Expanding Anfinsen’s Principle: Controlling Protein Structure by Altering Local Translation Rate

Abstract

Anfinsen's seminal experiments on the refolding of ribonuclease in vitro showed that the final native structure of a protein is determined by its primary amino acid sequence. In vivo, a nascent polypeptide chain can start to fold co-translationally, and the conformations adopted during chain synthesis can be distinct from the conformations adopted by free, full length chains refolded after dilution from denaturant. Moreover, altering protein translation rate has been shown to further affect co-translational protein folding mechanisms. Translation rate can be altered without changing the encoded amino acid sequence by altering synonymous codon usage. But common codons are typically translated faster than rare codons, and clusters of rare codons can cause significant pauses in translation. Here, we designed and tested a system to investigate the effects of a cluster of synonymous rare codons on a protein folding mechanism and its final folded structure. We designed a protein that consists of three half-domains, connected by flexible linkers. The N- and C-terminal half-domains compete to interact with the central half-domain. This protein therefore has the potential to form one of two mutually exclusive native structures. We altered synonymous codons at the 5’ end of the sequence encoding the C-terminal half-domain to adjust the rate at which the C-terminal half-domain appears outside the ribosome exit tunnel. The presence of rare codons at this location causes the N-terminal half-domain to pair with the central domain more often than the C-terminal half-domain. This effect is tunable, based on the rareness of the codons and the corresponding duration of the translation pause. These results suggest that Anfinsen's principle might need to be expanded, to include possible effects of translation rate on protein structure formation. Either