Evolution of Elongation Factor G

I got some good news the other day that my paper Evolution of elongation factor G and the origins of mitochondrial and chloroplast forms" has been accepted for publication in Molecular Biology and Evolution. Today it was published in an advance access format.
 Elongation factor G (EF-G) is a bacterial GTPase translation factor that binds to the ribosome and catalyses the movement of aminoacyl-tRNA from the A site to the P site (translocation). It also has a second function in ribosome recycling, where it interacts with RRF, the recycling factor, splitting the ribosomal subunits. EF-G is one of the most ancient proteins, with orthologues in archaea and eukaryotes (EF2), and it's also found in the eukaryotic organelles chloroplasts and mitochondria. Our paper is concerned with the bacterial and organellar versions.


I started working on EF-G during the early stages of my PhD in York with Sandra Baldauf. I found that there were two versions of EF-G in mitochondria (mtEFG1 and mtEFG2), and that these grouped with an odd selection of bacteria (spirochetes, planctomycetes and delta-proteobacteria, hence the name spdEFG1 and spdEFG2). The association of mitochondrial and spirochete EF-G had been noted briefly in Ford Doolittle's "Phylogenetic Classification and the Universal Tree" paper in Science , but there were no in depth analyses published. I presented my work at the SMBE meeting in Halifax, Nova Scotia in 2007, and we got the paper into shape for a manuscript. However, for several reasons (mainly that I was running out of time to submit my PhD and I needed to finish off my other thesis chapters), things got delayed and the paper was put on the shelf for a while. I came back to it when I started my post doc in 2008. In the meantime, the two paralogues were identified in yeast and spirochetes and biochemical investigations were carried out. This confirmed what we suspected: the two versions had become subfunctionalised for two facets of EF-G function, translocation and ribosome recycling. It was definitely high time to look at this in detail from an evolutionary perspective.


It's protein functional and structural evolution that I find most fascinating, so for me, the most exciting thing in our results is how asymmetric the subfunctionalisation has been. mt/spdEFG1 is very highly conserved, often with residues very different to classical EF-G (including a three amino acid insertion in the all-important Switch I region of the GTPase domain), whereas mt/spdEFG2 is much less conserved, even in the sites that are supposedly involved in its ribosome recycling function. It seems that for this function, all that's required is something roughly EF-G-shaped. Indeed, the most well conserved sites of spdEFG2 seem to be those essential for structural integrity. Just having this shadow, shell of an EF-G around to do the ribosome recycling seems to have freed spdEFG1 from some constraints and enabled it to become more highly specialised for its role in translocation.


So where to go from here with EF-G? Well we found that some delta-proteobacteria have all three types of EF-G (canonical EF-G, spdEFG1 and spdEFG2). I'm hoping that one day someone will check out these versions in depth in terms of recycling/translocation functionality and GTP hydrolysis activity. Maybe then we'll be be able to answer the burning question: just what does that Switch I insertion do?!

Atkinson GC, & Baldauf SL (2010). Evolution of elongation factor G and the origins of mitochondrial and chloroplast forms. Molecular biology and evolution PMID: 21097998

Doolittle, W. (1999). Phylogenetic Classification and the Universal Tree Science, 284 (5423), 2124-2128 DOI: 10.1126/science.284.5423.2124