A cracking discovery

Michel Milinkovitch is an evolutionary biologist who thinks outside the box. His exciting career recently led to the discovery that crocodile head scales form by physical cracking of the skin.

Not many people can say they’ve had close encounters with a 4-metre long crocodile, a near-extinct Galápagos tortoise and… a yeti. For Michel Milinkovitch, an evolutionary biologist at the University of Geneva, this is just part of the job. A typical day at work could be spent in his lab working with computer scientists, or it could be spent at the zoo studying crocodiles up close, very close. “Once we were working in the croc pit and a male charged,” he says laughing “I just shouted RUN and everyone just dropped everything and ran away. This was a 4-metre animal probably weighing half a ton, it’s like a bus passing in front of you”.

Juvenile Nile crocodile

Milinkovitch’s work on crocodile scale patterning recently made the cover of Science. His team found that crocodile head scales form by physical cracking of the skin, rather than being genetically defined like the scales in the body. In fact, this is the first time this cracking phenomenon has ever been observed in a biological process. “Cracking is something known for physicists but totally unknown for biologists.”

Not scales after all

The dogma was that all keratinised appendages like mammalian hairs and feathers and reptile scales, grow from developmental genetic units. But during a visit to a crocodile farm in France he noticed that, unlike the scales of snakes and lizards, the scale patterns on each side of the crocodile’s head did not mirror each other. “I was very surprised by this topology, which was not very organised. […] It didn’t fit with what we knew about the development of hair, feathers and scales.”

Milinkovitch and his team quantified this by making a high-resolution 3D reconstruction of the heads of 15 young Nile crocodiles, and then performed a mathematical analysis on the scales’ apparent chaotic pattern. He baffled at the results: the data didn’t fit any models used to describe biological processes, but were a perfect match with the statistical signatures of cracking phenomena, like when dried mud cracks.

To get to the bottom of this, the team looked in eggs during embryonic development. They found that the genetic units that give rise to the crocodile body scales never appear in the head. Instead, as the skin became thicker, grooves gradually appeared, grew deeper, longer and interconnected. “How can you grow skin and in the meanwhile make it stiff?” Milinkovitch proposes that the fast growth of the crocodile’s skull causes the thick keratinised skin to crack. When they had a closer look at the embryos, the team found that, where the grooves appear, there is increased cell proliferation, as if the skin heals as it breaks. Milinkovitch plans to focus his future research on the physics of biology, for instance, on understanding how mechanical forces, such as a tissue being pulled, control cell proliferation.

Crocodile head scales form by a physical process similar to how dried mud cracks. 

A smart move

His interest in physics goes a long way back. He had a brief affair with it during his undergraduate degree, but decided to stick with his passion for evolution and pursue a PhD at the Université Libre de Bruxelles (ULB) studying phylogeny using DNA hybridization. A chance encounter with Jeffrey Powell at a conference, however, changed his career path. He moved to the US to do a PhD with Powell at Yale University, where he discovered the power of DNA sequencing, which was booming at the time. He swiftly swapped DNA hybridization for DNA sequencing, a step that would place him at the forefront of the phylogenetics scene “It was the right time and the right place.” This was the start of a productive career that led to his first independent position back at ULB in 1996. In these early years, his lab focused on two rather different areas: conservation biology and phylogeny inference.

Saving the Galápagos tortoises

Milinkovitch used phylogenetic and population genetic approaches to investigate the influence of human factors on the genetic diversity of endangered species. His work helped managing south-American dusky dolphin populations threatened by illegal fishing, as well as Jamaican boas, Komodo dragons, amongst others. But his most remarkable conservation success story is the rescue of a near-extinct Galápagos giant tortoise.

The Galápagos are a group of islands on the Ecuador coast famous for inspiring Charles Darwin’s theory on the mechanisms of evolution. Each of the main 13 islands had a different giant tortoise species, but now only ten remain, mostly because their habitats were destroyed by cattle grazing, or because they were poached by whalers, sealers and pirates, who used them as a long-lasting supply of fresh meat (tortoises can survive months without eating or drinking). In the mid 1960’s there were only 14 tortoises left in Española Island, which were taken to the main island for captive breeding by the Charles Darwin Research Station and Galápagos National Park. Milinkovitch and colleagues joined the breeding program in early 1990s, and for ten years they did molecular genetic analysis of the tortoise offspring. Today, there are nearly 2000 giant tortoises in Española, and a significant proportion of these animals were born in the island. “This is one of the most successful repatriation breeding programs ever” he says enthusiastically “Our work shows that genetics can help, we can identify populations […] and maximize genetic diversity in the offspring”.

Michel Milinkovitch with Galapagos baby tortoises.

Frogs hop ‘out of India’

With the progress of computer and DNA sequencing technologies in the 1990’s, evolutionary biologists developed new methods to study phylogeny. Milinkovitch created various software for phylogeny inference, such as MetaPIGA and MANTiS, and worked out some intriguing phylogenetic trees. Of note is his work on the phylogeny of cetaceans, which revealed a close relationship between whales and artiodactyls like hippos, camels and pigs, and his ‘out-of-India’ theory to explain how frogs radiated from India.

Sixty-five million years ago, when dinosaurs were being wiped out because of a massive asteroid impact on the Yucatán coat of Mexico, colossal volcanism events lasting around 10 thousands years covered nearly half of India in lava, leading to mass extinction. Milinkovitch and his student Franky Bossuyt showed that frogs survived this cataclysm and spread around the world after India collided with Asia during continental drift. This and other phylogenetic studies on frogs planted a seed of curiosity that would later make Milinkovitch shift the main axis of his lab to evolutionary developmental biology (evo-devo).

From phylogeny to evo-devo

A problem arose in his mind when he started working on evo-devo. “If you want to study evolution of spines in mammals, or how scales develop, the mouse is not going to help.” To study the evolution of certain traits like spines or scales he would have to use atypical animal models, and this is quite complicated. He started small colonies of snakes and lizards at ULB, but didn’t feel he had enough support to expand this work. However, this fresh view of using non-model animals caught the eye of the evo-devo science community of the University of Geneva, who invited him to work there. Milinkovitch’s lab moved to Geneva in 2004, where he now runs several unique colonies of reptiles and other exotic mammals to study, for instance, how spines develop in unrelated species such as hedgehogs and tenrecs. Aware that breeding these animals is a mammoth (and costly) task that not many can take on, he opens his colonies to other labs around the world.

Milinkovitch's team on a field trip.


Research in Switzerland

What is research in Switzerland like? It seems that even when it comes to science the cliché stands. “It’s very well organised” Milinkovitch answers. He explains how the peer reviewing process that evaluates grant proposals is well managed and fair in Switzerland “What counts is excellence, nothing more”. Project proposals are judged on their quality, and there is little pressure to do ‘applied research’, in contrast to many funding agencies worldwide. “There is a big difference between encouraging interest in applying basic science and forcing work on applied stuff” he says. Another aspect he praises in Swiss research is the wide support for multidisciplinary projects. The Swiss government has recently established SystemsX, a large public research initiative to advance systems biology research in the country. He admits that multidisciplinarity is what excites him most in his research. He currently has working in his lab people with backgrounds in physics, computer science, engineering, evolutionary and developmental biology. This eclectic combination of expertises is not always easy to manage, he confesses, but it is incredibly prolific, and the recent Science paper on crocodile scale development is a clear example of that. “I think this has been possible here because there are specific grants for multidisciplinarity, but also because terrific students are open to these things”.

Reference:

Milinkovitch, M., Manukyan, L., Debry, A., Di-Poi, N., Martin, S., Singh, D., Lambert, D., & Zwicker, M. (2012). Crocodile Head Scales Are Not Developmental Units But Emerge from Physical Cracking Science, 339 (6115), 78-81 DOI: 10.1126/science.1226265

Image credits: LANE-Michel Milinkovitch

This article was published in Lab Times on the 7-02-13.