By Allison Shultz
In December of 2014, our knowledge of bird genomics made a huge leap forward – moving from about 8 published bird genomes to more than 48. This flock of bird genomes was accompanied by 28 publications looking at everything from genome characteristics of individual species to examining how characteristics of the genome have evolved in all birds.
One of the first steps, building the bird tree of life, is described in the video below. This tree, or phylogeny, is an essential tool and first step to then understand bird evolution. While these findings support many of the relationships found by Hackett et al. (2008), such as parrots and songbirds being each other’s closest relatives, they also discovered some new relationships, such as pigeons and flamingoes being sister clades. Surprisingly, even with entire genomes, some of the relationships are difficult to figure out because of something called incomplete lineage sorting, which happened because there was a rapid radiation of many major types of birds about 66 million years ago.
With this framework in place, researchers were able to look at the evolution not only of the structure of the genome itself, but of the gene sequences. Birds have much smaller genomes than other amniotes. The authors show that this is largely because repeat elements of the genomes like SINEs are 10 – 27x less than in other reptiles, genes are closer together, and protein coding genes have much shorter introns. This is possibly an adaptation to powered flight, as bats also show many of these characteristics. Why would a small genome go with flying? It may have something to do with high metabolic requirements. Birds also have some regions of the genome missing compared to other amniotes and more microchromosomes, possibly due to these large losses of genome regions.
Having genome sequences from almost all orders of birds also allows researchers to study genes associated with particular complex traits. One such trait is vocal learning (the ability to learn songs), which likely evolved two or three times – once in hummingbirds, and either once in the ancestor of parrots and passerines (songbirds), or once in each of those two groups. They not only identified faster than normal evolution in vocal learners for 227 genes, but found that many of the genes under selection were involved in neural connectivity, brain development and neural metabolism. They also found 822 regions of the genome that showed patterns of convergent accelerated evolution, and may represent some of the regions of the genome that control gene expression – many of these are actually near genes expressed in regions of the brain involved in vocal learning.
Researchers also studied the question of whether the loss of teeth (edentulism) occurred once in birds or multiple times, as had recently been proposed due to the discovery of several fossil birds possessing teeth. They discovered deletions in protein-coding exons, and the resulting inactivation of several genes important in enamel and dentin formation that were shared by all bird genomes they examined. Based on this shared history of these deletions and subsequent shift in mutation rates at the base of birds (because the genes were no longer functional and therefore were no longer under stabilizing selection) they propose that tooth loss and beak development evolved at the same time in the ancestor of all modern birds.
Feathers are another characteristic unique to birds, and researchers investigated the evolution of alpha and beta-keratins, both of which are coded by multigene families and are essential in the development of not only feathers, but also claws, scales, and keratinocytes (skin cells). They found that for alpha-keratins, which are found in all vertebrates, the number of genes is lower in birds than in mammals and non-avian reptiles (e.g. lizards and snakes), except for two genes that have expanded. However, when the researchers looked at beta-keratins, which are unique to reptiles (including birds), they found that the number of genes has expanded in birds compared to other non-avian reptiles. Furthermore, while the number of keratinocyte genes remain fairly constant in all species, genes associated with feathers and claws were associated with different lifestyles, like aquatic or predatory lifestyles.
The results discussed here have barely scratched the surface of what was found by analyzing the 48 bird genomes, but for anyone looking to learn more about these studies, all of the publications are available here: http://avian.genomics.cn/en/index.html. As these and even more genomes become available, we will continue to learn about what makes a bird a bird and how the amazing diversity of this lineage of animals is possible.
Allison Shultz is a PhD student in the Department of Organismic and Evolutionary Biology at Harvard University.