The common mormon butterfly (right) mimicking the common rose (left). Credit: Krushnamegh Kunte (left) and Khew Sin Khoon (right)
The common mormon butterfly (right) mimicking the common rose (left). Credit: Krushnamegh Kunte (left) and Khew Sin Khoon (right)

The Supergene That Paints a Liar

The females of the common mormon butterfly are masters of disguise. Some of them look like the black-and-white males, but others embellish their wings with white brushstrokes and red curlicues to mimic distantly related swallowtail butterflies that are toxic. By copying them, the delectable female mormons fool predators into thinking that they are similarly distasteful.

The picture below shows how varied these disguises can be. The right halves are all different butterfly species, and the left halves are all common mormon females.

View Images
The left halves are all females of the common mormon butterfly (Papilio polytes). The right halves are either common mormon males (top) or different species of swallowtails (bottom). Adapted from Kunte et al, 2014.

British scientists Sir Cyril Clarke and Philip Sheppard studied these butterflies in the 1960s and, through cross-breeding experiments, showed that the insects never mix and match their patterns. You don’t get intermediate butterflies with red streaks from one pattern and white blotches from another. Instead, each pattern is inherited as one.

The duo reasoned that the patterns were controlled by a “supergene”—a cluster of genes that each control different parts of the wings, but are inherited as a single block. Imagine a row of switches that control the appliances in your house, but are taped together so that flipping one flips them all.

The supergene concept has been very influential and scientists have identified many such clusters in plants, snails, and other butterflies. But until now, no one had found the common mormon supergene. Krushnamegh Kunte from the Tata Institute of Fundamental Research in India finally did it and his results are a complete surprise—one that both confirms and refutes Clarke and Sheppard’s hypothesis.

Kunte, together with Marcus Kronforst at the University of Chicago, compared non-mimetic females that resemble males, or mimetic ones that look like the common rose swallowtail. They searched for parts of the butterfly’s genome that were linked to the mimetic patterns, and eventually homed in on a small region containing five genes. Four were similar in all the females, regardless of their patterns.

The fifth gene, known as doublesex, was another story.

Kunte and Kronforst found over 1,000 mutations separating the versions of the gene (alleles) in the mimetic and non-mimetic females. This astonishing variety was all the more surprising since doublesex has a reputation for consistency. It’s much the same in flies, beetles, ants, and every other insect group. And yet, in the common mormon—and only between the mimetic and non-mimetic females—this typically conservative gene was a hotbed of evolution.

The butterflies don’t switch their wing patterns by inheriting different versions of a cluster of genes, as Clarke and Sheppard suggested, but by inheriting different versions of doublesex. The supergene is not a collective, but a single gene. There isn’t a row of switches all taped together; there’s just one switch that controls everything.

View Images
The varying patterns of the common mormon butterfly (Papilio polytes). Credit: Wei Zhang

The doublesex discovery is doubly surprising because this gene already has a well-defined role: it sends developing butterflies down either a male or female path. It’s like finding that the light in your bedroom also starts your car.

Kunte now wonders if scientists have been typecasting doublesex as a sexual differentiation gene, when it could potentially do much more.  “Throughout the animal kingdom, you see tremendously different males and females,” he said. Maybe this particular gene family, involved in sex determination throughout the animal kingdom, is also involved in making deer antlers or peacock tails.”

It’s still not clear how doublesex alleles produce the patterns on the butterfly’s wings. With a thousand mutations to look at, the team understandably had some trouble matching these to specific wing elements. Still, they found some hints about how the gene does its thing.

When genes are activated, the instructions encoded in their DNA are first converted into a related molecule called RNA. These RNA transcripts are then used to build proteins. Kunte and Kronforst found that doublesex RNA is sliced and rejoined into four distinct ‘isoforms’, like different paths through the same choose-your-own-adventure book. One of these isoforms is found in males, and the other three are found in females.

You might think that each of the three female isoforms corresponds to a different pattern, and that’s what Kunte and Kronforst first suspected too. They were wrong. Every female has all three forms, regardless of their pattern. Instead, it’s the way these isoforms are used that matters. The mimetic females make more of them than the non-mimetic ones, especially in their wings and especially in parts that produce white markings.

So, the same gene gets processed in different ways and switched on at different strengths in different parts of the butterfly’s wing to produce a variety of patterns.

But that doesn’t explain why the patterns are so stable. Whenever a new generation is born, different versions of the same gene line up and shuffle their DNA, creating new combinations. In the common mormon, the doublesex mutations that produce one mimetic pattern should eventually shuffle with those that produce another, producing new blends. Clearly, that doesn’t happen.

Kunte and Kronforst found out why: the mimetic version of doublesex is inverted relative to the non-mimetic one, so it sits in a different orientation in the genome. This inversion stops the two versions from lining up properly and prevents them from shuffling, ensuring that the gene’s thousand-plus mutations are all inherited together.

And that is, more or less, what Clarke and Sheppard thought!

They envisioned many elements acting in concert to produce the right copycat wings, and passing to the next generation as a single block. They were right except for one detail: those elements don’t have to be individual genes. They can be different parts of the same gene. As Mark Scriber from Michigan State University told me, it’s amazing that Cyril Clarke basically got the right answer 60 years ago, without any powerful genetic tools that today’s scientists can use.

Matthieu Joron from the CNRS was also impressed. He has found supergenes in a different group of butterflies—the long-winged Heliconius species. These are more in line with Clarke and Sheppard’s ideas: clusters of individual genes, locked together by a genomic inversion. Both lineages of butterflies—the longwings and the mormons—have evolved mimicry in a similar way.

That’s perhaps surprising, since the two lineages use mimicry for very different reasons. The mormons are Batesian mimics—they deceive predators into thinking that they are toxic by resembling species that genuinely are. But the longwings are Mullerian mimics—they are all actually toxic and they reinforce their warning colours by resembling each other, so that a predators which learns to avoid one species also knows to avoid them all.

The Batesian mormons are liars that mooch off the warning signals of their better-defended peers. The Mullerian longwings are honest communicators that find safety in their shared warnings. And yet, both groups have evolved their lookalike patterns in much the same way.

Note: This is an expanded version of a news story in The Scientist.

Reference: Kunte, Zhang, Tenger-Trolander, Palmer, Martin, Reed, Mullen & Kronforst. 2014. doublesex is a mimicry supergene. Nature