The most profound transition in the biochemistry of life is thought to have been the one from a primordial RNA-peptide world to the modern DNA-protein world. The reasons for this transition are a matter of speculation, but may have included changes in planetary conditions that disfavored long RNA molecules. Whatever the reasons, the original combination of both catalysis and replication in one class of molecules (RNA) gave way to the tripartite division of labor we still observe today, with proteins providing the catalytic activity, DNA serving as the information repository, and RNA mediating between them.
The change from RNA to DNA as the main repository of biological information seems chemically unchallenging, but there is a major obstacle on the path from peptides to proteins, the so-called protein folding problem. Chemical activity requires structure and whereas the peptides of the RNA-peptide world could use nucleic acids as scaffolds, proteins needed to start folding autonomously after most RNA scaffolds went extinct. Since folding is an exceedingly rare property in random polypeptides, an origin of folding by random concatenation of amino acids does not seem attractive. Instead, a few simple mechanisms have been substantiated that can lead from unstructured precursors to folded proteins. These mechanisms are a long-standing focus of our research.
Here we discuss an exploration of these mechanisms using fragments of ribosomal proteins as a starting point. The ribosome is the main survivor of the transition to the DNA-protein world, due to its irreplaceable function in the synthesis of new proteins, and its proteins provide the earliest examples of polypeptides on their way to folding autonomously. Our results suggest that repetition of building blocks was the most straight-forward path to structure, whereas the recombination of preoptimized building blocks was largely unsuccessful.