Where the wild things glow

Brown is not the colour that springs to mind when you hear the word ‘coral’. We are more accustomed to pictures of coral reefs with more aesthetically pleasing colours, like red or purple. That’s not an accurate representation of reality though: a large part of coral species has a more brownish colouring,  due to the endosymbiotic zooxanthellae that live inside them. These endosymbiotes provide them with energy and nutrients, and receive protection in return. But don’t let this brown colour fool you… When some of these corals are exposed to light of the right wavelength, they return the favor by fluorescing with amazing colours. The diversity of colours displayed by these fluorescent corals is remarkable, ranging from azure blue to a deep crimson red. The exact benefits of corals fluorescence are not fully known yet. The fluorescence might help provide the zooxanthellae with light in deeper waters or on cloudy days. But on places where light is abundant, the fluorescence is concentrated above the zoocanthellae to protect them from damaging UV rays. That’s some funky sunscreen if you ask me!

Some corals can be muddy brown by daylight and stunningly fluorescent when exposed to light of the right wavelength. (Copyright Anthony Calfo)

The fluorescence of corals can be traced back to a single class of fluorescent proteins (FPs). The ancestral form of this protein was green, so all other colours that we see in corals today must be derived from this ‘ancestor protein’.  In one of the coolest phylogenetic trees ever, Ugalde et al. showed the relatedness of FPs using a petri dish and some transfected bacteria. The interesting question of course is, how it is possible to evolve from one colour, to an entirely different one!

To investigate how red fluorescence might evolve from a green FP, you could compare green and red proteins that exist today and look for the differences. However, as you can see in the phylogenetic tree, the ancestor protein of red and green FPs had an intermediary orange colour. Current green FPs are therefore expected to share relevant mutations for acquiring a red phenotype. To clearly see what happened when red colour evolved, it makes more sense to compare the current red FPs to the ancestor of both green and red FPs (a).

Awesome phylogenetic tree of FPs using bacteria on a petri dish (Ugalde 2004).

This is exactly what Field and Matz did in a study published recently in the journal of Molecular Biology and Evolution. They faced a huge problem though: there were 37 amino acid substitutions between the reconstructed ancestral green FP and the current red FPs. While this doesn’t seem like that much, it means that there are 130 billion possible intermediates (2^37) between green and red!  Needless to say, it’ss impossible to reconstruct all these possible routes from green to red. So what Field and Matiz did, is construct a library of oligonucleotides, covering all possible differences between the ancestor and current red FPs.  By randomly combining these oligos, they can efficiently and quickly analyze a big bunch of possible evolutionary intermediates. A real important feature of this approach, is that it can detect epistatic interactions between mutations. (Epistasis: a combination of mutations has a bigger effect on phenotype than both mutations would have had alone. It’s the biological equivalent of the saying that ‘the whole is greater than the sum of its parts’.)

By looking at the phenotype of 20,000 bacterial colonies expressing random combinations of oligos, they could clearly see differences between colonies that fluoresced green and the ones that fluoresced yellow/orange. With 100 colonies showing either green or yellow/orange phenotypes, it becomes possible to invoke the great powers of statistics. By performing fisher exact tests Field and Matiz could determine the contribution of every single mutation to being red or being green. With this ranked list of mutations Field and Matiz tried to follow the evolutionary path from green to red in both directions. Introducing the 11 mutations into the ancestral green protein is enough to end up with a protein that is just as red as current-day red FPs. The reverse experiment was also successful: with 12 mutations the red FP could be reverted to the ancestral green colour. You can see this experiment for yourself in the nice figure below: those bacteria sure were put to work!

Retracing FP evolution: the green to red transition is visualized using bacteria. Every large dot represents a mutation in the ancestor green protein, eventually leading to the red protein of today.

As you can see, some mutations have a larger effect than others; the first few mutations merely change the hue of the green, whereas later mutations shift the green to a clear yellow and orange. However, if you introduce the later mutations by themselves in the ancestral green FP, no red colour will appear. So, the initial mutations are just as important as the colour shifting ones! This implies that natural selection is not the only force at work here. The background mutations could not be a product of natural selection, because these mutations do not affect colour at all. It is likely that other processes, like genetic drift, are also involved in red FP evolution. That’s a pretty cool if you ask me: it’s the adaptionist/genetic drift debate in a petri dish!

The evolutionary path presented here is of course not absolute. First there is the issue that the sequence of the ancestral green protein is reconstructed from FP proteins that we observe today, which might not necessarily be correct. Secondly, the order of mutations is for a large part arbitrary, because many mutations only have small effects (the ‘fine-tuners’) or are necessary for the right background, but not affect colour themselves. Nonetheless, I think it’s pretty impressive that scientists are able to reconstruct possible evolutionary histories of single proteins!

Field, S., & Matz, M. (2009). Retracing Evolution of Red Fluorescence in GFP-Like Proteins from Faviina Corals Molecular Biology and Evolution, 27 (2), 225-233 DOI: 10.1093/molbev/msp230

Coral Evolution: From Socialists to Soloists
The evolution of novelty through subtle tinkering
Taking it in: Bacterial Endocytosis