Periodic Reporting for period 1 - FLIGHTLOSS (The Genomic Underpinnings of Convergent Evolution: Repeated Loss of Flight in Island Rails, the Greatest Avian Colonizers)
Período documentado: 2020-10-06 hasta 2022-10-05
What is the first feature that comes to mind when you think of a bird? Probably that it can fly! However, this distinguishing ability is not shared among all 11,000 bird species, as there are well-known examples of flightless bird, such as ostriches, kiwis, or penguins. But when the ancestral birds evolved from one of the dinosaur lineages, they could indeed fly, which means that ostriches, kiwis, and penguins have lost this ability over the course of evolution.
Less well-known than ostriches and kiwis is the family of birds that comprises most flightless species today, namely the rails (Rallidae). These birds are often secretive and many live their lives in so dense vegetation that they are seldomly spotted. Among the volant species, some species are sedentary while others migrate, but migration happens during the night and therefore generally escapes our attention. Nevertheless, rails are the most incredible dispersers! They have a strong tendency to fly out across the ocean, because most islands around the globe have been colonized by rails at one point or another. If the island is isolated, over time the island rail population will diverge from its mainland ancestor and become a separate species, edemic to the island or archipelago. If the islands are not home to predatory mammals, the island rails lose their flight ability over time, and hundreds or even over a thousand flightless rail species are estimated to have thrived on islands several hundred years ago. Then came waves of human settlement, and with them the introduction of mammals, which led to the rapid extinction of flightless island birds.
But how can flight loss evolve repeatedly in the same group of birds? We know that it is not because it first evolves on one island and is then spread to another, because once the rails lose flight, they also lose their dispersal capability. Instead, when arriving to a predator-free island, the rails face similar conditions. Flying requires large pectoral muscles and is metabolically very costly, so if you do not need to, instead allocating energy into sturdier legs is beneficial. Thus, the evolution towards flight loss if predictable and convergent. But what does convergent morphological evolution look like at a molecular level? Are the same changes happening independently in the genomes of different island lineages evolving flight loss? Or do a myriad of different genomic changes lead to the same result? Whether evolution is predictable at a molecular level is still a fundamental question in evolutionary biology. It also specifically offers understanding of the unique evolution of island biota, and may be relevant in consideration of conservation measures.
This action exploited the natural experiment that rails offer. Focusing on the Laterallus, Zapornia, and Hypotaenidia rails, replicates of insular flightless species and their volant sister species on the mainland were studied. The most recent common ancestors of some of those species pairs were also sisters species.
Less well-known than ostriches and kiwis is the family of birds that comprises most flightless species today, namely the rails (Rallidae). These birds are often secretive and many live their lives in so dense vegetation that they are seldomly spotted. Among the volant species, some species are sedentary while others migrate, but migration happens during the night and therefore generally escapes our attention. Nevertheless, rails are the most incredible dispersers! They have a strong tendency to fly out across the ocean, because most islands around the globe have been colonized by rails at one point or another. If the island is isolated, over time the island rail population will diverge from its mainland ancestor and become a separate species, edemic to the island or archipelago. If the islands are not home to predatory mammals, the island rails lose their flight ability over time, and hundreds or even over a thousand flightless rail species are estimated to have thrived on islands several hundred years ago. Then came waves of human settlement, and with them the introduction of mammals, which led to the rapid extinction of flightless island birds.
But how can flight loss evolve repeatedly in the same group of birds? We know that it is not because it first evolves on one island and is then spread to another, because once the rails lose flight, they also lose their dispersal capability. Instead, when arriving to a predator-free island, the rails face similar conditions. Flying requires large pectoral muscles and is metabolically very costly, so if you do not need to, instead allocating energy into sturdier legs is beneficial. Thus, the evolution towards flight loss if predictable and convergent. But what does convergent morphological evolution look like at a molecular level? Are the same changes happening independently in the genomes of different island lineages evolving flight loss? Or do a myriad of different genomic changes lead to the same result? Whether evolution is predictable at a molecular level is still a fundamental question in evolutionary biology. It also specifically offers understanding of the unique evolution of island biota, and may be relevant in consideration of conservation measures.
This action exploited the natural experiment that rails offer. Focusing on the Laterallus, Zapornia, and Hypotaenidia rails, replicates of insular flightless species and their volant sister species on the mainland were studied. The most recent common ancestors of some of those species pairs were also sisters species.
Through a combination of previous field sampling that has yielded tissue samples deposited in museum collections, new field work in South America (carried out by local collaborators, as participation by the MSCA fellow was prevented by pandemic-related travel bans), and sampling of historic specimens at the Natural History Museum and several other natural history museums, both living and extinct species could be accessed. Large efforts were spent to acquire tissue samples from ornithological museum collections and multiple loans were obtained from Kansas University Biodiversity Institute and Natural History Museum, US; University of Washington Burke Museum, US; American Museum of Natural History, US; Museums Victoria, Australia; and the Smithsonian Institution, US.
The application of museomics allowed the extraction of recent and historic DNA and the construction of massively parallel sequencing libraries. Using linked short-read technologies, annotated high-quality reference genome assemblies could be produced, now available for thirteen species. From eleven more degraded samples, such as century-old toepad tissue from extinct species, short reads could be sequenced at intermediated depth, and those reads can be mapped onto the reference genome assemblies. Finally, non-focal species—i.e. those that are not among species pairs or groups comprising flightless island rails, but still provide perspective to the evolutionary history of the clade—were sequenced for short reads at lower depth. Given the enrichment of mitochondria in toepad tissue, this has allowed the assembly of full mitochondrial genomes at great depth, available for six species and ten subspecies of the great colonizer Buff-banded Rail (Hypotaenidia philippensis).
Sourcing tissue samples from all over the world during a global pandemic turned out challenging and imposed considerable logistic delays. It was therefore only towards the end of the action that the final set of genome data was finally available, and this massive dataset will therefore be analyzed following the end of the action. While the project has been featured at talks and in social media, the main dissemination of scientific results through peer-reviewed papers is yet to come.
The application of museomics allowed the extraction of recent and historic DNA and the construction of massively parallel sequencing libraries. Using linked short-read technologies, annotated high-quality reference genome assemblies could be produced, now available for thirteen species. From eleven more degraded samples, such as century-old toepad tissue from extinct species, short reads could be sequenced at intermediated depth, and those reads can be mapped onto the reference genome assemblies. Finally, non-focal species—i.e. those that are not among species pairs or groups comprising flightless island rails, but still provide perspective to the evolutionary history of the clade—were sequenced for short reads at lower depth. Given the enrichment of mitochondria in toepad tissue, this has allowed the assembly of full mitochondrial genomes at great depth, available for six species and ten subspecies of the great colonizer Buff-banded Rail (Hypotaenidia philippensis).
Sourcing tissue samples from all over the world during a global pandemic turned out challenging and imposed considerable logistic delays. It was therefore only towards the end of the action that the final set of genome data was finally available, and this massive dataset will therefore be analyzed following the end of the action. While the project has been featured at talks and in social media, the main dissemination of scientific results through peer-reviewed papers is yet to come.
Despite advances in genomics, less than a handful studies have addressed the loss of flight in birds, and all of them have limitations: (1) A study of the Galápagos cormorant found changes in regulation and protein sequence of genes affecting ciliary function, but compared a single flightless to three volant species, which restricts the power to detect conclusive patterns. (2) A study of ratites found independent losses of flight and convergent changes in gene regulation, but this tracks ancient changes dating up to 85 million years ago, posing challenges to detect genetic changes that may have subsequently been obscured by mutational turnover. A study of steamer-ducks found changes in a gene region associated with ciliary function, but this study system turned out to be a small recent radiation in which all flightless species originated in a single event of flight loss.
Importantly, none of the previous studies was able to explicitly compare independent events of flight loss in closely related lineages, which would enable more generalizable conclusions about convergent evolution. Surprisingly, rails have not been studied in this context, despite the unique suitability of the family, that comprises no less than 19 extant island-dwelling species with reduced or lost flight ability, corresponding to 16% of the total diversity of the family. During previous work, the fellow has identified recently evolved flightless/volant species sister pairs—some as young as <350,000 years—within the genera Laterallus, Hypotaenidia, and Zapornia.
While no conclusion is available yet, due to the delay of the action, it will move the state of the art, expanding analyses of parallel independent flight loss in recent evolutionary time, based on tens of sister species pair contrasts. It will offer an understanding of whether the loss of flight has been determined by mutations that change critical protein sequences or/and changes in the regulation of gene expression, e.g. through loss of regulatory elements. In addition to explicitly addressing the evolution of flight loss, the results are pertinent to the classical question of determinism versus historical contingency in evolution.
Importantly, none of the previous studies was able to explicitly compare independent events of flight loss in closely related lineages, which would enable more generalizable conclusions about convergent evolution. Surprisingly, rails have not been studied in this context, despite the unique suitability of the family, that comprises no less than 19 extant island-dwelling species with reduced or lost flight ability, corresponding to 16% of the total diversity of the family. During previous work, the fellow has identified recently evolved flightless/volant species sister pairs—some as young as <350,000 years—within the genera Laterallus, Hypotaenidia, and Zapornia.
While no conclusion is available yet, due to the delay of the action, it will move the state of the art, expanding analyses of parallel independent flight loss in recent evolutionary time, based on tens of sister species pair contrasts. It will offer an understanding of whether the loss of flight has been determined by mutations that change critical protein sequences or/and changes in the regulation of gene expression, e.g. through loss of regulatory elements. In addition to explicitly addressing the evolution of flight loss, the results are pertinent to the classical question of determinism versus historical contingency in evolution.