Your Bones Got a Little Genome

Genome size can be measured in a variety of ways. Classically, the haploid content of a genome was measured in picograms and represented as the C-value. People began to realize that the C-value was not correlated with any measures of organismal complexity and seemed to vary unpredictably between taxa. This was known as the C-value paradox, and it confused geneticists for quite a while. With an increased understanding of genome structure, however, came the resolution of the paradox: this measure of genome size does not correlate with gene content. The majority of many eukaryotic genomes consist primarily of non-genic (and, presumably, non-functional) sequences.

With the resolution of the C-value paradox came the introduction of the G-value paradox. Genome size may vary across taxa in orders of magnitude, but gene content has remained quite stable amongst animals (ranging from ~10,000 genes to ~30,000 genes). How could organisms with such different developmental complexities have such similar gene content? There are a few hypotheses, some of which invoke important roles for alternative splicing, gene expression machinery, or non-protein-coding RNAs in explaining organismal complexity.

Even though we now understand that the C-value paradox was partially the result of differences in the sizes of intergenic and non-coding regions (rather than gene content) and, in a few cases, due to polyploidy, there are still interesting questions regarding the evolution of genome size and complexity. For example, how did eukaryotic genomes accumulate introns and other features that distinguish them from bacterial genomes? One solution, presented by Mike Lynch, is that eukaryotic species have smaller effective population sizes, which means that natural selection is less efficient at purging slightly deleterious mutations. If fancy gene structure is maladaptive, but not bad enough to induce a large fitness cost, then it will evolve in organisms with small populations (eukaryotes), but not in those with large populations (prokaryotes). And if we look within eukaryotes, we should see more complex genomes in the eukaryotes with historically small populations (ie, multicellular organisms) than in those with large populations (single celled critters).

Another interesting observation is that, amongst amniotes (mammals and reptiles), birds have the smallest genomes (birds are reptiles). Birds are also the extant group of amniotes with the highest proportion of flying species -- there are a few flying mammals (ie, bats), but you won't see any flying crocodiles, turtles, or snakes (although there are gliding snakes). This has led to much speculation regarding the relationship of avian flight and genome size. One hypothesis is that small genomes reduce metabolic costs, thereby allowing birds to expend lots of energy on flapping their wings . . . or whatever the hell it is birds do all day. To test this hypothesis, a group of researchers decided to investigate the genomes of the closest relatives of birds: dinosaurs. Yes, you read that correctly -- they did dinogenomics.

If small genomes evolved in birds for the purpose of flight, then we would expect that the ancestors of birds and dinosaurs (and their other descendents) would have genome sizes comparable with those found in other amniotes (ie, larger than bird genomes). We can't very well collect dinosaur genomes and directly quantify their sizes (sorry kids, Jurassic Park is a fairy tale). But we can measure some features of dinosaurs that have been preserved in their bones. Chris Organ, along with colleagues in Scott Edwards' lab at Harvard and Mark Pagel's lab at the University of Reading, did just that.

Organ and colleagues took the observation that genome size and cell size are positively correlated -- organisms with large genomes have large cells, and those with small genomes have small cells -- and attempted to extend that to extinct taxa. They measured bone cell size in fossils from 31 dinosaur species and found that they fell into two main groups. Interestingly, each of the groups corresponds to one of the two groups of dinosaurs. The theropods (the dinosaur radiation containing the birds, T. rex, and other similar beasts) have smaller genomes than the ornithischians (which includes triceratops and other similar herbivores).

The researchers also found that bone cell size predicts genome size quite well for extant vertebrates. They then extended this analysis out to the extinct dinosaurs and found that the small genomes observed in birds were probably shared by almost all other theropods. The ornithischians, on the other hand, had, on average, larger genomes, although not as large as extant mammalian genomes. The following tree summarizes Organ et al's findings:

Click to embiggen.

The predominantly pink region of the tree (representing genome sizes in the small range) contains the theropods. If selection for small genomes due to metabolic restrictions of flight had occurred only in birds, we would expect birds to have smaller genomes than their closest relatives. But small genome size is a trait shared by all theropods, and one that distinguishes them from ornithischians and other amniotes. What caused this decrease in genome size? The researchers think that it's due to a decrease in repetitive elements (sort-of viral DNA) within theropods right after they split with ornithischians. It may be that the theropods found a way to wrangle up some of that junk DNA, decreasing the size of their genomes.

Furthermore, the evolution of flight cannot be used to explain the small genome size in birds. There had to be something else (other than flight) selecting for genome size in the common ancestor of theropods, after they split with other dinosaurs. I'll let the authors explain, in their own words, what they think is going on:

[B]ecause genome size in part affects cell size, it has direct consequences for the rate of cell division, transcriptional processes, and cellular respiration. Consequently, it is thought that physiological demands may have constrained the evolution of genome size in endothermic vertebrates by favouring smaller red blood cells that increase surface area to volume ratios, and therefore their ability to facilitate gas exchange (a constraint that mammals may have circumvented with enucleate red blood cells). Our results suggest that this component of endothermy in living birds may have originated early in the saurischian/theropod lineage with commensurate changes in genome size, a conjecture consistent with studies of dinosaurian growth physiology using bone palaeohistology. The later secondary expansion of genome size in flightless birds suggests that, even though flight and genome size may not have arisen together, the two may be functionally related, perhaps at a physiological level.
[References omitted from the original.]

This is still a metabolic explanation for reduced genome size, although it's unrelated to flight. Instead, the authors argue that gas exchange across the membranes of red blood cells is facilitated by decreasing the size of the cells (achieved through decreasing genome size). This allowed the common ancestor of theropods to maintain a constant body temperature (they were endothermic in this model). In other words, T. rex was warm blooded. Next think you know, they're gonna try and tell us that Velociraptor had feathers.

This is a story about dinosaurs, so a lot of internet folks have chimed in. Larry Moran writes:

The result indicates that small genome size in birds is not an adaptation for flight. Perhaps it is not an adaptation at all but simply an accident due to the fact that the ancestor of sauropods just happened to have a reduced genome.

It is possible that the small genome size in birds is the result of being descended from organisms with small genomes (ie, the trait is not currently under selection in birds), but what caused the initial decrease in genome size? And why were birds and other theropods able to maintain a small genome size while other amniotes could not decrease their genome size? Shouldn't the avian genomes have been invaded by transposable elements by now? There are two things at issue here: why the genome decreased in size and why it stayed small. Larry's quibble does not address the first and only partially answers the second.

I'll close with a quote from Carl Zimmer's excellent coverage of this article for Science. But, before I do, allow me to share an anecdote. I first heard about this research at the Evolution meeting last summer. The tiny room was packed for Organ's presentation, and I was forced to sit on the floor. After the talk, most of the audience cleared out, but I wanted to see the next presentation. I found an empty seat next to Scott Edwards (the senior author on the study) and told him that it was some pretty cool stuff. He looked at me, and said, "Yeah!" as he let out a proud chuckle.

Anyway, here is the quote from Zimmer's article:

"It's a cute paper, but I'm not terribly confident in the outcome," says Michael Lynch of Indiana University, Bloomington, who questions whether natural selection is responsible for driving genomes to different sizes to fine-tune metabolism. "There's a correlation of the two, but I don't know of any direct demonstration of causality."

Yes, that's the same Mike Lynch who argues in favor of relaxed selective constraint being responsible for increased genome size. What we're seeing here is the beginning of a neo-neutralist movement in genome evolution that critiques every positive selection story as a spandrel. I think I know which side Larry will take.

Lynch M. 2006. The origins of eukaryotic gene structure. Mol Biol Evol 23: 450-468. doi: 10.1093/molbev/msj050

Organ CL, Shedlock AM, Meade A, Pagel M and Edwards SV. 2007 Origin of avian genome size and structure in non-avian dinosaurs. Nature 446: 180-184. doi: 10.1038/nature05621

Taft RJ, Pheasant M and Mattick JS. 2007. The relationship between non-protein-coding DNA and eukaryotic complexity. BioEssays 29: 288-299. doi: 10.1002/bies.20544