The biologists had already suspected that optix played a role in activating the butterfly’s brown pigment. But they were surprised that the black pigment was turned on in the absence of optix.
A further surprise came when they turned off the optix gene in a second species, the buckeye. The butterfly’s usual browns and yellows disappeared, replaced by scales of a blazing iridescent blue. “That knocked our socks off,” Dr. Reed said.
A second group, led by Anyi Mazo-Vargas of Cornell University and Arnaud Martin of George Washington University, has explored the role of a gene called WntA, which plays a powerful role in the patterning of butterflies’ wings.
The standard pattern of nymphalid butterflies, a 90-million-year-old family of some 6,000 different species, consists of four bands, parallel to the body, that run between it and the edge of the wings. The second band, called the central symmetry system, contains the pattern in the middle of the wings, and the third band holds the eye spots.
Dr. Martin’s team found that when they delete the WntA gene with the Crispr technique, the central symmetry system band disappears entirely from the wings of the speckled wood and buckeye butterflies.
But in other species, the loss of WntA has very different effects, suggesting that the gene has been adapted many times to play different patterning roles as new butterfly species evolved.
In the monarch butterfly, for instance, loss of WntA affects an almost invisible white line that edges the distinctive black lines that delineate the wing’s veins. In the absence of WntA, the white lines expand into the areas between the veins, replacing the distinctive orange pigment.
The WntA gene becomes active in the caterpillar stage, impressing its patterning information on the embryonic wing structures. Dr. Martin sees the WntA gene as a sketching tool that defines the outline of the wing design, and the optix gene studied by Dr. Reed’s group as a “paintbrush” gene that fills in the color.
Dr. Reed hopes in time to understand the patterning mechanism so well that he will be able to recreate the pattern of one butterfly’s wings on those of a second species. But understanding butterfly wing patterning is just a step toward addressing larger questions in evolutionary biology.
One is the knotty question of how the string of information in a DNA molecule specifies the 3-D structures of the body. The butterfly wing presents this problem more tractably, in just two dimensions.
Another is that of how species evolve different forms. The work of these two groups shows that genes like WntA and optix — master genes that control the activity of other genes — can evolve very different roles in different species.
The nymphalid butterflies use WntA one way, the monarch for a quite another. The optix gene controls specialized wing scales in moths, but when butterflies evolved it somehow recruited a quite different set of servant genes, ones involved in generating pigment, not scale type.
“A big question in evolutionary biology is how do you rewire these gene networks,” Dr. Reed said.
Both Dr. Reed and Dr. Martin are enthralled by the ease and power of the Crispr gene-editing tool, invented in 2012. Before, they could infer what a gene might do but couldn’t prove it.
With Crispr, they can manipulate any gene of choice inside an egg and see exactly what it does.
“These are experiments we could only have dreamed of years ago,” Dr. Reed said. “The most challenging task in my career has become an undergraduate project overnight.”