New publication! The genetics of convergent structural colour evolution

In a paper published in Philosophical Transactions of the Royal Society, B, we demonstrate the genetic basis of structural colour evolution in the co-mimic butterflies, Heliconius erato and Heliconius melpomene. These species both show variation in colour from black to blue, due to differences in nanostructures on their scale wings, across a geographical cline in South to Central America. This brings together the work of three former and current PhD students in the group.

We performed crosses between blue and black subspecies of both species and used these to identify regions of the genome responsible for colour variation. Firstly, Melanie Brien used photographs to quantify colour variation between individuals. Juan Enciso-Romero, then used x-ray scattering data from the European Synchrotron Radiation Facility to quantify variation in the scale structures.

(A) Crosses between iridescent and non-iridescent morphs of Heliconius melpomene and Heliconius erato. In the F2 generation, there is continuous variation in blue colour. (B) Schematic of part of a scale showing the structures. We used x-ray scattering data to quantify variation in the spacing of the ridges, which has an effect on the brightness of the reflected colour.
(A) Crosses between iridescent and non-iridescent morphs of Heliconius melpomene and Heliconius erato. In the F2 generation, there is continuous variation in blue colour. (B) Schematic of part of a scale showing the structures. We used x-ray scattering data to quantify variation in the spacing of the ridges, which has an effect on the brightness of the reflected colour.

In Heliconius erato we largely find a single genomic region that controls variation in both blue colour and scale structure. This is on the sex-determining Z chromosome (males have two copies of this chromosome while females just have one)

There are peaks that cross the significance threshold on chr 20 and z in A, and on Z in B and C. The line does not cross the threshold in D
H. erato analysis with lines showing the strength of the association between genotype and phenotype. (A) is for a measure of blue colour, (B) is for overall brightness, (C) is for ridge spacing and (D) for cross-rib spacing. Dotted line shows p=0.05 significance level.

In Heliconius melpomene, by contrast, we find that quantified colour differences are associated with variation on chromosome 3, while variation in scale structure is associated with chromosome 7. These results suggest that the two species have convergently evolved similar colour using different genetic mechanisms.

There are peaks that cross the significance threshold on chr 3 in A, and  B and on chr 7 in C. The line does not cross the threshold in D
H. melpomene analysis with lines showing the strength of the association between genotype and phenotype. (A) is for a measure of blue colour, (B) is for overall brightness, (C) is for ridge spacing and (D) for cross-rib spacing. Dotted line shows p=0.05 significance level.

To try to find which genes in these genomic regions were controlling structural colour, Victoria Lloyd then compared the level of expression of genes within this mapped genomic regions. Using developing wing tissue from blue and black butterflies, she looked for genes that were expressed at a higher level in the blue wings. Interesting candidates include genes that are part of or interact with the cell cytoskeleton, which might have a role in shaping the structures.

This work is a first step in understanding how biological systems make and control structural colours. Something that humans have been trying to replicate for decades, with limited success.