Which came first, the bird or the smaller genome?

It’s easy to think of a genome as a collection of genes, perhaps because so many of the metaphors used to explain genes and genomes (blueprint, book of life, Rosetta Stone) can give one the impression that everything in a genome is useful or functional. But genomes are, in fact, packed with debris. Many genomes contain huge collections of fossil genes: genes that have been inactivated by mutation but were never discarded, sort of like the old cheap nonfunctional VCRs in my basement. And many genomes contain even more massive collections of another kind of fossil-like DNA: mobile elements, or their remnants. The human genome, for example, contains over 1 million copies of a single type of mobile genetic element, the Alu transposon. Together, the various types of mobile genetic elements comprise nearly half of the human genome.

Think about that. Almost half of the human genome is made up of known mobile elements, pieces of DNA that can move around, either within a genome or between genomes with the help of a virus. This extraordinary fact -- and many of the specifics surrounding it -- constitutes one of the most compelling sources of evidence in favor of common descent, the kind of data for which only common ancestry provides a complete (or even reasonable) explanation. I’ll come back to this topic regularly.

Now it turns out, not surprisingly, that differences in genome size among types of organisms are determined primarily by the numbers of these mobile elements, and not by the number of genes. In fact, there is wild variation in genome size among types of organisms, and the variation has little to do with the numbers of genes expressed by those organisms. Consider birds, the subject of this week’s Journal Club (“Origin of avian genome size and structure in non-avian dinosaurs,” Organ et al., Nature 446:180-184, 8 March 2007).

Birds have remarkably small genomes, averaging 1/2 to 1/3 of the size of typical mammalian genomes. (The chicken genome, for example, is less than half the size of the mouse genome.) Why might this be? In other words, how might we explain this difference? The authors point to two important ideas. First, the chicken genome has been fully sequenced and analyzed, and it contains far less of the debris mentioned above. It seems that the processes that create (or multiply) mobile genetic elements are significantly less active in birds than in mammals and other vertebrates. Second, small genome size is intriguingly correlated with flight. Bats, compared to other mammals, have small genomes, and flightless birds, compared to other birds, have larger genomes. This has led to the proposal that small genome size might offer a selective advantage to flying animals, by reducing the energy cost associated with hauling all that debris around. So, it seems that a smaller genome is advantageous for flying vertebrates, and that genome size can be reduced by restraining the production of mobile genetic elements. And this raises several interesting questions, including this one: did the reduction in genome size accompany the origin of bird flight, or did it happen in advance? In other words, we can propose at least two alternative scenarios:

• 1) flight drove the genome change, by favoring small genomes, or
  
• 2) the genome change happened first, and helped to get flight off the ground. ;-)

How can we even hope to distinguish between these possible explanations? We would need, somehow, to look at the genomes of the ancestors of birds. And all evidence indicates that the relevant ancestors of birds are dinosaurs; in fact, today's birds are considered to be flying dinosaurs. The recent description of protein sequences from T. rex bone provided strong confirmation of the birds-from-dinosaurs hypothesis, but no DNA was recovered from the samples, and no information about genome structure can be inferred from those otherwise fascinating studies. If only, a la Jurassic Park, we could get some dino DNA...

Enter Organ et al. with a wonderfully creative idea. It turns out that, in organisms alive today, cell size is strongly correlated with genome size. In other words, organisms with large genomes tend to have larger cells. This relationship was first described in red blood cells, but Organ et al. show that it holds quite well in bone cells as well. Using bones from living species, they created a statistical model that enabled them to infer genome size by looking at the size of bone cells. Then they combined their model with measurements of bone cell size from fossilized bones of long-extinct animals, and were able to estimate the genome size of dozens of extinct species, including 31 dinosaur species and several extinct bird species. Their results are remarkable: small genomes are found in the entire lineage (with one interesting exception, Oviraptor) that gave rise to birds, all the way back to the theropod dinosaurs that are the typical reference point in the dinosaur-to-bird story. Here's how the authors put it: "Except for Oviraptor, all of the inferred genome sizes for extinct theropods fall within the narrow range of genome sizes for living birds." Even if you don't have access to Nature, you can have a look at the cool family tree in Figure 2, which shows small genomes in red and larger ones in blue. It's a compelling image.

The results suggest that small genomes arose long before dinosaurs took to the air, and raise some interesting questions about the interplay of physiological function (e.g., energy consumption associated with flight) and genome structure. Certainly scenario #1 above is not favored by these findings: flight apparently arose in organisms that already had much smaller genomes than many of their earthbound cousins. The relationship between flight and small genome size, then, remains unclear and even mildly controversial. Organ et al. acknowledge that the two characteristics did not arise together, but after reference to the larger genomes in flightless birds, they conclude their paper by noting that "the two may be functionally related, perhaps at a physiological level." And they postulate that small genome sizes may have been favored by warm-bloodedness and its associated energetic demands. But a minireview of the paper raises several criticisms of these hypotheses, and it is clear that the evolutionary forces acting on genome size are complex and yet poorly understood.

Notwithstanding the unanswered questions regarding genome evolution, this paper is the kind of scientific article that should be carefully considered by those who deny common descent. Following are some aspects of the story that create interesting questions for creationists and/or design advocates.

Consider the results presented in Figure 2. Outside of common ancestry, how are we to account for these data? The strong correlation between flight and small genome size in living organisms might look like some kind of "design" to someone who favors that sort of thinking, but Organ et al. have conclusively uncoupled genome size and flight. Of course those of us who see the universe as a creation will be happy to marvel at the advantages presented by small genomes to flying organisms, and perhaps we'll all think of these wonders as evidence of brilliant "design." But it seems to me that "design" does not serve a significant explanatory role here. On the contrary, I maintain that the work of Organ et al. demonstrates the following: in dinosaur lineages, the best way to predict genome size in an extinct species is to know the ancestry of the species. Common design aspects don't help. Common descent explains the pattern.

And yet, I think it gets much worse than that for anti-evolution thinkers. I regularly see certain old-earth creationists (e.g. the folks at Reasons To Believe) and design proponents (e.g. William Dembski) arguing that "junk DNA" (which includes, but is not limited to, the 45% of the human genome composed of mobile elements and their debris) is not "junk" but can have important functions. (The arguments of these critics are flawed in several ways, which I'll detail some other time.) While it's true that mobile elements have contributed to the formation of new genes from time to time, and are thought to be significant sculptors of genomic evolution, it's also true that mobile elements are indiscriminate in their jumping, and their continued hopping about is a documented cause of harmful mutation. Here, though, is a significant quandary for a design advocate considering a bird genome: if these mobile elements have important functions in the organism, then how is it that birds can get by with 1/4 as many of them as, say, squirrels? Why, if these elements have important functions in the organism, do bats seem to need far fewer of them than, say, rats? (The genome of the big brown bat is 40% the size of the genome of the aardvark. Hello!) It seems to me that these facts are best understood when one considers the possibility that most of this DNA is essentially parasitic, and that some types of organisms have benefited by restraining its spread. A "design" perspective with regard to genome size is just not helpful, and if that perspective insists on excluding common ancestry, then it's worse than worthless.

Article(s) discussed in this post: Organ, C.L., Shedlock, A.M., Meade, A., Pagel, M. and Edwards, S.V. (2007) Origin of avian genome size and structure in non-avian dinosaurs. Nature 446:180-184.