Structural variability of multifunctional proteins indicates frequent stochastic evolution of protein oligomers.

Recently, it has been suggested that the evolution of many protein homomer complexes follows a neutral pattern, with little effect on their biochemical function. One of the strongest arguments in support of this hypothesis is the observation that homologous enzymes with the same catalytic function can have different quaternary structures in various species. However, in the case of proteins with multiple functions ("moonlighting" proteins), this pattern can also have an adaptive explanation if quaternary structure is responsible for their variable, non-canonical functions. To test whether moonlighting can be responsible for the variability of quaternary structure, here we examine the opposite of the "same function-multiple structures" pattern, and test whether orthogroups of moonlighting (multifunctional) and non-moonlighting proteins have similar quaternary structure variability. We show that there is very little association between moonlighting and homomer quaternary structure diversity, which is in agreement with the neutral expectation and the hypothesis that many homomers might be adaptive by shaping the biophysical characteristics of the cell and cytoplasm, rather than the biochemical function of the protein.