About the SA Blog Network

The Ocelloid

The Ocelloid

Through the eye of a microbe
The Ocelloid Home

The Lost Eukaryote: an introduction to cellular evolution

Email   PrintPrint

The most fundamental divide in the diversity of living creatures is arguably the one between prokaryotes (=bacteria*) and eukaryotes (the tiny island of cumbersomely complex cells that consists of protists. And a couple insignificant lineages that are hardly worth talking about). Much of the earth’s biota seems perfectly content with small, streamlined genomes and similarly small, streamlined cell architecture. All but one group, that for some odd reason ended up with a membrane-bound package of a junky genome we call the nucleus. The nucleus, in turn, is but a spokesperson-organelle for the massive changes in cellular architecture that occurred in the transition from prokaryotic to eukaryotic forms — a feature that most likely arose with the changes rather than initiating them. The most prominent features present in all eukaryotes are the actin and tubulin cytoskeleton, endomembrane trafficking (enabling phagocytosis) and mitochondria or some form thereof. Unfortunately (or fortunately, as it keeps us employed?), most of these features appear to have been already present and well-developed in the last common ancestor of all known eukaryotes, thereby depriving us of a convenient grade from which to infer how these structures actually evolved. Once upon a time, it was thought that some anaerobic eukaryotes lacked mitochondria and diverged from the aerobes before the mitochondrion was ‘enslaved’ through endosymbiosis in the latter (oddly enough, early ribosomal DNA trees even supported that, but that’s a story for another day). However, it later turned out that even the mitochondrion was already present in the last common ancestor, and thus when we work our way back to reconstruct the evolution of the eukaryotic cell, we are stuck with a fairly modern cell that seemingly erupts spontaneously from a bacterial sea. Odd and unsettling to say the least.

*Yeah, yeah, Archaea included, we can argue about that later…

As a protistologist and some sort of a cell biologist by modest training, I am particularly interested in cellular evolution. In other words, while some focus on the evolution of macroscopic structures like wings and organs, and others look at molecular evolution of proteins and DNA sequences, I am especially fascinated by the in-between, or how subcellular structures themselves evolve. Unlike molecular biologists, we don’t have the luxury of compressing the bulk of our data into sequences, and unlike developmental biologists, we can’t really fiddle with gene expression patterns and play with a variety of well-established mutants, both natural (visible diversity) and lab-generated. This is partly why there’s a chance you probably never heard of evolutionary cell biology as a field. The other big problem is that much of cellular diversity is, in fact, microbial, and microbial eukaryotes are barely studied (yeasts excluded — but they’re secondarily unicellular anyway, and really, really weird). It is in the unicellular protist realm where the cell is at its finest, for it cannot cower behind the multitudes of defective cell types of a multicellular organism to get by, and must be largely self-sufficient. (This is illustrated further by the higher average complexity (diversity of cell parts) in a unicellular cell than that of multicellular organisms (McShea 2002 Evolution)) Not only are these unicellular organisms cellularly complicated, they’re also quite diverse. Bacteria most definitely have a cell biology of their own, but that has become recognised only recently, with the advent of fluorescent, and now super-resolution, light microscopy — where one can finally track labelled proteins in a living cell. Thus, for the moment, evolutionary cell biology is ultimately the cell biology of protists in light of evolution.

Of course, just comparing cell structures and marvelling at their diversity isn’t really all there is to exploring the evolution of something. Even reconstructing ancestral states is just the beginning. Evolutionary biology ultimately pursues mechanisms — the more general, the better. We could simply assume evolution is adaptation  and make up stories as we go along (not entirely unpopular in some circles), but that wouldn’t be good science. Evolution involves introduction of variation through mutation (with its own associated biases) as well as sorting thereof nor only through selection, but also by drift and migration.

Furthermore, heritability is a key required component in evolutionary change, and here we may even get something interesting: transmission of information from one cell to the next (generationally) is not only genomic (or genetic), but also depends on a spatial component. If you simply express a genome in a lipid vesicle, the proteins will not magically self-assemble into a working cell. A chunk of necessary information is directed by the patterning in the cell preceding the division. Extra-nuclear (or extra-genomic) cellular inheritance is not a mere figment of speculative imagination — it has been demonstrated in ciliates in a landmark experiment by Tracy Sonneborn and Janine Beisson: a row of cilia was inverted surgically (presumably without affecting the genome, of course) in a Paramecium, and this strain with a backwards row of cilia persists to this very day, despite multiple genetic outcrossings (Beisson & Sonneborn 1965 PNAS)! Several of Sonneborn’s deciples have continued the work on cytoplasmic inheritance in ciliates, with some fascinating results. However, molecular work on poorly-established model organisms is difficult and frustrating, and until recently bordered on insanity.  Unfortunately, just as the tools for doing molecular and cell biology on more obscure organisms are greatly improving (10 years ago, you couldn’t just sequence a genome on a whim…), the field has largely…retired.

If there is a channel of inheritance that occurs in parallel with classical genetics, this opens up a whole new jungle of tantalising questions and models waiting to be described and later discarded in favour of better ones. While classical quantitative genetics (which studies the inheritance of visible, measurable traits from generation to generation) is a fairly established and well-studied field at this point, a parallel epigenetic system of heritability would call for expansion of the field to include non-genomic quantitative genetics, where it gets rather tricky due to lack of direct digital coding sequences. Of course, if such a thing were to be pursued and studied, it would have to be in unicellular organisms, for they don’t have that pesky bottleneck where the entire multi-million celled creature has to fit through a fertilised egg or seed for later re-patterning. Essentially, this would call for an evolutionary developmental biology of the single cell. While all cells go through something resembling classical development in principle in at least some stage of their lives, we don’t typically think of development on a cellular level. We really should.

Enough with the long-winded theoretical introduction. What, if anything, can we say about the grandest scale of eukaryotic cellular evolution, or that nagging question of how eukaryotes evolved? Unfortunately, as mentioned above, the picture is a little unsettling. That last common ancestor of ours was simply too complex! (creationist quotemining in 3…2…1)

The last common ancestor is not the same as the first common ancestor...!

It appears that the last eukaryotic common ancestor (LECA), of all currently known living eukaryotes, has been a fairly sophisticated cell with a nucleus and a mitochondrion, as well as elaborate cytoskeletal and membrane trafficking systems. Presumably, the first eukaryotic common ancestor was drastically less complicated, but its nature remains elusive, and all but one of its descendants...lost. (Also see Field & Dacks 2009 Curr Op Cell Biol)(abbreviations are for major eukaryotic supergroups, in no particular order: Ex - Excavates; Op - Opisthokonts; Am - Amoebozoans; SAR - Stramenopile-Alveolate-Rhizaria clade; Arch - Archaeplastids; Ha - "Hacrobia")

Not only does LECA appear to possess a mitochondrion and a modern nucleus, but it already has a sophisticated membrane trafficking system, a cytoskeleton, capacity to devour prey by phagocytosis, a eukaryotic cell cycle regulation system, meiotic sex, and even a flagellum. Not only does it have modern-looking structures, but it seems to have already used many of the same molecular components used in a variety of living eukaryotes today. As an aside, you may perhaps recall having learned cell biology going structure by structure: there’s an endoplasmic reticulum for making proteins and moving them, a Golgi for sorting them, vacuoles and lysosomes for storage and digestion, a nucleus for DNA… but it’s perhaps more productive, and less confusing even, to think of the cell as a network of systems (like the human body), the key ones being metabolic pathways, the genome, cell cycle, the membrane trafficking system and the cytoskeleton, with the rest of the cell emerging from them. (this list is by no means meant to be definitive)

Of course, the first eukaryote-like thing, FECA*, presumably emerged from the bacterial realm. Somehow in the interim, between FECA and LECA, our lineage lost many of its bacterial features (such as a murein wall — think Gram staining) and picked up all sorts of eukaryotic traits. One would imagine it not being a case where a single proto-eukaryote population just sits around and gradually eukaryifies until it becomes LECA and then explodes into a ton of supergroups — the pre-LECA eukaryotes were probably diverse and had numerous long-lost offshoots. But somehow, it appears that only one lineage survived to rapidly diversify into modern extant eukaryotes. What where those enigmatic lost eukaryotes? Why did only one lineage survive to bind them all in mystery?

* We could call it them the Lost Eukaryotic Common Ancestors, but the acronym would be confusing…

Unfortunately, where we have a sample size of one in the form of a single phylogenetic event, we are left with little else but mere speculation (the question of the origin of sex falls under the same category). We might be tempted to think the presence of a mitochondrion or its relics in every known eukaryote may allude to mitochondrial symbiosis doing something important. Perhaps a massive selective sweep because this new organelle was that damn awesome. While this may sound reasonable, we have no clear evidence pointing either way. If we knew roughly when eukaryotes arose, we could speculate on some massive environmental change, perhaps a mass extinction where just by chance a single lineage survived. But our estimates for the origin of eukaryotes range from 0.8-3.5 billion years ago, in the wildest estimates. The likeliest time period in my irrelevant opinion, based on fossils and molecular clocks, would be the early Mesoproterozoic or the late Paleoproterozoic (~1.2-1.8 billion years) — a time period still poorly understood. Hell, at times we can hardly tell whether a microfossil is even biotic in origin, let alone discern what made it!

I have probably convinced you by now that both the question of how cells evolve and the issue of the very origin of eukaryotes are thoroughly impossible to address. Usually when people write about science, the story works towards gradually clarifying one conundrum or another. Yes, there is often the occasional setback and an annoyingly unfitting data point that rudely asserts its foul presence in the midst of your otherwise beautiful hypothesis. But the topic of eukaryotic evolution is a whole other type of story — in fact, while the protistan phylogeny has been clearing up over the past decade, the question of how they got there in the first place slipped further and further away. And the recent adventures in protistan genomes and proteomes only make it worse — by rendering the Last Eukaryotic Common Ancestor unbearably complex.

But there is hope, and it lies in the bewildering diversity of eukaryotic cells — as protists. We can still learn how eukaryotic cells evolve, and work on those general principles and models that are the holy grail of evolutionary biology (as much as anything can be holy in science, but we try!). We could perhaps even extrapolate those principles back in time and use the few subtle clues we have to uncover some of the FECA’s descendents’ path to eukaryocy (fine, eukaryote-hood). In fact, in the next post we’ll look at once such case in the evolution of membrane trafficking machinery. We still have a vastness of post-LECA diversity and evolution to address.

Anywhere there is heritable diversity, there is an evolutionary system awaiting attention. Like culture and language, cells are no exception.

Psi Wavefunction About the Author: Psi Wavefunction is a graduate of the University of British Columbia working as a protist researcher (soon to be graduate student) at Dalhousie University in Halifax, Nova Scotia, and blogs about protists and evolution at The Ocelloid as well as at Skeptic Wonder. Follow on Twitter @Ocelloid.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 5 Comments

Add Comment
  1. 1. ColinBiss 3:11 pm 11/3/2011

    Sorry, but all that writing got me thinking about sex, instead of protists. I want to add a speculation about why it evolved, which occurred to me because of something said in the post about funneling down to egg and sperm.
    It occurs to me that the benefit of sex, other than diversity, is a reset. If you’re reset every generation, you have to evolve by modifying your software, not your hardware. Software modifications (genetic changes) are more durable than hardware modifications, and more reliably passed on to future generations. A hardware modification (like those Paramecia with the inverted cilia) could be lost easily with a single error in transmittal of the relevant part, and would be less reliably passed on.
    A reset has power. How would a bacterium, with the hardware modifications it has adopted over its life or over multiple generations, revert to the original form if the modifications stopped being beneficial? That process would surely be messy. The sex reset also has disadvantages, as all the benefit of experience is lost with each generation. Fortunately, or perhaps not accidentally at all, some of the lesser eukaryotes have evolved coping mechanisms for this loss, including parenting and Grade 7.

    Link to this
  2. 2. Christopher Taylor 6:53 pm 11/3/2011

    It seems reasonable that the origins of phagocytosis and of the nucleus might be connected, seeing as much the same vesicle/tubule system is involved in them both (I think; I’m quite willing to be corrected). Has anyone ever suggested that the nucleus might have developed as a way of protecting the chromosomes and other high-priority material during phagocytosis? Kind of like a microbial cup. Mind you, some planctomycetes have a nucleus-type thingy without phagocytosis.

    Eukaryotes are also wierd in their linear rather than circular chromosomes. I have a feeling that there are some bacteria with linear chromosomes, but I don’t remember which ones off the top of my head.

    Link to this
  3. 3. Torbjörn Larsson, OM 7:56 pm 11/3/2011

    This article is most auspiciously timed.

    “Presumably, the first eukaryotic common ancestor was drastically less complicated, but its nature remains elusive, and all but one of its descendants…lost.” *

    Except, as an article claimed a few weeks back, sequencing a relative to the Mimivirus makes the authors suggest that the NCLDV lineage evolved from an ancestral cellular genome by typical parasitic reduction. ["Distant Mimivirus relative with a larger genome highlights the fundamental features of Megaviridae", Arslan et al, PNAS 2011.] The cell-like gene set branches after the Archaea did in the Archaea-Eukaryote lineage, presumably after your FECA.

    My interest here as a part time student of astrobiology is eukaryote type events and its likelihood. As I understand it you can have multicellulars without access to the energy from having a mitochondrion type energy plant. But according to Lane et al energy hypothesis it is the ~ 10^5 larger energy from mitochondrions that makes for eukaryotes larger protein turnaround and enables them to add and support a vast functionality.

    While I wait for the scientists to hash out if this lineage is a valid cellular descendant and what it means, I note that the core set of 44 cell-like genes [Table S01], those that are less likely to be of horizontal gene transfer ancestry under the hypothesis, doesn’t seem to contain any mitochondrion homologous genes. That is rather unlikely.

    - Sequencing put SARS11 clade as mitochondrial ancestor. They have small genomes, the nearest complete genome I could find was HTCC7211 ~ 1.5 k genes. [ SARS11 ]

    Compare with a nuclear genome, say humans at ~ 25 k genes. Likelihood for a gene being mitochondrial when drawing from the whole set is then p ~ 0.06.

    For a first order guesstimate, assume unrealistically independent genes. The likelihood that all are non-mitochondrial becomes p = (1-0.06)^44 ~ 0.07.

    - To make a quick and dirty test, I can use a Poisson Distribution events test. Sees 0 events out of 44, expects ~ 3 events, which gives me a p < 0.1 (using StatPac calculator). This roughly corresponds with and is validated by the first test above.

    So there is, this layman thinks, at least ~ 90 % chance that the NCLDV ancestor originated before the LECA. Further sequencing should help establish if this is valid and if so hopefully increase the likelihood using biologist's phylogenetic methods.

    Also of interest would be to know what cell functionality it had at the time. Maybe the chaperonins can clue biologists in, but I note a "Vacuolar protein sorting-associated protein". Bacteria has vacuoles, but all the vacuolar protein sorting-associated proteins I get from a hasty googling are eukaryote associated with a Golgi network. Presumably then FECA descendants may have had an endomembrane system.

    I assume an endomembrane system may lead up to phagocytosis and then a protective nuclear membrane independent of any mitochondrion event. That event would be a potential lock in effect as you claim (see Lane et al hypothesis), so if it happened early after phagocytosis evolved it would predict that we see no more such events respectively why we only see NCLDVs as closest relatives after Archaea.

    Phagocytosis is also a requisite for at least the NCLDV Megaviridae lineage, so it is consistent and putatively validated to hypothesize that it would be present in a putative common NCDLV/Eukaryote ancestor.

    A biological just-so story would be that at least NCLDVs would spur the evolution of, I assume, protective measures like a nucleus. Unfortunately I don't see how that could be made testable.

    Again, an extended NCLDV cell-like gene core set would help resolve early eukaryote evolution. Also, Claverie et al suggests one should look at these large viruses as having the viral particle (VP) as the seed and the viral factory (VF) as the adult organism. ["Mimivirus: the emerging paradox of quasi-autonomous viruses", Claverie et al, TIGS 2010.]

    Extrapolating from the hour glass model of developmental constraints, the early VF should have a few robust features that may or may not (being parasitically reduced) evolved out of an original parasitic cell's functions and may be additively informative on it.

    Finally, awesome figure! I look forward to the next article in this series.

    * The figure text is partly hidden, at least in Firefox, and was accessed through the page source code.

    Link to this
  4. 4. Torbjörn Larsson, OM 8:08 pm 11/3/2011

    “Further sequencing should help establish if this is valid”.

    Ehm, yes, or more simply and much more likely checking those genes for actual ancestry and finding my haphazard reading is wrong. I should be able to do it myself now that the databases et cetera is out there it seems. But it would take some time to learn how.

    Link to this
  5. 5. Sigmatize 2:03 pm 11/4/2011

    Thanks, Psi, for the introduction, and for your enthusiasm for the subject matter. Eukaryote studies embody all the phenomena of natural history just as geography embodies all the specialties of academia. Arguably, if not obviously, all of evolutionary developments since LECA are merely variations on the themes originated in LECA. Even politics and morality can be seen as nascent recapitulations of the symbiosis demonstrated among organelles. I eagerly await future posts.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Scientific American MIND iPad

Give a Gift & Get a Gift - Free!

Give a 1 year subscription as low as $14.99

Subscribe Now >>


Email this Article