The Fossil Record of Early Eukaryotic Diversification

The Cambrian explosion can be thought of as the culmination of a diversification of eukaryotes that had begun several hundred million years before. Eukaryotes – one of the three domains of life — originated by late Archean time, and probably underwent a long period of stem group evolution during the Paleoproterozoic Era. A suite of taxonomically resolved body fossils and biomarkers, together with estimates of acritarch and compression fossil diversity, suggest that while divergences among major eukaryotic clades or ‘super-groups’ may have occurred as early as latest Paleoproterozoic through Mesoproterozoic time, the main phase of eukaryotic diversification took place several hundred million years later, during the middle Neoproterozoic Era. Hypotheses for Neoproterozoic diversification must therefore explain why eukaryotic diversification is delayed several hundred million years after the origin of the eukaryotic crown group, and why diversification appears to have occurred independently within several eukaryotic super-groups at the same time. Evolutionary explanations for eukaryotic diversification (the evolution of sex; the acquisition of plastids) fail to account for these patterns, but ecological explanations (the advent of microbial predators) and environmental explanations (changes in ocean chemistry) are both consistent with them. Both ecology and environment may have played a role in triggering or at least fueling Neoproterozoic eukaryotic diversification. ALTHOUGH THE CAMBRIAN explosion is certainly a significant event in itself, it can be seen as part of a larger diversification of eukaryotes that had begun millions of years before. Neither multicellularity nor biomineralization was invented in the Cambrian explosion; both were present in several clades long before the close of the Proterozoic (Allison and Hilgert, 1986; Butterfield, 2000; Porter et al., 2003). Similarly, predatory behavior, often cited as one of the possible triggers of Cambrian radiation (e.g., Stanley, 1973), first made its appearance several hundred million years before, albeit in microbial organisms. Eukaryotic diversification reached an acme in latest Neoproterozoic and early Cambrian time, but substantial increases in diversity characterize the earlier Neoproterozoic Era as well (Knoll 1994; Vidal and Vidal-Moczydlowska, 1997). Here I discuss the Proterozoic diversification of eukaryotes, focusing, in particular, on evidence from the late Mesoproterozoic and Neoproterozoic fossil record. WHAT IS A EUKARYOTE? Before discussing the fossil record of eukaryotes, it would be useful to define what a eukaryote is. Eukaryotes are one of three major groups or ‘domains’ of life. They are distinct from the other two domains – Archaea and Bacteria – in that they have a more complex cell structure: their cells possess a membrane-bound ‘true’ nucleus, a cytoskeleton, a complex endomembrane system, and organelles such as mitochondria and plastids. Eukaryotes are probably best known for their multicellular representatives — indeed, multicellularity evolved at least seven times within the eukaryotes, in animals, fungi, red algae, slime molds, brown algae, and at least twice in green algae. A substantial component of eukaryotic diversity, however, is microbial, comprising taxa traditionally grouped within the protozoa and algae. It is these microbial groups that primarily contribute to the early fossil record of eukaryotes. PALEONTOLOGICAL SOCIETY PAPERS, V. 10, 2004 36 CURRENT VIEWS OF EUKARYOTE PHYLOGENY Our view of eukaryote phylogeny has undergone significant change in the last decade. Early phylogenies based on rRNA painted a compelling picture: complex eukaryotes, such as animals, plants, and fungi, diverged rapidly and late in eukaryotic evolution, while simpler, often bizarre, mitochondrion-lacking eukaryotes diverged in regular succession much earlier (Sogin, 1991). Several studies now show that this pattern is an artifact of long branch attraction (e.g., Philippe et al., 2000), and that the early branching, amitochondriate clades are in fact highly derived, their mitochondria secondarily lost or reduced (e.g., Tovar et al., 2003; Sutak et al., 2004). Recent phylogenies based on molecular and ultrastructural data are starting to converge on a different pattern. These indicate that most eukaryotes fall within one of several ‘super-groups’ (Fig. 1; Simpson and Roger, 2002; Baldauf, 2003; Nikolaev et al., 2004): the opisthokonts, including the animals, choanoflagellates, and fungi; the Amoebozoa, including most slime molds and the lobose amoebae; the plants, including the red and green algae and the land plants; the chromalveolates, including the diatoms, brown algae, ciliates, dinoflagellates, chrysophyte algae, and, possibly, the haptophytes (the group that includes the coccolithophores); the Rhizaria, including the foraminifera, the filose testate amoebae, and the radiolaria; and the excavates, including the euglenids and several parasitic taxa. Relationships among these taxa are not well resolved, although recent work based on gene fusion data suggests that the tree can be rooted between two larger clades, the opisthokonts + amoebozoa on one side, and the rest of the eukaryotic super-groups on the other (Stechmann and Cavalier-Smith, 2002, 2003). If correct, this would represent a radical departure from the conventional view of eukaryotic evolution, implying that the group that includes the animals and fungi may have branched off early in eukaryotic evolution. THE EARLIEST EUKARYOTES Late Proterozoic rocks record the diversification of eukaryotic super-groups, but the eukaryotic clade itself appears to have originated much earlier. Sterane molecules, whose biochemical precursors, sterols, are widespread and diverse in living eukaryotes, have been found in ~2700 Ma shales from Australia (Brocks et al., 1999), suggesting that eukaryotes may have evolved by late Archean time. Sterols have also been found in several bacterial groups as well (see, e.g., Cavalier-Smith, 2002a), although none exhibit the structural diversity indicated by the Archean steranes (Brocks et al., 2003). Plausible eukaryotic body fossils — problematic macrofossils and structurally complex microfossils first appear ~1000 m. y. later, in late Paleoproterozoic and early Mesoproterozoic rocks. Problematic macrofossils include the ~1400-1800 Ma Grypania and ~10001600 Ma “strings of beads” structures, both interpreted as photosynthetic eukaryotes (Grey and Williams, 1990; Han and Runnegar, 1992; Kumar, 1995). If these interpretations are correct, they would imply that the eukaryotic super-groups must have already diverged by early Mesoproterozoic time, as plastids, the site of photosynthesis in eukaryotes, were acquired after the plant supergroup had diverged from its closest relative (although see Andersson and Roger, 2002). More convincingly eukaryotic fossils include the ~1500 Ma structurally complex microfossil Tappania plana, whose irregularly branching processes and bulbous protrusions suggest the presence of a cytoskeleton – a character unique to eukaryotes (Javaux et al., 2001; see also Javaux et al., 2003). With the possible exception of problematic macrofossils like Grypania, these early fossils may best be interpreted as stem group eukaryotes: taxa that branched off the ‘main line’ early in eukaryotic evolution, prior to the last common ancestor of all living eukaryotes [in contrast, ‘crown group eukaryotes’ refer to those eukaryotes that are descended from the last common ancestor of all living eukaryotes (Fig. 1); see, e.g., Budd and Jensen (2000) for further discussion of ‘stem’ and PORTER–EARLY EUKARYOTIC DIVERSIFICATION 37 FIGURE 1—A current view of eukaryote phylogeny, based on a consensus of molecular and ultrastructural data [modified from Simpson and Roger (2002), Baldauf (2003), and Nikolaev et al. (2004)]. Question marks associated with excavates and chromalveolates indicate that monophyly of these clade is in question. The dotted line indicates uncertainty with respect to the inclusion of the haptophytes in the chromalveolate clade. Rooting of the tree is based on gene fusion data from Stechmann and Cavalier-Smith (2002, 2003). The three-gene fusion uniting the opisthokont and Amoebozoa supergroups refers to genes involved in pyrimidine synthesis; the two-gene fusion uniting the rest of the super-groups refers to the dihydrofolate reductase and thymidylate synthase genes (Stechmann and Cavalier-Smith, 2003). Dates are derived from fossils discussed in the text, and in most cases represent broad estimates. As discussed in the text, many of these taxonomic assignments are uncertain; dates associated with the more questionable assignments are in italics; those associated with the more convincing assignments are larger and in boldface (see also Table 1). PALEONTOLOGICAL SOCIETY PAPERS, V. 10, 2004 38 ‘crown’ groups]. As stem groups, these fossils provide a minimum age constraint on the origin of the eukaryotic clade itself and on the acquisition of important eukaryotic synapomorphies (e.g., sterols and the cytoskeleton). The eukaryotic crown had originated by the time we see the first representatives of a eukaryotic super-group (~1200 Ma; though possibly as early as 1500-1700 Ma; see Fig. 1 and below), but not until the early to middle Neoproterozoic Era do fossils representing a diversity of eukaryotic super-groups appear. EUKARYOTIC DIVERSIFICATION Two lines of evidence indicate that the diversification of eukaryotes was underway by middle Neoproterozoic time. The first, and most compelling, comes from a suite of body fossils and molecular biomarkers that, as discussed in the following sections, can either be plausibly or convincingly assigned to modern eukaryotic clades (see Table 1). The second comes from acritarchs – problematic organic-walled microfossils — and macroscopic carbonaceous compression fossils. Although taxonomically unresolved, these provide useful information about general trends in Neoproterozoic eukaryote evolution. Taxonomically resolved body fossils.— The earliest body fossil that can be assigned with confidence to a eukaryotic super-group is late Mesoproterozoic in age: Bangiomorpha pubescens, a multicellular filament found in the ca. 1200 Ma Hunting Formation, arctic Canada (Fig. 2.1-2.2; Butterfield et al., 1990; Butterfield, 2000). Although Bangiomorpha specimens exhibit a number of characters found within both the red and green algal and cyanobacterial clades, they exhibit one feature – an unusual pattern of cell division (Fig. 2.1) – not known outside the red algal family Bangiaceae (Butterfield et al., 1990; Butterfield 2000). It is possible this character is “well within the capacity of a filamentous bacterium to evolve” (p. 38, Cavalier-Smith, 2002a), but it is more parsimonious (based on morphological characters alone) to suggest the fossils are indeed early representatives of red algae. Additional evidence for multicellular photosynthetic eukaryotes comes from the somewhat younger filamentous microfossils Paleovaucheria clavata (German, 1981) and Proterocladus sp. (Butterfield, 1994). Paleovaucheria (Fig. 2.3), which occurs in both the >1000 Ma Lakhanda Suite, Siberia (German, 1981, 1990; Woods, et al., 1998), and the ~750 Ma Svanbergfjellet Formation, Spitsbergen (Butterfield, 2004), exhibits several characters that collectively indicate an affinity with the vaucheriacean algae, relatives of the brown algae, and members of the chromalveolate clade (Potter et al., 1997; Fast et al., 2001). These include sparsely branching thalli with few septa that tend to concentrate at the bulbous termini of the filaments; circular openings at the ends of the termini; and filaments of two distinct size classes (German, 1990; Woods et al., 1998; Butterfield, 2004). Proterocladus sp. (Fig. 2. 5), also from the Svanbergfjellet Formation, is a simple, uniseriate filament that exhibits occasional branching and intercellular septa, morphology reminiscent of the modern green alga, Cladophora and Cladophoropsis, though also observed in some simple red algae (Butterfield et al., 1994). Vase-shaped microfossils (VSMs), abundant, diverse, and globally distributed in rocks ~750 Ma in age, provide evidence for two additional eukaryotic super-groups (Porter and Knoll, 2000; Porter et al., 2003). At least 17 species of VSMs have been described. These include Paleoarcella athanata (Fig. 2.7), whose test is indistinguishable from that of the modern lobose testate amoeba, Arcella, a member of the amoebozoan super-group (Fig. 1); and Melicerion poikilon (Fig. 2.6) whose test, apparently covered in mineralized scales, is closely comparable to those of euglyphid amoebae in the rhizarian super-group (Fig. 1). [Other VSM species (e.g., Figs. 2.8-2.9) exhibit characters found in both modern taxa and thus cannot be definitively assigned to either.] Thus, VSMs indicate that both the Rhizaria and the Amoebozoa had begun to diverge by ~750 Ma. They also provide the earliest definitive fossil evidence for heterotrophic eukaryotes. PORTER–EARLY EUKARYOTIC DIVERSIFICATION 39 T A B LE I— Ta xo no m ic al ly r es ol ve d bo dy fo ss ils a nd b io m ar ke rs p re se rv ed in P ro te ro zo ic r oc ks . M a = M ill io ns o f y ea rs . “C on fid en ce ” r ef er s to c on fid en ce in in te rp re ta tio n: ( + ) = in te rp re ta tio n is b as ed o n co m pe lli ng e vi de nc e; ( ~ ) = th e in te rp re ta tio n is p la us ib le . F or m or e de ta ils , s ee d is cu ss io n in te xt . l i s s o F e c n e r r u c c O e g A n o i t a t e r p r e t n I e c n e d i f n o C p u o r g r e p u S s e c n e r e f e R a h p r o m o i g n a B s n e c s e b u p a d a n a C , n o i t a m r o F g n i t n u H a M 0 0 2 1 ~ d e r e t y h p o i g n a B a g l a + s t n a l P 0 9 9 1 , . l a t e d l e i f r e t t u B 0 0 0 2 , d l e i f r e t t u B a i r e h c u a v o e a l a P a t a v a l c a i r e b i S , e t i u S a d n a h k a L n e g r e b s t i p S , m F t e l l e j f g r e b n a v S a M 0 0 0 1 ~ a M 0 5 7 ~ n a e c a i r e h c u a V a g l a + s e t a l o e v l a m o r h C t e s d o o W ; 0 9 9 1 , 1 8 9 1 , n a m r e G 4 0 0 2 , d l e i f r e t t u B ; 8 9 9 1 , . l a s u d a l c o r e t o r P . p s n e g r e b s t i p S , m F t e l l e j f g r e b n a v S a M 0 5 7 ~ n a e l a r o h p o d a l C a g l a n e e r g ~ s t n a l P 4 9 9 1 , . l a t e d l e i f r e t t u B a l l e c r a o e a l a P ( a t a n a h t a ) M S V A S U , p u o r G r a u h C a M 0 5 7 ~ e a b e o m a e t a t s e t + a o z o b e o m A 3 0 0 2 , . l a t e r e t r o P n o l i k i o p n o i r e c i l e M ) M S V ( A S U , p u o r G r a u h C a M 0 5 7 ~ e t a t s e t d i h p y l g u E a b e o m a + a i r a z i h R t e r e t r o P ; 0 0 0 2 , l l o n K & r e t r o P 3 0 0 2 , . l a s M S V r e h t O s e i t i l a c o l s u o r e m u N a M 0 5 7 ~ e s o l i f d n a e s o b o L e a b e o m a e t a t s e t ~ a i r a z i h R & a o z o b e o m A t e r e t r o P ; 0 0 0 2 , l l o n K & r e t r o P 3 0 0 2 , . l a a i n a p p a T . p s a d a n a C , p u o r g r e p u S r e l a h S a i l a r t s u A , p u o r G r e p o R a M 0 0 8 0 0 9 ~ a M 0 0 5 1 ~ i g n u F ~ at n o k o h t s i p O s s e r p n i , d l e i f r e t t u B ’ s l i s s o f o r c i m e l a c s ‘ a d a n a C , p u o r G r i d n i T a M 0 0 6 0 0 7 ~ , s m o t a i D & , s t e y h p o t p a h s e t y h p o s y r h c ~ s e t a l o e v l a m o r h C ; 6 8 9 1 , t r e g l i H & n o s i l l A 2 9 9 1 , . l a t e n a m f u a K e n a r e c a m m a G ) r e k r a m o i b ( A S U , p u o r G r a u h C a n i h C , m F i z n a h s n a u T a M 0 5 7 ~ a M 0 0 7 1 ~ s e t a i l i C ~ s e t a l o e v l a m o r h C ; 8 9 9 1 , . l a t e s n o m m u S ; 0 9 9 1 , r e t l a