What makes an eyespot? Cuticular proteins of course!

Close-up of an eyespot on a Papilio xuthus caterpillar [NOT the Eye of Sauron].
Photo from Futahashi et al. (2012) BMC Biology 10:46
My PhD research is generally focussed on the ecology and evolution of eyespots in caterpillars. That includes questions relating to how effectively these markings deter attacking birds?, but also marcoevolutionary questions like why did eyespots evolve in the species they did?

How are eyespots formed? however, is a different, yet equally important set of research questions which involes molecular techniques that I have little-to-no experience with. A research group out of Japan has recently identified a set of genes involved in producing markings on a caterpillar's cuticle, some of which are directly involved in the formation of eyespots. Their paper was published a paper in a prestigious open access journal called BMC Biology

Aside: I'm not talking about fingernails here! For insects, the cuticle refers to their exoskeleton - what caterpillar's have in place of "skin". The cuticle is made up of the epicuticle (thin, waxy, water-resistant outer layer containing no chitin), and a layer beneath it called the procuticle (chitinous and thicker). The procuticle itself is composed of the the exocuticle (rigid and hard) and the endocuticle (tough and flexible). In caterpillars the exocuticle is greatly reduced which gives them their soft-bodies.

Because the paper is "open access" it can be downloaded for free here:

Before this project had begun, the researchers had some idea of what parts of the genome code for proteins related to pigment colour in caterpillars, and that a switch in larval colour pattern from "bird-dropping" to "cryptic" within Papilio caterpillars was related to a decline in the amount of circulating juvenile hormone (JH) at the start of the 4th larval instar. Other than that, little was known about the genes related to specific colour patterns in caterpillars, despite a fair amount of similar work having already been done in adult butterflies. The researchers therefore set out to identify the genes associated with specific colour patterns expressed in butterfly caterpillars, patterns like the eyespots of Papilio caterpillars.

A generalized schematic showing the moulting process of a Papilio xuthus caterpillar from the 3rd to 5th instar. This image is from Figure 2 of Futahashi et al 2012.
One key result of this research was that some of the regulatory genes involved in caterpillar colour pattern were specific to the larval stage and differed from those genes associated with adult colour pattern, yet other genes participate in the formation of colour pattern in both larvae and adults. The researchers found at least two transcription factor genes that are associated with the Papilio caterpillar eyespots. Specifically they were able to show that the black part of the eyespot is related to a gene called "spalt", which has also been linked to black markings in butterfly wings. They also found that E75A and/or E75B regulate both the specific marking produced, and when (i.e., during what instar) that marking is produced. E75 is an ecdysteroid signal-related transcription factor and the ecdysteroid synthesis enzyme 3DE 3b-reductase are clearly associated with eyespot markings.
"What the heck is an ecdysteroid?" - said almost everyone. 
Aside: As insects grow they shed their skin. You might know this process as "moulting", but scientists have another term for it - ecdysis. Ecdysteroids are just hormones involved in the moulting process of insects. An "ecdysteroid synthesis enzyme" is biological molecule associated with manufacturing these hormones.

Two pieces of evidence allow the authors can be quite certain that the genes described above are involved in the coordinated genetic regulation of caterpillar colour pattern formation: i) the eyespot markings that these genes are associated with only occur only at certain instars, and 2) previous work has shown that these genes are related to the moulting process.

In addition to spalt and E75, three Papilio-specific genes were found: Px-0559, which was associated with black markings; Px-3233, associated with yellow markings; and Px-3244, associated with eyespot markings. These genes were expressed during the middle or late moulting period which coincides with the switch in colour formation from the "bird-dropping" colour pattern to the "green with eyespots" colour pattern. The relative amount of expression for each of these genes likely varies among species in such a way that it produces species-specific colour patterns.

One more thing - the researchers also looked at the specific areas of a caterpillar's cuticle where some of these genes specifically targeted to see if the structure of the exoskeleton in those areas might also be affected. They did this by looking at these areas with a high powered microscope called an electron microscope. They found that there were distinct structural differences in those areas of the caterpillar's cuticle that possessed distinct markings. In particular the eyespot region was easily recognized by the structure alone when examined using electron microscopy. Specifically, the black region of the eyespot was relatively fine, whereas yellowish green region around the eyespot was relatively course, and the red area was intermediate. Interestingly, the white stripe in the centre of the eyespot had "a very smooth surface", and apparently the authors have previously shown that there is a muscle attached to this white stripe region.
A close up of the eyespot region of a Papilio xuthus caterpillar using scanning electron microscopy. This image is from Figure 8C of Futahashi et al 2012.
Overall this piece of the research shows us that there is a tight association between colour pattern and surface structure. The function or adaptive value of this association is interesting to think about. Does a different texture affect how light reflects from these parts of the caterpillar's body? Could this help increase the salience of the eyespot to onlooking birds? Do these textural differences make the eyespot look more like a real eye? We know that some of the beautiful colours we see in butterflies and birds are actually created by specific surface structures, but might structure play additional, under-appreciated roles in crypsis or mimicry? I welcome any thoughts or comments you may have - you can post them in the comments section below.

*Minor corrections made Dec 17 2012


  1. This is fascinating work, and you included essential explanations & definitions without being condescending or boring. Not an easy task! Thanks for writing. Now I'm off to read the Futahashi et al. paper...