Homology, hierarchy and swimming sea slugs

Compare your arm to the arm of a monkey or the foreleg of a cat and you’ll probably find it as no surprise that they’re made up of similar bones arranged in similar ways. A commonplace observation for us, but for Darwin and other early naturalists, these similarities offered proof for the theory of evolution. The bones in our arms and the bones in the forelimbs of other vertebrates are organized in similar ways because they evolved from a common forelimb structure that was present in the ancestor of all tetrapods.

The idea that similar traits in disparate species are related through evolutionary history is commonly referred to as homology, a termed coined by the 19th century comparative anatomist Richard Owen. The concept of homology is easy enough to appreciate, and evolutionary theory has been greatly influenced by it. Nevertheless, there has been some contention over the details of its definition.

Consider, for instance, how homologous traits are formed during development. If we were to look under the hood, so to speak, and uncover the mechanistic paths by which homologous traits develop in different species, would we find that they too are homologous? In other words, would similar mechanisms govern the formation of similar traits in different species? Or perhaps homologous traits can in fact be non-homologous at developmental and genetic levels.

It’s intuitive to think that homologous traits would develop through similar mechanisms. After all, it is the simpler of the two scenarios. But some biologists have posited that Occam’s razor may not apply in this case. Biological systems are hierarchically organized, they argue. Phenotype, genotype and the developmental processes in between all represent different levels in the hierarchy. While each level in the hierarchy is causally linked to another level, there is no a priori reason why each level couldn’t evolve independently of others. And in such cases, homologous phenotypes would not necessarily be homologous at deeper levels.

Researchers in Paul Katz’s laboratory at Georgia State University have recently uncovered insights into this problem by studying the swimming behaviors of sea slugs. They discovered that while different sea slug species exhibit similar, homologous swimming behaviors, the neural mechanisms that generate their behaviors are remarkably different. Thus, homology at one level does not necessarily imply homology at another.

Now I’m going to guess that chances are you haven’t spent much time watching sea slugs swim. It’s a shame, really, because it’s quite a spectacular sight.

The sea slugs that Katz’s laboratory study swim by flattening and repeatedly flexing their bodies from side-to-side. The neural circuit that generates this ‘left-right’ swimming behavior was first identified in a certain sea slug species in the genera Melibe. It turns out that the circuit is remarkably simple: it’s composed of only four pairs of bilaterally symmetrical neurons that are wired up in such a way that their net pattern of activity produces the alternating, rhythmic left-right body flexions that allow the sea slug to swim.

Akira Sakurai, a research scientist in Katz’s group, was curious about this circuit in another sea slug species, Dendronotus iris, which also exhibits left-right swimming behaviors homologous to those of Melibe. When he probed the swim circuit in the Dendronotus species, he was up for a big surprise. The four pairs of neurons were there, as he expected. But amazingly, the neurons were wired up in a totally different way. Nevertheless, the net output of the circuit in Dendronotus produced the same swimming behaviors as those observed in Melibe sea slugs.

These discoveries are consonant with the hierarchical organization of biological systems. They demonstrate that homologous behaviors in different species are not necessarily generated by homologous neural mechanisms. But you must wonder, why? Why would these mechanisms diverge in the first place? There are many plausible reasons, but my favorite explanation comes from studies on the evolution of gene regulation in animals.

Imagine for a moment that neural circuits are to a certain extent buffered against perturbation and capable of accommodating subtle deleterious changes in structure or function that evolve randomly through drift. As changes accumulate, the function of the system would eventually be affected. However, the neural circuit would continue to maintain the appropriate output by evolving compensatory changes in other aspects of the circuit. Thus, over time, the ‘seesaw’ of deleterious and compensatory changes would cause the circuit mechanisms to diverge between species while the behavioral output would remain unchanged.

Regardless of the reasons why neural circuits may evolve the way they do, the results from the Katz laboratory spark many more questions. One wonders how many different configurations a given circuit can take on. If there is a finite number, what are the constraints? Exploring the swim circuit in more sea slug species will undoubtedly lead to many more insightful discoveries.