Posted by: Dan | October 4, 2006

Origins of Gene Structure

Looking back at the evolution of cells prior to the expansion of multi-cellularity, there was the advent of the eukaryotic cell. How did it arise? The concept of endosymbiosis is of course a well-known theory on the origins of the eukaryotic cell, but what about the genetic structure itself? One of the leading thinkers in this area appears to be Michael Lynch, who recently penned a review in Molecular Biology and Evolution on The Origins of Eukaryotic Gene Structure. In this review Lynch addresses the correlation between genomic complexity and organismal complexity, proposing that the early developments of complexity are driven primarily by non-adaptive stochastic forces, rather than by Natural Selection alone. He draws support from observations of a general negative relationship between selection efficiency and genome complexity, and while this concept is rather speculative, it shows great appeal as a plausible alternative – afterall, how can we explain how “intron splicing, untranslated regions, modular regulatory elements, and expansive intergenic regions that harbor diffuse control mechanisms” might have been selected for?

As his alternative, Lynch proposes that random genetic drift and mutational pressures account for the acquired robustness and diversity of storage of genetic information – something very much echoed by observations on gene duplication in divergence of protein-DNA interactions, and evolution of digital organisms and computer viruses.

Of course Lynch doesn’t completely dismiss Natural Selection, but adaptive mechanisms along the strict Darwinian or Modern Synthesis modes are insufficient to explain the genome-wide changes. Put more succinctly:

… it is worth emphasizing that the goal here is not to explain the expression of a gene in a new temporal or spatial context, but to understand how a gene that is initially under control of a ubiquitously expressed transcription factor comes to be regulated by spatially and/or temporally restricted transcription factors while initially retaining the same overall expression pattern. The process envisioned, subfunction fission, invokes gradual structural modifications of preexisting enhancers within a gene (descent with modification) rather than the saltatory appearance of entirely new regulatory modules (Force et al., 2005). Subfunction fission involves consecutive phases of regulatory-region expansion and contraction.

Here, Lynch’s “gradual structural modifications of preexisting enhancers” clearly has appeal in justifying a stochastic or drift-driven acquisition of complexity, but what does he mean by mutational pressure as the other driver of complexification? He explains this with his findings on the genomic perils of increased organism size, which cause substantial reductions in the efficiency of natural selection. This makes sense for a couple reasons that can be reduced to a discussion on how easily it is for mutations to generate variation in eukaryotes versus prokaryotes – it’s easy to see that prokaryotes have an enormous potential for generating diversity in a population, the reverse of which is the benefit eukaryotes gain when they acquire new mechanisms of converting mutations (be they base-subsituted or transposed mutations) to the phenotypic level for adaptive selection to act upon. This is an intuitive (if complex) concept, suggesting an origin for what is now popularly referred to as epigenetics, or multiple levels of regulation for gene expression.

This may be tough for some to wrap their head around, but the over-riding theme in the Natural Selection v. Neutral Theory debate is that life works on a mixture of stochastic (genetic drift) and deterministic (adaptive selection) forces to generate phenotypic diversity at a rate fast enough to survive in a world of limited resources and harsh conditions.

As I mentioned towards the top, related principles of self-managing complexity in reproducing populations has been demonstrated by work in genetic algorithms and digital organisms. It’s interesting, I think, that some individuals seize upon related concepts to promote a “Front-loading” explanation for Intelligent Design, but this is an illogical argument — it rests upon the presupposition that both the mathematical rules of complex systems and the origin of life have as their causation a higher power: a wildly speculative view that sees what it wants to see.


For more reading, I recommend looking up Motoo Kimura’s 1983 book “The Neutral Theory of Molecular Evolution,” Susumu Ohno’s 1970 book “Evolution by Gene Duplication,” and topics on “Digital Organisms.”



  1. You can’t talk about Lynch’s model without mentioning population size. It is, essentially, Ohta’s nearly neutral theory extended to the evolution of genomes. Natural selection is most effective at removing weakly deleterious mutations (and fixing mildly adaptive mutations) in larger populations. Lynch argues that introns, transposable elements, and complex regulatory structure are slightly deleterious. Because bacteria have larger population sizes than eukaryotes (and single celled eukaryotes tend to have larger pop sizes than multicellular eukaryotes), complex genomes have evolved in eukaryotes.

  2. I know – I didn’t quite emphasize population size to the degree that Lynch’s model applies it – and this is because I was trying to describe his model in the simplest, most intuitive way that I could think of, that I was trying to get at in the paragraph following the blockquote.

  3. The concept of endosymbiosis is of course a well-known theory on the origins of the eukaryotic cell…

    There is competition now. I find the viral hypothesis for the origin of eukaryotes to be very interesting. See for example:
    The two ages of the RNA world, and the transition to the DNA world: a story of viruses and cells
    Patrick Forterre
    Biochimie, Volume 87, Issues 9-10 , September-October 2005, Pages 793-803

  4. Ivy,
    Forterre’s hypothesis on viral-cell competition is a very good topic for discussion, but I don’t know that it constitutes a competing or alternative theory – it doesn’t contradict the argument that organelles originated via endosymbiosis at the prokaryote-eukaryote boundry. Instead it focuses on an earlier stage that hasn’t gotten much discussion time when it comes to early life: that of how a mixture of early nucleic acids (and maybe lipids and/or peptides) might have evolved in the RNA World scenario.

    I argue here that evolution of the RNA world taken place in a framework of competing cells and viruses (preys, predators and symbionts). I focus on the RNA-to-DNA transition and expand my previous hypothesis that viruses played a critical role in the emergence of DNA. The hypothesis that DNA and associated mechanisms (replication, repair, recombination) first evolved and diversified in a world of DNA viruses infecting RNA cells readily explains the existence of viral-encoded DNA transaction proteins without cellular homologues. It also potentially explains puzzling observations from comparative genomic, such as the existence of two non-homologous DNA replication machineries in the cellular world. I suggest here a specific scenario for the transfer of DNA from viruses to cells and briefly explore the intriguing possibility that several independent transfers of this kind produced the two cell types (prokaryote/eukaryote) and the three cellular domains presently known (Archaea, Bacteria and Eukarya).

    His ideas for RNA-to-DNA transition make some sense (although I’ll want to read up on that a bit mroe when I get a chance, I’m not overly familiar with this), but I’m very skeptical that “independent transfers” of this kind would be responsible for the prokaryote/eukaryote transition, the comparmentalization of genomic information into the nucleus and chromosomes, the occurence of organelle genomes, or the distinct cell membrane architectures amongst archaea, bacteria and eukarya.

    However, I think this viral hypothesis’ strengths are the diversity of viral mechanisms for transfer of genetic information. The occurence of ssRNA, dsRNA, ssDNA, dsDNA, reverse-transcribed, and other viral genomes seems to make more sense under this hypothesis. The one clue that viral competition might have been involved in the prokarya-eukarya transition, that I can think of, is that of Herpes-like viruses, which incorporate their DNA into eukaryotic chromosomes – this may have either been part of viral competition as part of Forterre’s hypothesis, or evolved later in the changing world.

    That’s my intial take on it, anyway – does that sound about right?

  5. However, I think this viral hypothesis’ strengths are the diversity of viral mechanisms for transfer of genetic information.

    Motive, method, opportunity. The viral hypothesis also provides motivation for such genetic changes, as hosts would be selected for resistance to infection, and virii would be selected for resistance to disablement. This could help explain apparently unnecessary complications, such as introns.

  6. […] Unfortunately it’s sometimes difficult to keep up with the pile of papers on my desk, and blogging about them. But I’m getting around to them – and one that I particularly wanted to come back to was originally recommended to me a while back by Ivy Privy: “The Two Ages of the RNA World, and the Transition to the DNA World: a Story of Viruses and Cells. […]


%d bloggers like this: