Vital Questions, Part 2

Review of The Vital Question by Nick Lane, Part 2 of 3

Last week, I covered the fundamentals, the outline of the history of life on earth, including some unexpected findings and downright mysteries.  I’ll continue here to talk about some of the broadbrush features in the history of eukaryotes as Lane paints the picture, and what he thinks the picture is telling us.  Next time, I’ll talk about sex and evolvability.

Eukaryotes encompass the full diversity of macroscopic life on earth, every animal and plant, from yeast cells to jellyfish to sequoia trees to you and me.  Strikingly, there are lots of things that all eukaryotes have in common (see last week’s column) that no bacteria or archaeans have.

It’s Lane’s ambition to explain some of the basic features of the eukaryotic lifestyle: birth and death, genetic recombination with two sexes, the association between sex and reproduction (which exists in almost all eukaryotes, but not in bacteria).  He begins with the event that created the eukaryotes and gave them their advantage–the domestication of a sugar-burning bacterium that became universal energy source for all eukaryotic cells.

This view of life’s history, a 4-billion-year story, places the mitochondria right at the centre of the evolution of the eukaryotic cell…

This is the biggest new idea in the book.  How well does it hold up?


Linear chromosomes with Introns

In prokaryotes, almost all the DNA comprises legitimate genes, that is, it codes for proteins.  In eukaryotes, less than 5% of the DNA codes for protein.  When this fact was first discovered in the 1970s, non-coding DNA was called “junk” or “parasitic DNA”.  The story was that it served no function except to replicate itself, along for a free ride.  Natural seleciton in prokaryotes is a race to the fastest reproducer, so they can’t afford parasitic DNA, and it has been selected out.  But eukaryotes play a more sophisticated game (so the story goes), where “fitness” depends on survival strategies, and not raw speed of growth and reproduction.  Hence eukaryotes don’t pay a high penalty for all this junk DNA, and it has accumulated in the genome.

I have never bought into this story.  Just because we don’t know the function of a stretch of DNA doesn’t mean that there is none.  And indeed, over the years, functions have been discovered for non-coding DNA.  Some of it is translated into RNAs (ribozymes) which biologically active in many of the same ways as proteins.  Some of it is “genetic capacitance”.  The genome has learned to behave like a packrate, holding on to genes that were useful in the evolutionary past but not now, because the genome has learned from long experience that conditions are likely to recur some generations in the future, and better to carry extra baggage for a few million years than to have to start all over and re-invent the wheel if the need should arise again.

But all of it affects the way in which chromosomes spool and unspool along their length, opening as euchromatin which is active and expressed, or closing up as heterochromatin which is temporarily inactive.  This is epigenetic control, an intricate logic that determines which genes are expressed where and when.  Prokaryotes don’t differentiate into diverse cell types, and they don’t have life cycles that demand growth, maturation and development.  But eukaryotes do, and hence every cell needs to express just that subset of the genome that it needs to do its job at the moment.  I don’t find it at all surprising that 95% of the genome is devolted to the decision when and where to deploy the other 5%.  An encyclopedia of human DNA by the ENCODE project estimated that 80% of genetic material has a purpose.

Lane believes that much of the non-coding DNA is indeed parasitic junk, and some of his argument about the origin of the cell nucleus, sex and aging hinge on this characterization.  Compellingly, he notes that the total quantity of DNA varies enormously from one plant or animal to another, and appears to bear no relationship to the complexity of what that species does or how it lives.  Looking at frogs alone, we see some species that have 100 times as much DNA as other frogs.

It can’t all be “necessary”.   But that doesn’t mean it’s parasitic.  Maybe there just isn’t much incentive to do the job of epigenetic regulation efficiently, so some species use much more DNA than others.

Lane describes a barrage of parasitic DNA that came with the mergers and invaders that formed the first eukaryotic cell.  It was to protect against this invasion that the cell’s DNA was sequestered into a nucleus and surrounded by a membrane.

Every time a eukaryotic gene is transcribed, DNA is first copied as messenger RNA, and the RNA is then translated into a unique protein.  But non-coding DNA is interspersed through as introns.  So the RNA must be edited, snipping out introns and re-attaching the free ends until just the coding portion remains, ready to be run through a ribosome where it becomes blueprint for a protein.  The editing takes place inside th nucleus, and the ribosomes are outside the nucleus, assuring that no un-edited RNAs are transcribed to protein.

In bacteria, it is efficient to have the DNA spread through the cell and conveniently interspersed with ribosomes.  But in eukaryotes, it would be a mistake to transcribe the RNA before the introns are snipped out.  So DNA is kept walled off in a nucleus, and RNA remains in the nucleus long enough for it to be snipped and edited.

Lane documents the evolutionary origin of introns and gene splicing, demonstrating convincinginly that introns originated with parasitic DNA that lived within the bacterial genome and politely provided their own scissors!  What he doesn’t say is that this system of “genes in pieces” has major advantages in complex organisms.  The self-splicing intron may have begun as a bacterial parasite and developed into a useful attribute in eukaryotes; or it may have evolved in eukaryotes as a useful innovation, and then crept back into bacteria as a parasite.  But the system has advantages for eukaryotes that make me think that it is the product of natural selection.  There is an economy in constructing genes from reusable modules that can be combined in different ways to perform different fuctions.  There is an advantage in efficiency of evolution, as it is easier to combine functional pieces for a new application than to design a new protein from scratch.  And, as I mentioned above, it may be that much of the DNA serves an epigenetic function, implementing a decision where and when to activate each particular gene.


One Wave of Cell Mergers, or Continuing Episodes?

Eukaryotes encompass the full diversity of macroscopic life on earth, from yeast cells to jellyfish to sequoia trees.  Strikingly, there are lots of things that all eukaryotes have in common (see last week’s column) that no bacteria or archaeans have.  There are also genes that clearly come from bacteria that characterize all eukaryotes, and so they were presumably present in LECA, the Last Eukaryotic Common Ancestor, “mother of us all”.  And yet these genes come from at least 50 different bacteria, on widely separated bacterial taxa.

In Lane’s picture, all the merging of different genomes took place in an intense wave at the origin of eukaryotic life.  In support of this picture, he notes that most of this diverse genetic legacy is amazingly well conserved across the full range of different eukaryotes–protists, animals and plants.

But I can’t resist digressing to tell you about the alternative picture, as presented in Lynn Margulis’s book with her son Dorion Sagan, Acquiring Genomes.  They cite evidence that mergers and acquisitions have been continuing in an ongoing series of events that are rare only in a relative sense.  Many, many genome mergers have taken place in the history of eukaryotic life, continuing into the recent past.

Many insects look entirely different in their larval form and their adult form.

House fly looks nothing like its larva or its pupa.

The transition, for example from a caterpillar to a butterfly, is not a continuous change in shape.  Rather, inside the crysalis, the caterpillar dissolves and its biomass is reformed into a butterfly [ref].

Caterpillar looks nothing like the adult monarch.  The caterpillar does not "grow into" a butterfly , but rather dissolves itself within the crysalis and starts over to build a new body.

Caterpillar looks nothing like the adult monarch. The caterpillar does not “grow into” a butterfly , but rather dissolves itself within the crysalis and starts over to build a new body.

Margulis and Sagan take this as evidence that larva and adult were once different species, and that the two species became one when their genomes merged.

But why?  What possible advantage can there be to making two species into one?  My answer has to do with stabilizing populations.  If the young and the mature form are in the same niche, competing directly with each other, then the larger, mature individuals have every advantage.  Predators or disease or famine–all of the common causes of death hit harder against the very young.  Very few of the young will have a chance to grow to maturity, and the population pyramid will be bottom-heavy with young ’uns, most of whom will never have a chance to grow up.  If the larvae and the adult are not in direct competition, then it is easier to establish a stable population structure.

And how could it come about that a maggot’s genome would merge with a housefly’s, or a caterpillar’s with a butterfly?  The answer is xenogenous fertilization.  It is rare, but not all that rare, that an egg of one species can be fertilized by a completely different species.

In the Liverpool University lab of Don Williamson, successful hybridizations were achieved between distant species to produce fertile offspring.  Sea urchins are echinoderms, ancient globes covered with spines.  Sea squirts are chordates, “almost vertebrates” like us.  This is one of the odd pairings that Williamson was able to hybridize.

Sea Squirt is a chordate


Sea Urchin is an echinoderm.

If Williamson could do it in a few years of futzing around in the lab, then it’s likely that over hundreds of millions of years in the vastness of the oceans and the air, it has happened dozens, maybe hundreds of times that an egg of one species has been fertilized by a sperm of a very different species.

Williamson’s story is included in Acquiring Genomes, and his story of the Web of Life to replace the Tree was nicely summarized in a series of New Scientist articles a few years ago.

So what are we to make of these two very different stories about the Tree of Life?  Both pictures seem compelling to me, and I would guess there is truth in both stories.  Perhaps there was an intense wave of endosymbiosis around the time of the first eukaryotes.  Perhaps there were many different competing mosaics of different bacteria and archaeans, unlikely hybrid monsters that struggled to keep peace among their parts, and struggled with one another.  It seems that one of these won out, and the others died off without a trace.  Her name was LECA.

This column is getting long, and there is more I have promised you: sex, parasitic DNA and a favorite topic of mine, the evolution of evolvability.

’Til next week…


8 thoughts on “Vital Questions, Part 2

  1. This is a really great post!

    Any idea what is going on with the publication of the book? I’ve had it on preorder at Amazon for months and finally got an email yesterday saying it was unavailable and they cancelling the order. Probably will buy the Kindle version.

    Regarding junk DNA I had the impression perhaps incorrectly that much of it was “boring” in the sense that it was just repetitive sequences of the same nucleotides in the same order. Wouldn’t this argue for it not having a significant function? Also, if hybridization has had a significant role in creating new species, would we not expect that we would end up with a lot of junk DNA (not the “boring” kind) since once two species merge each species would have DNA not useful to the new species.

    • Hi, James –
      Re availability: I think you’re right – the hard copy of the book is available only in the UK.

      Re the nature of non-coding DNA: I think we don’t know yet. My guess is that there are significant physical constraints that arise from the spooling and unspooling as different genes are activated in different combinations. Maybe some of it is just “spacer” but the spacing helps make it possible to get better control of gene expression. This is all conjecture, but I would guess that there is a lot of evolutionary pressure to get the time and place of gene expression carefully regulated, and not very much evolutionary pressure to economize on the size of the genome.

      Re hybridization: there are worms and insects that have 4 or 5 or more distinct stages of their maturation. It’s possible that these evolved separately and later merged, and much of the original genomes remain active and useful.

      – Josh

      • There is at least 2 more conjectures I’ve come up with quite a while ago to add to our growing list:

        1) Probabilistic Shield Against Catastrophic Mutations. If everything is coding, ANY mutation will be to coding genes. Which would only not be 100% harmful due to the redundancies of aminoacid coding. Thus ~30%. In fact, I think that’s another example of a Probabilistic Shield: Codons. To get from 30% of a 100% to 30% of 5% in humans is quite significant.

        We can measure the power of this selective pressure by correlating the genome/non-coding DNA for different species with the environments they evolved with. For example, the frog you mention, does it has (or had) shallower waters as habitat, or is less nocturnal than the other frogs? Latitude could also be a factor.

        What about that ugly bush that has that gigantic genome? =) Not seeing much complexity there… How extreme is its environment? This may also complement the other hypotheses of evolvability: “Will try to protect against mutation but, if we perceive lots of mutations / repair going on, then we better do something” (Anthropomorphizing :))

        2) Avoid particular base sequences (or sequence length) of RNA to fold in itself, specially in the nucleus where it has little machinery to untie that knot / clean up. It could also end up being connected to complementary sequences on the DNA itself, mucking up everything. Granted, the later is less likely but, as above, catastrophic. And since everything would be 21 aminoacids (codons help again), that would not be so rare if it DID find an open strand. Which would explain strange sequences (non existent aminos, except for artificial) in non-coding DNA. The different base in RNA from DNA, could be explained as this and diminish the power of this conjecture as an evolutionary pressure. The folding on itself still remains though.

        • Your #1 has an error in logic, I think. If the genome were 1/20th as big, there would be 1/20th as much to copy, so the number of errors would be 1/20th. The rate of error in the coding areas would be the same. Your idea only works if you can copy a genome that’s 20 times as big without making 20 times as many errors.

          The rate of copy errors in the germ line is extremely low, less than 1 nucleotide in a billion, I think. My guess is that error rate is highly optimized, offering a very tiny chance of experimenting with something new in unstressed conditions. Mutation rates go up in stress conditions, presumably because the genome “knows” that something new is needed.

          • Hi Josh,

            not sure I quite understand, help me =)

            here’s what I got:

            I agree with “The rate of error in the coding areas would be the same. “; PROVIDED the ratio of coding to (so-called) junk DNA was exactly the same.

            That was exactly my point =) Why we see “so much junk” compared to the stuff we “care about”. Precisely because we do care about those to protect them with this “shield”.

            am I still missing something?

            A little aside, I had UV radiation in mind, harming the DNA. But I guess you’re right, it would have to go through copying to propagate the damage. Although if they are different damage mechanisms then there is one more thing to figure in the calculation. =)

            Also, I agree and I think evolution (and thus changes to the genome, such as adding “junk”) probably entail A LOT of stress and death and what not…


            Crazier digression, feel free to skip: =)

            About the calculation itself, I guess we’d have to multiply by the rate this happens per second, by the number of places it happens in 1 cell, by the number of germ cells being produced (in males, btw, how protected vs mutations are eggs?), by the number of seconds in the organism lifetime to be able to say how low it really is.

            Although if LECA took a while to become multicellular and if it branched even before that, and until it was complex enough to have dedicated cells for reproduction, we’d also have to multiply by all the cells in the organism?

            Anyway, not to sound too Lamarckian but 🙂 epigenetics seem to have a system to speed up the figuring out of what should be evolved next. What in Computer Science we’d call a Heuristic. As in a mouse having effects from 2 generations back. In fact, I’m an example of this (except for being a pandimensional mouse, lol), my grandma had a life/death situation during war trying to protect her 7+ children plus her still unborn: my father. Apparently she got a kind of “allergic” reaction after the fact, which all her body got psoriasis-like scars. My father was born with psoriasis (or susceptibility thereof) highly exacerbated by alcohol. I rarely show any symptoms tho I seem to have inflammation dialed up (bummer =/) although I don’t have any allergies nor seem to get sick easily. I don’t quite seem to get drunk except for dizziness but the susceptibility may be even higher, if all those calories are being converted to glucose quickly then it would probably be an effect comparable to drinking raw corn syrup? 🙂 (I mean, skin cells reproducing too fast without the proper nutrients, like folate? sounds like a possible mechanism of psoriasis to me)

            Granted, just by being here entails I’m concerned about health but, it shows an epigenetic heritability. None of his brothers or sisters have this. And perhaps even more curiously, he’s not the younger one! So it did not even involve germline cells!

  2. -Concerning non coding DNA you say “it is easier to combine functional pieces for a new application than to design a new protein from scratch”. I agree. That is just another example (like symbiogenesis leading to eukaryotic cells) in which COMBINATION (instead of competition leading to the contrary thing, “separation”) is another important mechanism of evolution of complexity.

    -“(temporarily)Silent genes” is another possible function (what you call “genetic capacitance”) for “non-coding DNA. I also agree. A good example could be the genes finally controlling levels of metilamines (like OTA) ? (functionally compensating for urea accumulation) in the evolution of vertebrates from sharks to frogs, to mammals.

    -But what about other possible functions for “non-coding DNA”?, eg. trasposon-like mechanisms to create new combinations (variation, finally, on which natural selection can act). Especially since various forms of stress seem to increase TE-like movement.

    I mention many other possible uses of “non-coding DNA” as well as evolutionary mechanisms to create new variation on Chapter 5 of my last book (Longevity en Evolution, Nova Sci. 2010, 2011)

  3. I have a scientist friend at a top University who says he has evidence of 200,000 – 300,000 short transcripts coming off from the introns, many of which are actually translated. Many of them are antisense sequence of exons and may serve some regulatory function (not known – they are different from microRNAs). Mutations in these genes could lead to diseases. It’s a wild story and is not published. Hard to publish such radical theories — or maybe it is an artifact. Hard to believe that people have missed this for so long.

Leave a Reply

Your email address will not be published. Required fields are marked *