The origin of complex life – it was all about energy

By Ed Yong | October 20, 2010 1:00 pm


The 21st century is all about conserving energy. The push towards energy-efficient buildings, vehicles and lifestyles is both fashionable and necessary, but it’s also ironic. Our pattern of ever-increasing energy consumption is deeply rooted in our history, not just since the Industrial Revolution, but since the origin of all complex life on Earth.

According to a new hypothesis, put forward by Nick Lane and Bill Martin, we are all natural-born gas-guzzlers. Our very existence, and that of every animal, plant and fungus, depended on an ancient partnership, forged a few billion years ago, which gave our ancestors access to unparalleled supplies of energy and allowed them to escape from the shackles of simplicity.

To Lane and Martin, energy supply is the key factor that separates the two major types of cells on the planet. The first group – the simple prokaryotes, such as bacteria and archaea – are small, consist entirely of single cells (or at most, simple colonies), and have little in the way of internal structure. They are very different to the eukaryotes, the group that includes all complex life on the planet, including every animal, plant, fungus and alga. Their cells are large, structured, and filled with many internal compartments. These include the nucleus, where DNA is stored, and the mitochondria, which act as tiny powerhouses (more on these later).

Prokaryotes can do many incredible things. They can eat food as unappetising as oil and live in places where literally not a single other living thing can thrive. But despite their boundless innovations, they have always remained simple. While eukaryotic cells have evolved into large and complex forms like animals and plants on at least six separate occasions, prokaryotes have always remained simple. Some have nudged into more complex territory – for example, by becoming incredibly big – but all of these pioneers have stopped short. Only once in history have simple prokaryotes made the transition to complex eukaryotes. Why?

Lane and Martin think that the answer lies within the mitochondria. They were once prokaryotes themselves. In the depths of history, a free-living bacterium was engulfed by a larger cell and was neither digested nor destroyed. Instead, it was domesticated. It forged a unique and fateful partnership with its host, eventually becoming the mitochondria of today. All of this happened just once in life’s history and all of today’s eukaryotes are descended from that fused cell. Indeed, many scientists view the origin of mitochondria as the origin of the eukaryotes themselves.

Mitochondria are the power centres of eukaryotic cells. Within their walls, proteins carry out chemical reactions that combine food with oxygen to produce ATP, the molecule that acts as a cell’s energetic currency. These proteins sit inside the mitochondrion’s inner membrane, which is repeatedly folded like ruched fabric. These folds provide a greater surface area for energy-producing reactions to occur, allowing the mitochondria to produce a substantial supply to its host. That gives eukaryotes a major advantage over their prokaryotic peers: with more available energy, they can afford to have more genes.

The transition from a simple prokaryotic cell to a complex eukaryotic one was accompanied by a large rise in the number of genes. The average bacterium only has around 5,000 genes but even the smallest eukaryotes have an average of 20,000. But having more genes comes at a cost. The simple act of reading the gene and translating it into a protein (biologists would say “expressing” it) requires energy. This process takes up a whopping 75% of a cell’s energy budget.

In the same way that a gadget-hoarding human would ramp up a sizeable electricity bill, a cell with a larger genome would face a substantial energy burden. And just like the gadget fanatic has a limited budget to spend on their escalating bills, a bacterium has only so much energy to devote to expressing its genes. Every extra gene reduces the amount of available energy per gene. Its only option would be to produce fewer proteins, which would put it at a disadvantage compared to its peers.

So prokaryotes are stuck in an energetic rut. They sit at the bottom of a deep evolutionary canyon, surrounded by steep walls that require a massive influx of energy to scale. Unable to surmount these barriers, they are stuck with small genomes and simple structures. Indeed, evolution tends to push bacteria towards ever more compact genomes, mercilessly pruning away superfluous genes. Today, in a million ‘letters’ of DNA, eukaryotes have around 12 genes while the average bacterium has around 1,000!

Eukaryotes, however, are not so constrained. Thanks to their mitochondria, they have energy to spare. The average eukaryote can support a genome that’s 200,000 times larger than that of a bacterium, and still devote a similar amount of energy to each of its genes. As Lane and Martin say, “Put another way, a eukaryotic gene commands some 200,000 times more energy than a prokaryotic gene.”

The eukaryotic genome is like a gas-guzzling monster truck, compared to the sleek, sports-car genomes of prokaryotes. The benefits of this lumbering size can’t be overstated. By having enough energy to support more genes, they have room to experiment. It’s no surprise that the diversity of eukaryotic genes vastly outstrips that of prokaryotic ones. The last common ancestor of all eukaryotes had already evolved at least 3,000 entire families of genes that the prokaryotes lack, and it had complex ways of controlling and regulating these newcomers.

But why haven’t prokaryotes evolved a workaround that produces the same benefits as mitochondria? If all it takes is an internal, intensely-folded compartment, then bacteria should have been able to evolve that. Indeed, some have evolved internal folds like those of mitochondria. Why are they still stuck in their energetic canyon?

The answer, according to Lane and Martin, is that mitochondria give eukaryotic cells something special that bacteria will never have, no matter how many folds they develop – an extra set of DNA. Having evolved from free-living bacteria, mitochondria have a tiny genome of their own. Most of the genes from the original bacteria have emigrated to the host cell’s main genome but those that remained in the mitochondria include those that are responsible for liberating energy from food and oxygen.

Having these energy-production genes close at hand means that mitochondria can react very quickly to any changes in their folded membrane that would hamper their abilities to fuel their host cell. Put simply, eukaryotes cells need the tiny amounts of DNA in their mitochondria in order to get a steady energy supply. Lose that DNA, and catastrophic blackouts ensue. Without this close association between extra membranes and energy-producing genes, prokaryotes cannot hope to achieve the huge and stable supplies necessary to become bigger and more complex.

In some ways, the exceptions here prove the rule. Epulopiscium fishelsoni is a giant bacterium that’s about as big as the full stop at the end of this sentence, and certainly a match for many eukaryotes in size. It has solved the problems posed by giant size by having as many as 600,000 copies of its full genome in every cell, dotted around its edges. Even this giant prokaryote needs to have genes in close proximity to its membrane.

But this strategy would never allow prokaryotes to achieve eukaryote-style complexity. It’s a false economy. The problem with Epulopiscium’s strategy is that it had hundreds of thousands of copies of its entire genome and every time the bacterium divides, all of that DNA needs to be copied. That is a massive energy drain that leads to the exact same problem that smaller bacteria face – the amount of available energy per gene is tiny. Faced with the same supply problem, Epulopiscium will remain a prokaryote.

By contrast, mitochondria have jettisoned the vast majority of their genes, so that copying their tiny remaining genomes is a cinch. They give a lot, but require little in return. They provided the first eukaryote with the equivalent of thousands of tiny batteries, giving them the extra power they needed to expand, evolve and experiment with new genes and proteins. Indeed, the rise of the eukaryotes was the greatest period of genetic innovation since the origin of life itself. As Lane and Martin write, “If evolution works like a tinkerer, evolution with mitochondria works like a corps of engineers.”

If Lane and Martin are correct, then their ideas on the importance of mitochondria have big implications for the evolution of eukaryotes. There are two general schools of thought on this (which I covered in greater depth in a previous post). One says that eukaryotes are descended from bacterial ancestors, and that they were well on the way towards evolving a complex structure before one of them engulfed the bacterium that would eventually become a mitochondrion.

But if mitochondria were the key the eukaryotic complexity, as Lane and Martin argue, then that model can’t be right. Instead, it’s the second model that is more plausible: that the first eukaryote was forged from a chance encounter between two prokaryotes. One swallowed the other and it was at this very moment that the first eukaryote came into being. Only then, with a surge of power, did all the characteristic features of eukaryotes start to evolve. It was a singular evolutionary step, when prokaryotes leapt out of their energetic canyon into the plateaus of complexity lying beyond, literally in a single bound.

Reference: Nature:

Related posts:

CATEGORIZED UNDER: Bacteria, Evolution, Select

Comments (25)

  1. Maria

    Fascinating stuff, and fabulously written – I particularly love the app analogy.

    Minor edit suggestion: “eukaryotes have around 12 genes” should probably read “prokaryotes have around 12 genes”.

    Thanks for your hard work!

    [Thanks, but the sentence you singled out is right. Eukaryotes are less gene-dense – Ed]

  2. dcwarrior

    I guess the eukaryote revolution was a big, non-trivial thing – it took what, 3 billion years after the first prokaryote for the first ones to appear? And then only a couple hundred million years after that for multicellular life to appear?

  3. posts like this is why they pay u the big bux dawg 😉

    I guess the eukaryote revolution was a big, non-trivial thing – it took what, 3 billion years after the first prokaryote for the first ones to appear? And then only a couple hundred million years after that for multicellular life to appear?

    i’d always assumed that the emergence of mitochondria via symbiogenesis wasn’t that much of a bottleneck, but you know what, perhaps it is. perhaps the universe is filled with “prokaryotic” life?

  4. Aleksandar Kuktin

    The problem with this hypothesis, like so many other, is that it can be abbreviated into the following formula:
    step 1. prokaryotes
    step 2. [magic] (single spell, maybe two)
    step 3. prokaryotes and eukaryotes

    The usual justification for step 2 being something along the lines of “yeeaaah, it is similar.. kind of.. I think I can imagine it.. yeah, it must be it! It’s good!”.

    In this case, step 2 is most obvious with the construction “the first eukaryote was forged from a chance encounter between two prokaryotes. One swallowed the other and it was at this very moment [notice the implicit reference to magic?] that the first eukaryote came into being.”

    It doesn’t work like that. You need to have extensive pre- and post-fusion (co)evolution of the two organisms for it to work. And even then, the two organisms must have a compelling reason to stay together. Knowing evolution, the reason that made them come together in the first place is different from the reason that keeps them together today.

    On a related note, the following does not compute:
    “The average eukaryote can support a genome that’s 200,000 times larger than that of a bacterium, and still devote a similar amount of energy to each of its genes. As Lane and Martin say, “Put another way, a eukaryotic gene commands some 200,000 times more energy than a prokaryotic gene.””

    If (a) the eukaryote can supply 200000 times more energy to the genome, and keep the amount of energy per gene the same; then how (b) can Lane and Martin say that a single eukaryotic gene commands 200000 more energy then a prokaryotic gene?
    Looks like a classical case of an equation mixup (someone multiplied when he/she needed to divide or vice versa).

    Oh, and the third thing is the fact that MAYBE the prokaryotes don’t NEED to change. 😉 Just like sharks – their bodyplan did not change for eons. Why? Because it’s perfect. Maybe bacteria are perfect too, but we just fail to grasp their genius? 😉

  5. Steve Morrison

    Some words seem to be missing from the start of the third paragraph.

    [Fixed – Ed]

  6. @Razib – The last paragraph of the paper speaks to your point. “The transition to complex life on Earth was a unique event that hinged on a bioenergetic jump afforded by spatially combinatorial relations between two cells and two genomes (endosymbiosis), rather than natural selection acting on mutations accumulated gradually among physically isolated prokaryotic individuals. Given the energetic nature of these arguments, the same is likely to be true of any complex life elsewhere.”

  7. Wayne

    The Halmark event, Cell enveloping Cell to create a new form of complexity, has happened on two occasions by my count.
    The first was mitocondra as described here, the second is chloroplasts.

  8. Dan

    I’d of course extend this position to the origin of not just complex life, but life in the first place. But then again I favor the Metabolism-first hypothesis of the origin of life. You just can’t have life without some rudimentary form of chemical energy consumption.

  9. amphiox

    re #4;

    If I recall correctly, in at least 2 of Lane’s books for general audiences he spends quite a lot of time describing in detail hypotheses concerning how step 2 may have came about. (Hint: magic is not involved)

  10. amphiox

    re #7;

    Once phagocytosis evolved, additional endosymbiosis events become much more routine. (Consider that chloroplasts actually constitute several, not one, endosymbiotic events).

    But phagocytosis is a very energy-intensive process (for one thing it requires a dynamic cytoskeleton, which itself is a very energy-intensive thing to maintain), and very likely required the pre-existence of mitochondria to evolve in the first place.

    Basically, once the first endosymbiotic merger occurred, resolving the energy-generation problem, further endosymbiotic events that increase complexity even further become much easier. But the question remains as to how the very first endosymbiotic event happened.

  11. Excellent, one of your best! I have a minor criticism though: when of Epulopiscium fishelsoni you write”It has solved the problems posed by giant size by having as many as 600,000 copies of its full genome in every cell”, isn’t “in every cell” a bit confusing? It makes me think of organism with many cells, which a prokaryote by definition isn’t… (ok, not exactly by definition, pro karyon means “before the nucleus”, not “one cell”, but still…)
    It’s not relevant to this article, but I’m going to write it anyway: this week I had to read a lot of papers on a subject which was supposed to be interesting, and in the end it really wasn’t, mostly because many of the papers were so badly written… I often thought “How, I wish Ed Yong had summarised it” 😀 (but please promise you won’t ever cover the Tilman-Grime debate on plant communities in this blog)

  12. “All of this happened just once in life’s history and all of today’s eukaryotes are descended from that fused cell.”

    The second half of this sentence can be correct, without requiring the first half of the sentence to be correct. IOW, there could have been multiple “domestications” of proto-mitochondria, but only one prevailed in the end. That makes the probability of the event seem less extreme.

    Much more speculatively: bacteria trade genes alot, and these early eukaryotes may have done the same. Multiple domestications of proto mitochondria could have traded genes for a while before resulting in a jumbled result that was the last common ancestor of all eukaryotes. My speculation is that this trading gene period could have been important but still too short to be teased out of the genetic record. Maybe it wasn’t a single event that resulted in eukaryotes.

  13. From my point of view, this is pretty dramatic stuff. I see a kid’s book in it. There aren’t enough picture books on science.

  14. Tom

    Great blog, glad I stumbled here from a friends recommendation. I am stretching back into the archives of my mind, and have not credible source to recall, but I seem to remember that the first symbiotic event was that of chlorophyll, and that the domestication of mitochondria came later? I am now recalling a diagram in the text book ‘Biology’ by Campbell and Reece, and the ejection of chlorophyll from non-autotrophs to leave only mitochondria? It may have been a hypothesis on the authors behalf though, or just my incorrect recollection.

    Cheers again, good reading.

  15. 1) Ed, you write wonderfully.

    2) This hypothesis explains a lot of things in genomics. It’s a simple concept though: I wonder why nobody had thought about that before!

    3) If we fill our cells with mitochondria will we speed up evolution? :-)

  16. I’m surprised to see this described as a new hypothesis due to Lane and Martin. I thought this was Lynn Margulis’ theory of endosymbiosis of some time ago. For example see this and this.

  17. amphiox

    Russ Abbott, I believe it is the energy considerations that is the new idea from Lane and Martin. Margulis established that mitochondria and chloroplasts originated from endosymbiotic bacteria, and this is no longer in any dispute. But the details with respect to the actual sequence of events that resulted ultimately in the eukaryotic cell remain open for debate, and specifically where the mitochondrial endosymbiosis falls in that sequence. On one end of the spectrum you have the so-called “Fateful encounter” hypotheses, which propose that the mitochondrial endosymbiosis occurred very early, between two true prokaryotes, and the subsequent development of the other uniquely eukaryotic traits (such as large size, complex internal membranes, nucleus, active cytoskeleton, etc) arose as a consequence of, or were enabled by, that initial merger. On the other end of the spectrum are the “Primite phagocyte” hypotheses, in which it is supposed that the mitochondrial endosymbiotic event occurred late in the sequence, that the host cell was already essentially a protoeukaryote, with an active cytoskeleton and capable of phagocytosis (all evolved without the presence of mitochondria), which went on to engulf the mitochondrial ancestor to produce the modern eukaryotic cell.

    Lane and Martin are essentially using energetics considerations to argue in favor of the “fateful encounter” end of the spectrum.

  18. @amphiox.

    OK. It’s a very good point that energy is essential to maintaining more complex structures. And I like the discussion of the need to have a separate and small genome to process energy. It’s not clear to me how this tells us whether the merger occurred early or late. All it seems to tell us is that the merger occurred before prokaryotes reached the limit of complexity imposed by their pre-merger energy structure.

    From my perspective as a non-biologist it’s not really all that interesting when the merger occurred. What’s more interesting (at least to me) is the fact that it was a necessary pre-requisite to further development.

    Did Lane and Martin make that point? (That’s a real question; I don’t know.) Did Margulis not know that the merger and a small energy genome was necessary for complexity? (Again, that’s a real question.)

    These do seem to be separate issues: (a) that a small genome for energy is required and (b) how it came about. An alternative for (b) — at least in the abstract — might have been that a portion of a prokaryote genome somehow broke off and persisted independently in the cell. In that case it would have been an internal division rather than a merger. (I have no idea whether that’s even biologically possible! It’s a top-of-the-head thought.)

  19. hypatiaa

    I don’t think endosymbiosis of this kind only happened once. There’s a hypothesis about how the double membrane gram-negative bacteria evolved from an ancient actinobacterium and an ancient clostridium (and thus setting the stage for an oxygen-rich atmosphere), that also involves endosymbiosis. This union came before the evolution of the bacteria which evolved into the mitochondria organelle. While it is a hypothesis, I think it’s a strong one and would drive a nail into the idea that this kind of endosymbiosis resulting in a fused cell is rare.

    Also, as far as inbetween forms – as far as I know, endosymbiosis does not happen in a single generation – there would be a long period where one cell type would likely be an ectosymbiont or even a parasite (as mitochondria’s closest bacterial relatives are). There is a recent example of an ectosymbiosis between a giant archaea and sulfur-oxidizing bacteria that results in a large multicellular archaeal filament coated in sulfur oxidizing bacteria, putting an interesting twist to how multicellularity may have evolved in its earliest stages.

    I’m not a biologist, but simple logic would dictate that an energy revolution resulting in eukaryotic cells would enable them to multiply exponentially to occupy all available niches for that sort of critter pretty fast, and thus make it difficult (but not impossible) for this type of evolution to happen a second time in a short time frame. So my guess is that it would be slow to evolve an endosymbiosis, but probably not hard, but difficult to do again once the first critter with such an energy advantage got a head start.

    All in all, I find the research fascinating and I find the hypothesis about the energy revolution plausible in the main, I only question the assumption that endosymbiosis between two prokaryotes would be rare.

  20. @ hypatiaa

    Great stuff. Do you have a reference (accessible to a non-biologist) for this:

    as far as I know, endosymbiosis does not happen in a single generation – there would be a long period where one cell type would likely be an ectosymbiont or even a parasite (as mitochondria’s closest bacterial relatives are). There is a recent example of an ectosymbiosis between a giant archaea and sulfur-oxidizing bacteria that results in a large multicellular archaeal filament coated in sulfur oxidizing bacteria, putting an interesting twist to how multicellularity may have evolved in its earliest stages.

  21. amphiox

    @Russ Abbot;

    Lane has a very thorough and easy to understand description of the basic reasoning behind this theory in his book “Power, Sex, Suicide: Mitochondria and the Meaning of Life”. Further information can be found in Lane’s third book “Life Ascending” in the chapter on the eukaryotic cell.

    The thing to remember is that phagocytosis has never, ever (at least to date) been observed in a prokaryote, and is currently considered to be a primitive eukaryotic trait (all eukaryotes either have it or secondarily lost it). It is also intimately associated with a dynamic cytoskeleton, which is also very much a signature eukaryotic trait.

    The question at hand is did the mitochondrial endosymbiosis occur before or after phagocytosis evolved. Phagocytosis provides an easy starting mechanism for multiple instances of endosymbiosis, so if the mitochondria were acquired after phagocytosis, then it is just one (perhaps the first) of many similar events, and there is nothing unique of special about it. If on the other hand, the mitochondrial endosymbiosis occurred before phagocytosis, then it must have arisen from some other mechanism, which given the paucity of known endosymbiosis in prokaryotes, might have been a very rare and unlikely event.

    Lane and Martin are essentially arguing that due to the energetics involved, phagocytosis can not evolve in cells that do not already have mitochondria, and so the mitochondrial endosymbiosis could not have occurred simply as the result of a phagocytotic event, and must have happened before the evolution of phagocytosis.

    In the most extreme version of their theory, the mitochondrial endosymbiosis is not just an “early” event, it is the enabling event. In their theory virtually everything unique about the eukaryotic cell, from large genome, to multiple introns, to straight chromosomes, to large amounts of noncoding “junk” DNA, to dynamic cytoskeleton, to phagocytosis, to large size, to loss of the bacterial cell wall, all of it, arose as a direct consequence of the fateful mitochondrial merger.

    Once the fully formed eukaryotic cell evolved, further endosymbiotic events were relatively easy, and occurred multiple times. But the first one was special, unlikely, and unique.

    This hypothesis can of course be falsified by a single demonstration of actual phagocytosis in a prokaryote.

  22. amphiox

    Another interesting thing to contemplate with respect to the “uniqueness” of the mitochondrial endosymbiosis is the fact that it occurred between an archaeon and a bacterium. It could be that it was a merger of two such different and in many ways incompatible genomes (for example, they have completely different pathways for the replication of DNA, very different ribosomes and thus protein synthesis, and lipid membranes composed of wholly different components) that provided the evolutionary opportunity for eukaryotes to develop complexity. The merger would provide the new proto-eukaryote with two wholly different and independent mechanisms for a large number of its most essential life processes. This would have provided both enormous genetic raw material (for all intents and purposes it was functionally like a giant macromutation, a genome duplication plus instant massive changes to all the duplicated genes plus the sudden appearance of a huge amount of completely new genes) and a very powerful selection pressure in the challenges of merging the two initially independently regulated genomes into something that can actually functionally work. It’s easy to imagine how such an endosymbiotic event could result in a complete, non-functional mess, leading to the mutual destruction of both symbionts, and quite possibly, many did result in just such a disaster, without only a few, or one, surviving.

    Possibly an endosymbiosis between two bacteria (a few I think are known or suspected) or between two archaea (of which I’m not sure if any are known, but of course we currently know less about archaeans than we do about bacteria) would not have produced the same kind of evolutionary opportunity.

  23. Dan Mills

    You need to define “complexity,” bud.

  24. MarkR

    EXACTLY! amphiox, Exactly.

    That would make life elsewhere in the galaxy even less likely than just the fateful encounter, because it could only occur on those wet, rocky, alkaline hydrothermal vent filled planets that produced at least TWO different lines of free living single celled biochemical machines, er, lifeforms. That last sentence is to highlight that, no matter what religious people (and others) believe, biochemistry is only microminiturized geochemistry, and that life is only different from energetic geochemical systems in that size and free floating status.

  25. MarkR

    Sorry, that would make complex cells and multicellular life, and therefore intelligent aliens, even less likely.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Not Exactly Rocket Science

Dive into the awe-inspiring, beautiful and quirky world of science news with award-winning writer Ed Yong. No previous experience required.

See More

Collapse bottom bar