Flexible Proteins at the Origin of Life

Abstract

Almost all modern proteins possess well-defined, relatively rigid scaffolds that provide structural preorganization for desired functions. Such scaffolds require the sufficient length of a polypeptide chain and extensive evolutionary optimization. How ancestral proteins attained functionality, even though they were most likely markedly smaller than their contemporary descendants, remains a major, unresolved question in the origin of life. On the basis of evidence from experiments and computer simulations, we argue that at least some of the earliest water-soluble and membrane proteins were markedly more flexible than their modern counterparts. As an example, we consider a small, evolved in vitro ligase, based on a novel architecture that may be the archetype of primordial enzymes. The protein does not contain a hydrophobic core or conventional elements of the secondary structure characteristic of modern water-soluble proteins, but instead is built of a flexible, catalytic loop supported by a small hydrophilic core containing zinc atoms. It appears that disorder in the polypeptide chain imparts robustness to mutations in the protein core. Simple ion channels, likely the earliest membrane protein assemblies, could also be quite flexible, but still retain their functionality, again in contrast to their modern descendants. This is demonstrated in the example of antiamoebin, which can serve as a useful model of small peptides forming ancestral ion channels. Common features of the earliest, functional protein architectures discussed here include not only their flexibility, but also a low level of evolutionary optimization and heterogeneity in amino acid composition and, possibly, the type of peptide bonds in the protein backbone.

Keywords: ancestral enzyme; ancestral membrane protein; flexible protein; ion channels; primordial protein structure; protein ligase.

Figure 1

The structure of the ligase discovered through in vitro evolution [61]. (a) Superimposed NMR structures demonstrating the flexibility catalytic loop in orange, the rigid hydrophilic core in gray, and the flexible termini in blue; (b) The zinc-binding residues of the rigid hydrophilic core. Zinc atoms are grey balls. Residues coordinating them are marked; (c) Schematic of the zinc-binding residues. Strands that connect zinc atoms are in blue. The residues considered to be involved in coordinating zinc atoms are represented explicitly. The catalytic loop is in green. The red rectangle indicates the region presumed to contain catalytic residues.

Figure 2

A representative snapshot of the hydrophilic core of the ligase from Molecular Dynamics simulations. The point of view is that of the flexible termini. Protein residues in the core are colored yellow, zinc ions atoms are grey balls, and water molecules within 5 Å of zinc ions are red.

Figure 3

Root-mean-square deviation (RMSD) of the backbone atoms of the ligase over 550 ns of the molecular dynamics simulation.

Figure 4

The structure of the double mutant E28A/D29A. Tails (residues 1–17 and 66–87) are gray, loop (residues 35 to 59) is green, residues 18–34 and 60–65 near Zn atoms are blue. White, dashed lines indicate salt bridges Asp55-Lys17 (on the left) and Asp43-Arg61 (on the right).

Figure 5

Snapshot from a simulation of the AAM channel after 3.6 μs (trajectory T2). (a) View of the channel from above. Each monomer of the hexametric structure is in a different color; (b) Side view of the channel represented as ribbons and water molecules filling the pore. All other components of the system were removed for clarity.

Figure 6

Cumulative transmembrane ion fluxes in simulations of AAM. The total number of Cl⁻ ions (red) and K⁺ ions (green) that cross the channel as a function of time for trajectory T1 (A) and T2 (B). The simulations were carried out at an ionic strength of 1 M (see Table 1).