poorly 

resolved (Fig. 3C).

98 CHAPTER 6 

Fig 3. Phylogenetic relationships among the siphonous green algae inferred from a five-locus DNA alignment 

using Bayesian analysis under a partitioned, unrooted model. A. Majority-rule consensus phylogram of postburnin 

trees. B. Relationships among families of the Bryopsidineae. C. Relationships among families of the core 

Halimedineae. Numbers at nodes indicate statistical support, posterior probabilities before the dash and ML 

bootstrap proportions after (both given as percentages). Encircled letters indicate calibration points (Table 3). 

The scale bar only applies to panel A.

Time-calibrated phylogeny 


We compiled a list of fossils that are thought to represent early records of lineages observed in our 

phylogenetic tree (Table 3). Prior to carrying out the main calibration analysis, we ran several 

analyses to test the sensitivity of results to the choice of certain fossils as calibration points and of 

the maximum age constraint imposed on the root of the siphonous algae. The exact combination of 

constraints used in these trial runs can be found in the online appendix. 

Analyses with Dimorphosiphon (min. 460.9 my) constrained at node f resulted in considerably older 

node age estimates throughout the tree than analyses in which this constraint was not imposed. 

Dimorphosiphon has been regarded as an ancestral taxon of the Halimedineae (Dragastan, Littler, 

and Littler 2002) but its combination of characters did not allow unambiguous placement on any 

node of the tree. From the incompatibility of this calibration point with the others, we concluded 

that our admittedly speculative assignment of this taxon to node f was unjustified. With 

Dimorphosiphon excluded from analyses, various combinations of the remaining calibration points 

led to very similar results. 

The analysis was sensitive to the maximum age constraint imposed on the phylogeny. By default, we 

used a maximum age constraint of 635 my on the root of the siphonous algae (node a) because there 

is no clear evidence for either Bryopsidales or Dasycladales in Ediacaran Konservat-Lagerstätten, in 

which macroalgae are abundant and well preserved (Xiao et al., 2002; Zhang et al., 1998). These 

deposits do, however, contain simple cylindrical and spherical forms (e.g., Sinospongia, 

Beltanelliformis) that may be ancestral to these extant lineages (Xiao et al. 2002). Changing the 

maximum age constraint to 800 my as a sensitivity experiment increased average node ages and 

widened their 95% confidence intervals. Analogously, lowering the maximum age constraint to 500 

my decreased the average ages and narrowed the confidence intervals. The chronogram presented 

in Fig. 4 resulted from the analysis without Dimorphosiphon and with a maximum age constraint of 

635 my; more specifically, the analysis was carried out with calibration points a1, a2, b1, c, d2, e1 and 

h1 (Table 3). Repeating this analysis without age constraint on node b resulted in a chronogram that 

was very similar. 

As an alternative strategy to using a maximum age constraint on the root of the siphonous algae 

(node a), we constrained the age of node b to be Early Devonian (416.0–397.5 my) based on the first 

occurrence of fossils with thalli comparable to those of Dasycladaceae, in terms of both general 

thallus form and reproductive structures, in strata of this age (Uncatoella: Kenrick and Li, 1998). This 

analysis, which is illustrated in the online appendix, should be regarded as a conservative alternative 

to the analysis presented in Fig. 4 and node ages in this chronogram should be interpreted as 

minimum values rather than realistic estimates. We will use the chronogram in Fig. 4 as the preferred 

reference time frame for the remainder of the paper because the maximum age constraint used for 

this analysis has an empirical basis (although it is based on absence of evidence rather than evidence 

of absence) and because we consider the intermediate-sized confidence intervals on the node ages 

of this tree to yield a fairly realistic image of the true uncertainty surrounding the node ages. In what 

follows, we report node age estimates as the mean node age followed by the 95% confidence 

interval in square brackets.

100 CHAPTER 6 

The origin of the orders Dasycladales and Bryopsidales was inferred to be in the Neoproterozoic (571 

my [628–510]). The earliest divergence between extant Dasycladaceae lineages occurred during the 

Ordovician or Silurian (458 my [517–407]) and the family diversified into its main extant lineages 

during the Devonian. The Polyphysaceae originated from the Dasycladaceae during the second half 

of the Paleozoic (367 my [435–306]). In the Bryopsidales, the suborders Bryopsidineae and 

Halimedineae diverged from each other in the Early Paleozoic (456 my [511–405]) and diversified 

into their component families during the second half of the Paleozoic. The families belonging to the 

core Halimedineae appear to have diverged from one another during the Permian. 

Discussion 

Taxonomic implications 

The increased sampling of taxa and loci compared to previous studies has produced some results 

that deserve mention from a taxonomic viewpoint. First, the lineage leading to the limestone-boring 

genus Ostreobium seems to deserve recognition at the suborder level, hence the tentative clade 

name Ostreobidineae. The bryopsidalean family Udoteaceae disintegrates. Besides the transfer of 

Avrainvillea and Cladocephalus to the Dichotomosiphonaceae, which was previously indicated by 

Curtis et al. (2008), a number of Rhipilia and Rhipiliopsis species form a lineage of their own and the 

remainder of the Udoteaceae receives little statistical phylogenetic support. We have applied the 

name Rhipiliaceae to the clade of Rhipilia and Rhipiliopsis species. This family name was proposed 

earlier but its phylogenetic relevance had not yet been demonstrated (Dragastan et al. 1997; 

Dragastan and Richter 1999). The same authors proposed the family Avrainvilleaceae to harbor the 

extant genera Avrainvillea and Cladocephalus and some fossil genera. However, the ultrastructural 

affinities between Dichotomosiphon, Avrainvillea and Cladocephalus (Curtis, Dawes, and Pierce 2008) 

and the limited DNA sequence divergence between Dichotomosiphon and Avrainvillea shown here 

suggest that transferring Avrainvillea and Cladocephalus to the Dichotomosiphonaceae would be 

more appropriate. Although the fossil taxa in the Avrainvilleaceae are more difficult to evaluate 

because no ultrastructural evidence is available to link them unequivocally to the 

Dichotomosiphonaceae, we nevertheless propose their transfer to this family for taxonomic 

convenience. We refrain from formally describing a suborder Ostreobidineae for Ostreobium because 

we feel that the description of such high-level taxa should be accompanied by detailed ultrastructural 

observations. 

Our results for the phylogeny of the Dasycladales are consistent with previous studies and do not add 

much insight into their pattern of diversification. One thing worth mentioning is that the close 

relationship between Batophora and Chlorocladus, which was suggested by previous molecular 

phylogenetic studies (Olsen et al. 1994; Zechman 2003), is confirmed by our multi-locus dataset. As 

reported in the aforementioned papers, this finding contradicts the traditional tribe-level 

classification by Berger and Kaever (1992).


Fig. 4. Chronogram of the siphonous green algae. Node ages were inferred using Bayesian inference assuming a 

relaxed molecular clock and a set of node age constraints derived from the fossil record. Values at nodes 

indicate average node ages and bars represent 95% confidence intervals. The calibration points used for this 

analysis are a1, a2, b1, c, d2, e1 and h1 (Table 3).

102 CHAPTER 6 

Dasycladalean diversification 

Of the five dasycladalean families, only Dasycladaceae and Polyphysaceae have recent 

representatives and Seletonellaceae, Triploporellaceae and Diploporellaceae are entirely extinct 

(Berger and Kaever 1992). The Seletonellaceae includes the oldest fossils assigned to Dasycladales, 

Yakutina aciculata from the Middle Cambrian (Kordé 1973; Riding 1994; Riding 2001) and Seletonella 

mira from the Upper Cambrian (Kordé 1950a; Riding 1994; Riding 2001). Cambroporella from the 

Lower Cambrian of Tuva, was initially described as a dasyclad (Kordé 1950b) but was reinterpreted as 

a probable bryozoan (Elias 1954). Unlike living dasyclads, Seletonellaceae contained gametes within 

the main axis (endospore reproduction) as opposed to within gametophores (sensu Dumais and 

Harrison 2000), and developed laterals irregularly along the length of the main axis (aspondyl 

structure), rather than in whorls as is the case in living representatives (euspondyl structure). 

Nonetheless, two lines of evidence support a dasyclad affinity for the Seletonellaceae. First, in terms 

of reproduction, development of reproductive cysts within the main axis is known as a teratology 

among living dasyclads (Valet 1968) and all living dasyclads pass through an "endospore stage" as 

swarms of haploid secondary nuclei generated by meiosis of the large primary nucleus in the rhizoid 

migrate upward through the main axis towards the gametophores (Elliott 1989; Berger and Kaever 

1992). Second, an aspondyl structure is a predicted outcome of a detailed reaction-diffusion model 

of dasyclad whorl morphogenesis (Dumais and Harrison 2000). 

Our chronogram suggests a late Neoproterozoic origin for the Dasycladales, with the 95% confidence 

interval ranging into the Cambrian: 571 my [628–510] (Fig. 4). The upper boundary on this node age 

(628 my) should not be overinterpreted because it is largely determined by the upper age constraint 

on this node (635 my). The Middle Cambrian lower boundary on this node age (510 my), however, 

can be taken as a fairly safe minimum age estimate for the Dasycladales as well as the Bryopsidales. 

Unlike Yakutina and Seletonella, most early Paleozoic dasyclad taxa did not biomineralize (LoDuca, 

Kluessendorf, and Mikulic 2003). This fact, when taken together with the results of the present 

analysis, suggests that noncalcified dasyclads may yet be discovered within Konservat-Lagerstätten of 

latest Neoproterozoic or Early Cambrian age. 

Taxa assigned to Dasycladales are both abundant and diverse in Ordovician strata, and include 

euspondyl (e.g., Chaetocladus) as well as aspondyl (e.g., Dasyporella) forms (Berger and Kaever, 

1992; LoDuca, 1997; LoDuca et al., 2003). The earliest euspondyl forms, like aspondyl taxa, were 

endospore. Later, however, at least one lineage of euspondyl forms emerged in which 

gametogenesis occurred within the laterals (cladospore reproduction). Collectively, endospore and 

cladospore taxa with a euspondyl structure comprise the family Triploporellaceae (Berger and Kaever 

1992; LoDuca 1997). Triploporellaceae are though to have originated from the Seletonellaceae, 

indicating that the latter is a paraphyletic group. 

The family Dasycladaceae is characterized by euspondyl thalli with spherical gametophores along the 

sides or tips of laterals (choristospore reproduction). It is though to have originated from the 

Triploporellaceae, in which case the latter group, like Seletonellaceae, is paraphyletic. Studies 

indicate that the dasycladacean gametophore, rather than being a modified lateral, is instead a 

separate structure with its own morphogenetic identity (Dumais and Harrison 2000). The oldest 

taxon with gametophores and associated reproductive cysts comparable to those of living 

Dasycladaceae is Uncatoella verticillata from the Lower Devonian (416–397 my) of China (Kenrick


and Li 1998). This taxon is somewhat problematic, however, in that the asymmetrical manner of 

lateral branching and instances of pseudodichotomies of the main axis are otherwise unknown 

among dasyclads (Kenrick and Li 1998). Uncatoella is the only taxon known from Paleozoic strata 

with gametophores and associated reproductive cysts comparable to those of living Dasycladaceae. 

Our molecular results suggest that the earliest divergence between the extant Dasycladaceae 

lineages occurred during the Ordovician or Silurian (458 my [517-407]). Notably, this result emerges 

regardless of whether Uncatoella is used to constrain the minimum age of node b. 

Archaeobatophora, from the Upper Ordovician of Michigan (Nitecki 1976), is of interest in this 

regard, as it bears a striking resemblance to the vegetative thallus of the extant dasycladacean 

Batophora. However, gametophores are not present among the few known specimens (all on a single 

small slab), and thus the status of this form as an early dasycladacean remains uncertain. Our results 

suggest that the main extant lineages belonging to the Dasycladaceae originated during the 

Devonian. This is somewhat older than expected because, with the possible exception of the 

Batophoreae (Archaeobatophora), the oldest fossils assigned to extant tribes within the 

Dasycladaceae are from the Mesozoic (Berger and Kaever 1992). Similarly, the age estimate for the 

split between Cymopolia and Neomeris, two calcified genera known from numerous occurrences in 

the fossil record, is somewhat older than anticipated: 211 my [300–152] vs. a Barremian age for the 

earliest Neomeris fossil (130–125 my) (Sotak and Misik 1993). Two factors could explain these 

discrepancies. First, it is possible that many early Dasycladaceae did not biomineralize. This is 

certainly true for most early Triploporellaceae and Selenonellaceae (LoDuca, Kluessendorf, and 

Mikulic 2003), and applies to Uncatoella and several living Dasycladaceae as well (e.g., Batophora). 

The lack of biomineralization would have severely limited the preservation potential of these taxa. 

Second, it may be that some Paleozoic and early Mesozoic calcified taxa previously inferred on the 

basis of lateral morphology to have been endospore or cladospore were in fact choristospore. A 

choristospore condition for such forms could be masked if calcification was restricted to the 

outermost part of the thallus, beyond the level of the gametophores, as is known for the living taxon 

Bornetella. 

The family Polyphysaceae is characterized by the development of clavate gametophores arranged in 

whorls (Berger and Kaever 1992; Berger et al. 2003). The distinctive gametophores of this group 

appear to have originated through modification of the spherical gametophores of Dasycladaceae 

(Dumais and Harrison 2000). Our results suggest that the Polyphysaceae originated and began their 

diversification in the second half of the Paleozoic. An Early Carboniferous origin of the family had 

been suggested based on the fossil taxon Masloviporella (Deloffre 1988; Berger and Kaever 1992). 

Eoclypeina and Clypeina are taxa classified as polyphysaceans from the Permian and Triassic, 

respectively (Berger and Kaever 1992). The extant polyphysacean Halicoryne has been reported from 

the Triassic (Misik, 1987; Tomasovych, 2004). However, because these reports are based solely on 

isolated reproductive elements, both the genus- and family-level assignment of this material must be 

regarded as tentative (Barattolo and Romano 2005). 

According to the molecular clock results, the polyphysacean taxon Acetabularia originated as early as 

the early Mesozoic. The oldest Acetabularia fossils, however, are from the Oligocene (Berger and 

Kaever 1992). Here, too, this discrepancy may reflect poor representation of the group in the fossil 

record, as living members of the genus are only very weakly calcified. Material from Mesozoic-age

104 CHAPTER 6 

strata has been assigned to the closely related extant taxon Acicularia (Iannace, Radiocic, and 

Zamparelli 1998). As is the case for reports of Halicoryne from the Triassic, however, such an 

assignment must be regarded as equivocal because the material at hand comprises only isolated 

reproductive elements (Barattolo and Romano 2005). 

Overall, the results of the relaxed molecular clock analysis line up rather well against major aspects 

of dasyclad evolution inferred from the dasyclad fossil record. Nonetheless, our results indicate that 

the fossil record of certain aspects of the evolutionary history of the Dasycladaceae may be more 

incomplete than previously suspected (see Kenrick and Li 1998). The node age estimates obtained in 

our relaxed molecular clock analysis are older than those recovered by Olsen et al. (1994). Causes for 

this discrepancy are difficult to pinpoint because differences between the studies are manifold: the 

molecular clock method, calibration technique, molecular dataset and selection of fossils all differ. 

Whereas the previous study relied exclusively on 18S rDNA sequences, our dataset includes several 

additional loci, which should yield more reliable results (Magallón and Sanderson 2005). 

Furthermore, at the time the previous study was published, only uniform (strict) molecular clock 

methods were available. Based on our alignment, 18S rDNA violates the uniform molecular clock 

(LRT: –2 lnL = 399.403, p < 0.0001), a statement also true for the other loci in our dataset. As a 

consequence, node age estimates from our analysis with relaxed molecular clock models should 

match the true divergence times more closely. The calibration method differs in that fossils are used 

as minimum age estimates for stem groups in our analyses, whereas they were used as point 

estimates to determine the rate of 18S rDNA evolution in Olsen et al. (1994). Finally, we use a more 

extensive list of fossil calibration points, including some recently described fossils that represent 

older occurrences of some dasyclad clades. Several of the aspects mentioned above would predict 

our age estimates to be older than those of Olsen et al. (1994), which is in agreement with the 

empirical results. 

Bryopsidalean diversification 

The fossil record of the Bryopsidales is not as well characterized as that of the Dasycladales. Many 

fossil taxa that are generally considered to belong to the order have been described (reviewed by 

Bassoullet et al., 1983; Dragastan et al., 1997; Dragastan and Schlagintweit, 2005), but the taxonomic 

placement of these taxa is often ambiguous (Mu 1990). Dragastan and Schlagintweit (2005) 

presented an evolutionary scenario of bryopsidalean diversification based on their interpretation of 

calcified fossil taxa. They proposed a temporal succession of three main lineages. The primitive 

Dimorphosiphonaceae, characterized by a single medullar siphon and a simple cortex, originated in 

the Neoproterozoic or early Paleozoic and persisted through the Devonian (± 360 my). The 

Protohalimedaceae, characterized by multiple medullar siphons and a cortex of variable complexity, 

diverged from the Dimorphosiphonaceae in the early Silurian (± 440 my) and thrived up to the PTboundary 

(± 250 my), when it suffered major losses, and continued into the Mesozoic to go extinct 

during the Cretaceous. The extant family Halimedaceae, which features multiple medullar siphons 

and a complex cortex, was thought to have diverged from the Protohalimedaceae in the second half 

of the Permian (270–250 my) and diversified through the Mesozoic and Cenozoic.


In our opinion, relationships between fossil taxa and lineages in the phylogeny of extant taxa are not 

evident. This uncertainty is reflected in our study by the lower number of fossil calibration points 

used within the Bryopsidales. Our time-calibrated phylogeny indicates that after the initial 

diversification of the order into its suborders during the early Paleozoic, current families originated in 

the second half of the Paleozoic. It also suggests that calcification is a relatively recent phenomenon 

in the extant lineages of the Bryopsidales because the Halimedaceae and Udoteaceae, the only 

extant families with calcified, corticated representatives, originated during the Permian (± 300–250 

my) and diversified during the Mesozoic (Fig. 4). All other lineages of the Halimedineae, including 

some recently discovered lineages (Verbruggen et al. 2009), are not calcified. As a consequence, the 

presence of the calcified, corticated families Dimorphosiphonaceae and Protohalimedaceae in older 

deposits is difficult to interpret in the context of our phylogeny. Based on our results, the classical 

paleontological interpretation that the Dimorphosiphonaceae and Protohalimedaceae are direct 

ancestral forms of the serial-segmented Halimedaceae seems doubtful. First, our phylogenetic results 

show no indication of the presence of calcification in Bryopsidales prior to the Permian. Second, it 

follows from our phylogeny that the internal architecture of thalli was relatively simple up until the 

late Paleozoic. Thalli consisting of a medulla and a cortex appear to have evolved from simple, 

siphonous thalli several times independently during the Permian–Triassic period. They evolved once 

in the Bryopsidineae (Codium), a second time in the Dichotomosiphonaceae (Avrainvillea) and a third 

time in the core Halimedineae (Halimedaceae, Udoteaceae, Pseudocodiaceae). 

Some alternative hypotheses may be posited to explain this disparity of results, all of which should 

be regarded as speculative. Assuming that the Dimorphosiphonaceae and Protohalimedaceae are 

genuine Bryopsidales, these two families could very well represent an early-diverging bryopsidalean 

lineage that went extinct. Alternatively, they could represent a collection of taxa that branched off at 

various places along the lineage leading from the origin of the Halimedineae to the Halimedaceae 

and Udoteaceae. 

Perspectives 

The genera Codium and Halimeda have been proposed as model systems for studying marine algal 

speciation, biogeography and macroevolution (Kooistra et al., 2002; Verbruggen et al., 2007). Both 

these genera are species-rich, ecologically diverse and geographically widespread, making them ideal 

case studies for a spectrum of evolutionary questions. Furthermore, these genera contrast in their 

climatic preferences, Halimeda being mostly tropical and Codium being more diverse in temperate 

seas. Further development of these case studies will greatly benefit from the temporal framework 

provided here. Both genera diverged from their respective sister lineages in the late Paleozoic 

(Halimeda: 273 my [327–223]; Codium: 307 my [370–245]). The most recent common ancestor of the 

extant species in both genera are highly comparable and can be situated in the Late Jurassic – Early 

Cretaceous (Halimeda: 147 my [198–103]; Codium: 153 my [217–97]). To get an initial idea of the 

time-frame of evolutionary diversification we have included representatives of each of the five 

sections of Halimeda (Verbruggen and Kooistra 2004) and each of the three major lineages of Codium 

(Verbruggen et al. 2007). It follows from our results that both genera spawned their major extant 

lineages during the Cretaceous. This close match between the timeframes of evolutionary 

diversification of both genera is convenient for comparative studies between them.

106 CHAPTER 6 

One advantage of using the calcified genus Halimeda as a model for marine evolution is that 

phylogenetic results can be contrasted with its extensive fossil record. An initial comparison of our 

results to interpretations of the fossil record marks a different timeframe of diversification: whereas 

the fossil record suggests diversification of the genus into its main lineages (taxonomic sections) in 

the Eocene (roughly 56–34 my) (Dragastan and Herbig 2007), our results indicate a considerably 

older, Cretaceous divergence (between approximately 147 and 97 my). The contradiction of these 

results cannot be explained at present, but three hypotheses are suggested. First, a deviation of the 

true molecular evolutionary process from the relaxed molecular clock model used here could cause 

the contradiction. Second, misinterpretation of the fossil record used to calibrate our tree may be at 

the basis. Third, a significant gap could be present in the fossil record. To distinguish between these 

alternative scenarios, additional paleontological as well as molecular phylogenetic research is 

needed. Paleontologists should document additional time-series in the fossil record, especially 

through the Cretaceous and Paleogene systems (145–23 my). Molecular systematists should evaluate 

the fit of different clock relaxation models to the molecular data (e.g., Lepage et al. 2007) and take 

different calibration approaches, either top-down (as was done here) or bottom-up (from calibration 

points within the genus). 

The literature on the genus Halimeda is marked by a contradiction of species ages inferred from 

paleontological and molecular phylogenetic data. In the paleontological literature, extant species are 

commonly reported from Miocene (> 5 my) and Paleogene deposits (> 23 my), and some are even 

thought to date back to the Triassic (> 200 my) (Dragastan, Littler, and Littler 2002; Dragastan and 

Herbig 2007). In contrast, interpretation of vicariance patterns in molecular phylogenetic trees 

implied that extant species were younger than 3 my, and reports of extant species in the fossil record 

were interpreted as the result of iterative convergent evolution (Kooistra, Coppejans, and Payri 

2002). Because of this possibility, we have not used any calibration points within the genus; only its 

earliest appearance was used to calibrate the clock. Even though the present study was not designed 

to answer questions about species ages in Halimeda, we can glean some information from its results. 

These indicate that the five sections of Halimeda diverged from one another during the first half of 

the Cretaceous (between approximately 147 and 97 my). Because each of the sections diversified 

relatively soon after they originated, we conclude that species ages must be considerably older than 

implied by Kooistra et al. (2002). This result also implies that the iterative morphological convergence 

hypothesis should be re-evaluated. It should be clear that we do not imply that the hypothesis is 

false, and records of extant species in Miocene and older sediments should certainly not be accepted 

without scrutiny. For example, Dragastan et al. (2002) synonymized several Mesozoic taxa, including 

some Triassic ones (> 200 my) with the extant taxon H. cylindracea because of similar segment 

shape. It follows directly from our results that no extant species can be of Triassic age and we are of 

the opinion that using extant species as form taxa for fossils is undesirable. Comparative studies of 

extant taxa in a molecular phylogenetic framework have presented unambiguous evidence that 

morphological convergence, especially of segment shape, has occurred during Halimeda evolution 

(Kooistra et al., 2002; Verbruggen et al., 2005). Consequently, interpretation of fossils as extant 

species should follow only from statistically sound morphometric analyses, preferably using timeseries 

in various parts of the world.

Acknowledgments 


We thank Fabio Rindi and Juan Lopez-Bautista for providing a Trentepohlia rbcL sequence. We are 

grateful to Barrett Brooks, Roxie Diaz, Cristine Galanza, John Huisman, Gerry Kraft, Tom and Courtney 

Leigh, Dinky Olandesca, Tom Schils and John West for collecting specimens or providing assistance in 

the field. We thank Fabio Rindi and an anonymous referee for their constructive comments on a 

previous version of the manuscript. Funding was provided by FWO-Flanders (research grant 

G.0142.05, travel grants and post-doctoral fellowships to HV and FL), NSF (DEB-0128977 to FWZ), the 

Ghent University BOF (doctoral fellowship to EC and travel grant to FWZ), the Smithsonian Marine 

Station at Fort Pierce, Florida (SMS Contribution 760), the Flemish Government (bilateral research 

grant 01/46) and the King Leopold III Fund for Nature Exploration and Conservation.

108 CHAPTER 6 

Table 1. Taxon list with Genbank accessions and voucher numbers (in parentheses). Vouchers are deposited in the Ghent University Herbarium, the US 

National Herbarium or the Zechman lab herbarium (CSU Fresno). 

taxon rbcL tufA atpB 16S 18S 

Acetabularia acetabulum AY177738 FJ535854 (HV1287) Z33461 

Acetabularia calyculus FJ535855 (HV389) 

Acetabularia crenulata AY177737 FJ539159 Z33460 

Acetabularia dentata AY177739 FJ480413 Z33468 

Acetabularia peniculus AY177743 FJ539163 Z33472 

Acetabularia schenkii AY177744 Z33470 

Avrainvillea lacerata FJ432635 (HV599) FJ432651 (HV599) FJ535833 (HV599) 

Avrainvillea nigricans FJ432636 (HV891) FJ432652 (HV891) FJ535834 (HV891) 

Batophora occidentalis AY177747 FJ539160 Z33465 

Batophora oerstedii AY177748 Z33463 

Bornetella nitida AY177746 FJ480414 Z33464 

Bornetella sp. FJ535850 (LB1029) 

Bryopsidella neglecta AY004766 

Bryopsis hypnoides AY942169 AY221722 

Bryopsis plumosa FJ432637 (HV880) FJ432653 (HV880) FJ480417 FJ535835 (HV880) FJ432630 (HV880) 

Caulerpa flexilis AJ512485 DQ652532 

Caulerpa sertularioides AY942170 FJ432654 (HV989) AY389514 AF479703 

Caulerpa taxifolia AJ316279 AJ417939 FJ539164 

Caulerpa verticillata AJ417967 

Caulerpella ambigua FJ432638 (TS78) FJ432655 (TS24) FJ535836 (TS78) 

Chalmasia antillana FJ539161 AY165785 

Chlorocladus australasicus AY177750 Z33466 

Chlorodesmis fastigiata FJ432639 (HV102) FJ535837 (HV102) AF416396 

Codium lineage 1 AB038481 FJ432662 (H0882) FJ535838 (SD0509370) 

Codium lineage 2 M67453 U09427 U08345 FJ535848 (KZN2K4.1) 

Codium lineage 3 EF108086 (DHO2.178) FJ535856 (KZN2K4.10) FJ535839 (H0773) FJ535849 (KZN2K4.10) 

Cymopolia spp. FJ535851 (WP011) Z33467 

Derbesia marina AF212142 

Derbesia tenuissima FJ535852 (H0755) FJ535857 (H0755)

Dichotomosiphon tuberosus AB038487 

Flabellia petiolata FJ432640 (HV1202) FJ535847 (HV1202) AF416389 

Halicoryne wrightii AY177745 FJ535858 (HV565) AY165786 

Halimeda discoidea AB038488 AY826360 (SOC299) FJ480416 AF407254 (SOC299) 

Halimeda gracilis AM049965 (HV317) AF407257 (HEC11839) 

Halimeda incrassata AY942167 AM049958 (H0179) AF407233 (H0179) 

Halimeda micronesica AM049964 (WLS420-02) AF407243 

Halimeda opuntia AB038489 AM049967 (HV61) AF407267 (H0484) 

Neomeris dumetosa Z33469 

Oltmannsiellopsis viridis DQ291132 DQ291132 DQ291132 DQ291133 D86495 

Ostreobium sp. FJ535853 (H0753) FJ535859 (H0753) FJ535840 (H0753) 

Parvocaulis exigua AY177740 FJ539162 

Parvocaulis parvula AY177741 Z33471 

Pedobesia spp. AY004768 FJ535841 (HV1201) 

Penicillus capitatus FJ432641 (HV338) AF416404 

Penicillus dumetosus AY942175 AF416406 

Pseudendoclonium akinetum AY835431 AY835432 AY835433 AY835434 DQ011230 

Pseudocodium floridanum AM909692 (NSF.I23) AM909697 (NSF.I23) FJ432631 (NSF.I23) 

Pseudocodium natalense AM909693 (KZNb2241) AM049969 (KZNb2241) FJ535842 (KZNb2241) FJ432632 (KZNb2242) 

Rhipidosiphon javensis FJ432644 (DML40134) FJ535843 (DML40134) 

Rhipilia crassa FJ432645 (H0748) FJ432657 (HV738) FJ535844 (H0748) 

Rhipilia nigrescens FJ432646 (H0847) FJ432658 (HV788) FJ432633 (HV788) 

Rhipilia tomentosa AY942164 

Rhipiliopsis profunda FJ432647 (DML51973) FJ432659 (DML51973) FJ535845 (DML51973) 

Rhipocephalus phoenix FJ432648 (HV404) FJ535846 (HV404) AF416402 

Trentepohlia aurea FJ534608 DQ399590 

Tydemania expeditionis AY942161 FJ432661 (HV873) FJ432634 (HV873) 

Udotea flabellum AY942166 AF407270 

Udotea glaucescens FJ432650 (H0862) 

Udotea spinulosa AY942160 

Ulothrix zonata AF499683 Z47999 

Ulva intestinalis AB097617 AY454399 AJ000040

110 CHAPTER 6 

Table 2. Selection of partitioning strategy and model of sequence evolution using the Bayesian 

Information Criterion (BIC). The log-likelihood, number of parameters and BIC score are listed for 

various combination of partitioning strategies and models of sequence evolution. Each partition had 

its own copy of the model of sequence evolution with a separate set of model parameters. Lower BIC 

values indicate a better fit of the model to the data. The lowest BIC (in boldface) was observed for 

the partitioning strategy with four partitions (ribosomal DNA and separate codon positions for each 

gene), and GTR+ 8 models applied to each partition. 

model lnL # par BIC 

single partition 

F81 -56405.2 122 113863 

F81+ 8 -51296.8 123 103655 

HKY -56009.6 123 113080 

HKY+ 8 -50707.8 124 102486 

GTR -54913.6 127 110923 

GTR+ 8 -50224.1 128 101553 

2 partitions: ribosomal DNA, protein coding 

F81 -55852.3 126 112792 

F81+ 8 -50702.4 128 102509 

HKY -55249.8 128 111604 

HKY+ 8 -49969.8 130 101061 

GTR -54303.0 136 109779 

GTR+ 8 -49774.3 138 100739 

5 partitions: one for each locus (atpB, rbcL, tufA, 16S, 18S) 

F81 -55772.5 138 112736 

F81+ 8 -50769.0 143 102772 

HKY -55160.7 143 111555 

HKY+ 8 -49913.6 148 101104 

GTR -54145.5 163 109697 

GTR+ 8 -49483.9 168 100417 

4 partitions: ribosomal DNA, separate codon positions for protein coding 

F81 -52199.0 134 105554 

F81+ 8 -49807.4 138 100805 

HKY -50927.6 138 103046 

HKY+ 8 -48121.3 142 97468 

GTR -50455.9 154 102241 

GTR+ 8 -47763.5 158 96890 

5 partitions: 16S, 18S, separate codon positions for protein coding 

F81 -52190.9 138 105572 

F81+ 8 -49798.1 143 100830 

HKY -50918.0 143 103070 

HKY+ 8 -48109.4 148 97496 

GTR -50160.4 163 101727 

GTR+ 8 -47769.0 168 96988 

11 partitions: 16S, 18S, separate codon positions for each gene 

F81 -52087.6 162 105573 

F81+ 8 -49742.6 173 100978 

HKY -50825.1 173 103143 

HKY+ 8 -48071.9 184 97731 

GTR -50188.9 217 102250 

GTR+ 8 -47561.4 228 97090

Table 3. List of calibration points used to date the phylogenetic tree. The nodes to which the age constraints are applied (first column) are indicated on the 

phylogram in Fig. 3. If more than one age constraint was identified for a node, these are listed as different calibrations (e.g. calibrations a1, a2 and a3 all 

apply to node a). The 'age' column indicates the type of constraint (minimum vs. maximum) and the value of the constraint (in million years). Ages of epochs 

and stages follow the International Stratigraphic Chart (ICS 2008). 

node name calibration fossil period age (my) reference 

a siphonous algae crown a1 absence of siphonous fossils Ediacaran max 635 Zhang et al. (1998) 

a2 Yakutina aciculata Middle Cambrian min 499 Korde (1957) 

a3 Chaetocladus plumula Middle Ordovician min 460.9 Whitfield (1894) 

b Batophoreae stem b1 Uncatoella verticillata Lower Devonian min 397.5 Kenrick and Li (1998) 

b2 Archaeobatophora typa Upper Ordovician min 443.7 Nitecki (1976) 

max 

b3 Uncatoella verticillata Lower Devonian 416.0 Kenrick and Li (1998) 

c Bornetelleae stem c Zittelina hispanica Hauterivian min 130.0 Masse et al. (1993) 

d Neomereae crown d1 Neomeris cretacea Albian min 99.6 Granier and Deloffre (1993) 

d2 Neomeris cretacea Barremian min 125.0 Sotak & Misik (1993) 

d3 Pseudocymopolia jurassica Portlandian min 142.0 Dragastan (1968) 

e Acetabularieae stem e1 Acicularia heberti Danian min 61.1 Morellet and Morellet (1922) 

e2 Acicularia boniae Middle Triassic min 228.7 Iannace et al. (1998) 

core Halimedineae 

Dimorphosiphon 

f stem f 

rectangulare Middle Ordovician min 460.9 Hoeg (1927) 

g Caulerpaceae stem g Caulerpa sp. Wolfcampian min 280.0 Gustavson and Delevoryas (1992) 

h Halimeda stem h1 Halimeda marondei Norian min 203.6 Flügel (1988) 

h2 Halimeda soltanensis Upper Permian min 251.0 Poncet (1989)

112 CHAPTER 6 

Appendix 

Alternative sets of calibration points used to constrain relaxed molecular clock analyses. Node ages 

were inferred for each of these conditions to evaluate the impact of some potentially erroneous 

fossil assignments. 

condition combination description 

01 a1 a2 b1 c d2 e1 f g 

h1 

moderately conservative 

02 a1 a3 b1 c d1 e1 f g 

h1 

ultraconservative youngest 

03 a1 a2 b2 c d3 e2 f g 

h2 

ultraliberal oldest 

04 a1 a2 b1 c d2 e1 f h1 moderately conservative, no Caulerpa 

05 a1 a2 b1 c d2 e1 g h1 moderately conservative, no Dimorphosiphon 

06 a1 a3 b1 c d1 e1 g h1 ultraconservative youngest, no Dimorphosiphon 

07 a1 a2 b2 c d3 e2 g h2 ultraliberal oldest, no Dimorphosiphon 

08 a1 a2 b1 c d2 e1 h1 moderately conservative, no Caulerpa, no Dimorphosiphon 

09 a2 b1 b3 c d2 e1 h1 very conservative young: strongly constrained age for crown 

Dasycladaceae based on Uncatoella 

10 a2 b1 b3 c d2 e1 h2 very conservative young: strongly constrained age for crown 

Dasycladaceae based on Uncatoella 

11 a1 a2 c d2 e1 h1 condition 08 without calibrations on node b 

As mentioned in the text, analyses were also repeated with different maximum age constraints on 

the root node to evaluate the sensitivity to this particular assumption. These analyses were variants 

of condition 08 in which the maximum age constraint a1 was replaced with 800 my and 500 my. 

The chronogram resulting from analysis 08 is shown in Fig. 4. 

The chronogram resulting from analysis 09 is shown in the online appendix (i.e. below).

A MULTI-LOCUS TIME-CALIBRATED PHYLOGENY OF THE SIPHONOUS GREEN ALGAE 113

7 

Systematics of the marine microfilamentous green algae Uronema 

curvatum and Urospora microscopica (Chlorophyta) 1 

Abstract 

The microfilamentous green alga Uronema curvatum is widely distributed along the western and 

eastern coasts of the north Atlantic Ocean where it typically grows on crustose red algae and on 

haptera of kelps in subtidal habitats. The placement of this marine species in a genus of freshwater 

Chlorophyceae had been questioned. Molecular phylogenetic analysis of nuclear-encoded small and 

large subunit rDNA sequences reveal that U. curvatum is closely related to the ulvophycean order 

Cladophorales with which it shares a number of morphological features, including a siphonocladous 

level of organization and zoidangial development. The divergent phylogenetic position of U. 

curvatum, sister to the rest of the Cladophorales, along with a combination of distinctive 

morphological features, such as the absence of pyrenoids, the diminutive size of the unbranched 

filaments and the discoid holdfast, warrants the recognition of a separate genus, Okellya, within a 

new family of Cladophorales, Okellyaceae. The epiphytic Urospora microscopica from Norway, which 

has been allied with U. curvatum, is revealed as a member of the cladophoralean genus 

Chaetomorpha and is herein transferred to that genus as C. norvegica nom. nov. 

Key words 

Chlorophyta, Cladophorales, green algae, marine, molecular phylogenetics, systematics, taxonomy, 

Ulvophyceae 

1 Accepted manuscript: Leliaert, F., J. Rueness, C. Boedeker, C. A. Maggs, E. Cocquyt, H. Verbruggen, 

and O. De Clerck. 2009. Systematics of the marine microfilamentous green algae Uronema curvatum 

and Urospora microscopica (Chlorophyta). European Journal of Phycology in press.

116 CHAPTER 7 

Introduction 

Green algae display a wide diversity of thallus organization, ranging from flagellate or coccoid unicells 

to colonial forms and various levels of multicellular organization. This morphological variation has 

been the basis for conventional green algal classification (Round 1984). For example flagellates were 

commonly grouped in the order Volvocales, coccoids in the Chlorococcales, and unbranched 

filaments in the Ulotrichales (Bold and Wynne 1985). Ultrastructural work, comparative biochemistry 

and life-history studies have demonstrated that a filamentous nature (and various other vegetative 

features) are independently derived in different lineages within the green algae (Mattox and Stewart 

1984). Molecular systematics have largely corroborated these findings, showing that convergent 

evolution is responsible for the presence of unbranched filaments in distantly related green algal 

lineages, such as the chlorophytan classes Ulvophyceae (Ulothrix, Urospora and Chaetomorpha) and 

Chlorophyceae (Microspora, Oedogonium, Uronema), and the streptophytan classes 

Klebsormidiophyceae (Klebsormidium) and Zygnematophyceae (Spirogyra and other genera) (Lewis 

and McCourt 2004, Pröschold and Leliaert 2007). Most notably, the genus Ulothrix, which is often 

regarded as the morphological archetype of the unbranched uniseriate filamentous morphology, has 

been shown to be polyphyletic, with its various members belonging to different green algal classes 

(O'Kelly 2007). 

Although the phylogenetic position of numerous unbranched filamentous species has now been 

resolved based on ultrastructural and molecular evidence (e.g. Booton et al. 1998, Leliaert et al. 

2003, O'Kelly et al. 2004), several taxa have remained largely unstudied. Amongst them are the 

marine microfilamentous species Uronema curvatum Printz and Urospora microscopica Levring. 

The genus Uronema Lagerheim (1887) is characterized by unbranched filaments of uninucleate cells 

with a single chloroplast and 1-4 pyrenoids (Chaudhary 1979). The filaments are attached by a basal 

discoid gelatinous holdfast, and apical cells are typically acuminate. Thalli reproduce asexually by a 

single zoospore (sometimes two) formed per cell. The genus is currently placed in the chlorophycean 

order Chaetophorales, based on ultrastructural evidence and 18S rDNA sequence data (Booton et al. 

1998). Uronema includes about 17 species (Guiry and Guiry 2009), all but two restricted to 

freshwater or damp terrestrial habitats. The only marine species are the North Atlantic U. curvatum 

and the south-west Pacific U. marinum Womersley, both inconspicuous algae that have probably 

passed unnoticed in many investigations. Uronema curvatum was originally described from 

Trondheim Fjord, Norway (Printz 1926), and has since been reported from scattered localities along 

the eastern and western coasts of the north Atlantic Ocean (Feldmann 1954, Rueness 1977, South 

and Tittley 1986, Rueness 1992, Kornmann and Sahling 1994, Maggs and O' Kelly 2007). The species 

grows in subtidal habitats on non-calcified crustose red algae (such as Peyssonnelia and Cruoria), 

crustose Cyanobacteria on pebbles, and on haptera of kelps. Uronema curvatum differs from the 

freshwater representatives of the genus in zoosporganial and apical cell morphology, and its generic 

placement has been debated by Rueness (1992) and Kornmann and Sahling (1994) who suggested a 

relationship with Cladophorales or Ulotrichales (Acrosiphoniales).

SYSTEMATICS OF MARINE MICROFILAMENTOUS GREEN ALGAE 117 

Figures 1-11. Uronema curvatum (= Okellya curvata, comb. nov.). Figs 1, 2. Field-collected sample, growing as 

an epiphyte on a red crust, on the haptera of Laminaria hyperborea (diameter of filaments 7-8 µm). Figs 3-11. 

Cultures. Fig. 3. Terminal sporangium with spores. Fig 4, 5. Vegetative filaments showing chloroplasts before 

and after staining with iodine solution (no pyrenoids are visible). Fig. 6. Irregular surface of chloroplast. Figs 7- 

10. Terminal sporangia with spores, some of which germinate in situ. Arrows indicate exit pore. Fig. 11. DAPIstained 

vegetative cells with two to four nuclei. 

Urospora Areschoug (1866) is a genus of cold water marine green algae characterized by an 

unbranched filamentous gametophyte composed of multinucleate cells with a parietal reticulate 

chloroplast, and a unicellular, club-shaped, uninucleate sporophyte (Codiolum phase). The uniseriate 

gametophytes are attached to the substrate by multicellular rhizoids arising from the basal cells. 

Urospora has a complex nomenclatural history and the taxonomic position of the genus has long 

been uncertain (Lokhorst and Trask 1981). The genus has been placed in the Cladophorales or 

Siphonocladales based on the multinucleate cells (Wille 1890, Rosenvinge 1893, Setchell and Gardner 

1920, Printz 1932) but is now recognized as a close relative of Acrosiphonia and Spongomorpha in the

118 CHAPTER 7 

ulvophycean order Ulotrichales based on morphological, ultrastructural and life-history features, and 

molecular data (Jorde 1933, Kornmann 1963, Floyd and O' Kelly 1984, van Oppen et al. 1995, Jónsson 

1999, Lindstrom and Hanic 2005). The genus includes about 12 species worldwide (Guiry and Guiry 

2009), four of which are common along western European shores (Lokhorst and Trask 1981). The 

epiphytic Urospora microscopica from Norway has been distinguished from other species in the 

genus by its minute filaments (Levring 1937). Since its description, the species has remained largely 

unnoticed and its systematic position uncertain (Rueness 1992, Lein et al. 1999). 

In the present study we aim to assess the phylogenetic position of the enigmatic microfilamentous 

species Uronema curvatum and Urospora microscopica by molecular phylogenetic analysis of 

nuclear-encoded small and large subunit rDNA sequences. 

Material and methods 

Specimens of Uronema curvatum, growing as epiphytes on Peyssonnelia dubyi P.L. Crouan & H.M. 

Crouan, were collected from Vega (county of Nordland, Norway) on 26 October 1990 (Rueness 1992). 

Specimens of Urospora microscopica, growing epiphytic on Cystoclonium purpureum at a depth of 3- 

5 m, were collected from Busepollen in Austevoll (county of Hordaland, Norway) in September 1994. 

Unialgal cultures of both species were obtained as described in Rueness (1992) and have been 

deposited in the Culture Collection of Algae and Protozoa (CCAP) as CCAP 455/1 (Uronema 

curvatum) and CCAP 504/1 (Urospora microscopica). Specimens were examined with a Nikon Eclipse 

TE 300 (Nikon Co., Tokyo, Japan) and Olympus BX51 (Olympus Co., Tokyo, Japan) bright field light 

microscopes. Photographs were taken with a Nikon DS-5M or Olympus E410 digital camera mounted 

on the microscope. Pyrenoids were stained with Lugol's iodine. DAPI nuclear staining was performed 

as described by Rueness (1992). 

Molecular phylogenetic analyses were based on nuclear-encoded small subunit (SSU) and partial 

large subunit (LSU) rDNA sequences. DNA extraction, PCR amplification and sequencing were 

performed as described in Leliaert et al. (2007). Taxa for which new sequences were generated are 

listed in Table S1 of the online supplementary material. Sequences have been deposited to 

EMBL/GenBank under accession numbers FN257507- FN257512. 

Two alignments were created for phylogenetic analyses. The first one was assembled to assess the 

phylogenetic position of Uronema curvatum and Urospora microscopica within the Chlorophyta. This 

alignment consisted of 30 SSU rDNA sequences, including other Uronema and Urospora 

representatives and exemplar taxa from a broad representation of chlorophytan classes for which 

SSU rDNA sequences have been deposited in GenBank. Two prasinophycean algae were used as 

outgroup taxa. Although no sequence data are available for the type species of Uronema (U. 

confervicola Lagerheim) and Urospora (U. mirabilis Areschoug), there is indirect evidence that 

Uronema belkae and Urospora neglecta (included in our phylogenetic analyses) are related to the 

types of the respective genera. Schlösser (1987) showed that the autolysin of U. confervicola reacts 

in bioassays on strains of U. belkae, suggesting that the two species are closely allied (Pröschold and 

Leliaert 2007). Urospora mirabilis is currently regarded as a synonym of Urospora penicilliformis


(Roth) J.E. Areschoug, which has been found to be related to Urospora neglecta based on 18S rRNA 

gene sequence data (Lindstrom and Hanic 2005). 

Based on the results of the phylogenetic analysis inferred from the SSU rDNA alignment, a second 

dataset of partial LSU rDNA sequences was assembled and analysed to examine the phylogenetic 

position of the two species within the Cladophorales with more confidence. In the Cladophorales, 

partial LSU rDNA sequences (first ca. 500 bp) are known to be more phylogenetically informative 

than SSU rDNA sequences (Leliaert et al. 2003). The LSU rDNA alignment consisted of 20 

cladophoralean sequences with Caulerpa and Chlorodesmis (Bryopsidales), Acrosiphonia 

(Ulotrichales) and Ulva (Ulvales) as outgroup taxa. Sequences were aligned using MUSCLE (Edgar 

2004), and visually inspected. 

Evolutionary models for the two alignments were determined by the Akaike Information Criterion in 

PAUP/Modeltest 3.6 (Posada and Crandall 1998, Swofford 2002). Both datasets were analysed with 

maximum likelihood (ML) and Bayesian inference (BI), using PhyML v2.4.4 (Guindon and Gascuel 

2003) and MrBayes v3.1.2 (Ronquist and Huelsenbeck 2003) respectively. The SSU rDNA dataset was 

analysed under a general time-reversible model with a proportion of invariable sites and gamma 

distribution split into 4 categories (GTR+I+G4). The LSU rDNA alignment was analysed under a 

general time-reversible model with gamma distribution split into 4 categories and no separate rate 

class for invariable sites (GTR+G4). BI analyses consisted of two parallel runs of four incrementally 

heated chains each, and 4 million generations with sampling every 1000 generations. A burnin 

sample of 2000 trees was removed before constructing the majority rule consensus tree. For the ML 

trees, the reliability of internal branches was evaluated with non-parametric bootstrapping (1000 

replicates). 

Results 

Morphology 

Thalli of Uronema curvatum form minute epiphytic turfs of curved, uniseriate, unbranched filaments, 

composed of 3-10 cells, 100-180 µm long (in culture, filaments may grow up to 100 cells and 700 µm 

long), with an increasing diameter towards the apex (Figs 1, 2). Thallus is dull, yellowish green in 

colour. Filaments are attached to the substrate by a basal discoid holdfast. Vegetative cells are 

subcylindrical, 3.5-6 µm in diameter at the base, increasing to 7-10 µm at the apex, 1.5-6 times as 

long as broad, up to 21 µm long. The chloroplast is parietal and lobed and occupies most of the cell 

wall (Figs 4-6); transmission electron microscopy showed that more than one chloroplast might be 

present per cell (Rueness 1992); pyrenoids are absent (Fig. 5). Cells are multinucleate, containing (1-) 

2-4 (-8) nuclei (Fig. 11). Thalli become reproductive before the filaments reach about 10 cells (in 

culture, unattached filaments may become longer). Prior to differentiation into sporangia, cells 

contain 8-16 (-32) nuclei (Rueness 1992). Zoids develop by transformation of apical and subapical 

cells into slightly swollen zoosporangia (Figs 3, 7); (8-) 16-32 zoids are formed per cell, which emerge 

through a domed pore in the upper part of the cell, on the outer face relative to the curvature of 

filaments (Figs 7, 8) (Rueness 1992). In culture, spores may germinate within the parent cell (Figs 9,

120 CHAPTER 7 

10). Filaments that were isolated into unialgal culture in 1990 have since been reproducing asexually 

by spores. 

Figures 12-21. Urospora microscopica (= Chaetomorpha norvegica, nom. nov.), culture. Fig. 12. Filament with 

basal cell with attachment disc. Fig. 13. Apical portion of filament with sporangia and lateral exit pores. Fig. 14. 

Parietal, lobed chloroplast. Fig. 15. Cell showing pyrenoids following staining with iodide solution. Figs 16, 17. 

The same filaments after DAPI-staining as seen under light field and fluorescence microscopy, showing 

multinucleate cells with four nuclei in axial arrangement (note one spore attached outside the filament and a 

few spores left in sporangium). Fig. 18. Lobed attachment disc with germinating spores. Fig. 19. Sporangia with 

exit pore (top filament) and start of exit pore formation (bottom filament, arrow) (fixed material, without cell 

contents). Figs 20, 21. Sporangia with spores, red eye spot visible (arrows).


Thalli of Urospora microscopica (Figs 12-21) form straight or curved, uniseriate, unbranched 

filaments up to 1750 µm long, composed of cylindrical cells, 10-20 µm in diameter (Fig. 12). Thallus is 

bright, grass green in colour. Filaments are attached to the substrate by a basal hyaline lobed 

holdfast (Figs 12, 18). Cells contain a parietal, lobed chloroplast with several pyrenoids (ca. 5) (Figs 

14, 15). Cells are multinucleate, containing 4 nuclei in axial arrangement (Figs 16, 17). Zoids develop 

by transformation of apical and subapical cells into zoosporangia; 10-35 zoids are formed per cell, 

which emerge through a domed pore in the middle part of the cell (sometimes or subapical or subbasal) 

(Figs 19-21). Filaments that were isolated into unialgal culture in 1994 have since been 

reproducing asexually by spores. 

Molecular phylogeny 

Specifications of the SSU and LSU rDNA sequence alignments and evolutionary models applied are 

given in Table S2 (online supplementary material). 

Phylogenetic analysis of the SSU rDNA dataset resulted in a chlorophytan tree with a poorly resolved 

backbone, in which the monophyly of the Ulvophyceae, Chlorophyceae and Trebouxiophyceae, and 

the relationships among these classes were weakly supported (Fig. 22). Even so, the phylogenetic 

positions of U. curvatum and U. microscopica could be determined with high support. Uronema 

curvatum is unrelated to the freshwater Uronema belkae (or any other member of the 

chlorophycean order Chaetophorales), but instead sister to the Cladophorales. Urospora 

microscopica is not allied with the Urospora neglecta or any other member of Ulvales but is placed 

within the Cladophorales clade. 

Phylogenetic analysis of the Cladophorales LSU rDNA alignment resulted in three well-supported 

clades, termed Cladophora, Siphonocladus and Aegagropila clades (Fig. 23). Concordant with the SSU 

rDNA tree, U. curvatum is sister to the Cladophorales with high support. U. microscopica falls within 

the Cladophora clade. It is most closely related to Chaetomorpha, although its exact phylogenetic 

position could not be determined with satisfactory statistical support. 

Discussion 

The placement of the marine species Uronema curvatum in a genus of freshwater Chlorophyceae had 

been questioned. Rueness (1992) examined the species in culture, and suggested a relationship with 

the cladophoralean genera Chaetomorpha and Rhizoclonium, or with the ulotrichalean Urospora 

based on the multinucleate cells. Kornmann and Sahling (1994) formally transferred the species to 

Urospora based on zoospore morphology, but this transfer was not widely adopted (Bartsch and 

Kuhlenkamp 2000, Nielsen and Gunnarsson 2001, Maggs and O' Kelly 2007). The present study shows 

that U. curvatum is closely allied to the ulvophycean order Cladophorales. 

U. curvatum shares a number of ecological and morphological features with the green macroalgal 

order Cladophorales. Like most members of this order, U. curvatum occurs in benthic marine coastal 

habitats. The assumption that the Cladophorales are an originally marine clade, which successfully

122 CHAPTER 7 

invaded freshwater habitats at least two times independently (Hanyuda et al. 2002), is reinforced by 

the phylogenetic position of U. curvatum, sister to the rest of the Cladophorales. 

Morphologically, U. curvatum shares the typical siphonocladous level of organization of the 

Cladophorales, i.e. multicellular thalli composed of multinucleate cells. The number of nuclei in 

cladophoralean species is highly variable and generally proportional to cell size. Most cladophoralean 

taxa have relatively large cells (ranging from several µm to several mm across) with hundreds or even 

thousands of nuclei, arranged in cytoplasmic domains (Kapraun and Nguyen 1994). The cells of U. 

curvatum typically contain 2-4 nuclei (Rueness 1992, Maggs and O' Kelly 2007), comparable to 

numbers found in Rhizoclonium riparium (Roth) Harvey, which has cell dimensionsof the same order 

of magnitude as U. curvatum (mostly 5-20 µm in diameter) (Leliaert and Boedeker 2007). 

Figure 22. ML tree of the core chlorophytes (Ulvophyceae, Trebouxiophyceae, Chlorophyceae), rooted with 

two prasinophytes, inferred from SSU rDNA sequences. The phylogenetic position of Uronema curvatum (= 

Okellya curvata, comb. nov.) and Urospora microscopica (= Chaetomorpha norvegica, nom. nov.), along with 

other members of the two genera are shown. ML bootstrap values (> 50) and BI posterior probabilities (> .90) 

are indicated at branches. The branches leading to Acetabularia and Caulerpa are scaled 50% (*) and 25% (**).


Thallus organization in the Cladophorales ranges from branched or unbranched uniseriate filaments 

to more complex architectural types (one notable exception being the coccoid Spongiochrysis in the 

Aegagropila clade, Rindi et al. 2006). Unbranched filamentous thalli, assigned to Chaetomorpha or 

Rhizoclonium, have evolved from branched forms several times independently within the 

Cladophorales (Hanyuda et al. 2002, Leliaert et al. 2003). U. curvatum thus represents another 

unbranched filamentous lineage of Cladophorales. U. curvatum attaches to the substrate by a basal 

discoid holdfast and hence differs from most Cladophorales, which are attached to the substrate by 

branched or unbranched rhizoids that develop from basal or intercalary cells. The diminutive 

Cladophora pygmaea Reinke also attaches by a similar basal discoid holdfast. Based on this feature it 

was placed in a separate section of the genus by van den Hoek (1963), but a molecular phylogenetic 

study refuted the separate placement of C. pygmaea and showed that mode of attachment is not an 

evolutionarily conserved character and has little taxonomic value above the species level (Leliaert et 

al. 2009). 

U. curvatum shares the typical zoidangial development and exit aperture of the Cladophorales: after 

vegetative growth ceases, the apical cells slightly swell and the cytoplasm is divided into zoids that 

are dispensed through a domed pore at the upper end of the cell. Some cladophoralean taxa display 

variation in zoidangial morphology. For example, in Wittrockiella (Aegagropila clade) the spores are 

released through extremely elongated exit tubes, resembling colourless hairs (Leliaert and Boedeker 

2007), and many taxa of the Siphonocladus clade have large cells that form numerous lateral exit 

pores (Hori 1994). 

The chloroplast of U. curvatum is parietal and lobed and lines most of the cell wall. In contrast, the 

cells of most other Cladophorales contain numerous chloroplasts interconnecting by delicate strands 

to form a continuous layer or a parietal network. In some taxa, such as Rhizoclonium riparium, cells 

contain a single or few parietal, lobed chloroplasts, similar to U. curvatum (Leliaert and Boedeker 

2007). Uronema curvatum differs from other Cladophorales by the lack of pyrenoids, although TEM 

observations by Rueness (1992) suggest the presence of starch grains inside the chloroplasts. In 

other Cladophorales most of the chloroplasts in a cell contain a single pyrenoid. The majority of 

species have bilenticular pyrenoids, i.e. each pyrenoid consists of two hemispheres, separated by a 

single thylakoid and each hemisphere is capped by a bowl-shaped starch grain. This pyrenoid 

structure was initially thought to be uniform within the order (Jónsson 1962, van den Hoek et al. 

1995), but several exceptions to this pattern have been reported, mainly in species of the 

Aegagropila clade (Matsuyama et al. 1998, Miyaji 1999, Hanyuda et al. 2002). 

Another marine species of Uronema, U. marinum, has been described as an epiphyte on green and 

(non-crustose) red seaweeds in shallow subtidal habitats from southern and western Australia, the 

Great Barrier Reef, Lord Howe Island, Micronesia and Hawaii (Womersley 1984, Abbott and Huisman 

2004, Kraft 2007). This species resembles U. curvatum in size, cell dimensions and mode of 

attachment, but differs in having one or two pyrenoids per cell. The taxonomic position of Uronema 

marinum is left undecided at this stage, and the name is retained pending molecular investigations. 

The micro-filamentous species, Urospora microscopica, was described from Osund, Norway, growing 

epiphytically on Nitophyllum and Cystoclonium (Levring 1937). Since its original description, the 

species has remained largely unnoticed (Lein et al. 1999). Rueness (1992), who re-examined the type 

material, found that U. microscopica differed from U. curvatum in cell dimensions, the straight

124 CHAPTER 7 

filaments, and the position of the sporangial pore, which is lateral or sub-basal in U. microscopica 

versus subapical in U. curvatum. In the present study we examined recent collections and cultures of 

U. microscopica, which made it possible to investigate this species in more detail and assess its 

phylogenetic position based on molecular data. The phylogenetic analysis clearly shows that U. 

microscopica is a member of the Cladophora clade. The species seems to be most closely related to 

Chaetomorpha, although the presence of four nuclei and sporangia with a single lateral pore would 

suggest a relationship with Rhizoclonium riparium (Leliaert and Boedeker 2007). 

Figure 23. ML tree of the Cladophorales inferred from partial LSU rDNA sequences, showing the phylogenetic 

position of Uronema curvatum (= Okellya curvata, comb. nov.) and Urospora microscopica (= Chaetomorpha 

norvegica, nom. nov.). ML bootstrap values (> 50) and BI posterior probabilities (> .90) are indicated at 

branches. 

Chaetomorpha is a marine genus of attached or unattached unbranched macro-filaments. More than 

200 species and infraspecific taxa have been described worldwide (Index Nominum Algarum), of 

which only about 50 are currently accepted (Guiry and Guiry 2009). Morphological features used to 

delimit species within the genus are growth form, cell dimensions and shape of the basal attachment 

cell. However, the extensive variability of these morphological characters depending on 

environmental conditions accounts for a great deal of taxonomic confusion, and the genus is clearly 

in need of revision (Leliaert and Boedeker 2007). 

Urospora microscopica differs from other members of the genus Chaetomorpha in the diminutive 

growth form, small cell size (10-20 µm in diameter) and low number of nuclei per cell. Most


Chaetomorpha species are much more robust, forming macroscopic thalli with cells ranging from ca. 

60 µm in diameter in C. ligustica (Kützing) Kützing to one or several mm across (e.g. C. melagonium 

(Weber & Mohr) Kützing, C. coliformis (Montagne) Kützing). Only a few other minute Chaetomorpha 

species have been described. Chaetomorpha sphacelariae Foslie (1881) is a diminutive epiphyte on 

Sphacelaria from Norway, which has remained unnoticed since its original description. 

Chaetomorpha minima Collins & Hervey (1917) has been morphologically associated with C. 

sphacelariae. This species from the north-west Atlantic Ocean and Caribbean Sea also grows 

epiphytically on algae, seagrasses and salt-marsh plants, and forms inconspicuous filaments, 

composed of long cells which are 10-27 µm across (Schneider and Searles 1991, Dawes and 

Mathieson 2008, Littler et al. 2008). Another minute species, C. recurva Scagel (1966), is found along 

the Pacific coast of North America and forms minute thalli with filaments 6-10 µm across. Given that 

morphological features, especially cell dimensions, are now known to be poor indicators of 

phylogenetic relationships in the Cladophorales (and green algae in general) (Leliaert et al. 2007), 

these species will need to be further examined using molecular tools to assess their phylogenetic 

affinity. 

Taxonomic conclusions 

The divergent phylogenetic position of U. curvatum, sister to the rest of the Cladophorales, along 

with a combination of distinctive morphological features such as the absence of pyrenoids, the 

diminutive size of the unbranched filaments and a discoid holdfast, warrants the recognition of a 

separate genus within a new family of Cladophorales: 

Okellyaceae Leliaert et Rueness, familia nov. 

Algae benthicae marinae, filamentibus simplicibus erectis curvatis, disco basali ad substratum affixae. 

Cellularum divisio intercalaris. Filamenta 3-8 cellulibus, usque 200 µm longa (in cultura usque 100 

cellulibus, 1 mm longa), diametro dilatato apicem versus. Cellulae apicales cylindricae obtusae, 

cellulae intercalares cylindricae, 3-20 µm latae, (1-) 2-4 (-8) nucleatae. Chloroplastus parietalis, 

pyrenoide absenti autem granis amyloideis. Zoosporangia apicalia et subapicalia, zoosporibus 

pluribus (8-32), e sporangioporo singulari emergentibus. Thalli epiphytici super algas crustosas in 

zona sublitorali. 

Marine benthic algae forming minute tufts of slightly curved, erect unbranched uniseriate filaments, 

attached by a basal discoid holdfast. Growth by intercalary cell divisions. Filaments composed of 3-8 

cells, up to 200 µm long (in culture sometimes up to 100 cells and 1 mm or more), diameter 

increasing towards the apex. Apical cell cylindrical with obtuse tip, intercalary cells cylindrical, 3-20 

µm in diameter, containing (1-) 2-4 (-8) nuclei. Chloroplast parietal, lacking pyrenoids but including 

starch grains. Multiple zoospores (8-32) developing by transformation of apical and subapical cells 

into zoosporangia, emerging through a single pore in the upper part of the cell. Thalli epiphytic on 

various crustose algae in the subtidal zone. 

Type genus: Okellya Leliaert et Rueness, gen. nov. 

Cum characteribus familia. Characters as for family.

126 CHAPTER 7 

Type species: Okellya curvata (Printz) Leliaert et Rueness, comb. nov. 

Basionym: Uronema curvatum Printz, Algenveg. Trondhjemsfj.: 233, pl. VII: figs 105-114 (1926). 

Etymology: named in honour of Charles J. O’Kelly for his pioneering and influential work on green 

algal systematics. 

Urospora microscopica is most closely related to Chaetomorpha (type: Chaetomorpha linum) in the 

Cladophorales and is therefore transferred to that genus. Since the combination Chaetomorpha 

microscopica has already been made by Meyer (1927) for a freshwater filamentous cladophoralean 

species (later transferred to Cladochaete and Chaetocladiella), a new species epithet must be chosen. 

Chaetomorpha norvegica Leliaert et Rueness, nom. nov. 

Basionym: Urospora microscopica Levring, Lunds Univ. Årsskr. 2, 33(8): 30, fig. 2e-k (1937). 


We are grateful to Caroline Vlaeminck for generating the sequence data. We thank Paul Goetghebeur 

for help with the Latin diagnosis. Funding was provided by FWO-Flanders (research grant G.0142.05, 

and post-doctoral fellowships to HV and FL) and the Ghent University BOF (doctoral fellowship to EC).

Supplementary material 


Table S1. Specimens for which new sequences were generated with collection data (location, collector, date of 

collection and voucher information) and EMBL accession numbers. 

Species Collection and voucher information SSU rDNA partial LSU 

rDNA 

Uronema curvatum Printz Norway: county of Nordland: Vega, subtidal, epiphytic on 

Peyssonnelia dubyi (Jan Rueness, 26 Oct 1990) 1 

Urospora microscopica Levring Norway: county of Hordaland: Busepollen in Austevoll, 

subtidal 3-5 m deep, epiphytic on Cystoclonium purpureum 

(Jan Rueness, Sep 1994) 1 

Chaetomorpha melagonium (F. 

Weber & Mohr) Kützing 

Wittrockiella lyallii (Harvey) 

Hoek, Ducker & Womersley 

Norway: Spitsbergen: Kongsfjord, subtidal, epilithic (Kai 

Bischof, 14 July 2005, L: L0793540 / A88) 

New Zealand: Fiordland: Rum River estuary, high intertidal, 

on fallen tree (Svenja Heesch, 30 Sept 2005, WELT: UPN626 

/ H67) 2 

FN257508 FN257507 

FN257510 FN257509 

FN257511 

FN257512 

1 

Cultures maintained by JR and FL, and deposited in Culture Collection of Algae and Protozoa (CCAP): 

http://www.ccap.ac.uk/ 

2 

UPN stands for “Ulva project number”: http://www.maf.govt.nz/mafnet/publications/biosecurity-technical- 

papers/ulva/ 

Table S2. Specification of the SSU and LSU rDNA sequence alignments and summary of models and model 

parameters obtained. 

SSU rDNA (Chlorophyta) LSU rDNA (Cladophorales + 

ulvophycean outgroups) 

Ingroup 28: Ulvophyceae, Chlorophyceae, 


20: Cladophorales 

Outgroup 2: Prasinophyceae 4: Ulvophyceae 

Alignment length / variable sites / 

parsimony informative sites 

Model estimated by the Akaike 

information criterion (AIC) 

1787 / 867 / 557 647 / 388 / 311 

GTR+I+G4 GTR+G4 

Estimated base frequencies (A/C/G/T) 0.24 / 0.21 / 0.28 / 0.27 0.21 / 0.24 / 0.32 / 0.23 

Estimated substitution rates 

(AC / AG / AT / CG / CT / GT) 

Among-site rate variation: proportion of 

invariable sites (I) / gamma distribution 

shape parameter (G) 

1.00 / 2.60 / 1.16 / 1.16 / 4.76 / 1.00 0.65 / 2.15 / 1.28 / 0.75 / 5.11 / 1.00 

0.27 / 0.46 0 / 0.39

8 

General discussion 

This thesis focuses on phylogenetic relationships among green algae and molecular evolutionary 

processes such as gain-loss dynamics of elongation factor genes, distribution of a non-canonical code 

and patterns in genome bias. 

Phylogeny of the green algae 

The green algae have long been recognized as a natural group, well differentiated from all other 

groups of algae by a number of shared characters, including intraplastidial starch storage, doublemembraned 

plastids containing chlorophylls a and b, stacked thylakoids, isokont zoids, and a stellate 

transition zone between the flagellar axoneme and the basal body. All these characters are also 

shared by land plants (at least by the ones with flagellate stages in their life cycles), and hence the 

close relationship between green algae and land plants has been noted for many decades. The first 

molecular phylogenetic studies based on SSU nrDNA sequences have confirmed the monophyly of 

the green plants (Sogin et al. 1986, Gunderson et al. 1987). More recently, phylogenies based on 

nuclear and plastid genes have provided clear evidence that the green plants, red algae and 

glaucophytes diverged after primary endosymbiosis of a non-photosynthetic eukaryotic host cell and 

a cyanobacterium had taken place (Rodriguez-Ezpeleta et al. 2005). 

Traditional evolutionary schemes of green algae have been largely based on thallus organization and 

it was assumed that ancestral coccoid or flagellated unicells have given rise to distinct lines of 

increasing size and complexity (Fott 1971, see Chapter 1). Since the 1970’s a vast amount of 

ultrastructural information has gradually been accumulated. The hypotheses formulated from 

comparative ultrastructure studies were highly inconsistent with the traditional systems of 

classification; they showed that various vegetative features have evolved numerous times in the 

green algae (Irvine and John 1984). Molecular phylogenetic studies, mainly based on SSU nrDNA 

sequences, by and large corroborated the ultrastructural data, showing that convergent evolution is 

responsible for the presence of near-identical morphologies in distantly related green algal lineages 

(see Chapter 7). DNA sequence data have further revolutionized our understanding of green algal 

evolution. Current hypotheses of the evolution of the Viridiplantae posit the early divergence of two 

main lineages from an ancestral green flagellated unicell (Lewis and McCourt 2004, Becker and Marin 

2009). One lineage, the Streptophyta, includes a paraphyletic assemblage of green algae and their 

descendents, the land plants. The second lineage, the Chlorophyta, includes the majority of green 

algae. This dichotomy is supported by fundamental differences in the ultrastructure of the flagellated 

cells (e.g. cruciate flagellar roots in Chlorophyta vs. unilateral flagellar root with multilayered 

structures in Streptophyta). Considerable progress has been made in clarifying the relationships 

among the streptophyte green algae and land plants based on multi-marker and genome-scale 

datasets (Parkinson et al. 1999, Turmel et al. 2003, Turmel et al. 2006, Lemieux et al. 2007, Moore et 

al. 2007, Rodriguez-Ezpeleta et al. 2007, Saarela et al. 2007). Conversely, the evolutionary history of

130 CHAPTER 8 

the Chlorophyta has received much less attention. Phylogenetic relationships among three 

chlorophytan clades (Trebouxiophyceae, Chlorophyceae and Ulvophyceae) have been the subject of 

long-standing debates, and the relationships within several major clades (e.g. the Ulvophyceae) have 

been difficult to elucidate. 

Figure 1. Phylogenetic signal of each site in the SSU alignment for the three alternative UTC topologies. Only a 

few sites contain phylogenetic signal concerning UTC topologies, i.e. colored sites, and all these sites favor a 

sister relationship between Chlorophyceae and Ulvophyceae, T(CU) topology. The phylogenetic signal of the 5% 

strongest signals is decreased to the 95 percentile of signal strength.

SSU nrDNA phylogenies 

GENERAL DISCUSSION 131 

SSU nrDNA phylogenies showed conflicting results regarding the relationships between Ulvophyceae, 

Trebouxiophyceae and Chlorophyceae (UTC classes). A majority of SSU nrDNA molecular phylogenies 

showed a sister relationship between the Trebouxiophyceae and Chlorophyceae (Friedl 1995, 

Bhattacharya et al. 1996, Krienitz et al. 2001, Lopez-Bautista and Chapman 2003). Alternatively, the 

Chlorophyceae and Ulvophyceae are resolved as sisters clades in other studies (Friedl and O'Kelly 

2002, Lewis and Lewis 2005, Watanabe and Nakayama 2007). Although the latter studies are more 

trustworthy due to a broader taxon sampling and the use of likelihood based methods with more 

realistic models of sequence evolution, these studies only included a small number of ulvophycean 

representatives, and almost never resulted in a comprehensive taxon sampling (i.e. including the 

Dasycladales, Bryopsidales, Cladophorales and Trentepohliales). 

We checked the phylogenetic signal of each site in the SSU alignment for the three alternative UTC 

topologies (Fig. 1). This shows that only a few sites contain phylogenetic signal concerning UTC 

topologies, i.e. colored sites, and that the great majority of these sites favor a sister relationship 

between Chlorophyceae and Ulvophyceae. This strong phylogenetic signal is contradictory to SSU 

phylogenies showing conflicting results. The source of these conflicting topologies can possibly be 

attributed to alignment differences and the use of other phylogenetic methods. Comparison of the 

signal maps in Fig. 1 with rate maps (Ben Ali et al. 2001, Wuyts et al. 2001) indicates that signal 

concerning the UTC relationships is mainly found in the faster-evolving parts of the molecule. 

Chloroplast phylogenomics 

Organellar genomes have been shown to be particularly useful for phylogenomic reconstruction 

because of their relatively high gene content, condensed in comparison to nuclear genomes. Also, 

organellar genes are typically single-copy, in contrast to many nuclear genes that are multi-copy in 

nature, which can have a confounding effect on phylogenetic reconstruction if paralogous copies are 

being analyzed. Although chloroplast genes have been widely employed to reconstruct phylogenetic 

relationships among relatively closely related green algae (e.g. within orders or genera, Verbruggen 

et al. 2007, De Clerck et al. 2008, Verbruggen et al. 2009), only a few studies have used chloroplast 

genes to reconstruct phylogenies across the entire green plant lineage (e.g. Daugbjerg et al. 1995). 

During the last decade a large number of complete chloroplast genomes from a wide range of green 

algae have been sequenced, and chloroplast phylogenomic studies have been valuable to resolve 

problematic relationships among green algae (Qiu et al. 2006, Jansen et al. 2007, Lemieux et al. 2007, 

Turmel et al. 2008, Turmel et al. 2009). Despite the plethora of answers these datasets have 

provided, they have not been able to resolve the branching order of the UTC classes conclusively. A 

phylogenetic analysis of 58 concatenated chloroplast genes support a sister relationship between 

Ulvophyceae and Trebouxiophyceae, while Ulvophyceae and Chlorophyceae were sisters in a 

phylogenetic analysis of seven mitochondrial genes (Pombert et al. 2004, Pombert et al. 2005). The 

latter topology is also supported by chloroplast gene order data and genomic structural features 

(shared gene losses and rearrangements within conserved gene clusters). These phylogenomic 

analyses included two Ulvophyceae, Pseudendoclonium akinetum and Oltmannsiellopsis viridis, that 

belong to or are allied with the Ulvales—Ulotrichales group, respectively. Only recently, 23

132 CHAPTER 8 

chloroplast genes of the siphonous ulvophycean seaweed Caulerpa filiformis (Bryopsidales) have 

been sequenced (Zuccarello et al. 2009). Phylogenetic analyses of these new data suggested that 

neither Ulvophyceae nor Trebouxiophyceae are monophyletic, and showed that Caulerpa is more 

closely related to the trebouxiophyte Chlorella than to the ulvophytes Oltmannsiellopsis and 

Pseudendoclonium (Zuccarello et al. 2009). Explanations for this unsuspected relationship have not 

yet been provided. Another remarkable observation is that it has proven impossible to amplify 

chloroplast genes in the Cladophorales despite considerable efforts in various labs. Chloroplast 

isolation has also proven difficult. 

Mitochondrial phylogenomics 

Complete mitochondrial genomes are available for a number of green algae (see Chapter 1), 

providing an independent opportunity to study problematic relationships among green algae. 

Phylogenetic analysis of seven mitochondrial genes supported a sister relationship between 

Ulvophyceae and Chlorophyceae (Pombert et al. 2004). Similar to chloroplast phylogenomic studies, 

this study is based on a limited sample of taxa: one prasinophyte, one trebouxiophyte, a few 

Chlorophyceae and Pseudendoclonium as the sole ulvophyte. Mesostigma was revealed as the oldest 

lineage within the Streptophyta in a phylogenetic study of 33 mitochondrial genes (Rodriguez- 

Ezpeleta et al. 2007). 

Amplifying nuclear markers: not a sinecure 

Only a few studies have employed nuclear markers other than SSU rDNA to reconstruct green algal 

phylogenies. Single gene phylogenies have been made for actin (An et al. 1999, Bhattacharya et al. 

2000), glucose-6-phosphate isomerase (Grauvogel et al. 2007), glyceraldehyde-3-phosphate 

dehydrogenase (Petersen et al. 2006, Robbens et al. 2007) and elongation factor genes (Keeling and 

Inagaki 2004, Noble et al. 2007). Recently, a multigene phylogeny (Rodriguez-Ezpeleta et al. 2007) 

has been made. In all case, these studies included only a few chlorophyte green algae and often no 

ulvophyte. 

We chose to amplify nuclear genes in the first place because phylogenetic analyses inferred from SSU 

nrDNA, chloroplast or mitochondrial genes showed conflicting results with respect to the radiation of 

UTC classes and relationships between ulvophycean orders. The amplification failure of chloroplast 

genes in Cladophorales poses an additional problem hampering the use of chloroplast genes for 

green algal phylogenetics. Furthermore, nuclear genes offer the opportunity to study multiple 

independent loci. 

We can think of four principal reasons why nuclear genes are barely used in green algal phylogenetic 

studies: (1) the limited availability of genomic data for green algae pose a restriction on the range of 

taxa which can be used directly as a data source or indirectly for primer design; (2) amplification and 

characterization of single copy nuclear genes is more difficult compared with chloroplast and SSU 

nrDNA sequences that can be relatively easily amplified with universal primers (Small et al. 2004). For 

SSU nrDNA there are numerous copies per cell present, whereas only a single copy of each singly


copy nuclear gene is present in each nucleus; (3) presence of multiple, large introns at variable 

positions hampering PCR amplification at the genomic DNA level, and (4) assessment of orthology is 

difficult due to the presence of evolutionary processes such as gene duplications, incomplete lineage 

sorting and lateral gene transfer. 

We anticipated all these challenges and have succeeded to amplify ten nuclear markers among the 

different green algal classes: actin, glucose-6-phosphate isomerase (GPI), glyceraldehyde-3phosphate 

dehydrogenase (GapA), oxygen-evolving enhancer protein (OOE1), 40S ribosomal protein 

S9, 60S ribosomal proteins L3 and L17, histone and elongation factor-1 alpha (EF-1α) or elongation 

factor-like (EFL). To overcome the limited availability of genomic data among green algae, an EST 

library for the siphonocladous ulvophyte Cladophora coelothrix was generated. The obtained EST 

sequences considerably contributed to our success since they allowed primer design and successful 

amplification of OOE1, three ribosomal proteins, histone and EF-1α genes among green algae. 

We have not been able to amplify nuclear genes starting from DNA material, most likely due to the 

presence of large introns (e.g. observed in available actin genes). Therefore, we worked on mRNA of 

actively dividing algal cells. Amplification of the targeted nuclear genes may have been facilitated by 

their high expression and mRNA levels. Actin is the monomeric subunit of microfilaments, one of the 

three major component of the cytoskeleton, which provides cell shape and mechanical support and 

which is involved during intracellular transport and cell division. Histones are the chief protein 

components of chromatin, act as spools around which DNA winds and play an important role in gene 

regulation. Elongation factor-1 alpha (EF-1α) and elongation factor-like (EFL) are two key genes of the 

translational apparatus. GapA is an important enzyme of the Calvin cycle and glycolysis. Glucose-6phosphate 

isomerase (GPI) is an essential enzyme for gluconeogenesis and glycolysis. Oxygenevolving 

enhancer protein (OOE1) is an auxiliary component of the photosystem II manganese 

cluster. Eukaryotic ribosomes are made of four ribosomal RNAs and approximately 80 ribosomal 

proteins, of which we amplified three (40S ribosomal protein S9 and 60S ribosomal proteins L3 and 

L17). 

A phylogenetic tree for each individual gene was made to check the orthology assumption. Histone 

genes were excluded from our final concatenated dataset because the orthology assumption was 

violated for the phylogenetic tree inferred from this gene. This is consistent with the observation that 

histone genes are highly duplicated across genomes (Nei and Rooney 2005; Wahlberg and Wheat 

2008). Elongation factor genes EF-1α and EFL were also excluded from the concatenated alignment 

because of their mutually exclusive distribution in the green plant lineage (Cocquyt et al. 2009). All 

other individual gene trees did not violate the orthology assumption, although some of the genes 

were known to be duplicated in the green plant lineage (e.g. GapA/B, Petersen et al. 2006). 

Conventional actin genes form complex gene families in various groups of complex multicellular 

organism (e.g., animals and land plants), but are single copy in most green algae. The common 

ancestor of the Viridiplantae most likely contained a single actin gene, followed by independent 

duplications of actin genes in the Ulvophyceae: Acetabularia cliftonii contains 3 actin genes, and in 

the Trebouxiophyceae: Nannochloris maculate, N. atomus and Chlorella vulgaris each contain 2 actin 

genes (An et al. 1999, Yamamoto et al. 2001, Yamamoto et al. 2003). Despite the presence of 

multiple actin gene duplications events in green algae, we believe that actin is a useful marker for 

deep phylogenetic reconstruction because these duplications appear to have occurred relatively

134 CHAPTER 8 

recently, often within a single species (e.g. several Nannochloris species only contain a single actin 

gene). The Actin Related Proteins Annotation server (ARPAnno) was used to check if all actin 

sequences were conventional actins (Goodson and Hawse 2002, Muller et al. 2005). Land plants have 

cytosolic and plastid glucose-6-phoshate isomerase (GPI). The former is thought to be vertically 

inherited while the latter is of cyanobacterial origin. The chlorophytes Chlamydomonas and Volvox 

contain a single, nuclear encoded GPI that is related to the cytosolic GPI of land plants, which 

subsequently acquired a transit peptide for transport to the plastids (Grauvogel et al. 2007). All our 

GPI sequences are related to the cytosolic GPI genes of land plants and to the plastid-targeted GPI 

gene of Chlamydomonas and Volvox. Glyceraldehyde-3-phosphate dehydrogenases (GAPDH) are 

prominent examples of homologous isoenzymes. In green algae, there are several nuclear encoded 

GAPDHs with different evolutionary origins. Glycolytic GapC genes probably have a proteobacterial 

(mitochondrial) origin, whereas photosynthetic GapA was acquired from the cyanobacterial ancestor 

of the chloroplast. Both genes have been duplicated: GapCp and GapB, respectively. GapB originated 

from a GapA gene duplication, but clearly differs from GapA by the presence of a specific C-terminal 

extension and several GapB specific amino acid insertions and positions. GapB is found in 

Streptophytes and in the genome of the unicellular prasinophyte Ostreoccoccus, but is absent in the 

genomes of Chlamydomonas and Volvox (Petersen et al. 2006, Robbens et al. 2007). For all green 

algae sequenced in this study, we only retained homologous GapA sequences, which are clearly 

different from GapC sequences. GapB sequences were not amplified in any of the sampled green 

algae. 

Resolving green algal phylogenies: contribution of nuclear genes 

In our opinion, the progress in our understanding of the phylogeny of green algae achieved in this 

study is due to (1) a good balance between taxon and gene sampling; (2) the use of model-based 

techniques (ML and BI), paying careful attention to the selection of suitable partitioning strategies 

and models of sequence evolution; and (3) the removal of fast-evolving sites to improve phylogenetic 

signal. The dataset on which our analyses were based consisted of seven single-copy nuclear markers 

(actin, GPI, GapA, OOE1, 40S ribosomal protein S9 and 60S ribosomal proteins L3 and L17), SSU 

nrDNA and two plastid genes (rbcL and atpB) for 43 taxa representing the major lineages of the 

Viridiplantae. 

We obtained high support across the topology of the Viridiplantae (see Chapter 2). We reveal the 

monophyly of the UTC classes and provided evidence for a sister relationship between the 

Ulvophyceae and Chlorophyceae with high support. We also inferred the relationships among the 

Ulvophyceae, which was found to consist of two main clades. The first clade contains the orders 

Ulvales and Ulotrichales. The second clade includes the early diverging genus Ignatius and a clade 

comprising the orders Trentepohliales, Bryopsidales, Dasycladales and Cladophorales, along with 

Blastophysa. In this clade, the Bryopsidales and Dasycladales are sisters, Blastophysa is most closely 

related to the Cladophorales, while the phylogenetic position of the Trentepohliales remains 

uncertain.


Figure 2. Substitution rates for SSU nrDNA, plastid and nuclear protein coding genes, measured as average 

substition rate in a sliding window across the alignment were calculated with HyPhy. Average substitution rates 

calculated only using variable sites (thick lines) are higher than average substitution rates based on all sites 

(thin lines) because the latter includes a number of constant sites that decrease the average substitution rate. 

(A) Average substitution rates based on the complete alignment. Nuclear genes have the highest substitution 

rates (the end of 60S ribosomal protein L3 is extremely fast), followed by plastid genes. The substitution 

rates for SSU nrDNA are generally low, except for a few positions. 

(B) Average substitution rates based on the 25% site stripped dataset, when looking at the rates of variable 

sites, average substitution rates for plastid and nuclear genes are higher than those for SSU nrDNA. 

Removal of fast-evolving sites: site stripping 

The rationale of removing fast-evolving sites (site stripping) is to remove noise from the data by 

removing those sites that are most likely to contain homoplasy and focusing on the more informative 

slow-evolving positions for phylogeny reconstruction (Philip et al. 2005). The usefulness of site 

stripping has been demonstrated in a number of studies that aimed to resolve deep nodes, for 

example a phylogenomic study of bilaterian animals (Delsuc et al. 2005). ML analysis of the complete 

bilaterian dataset (almost 150 genes) support the Coelomata hypothesis (arthropods + 

deuterostomes), most likely due to a long branch attraction artifact between the fast-evolving 

nematodes and the distant fungal outgroup. Progressive removal of fast-evolving sites resulted in 

increasing support for the Ecdysozoa hypothesis (arthropods + nematodes). This study showed that 

even advanced phylogenetic methods such as ML can be misled by differences in evolutionary rates 

among species. 

An analysis of sitewise substitution rates in our complete alignment shows that nuclear genes evolve 

fastest, followed by plastid genes and SSU nrDNA, which generally has very low substitution rates 

(Fig. 2 A). This is consistent with Fig. 1, which shows that the phylogenetic signal at many sites of the 

SSU nrDNA molecule is too low for resolving deep phylogenetic relationships. On the other hand, 

phylogenetic signal in some parts of the nuclear genes may be lost due to high sitewise substitution 

rates (e.g. end of 60S ribosomal protein L3). To counteract this erosion of ancient phylogenetic signal 

in our dataset, we removed the 25% fastest-evolving sites to increase the signal to noise ratio. This 

did not change phylogenetic relationships but it improved the phylogenetic signal for the branching

136 CHAPTER 8 

order of the UTC classes and among the Ulvophyceae. The inferred relationships between the 

radiation of the UTC classes and the fast-evolving Bryopsidales—Dasycladales—Cladophorales orders 

are thus stable even when only the most reliable phylogenetic characters are used. Figure 2 B shows 

that for the 25% site-stripped dataset sitewise substitution rates for variable sites are higher in 

plastid and nuclear genes than in SSU nrDNA. Likewise plastid and nuclear genes may contain more 

phylogenetic signal than SSU nrDNA. 

The comparison of the signal maps in Fig. 1 with rate maps (Ben Ali et al. 2001; Wuyts, Van de Peer, 

and Wachter 2001) indicates that signal concerning the UTC relationships is mainly found in the 

faster-evolving parts of the molecule. Because our site stripping approach removes these fastevolving 

sites, one may expect that most sites containing phylogenetic signal concerning the UTC 

relationships will be removed from the SSU nrDNA alignment. However, about 60% of the sites 

containing phylogenetic signal concerning the UTC relationships are retained in the 25% site stripped 

dataset. This relatively high number of retained fast-evolving sites can be due to the lower sitewise 

substitution rates for SSU nrDNA compared to plastid and nuclear genes (Fig. 2). Nevertheless, SSU 

nrDNA alone cannot resolve the UTC relationship but it is generally accepted that the information 

contained in a single genes is not enough to resolve deep relationships (Philippe et al. 2005), which is 

one of the reason why we based this study on 10 markers and not on a single gene. 

Figure 3. Difference between removal of the 25% fastest-evolving sites (site stripping) and random removal of 

25% of the sites. 

(A) Rate-smoothed tree with indication of the epoch in which the relationships of interest are situated (gray 

box across figure). 

(B) The strength of the phylogenetic signal in the site-stripped alignment (black line) is much higher than that 

of the randomly stripped alignments (blue band) in the epoch of interest. The gray area represents the 95 

percentile of signal strength (measured as average bootstrap values in a sliding window across the tree) 

obtained from the 100 alignments from which 25% of the sites were removed at random.


Our site stripping approach is also validated by a comparison between random removal of sites and 

selective removal of fast-evolving site. This exercise shows that only in the latter case phylogenetic 

signal is increased (Fig. 3). Random removal of one fourth of the positions in the alignment results in 

a global decrease of phylogenetic signal, as measured by bootstrap support. 

Dating the green tree of life 

To interpret the ecological, morphological and cytological diversification of the green algae in a 

geological timeframe, we dated our phylogeny of the Viridiplantae using relaxed molecular clock 

methods and a combination of the fossil record and previous molecular clock results (Fig. 4). We 

used the lognormal clock relaxation approach (Thorne et al. 1998) in a Bayesian framework as 

implemented in PhyloBayes (Lartillot et al. 2007). This method assumes autocorrelated rates of 

molecular evolution, more specifically that the logarithm of the rate of molecular evolution can be 

described by a Brownian motion process. This model indicates a rapid radiation of the UTC classes 

within a timeframe of ca. 20 my during the first period of the Neoproterozoic (between 900 my [852- 

957] and 881 my [835-935]). This rapid radiation of the UTC classes complicates the interpretation of 

ecological diversifications. A broader taxon sampling within the Trebouxiophyceae and 

Chlorophyceae may allow predictions about the number of times transitions from marine to 

freshwater and terrestrial habitats occurred. The inferred radiation time for the UTC classes precedes 

the Cryogenian period (850-635 my), in the middle of the Neoproterozoic, which is known as a period 

of severe glaciations during which ice sheets reached the equator, events better known as “Snowball 

Earth”. 

We show a Neoproterozoic diversification of all ulvophycean orders in a timeframe of ca. 130 my 

(between 826 my [792-870] and 596 my [500-686]). Originating from an ancestral marine 

prasinophyte, our phylogenetic tree indicates that transitions to freshwater and terrestrial habitats 

not only occurred in the Chlorophyceae and Trebouxiophyceae, but also several times independently 

within the predominantly marine Ulvophyceae (Lopez-Bautista et al. 2007, Becker and Marin 2009). 

A number of representatives within the Ulvales, Ulotrichales and Cladophorales are adapted to 

freshwater environments. Ignatius, all members of the Trentepohliales and two species of 

Cladophorales (Cladophorella and Spongiochrysis) are growing in subaerial habitats (Fritsch 1944, 

Lopez-Bautista and Chapman 2003, Rindi et al. 2006, Watanabe and Nakayama 2007). 

Multicellularity evolved independently in the UTC classes. Within the Ulvophyceae, multicellularity 

arose independently in the Ulvales—Ulotrichales. In chapter 6, we calibrated a five-locus phylogeny 

of the siphonous orders Dasycladales and Bryopsidales using relaxed molecular clock methods 

calibrated with the fossil record. These models indicate a late Neoproterozoic or early Cambrian 

origin of the Dasycladales and Bryopsidales (571 million years [628–510]) and a diversification of the 

orders into their families during the Paleozoic. The siphonous thallus structure, which is essentially 

composed of a single giant cell containing numerous nuclei, is most likely derived from a 

multinucleate and multicellular common ancestor of the Bryopsidales—Dasycladales—Cladophorales 

(BCD clade). Additional support for the evolution of siphonous algae from a multicellular ancestor is 

provided by the occurrence of a cross wall at the base of each reproductive structure in one of the 

two major bryopsidalean lineages (the Bryopsidineae).

138 CHAPTER 8 

Figure 4. Chronogram of the Viridiplantae. Node ages were inferred using Bayesian inference assuming a 

relaxed molecular clock and a set of node age constraints derived from the fossil record and from Verbruggen 

et al. (2009). These calibration points are indicated on the tree. Values at nodes indicate average node ages and 

bars represent 95% confidence intervals.


In the light of this interpretation, a reinterpretation of the fossil Proterocladus seems in place. This 

fossil, which is characterized by large cells with cross-walls, has been interpreted as an ancestor of 

the Cladophorales (Butterfield et al. 1994). Our results now suggest that the ancestors of the BCD 

clade also featured this morphology and we propose that Proterocladus should be regarded as an 

ancestor of the BCD clade rather than of the Cladophorales. This is the interpretation we have used 

in our molecular clock study. 

Phylogenetic position of Blastophysa warrants the recognition of a new order 

The enigmatic, endophytic green alga Blastophysa, is usually placed in the cladophoralean family 

Chaetosiphonaceae based on morphological, ultrastructural, cytological, and biochemical features 

(O'Kelly and Floyd 1984). However, the phylogenetic position of Blastophysa has been doubted since 

its establishment. First, the monophyly of the family Chaetosiphonaceae is questionable due to 

differences in ultrastuctural features of motile cells between Blastophysa and Chaetosiphon. 

Ultrastructural features of the motile cells of Blastophysa are almost identical to these of the 

Cladophorales and Dasycladales and have less features in common with the Bryopsidales (Chappell et 

al. 1991). Blastophysa has often been allied with the Bryopsidales in taxonomic treatments of 

seaweed flora’s (e.g. Burrows 1991, Brodie et al. 2007, Kraft 2007). Although, Parker (1970) 

suggested that Blastophysa is not related to the Bryopsidales based on the presence of cellulose I in 

the cell walls (Burrows 1991). Finally, a phylogeny based on SSU and LSU nrDNA placed Blastophysa 

at the base of the Cladophorales, Dasycladales and Bryopsidales (Zechman et al. 1990). 

Our multi-locus phylogeny of the Viridiplantae (Chapter 2), shows a sister relation between 

Blastophysa and the order Cladophorales. The unique ultrastructural features, along with the 

divergent phylogenetic position of Blastophysa would warrant the recognition of a separate order of 

Ulvophyceae. 

Molecular evolution in green algae 

There are several indications that profound changes to the translational system of the Ulvophyceae 

have occurred, more specifically in the siphonous and siphonocladous seaweed Bryopsidales, 

Cladophorales and Dasycladales (BCD clade) and their sister lineages Ignatius and Trentepohliales. 

First, we showed that their elongation factors, key genes of the translational apparatus that bind to 

the tRNAs before they attach to the ribosome, are fundamentally different than those of other green 

algae. In Chapter 3, we showed that elongation factor-1 alpha (EF-1α) and elongation factor-like 

(EFL) have an almost mutually exclusive distribution in the green plant lineage. The Streptophyta 

posses EF-1α except Mesostigma, which has EFL. All Chlorophyta encode EFL except the 

Ulvophyceae of the BCD clade and Ignatius. Due to amplification failure we could not find out which 

gene is present in Trentepohlia but its phylogenetic position strongly suggests the presence of EF-1α. 

The fact that the BCD clade and Ignatius use another elongation factor gene is in itself an indication 

for a change in the translational system. In Chapter 4, we discussed the presence and distribution of 

a non-canonical code, which translates stop codons TAG and TAA to glutamine. This non-canonical

140 CHAPTER 8 

code is found in Dasycladales, Cladophorales + Blastophysa and Trentepohliales. The remaining 

Ulvophyceae and green algae use the standard code. The presence of a different code requires 

changes in translation termination, tRNA-species and tRNA synthetase specificity, all factors 

influencing the translational system. In Chapter 5, we analyzed the evolution of codon usage bias 

and showed that codon usage is less biased in multicellular and macroscopic species compared to 

small unicells, indicating a more uniform use of synonymous codons in the former organisms. Within 

the Chlorophyta, the prasinophytes, Trebouxiophyceae and Chlorophyceae have strong codon usage 

bias, whereas codon usage bias is low in all Ulvophyceae, except for the small colonial Ignatius. The 

low codon usage bias observed for most Ulvophyceae indicates a uniform use of synonymous 

codons, implying that the tRNA pool is diverse, which impacts the efficiency of the translational 

system. In Chapter 5, we also showed that GC content shows a similar trend as codon usage bias, the 

Ulvophyceae having lower GC content than the prasinophytes, Trebouxiophyceae and 

Chlorophyceae. This trend is most pronounced in the Trentepohliales, Dasycladales and Bryopsidales. 

The link with the translational system is not directly clear but a low GC content may facilitate 

denaturation of the DNA during replication and transcription. Finally, rates of molecular evolution 

are high in the Trentepohliales and BCD clade as indicated by the long root-to-tip path lengths in our 

phylogeny. These high rates of molecular evolution suggest that more mutations are fixed in these 

groups. It is striking that root-to-tip paths are particularly longer for the Trentepohliales and BCD 

clade in SSU nrDNA phylogenies compared to plastid gene phylogenies (data not shown). Because 

SSU nrDNA is the structural RNA for the small component of eukaryotic cytoplasmic ribosomes, this 

provides another indication for changes in the translational apparatus. Taken together, this evidence 

points to profound changes in the translational apparatus. 

Perspectives for future research 

The approach applied in this study started with mining the available green algal genomic data (e.g. 

EST, complete genomes) and a newly generated cDNA library of a siphonocladous ulvophyte. This 

allowed designing primers that could amplify nuclear genes in a wide range of green algae. 

Amplification starting from genomic DNA was difficult, if not impossible, mainly due to the 

omnipresence of introns. The introns hampered the design of primers applicable to a wide range of 

taxa at the genomic DNA level because length and position of introns are highly variable between 

different green algal species (e.g. actin). De novo sequencing of nuclear genomes of a selection of 

green algae (e.g. with Roche 454 technology) could alleviate this need by permitting the 

identification of intronless genes and regions suitable for exon-primed intron crossing (EPIC). 

Next-generation sequencing technologies (NGST) parallelize the sequencing procedure, allowing 

rapid sequencing of billions of nucleotides simultaneously (e.g. Roche 454 technology, Rokas and 

Abbot 2009). NGST data for a balanced set of green algae is likely to provide a major boost to the 

assembly of the green tree of life. This brute-force approach is probably more efficient than primer 

design and amplification of nuclear genes as was attempted in this study. In addition to serving the 

resolution of the tree of life, NGST data can also be used to explore several other aspects of genome 

evolution. NGST data are appropriate to evaluate the importance and relative contributions of 

processes such as lineage sorting, hybridization and lateral gene transfer in shaping the respective


genomes (Rokas and Abbot 2009). Also the origin of new genes or gene clusters and the evolution of 

introns and repetitive elements can be studied. 

NGST data for Ulvophyceae, especially siphonous and siphonocladous species, as well as 

Trebouxiophyceae would be particularly useful since genomic data for these groups are lacking or 

sparse. Most available green algal genomic data come from prasinophytes and Chlorophyceae, e.g. 

the recently released or nearly finished genomes of Chlamydomonas and Volvox (Chlorophyceae) 

and Bathycoccus and Micromonas (prasinophytes). 

With respect to my study, NGST data would allow testing the reliability of the observed differences in 

codon usage bias and GC content among green algae (Chapter 5). A comparison of genes that 

regulate translation (e.g. eukaryotic release factor, glutamine tRNA genes, genes responsible for 

nonsense-mediated mRNA decay) among the various ulvophyte lineages would be particularly 

interesting because this will further increase our knowledge about the profound changes in 

translation machinery that occurred in these lineages as is indicated by the presence of a noncanonical 

genetic code, a different elongation factor gene, fast rates of molecular evolution 

especially of the ribosomal RNA (SSU nrDNA) and a more balanced codon usage. 

Despite that NGST data for a balanced set of green algae would provide a lot of useful information, 

continuing the approach followed in this study can in the short term increase our knowledge about 

green algal phylogenetic. In a next stage it is advisable to incorporate more prasinophytess, 

Trebouxiophyceae and Chlorophyceae. A broader taxon sampling within the Trebouxiophyceae and 

Chlorophyceae would allow definitive conclusion about the monophyletic nature of these classes and 

could demonstrate whether the relationship between the UTC classes is robust to differences in 

taxon sampling. Although the class Ulvophyceae was covered by 21 species, also here improvements 

in taxon sampling can be beneficial. For example, Oltmannsiellopsis is considered to be an ulvophyte 

based on SSU nrDNA and complete chloroplast genome sequences (Pombert et al. 2006) despite 

having ultrastructural features reminiscent of prasinophytes, especially of the genus Tetraselmis 

(O’Kelly personal communication). Incorporation of Oltmannsiellopsis would therefore provide 

further insight into the evolution of the ulvophyte lineages. 

With respect to the non-canonical genetic code in some ulvophyceaen lineages, studying 

selenoproteins could also prove interesting. The amino acid selenocysteine is encoded by the stop 

codon TGA in conjunction with a selenocysteine insertion sequence (SECIS) element in the 3’UTR 

(Lobanov et al. 2007). The non-canonical code of some Ulvophyceae involves the reassignment of 

stop codons TAG and TAA to glutamine, leaving TGA as the only stop codon. Because the 

Ulvophyceae only use TGA as stop codon, the last stop codon in a selenoprotein should still 

terminate translation while the in-frame TGA codons should encode selenocysteine in the presence 

of a SECIS element. In this context, it would be useful to know whether selenoproteins are converted 

back to conventional cysteine-containing proteins in Ulvophyceae with a non-canonical code.

References 

Abascal F, Zardoya R, and Posada D. 2005. ProtTest: selection of best-fit models of protein evolution. 

Bioinformatics 21:2104-2105. 

Abbott IA and Huisman JM. 2004. Marine green and brown algae of the Hawaiian Islands. Bishop 

Museum Press, Honolulu. 

Adl SM, Simpson AGB, Farmer MA, Andersen RA, Anderson OR, Barta JR, Bowser SS, Brugerolle G, 

Fensome RA, Fredericq S, James TY, Karpov S, Kugrens P, Krug J, Lane CE, Lewis LA, Lodge J, 

Lynn DH, Mann DG, McCourt RM, Mendoza L, Moestrup Ø, Mozley-Standridge SE, Nerad TA, 

Shearer CA, Smirnov AV, Spiegel FW, and Taylor M. 2005. The new higher level classification 

of eukaryotes with emphasis on the taxonomy of protists. Journal of Eukaryotic Microbiology 

52:399-451. 

Akashi H. 1995. Inferring weak selection from patterns of polymorphism and divergence at silent 

sites in Drosophila DNA. Genetics 139:1067-1076. 

An SS, Mopps B, Weber K, and Bhattacharya D. 1999. The origin and evolution of green algal and 

plant actins. Molecular Biology and Evolution 16:275-285. 

Andersen R. 2005. Algal culturing techniques. Elsevier Academic Press. 

Andersson JO, Sjogren AM, Davis LAM, Embley TM, and Roger AJ. 2003. Phylogenetic analyses of 

diplomonad genes reveal frequent lateral gene transfers affecting eukaryotes. Current 

Biology 13:94-104. 

Andersson JO. 2005. Lateral gene transfer in eukaryotes. Cellular and Molecular Life Sciences 

62:1182-1197. 

Angellotti MC, Bhuiyan SB, Chen G, and Wan XF. 2007. CodonO: codon usage bias analysis within and 

across genomes. Nucleic Acids Research 35:W132-W136. 

Archibald JM. 2005. Jumping genes and shrinking genomes - Probing the evolution of eukaryotic 

photosynthesis with genomics. IUBMB Life 57:539-547. 

Archibald JM. 2008. Plastid evolution: Remnant algal genes in ciliates. Current Biology 18:R663-R665. 

Areschoug JE. 1866. Observationes phycologicae. Particula prima. De Confervaceis nonnullis. Nova 

Acta Regiae Societatis Scientiarum Upsaliensis Ser 3, 6(2):1-26. 

Badour SS. 1981. The inorganic carbon requirements of Chlamydomonas segnis (Chlorophyceae) for 

cell development in synchronous cultures. Journal of Phycology 17:293-299. 

Baker KE and Parker R. 2004. Nonsense-mediated mRNA decay: terminating erroneous gene 

expression. Current Opinion in Cell Biology 16:293-299. 

Baldauf SL, Palmer JD, and Doolittle WF. 1996. The root of the universal tree and the origin of 

eukaryotes based on elongation factor phylogeny. Proceedings of the National Academy of 

Sciences of the United States of America 93:7749-7754. 

Baldauf SL and Doolittle WF. 1997. Origin and evolution of the slime molds (Mycetozoa). Proceedings 

of the National Academy of Sciences of the United States of America 94:12007-12012. 

Baldauf SL. 2008. An overview of the phylogeny and diversity of eukaryotes. Journal of Systematics 

and Evolution 46:263-273. 

Barker D, Meade A, and Pagel M. 2007. Constrained models of evolution lead to improved prediction 

of functional linkage from correlated gain and loss of genes. Bioinformatics 23:14-20.

144 REFERENCES 

Bartsch I and Kuhlenkamp R. 2000. The marine macroalgae of Helgoland (North Sea): an annotated 

list of records between 1845 and 1999. Helgoland Marine Research 54:160-189. 

Becker B and Marin B. 2009. Streptophyte algae and the origin of embryophytes. Annals of Botany 

103:999–1004. 

Beier H and Grimm M. 2001. Misreading of termination codons in eukaryotes by natural nonsense 

suppressor tRNAs. Nucleic Acids Research 29:4767-4782. 

Beltran M, Jiggins CD, Brower AVZ, Bermingham E, and Mallet J. 2007. Do pollen feeding, pupalmating 

and larval gregariousness have a single origin in Heliconius butterflies? Inferences 

from multilocus DNA sequence data. Biological Journal of the Linnean Society 92:221-239. 

Ben Ali A, De Baere R, Van der Auwera G, De Wachter R, and Van de Peer Y. 2001. Phylogenetic 

relationships among algae based on complete large-subunit rRNA sequences. International 

Journal of Systematic and Evolutionary Microbiology 51:737-749. 

Berger S and Liddle LB. 2003. The life cycle of Acetabularia (Dasycladales, Chlorophyta): textbook 

accounts are wrong. Phycologia 42:204-207. 

Berney C and Pawlowski J. 2006. A molecular time-scale for eukaryote evolution recalibrated with 

the continuous microfossil record. Proceedings of the Royal Society B-Biological Sciences 

273:1867-1872. 

Bhattacharya D, Stickel SK, and Sogin ML. 1993. Isolation and molecular phylogenetic analysis of actin 

coding regions from Emiliania huxleyi, a prymnesiophyte alga, by reverse transcriptase and 

PCR methods. Molecular Biology and Evolution 10:689-703. 

Bhattacharya D, Friedl T, and Damberger S. 1996. Nuclear-encoded rDNA group I introns: Origin and 

phylogenetic relationships of insertion site lineages in the green algae. Molecular Biology and 

Evolution 13:978-989. 

Bhattacharya D, Weber K, An SS, and Berning-Koch W. 1998. Actin phylogeny identifies Mesostigma 

viride as a flagellate ancestor of the land plants. Journal of Molecular Evolution 47:544-550. 

Bhattacharya D, Aubry J, Twait EC, and Jurk S. 2000. Actin gene duplication and the evolution of 

morphological complexity in land plants. Journal of Phycology 36:813-820. 

Bold HC and Wynne MJ. 1985. Introduction to the Algae, 2nd edition. Prentice-Hall, New Jersey. 

Booton GC, Floyd GL, and Fuerst PA. 1998. Origins and affinities of the filamentous green algal orders 

Chaetophorales and Oedogoniales based on 18S rRNA gene sequences. Journal of Phycology 

34:312-318. 

Brinkmann H, Van der Giezen M, Zhou Y, De Raucourt GP, and Philippe H. 2005. An empirical 

assessment of long-branch attraction artefacts in deep eukaryotic phylogenomics. Systematic 

Biology 54:743-757. 

Brodie J, Maggs CA, and John DM. 2007. Green seaweeds of Britain and Ireland. British Phycological 

Society, London. 

Brown JM and Lemmon AR. 2007. The importance of data partitioning and the utility of bayes factors 

in Bayesian phylogenetics. Systematic Biology 56:643-655. 

Bulmer M. 1991. The selection-mutation-drift theory of synonymous codon usage. Genetics 129:897- 

907. 

Burrows EM. 1991. Seaweeds of the British Isles. Natural History Museum Publications, London. 

Butterfield NJ, Knoll AH, and Swett K. 1994. Paleobiology of the Neoproterozoic Svanbergfjellet 

formation, Spitsbergen. Fossils and Strata 34:1-84.

Castresana J. 2000. Selection of conserved blocks from multiple alignments for their use in 

phylogenetic analysis. Molecular Biology and Evolution 17:540-552. 

Cavalier-Smith T. 1981. Eukaryote kingdoms: Seven or nine? Biosystems 14:461-481. 

REFERENCES 145 

Chappell DF, O'Kelly CJ, and Floyd GL. 1991. Flagellar apparatus of the biflagellate zoospores of the 

enigmatic marine green alga Blastophysa rhizopus. Journal of Phycology 27:423-428. 

Chaudhary BT. 1979. Some observations on the morphology, reproduction and cytology of the genus 

Uronema Lagh. (Ulotrichales, Chlorophyceae) (Note). Phycologia 18:299-302. 

Chen SL, Lee W, Hottes AK, Shapiro L, and McAdams HH. 2004. Codon usage between genomes is 

constrained by genome-wide mutational processes. Proceedings of the National Academy of 


Chisholm JRM, Dauga C, Ageron E, Grimont PAD, and Jaubert JM. 1996. 'Roots' in mixotrophic algae. 

Nature 381:382-382. 

Cocquyt E, Verbruggen H, Leliaert F, Zechman FW, Sabbe K, and De Clerck O. 2009. Gain and loss of 

elongation factor genes in green algae. BMC Evolutionary Biology 9:39. 

Cocquyt E, Gile GH, Leliaert F, Verbruggen H, Keeling P, and De Clerck O. in prep. Complex 

phylogenetic distribution of a non-canonical genetic code in green algae. 

Cocquyt E, Verbruggen H, Leliaert F, and De Clerck O. submitted. Ancient relationships among green 

algae inferred from nuclear and chloroplast genes. 

Cohen J and Adoutte A. 1995. Why does the genetic code deviate so easily in ciliates? Biology of the 

Cell 85:105-108. 

Collins FS and Hervey AB. 1917. The algae of Bermuda. Proceedings of the National Academy of 

Sciences of the United States of America 53. 

Coppejans E. versie 1998-1999. Syllabus Morfologie en Systematiek van de “Lagere Planten”. Partim 

Algologie. Universiteit Gent. 

Cutter AD, Wasmuth JD, and Blaxter ML. 2006. The evolution of biased codon and amino acid usage 

in nematode genomes. Molecular Biology and Evolution 23:2303-2315. 

Daugbjerg N, Moestrup Ø, and Arctander P. 1995. Phylogeny of genera of Prasinophyceae and 

Pedinophyceae (Chlorophyta) deduced from molecular analysis of the rbcL gene. 

Phycological Research 43:203–213. 

Dawes CJ and Mathieson AC. 2008. The seaweeds of Florida. University Press of Florida, Gainesville, 

Florida. 

De Clerck O, Verbruggen H, Huisman JM, Faye EJ, Leliaert F, Schils T, and Coppejans E. 2008. 

Systematics and biogeography of the genus Pseudocodium (Bryopsidales, Chlorophyta), 

including the description of P. natalense sp nov from South Africa. Phycologia 47:225-235. 

de Koning AP, Noble GP, Heiss AA, Wong J, and Keeling PJ. 2008. Environmental PCR survey to 

determine the distribution of a non-canonical genetic code in uncultivable oxymonads. 

Environmental Microbiology 10:65-74. 

De Rijk P and De Wachter R. 1993. DCSE, an interactive tool for sequence alignment and secondary 

structure research. Comput. Appl. Biosci. 9:735-740. 

De Rijk P, Robbrecht E, de Hoog S, Caers A, Van de Peer Y, and De Wachter R. 1999. Database on the 

structure of large subunit ribosomal RNA. Nucleic Acids Research 27:174-178. 

Delsuc F, Brinkmann H, and Philippe H. 2005. Phylogenomics and the reconstruction of the tree of 

life. Nature Reviews Genetics 6:361-375.


Derelle E, Ferraz C, Rombauts S, Rouze P, Worden AZ, Robbens S, Partensky F, Degroeve S, Echeynie 

S, Cooke R, Saeys Y, Wuyts J, Jabbari K, Bowler C, Panaud O, Piegu B, Ball SG, Ral JP, Bouget 

FY, Piganeau G, De Baets B, Picard A, Delseny M, Demaille J, Van de Peer Y, and Moreau H. 

2006. Genome analysis of the smallest free-living eukaryote Ostreococcus tauri unveils many 

unique features. Proceedings of the National Academy of Sciences of the United States of 

America 103:11647-11652. 

DOE Joint Genome Institute. www.jgi.doe.gov 

dos Reis M and Wernisch L. 2009. Estimating translational selection in eukaryotic genomes. 

Molecular Biology and Evolution 26:451-461. 

Douzery EJP, Snell EA, Bapteste E, Delsuc F, and Philippe H. 2004. The timing of eukaryotic evolution: 

Does a relaxed molecular clock reconcile proteins and fossils? Proceedings of the National 

Academy of Sciences of the United States of America 101:15386-15391. 

Drew EA and Abel KM. 1992. Studies on Halimeda IV. An endogenous rhythm of chloroplast 

migration in the siphonous green-alga, H. distorta. Journal of Interdisciplinary Cycle Research 

23:128-135. 

Driskell AC, Ane C, Burleigh JG, McMahon MM, O'Meara BC, and Sanderson MJ. 2004. Prospects for 

building the tree of life from large sequence databases. Science 306:1172-1174. 

Duret L and Mouchiroud D. 1999. Expression pattern and, surprisingly, gene length shape codon 

usage in Caenorhabditis, Drosophila, Arabidopsis. Proceedings of the National Academy of 


Dybvig K, Cao Z, French CT, and Yu HL. 2007. Evidence for type III restriction and modification 

systems in Mycoplasma pulmonis. Journal of Bacteriology 189:2197-2202. 

Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. 

Nucleic Acids Research 32:1792-1797. 

Fawley MW, Yun Y, and Qin M. 2000. Phylogenetic analyses of 18S rDNA sequences reveal a new 

coccoid lineage of the Prasinophyceae (Chlorophyta). Journal of Phycology 36:387-393. 

Feldmann J. 1954. Inventaire de la flore marine de Roscoff. Algues, champignons, lichens et 

spermatophytes. Trav. St. Biol. Roscoff, Nouv. Sér. Suppl. 6:152. 

Felsenstein J. 1985. Confidence-limits on phylogenies - an approach using the bootstrap. Evolution 

39:783-791. 

Felsenstein J. 2005. PHYLIP (Phylogeny Inference Package) version 3.6. 

http://evolution.genetics.washington.edu/phylip/ 

Fenchel T and Finlay BJ. 2004. The ubiquity of small species: Patterns of local and global diversity. 

Bioscience 54:777-784. 

Finlay BJ. 2002. Global dispersal of free-living microbial eukaryote species. Science 296:1061-1063. 

Floyd GL and O' Kelly CJ. 1984. Motile cell ultrastructure and the circumscription of the orders 

Ulotrichales and Ulvales (Ulvophyceae, Chlorophyta). American Journal of Botany 71:111– 

120. 

Foslie M. 1881. Om nogle nye arctiske havalger. Christiania Vidensk.-Selsk. Forh. 14:1-14. 

Fott B. 1971. Algenkunde. 2nd Ed. VEB Fischer, Jena. 

Friedl T. 1995. Inferring taxonomic positions and testing genus level assignments in coccoid green 

lichen algae - a phylogenetic analysis of 18S ribosomal-RNA sequences from Dictyochloropsis 

reticulata and from members of the genus Myrmecia (Chlorophyta, Trebouxiophyceae cl. 

nov.). Journal of Phycology 31:632-639.


Friedl T and O'Kelly CJ. 2002. Phylogenetic relationships of green algae assigned to the genus 

Planophila (Chlorophyta): evidence from 18S rDNA sequence data and ultrastructure. 

European Journal of Phycology 37:373-384. 

Fritsch FE. 1944. Cladophorella calcicola nov. gen. et sp., a terrestrial member of the Cladophorales. 

Annals of Botany N.S. 8:157-171. 

Galtier N. 2007. A model of horizontal gene transfer and the bacterial phylogeny problem. Systematic 

Biology 56:633-642. 

Geuten K, Massingham T, Darius P, Smets E, and Goldman N. 2007. Experimental design criteria in 

phylogenetics: where to add taxa. Systematic Biology 56:609-622. 

Gile GH, Patron NJ, and Keeling PJ. 2006. EFL GTPase in cryptomonads and the distribution of EFL and 

EF-1α in chromalveolates. Protist 157:435-444. 

Gile GH, Novis P, Cragg D, Zuccarello GC, and Keeling P. 2009. The distribution of elongation factor- 

1alpha (EF-1α), elongation factor-like (EFL), and a non-canonical genetic code in the 

Ulvophyceae: discrete genetic characters support a consistent phylogenetic framework. 

Journal of Eukaryotic Microbiology in press. 

Gogarten JP. 2003. Gene transfer: Gene swapping craze reaches eukaryotes. Current Biology 13:R53- 

R54. 

Goodson HV and Hawse WF. 2002. Molecular evolution of the actin family. Journal of Cell Science 

115:2619-2622. 

Graham LE, Graham JM, and Wilcox LW. 2009. Algae. Benjamin Cummings, San Francisco, Boston, 

New York, Cape Town, Hong Kong, London, Madrid, Mexico City, Montreal, Munich, Paris, 

Singapore, Sydney, Tokyo, Toronto. 

Grauvogel C, Brinkmann H, and Petersen J. 2007. Evolution of the glucose-6-phosphate isomerase: 

The plasticity of primary metabolism in photosynthetic eukaryotes. Molecular Biology and 

Evolution 24:1611-1621. 

Guillou L, Eikrem W, Chretiennot-Dinet MJ, Le Gall F, Massana R, Romari K, Pedros-Alio C, and Vaulot 

D. 2004. Diversity of picoplanktonic prasinophytes assessed by direct nuclear SSU rDNA 

sequencing of environmental samples and novel isolates retrieved from oceanic and coastal 

marine ecosystems. Protist 155:193-214. 

Guindon S and Gascuel O. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies 

by maximum likelihood. Systematic Biology 52:696-704. 

Guiry MD and Guiry GM. 2009. AlgaeBase. World-wide electronic publication. National University of 

Ireland, Galway. http://www.algaebase.org 

Gunderson JH, Elwood HJ, Ingold A, Kindle KL, and Sogin ML. 1987. Phylogenetic relationships 

between chlorophytes, chrysophytes, and Oomycetes. Proceedings of the National Academy 

of Sciences of the United States of America 84:5823-5827. 

Hall JD and Delwiche CF. 2007. In the shadow of giants: systematics of the charophyte green algae. 

Pp. 155-169 in J. Brodie, and J. Lewis, eds. Unravelling the algae: the past, present, and future 

of algal systematics. The Systematics Association Special Volume Series, 75. CRC Press, Boca 

Raton, FL, USA. 

Hanyu N, Kuchino Y, Nishimura S, and Beier H. 1986. Dramatic events in ciliate evolution - alteration 

of UAA and UAG termination codons to glutamine codons due to anticodon mutations in two 

Tetrahymena tRNAs-Gln. EMBO Journal 5:1307-1311. 

Hanyuda T, Arai S, and Ueda K. 2000. Variability in the rbcL introns of Caulerpalean algae 

(Chlorophyta, Ulvophyceae). Journal of Plant Research 113:403-413.


Hanyuda T, Wakana I, Arai S, Miyaji K, Watano Y, and Ueda K. 2002. Phylogenetic relationships within 

Cladophorales (Ulvophyceae, Chlorophyta) inferred from 18S rRNA gene sequences, with 

special reference to Aegagropila linnaei. Journal of Phycology 38:564-571. 

Hashimoto T, Nakamura Y, Nakamura F, Shirakura T, Adachi J, Goto N, Okamoto K, and Hasegawa M. 

1994. Protein phylogeny gives a robust estimation for early divergences of eukaryotes: 

phylogenetic place of a mitochondria-lacking protozoan, Giardia lamblia. Molecular Biology 

and Evolution 11:65-71. 

Hedges SB, Blair JE, Venturi ML, and Shoe JL. 2004. A molecular timescale of eukaryote evolution and 

the rise of complex multicellular life. BMC Evolutionary Biology 4. 

Herron MD, Hackett JD, Aylward FO, and Michod RE. 2009. Triassic origin and early radiation of 

multicellular volvocine algae. Proceedings of the National Academy of Sciences of the United 

States of America 106:3254-3258. 

Hershberg R and Petrov DA. 2008. Selection on codon bias. Annual Review of Genetics 42:287-299. 

Higgs PG and Ran WQ. 2008. Coevolution of codon usage and tRNA genes leads to alternative stable 

states of biased codon usage. Molecular Biology and Evolution 25:2279-2291. 

Holder M and Lewis PO. 2003. Phylogeny estimation: Traditional and Bayesian approaches. Nature 

Reviews Genetics 4:275-284. 

Hori T. 1994. An illustrated atlas of the life history of algae. Volume 1. Green Algae. Uchida Rokakuho 

Publishing Co., Ltd., Tokyo. 

Horowitz S and Gorovsky MA. 1985. An An unusual genetic code in nuclear genes of Tetrahymena. 

Proceedings of the National Academy of Sciences of the United States of America 82:2452- 

2455. 

Inagaki Y, Blouin C, Doolittle WF, and Roger AJ. 2002. Convergence and constraint in eukaryotic 

release factor 1 (eRF1) domain 1: the evolution of stop codon specificity. Nucleic Acids 

Research 30:532-544. 

Index Nominum Algarum UH, University of California, Berkeley. Compiled by Paul Silva. Available 

online at http://ucjeps.berkeley.edu/INA.html. 

Ingvarsson PK. 2008. Molecular evolution of synonymous codon usage in Populus. BMC Evolutionary 

Biology 8:307. 

Irvine D and John D. 1984. The Systematics of Green Algae. Academic Press, London. 

James TY, Kauff F, Schoch CL, Matheny PB, Hofstetter V, Cox CJ, Celio G, Gueidan C, Fraker E, 

Miadlikowska J, Lumbsch HT, Rauhut A, Reeb V, Arnold AE, Amtoft A, Stajich JE, Hosaka K, 

Sung GH, Johnson D, O'Rourke B, Crockett M, Binder M, Curtis JM, Slot JC, Wang Z, Wilson 

AW, Schussler A, Longcore JE, O'Donnell K, Mozley-Standridge S, Porter D, Letcher PM, 

Powell MJ, Taylor JW, White MM, Griffith GW, Davies DR, Humber RA, Morton JB, Sugiyama 

J, Rossman AY, Rogers JD, Pfister DH, Hewitt D, Hansen K, Hambleton S, Shoemaker RA, 

Kohlmeyer J, Volkmann-Kohlmeyer B, Spotts RA, Serdani M, Crous PW, Hughes KW, 

Matsuura K, Langer E, Langer G, Untereiner WA, Lucking R, Budel B, Geiser DM, Aptroot A, 

Diederich P, Schmitt I, Schultz M, Yahr R, Hibbett DS, Lutzoni F, McLaughlin DJ, Spatafora JW, 

and Vilgalys R. 2006. Reconstructing the early evolution of Fungi using a six-gene phylogeny. 

Nature 443:818-822. 

Jansen RK, Cai Z, Raubeson LA, Daniell H, Depamphilis CW, Leebens-Mack J, Muller KF, Guisinger- 

Bellian M, Haberle RC, Hansen AK, Chumley TW, Lee SB, Peery R, McNeal JR, Kuehl JV, and 

Boore JL. 2007. Analysis of 81 genes from 64 plastid genomes resolves relationships in


angiosperms and identifies genome-scale evolutionary patterns. Proceedings of the National 

Academy of Sciences of the United States of America 104:19369-19374. 

Jobb G, von Haeseler A, and Strimmer K. 2004. TREEFINDER: a powerful graphical analysis 

environment for molecular phylogenetics. BMC Evolutionary Biology 4. 

Jobb G. 2008. TREEFINDER version of March 2008. www.treefinder.de 

Jónsson S. 1962. Recherches sur des Cladophoracées marines: structure, reproduction, cycles 

comparés, conséquences systématiques. Annales des Sciences Naturelles, Botanique 12:25- 

263. 

Jónsson S. 1999. The status of the Acrosiphoniales (Chlorophyta). Rit Fiskideildar 16:187–196. 

Jorde I. 1933. Untersuchungen über den Lebenszyklus von Urospora Aresch. und Codiolum A. Braun. 

Nyt Magazin for Naturvidensk 73:1-19. 

Jukes TH and Osawa S. 1993. Evolutionary changes in the genetic code. Comparative Biochemistry 

and Physiology - Part B: Biochemistry & Molecular Biology 106:489-494. 

Kamikawa R, Inagaki Y, and Sako Y. 2008. Direct phylogenetic evidence for lateral transfer of 

elongation factor-like gene. Proceedings of the National Academy of Sciences of the United 


Kapraun DF and Nguyen MN. 1994. Karyology, nuclear DNA quantification and nucleus-cytoplasmic 

domain variations in some multinucleate green algae (Siphonocladales, Chlorophyta). 

Phycologia 33:42-52. 

Karol KG, McCourt RM, Cimino MT, and Delwiche CF. 2001. The closest living relatives of land plants. 

Science 294:2351-2353. 

Karsten U, Friedl T, Schumann R, Hoyer K, and Lembcke S. 2005. Mycosporine-like amino acids and 

phylogenies in green algae: Prasiola and its relatives from the Trebouxiophyceae 

(Chlorophyta). Journal of Phycology 41:557-566. 

Kass RE and Raftery AE. 1995. Bayes Factors. Journal of the American Statistical Association 90:773- 

795. 

Kawahara Y and Imanishi T. 2007. A genome-wide survey of changes in protein evolutionary rates 

across four closely related species of Saccharomyces sensu stricto group. BMC Evolutionary 

Biology 7:9. 

Keeling PJ and Doolittle WF. 1996. A non-canonical genetic code in an early diverging eukaryotic 

lineage. EMBO Journal 15:2285-2290. 

Keeling PJ and Leander BS. 2003. Characterisation of a non-canonical genetic code in the oxymonad 

Streblomastix strix. Journal of Molecular Biology 326:1337-1349. 

Keeling PJ and Inagaki Y. 2004. A class of eukaryotic GTPase with a punctate distribution suggesting 

multiple functional replacements of translation elongation factor 1α. Proceedings of the 

National Academy of Sciences of the United States of America 101:15380-15385. 

Keeling PJ and Palmer JD. 2008. Horizontal gene transfer in eukaryotic evolution. Nature Reviews 

Genetics 9:605-618. 

Kim GH, Klotchkova TA, and Kang YM. 2001. Life without a cell membrane: regeneration of 

protoplasts from disintegrated cells of the marine green alga Bryopsis plumosa. Journal of 

Cell Science 114:2009-2014. 

Knight R, Freeland S, and Landweber L. 2001a. A simple model based on mutation and selection 

explains trends in codon and amino-acid usage and GC composition within and across 

genomes. Genome Biology 2:1-13.


Knight RD, Freeland SJ, and Landweber LF. 2001b. Rewiring the keyboard evolvability of the genetic 

code. Nature Reviews Genetics 2:49-58. 

Kolisko M, Cepicka I, Hampl V, Leigh J, Roger AJ, Kulda J, Simpson AGB, and Flegr J. 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. 

Kornmann P. 1963. Die Ulotrichales, neu geordnet auf der Grundlage entwicklungsgeschichtlicher 

Befunde. Phycologia 3:60-68. 

Kornmann P and Sahling P-H. 1994. Meeresalgen von Helgoland: Zweite Ergänzung. Helgoländer 

Meeresuntersuchungen 48:365-406. 

Kraft GT. 2007. Algae of Australia: marine benthic algae of Lord Howe Island and the Southern Great 

Barrier Reef, 1: Green Algae. Australian Biological Resources Study & CSIRO Publishing, 

Canberra & Melbourne. 

Krienitz L, Ustinova I, Friedl T, and Huss VAR. 2001. Traditional generic concepts versus 18S rRNA 

gene phylogeny in the green algal family Selenastraceae (Chlorophyceae, Chlorophyta). 

Journal of Phycology 37:852-865. 

Krienitz L, Hegewald E, Hepperle D, and Wolf M. 2003. The systematics of coccoid green algae: 18S 

rRNA gene sequence data versus morphology. Biologia 58:437-446. 

Lagerheim G. 1887. Note sur l'Uronema, nouveau genre des algues d'eau douce. Malpigia 1:517-523. 

Lake JA. 2008. Reconstructing evolutionary graphs: 3D parsimony. Molecular Biology and Evolution 

25:1677-1682. 

Lane CE and Archibald JM. 2008. The eukaryotic tree of life: endosymbiosis takes its TOL. Trends in 

Ecology & Evolution 23:268-275. 

Lartillot N, Blanquart S, and Lepage T. 2007. PhyloBayes v2.3. 

http://www.lirmm.fr/mab/article.php3?id_article=329 

Lein TE, Bruntse G, Gunnarsson K, and Nielsen R. 1999. New records of benthic marine algae for 

Norway, with notes on some rare species from the Florø district, western Norway. Sarsia 

84:39-53. 

Leliaert F, Rousseau F, De Reviers B, and Coppejans E. 2003. Phylogeny of the Cladophorophyceae 

(Chlorophyta) inferred from partial LSU rRNA gene sequences: is the recognition of a 

separate order Siphonocladales justified? European Journal of Phycology 38:233-246. 

Leliaert F and Boedeker C. 2007. Cladophorales. Pp. 131–183 in J. Brodie, C. A. Maggs, and D. M. 

John, eds. Green seaweeds of Britain and Ireland. British Phycological Society, London, UK. 

Leliaert F, De Clerck O, Verbruggen H, Boedeker C, and Coppejans E. 2007. Molecular phylogeny of 

the Siphonocladales (Chlorophyta : Cladophorophyceae). Molecular Phylogenetics and 

Evolution 44:1237-1256. 

Leliaert F, Boedeker C, Pena V, Bunker F, Verbruggen H, and De Clerck O. 2009. Cladophora 

rhodolithicola sp. nov. (Cladophorales, Chlorophyta), a diminutive species from European 

maerl beds. European Journal of Phycology 43: in press. 

Lemieux C, Otis C, and Turmel M. 2000. Ancestral chloroplast genome in Mesostigma viride reveals 

an early branch of green plant evolution. Nature 403:649-652. 

Lemieux C, Otis C, and Turmel M. 2007. A clade uniting the green algae Mesostigma viride and 

Chlorokybus atmophyticus represents the deepest branch of the Streptophyta in chloroplast 

genome-based phylogenies. BMC Biology 5:2.


Levring T. 1937. Zur Kenntnis der Algenflora der Norwegischen Westküste. Lunds Univ. Årsskr. 33:1- 

147. 

Lewis LA and McCourt RM. 2004. Green algae and the origin of land plants. American Journal of 

Botany 91:1535-1556. 

Lewis LA and Lewis PO. 2005. Unearthing the molecular phylodiversity of desert soil green algae 

(Chlorophyta). Systematic Biology 54:936–947. 

Li CH, Lu GQ, and Orti G. 2008. Optimal data partitioning and a test case for ray-finned fishes 

(Actinopterygii) based on ten nuclear loci. Systematic Biology 57:519-539. 

Lindstrom SC and Hanic LA. 2005. The phylogeny of North American Urospora (Ulotrichales, 

Chlorophyta) based on sequence analysis of nuclear ribosomal genes, introns and spacers. 


Littler DS, Littler MM, and Hanisak MD. 2008. Submersed plants of the Indian River Lagoon: A floristic 

inventory and field guide. OffShore Graphics, Inc., Washington, DC. 

Littler MM, Littler DS, and Lapointe BE. 1988. A comparison of nutrient-limited and light-limited 

photosynthesis in psammophytic versus epilithic forms of Halimeda (Caulerpales, 

Halimedaceae) from the Bahamas. Coral Reefs 6:219-225. 

Lobanov AV, Fomenko DE, Zhang Y, Sengupta A, Hatfield DL, and Gladyshev VN. 2007. Evolutionary 

dynamics of eukaryotic selenoproteomes: large selenoproteomes may associate with aquatic 

life and small with terrestrial life. Genome Biology 8:R198. 

Lokhorst GM and Trask BJ. 1981. Taxonomic studies on Urospora (Acrosiphoniales, Chlorophyceae) in 

western Europe. Acta Botanica Neerlandica 30:353-431. 

Lopez-Bautista JM and Chapman RL. 2003. Phylogenetic affinities of the Trentepohliales inferred 

from small-subunit rDNA. International Journal of Systematic and Evolutionary Microbiology 

53:2099-2106. 

Lopez-Bautista JM, Rindi F, and Casamatta D. 2007. The subaerial algae. Pp. 601-617 in J. Seckbach, 

ed. Algae and Cyanobacteria in Extreme Environments. Springer. 

Lozupone CA, Knight RD, and Landweber LF. 2001. The molecular basis of nuclear genetic code 

change in ciliates. Current Biology 11:65-74. 

Maggs CA and O' Kelly CJ. 2007. Uronema. Pp. 209-210 in J. Brodie, C. A. Maggs, and D. M. John, eds. 

Green seaweeds of Britain and Ireland. British Phycological Society, London, UK. 

Mann DG. 1996. Crossing the Rubicon: the effectiveness of the marine/freshwater interface as a 

barrier to the migration of diatom germplasm. Pp. 1-21 in S. Mayama, M. Idei, and I. Koizumi, 

eds. Fourtheenth International Diatom Symposium. Koeltz Scientific Books, Tokyo, Japan. 

Maquat LE. 2004. Nonsense-mediated mRNA decay: Splicing, translation and mRNP dynamics. Nature 

Reviews Molecular Cell Biology 5:89-99. 

Marin B and Melkonian M. 1999. Mesostigmatophyceae, a new class of streptophyte green algae 

revealed by SSU rRNA sequence comparisons. Protist 150:399-417. 

Martins EP. 1999. Estimation of ancestral states of continuous characters: A computer simulation 

study. Systematic Biology 48:642-650. 

Matsuyama K, Matsuoka T, Miyaji K, Tanaka J, and Aruga Y. 1998. Ultrastructure of the pyrenoid in 

the family Cladophoraceae (Cladophorales, Chlorophyta). Journal of Japanese Botany 73:279- 

286.


Mattox K and Stewart K. 1984. Classification of the green algae: a concept based on comparative 

cytology. Pp. 29–72 in D. Irvine, and D. John, eds. Systematics of the green algae. The 

Systematics Association Special Volume 27, Academic Press, London and Orlando. 

McCourt RM, Delwiche CF, and Karol KG. 2004. Charophyte algae and land plant origins. Trends in 

Ecology & Evolution 19:661-666. 

Melkonian M. 1989. Flagellar Apparatus Ultrastructure in Mesostigma viride (Prasinophyceae). Plant 

Systematics and Evolution 164:93-122. 

Menzel D. 1988. How do giant plant cells cope with injury? - The wound response in siphonous green 

algae. Protoplasma 144:73-91. 

Meyer KI. 1927. Algae of the northern end of lake Baikal. Archives Russes de Protisologie 6:93-118 [In 

Russian, with French summary]. 

Mishler BD, Lewis LA, Buchheim MA, Renzaglia KS, Garbary DJ, Delwiche CF, Zechman FW, Kantz TS, 

and Chapman RL. 1994. Phylogenetic relationships of the "green algae" and "bryophytes". 

Annals of the Missouri Botanical Garden 81:451-483. 

Miyaji K. 1999. A new type of pyrenoid in the genus Rhizoclonium (Cladophorales, Chlorophyta). 


Moestrup Ø. 1978. Phylogenetic validity of flagellar apparatus in green algae and other chlorophyll a 

and b containing plants. Biosystems 10:117-144. 

Moore MJ, Bell CD, Soltis PS, and Soltis DE. 2007. Using plastid genome-scale data to resolve 

enigmatic relationships among basal angiosperms. Proceedings of the National Academy of 


Muller J, Oma Y, Vallar L, Friederich E, Poch O, and Winsor B. 2005. Sequence and comparative 

genomic analysis of actin-related proteins. Molecular Biology of the Cell 16:5736-5748. 

Negrutskii BS and El'skaya AV. 1998. Eukaryotic translation elongation factor 1α : Structure, 

expression, functions, and possible role in aminoacyl-tRNA channeling. Pp. 47-78. Progress in 

Nucleic Acid Research and Molecular Biology, Volume 60. 

Nei M and Rooney AP. 2005. Concerted and birth-and-death evolution of multigene families. Annual 

Review of Genetics 39:121-152. 

Nielsen R and Gunnarsson K. 2001. Seaweeds of the Faroe Islands. An annotated checklist. 

Fródskaparrit 49:45-108. 

Noble GP, Rogers MB, and Keeling PJ. 2007. Complex distribution of EFL and EF-1α proteins in the 

green algal lineage. BMC Evolutionary Biology 7:82. 

Nosenko T, Lidie KL, Van Dolah FM, Lindquist E, Cheng JF, and Bhattacharya D. 2006. Chimeric plastid 

proteome in the florida "red tide" dinoflagellate Karenia brevis. Molecular Biology and 

Evolution 23:2026-2038. 

Nozaki H, Matsuzaki M, Takahara M, Misumi O, Kuroiwa H, Hasegawa M, Shin-i T, Kohara Y, 

Ogasawara N, and Kuroiwa T. 2003a. The phylogenetic position of red algae revealed by 

multiple nuclear genes from mitochondria-containing eukaryotes and an alternative 

hypothesis on the origin of plastids. Journal of Molecular Evolution 56:485-497. 

Nozaki H, Ohta N, Matsuzaki M, Misumi O, and Kuroiwa T. 2003b. Phylogeny of plastids based on 

cladistic analysis of gene loss inferred from complete plastid genome sequences. Journal of 

Molecular Evolution 57:377-382. 

Nylander JAA, Ronquist F, Huelsenbeck JP, and Nieves-Aldrey JL. 2004. Bayesian phylogenetic analysis 

of combined data. Systematic Biology 53:47-67.


O'Kelly CJ and Floyd GL. 1984a. Flagellar apparatus absolute orientations and the phylogeny of the 

green algae. Biosystems 16:227-251. 

O'Kelly CJ and Floyd GL. 1984b. Correlations among patterns of sporangial structure and 

development, life histories, and ultrastructural features in the Ulvophyceae. Pp. 121-156 in 

D. Irvine, and D. John, eds. Systematics of the green algae. The Systematics Association 

Special Volume 27, Academic Press, London and Orlando. 

O'Kelly CJ, Wysor B, and Bellows WK. 2004. Gene sequence diversity and the phylogenetic position of 

algae assigned to the genera Phaeophila and Ochlochaete (Ulvophyceae, Chlorophyta). 


O'Kelly CJ. 2007. The origin and early evolution of green plants. Pp. 287-309 in P. G. Falkowski, and A. 

H. Knoll, eds. Evolution of primary producers in the sea. Elsevier Academic Press, Amsterdam, 

The Netherlands. 

O'Neil RM and La Claire JW. 1984. Mechanical wounding induces the formation of extensive coated 

membranes in giant algal cells. Science 225:331-333. 

Pagel M. 1994. Detecting correlated evolution on phylogenies: a general method for the comparative 

analysis of discrete characters. Proceedings of the Royal Society of London Series B-Biological 

Sciences 255:37-45. 

Pagel M. 1999. The maximum likelihood approach to reconstructing ancestral character states of 

discrete characters on phylogenies. Systematic Biology 48:612-622. 

Pagel M and Meade A. 2004. A phylogenetic mixture model for detecting pattern-heterogeneity in 

gene sequence or character-state data. Systematic Biology 53:571-581. 

Pagel M and Meade A. 2006. Bayesian analysis of correlated evolution of discrete characters by 

reversible-jump Markov chain Monte Carlo. American Naturalist 167:808-825. 

Palmer JD. 2003. The symbiotic birth and spread of plastids: How many times and whodunit? Journal 

of Phycology 39:4-11. 

Paradis E, Claude J, and Strimmer K. 2004. APE: Analyses of Phylogenetics and Evolution in R 

language. Bioinformatics 20:289-290. 

Parker B. 1970. Significance of cell wall chemistry to phylogeny in the green algae. Annals of the New 

York Academy of Sciences 175:417-428. 

Parkinson CL, Adams KL, and Palmer JD. 1999. Multigene analyses identify the three earliest lineages 

of extant flowering plants. Current Biology 9:1485-1488. 

Petersen J, Teich R, Becker B, Cerff R, and Brinkmann H. 2006. The GapA/B gene duplication marks 

the origin of streptophyta (Charophytes and land plants). Molecular Biology and Evolution 

23:1109-1118. 

Philip GK, Creevey CJ, and McInerney JO. 2005. The Opisthokonta and the Ecdysozoa may not be 

clades: stronger support for the grouping of plant and animal than for animal and fungi and 

stronger support for the Coelomata than Ecdysozoa. Molecular Biology and Evolution 

22:1175-1184. 

Philippe H, Snell EA, Bapteste E, Lopez P, Holland PWH, and Casane D. 2004. Phylogenomics of 

eukaryotes: Impact of missing data on large alignments. Molecular Biology and Evolution 

21:1740-1752. 

Philippe H, Delsuc F, Brinkmann H, and Lartillot N. 2005a. Phylogenomics. Annual Review of Ecology 

Evolution and Systematics 36:541-562.


Philippe H, Lartillot N, and Brinkmann H. 2005b. Multigene analyses of bilaterian animals corroborate 

the monophyly of Ecdysozoa, Lophotrochozoa, and Protostomia. Molecular Biology and 

Evolution 22:1246-1253. 

Pickett-Heaps JD and Marchant HJ. 1972. The phylogeny of the green algae: a new proposal. Cytobios 

6:255-264. 

Pombert JF, Otis C, Lemieux C, and Turmel M. 2004. The complete mitochondrial DNA sequence of 

the green alga Pseudendoclonium akinetum (Ulvophyceae) highlights distinctive evolutionary 

trends in the Chlorophyta and suggests a sister-group relationship between the Ulvophyceae 

and Chlorophyceae. Molecular Biology and Evolution 21:922-935. 

Pombert JF, Otis C, Lemieux C, and Turmel M. 2005. Chloroplast genome sequence of the green alga 

Pseudendoclonium akinetum (Ulvophyceae) reveals unusual structural features and new 

insights into the branching order of chlorophyte lineages. Molecular Biology and Evolution 

22:1903-1918. 

Pombert JF, Lemieux C, and Turmel M. 2006. The complete chloroplast DNA sequence of the green 

alga Oltmannsiellopsis viridis reveals a distinctive quadripartite architecture in the 

chloroplast genome of early diverging ulvophytes. BMC Biology 4. 

Pond SLK, Frost SDW, and Muse SV. 2005. HyPhy: Hypothesis testing using Phylogenies. 

Bioinformatics 21:676-679. 

Posada D and Crandall KA. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 

14: 817-818. 

Powell JR and Moriyama EN. 1997. Evolution of codon usage bias in Drosophila. Proceedings of the 

National Academy of Sciences of the United States of America 94:7784-7790. 

Printz H. 1926. Die Algenvegetation des Trondjemsfjordes. Skr. Norske Vidensk.-Akad. Oslo, Mat.naturv. 

Kl. 1926 5:1- 273. 

Printz H. 1932. Observations on the structure and reproduction in Urospora. Aresch. Nyt. Mag. 

Naturvidensk. 70:274-287. 

Pröschold T and Leliaert F. 2007. Systematics of the green algae: conflict of classic and modern 

approaches. Pp. 123-153 in B. J., and L. J.M., eds. Unravelling the algae: the past, present, 

and future of algal systematics. The Systematics Association Special Volume Series, 75. CRC 

Press, Boca Raton, FL, USA. 

Qiu YL, Li LB, Wang B, Chen ZD, Knoop V, Groth-Malonek M, Dombrovska O, Lee J, Kent L, Rest J, 

Estabrook GF, Hendry TA, Taylor DW, Testa CM, Ambros M, Crandall-Stotler B, Duff RJ, Stech 

M, Frey W, Quandt D, and Davis CC. 2006. The deepest divergences in land plants inferred 

from phylogenomic evidence. Proceedings of the National Academy of Sciences of the United 


Rambaut A and Drummond A. 2007. Tracer: MCMC Trace Analysis Tool. 

http://tree.bio.ed.ac.uk/software/tracer/ 

Reyes-Prieto A and Bhattacharya D. 2007. Phylogeny of nuclear-encoded plastid-targeted proteins 

supports an early divergence of glaucophytes within plantae. Molecular Biology and 

Evolution 24:2358-2361. 

Reyes-Prieto A, Weber APM, and Bhattacharya D. 2007. The origin and establishment of the plastid in 

algae and plants. Annual Review of Genetics 41:147-168. 

Rindi F, Guiry MD, and Lopez-Bautista JM. 2006a. New records of Trentepohliales (Ulvophyceae, 

Chlorophyta) from Africa. Nova Hedwigia 83:431-449.


Rindi F, López-Bautista JM, Sherwood AR, and Guiry MD. 2006b. Morphology and phylogenetic 

position of Spongiochrysis hawaiiensis gen. et sp. nov., the first known terrestrial member of 

the order Cladophorales (Ulvophyceae, Chlorophyta). International Journal of Systematic and 

Evolutionary Microbiology 56:913-922. 

Robbens S, Petersen J, Brinkmann H, Rouze P, and Van de Peer Y. 2007. Unique regulation of the 

Calvin cycle in the ultrasmall green alga Ostreococcus. Journal of Molecular Evolution 64:601- 

604. 

Rodriguez-Ezpeleta N, Brinkmann H, Burey SC, Roure B, Burger G, Löffelhardt W, Bohnert HJ, Philippe 

H, and Lang BF. 2005. Monophyly of primary photosynthetic eukaryotes: green plants, red 

algae, and glaucophytes. Current Biology 15:1325-1330. 

Rodriguez-Ezpeleta N, Philippe H, Brinkmann H, Becker B, and Melkonian M. 2007. Phylogenetic 

analyses of nuclear, mitochondrial, and plastid multigene data sets support the placement of 

Mesostigma in the Streptophyta. Molecular Biology and Evolution 24:723-731. 

Roger AJ, Sandblom O, Doolittle WF, and Philippe H. 1999. An evaluation of elongation factor 1 alpha 

as a phylogenetic marker for eukaryotes. Molecular Biology and Evolution 16:218-233. 

Roger AJ and Hug LA. 2006. The origin and diversification of eukaryotes: problems with molecular 

phylogenetics and molecular clock estimation. Philosophical Transactions of the Royal 

Society B-Biological Sciences 361:1039-1054. 

Rogers MB, Watkins RF, Harper JT, Durnford DG, Gray MW, and Keeling PJ. 2007. A complex and 

punctate distribution of three eukaryotic genes derived by lateral gene transfer. BMC 

Evolutionary Biology 7:89. 

Rokas A and Abbot P. 2009. Harnessing genomics for evolutionary insights. Trends in Ecology & 


Ronquist F and Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed 

models. Bioinformatics 19:1572-1574. 

Rosenvinge LK. 1893. Grønlands Havalger. Medd. Grønland 3:763-981. 

Round FE. 1984. The systematics of the Chlorophyta: an historical review leading to some modern 

concepts. in I. D.E.G., and D. M. John, eds. The Systematics of Green Algae. The Systematics 

Association Special Volume 27, Academic Press, London and Orlando. 

Rueness J. 1977. Norsk Algeflora. Universitetsforlaget, Olso. 

Rueness J. 1992. Field and culture observations on Uronema curvatum Printz (Chlorophyta). Acta 

Phytogeogr. Suec. 78:125-130. 

Saarela JM, Rai HS, Doyle JA, Endress PK, Mathews S, Marchant AD, Briggs BG, and Graham SW. 2007. 

Hydatellaceae identified as a new branch near the base of the angiosperm phylogenetic tree. 

Nature 446:312-315. 

Sakaguchi M, Suzaki T, Khan S, and Hausmann K. 2002. Food capture by kinetocysts in the heliozoon 

Raphidiophrys contractilis. European Journal of Protistology 37:453-458. 

Sakaguchi M, Takishita K, Matsumoto T, Hashimoto T, and Inagaki Y. 2008. Tracing back EFL gene 

evolution in the cryptomonads–haptophytes assemblage: Separate origins of EFL genes in 

haptophytes, photosynthetic cryptomonads, and goniomonads. Gene in press. 

Salim HMW and Cavalcanti ARO. 2008. Factors influencing codon usage bias in genomes. Journal of 

the Brazilian Chemical Society 19:257-262. 

Sambrook J, Fritch EF, and Maniatis T. 1989. Molecular cloning: a laboratory manual. Cold Spring 

Harbor Laboratory Press, New York.


Sanderson MJ. 2002. Estimating absolute rates of molecular evolution and divergence times: A 

penalized likelihood approach. Molecular Biology and Evolution 19:101-109. 

Sanderson MJ. 2003. r8s: inferring absolute rates of molecular evolution and divergence times in the 

absence of a molecular clock. Bioinformatics 19:301-302. 

Santos MAS, Moura G, Massey SE, and Tuite MF. 2004. Driving change: the evolution of alternative 

genetic codes. Trends in Genetics 20:95-102. 

Scagel RF. 1966. Marine algae of British Columbia and northern Washington, Part I: Chlorophyceae 

(green algae). Bulletin National Museum of Canada 207:1-257. 

Schlösser UG. 1987. Action of cell wall autolysins in asexual reproduction of filamentous green algae: 

evidence and species specificity. Pp. 75–80 in W. Wiessner, D. G. Robinson, and S. R.C., eds. 

Algal Development: Molecular and Cellular Aspects. Springer, Berlin, Heidelberg, Germany. 

Schluter D, Price T, Mooers AO, and Ludwig D. 1997. Likelihood of ancestor states in adaptive 

radiation. Evolution 51:1699-1711. 

Schluter PM, Stuessy TF, and Paulus HF. 2005. Making the first step: practical considerations for the 

isolation of low-copy nuclear sequence markers. Taxon 54:766-770. 

Schneider CW and Searles RB. 1991. Seaweeds of the southeastern United States. Cape Hatteras to 

Cape Canaveral. Duke University Press, Durham & London. 

Schneider SU, Leible MB, and Yang XP. 1989. Strong homology between the small subunit of ribulose- 

1,5-bisphosphate carboxylase/oxygenase of two species of Acetabularia and the occurrence 

of unusual codon usage. Molecular & General Genetics 218:445-452. 

Schneider SU and de Groot EJ. 1991. Sequences of two rbcS cDNA clones of Batophora oerstedii: 

structural and evolutionary considerations. Current Genetics 20:173-175. 

Schöniger M and Von Haeseler A. 1994. A stochastic model for the evolution of autocorrelated DNA 

sequences. Molecular Phylogenetics and Evolution 3:240-247. 

Schultz DW and Yarus M. 1994. Transfer RNA mutation and the malleability of the genetic code. 

Journal of Molecular Biology 235:1377-1380. 

Sengupta S and Higgs PG. 2005. A unified model of codon reassignment in alternative genetic codes. 

Genetics 170:831-840. 

Seo TK and Kishino H. 2008. Synonymous substitutions substantially improve evolutionary inference 

from highly diverged proteins. Systematic Biology 57:367-377. 

Setchell WA and Gardner NL. 1920. The marine algae of the Pacific coast of North America. Part II. 

Chlorophyceae. University of California Publications in Botany 8:139-374. 

Shapiro B, Rambaut A, and Drummond AJ. 2006. Choosing appropriate substitution models for the 

phylogenetic analysis of protein-coding sequences. Molecular Biology and Evolution 23:7-9. 

Shimodaira H and Hasegawa M. 2001. CONSEL: for assessing the confidence of phylogenetic tree 

selection. Bioinformatics 17:1246-1247. 

Shimodaira H. 2002. An approximately unbiased test of phylogenetic tree selection. Systematic 

Biology 51:492-508. 

Sluiman HJ. 1989. The green algal class Ulvophyceae. An ultrastructural survey and classification. 

Cryptogamic botany 1:83-94. 

Small RL, Cronn RC, and Wendel JF. 2004. Use of nuclear genes for phylogeny reconstruction in 

plants. Australian Systematic Botany 17:145-170.


Sogin ML, Elwood HJ, and Gunderson JH. 1986. Evolutionary diversity of eukaryotic small subunit 

rRNA genes. Proceedings of the National Academy of Sciences of the United States of 

America 83:1383-1387. 

Soltis PS and Soltis DE. 2003. Applying the bootstrap in phylogeny reconstruction. Statistical Science 

18:256-267. 

South GR and Tittley I. 1986. A checklist and distributional index of the benthic marine algae of the 

North Atlantic Ocean. British Museum (Natural History) and Huntsman Marine Laboratory, 

London and St. Andrews, New Brunswick, Canada. 

Steinkötter J, Bhattacharya D, Semmelroth I, Bibeau C, and Melkonian M. 1994. Prasinophytes form 

independent lineages within the Chlorophyta: evidence from ribosomal RNA sequence 

comparison. Journal of Phycology 30:340-345. 

Stiller JW, Riley J, and Hall BD. 2001. Are red algae plants? A critical evaluation of three key molecular 

data sets. Journal of Molecular Evolution 52:527-539. 

Stiller JW and Harrell L. 2005. The largest subunit of RNA polymerase II from the Glaucocystophyta: 

functional constraint and short-branch exclusion in deep eukaryotic phylogeny. BMC 

Evolutionary Biology 5. 

Subramanian S. 2008. Nearly neutrality and the evolution of codon usage bias in eukaryotic genomes. 

Genetics 178:2429-2432. 

Sung GH, Sung JM, Hywel-Jones NL, and Spatafora JW. 2007. A multi-gene phylogeny of 

Clavicipitaceae (Ascomycota, Fungi): Identification of localized incongruence using a 

combinational bootstrap approach. Molecular Phylogenetics and Evolution 44:1204-1223. 

Swofford DL. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. 

Sinauer Associates, Sunderland, Massachusetts. 

Tartar A, Boucias DG, Adams BJ, and Becnel JJ. 2002. Phylogenetic analysis identifies the invertebrate 

pathogen Helicosporidium sp. as a green alga (Chlorophyta). International Journal of 

Systematic and Evolutionary Microbiology 52:273-279. 

Telford MJ, Wise MJ, and Gowri-Shankar V. 2005. Consideration of RNA secondary structure 

significantly improves likelihood-based estimates of phylogeny: Examples from the Bilateria. 

Molecular Biology and Evolution 22:1129-1136. 

Than C, Ruths D, Innan H, and Nakhleh L. 2006. Identifiability issues in phylogeny-based detection of 

horizontal gene transfer. Pp. 215-229. Comparative Genomics, Proceedings. 

Thorne JL, Kishino H, and Painter IS. 1998. Estimating the rate of evolution of the rate of molecular 

evolution. Molecular Biology and Evolution 15:1647-1657. 

Tourancheau AB, Tsao N, Klobutcher LA, Pearlman RE, and Adoutte A. 1995. Genetic code deviations 

in the ciliates: evidence for multiple and independent events. EMBO Journal 14:3262-3267. 

Turmel M, Ehara M, Otis C, and Lemieux C. 2002a. Phylogenetic relationships among streptophytes 

as inferred from chloroplast small and large subunit rRNA gene sequences. Journal of 

Phycology 38:364-375. 

Turmel M, Otis C, and Lemieux C. 2002b. The complete mitochondrial DNA sequence of Mesostigma 

viride identifies this green alga as the earliest green plant divergence and predicts a highly 

compact mitochondrial genome in the ancestor of all green plants. Molecular Biology and 


Turmel M, Otis C, and Lemieux C. 2003. The mitochondrial genome of Chara vulgaris: Insights into 

the mitochondrial DNA architecture of the last common ancestor of green algae and land 

plants. Plant Cell 15:1888-1903.


Turmel M, Otis C, and Lemieux C. 2006. The chloroplast genome sequence of Chara vulgaris sheds 

new light into the closest green algal relatives of land plants. Molecular Biology and Evolution 

23:1324-1338. 

Turmel M, Brouard JS, Gagnon C, Otis C, and Lemieux C. 2008. Deep division in the Chlorophyceae 

(Chlorophyta) revealed by chloroplast phylogenomic analyses. Journal of Phycology 44:739- 

750. 

Turmel M, Gagnon M-C, O'Kelly CJ, Otis C, and Lemieux C. 2009. The chloroplast genomes of the 

green algae Pyramimonas, Monomastix, and Pycnococcus shed new light on the evolutionary 

history of prasinophytes and the origin of the secondary chloroplasts of euglenids. Molecular 

Biology and Evolution 26:631-648. 

Van de Peer Y, Taylor JS, Braasch I, and Meyer A. 2001. The ghost of selection past: Rates of evolution 

and functional divergence of anciently duplicated genes. Journal of Molecular Evolution 

53:436-446. 

van den Hoek C. 1963. Revision of the European species of Cladophora, Brill, Leiden, The 

Netherlands. 

van den Hoek C, Mann DG, and Jahns HM. 1995. Algae. An Introduction to Phycology. Cambridge 

University Press, Cambridge. 

van Oppen MJH, Olsen JL, and Stam WT. 1995. Genetic variation within and among North Atlantic 

and Baltic populations of the benthic alga Phycodrys rubens (Rhodophyta). European Journal 

of Phycology 30:251-260. 

Venditti C, Meade A, and Pagel M. 2008. Phylogenetic mixture models can reduce node-density 

artifacts. Systematic Biology 57:286-293. 

Verbruggen H, Leliaert F, Maggs CA, Shimada S, Schils T, Provan J, Booth D, Murphy S, De Clerck O, 

Littler DS, Littler MM, and Coppejans E. 2007. Species boundaries and phylogenetic 

relationships within the green algal genus Codium (Bryopsidales) based on plastid DNA 

sequences. Molecular Phylogenetics and Evolution 44:240-254. 

Verbruggen H. 2008. TreeGradients v1.02. http://www.phycoweb.net 

Verbruggen H and Theriot EC. 2008. Building trees of algae: some advances in phylogenetic and 

evolutionary analysis. European Journal of Phycology 43. 

Verbruggen H. 2009. TreeGradients v1.02. http://www.phycoweb.net 

Verbruggen H, Ashworth M, LoDuca ST, Vlaeminck C, Cocquyt E, Sauvage T, Zechman FW, Littler DS, 

Littler MM, Leliaert F, and De Clerck O. 2009. A multi-locus time-calibrated phylogeny of the 

siphonous green algae. Molecular Phylogenetics and Evolution 50:642-653. 

Vetsigian K and Goldenfeld N. 2009. Genome rhetoric and the emergence of compositional bias. Pp. 

215-220. 

Vroom PS and Smith CM. 2003. Reproductive features of Hawaiian Halimeda velasquezii 

(Bryopsidales, Chlorophyta), and an evolutionary assessment of reproductive characters in 

Halimeda. Cryptogamie Algologie 24:355-370. 

Waddell PJ, Cao Y, Hauf J, and Hasegawa M. 1999. Using novel phylogenetic methods to evaluate 

mammalian mtDNA, including amino acid-invariant sites-LogDet plus site stripping, to detect 

internal conflicts in the data, with special reference to the positions of hedgehog, armadillo, 

and elephant. Systematic Biology 48:31-53. 

Wahlberg N and Wheat CW. 2008. Genomic outposts serve the phylogenomic pioneers: designing 

novel nuclear markers for genomic DNA extractions of Lepidoptera. Systematic Biology 

57:231-242.


Wan XF, Zhou J, and Xu D. 2006. CodonO: a new informatics method for measuring synonymous 

codon usage bias within and across genomes. International Journal of General Systems 

35:109-125. 

Watanabe S, Kuroda N, and Maiwa F. 2001. Phylogenetic status of Helicodictyon planctonicum and 

Desmochloris halophila gen. et comb. nov and the definition of the class Ulvophyceae 

(Chlorophyta). Phycologia 40:421-434. 

Watanabe S and Nakayama T. 2007. Ultrastructure and phylogenetic relationships of the unicellular 

green algae Ignatius tetrasporus and Pseudocharacium americanum (Chlorophyta). 

Phycological Research 55:1-16. 

Wiens JJ. 2005. Can incomplete taxa rescue phylogenetic analyses from long-branch attraction? 

Systematic Biology 54:731-742. 

Wiens JJ, Kuczynski CA, Smith SA, Mulcahy DG, Sites JW, Townsend TM, and Reeder TW. 2008. 

Branch lengths, support, and congruence: testing the phylogenomic approach with 20 

nuclear loci in snakes. Systematic Biology 57:420-431. 

Wiens JJ and Moen DS. 2008. Missing data and the accuracy of Bayesian phylogenetics. Journal of 

Systematics and Evolution 46:307-314. 

Wille N. 1890. Cladophoraceae. Pp. 114-119 in A. Engler, and K. Prantl, eds. Die natürlichen 

Pflanzenfamilien. I.Teil, Abt. 2, Leipzig. 

Wolf PG. 1997. Evaluation of atpB nucleotide sequences for phylogenetic studies of ferns and other 

pteridophytes. American Journal of Botany 84:1429-1440. 

Womersley HBS. 1984. The marine benthic flora of southern Australia. Part I. Government Printer, 

South Australia, Adelaide. 

Worden AZ, Lee J-H, Mock T, Rouze P, Simmons MP, Aerts AL, Allen AE, Cuvelier ML, Derelle E, 

Everett MV, Foulon E, Grimwood J, Gundlach H, Henrissat B, Napoli C, McDonald SM, Parker 

MS, Rombauts S, Salamov A, Von Dassow P, Badger JH, Coutinho PM, Demir E, Dubchak I, 

Gentemann C, Eikrem W, Gready JE, John U, Lanier W, Lindquist EA, Lucas S, Mayer KFX, 

Moreau H, Not F, Otillar R, Panaud O, Pangilinan J, Paulsen I, Piegu B, Poliakov A, Robbens S, 

Schmutz J, Toulza E, Wyss T, Zelensky A, Zhou K, Armbrust EV, Bhattacharya D, Goodenough 

UW, Van de Peer Y, and Grigoriev IV. 2009. Green evolution and dynamic adaptations 

revealed by genomes of the marine picoeukaryotes Micromonas. Science 324:268-272. 

Wuyts J, Van de Peer Y, and Wachter RD. 2001. Distribution of substitution rates and location of 

insertion sites in the tertiary structure of ribosomal RNA. Nucleic Acids Research 29:5017- 

5028. 

Wuyts J, Perriere G, and Van de Peer Y. 2004. The European ribosomal RNA database. Nucleic Acids 

Research 32:D101-D103. 

Yamamoto M, Nozaki H, and Kawano S. 2001. Evolutionary relationships among multiple modes of 

cell division in the genus Nannochloris (Chlorophyta) revealed by genome size, actin gene 

multiplicity, and phylogeny. Journal of Phycology 37:106-120. 

Yamamoto M, Nozaki H, Miyazawa Y, Koide T, and Kawano S. 2003. Relationship between presence 

of a mother cell wall and speciation in the unicellular microalga Nannochloris (Chlorophyta). 


Yoon HS, Hackett JD, Ciniglia C, Pinto G, and Bhattacharya D. 2004. A molecular timeline for the origin 

of photosynthetic eukaryotes. Molecular Biology and Evolution 21:809-818.


Zechman FW, Theriot EC, Zimmer EA, and Chapman RL. 1990. Phylogeny of the Ulvophyceae 

(Chlorophyta): cladistic analysis of nuclear-encoded rRNA sequence data. Journal of 

Phycology 26:700-710. 

Zhang HC and Qiao GX. 2007. Systematic status of the genus Formosaphis Takahashi and the 

evolution of galls based on the molecular phylogeny of Pemphigini (Hemiptera: Aphididae: 

Eriosomatinae). Systematic Entomology 32:690-699. 

Zhang RF, Cui ZL, Zhang XZ, Jiang JD, Gu JD, and Li SP. 2006. Cloning of the organophosphorus 

pesticide hydrolase gene clusters of seven degradative bacteria isolated from a methyl 

parathion contaminated site and evidence of their horizontal gene transfer. Biodegradation 

17:465-472. 

Zimmer A, Lang D, Richardt S, Frank W, Reski R, and Rensing SA. 2007. Dating the early evolution of 

plants: detection and molecular clock analyses of orthologs. Molecular Genetics and 

Genomics 278:393-402. 

Zuccarello GC, Price N, Verbruggen H, and Leliaert F. 2009. Analysis of a plastid multigene dataset 

and the phylogenetic position of the marine macroalga Caulerpa filiformis (Chlorophyta). 

Journal of Phycology in press. 

Zuker M. 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids 

Research 31:3406-3415. 

Zwickl DJ and Hillis DM. 2002. Increased taxon sampling greatly reduces phylogenetic error. 

Systematic Biology 51:588-598.

Summary 

Green algae are distributed worldwide and can be found in almost every habitat, ranging from polar 

to tropical marine, freshwater and terrestrial environments, and as symbionts. They exhibit a 

remarkable morphological and cytological diversity ranging from unicells and colonial forms, over 

multicellular filaments and foliose blades, to siphonous life forms that are essentially composed of a 

single giant cell containing countless nuclei. Together with land plants, green algae form the green 

lineage or Viridiplantae. Morphological and molecular studies have identified a major split within the 

Viridiplantae giving rise to two monophyletic lineages, the Chlorophyta and the Streptophyta. The 

Streptophyta consists of several lineages of freshwater green algae from which land plants evolved 

approximately 470 million years ago. Whereas considerable progress has been made during the past 

decade in clarifying the relationships among the streptophyte green algae and land plants, the 

phylogeny and evolutionary history of the Chlorophyta has been more difficult to elucidate. 

The Chlorophyta consists of a paraphyletic assemblage of early diverging unicellular green algae, 

termed the prasinophytes, which gave rise to three main clades, the classes Ulvophyceae, 

Trebouxiophyceae and Chlorophyceae (UTC). The relationships among these three classes have been 

at the center of a long-standing debate. Based on ultrastuctural characters, and SSU nrDNA, 

chloroplast and mitochondrial phylogenies all possible relationships between UTC classes have been 

hypothesized. The unstable relationships exhibited among these three classes are likely due to a 

combination of their ancient age and the short time span over which they diverged from one 

another. Determining relationships among the different orders of the Ulvophyceae poses a similar 

problem. Up to now, green algal phylogenies were either based on limited gene or taxon sampling. 

The slow progress in green algal phylogenetics is in part due to difficulties in amplifying single-copy 

nuclear markers, as a consequence of the limited availability of genomic data for green algae and the 

presence of large introns in their genes. 

In Chapter 2, we performed phylogenetic analyses (ML and BI) of seven nuclear genes, SSU nrDNA 

and two plastid markers with carefully chosen partitioning strategies and models of sequence 

evolution. We obtained high support across the topology of the Chlorophyta, show the monophyly of 

the UTC classes and reveal a sister relationship between Chlorophyceae and Ulvophyceae. Even 

though topology tests (AU) do not exclude an alternative branching order of UTC classes, we showed 

that moderate removal of fast-evolving sites improves the phylogenetic signal in the desired epoch. 

We also inferred the relationships among the Ulvophyceae, which was found to consist of two main 

clades. The first clade contains the orders Ulvales and Ulotrichales. The second clade includes the 

early diverging genus Ignatius and a clade comprising the orders Trentepohliales, Bryopsidales, 

Dasycladales and Cladophorales, along with Blastophysa. In this clade, the Bryopsidales and 

Dasycladales are sisters, Blastophysa is most closely related to the Cladophorales, while the 

phylogenetic position of the Trentepohliales remains uncertain. The inferred relationships provide 

novel insights into the evolution of multicellularity and multinucleate cells in the green tree of life. 

Multicellularity evolved multiple times independently in the Streptophyta and Chlorophyta, and at 

least twice in the two main clades of the Ulvophyceae. Siphonous thallus structures are most likely

162 SUMMARY 

derived from a multinucleate and multicellular common ancestor of the Bryopsidales— 

Dasycladales—Cladophorales clade. 

In order to better understand the timeframe in which key evolutionary events took place in the green 

algae, we dated our multilocus phylogeny using relaxed molecular clock methods and the fossil 

record (Chapters 6 and 8). This dated phylogeny shows a rapid radiation of the UTC classes in ca. 20 

my during the first period of the Neoproterozoic (between 900—881 my) and a Neoproterozoic 

diversification of all ulvophycean orders in a timeframe of ca. 130 my (between 826—596 my). 

Guided by this improved green algal phylogenetic tree, we addressed various topics relating to 

molecular evolution of the Chlorophyta. 

In Chapter 3, we studied the distribution and gain-loss patterns of elongation factor genes. 

Elongation factor-1 alpha (EF-1α) and elongation factor-like (EFL), two key genes of the translational 

apparatus, have an almost mutually exclusive distribution in eukaryotes. All streptophytes except 

Mesostigma encode EF-1α. In the Chlorophyta, the prasinophytes, Trebouxiophyceae, Chlorophyceae 

and ulvophycean orders Ulvales and Ulotrichales have EFL. We showed that not only the ulvophyte 

Acetabularia (Dasycladales) but also closely related lineages, i.e. Ignatius, Cladophorales + 

Blastophysa, Bryopsidales and other Dasycladales, possess EF-1α. 

In order to gain more insight in the evolution of EF-1α and EFL in the Viridiplantae we analyzed their 

gain-loss dynamics in a maximum likelihood framework using continuous-time Markov models. These 

models revealed that the presence of EF-1α, EFL or both genes along the backbone of the green plant 

phylogeny is highly uncertain due to sensitivity to branch lengths and lack of prior knowledge about 

ancestral states or rates of gene gain and loss. Model refinements based on insights gained from the 

EF-1α phylogeny reduce uncertainty but still imply several equally likely possibilities: a primitive EF- 

1α state with multiple independent EFL gains or coexistence of both genes in the ancestor of the 

Viridiplantae or Chlorophyta followed by differential loss of one or the other gene in the various 

lineages. 

The genetic code, which translates nucleotide triplets (codons) into amino acids, is virtually identical 

in all living organisms. However, a small number of eubacterial, and eukaryotic nuclear and 

mitochondrial genomes have evolved slight variations on this universal code. A non-canonical code, 

in which TAG and TAA have been reassigned from stop codons to glutamine, has previously been 

reported for the green algal order Dasycladales. Based on the amplified housekeeping nuclear genes, 

we demonstrate in chapter 4 that this non-canonical genetic code is shared with the related clades 

Trentepohliales, Cladophorales and Blastophysa, but not with the sister clade of the Dasycladales, 

the Bryopsidales. We favor a stepwise acquisition model for the evolution of a non-canonical code, 

whereby the alternative codes observed in these green algal orders share a single origin. Stop codon 

reassignment is a gradual process requiring changes to tRNA and eukaryotic release factor (eRF1) 

genes. We suggest that mutations in the anticodons of canonical glutamine tRNAs occurred once 

along the branch leading to the orders Trentepohliales, Dasycladales, Bryopsidales, Cladophorales 

and the genus Blastophysa. The presence of these mutated tRNAs allow TAG and TAA codons to be

SUMMARY 163 

translated to glutamine instead of terminating translation. At this step, the mutated tRNAs compete 

with eRF1 for the TAA and TAG codons. To complete the transition to the non-canonical code, a 

subsequent mutation of eRF1 that prevents binding of eRF1 with TAG and TAA is required. The 

complex distribution of the non-canonical code in the Ulvophyceae could then be explained by three 

independent mutations of eRF1 in the Trentepohliales, Dasycladales and Cladophorales + 

Blastophysa, along with a decrease in importance of the mutated tRNAs or their extinction through 

selection or drift in the branch leading to the Bryopsidales. A detailed comparison of eukaryotic 

release factors (eRF1) and glutamine tRNAs in the respective clades of the Ulvophyceae is, however, 

needed to test this evolutionary scenario. 

In Chapter 5, we studied differences in synonymous codon usage bias and GC content among green 

algae based on our housekeeping nuclear genes and analyzed their evolution in a phylogenetic 

framework. We observe stronger codon usage bias in the ancestral streptophytes Mesostigma and 

Chlorokybus than in the remainder of the Streptophyta. Within the Chlorophyta, the prasinophytes, 

Trebouxiophyceae and Chlorophyceae have markedly stronger codon usage bias than the 

Ulvophyceae. One exception is the ulvophyte Ignatius, which has a markedly stronger codon usage 

bias than other members of the Ulvophyceae. GC content patterns show congruent trends, species 

with strong codon usage bias having high GC content. 

We interpret these results along with the biology of the organisms in the framework of two models: 

the mutation-selection-drift model and the co-evolutionary model of genome composition and 

resource allocation. It is remarkable that unicellular organisms and colony-forming species have 

much more pronounced GC and codon usage biases as compared to multicellular and macroscopic 

species. This may follow from unicells having large population sizes, which leads to more codon 

usage bias due to stronger selection as compared to species with smaller population sizes where drift 

can more rapidly fix mutation. Their large population sizes are revealed by characteristics for rselected 

species: small body size, fast growth rate and short generation times. We observe a negative 

correlation between rate of molecular evolution and codon usage bias and a positive correlation 

between GC content and codon usage bias. 

Chapters 6 and 7 focus on the phylogenetic relationships and evolution within two specific 

ulvophycean clades. In Chapter 6, we dated a five-locus phylogeny of the siphonous orders 

Dasycladales and Bryopsidales using relaxed molecular clock methods calibrated with the fossil 

record. These models indicate a late Neoproterozoic or early Cambrian origin of the Dasycladales and 

Bryopsidales (571 million years [628–510]) and a Paleozoic diversifications of the different families 

within each order. In Chapter 7 we studied phylogenetic relationships within the Cladophorales 

based on small and large subunit nrDNA sequences. We found that Uronema curvatum, a marine 

microfilamentous species placed in the Chlorophyceae, is sister to the rest of the Cladophorales. 

Based on the divergent phylogenetic position of U. curvatum we described a new genus and family of 

Cladophorales (Okellya, Okellyaceae).

Samenvatting 

Groenwieren worden wereldwijd aangetroffen in ongeveer elk mogelijke habitat, zowel in polaire 

gebieden alsook in tropische zeeën, zoetwatermeren, terrestrische omgevingen en zelfs als 

symbiont. Morfologische en cytologische diversiteit tussen groenwieren is groot, gaande van 

eencelligen en kolonies naar meercellige filamenten en bladachtige thalli, tot sifonale levensvormen 

die bestaan uit één enkele reusachtige cel met een ontelbaar aantal kernen. 

Samen met de landplanten vormen de groenwieren de groene planten of Viridiplantae. 

Morfologische en moleculaire studies toonden aan dat de Viridiplantae opgesplitst worden in 2 

monofyletische groepen, de Chlorophyta en Streptophyta. De Streptophyta bestaan uit enkele 

groepen van zoetwater groenwieren die ongeveer 470 miljoen jaar geleden aan de oorsprong lagen 

van de landplanten. De laatste 10 jaar is er veel vooruitgang geboekt in de opheldering van de 

verwantschap tussen de streptophyte groenwieren en de landplanten. De fylogenie en evolutionaire 

geschiedenis van de Chlorophyta blijken echter moeilijker te onderzoeken. 

De Chlorophyta bestaan uit een parafyletische groepering van reeds vroeg gedivergeerde eencellige 

groenwieren, de zogenaamde prasinophyten, die aan de basis liggen van drie belangrijke groepen: de 

klassen Ulvophyceae, Trebouxiophyceae and Chlorophyceae (UTC). De verwantschap tussen deze 

drie klassen is niet eenduidig en vormt de basis voor een langdurig debat. 

Gebaseerd op ultrastructurele kenmerken en fylogenieën op basis van SSU nrDNA, chloroplast en 

mitochondriale genen, zijn alle mogelijke verwantschappen tussen de UTC klassen vastgesteld. Deze 

onstabiele verwantschappen worden waarschijnlijk veroorzaakt door een combinatie van de 

ouderdom van deze klassen en de korte tijdsspanne waarin ze van elkaar divergeerden. Het bepalen 

van de verwantschap tussen de verschillende ordes binnen de Ulvophyceae vormt een gelijkaardig 

probleem. Totnogtoe waren groenwier fylogenieën steeds gebaseerd op een beperkt aantal genen of 

taxa. De trage vooruitgang in het ophelderen van de groenwier fylogenie is waarschijnlijk te wijten 

aan moeilijkheden bij de amplificatie van nucleaire merkers, waarvan slechts een kopie per genoom 

aanwezig is als gevolg van de beperkte beschikbaarheid van genomische data voor groenwieren en 

de aanwezigheid van lange introns. 

In hoofdstuk 2, voeren we een fylogenetische analyse (ML and BI) uit op basis van zeven nucleaire 

genen, SSU nrDNA en twee chloroplast merkers waarbij veel aandacht gaat naar de 

partitioneringsstrategie en de keuze van evolutionaire sequentiemodellen. We verkrijgen zo een 

goede ondersteuning van de volledige fylogenie, tonen de monofylie van de UTC klassen aan en 

bewijzen een zusterverwantschap tussen de Chlorophyceae en Ulvophyceae. Alhoewel dat topologie 

testen (AU) een alternatieve vertakkingvolgorde tussen de UTC klassen niet uitsluit, tonen we aan dat 

het verwijderen van een beperkt aantal snel evoluerende posities het fylogenetisch signaal versterkt 

in de gewenste periode. 

De studie toont eveneens de verwantschap aan tussen de verschillende ordes binnen de 

Ulvophyceae. Deze bestaan uit twee groepen. De eerste groep omvat de Ulvales en Ulotrichales. De 

tweede groep bestaat uit het reeds vroeg ontstane genus Ignatius en uit een groep die de ordes 

Trentepohliales, Bryopsidales, Dasycladales en Cladophorales, samen met het genus Blastophysa

166 SAMENVATTING 

omvat. Binnen deze laatste groep zijn de Bryospsidales en Dasycladales zusters, Blastophysa is het 

nauwst verwant aan de Cladophorales, terwijl de fylogenetische positie van de Trentepohliales 

onzeker blijft. 

De afgeleide verwantschappen verschaffen ons nieuwe inzichten over de evolutie van meercellige 

organismen en cellen met meerdere kernen. Meercellige organismen zijn verschillende keren 

onafhankelijk van elkaar ontstaan: minstens eenmaal in de Streptophyta en eenmaal in Chlorophyta 

en tenminste eenmaal in beide hoofdgroepen binnen de Ulvophyceae. Sifonale thallus structuren zijn 

hoogstwaarschijnlijk afgeleid van een multinucleate, meercellige voorouder van de Bryopsidales— 

Dasycladales—Cladophorales groep. 

Om de evolutionaire tijdschaal waarbinnen deze gebeurtenissen plaatsvonden beter te begrijpen, 

hebben we onze multilocus groenwier fylogenie gedateerd aan de hand van ‘relaxed molecular clock’ 

methodes en fossiele data (hoofdstukken 6 en 8). Deze gedateerde fylogenie geeft een snelle 

radiatie van de UTC klassen aan in ca. 20 miljoen jaar (mj) gedurende de eerste periode van het 

Neoproterozoic (tussen de 900—881 mj) en een Neoproterozoische diversificatie van alle 

ulvophyceae ordes in een tijdsspannen van ca. 130 mj (826—596 mj). 

In het kader van deze verbeterde groenwier fylogenie worden verschillende topics gerelateerd aan 

moleculaire evolutie binnen de Chlorophyta besproken. 

In hoofdstuk 3 bestuderen we de distributie en evolutie van ‘elongation factor’ genen. Elongation 

factor-1 alpha (EF-1α) en elongation factor-like (EFL), twee sleutelgenen van het translatie apparaat, 

hebben een onderling exclusief distributiepatroon in eukaryoten. Alle Streptophyta, behalve 

Mesostigma, hebben EF-1α. In de Chlorophyta hebben de prasinofyten, Trebouxiophyceae, 

Chlorophyceae en ulvophyte ordes Ulvales en Ulotrichales het EFL gen. Wij tonen aan dat niet alleen 

de ulvofyt Acetabularia (Dasycladales), maar ook nauw verwante groepen, in het bijzonder Ignatius, 

Cladophorales + Blastophysa, Bryopsidales en andere Dasycladales, EF-1α bezitten. 

Om meer inzicht te krijgen in de evolutie van EF-1α en EFL binnen de Viridiplantae, hebben we de 

dynamiek van winst en verlies van deze genen in een maximum likelihood kader geanalyseerd met 

behulp van ‘continuous-time Markov’ modellen. Deze modellen tonen aan dat de aanwezigheid van 

EF-1α, EFL of beide genen langsheen de fylogenie van de groene planten onzeker is tengevolge van 

gevoeligheid voor taklengtes en een gebrek aan voorafgaande kennis omtrent voorouderlijke 

kenmerktoestanden en de snelheid van aanwinst en verlies van deze genen. Inzichten afgeleid van de 

EF-1α fylogenie maakten het mogelijk om ons model te verfijnen en verminderden de onzekerheid 

omtrent de aanwezigheid van EF-1α, EFL of beide genen. Desondanks blijven er nog steeds 3 

mogelijke scenario’s over: EF-1α als voorouderlijke toestand met meerdere onafhankelijke 

aanwinsten van EFL of een co-existentie van beide genen in de voorouder van de Viridiplantae of 

Chlorophyta gevolgd door differentieel verlies van een van beide genen in de verschillende 

evolutionaire lijnen. 

De genetische code, die nucleotide triplets (codons) vertaalt naar aminozuren, is bijna identiek in alle 

levende organismen. Een klein aantal eubacteriële en eukaryote genomen ontwikkelden echter

SAMENVATTING 167 

kleine variaties op deze universele code. Een alternatieve code, waarbij TAG en TAA veranderen van 

stop codons naar glutamine coderende codons, was reeds aangetoond bij de groenwier orde 

Dasycladales. Op basis van de geamplificeerde nucleaire huishoudgenen, tonen we in hoofdstuk 4 

aan dat deze alternatieve code ook voorkomt bij de nauw verwanten groepen Trentepohliales, 

Cladophorales en Blastophysa, maar niet bij de zustergroep van de Dasycladales, de Bryopsisales. 

We verkiezen een ‘stepwise acquisition’ model voor de evolutie van de alternatieve code, waarbij de 

alternatieve code, geobserveerd in deze groenwier ordes, een gemeenschappelijk oorsprong heeft. 

De verandering van stop codon naar aminozuur coderend codon is een geleidelijk proces dat tevens 

verandering in tRNAs en ‘eukaryotic release factor’ (eRF1) genen vergt. We veronderstellen dat 

mutaties in het anticodon van gewone glutamine tRNAs eenmaal plaatsvond in de tak leidend naar 

de groepen Trentepohliales, Dasycladales, Bryopsidales, Cladophorales and the genus Blastophysa. 

De aanwezigheid van gemuteerde tRNAs (met anticodons complementair aan TAG en TAA) maakt 

het dan mogelijk dat TAG en TAA codons worden vertaald in glutamine in plaats van de translatie te 

beëindigen. Op dat moment zijn de gemuteerde tRNAs in competitie met eRF1 om te binden op TAG 

en TAA codons. Om de overgang naar een alternatieve code te vervolledigen is een mutatie in het 

eRF1 gen, die de binding met TAG en TAA verhindert, noodzakelijk. De complexe distributie van de 

alternatieve code binnen de Ulvophyceae kan verklaard worden door drie onafhankelijk mutaties van 

eRF1 in de Trentepohliales, Dasycladales, Cladophorales + Blastophysa tezamen met een afname in 

belang of verlies van de gemuteerde tRNAs door selectie of genetische drift in de tak leidend naar de 

Bryopsidales. Een nauwgezette vergelijking van eRF1 genen en glutamine tRNAs in de verschillende 

groepen binnen de Ulvophyceae is echter noodzakelijk om dit scenario te testen. 

In hoofdstuk 5 bestuderen we verschillen in codon gebruik en GC % tussen groenwieren gebaseerd 

op onze huishoudgenen en analyseren we de evolutie van codon gebruik en GC % in een 

fylogenetisch kader. We stellen een sterkere bias in codon gebruik vast bij de streptofyten 

Mesostigma en Chlorokybus dan bij de rest van de Streptophyta. Binnen de Chlorophyta hebben de 

prasinofyten, Trebouxiophyceae en Chlorophyceae veel meer bias in codon gebruik dan de 

Ulvophyceae. Een uitzondering is de ulvofyt Ignatius die veel meer bias in codon gebruik heeft dan 

de rest van de Ulvophyceae. GC waarden vertonen een gelijkaardig patroon, soorten met veel bias in 

codon gebruik hebben een hoger GC gehalte. 

Deze resultaten worden gerelateerd aan de biologie van de organismen door middel van twee 

modellen: het mutatie-selectie-drift model en het co-evolutie model tussen genoomsamenstelling en 

bouwsteen allocatie. Het valt op dat unicellulaire organismen en kolonievormende soorten veel meer 

bias in GC % en codon gebruik vertonen dan multicellulaire en macroscopische organismen. Dit kan 

het gevolg zijn van de grotere populatiegrootte van unicellulaire organismen waarbij een hogere 

selectiedruk leidt tot een sterkere bias in codon gebruik. Bij soorten met een kleine populatiegrootte 

zal de genetische drift er namelijk voor zorgen dat mutaties snel fixeren. De grote populatiegrootte is 

kenmerkend voor r-geselecteerde soorten. R-geselecteerde sooten zijn tevens klein, groeien snel en 

hebben een korte generatietijd. We stellen een negatieve correlatie vast tussen de snelheid van 

moleculaire evolutie en bias in codon gebruik enerzijds en een positieve correlatie tussen GC % en 

bias in codon gebruik anderzijds.

168 SAMENVATTING 

In hoofdstukken 6 en 7 ligt de nadruk op de fylogenetisch verwantschap en evolutie binnen enkele 

ulvophyt groepen. Gebaseerd op 5 loci, dateren we in hoofdstuk 6 een fylogenie van de sifonale 

ordes Dasycladales en Bryopsidales gebruik makend van ‘relaxed molecular clock’ methodes 

gekalibreerd met fossiele gegevens. Volgens deze modellen ontstonden de Dasycladales en 

Bryopsidales in het late Neoproterozoic of het vroege Cambrium (571 miljoen jaar [628–510]) en 

diversifieerden de verschillende families binnen deze ordes gedurende het Paleozoic. In hoofdstuk 7 

besturen we de fylogenetische verwantschappen binnen de Cladophorales gebaseerd op small en 

large subunit nrDNA sequenties. We stellen vast dat Uronema curvatum, een mariene soort 

bestaande uit kleine filamenten die normaalgezien binnen de Chlorophyceae wordt geplaatst, nauw 

verwant is aan de rest van de Cladophorales. Gebaseerd op deze afwijkende fylogenetische positie 

beschrijven we een nieuw genus en familie binnen de Cladophorales voor U. curvatum (Okellya, 

Okellyaceae).

Phylogeny and molecular evolution of green algae - Phycology ...

Create successful ePaper yourself

Delete template?

Save as template?