As early as 1879, naturalist Fritz Müller noted that many of the Heliconius butterflies he found in the Amazon shared the exact same blazing black, red, and white wing color patterns, although they were different species. He reasoned that the butterflies had come to resemble each other’s striking coloration—indicating to birds that they were toxic and not to be eaten—aiding the species’ survival because the more individuals with these colorations, the faster predators learn to avoid them, an idea that became enshrined in textbooks as “Müllerian mimicry.”
How the butterflies evolved to resemble one another has long been a mystery. One pressing question for evolutionary biologists is whether pairs of lookalike butterfly species took the same paths to arrive at the same color pattern, using the same genetic and developmental machinery every time, or did they effectively reinvent the wheel, coming up with new mechanisms from scratch?
A new study appearing today (November 14) in Current Biology takes a stab at answering that question. The research demonstrates that while lookalike Heliconius species employ the same gene to guide wing coloration, it is likely under different epigenetic control—ultimately suggesting there’s more than one way to arrive at the same coloration. The findings complement those of another study in PNAS this week that takes a deep dive into the regulatory networks that underlie color patterning in the Heliconius genus, illustrating how the insects use relatively few genes to generate such striking diversity.
“It’s cool stuff. . . . You can do the same thing in a number of different ways,” remarks James Mallet, an evolutionary biologist at Harvard University. “There are maybe quite a variety of different ways that you can twiddle the knobs to get the same common pattern.”
The international collaboration of researchers conducting the Current Biology study decided to focus their investigations on WntA, one of four main genes known to play important roles in wing color patterns across the 40-species-strong Heliconius genus. Early in wing development, caterpillars produce WntA, a protein that spreads across the wing, producing a concentration gradient that instructs other proteins and genes to create patterns of particular colors on the wing. WntA is known to be important for generating the characteristic stripes in many Heliconius species.
Using CRISPR-Cas9, the team knocked out the WntA gene in three pairs of distantly related, but seemingly identical species, to investigate the effect on coloration. The researchers’ rationale was that if WntA played the exact same role in wing patterning in lookalike butterflies, then the knockout insects should also end up looking the same, explains Riccardo Papa, an evolutionary biologist at the University of Puerto Rico and a coauthor on the study.
That didn’t turn out to be the case. For instance, the distantly related Central American species H. hewitsoni and H. pachinus, are both black and bear two characteristic thin white stripes on the forewing and one on the hindwing. Once WntA was knocked out, H. hewitsoni’s double stripe had merged into one large white patch on the lower forewing, while the H. pachinus mutant carried two large splotches at the base and at the tip of the wing.
The team observed similar differences in the boundaries of wing coloration between once-lookalike pairs H. erato and H. melpomene, and between H. sapho sapho and H. cydno chioneus: patches of black would switch to red or yellow. Despite no difference in the sequence of WntA itself across the Heliconius genus, knocking it out had variable effects on lookalike species. WntA must therefore be acting differently in lookalike species to arrive at the same appearance.
To Papa, one likely explanation for these differences is that WntA may be subject to different genetic regulation. Consistent with that idea, the team also observed differences between wildtype individuals of lookalike species in WntA RNA expression patterns across the wing. In other words, the butterflies don’t entirely reinvent the wheel to arrive at the same color pattern—they all use the same gene but they use it differently. “As genomes diverge [in species evolution], changes can happen in the way that these genes are being utilized,” says Papa. “Morphologies that are pretty much identical can be achieved by using genes in different ways.”
Krzysztof Kozak, an evolutionary biologist at the Smithsonian Tropical Research Institute whose colleague participated in the research but he himself was not involved, agrees with Papa’s interpretation. He adds that it’s somewhat surprising for the lookalike butterflies to have evolved different mechanisms for producing wing color patterns in a relatively short space of time; some species only diverged 12 million years ago.
However, he’d like to see more work on the mechanisms involved. The study “doesn’t really tell us much about what other genes are there and how do these come into play, how do they actually form these different [regulatory] networks?” he asks.
This question was the focus of the PNAS study published earlier this week, in which an overlapping group of scientists took a close look at the gene optix, which is responsible for producing red color patterns in Heliconius butterflies. Through several specialized assays, the researchers identified five regulatory elements that enhance the expression of optix, and knocking them out sequentially revealed a complex genetic architecture, with enhancers interacting with one another and controlling multiple aspects of red color patterns. “The complexity is not really in the number of genes, which are very few, but actually in the evolution of those regulatory elements that determine [the diversity across] the beautiful radiation of butterflies,” notes Papa, a coauthor on this study as well.
Other phenotypes in nature often appear to be determined by very large numbers of genes scattered around the genome, “yet in these Heliconius butterflies, we have a system where only a few genes seem to be very key in switching color patterns,” thanks to complex regulatory architecture, Mallet notes.
One possible reason why Heliconius species use a relatively small number of genes to control color patterns is that they interbreed a lot with one another. In Science this month, Mallet, along with Papa and other colleagues, illustrated through an assembly of 20 Heliconius genomes frequent interbreeding within the major clades of the genus. Recurring bursts of recombination with other species would easily break up large numbers of genes that are sprawled across the genome, whereas that would be less likely to happen if genes controlling color variation are restricted to a few loci, Mallet explains. “If introgression is a major feature of their evolution, then that will mess up color patterns a lot unless you’ve got relatively few loci,” he says.
“All these papers together form a very great example of how complex [a] picture we can now develop of basically any form of life you want—as long as you can cultivate it and get its DNA, you can develop incredibly sophisticated studies,” notes Kozak. In the future, he says, he hopes this ability will help researchers answer questions about the predictability of evolution more broadly. “I think in the next five, ten years, we’ll just see a proliferation of studies like this in all forms of life. And only then can we start generalizing about predictability of evolution, the importance of gene regulatory networks, how they are structured, and so on.”
C. Concha et al., “Interplay between developmental flexibility and determinism in the evolution of mimetic Heliconius wing patterns,” Current Biology, doi:10.1016/j.cub.2019.10.010, 2019.
N.B. Edelman et al., “Genomic architecture and introgression shape a butterfly radiation,” Science, doi:10.1126/science.aaw2090, 2019.
Katarina Zimmer is a New York–based freelance journalist. Find her on Twitter @katarinazimmer.