What Snakes Teach Us About the Mouse Body Plan

Make your way along the length of a vertebrate and it becomes pretty clear that the backbone is divided into regions. The vertebrae of the ribcage, for example, have a distinct shape and can easily be told apart from the lumbar vertebrae further along. What about snakes, though? Like their body, their backbone isn't obviously divided into regions, though the divisions seem clear in lizards, their closest relatives. Did snake bodies "deregionalize" during evolution, losing the morphological differences along their backbone? If so, how? Jason Head and David Polly took on the challenge of understanding how a lizard-like body evolves into a snake-like one.

The duo tackled the question by analysing the variation in the shape of individual vertebrae along the bodies of around 50 different species of lizards and snakes, along with the alligator Alligator mississippiensis. They used a statistical analysis to estimate the number of regions along each vertebral column and the boundaries of each region. Overlaying that onto a phylogenetic tree, they asked a simple question: Given the distribution of the data, is it likely that snakes have "lost" regions during their evolution -- that is, do they have fewer regions than lizards? The alligator skeleton is well-documented as having four regions, so the researchers included it as a known point of comparison.

(It may seem odd to use alligators as an outgroup for comparison, but they aren't actually lizards. Together with birds and crocodilians, they form a reptile clade called Archosauria, which is a sister group to the clade which includes snakes and lizards. In other words, an alligator is more closely related to a parakeet than to a snake. Really.)

The results were pretty clear. Head & Polly wrote that a "model in which limbed squamates [ie, lizards] and Alligator have four regions and this is reduced to three in snake-like taxa was no better supported than the hypothesis that all squamates share three regions and Alligator has four". In other words, there's no evidence that lizard bodies have more regions than snake bodies, and no reason to believe that snakes have "lost" regions during their evolution.

In fact, a careful analysis turns the picture on its head. The skeletons of snakes and lizards fit in well with the general pattern of reptiles and ancient mammals. In all of these groups, the shape of the vertebral column and ribs changes only gradually along the length of the body. This differs from modern mammals and archosaurs, whose skeletons have distinct changes along their length dividing them into regions. One of the figures in the paper shows this nicely:

spacer
This tree shows the skeletons of a few amniotes; mammals branch off node number 3 and reptiles off node number 2. The mouse (Mus) and alligator (Alligator) skeletons stand out because they're much more clearly divided into regions than the other skeletons, which show a much smoother change in morphology along their length. The general picture that emerges is one where the ancestral condition for amniotes is subtle gradations in vertebral morphology. This is still the pattern in snakes and lizards, while mammals and archosaurs have adapted the same machinery to produce regions with distinct boundaries. Head and Polly argue that the major innovation in the evolution of snake-like bodies wasn't the loss of regions, but a developmental decoupling between the parts of the skeleton along the belly- and the back-side, and suggest that future research should focus on understanding how that happened.

To me, this paper turned out to be a great example of the problems that can arise from focusing our research on a handful of organisms. Life is rich and diverse and very, very weird, and it's easy to end up with a skewed view of it because we look at creatures that are familiar or accessible or popular. Of course, we also learn a lot from this approach, but it's important to step back and have a broader look every once in a while. Even though mice and alligators show similar body patterning during development, it turns out that their ancestors converged on the same approach. Everyone in between the two groups does it differently. Our ancestors figured out how to sharpen the boundaries between body regions and gained an extra region during their evolution; snakes didn't lose any. They're the normal ones.