Stalk-Eyed Fly: Stretching the Imagination of Axon Elongation

Artwork by Matthew Ross

If you think the hammerhead shark has a weird head, you should check out the stalk-eyed fly. The stalk-eyed fly (Cyrtodiopsis whitei) is frequently showcased in nature documentaries, like the series Life, to show off the somewhat disconcerting development of its alien appearance. If this sounds unfamiliar to you, allow me to describe why this little insect has garnered so much attention. When the fly emerges from the pupa as an adult, it has an eerily translucent, roundish-head, with a bulbous maroon eye on either side. Soon after completing metamorphosis, the fly begins swallowing air and forcing bubbles through its cranium into each eye stalk. As the pressure of the air builds, both stalks expand until the fly looks like the offspring of a praying mantis and a hammerhead shark. Remarkably, this process is completed in a mere 15 minutes (Buschbeck & Hoy, 2005). After a short while in the sun, the exoskeleton hardens and the stalk length becomes permanent. The unusually long stalk length in this species is important for female mate choice (Hingle et al., 2001). The further apart the eyes, the luckier the guy.

This extension of the eyes away from the head creates a neurological problem. How does the fly maintain contact between the retina, the part of the eye containing light-capturing cells called photoreceptors, and the brain, which perceives the visual information collected by the photoreceptors? Neurons in the retina collect information from photoreceptors and send the information via a long projection, called an axon, to the brain. All these axons bundle together to form a rope-like nerve—the optic nerve. Given the rapid separation between brain and eye as the eye stalk expands, it would seem as though the optic nerve is put under a lot of stress. But the way the stalk-eyed fly solves this problem is quite curious.

Axons are viscoelastic (Tang-Schomer et al, 2010), meaning they behave somewhat like a viscous substance, somewhere in between a liquid and a solid. They can stretch under relatively minor forces in day-to-day activity (like when your fingers bend), but snap easily like a brittle twig when stretched suddenly and with greater force. This is kind of like the properties of a cornstarch and water mixture (Google "oobleck" later if you haven’t heard of it). With a small ball of “oobleck”, you could stretch it into a long band if you let it droop slowly; however if you pulled each end rapidly, the concoction would break in two. Could the stalk-eyed fly lengthen its optic nerves at a rate that doesn’t cause snapping to reach the full width of the stalk in 15 minutes?

In 2004, Pfister and colleagues estimated axons could grow a maximum of 8 millimeters per day, about the width of your pinky nail. With continuous pulling, the rate would come out to about 5.5 micrometers per minute. But in a study published by Magdesian and colleagues in January this year, we learned neurons can be pushed to a much faster rate of elongation. Axons of neurons grown in a dish were pulled with miniscule probes at a maximum speed of 100 micrometers per minute. Almost twenty times faster than the original estimates. In fifteen minutes at this speed, the stalk-eyed fly could only make a stalk that is 1.5 millimeters. With some flies making stalks that are 4-5 millimeters long, even a highly artificial lab setting couldn’t pull off the stalk-eyed fly’s amazing biology.

So, the stalk-eyed fly’s axons can’t be stretched. And, since the fly is able to interact with its environment soon after creating its shapely stalks, we can also rule out the possibility that neurons in the retina extend axons after the stalk has reached its final length. The final option is that the optic nerve develops prior to the stalks extending and extra nerve length generates "slack" in the cranium that uncoils when the stalk extends. This last option, amazingly, is exactly what the stalk-eyed fly does.

Example drawings of the stalk-eyed fly’s coiled optic nerve from Buschbeck and Hoy’s Figure 3

As a pupa, the axons from the retina grow to meet their targets and then begin extending. Since the pupa's head is confined within a consistent space throughout development, the optic nerve bunches up in the area of the compressed stalk. Imagine throwing a grappling hook attached to a coiled rope on the ground. As the hook sails, the rope rapidly uncoils and launches with the hook. Intriguingly, Buschbeck and Hoy found that the shape of the coiled nerve differs between flies, and even differs between stalks in the same fly. The remaining mystery is how the axons of the optic nerve elongate as a unit in the absence of tension, a factor thought by many experts to be required for axon elongation.

Here, I would like to define what I mean by "elongation." When an axon grows to its target recipient neurons in early development, the fan-shaped tip of the axon (called the growth cone), feels its way through a meshwork of other cells, following a breadcrumb trail of molecules to its target. As the growth cone pushes forward, the axon lengthens as a result. Once the axon reaches its target, subsequent growing of the organism requires that the axon "elongate" to compensate and still remain connected to the target. Hence, the axon does not quite "stretch" like a rubber band, but rather increases its length by adding more material to itself.

As an example, consider neurons that control the muscles in your toes reside in the spinal cord and therefore must have an axon that extends from the bottom of the spinal cord, through your leg, to your toe. These axons do not start growing toward your toes after your legs fully develop, though. Instead, they grow to meet their muscle targets very early in development, when your leg is just a lump of tissue early in gestation. From this point until you stop growing in adolescence, the axon elongates while the neuron's cell body remains in the spinal cord, and the end of the axon remains connected to your toe muscles. In this case, the axon remains taut throughout elongating, resulting in tension that seems to serve as the factor driving the axon to increase its length. So, how does the stalk-eyed fly get around this?

Buschbeck and Hoy hypothesized that proliferating glial cells, supporting cells of the nervous system, caused compression around the optic nerve, and stimulated elongation. While the authors found that glial cells indeed divide during elongation in this region, they never determined whether these dividing cells were sufficient to elongate the axons. However, later work by a team of researchers from Zhejiang University compared the viscoelastic properties, in this case the “softness,” of neurons and glia. The study determined that glia were softer and squishier than neurons (Lu et al, 2006). This probably indicates that a compression-related elongation of the nerve is unlikely, since the glia would compress themselves long before the axon would become compressed.

So what do we know now about the mechanisms of axon elongation? The most recent consensus is that a combination of chemical factors and mechanical forces interact to elicit the elongation of axons as body structures grow (Athamneh & Suter, 2015). A barrage of studies investigated how axons lengthened in response to mechanical force by observing how axons elongate in growing limbs of animals/insects. Some studies involve attaching neurons to two tiny platforms joined by their axons across a gap and moving the platforms apart (Pfister et al, 2004), like some apocalyptic chasm in the earth forming between two groups of people trying to hold on to each other by ropes. Tension is sufficient to elongate axons, and slackening after this artificial lengthening can cause the axons to retract back to tautness (Dennerll et al, 1989), creating a paradox for our humble stalk-eyed fly. Seemingly, the stalk-eyed fly’s ability to "overextend" its optic nerve without an obvious means of tension and to overcome inhibitory influence against slackness is an outlier in the scheme of typical axon elongation in other animals. As if it wasn’t weird enough already…


Athamneh A, Suter D (2015) Quantifying mechanical force in axonal growth and guidance. Front Cell Neurosci 9

Buschbeck E, Hoy R (2005) The development of a long, coiled, optic nerve in the stalk-eyed fly Cyrtodiopsis whitei. Cell Tissue Res 321:491–504.

Hingle A, Fowler K, Pomiankowski A (2001) Size-dependent mate preference in the stalk-eyed fly Cyrtodiopsis dalmanni. Anim Behav 61:589–595.

Lu Y-BB, Franze K, Seifert G, Steinhäuser C, Kirchhoff F, Wolburg H, Guck J, Janmey P, Wei E-QQ, Käs J, Reichenbach A (2006) Viscoelastic properties of individual glial cells and neurons in the CNS. Proc Natl Acad Sci USA 103:17759–17764.

Magdesian MH, Lopez-Ayon GM, Mori M, Boudreau D, Goulet-Hanssens A, Sanz R, Miyahara Y, Barrett CJ, Fournier AE, De Koninck Y, Grütter P (2016) Rapid Mechanically Controlled Rewiring of Neuronal Circuits. J Neurosci 36:979–987.

Pfister B, Iwata A, Meaney D, Smith D (2004) Extreme Stretch Growth of Integrated Axons. J Neurosci 24:7978–7983.

Tang-Schomer M, Patel A, Baas P, Smith D (2009) Mechanical breaking of microtubules in axons during dynamic stretch injury underlies delayed elasticity, microtubule disassembly, and axon degeneration. Faseb J Official Publ Fed Am Soc Exp Biology 24:1401–1410.

Written by: Cody Call

Cody is a neuroscientist at Johns Hopkins University School of Medicine. When not doing research you can find him sipping fine teas while playing a game of go.