How Butterflies Pattern Top and Bottom Wing Surfaces Differently

By Viviane Callier

A butterfly wing is often described as a canvas, but it’s actually two: The patterns on dorsal and ventral surfaces are typically different because they have evolved different functions. The ventral side (underside, seen when wings are closed together) is often used for camouflage and predator avoidance, whereas the dorsal side (top side, seen when wings are open) is used for sexual signaling.

A gene called apterous A appears essential for the differentiation between dorsal and ventral wing surfaces, according to a new study in Proceedings of the Royal Society B. The gene, which is also present in beetles and flies, evolved this novel patterning function in butterflies.

The transcription factor apterous is expressed only on the dorsal wing surface in flies and is required for normal wing development: If you knock out the gene, the fly does not develop wings. Butterflies have duplicate copies of the gene, apterous A and apterous B.

“This double dose of diversity […] means that butterfly wings, when developing, must be able to tell which side is which early in development, to initiate distinct patterning processes and, ultimately, coloration mechanisms,” says Arnaud Martin, butterfly biologist at George Washington University, who was not involved in the work.

To investigate the role of each apterous copy, Anupama Prakash, a graduate student in the lab of Antónia Monteiro, Ph.D., at the National University of Singapore, used the CRISPR/Cas9 gene-editing method to knock them out. Prakash found that knocking out either copy of apterous in some individuals prevented wing development entirely, but, interestingly, knocking out apterous A in other individuals—presumably in smaller clones that still allowed wings to grow to a normal size—caused the ventral wing pattern in Bicyclus anynana butterflies to appear on the dorsal wing surface. “So, while both apterous copies have a conserved function in wing development, apterous A evolved a new function in pattern specification,” Prakash explains.

The researchers also looked at where apterous A and B were being expressed. Like in flies and beetles, apterous A and apterous B were expressed only on the dorsal surface of the wing. Interestingly, apterous expression was absent in the area of the wing that would become an eyespot center on the dorsal wing surface. Eyespots evolved approximately 90 million years ago on the ventral wing surfaces; eyespots on the dorsal wing surfaces, where apterous is expressed, evolved much later, only 50 million years ago. So, it appears as though apterous A was repressing eyespot development, and about 50 million years ago butterflies found a way to eliminate apterous in the specific locations where they eventually evolved eyespots.

“What is very interesting is that in mutants with extra eyespots on the dorsal surface, these mutants have extra spots of apterous A repression, suggesting that eliminating apterous is a required step in eyespot number evolution on the dorsal surface,” says Monteiro.

The researchers are now looking at how apterous A evolved its new function and how it acquired new downstream targets. “I am looking forward to seeing how apterous A, in particular, is plugged into developmental instructions that further make the two wing sides so different,” Martin says. “It’s elegant work in a terrific model system.”

Viviane Callier is trained as an insect physiologist and is now a freelance science writer in Washington, DC. Twitter: @vcallier. Email: vcallier@gmail.com