| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Invited Special Papers |
2Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, Connecticut 06269 USA; 3Department of Botany, Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103 USA
Received for publication January 15, 2004. Accepted for publication June 15, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONTHE DATA REVOLUTION(S)MAJOR CLADES OF GREEN...RELATIONSHIPS OF MAJOR GROUPS...THE ORIGIN OF LAND...CONCLUSIONSLITERATURE CITED |
|---|
Key Words: Chlorophyta • Charophyta • DNA • Mesostigma • Streptophytina • ultrastructure
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONTHE DATA REVOLUTION(S)MAJOR CLADES OF GREEN...RELATIONSHIPS OF MAJOR GROUPS...THE ORIGIN OF LAND...CONCLUSIONSLITERATURE CITED |
|---|
), a field that was undergoing rapid and fascinating changes, both in content and theory. "The present period may be termed the ‘Age of Ultrastructure’ in green algal systematics," wrote Frank Round (1984
, p. 7) in the introductory chapter, which summarized the history and state of the art. Round (1984)
argued that light microscopy had laid the foundation in the preceding two centuries, but that the foundation was largely descriptive—alpha taxonomy in the most restricted sense. Ultrastructure, he asserted, had enlarged and presumably would continue to expand our horizons to unify systematics of green algae and overcome the fragmented alpha taxonomy that had dominated the field. Little did Round know that this golden age of green algal systematics was about to go platinum. Molecular systematics, in concert with a rigorous theoretical approach to data analysis and hypothesis testing (Theriot, 1992
; Swofford et al., 1996
), would at first complement and then transform the age of ultrastructure and usher in the "Age of Molecules." In this article, we review the major advances in green algal systematics in the past 20 years, with a focus on well-supported, monophyletic taxa and the larger picture of phylogeny and evolution of green algae. We will review the types of data that have fueled these advances. As will become obvious, this perspective entails discussion of some embryophytes as well as their closest green algal relatives. In addition, we will point out major uncertainties in green algal systematics, which pose some of the most provocative areas for further research.
Deconstructing hypotheses of relationships of the green algae and land plants
A link between green algae and land plants has been clear to biologists for centuries, since before Darwin and the advent of evolutionary thinking and phylogenetics (Smith, 1950
; Prescott, 1951
). Recent new data on morphology, genes, and genomes, as well as new ways of analyzing and synthesizing information, are only the most recent in a long history of change in our understanding of these so-called "primitive" plants. This review focuses primarily on research that has led to both some radical restructuring of the classification of algae and some satisfying confirmations of the careful observations of earlier workers.
First and foremost, green algae, the division Chlorophyta of Smith (1950)
, are undoubtedly monophyletic with embryophyte green plants, although the Chlorophyta in this sense is paraphyletic (Mattox and Stewart, 1984
; Mishler and Churchill, 1985
; McCourt, 1995
). Embryophytes (land plants; bryophytes and vascular plants) are clearly descended from green algal-like ancestors, but the sister of the embryophytes includes only a few green algae. The remainder of Chlorophyta constitutes a monophyletic group. This major bifurcation in green plant evolution implies a single common ancestor to the two lineages, but, given the diversity of unicellular green algae and our growing understanding of them, there may be additional lineages outside this major bifurcation.
What are green algae?
The term algae is not phylogenetically meaningful without qualifiers. Algae in general and green algae in particular are difficult to define to the exclusion of other phylogenetically related organisms that are not algae. This difficulty is a reflection of recent data on algae as well as the way phylogenetic thinking has permeated classification. Green algae are photosynthetic eukaryotes bearing double membrane-bound plastids containing chlorophyll a and b, accessory pigments found in embryophytes (beta carotene and xanthophylls), and a unique stellate structure linking nine pairs of microtubules in the flagellar base (Mattox and Stewart, 1984
; Sluiman, 1985
; Bremer et al., 1987
; Kenrick and Crane, 1997
). Starch is stored inside the plastid and cell walls when present are usually composed of cellulose (Graham and Wilcox, 2000a
).
The plastids of green algae are descended from a common prokaryotic ancestor (Delwiche, 1999
; Delwiche et al., 2004
), for which descendants are endosymbiotic in the host cells of a number of other eukaryotic lineages. These plastids are termed primary, i.e., derived directly from a free-living prokaryotic ancestor (Delwiche and Palmer, 1997
; Delwiche, 1999
), although a secondary origin has been proposed (Stiller and Hall, 1997
; see Keeling, 2004
, in this issue for an overview of this process and variations on the theme of endosymbiosis). Plastid-bearing lineages permeate all of the other major clades of algae (see also in this issue Andersen, 2004
; Hackett et al., 2004
; Saunders and Hommersand, 2004
).
Green algal diversity
Mostly microscopic and rarely more than a meter in greatest dimension, the green algae make up for their lack in size with diversity of growth habit (Figs. 1– 17) and fine details of their cellular architecture. Body (thallus) size and habit ranges from microscopic swimming or nonmotile forms (e.g., nanoplankton, benthos, or lichen phycobionts) to macroscopic (benthic attached forms). Thallus structure runs the gamut of complexity, from swimming and nonmotile unicells, to filaments, colonies, and various levels of tissue organization (pseudoparenchymatous, parenchymatous, or thalloid) and branching morphologies. Unicells are spherical to elongate, with or without flagella, scales, and wall layers or other coverings (e.g., loricas). Filaments generally exhibit cylindrical cells arranged end-to-end, although chains of irregularly shaped cells are known. Unbranched (Oedogonium) and branching (Draparnaldia) forms are known, and many branching forms have attenuated terminal filament tips (Chaetophora). Colonies of various sizes occur, from pairs of cells (Euastropsis) to thousands (Hydrodiction). Cells in colonies may be joined by gelatinous strands or share a common parental wall. Colonies range in form from small sarcinoid packets (nonlinear clusters of cells; Chlorokybus) to aggregates of thousands of swimming cells (Volvox). Branching forms may be simple bifurcating or reticulating networks of filaments, but a few achieve a complexity that can be called tissuelike (Nitella). Cells may be uninucleate or coenocytic, in which many nuclei are dispersed throughout the cytoplasm of so-called giant cells (Caulerpa).
|
|
THE DATA REVOLUTION(S) |
|---|
|
TOPABSTRACTINTRODUCTIONTHE DATA REVOLUTION(S)MAJOR CLADES OF GREEN...RELATIONSHIPS OF MAJOR GROUPS...THE ORIGIN OF LAND...CONCLUSIONSLITERATURE CITED |
|---|
). Light microscopy opened the windows on the microstructures of algae and provided the bulk of observational data for a long time. As described earlier, electron microscopy and molecular data have revolutionized our understanding of green algal phylogeny, which is reflected in modern classification. Table 1 lists the features and sources of morphological and genetic data, along with representative publications and reviews.
|
). Motile ancestral green algal unicells were hypothesized to have given rise to distinct lines of increasing size and complexity, with each exhibiting a variation on a theme. One line resulted in motile colonies, another in nonmotile branching thalli or colonies (with motile reproductive cells retained in the life cycle), and a third in large nonmotile coenocytic forms (retaining motile uninucleate gametes; Smith, 1950
; Bold and Wynne, 1985
; McCourt, 1995
).
The view that the cellular features involved in the vital processes of cell division and swimming (of gametes or asexual zoospores) would be highly conserved evolutionarily led to numerous comparative studies targeting the mitotic, cytokinetic, and swimming apparatus of the cell (e.g., Stewart and Mattox, 1975
). The flagellar apparatus, with its flagellar basal bodies and axonemes and rootlets of microtubules, has been painstakingly compared across a large number of green algae. To a lesser extent, plastid structure has been important for diagnosis of some groups. This focus came about through research by early workers (Pickett-Heaps and Marchant, 1972
, and others; Table 1) that revealed evolutionarily conservative characters that cut across misleading convergence in vegetative morphology. In the age of molecular systematics, we are evaluating hypotheses formulated from comparative ultrastructure studies over the last 30 years, and adding new hypotheses as well.
Ultrastructural work showed that filamentousness, coloniality, coccoid habit (nonmotile, lacking flagella), and many other vegetative features evolved numerous times and were generally unreliable as characters marking monophyletic groups. The simple sequencing of forms, flagellate 
coccoid unbranched filament branched filament tissuelike (with a cul de sac towards siphonous thalli from coccoids), may have occurred repeatedly in many lineages (Mattox and Stewart, 1984
; Round, 1984
).
Close on the heels of this first wave of new data and new methods of analysis employing a phylogenetic approach (Theriot, 1992
) came a molecular revolution. By and large, the refutation of the classical tendencies as organizing principals in algal evolution and classification was confirmed, and a new dogma arose, which has been refined and enlarged but retains the major scaffolding erected by ultrastructure and biochemistry.
Molecular studies in green algae have been largely driven by slightly earlier studies in embryophytes and, to a lesser extent, cyanobacteria (Palmer, 1985
). This is due largely to the development of primers for many genes that were first studied in embryophytes (Table 1). The green alga lineage (eukaryotes containing primary green plastids) originated as much as 1500 million years ago (mya; Yoon et al., 2004
), and the divergence of land plants occurred perhaps 700 mya (Heckman et al., 2001
) or more likely 425–490 mya (Sanderson, 2003
). Despite the antiquity of a shared common ancestor of land plants and their nearest relatives (Karol et al., 2001
), ribosomal DNA and many plastid genes are recognizable as homologs in green plants and algae. As a result, molecular phylogenetics of green algae expanded rapidly as methodologies and approaches were transferred to algal taxa.
The complete nuclear genome for Chlamydomonas reinhardtii (Grossman et al., 2003
) was released in early 2004, and annotation is an ongoing process (E. Harris, Duke University, personal communication). Organellar genomes to date include five plastid (Wakasugi et al., 1997; Turmel et al., 1999b
; Lemieux et al., 2000
; Maul et al., 2002
; Turmel et al., 2002b
) and 10 mitochondrial genomes (Gray, 1993, direct submission to NCBI; Wolff et al., 1993; Denovan-Wright et al., 1998
; Kroyman and Zetsche, 1998
; Turmel et al., 1999a
; Nedelcu et al., 2000
; Turmel et al., 2002b
, 2002c
; Turmel et al., 2003
), and several other green algal plastid or mitochondrial genomes should soon be published (C. Lemieux, Lavall University, personal communication). Several complete genome sequencing projects are also underway (A. Grossman, Stanford University; B. Palenik, Scripps Institution of Oceanography, personal communications).
With the advent of molecular data has come the possibility of finding characters that transcend rampant morphological homoplasy revealed by ultrastructure. However, a recurrent theme of many molecular studies has been a pattern of relatively well-resolved distal branches of a phylogenetic tree with weak or poor support for the internal or deeper divergences. Chapman et al. (1998)
blamed this result on the limitations of sequence data and the antiquity of green algae: a stochastically changing molecule cannot be expected to retain enough signal to resolve events that happened nearly simultaneously in the ancient past (Lanyon, 1988
). We suspect that a major reason is that most studies have employed only one or two genes and that more data can resolve these major ambiguities.
|
MAJOR CLADES OF GREEN ALGAE |
|---|
|
TOPABSTRACTINTRODUCTIONTHE DATA REVOLUTION(S)MAJOR CLADES OF GREEN...RELATIONSHIPS OF MAJOR GROUPS...THE ORIGIN OF LAND...CONCLUSIONSLITERATURE CITED |
|---|
).
|
A working classification of green algae and plants
For the purposes of this review, we will use the classification shown in Table 2, which gives division, class, and order names of major groups with informal names in parentheses. This is not intended to be a definitive taxonomic revision of green algal classification, but we anticipate that such a revision will incorporate the basic scheme used here (C. F. Delwiche, University of Maryland, personal communication). The use of some terms or prefixes (e.g., charo-) is inevitably confusing because of the historical claim that such terms have on us. In other cases, paraphyly of traditional classes, orders, families, genera, or even species, makes classification difficult.
|
assigned the groups class-level rank, i.e., Chlorophyceae, Pleurastrophyceae (now at least in part Trebouxiophyceae), and Ulvophyceae. All members of this clade have swimming cells with two or four anterior flagella. Within the cell, the flagellar basal bodies are associated with four microtubular rootlets. All three groups share the character of cruciately arranged rootlets that alternate between two and higher numbers of microtubules (X-2-X-2). From group to group, differences in the offset from this cruciate pattern are seen, and arrangements are categorized according to relative positioning (O'Kelly and Floyd, 1984
). When viewed from above the cell, the basal bodies and rootlets can have a perfect cruciate pattern (i.e., with basal bodies directly opposed, DO) or they are offset in a counterclockwise (CCW) or clockwise (CW) position. The ancestral condition has been inferred to be a CCW offset (Trebouxiophyceae and Ulvophyceae; Mattox and Stewart, 1984
), with CW and DO the derived states (chlorophytes).
Molecular work, primarily on the small subunit of ribosomal DNA (18S rDNA) has strongly supported the monophyly of this triad of green algal groups and shows the ulvophytes as sister to a clade containing chlorophytes and trebouxiophytes. A combined analysis of morphology and 18S rDNA data strongly supported the monophyly of the three groups (Mishler et al., 1994
). Later studies provided more evidence for this topology, often with high support, while greatly increasing taxon sampling (Friedl and Zeltner, 1994
; Friedl, 1995
; Bhattacharya et al., 1996
; Krienitz et al., 2001
).
Charophyte clade
This group contains a number of green algae plus a large number of what are considered to be the mostly highly derived green autotrophs, the land plants (Graham, 1993
). Nomenclaturally, the group has led a confused life. Mattox and Stewart (1984)
placed the algae in this group in Charophyceae, although the exclusion of the land plants made this taxonomic arrangement paraphyletic. Graham and Wilcox (2000a)
acknowledged the paraphyly of the group and used the term "charophyceans" to refer to them. Bremer et al. (1987)
assigned the division name Streptophyta to the green algae plus land plants, although Jeffrey (1982)
had used this name more restrictedly, including only stoneworts (Charales) and embryophytes (archegoniate land plants). We will refer to the green algal groups of the charophyte clade as charophyte algae. The clade (charophyte + embryophytes) is characterized by biflagellate cells (when motile cells are present), with asymmetrically inserted flagella and two dissimilar flagellar roots (including a multilayered structure, or MLS, and a smaller root), persistent mitotic spindles, open mitosis, and several enzyme systems not found in other green algae (Mattox and Stewart, 1984
; Graham and Wilcox, 2000a
).
|
RELATIONSHIPS OF MAJOR GROUPS OF GREEN ALGAE |
|---|
|
TOPABSTRACTINTRODUCTIONTHE DATA REVOLUTION(S)MAJOR CLADES OF GREEN...RELATIONSHIPS OF MAJOR GROUPS...THE ORIGIN OF LAND...CONCLUSIONSLITERATURE CITED |
|---|
Prasinophytes
Prasinophyte algae have received attention recently because they are some of the important bloom-forming marine planktonic algae (O'Kelly et al., 2003
) that are known mixotrophs and can account for a significant component of biomass in marine planktonic systems (Diez et al., 2001
). Prasinophyceae also occupy a critical position at the base of the green algal tree of life and are viewed as the form of cell most closely representing the first green alga, or "ancestral green flagellate" (AGF). One of the first green algae to be examined using transmission electron microscopy was a member of class Prasinophyceae (Manton and Parke, 1960
). Prasinophyceae (Micromonadophyceae of Mattox and Stewart, 1984
) was proposed to segregate flagellate unicellular green algae with organic body scales from the other green flagellates (Christiansen, 1962
; Moestrup and Throndsen, 1988
; Melkonian, 1990c
).
Members of Prasinophyceae have diverse morphologies, ranging from bean- to star-shaped cells with one to eight flagella often inserted in a flagellar pit and up to seven distinct types of organic scales. Recently, coccoid forms have been added to this group (Fawley et al., 2000
). They occur in marine and brackish water, although some members live in freshwater habitats (e.g., Pedinomonas, Mesostigma). Prasinophyte algae are among the smallest of the eukaryotic planktonic marine flagellates (Zignone et al., 2002
). Sexual reproduction has only been demonstrated in Nephroselmis olivacea (Suda et al., 1989
).
Scale morphology has been used to differentiate the major groups of prasinophytes (Norris, 1980
; Melkonian, 1984
; Moestrup, 1984
). In all but a few genera, organic scales are produced in the Golgi apparatus and coat the body and flagella. Some taxa possess up to seven distinct scale types, but others have a single type of scale. Flagellar behavior and morphology also reflect a tremendous diversity in this group. Some prasinophyte taxa have cruciately arranged rootlets, but others have asymmetrical rootlets. Some cells push with undulating flagella, and others swim with flagella forward. The number of rootlets varies between two and four and can include associated system I (SMAC) and system II (rhizoplast) fibers. Pyramimonas octopus has eight flagella and at least 60 flagellar structures connecting the basal bodies (Moestrup and Hori, 1989
). Multilayered structures (MLSs) are found in flagellate sperm cells of embryophytes and occur in Mesostigma, a putative charophyte alga, and Pyramimonadales but are absent from all other groups of prasinophytes and all other taxa in the chlorophyte clade (assuming that the MLS-like structures of Trentopohliales are not homologous; Chapman, 1984
). MLSs have also been described in dinoflagellates and jakobids (Wilcox, 1989
; O'Kelly, 1992
), and if these are homologous to those in green plants, then the presence of an MLS can be interpreted as the ancestral condition in the green algae. The structure of the flagellar transition region has been used to distinguish among some of the orders and provides evidence for the relationship between Mesostigma and charophytes (Melkonian, 1984
). Characteristic flagellar hairs can also be used to distinguish among the main groups (Marin and Melkonian, 1994
). Lateral flagellar hairs occur in all studied members, but tip hairs do not occur on flagella of Chlorodendrales and Nephroselmis. Ultrastructural differences in the manner of mitosis and cytokinesis are also apparent. Most members have persistent telophase spindles, the exception being members of Chlorodendrales.
The diversity in cell shape, number of flagella, flagellar apparatus organization, organic scales covering the cell surface and flagella, and differences in cell division led to the conclusion that Prasinophyceae as proposed are not a monophyletic group (Mattox and Stewart, 1984
; Mishler and Churchill, 1984
, 1985
; but see Melkonian, 1984
). Various taxonomic treatments based on morphological and ultrastructural data have been proposed. Moestrup (1991)
segregated the uniflagellate, nonscaled taxa Pedinomonas, Marsupiomonas, Resultor, and Scourfieldia sp. M561 into the class Pedinophyceae. Moestrup and Throndsen (1988)
reclassified Prasinophyceae (excluding Micromonas).
Such dramatic morphological variation in unicells led to striking differences in opinion regarding the AGF. Some authors hypothesized that smaller taxa such as Pedinomonas, with their simpler flagellar apparatus and scales, were the best candidates (Moestrup, 1991
). Other authors proposed that the AGF was a larger, more complex, and multiflagellate taxon. O'Kelly (1992)
suggested that the complex flagellar apparatuses found in the larger, multiflagellate taxa were adaptations for prey capture in these algae or in their recent ancestors. Food particle ingestion and digestion has been shown in members of the Pyramimonadales (Bell and Laybourn-Parry, 2003
) and has also been suggested (Delwiche, 1999
) as a means of obtaining a cyanobacterial symbiont that ultimately became permanently incorporated as the plastid.
The advent of molecular systematics permitted evaluation of many morphologically and ultrastructurally based hypotheses regarding diversity and evolution of scaly, green flagellates, most of which are detailed in three comprehensive reviews (Melkonian, 1984
; O'Kelly, 1992
; Sym and Pienaar, 1993
). Molecular analyses of 18S rDNA data have echoed the morphological diversity seen in prasinophytes, identifying at minimum seven separate lineages that form a grade at the base of the green tree of life (Kantz et al., 1990
; Marin and Melkonian, 1994
; Steinkötter et al., 1994
; Nakayama et al., 1998
; Fawley et al., 2000
; Zignone et al., 2002
). The major lineages that have been recovered by molecular data are discussed next (Fig. 18). It is evident that most (perhaps all) of the remaining orders require further attention and should be reclassified.
Pyramimonadales
This monophyletic order is usually considered as sister to the rest of Prasinophytes and includes the quadriflagellate taxa Halosphaera, Cymbomonas, Pyramimonas, and Pterosperma. Some taxa have complicated scales. Moestrup et al. (2003)
detailed the ultrastructural evidence for a close relationship of Cymbomonas tetramitiformis with Halosphaera. However, these authors also concluded that Cymbomonas possesses scales similar to Mamiella. It may be that the presence/absence of scales or a theca is a phylogenetically informative character, whereas scale morphology is not phylogenetically useful. Some members of this order, such as Halosphaera, Pterosperma, and Cymbomonas (Moestrup et al., 2003
), are known to produce a cyst or phycoma stage, which at least in Cymbomonas contains two chloroplasts and thus is likely the result of sexual reproduction. The resistant walls of phycomata appear in the fossil record in the Early Cambrian and perhaps earlier (Tappan, 1980
).
Mamiellales
This order includes some of the smallest eukaryotes known, e.g., Crustomastix (3–5 µm long). Cells have one or two laterally inserted flagella and a total of two flagellar roots per cell. Some members lack scales, but most have a single form of scale over the body and flagella (Nakayama et al., 2000
). Molecular data indicate that this group also includes coccoid taxa. Presumably all taxa contain the pigment prasinoxanthin, but Zignone et al. (2002)
identified siphonoxanthin in Crustomastix. Mamiellales are usually reconstructed as sister to the rest after Pyramimonadales.
Mesostigmatophyceae
Mesostigma viride is the only member of this class. This asymmetrical cell with two laterally inserted flagella was recently placed in a separate class with the charophyte Chaetosphaeridium based on molecular and cellular evidence (e.g., presence of similar maple-leaf-shaped scales on flagella). The flagellar rootlet system of Mesostigma has an MLS, unlike most other prasinophytes (except for Pyramimonadales). Although a separate class is warranted, the inclusion of Chaetosphaeridium has been refuted. The placement of Mesostigma is further discussed in the section on relationships of the charophyte algae and embryophytes.
Pedinophyceae
This class includes Pedinomonas and Resultor, tiny (<3 µm) uniflagellate cells without scales. The flagellar apparatus is unusual among prasinophytes, consisting of one emergent flagellum and an additional basal body. The rootlets have an X-2-X-2 pattern, and the two basal bodies are offset in a counterclockwise orientation (Moestrup, 1991
). Dividing cells have a persistent telophase spindle. Given this combination of characteristics, the phylogenetic position of Pedinophyceae still remains a puzzle. Contrasting hypotheses are that Pedinomonas represents either a reduced member of Mamiellales or a reduced ulvophycean taxon (Melkonian, 1990b
). The only molecular study to include Pedinomonas placed it as sister of the green algae (Kantz et al., 1990
), but because this analysis lacked an outgroup, there is no way to interpret its phylogenetic position. To date, no phylogenetic studies have included Resultor, although one rbcL (Rubisco large subunit) sequence from this taxon is accessioned in GenBank (www.ncbi.nlm.nih.gov).
Pseudoscourfieldiales
Members of this order (Pseudoscourfieldia marina, Nephroselmis pyriformis, and N. olivacea) have two flagella of unequal length, two body scale layers, and two flagellar scale layers. The flagellar apparatus is complex, having three flagellar roots. Molecular data resolve this group into either one or two clades of flagellate and coccoid taxa, both of which are usually placed as sister to Mamiellales.
Chlorodendrales
There is strong support for the monophyly of this order of flagellates and evidence that they share characters with the UTC clade, rather than with the other Prasinophyceae. The members of this group are distinct from other prasinophytes in many morphological and ultrastructural features: they possess a metacentric spindle that collapses at telophase; paired flagella that beat in a breast stroke pattern; and an X-2-X-2 configuration of the flagellar rootlets. These algae are also distinct in possessing a theca, which is formed by fusion of the outer layer of scales. The striking similarity between some members of the UTC clade and Chlorodendrales led Mattox and Stewart (1984)
to reclassify Tetraselmis into a new class Pleurastrophyceae, along with Pleurastrum, Trebouxia, and Pseudotrebouxia (the latter genera are now placed in Trebouxiophyceae, Friedl, 1995
, 1996
). Analyses of molecular data always place Tetraselmis as sister to the morphologically more complex UTC clade.
Coccoids
In addition to the aforementioned lineages, Fawley et al. (2000)
identified at least two well-supported, yet unnamed, clades of coccoid taxa from existing culture collection isolates that had not previously been sampled. These taxa were suspected members of the Prasinophyceae because they possess prasinophyte-specific pigments. It is evident that molecular data have corroborated and even enhanced our understanding of the diversity and nonmonophyly of prasinophyte algae evident from ultrastructural data. Molecular data also have been particularly important in demonstrating that the earliest prasinophyte lineages (closest relatives of the AGF) were large, complex, scaly, and multiflagellate rather than small and naked. Molecular analyses have also confirmed a close phylogenetic relationship of Tetraselmis and the UTC clade. As more taxa and genes are sampled, we are gaining greater detail about the evolution of this grade of green algae; however, important questions remain about the influence of data analysis, data set composition, and taxon sampling on the phylogenetic placement of these individual lineages and taxa. Taxonomic revisions to classify the phylogenetic lineages now treated as orders within the Prasinophyceae should be forthcoming as greater resolution is achieved and analyses based on data from more than a single gene are published.
Ulvophytes
This group was named the class Ulvophyceae by Mattox and Stewart (1984)
, but because of the uncertain status of the class as a clade, we will refer to the group collectively as ulvophytes. Ulvophytes are diverse morphologically and ecologically and comprise some of the more strikingly beautiful green algae, extant or otherwise (Berger and Kaever, 1992
). The group is predominantly marine and includes some of the best-known green seaweeds, such as the sea lettuce Ulva (Hayden and Waaland, 2002
), the weedy Codium (Goff et al., 1992
), Caulerpa (Meinesz, 1999
), and the model organism Acetabularia (Mandoli, 1998
). Other filamentous genera dominate localized freshwater habitats (Cladophora, Rhizoclonium, and Pithophora), sometimes to the detriment of human use (Lembi and Waaland, 1988
).
Thallus forms include nonmotile unicells, branched and unbranched filaments, filmy membranes (mono- or distromatic), and cushiony forms of compacted tubes. Many ulvophytes have multinucleate thalli and a siphonous construction, i.e., with few or no cross walls, which makes the thallus one giant, multinucleate cell. The surfaces of thalli of many marine ulvophytes are lightly to heavily calcified, and some species are important contributors to coral reef structure and common in the fossil record (Butterfield et al., 1988
; Berger and Kaever, 1992
). At the ultrastructural level, bi- and quadriflagellate motile cells have a cruciate (X-2-X-2) flagellar root system with CCW offset and overlapped basal bodies, with scales and rhizoplasts. Cytokinesis occurs by furrowing, with a closed persistent spindle (Mattox and Stewart, 1984
; O'Kelly and Floyd, 1984
; Sluiman, 1989
).
The diplobiontic life cycle (free-living gametophyte and sporophyte phases, which may be iso- or heteromorphic; Bold and Wynne, 1985
) is found in four of the five groups of ulvophytes. This type of life cycle, though neither uniform nor universal in the group, contrasts to that of other green algae, which generally have a predominant haploid vegetative phase and a single-celled, often dormant, zygote as the diploid stage (haplobiontic, Bold and Wynne, 1985
). The diplobiontic life cycle is thus largely absent from green algae in freshwater habitats.
Because of the lack of clear synapomorphies for the ulvophytes, monophyly has been an open question since the group was established (O'Kelly and Floyd, 1984
; Bremer, 1985
; Mishler and Churchill, 1985
). Analyses using molecular data are lacking. Preliminary studies of 18S and 26S rRNA (Zechman et al., 1990
) indicated that Ulvophyceae of Mattox and Stewart (1984)
are not monophyletic, but a verdict awaits addition of further taxa and genes. Antiquity of the group and possible large numbers of extinctions may make recovery of the relationships among the major clades within the ulvophytes difficult with a single gene such as 18S rDNA.
Within the ulvophyte lineage, five well-demarcated groups are recognized (O'Kelly and Floyd, 1984
), which are usually ranked as orders, although their elevation to class is favored by some authors (e.g., van den Hoek et al., 1995
). Figure 19 shows a phylogenetic tree of the ulvophytes based on Hayden and Waaland (2002)
, O'Kelly et al. (2004),
and a preliminary analysis of 18S rDNA sequences (F. Zechman, California State University at Fresno, personal communication).
|
; Mattox and Stewart, 1984
). Assignment to Klebsormidium was confirmed by molecular data (Karol et al., 2001
).
Ulvales
This order contains several seaweed genera familiar worldwide, including Ulva and Enteromorpha, with membranous or tubular thalli, respectively. Recent work indicates that these genera are paraphyletic and Enteromorpha has been synonymized with Ulva (Hayden and Waaland, 2002
; Hayden et al., 2003
). Morphogenetic switching between the tubular and membranous form may have evolved multiple times in the group (Tan et al., 1999
).
Cladophorales (including Siphonocladales)
Cladophorales contain branched and unbranched siphonous (multinucleate) green algae. This group includes one of the most commonly encountered freshwater and marine genera, the branching filamentous Cladophora. The genus has been monographed by van den Hoek and colleagues (1963
, 1982
, 1984
; van den Hoek and Chihara, 2000
), who placed it in a separate order, Cladophorales. Several recent studies of 18S rDNA sequences (Bakker et al., 1994
; Hanyuda et al., 2002
) indicate that Cladophora is paraphyletic or polyphyletic to other genera in the group (Chaetomorpha, Rhizoclonium, Microdictyon, and Pithophora), although additional data are needed to resolve relationships of this group. Hanyuda et al. (2002)
provided clear evidence that marine Cladophora has given rise to freshwater forms several times.
Caulerpales and Dasycladales
Caulerpales and Dasycladales are two marine groups of distinctive and often beautiful, siphonous seaweeds. The highly invasive Caulerpa taxifolia is a member of Caulerpales. Molecular studies of Caulerpa using the tufA gene have indicated that many morphological species are not monophyletic and that later diverging lineages in the genus are diversifying faster than ancient ones (Fama et al., 2002
).
Dasycladales comprise two families, Dasycladaceae and Polyphysaceae (formerly Acetabulariaceae, Silva et al., 1996
), that are estimated to have diverged some 400 mya based on fossil evidence (Berger and Kaever, 1992
); molecular data indicate a more recent split approximately 265 mya (Olsen et al., 1994
). The group includes the model organism, mermaid's wine glass (Acetabularia spp.). Zechman et al. (1990)
studied rDNA sequences derived from RNA transcripts, and later Olsen et al. (1994)
studied 18S rDNA sequences in Dasycladaceae and Polyphysaceae. Both analyses found support for monophyly of the former family. All dasyclads possess a stemloop deletion unique among green algae (Olsen et al., 1994
). Olsen et al. (1994)
used distance and parsimony methods and supported monophyly of the two families, but only when the phylogenetic tree was unrooted; the authors suggested that rooting the tree with ulvophycean outgroups removed signal and resolution because the outgroups were so distantly related to the ingroup families. Olsen et al. (1994)
also found that several genera, including Acetabularia, were paraphyletic.
Berger et al. (2003)
sampled 17 of 19 species from Polyphysaceae (Acetabulariaceae) in a study of 18S rDNA sequences and corroborated the hypothesis of monophyly of Polyphysaceae and paraphyly of Dasycladaceae. Zechman (2003)
sampled a different gene, the plastid rbcL, in taxa from the two families and performed parsimony, likelihood, and Bayesian analyses. In contrast to the earlier 18S rDNA studies, Zechman (2003)
concluded that both families were paraphyletic, not just Dasycladaceae. He also found a similar problem of paraphyly of genera. Berger et al. (2003)
mapped features of cap formation onto the 18S rDNA tree and proposed a series of name changes in genera that would reconcile monophyletic groups on the tree and names of genera. Clearly, there is much to be done in this group, and additional multigene studies to resolve the phylogeny of the two families are underway (F. Zechman, California State University at Fresno, personal communication).
Trentopohliales
With an interesting mixture of ultrastructural characters, this group is sometimes omitted from the ulvophytes or included with several disparate green algal lineages (Chapman, 1984
). Although entirely terrestrial, the group has marine relatives. A sea-to-land transition or vice versa would be unique among green, red, and brown algae (Graham and Wilcox, 2000). Trentopohlians have phragmoplast-like cytokinesis (Chapman and Henk, 1986
; Chapman et al., 2001
), an MLS-like flagellar structure and unusual zoospores (Graham, 1984
), and "primitive"-type plasmodesmata, three characters reminiscent of charophyte algae (Chapman, 1984
; Chapman et al., 1998
). However, 18S rDNA sequence data place these algae firmly in the chlorophyte clade, most likely in the ulvophytes. Clearly, this is one group that bears further investigation of both morphology and DNA sequences.
Chlorophyceae
Chlorophyceae are a monophyletic group that includes some of the most familiar of the microscopic green algae, including many model organisms. The unicellular flagellate Chlamydomonas has been used to study flagellar motion and swimming (Mitchell, 2000
), photosynthesis mutations (Niyogi, 1999
), and plastid genome modifications in secondarily nonphotosynthetic taxa (Vernon et al., 2001
). Colonial green algae such as Volvox have been models for the evolution of multicellularity, cell differentiation, and colony motility (Hoops, 1997
; Kirk, 2003
). Chlorophycean algae Scenedesmus and Pediastrum are also important paleoecological or limnological indicators (Nielsen and Sorensen, 1992
; Komarek and Jankovska, 2001
).
Green algae in this class have a great range of vegetative morphology, from coccoid to swimming unicells, colonies, and simple flattened thalli to unbranched and branched filaments. All have a haplobiontic life cycle (zygotic meiosis). Orders previously recognized were defined on their vegetative morphology, form of sexual reproduction (either isogamy, anisogamy, or oogamy), and mode of asexual reproduction (zoosporic or autosporic). During cell division, mitosis is closed and cytokinesis involves a phycoplast system of microtubules, sometimes combined with furrowing. Swimming cells are vegetative cells, zoospores (asexual), or gametes, with two, four, or hundreds of flagella. Cells with two or four flagella have cruciate (X-2-X-2) rootlets and flagella that are displaced in a "clockwise" (CW, 1–7 o'clock) direction or are "directly opposed" (DO, 12–6 o'clock). In some swimming colonial forms such as Volvox, the flagellar apparatus undergoes developmental modification and the flagella are reoriented for colony swimming (Hoops, 1997
). Other variants include taxa with two unequal flagella (Heterochlamydomonas) or flagella emerging from a pit (Hafniomonas).
A flood of primarily 18S rDNA data collected from chlorophycean green algae in the last two decades has given rise to dramatic modifications at every level of classification (Fig. 20). A number of the traditional orders (Chlorellales, Chlorococcales, Chlorosarcinales), originally circumscribed using vegetative morphology, are now known to contain phylogenetically unrelated taxa. This is especially true for the groups that are morphologically depauperate and exhibit convergent evolution toward reduced morphology or cases in which absence of a trait was used as a main distinguishing feature of an order. Numerous evolutionary losses of motile cells are found across the chlorophycean green algae. Members of the Chlorellales (now mostly in the Trebouxiophyceae) completely lack motile cells, and thus the phylogenetically useful features that come from flagellar ultrastructure cannot be scored for these algae (Huss and Sogin, 1990
).
|
; Huss et al., 1999
). Buchheim et al. (1990)
showed that Chlamydomonas is not monophyletic. Since then, many additional genera have been shown to be highly polyphyletic. Congeners of Neochloris and Characium are now in three classes (Watanabe and Floyd, 1989
; Lewis et al., 1992
; Kouwets, 1995
; Watanabe et al., 2000
). Friedl and O'Kelly (2002)
used ultrastructural and molecular data to distribute four species of the chlorosarcinoid genus Planophila into Ulvophyceae (two species into two distinct clades), one into Trebouxiophyceae, and a fourth species into Chaetopeltidales (Chlorophyceae). Species of Chlorococcum were separated into Chlamydmonadales and Ulvophyceae (Watanabe et al., 2001
; Krienitz et al., 2003
). Thus, molecular data have provided evidence of convergent morphological evolution in many of the characters used to distinguish unicellular genera of chlorophycean green algae. The results of these studies only emphasize the need for additional molecular investigations, especially those that explore morphological character evolution in Chlorophyceae. Two overarching themes are evident in Chlorophyceae. First, superficial similarities in body plans can be misleading, especially for simpler forms, and second, a range of morphology is represented within a single clade, with the coccoid form being nearly ubiquitous. To paraphrase J. B. S. Haldane (cited in Evans et al., 2000
), nature must have had an inordinate fondness for coccoids.
|
; Wilcox et al., 1992
; Buchheim et al., 1994
; Nakayama et al., 1996a
, b
; Watanabe et al., 1996
; Booton et al., 1998a
, b
; Krienitz et al., 2003
). Vegetative morphology includes biflagellate unicells and colonies, quadraflagellate unicells, coccoid unicells, weakly branching filaments, or packets that are sometimes in a gelatinous matrix. All of the taxa possess the CW orientation of flagella and rootlets.
Buchheim et al. (1996
, 1997), among others, demonstrated that the large and well-recognized genus Chlamydomonas represents at least five lineages within the CW clade. Some of the lineages share organismal characters, such as the pigment loroxanthin (Fawley and Buchheim, 1995
) and autolysin groups (Buchheim et al., 1990
). Wall-less unicellular taxa in the related order Dunaliellales (sensu Ettl) were also shown to be nonmonophyletic, but instead interspersed in the CW clade. The unusual Hafniomonas, previously interpreted as having a modified CCW flagellar apparatus orientation or as a relative of Chaetopeltidales, is also a member of the CW clade (Nakayama et al., 1996a
).
The colonial flagellates previously placed in Volvocales are not monophyletic based on molecular data (Buchheim and Chapman, 1991
; Buchheim et al., 1994
) and do not represent a tidy progression of evolution by building ever larger colonies from Chlamydomonas-like ancestors with developmentally modified flagellar apparatus ultrastructure. Some colonial genera (Gonium) have been shown to be monophyletic, although most colonial forms with larger cell numbers are not (Nozaki et al., 2000
).
Sphaeropleales (DO clade)
As emended by Deason et al. (1991)
, this order includes vegetatively nonmotile unicellular or colonial taxa that have biflagellate zoospores with the DO flagellar apparatus arrangement: Sphaeroplea, Atractomorpha, Neochloris, Hydrodictyon, and Pediastrum. All of these taxa possess basal body core connections (Wilcox and Floyd, 1988
). With an increase in the number of taxa for which sequence data are available, there is evidence of an expanded DO clade that includes additional zoosporic (Bracteacoccus, Schroederia; Buchheim et al., 2001
; Lewis, 1997
) and some strictly autosporic genera such as Ankistrodesmus, Scenedesmus, Selanastrum, Monoraphidium, and Pectodictyon (Krienitz et al., 2001
, 2003
). Monophyly of the DO clade is generally weakly supported by phylogenetic analysis of molecular data, even those that have used data from two genes (Buchheim et al., 2001
; Wolf et al., 2002
; Shoup and Lewis, 2003
).
Oedogoniales
This order includes three genera, Oedogonium, Oedocladium, and Bulbochaete, and approximately 600 species, all of which grow attached to submerged surfaces in freshwater habitats. All members of this order form simple or branched filaments. The familiar Oedogonium is often used as an example of oogamous sexual reproduction in green algae. All genera produce unusual motile cells (either asexual zoospores or male gametes) with an anterior ring of flagella (stephanokont). The flagellar apparatus of these cells has been studied extensively (Pickett-Heaps, 1975
) and is clearly unlike the swimming cells in other groups of green algae. Given the strikingly unusual ultrastructure of the swimming cells, homology assessment with flagellar characters found in the other groups is difficult. Pickett-Heaps (1975)
hypothesized that this group represents a "basal" lineage that gave rise to other filamentous forms. Analyses of molecular data (18S and 26S rDNA) have indicated that the order is clearly monophyletic (Booton et al., 1998b
; Buchheim et al., 2001
) and is often placed sister to the rest of Chlorophyceae. However, this placement varies, often with differing resolutions that include the Chaetopeltidales and Chaetophorales. Because all three orders represent "long branches" in the tree, a robust placement is not often obtained or depends on method of analysis.
Chaetopeltidales
This order of four genera was proposed by O'Kelly (1994) to include taxa that produce quadriflagellate motile cells with a perfectly cruciate (DO) flagellar apparatus orientation. The vegetative morphologies found in this order include flattened thalli (Chaetopeltis) and a sarcinoid genus Floydiella (formerly Planophila; Friedl and O'Kelly, 2002
). Analyses of 18S rDNA data place this group, albeit weakly, with the biflagellate DO taxa (Booton et al., 1998a
; Krienitz et al., 2003
). Analyses of combined 18S and 26S rDNA data were also inconclusive, but they often resulted in topologies with this order outside the two clades of biflagellate taxa, closer to the base of Chlorophyceae (Buchheim et al., 2001
; Shoup and Lewis, 2003
).
Chaetophorales
Members of the order have unbranched or branched filamentous vegetative bodies and produce quadriflagellate motile cells with upper and lower pairs of basal bodies in a CW + CW arrangement. Phylogenetic analysis of 18S rDNA data by Nakayama et al. (1996a)
placed Chaetophora incrassata as sister to the rest of Chlorophyceae. With expanded sampling, Booton et al. (1998b)
placed Chaetophorales nearest the biflagellate taxa with the CW orientation, although this placement was very weakly supported. As with Oedogoniales and Chaetopeltidales, the monophyly of the order is supported, but its relative placement remains unresolved.
Incertae Sedis
Topologies resulting from the analysis of two genes (18S and 26S rDNA) have indicated a distinctive chlorophycean clade, adjacent to Sphaeropleales, which includes the filamentous genus Cylindrocapsa and the unicellular Trochiscia and Treubaria. Previous hypotheses indicated an alliance of Cylindrocapsa with Sphaeropleales based on morphology of the pyrenoid (region within plastid containing a high concentration of the enzyme ribulose-1, 5-bisophosphate carboxylase [rubisco] and associated with starch synthesis), but this relationship is not supported with molecular data (Buchheim et al., 2001
). Another unnamed group of freshwater picoplanktonic taxa, the Mychonastes clade (Krienitz et al., 2003
), forms a sister group to Oedogoniales at the base of Chlorophyceae. A new order will need to be established eventually to accommodate this clade. In addition, several flagellate taxa, including species of Carteria and Hafniomonas, form a grade at the base of the CW clade and do not correspond to a named group (Nakayama et al., 1996a
; Hoham et al., 2002
).
Trebouxiophytes
In their 1984 classification, Mattox and Stewart proposed the class Pleurastrophyceae to accommodate freshwater algae with the CCW flagellar apparatus orientation, a metacentric spindle, and phycoplast-mediated cytokinesis. This class included the nonmotile unicells Pleurastrum and Trebouxia, the unicellular flagellate Tetraselmis, and Microthamnion, a filamentous alga with short branches. Conversely, Melkonian (1990a)
chose to group these same taxa (excluding the flagellate Tetraselmis) into a separate order Microthamniales. Additional studies expanded membership of this group to include the terrestrial alga Leptosira, which produces unbranched uniseriate filaments (Lokhorst and Rongen, 1994
). An early molecular analysis based on just a few taxa by Kantz et al. (1990)
provided evidence for the nonmonophyly of Pleurastrophyceae.
Friedl and Zeltner (1994)
