Genome sequencing of butterflies resolves centuries-old conundrum

When conditions are just right, organisms can undergo rapid bursts of diversification, and what starts out as one species can end up as an entire family tree in the evolutionary equivalent of the blink of an eye.

A new study published in the journal Proceedings of the National Academy of Sciences shows this is what recently happened to a group of tropical South American butterflies called glasswings, which partially explains why they all tend to look alike and why scientists have historically had such a hard time telling them apart.

“These butterflies have puzzled and exasperated lepidopterists, taxonomists and museum curators for centuries, including me personally for the last three decades,” said study co-author Keith Willmott, a curator at the Florida Museum’s McGuire Center for Lepidoptera and Biodiversity. “Apparently these species have evolved very recently and hybridized frequently.”

Rapid diversification events can be a headache for the scientists who study them. With so much happening all at once, it can be difficult or impossible for scientists to look back and figure out how everything was connected and which order events took place in.

In the case of glasswing butterflies, dozens of new species evolved within just the span of 1-2 million years, coinciding with drastic changes in climate that occurred during the Pleistocene ice ages. There were as many as 17 periods of intense cold and glaciation during the Pleistocene interspersed with prolonged stretches of balmy weather.

These chaotic shifts in climate resulted in the diversification of several different groups. Florida scrub mints, for example, were separated and later reunited several times as the state alternately flooded when sea levels rose and resurfaced when they sank. Scrub mints were isolated on islands when the waters were high, during which time many evolved into new species. When the waters were low, scrub mints expanded their distributions until they overlapped, resulting in widespread hybridization.

The same pattern seems to have played out in glasswing butterflies, only instead of being separated by an ocean, they may have become isolated in patches of forest or on opposite sides of the Andes as changes in rainfall and temperature disrupted their habitats. In other cases, closely related species with similar distributions show a preference for different elevations that now keep them separate.

The authors discovered this by sequencing the genomes of almost all species of two particularly fast radiations of glasswing butterflies to remap their evolutionary trees. Of those species, 10 were sequenced to the gold standard of “reference quality” genomes that are freely available to the research community.

By genetically mapping these butterflies, the team found that six geographic populations were more genetically distinct than previously thought, leading to them being recognized as new species. Also, understanding the species from a genomic perspective enables experts to highlight any visual differences that could be used to identify and track the different species, now that they are confirmed as genetically distinct.

“With this new genetically informed evolutionary tree, and multiple new reference genomes, we hope that it will be possible to advance biodiversity and conservation research around the world and help protect the butterflies and other insects that are crucial to many of Earth’s ecosystems,” said first author Eva van der Heijden, a doctoral student at the University of Cambridge.

Glasswings also played fast and loose with the number and arrangement of their chromosomes. These variations likely fanned the flames of diversification, since any two butterflies with different chromosome configurations would not have been able to produce viable offspring. These sorts of chromosome rearrangements have the potential to produce new species more or less instantaneously.

Rapid radiations tend to produce species that all look very similar as well, which is one of the reasons scientists have had such a hard time classifying glasswings. Species in this group also form complex mimicry rings, which means they look even more similar than they otherwise would.

But the butterflies themselves have a surefire way of recognizing each other.

“Chemical communication is likely to be especially important in butterflies that mimic each other, like glasswings, where recognizing members of the same species based on wing patterns is presumably more challenging,” Willmott said.

Glasswing butterflies feed on certain flowers and withered plants that contain a type of bitter alkaloid that birds find distasteful. They take advantage of this by storing as much of it as they can in their bodies. Birds quickly learn from trial and error which butterflies to associate with the toxin and leave them alone. Different species can reduce their losses by all sharing the same warning color pattern, a natural phenomenon called Müllerian mimicry, so that fewer individuals are attacked as predators learn what to avoid – once a bird learns one species, it will avoid them all.

But the alkaloids can also be used to make highly aromatic compounds, and glasswings have evolved a specialized organ to diffuse the scent.

“A defining feature of these butterflies is the presence of a ‘hair-pencil’ on the male hindwing, which resembles a paint-brush and is composed of highly elongate wing scales that disperse scents,” Willmott said.

The authors conducted an analysis of the perfume from three closely related species and found that all of them had distinct chemical profiles. Moths and butterflies, in general, have a keen sense of smell. Indian moon moths, for example, can follow the pheromone trail of a female from several miles away. It’s likely that glasswings rely on their distinct chemical calling cards rather than trying to visually identify potential mates in a sea of similar-looking butterflies.

Additional co-authors of the study are: Karin Näsvall, Patricio Salazar-Carrión, Jonah Walker, Nicol Rueda-M, Dominic Absolon, Thomas Mathers, Camilla Santos, Shane McCarthy, Jonathan Wood, Joana Meier and Marianne Elias of the Wellcome Sanger Institute; Fernando Seixas of Harvard University; Carlos Eduardo Beserra Nobre and Artur Campos Maia of the Federal University of Pernambuco; Daiane Szczerbowski and Stefan Schulz of the Technische Universität Braunschweig; Ian Warren and Chris Jiggins of the University of Cambridge; Kimberly Gabriela Gavilanes Córdova, María José Sánchez-Carvajal, Franz Chandi, Alex Arias-Cruz and Caroline Bacquet of the Universidad Regional Amazónica Ikiam; Camilo Salazar of the Universidad del Rosario, Bogotá; Kanchon Dasmahapatra of the University of York; Stephen Montgomery of the University of Bristol; Melanie McClure of the Université de Guyane; Gerardo Lamas of the Universidad Nacional Mayor de San Marcos; and André Victor Lucci Freitas of the Universidade Estadual de Campinas.