Domains in Evolution: Modularity and Combinatorial Protein Structure

Development of biochemical, biophysical, and bioinformatic tools to identify and classify protein domains is a means of linking and understanding protein structure in the context of evolution. Another way of describing domains is as structural modules that lead to functional diversity in evolution (Moore et al., 2008). The smallest proteins often possess only a single domain, although many proteins contain multiple domains.

The fundamental idea that proteins evolved to have either repeated domains or multiple domains has been postulated for decades. Recent advances in bioinformatic resources, however, provide a clearer view of structural similarity and homology with functionality across genomes and over evo­lutionary timescales. This evidence, then, suggests evolution exploits the modularity of protein domains to create multi­domain proteins with more complex and functions.

Multi-domain protein families evolved to perform complex biochemical and cell signaling processes. A classic example of a multi-domain protein is pyruvate kinase, which catalyzes the conversion of phosphoenolpyruvate to pyruvate in glycolysis (Gupta and Bamezai, 2010). Pyruvate kinase contains three modular domains: an a/ß-barrel domain, a ß-barrel domain, and an a/ß/a-sandwich domain.

These three domains work in concert to catalyze the transfer of a phosphate group from phosphoenolpyruvate to adenosine diphosphate (ADP) to generate pyruvate and adenosine triphosphate (ATP). Inter­estingly, this three-domain protein is retained throughout al­most all organisms and the modular structure of the protein is also largely conserved.

Protein domains can also greatly expand the diversity of function through reorganization of functional modules in a combinatorial manner. For example, non-ribosomal peptide synthetases (NRPS) are large proteins that possess multiple enzymatic domains with each domain catalyzing a specialized reaction (Strieker et al., 2010).

These large proteins synthesize non-ribosomal peptides, which are often important natural products with antimicrobial or other qualities. Importantly, these proteins evolved as gene clusters that encode an ordered series of domains capable of performing intricate biosynthetic steps, including adenylation, cyclization, condensation, epimerization, methylation, and oxidation/reduction reactions for the synthesis of non-ribosomal peptides. Moreover, the diverse combination of NRPS domains directly results in chemical diversity in the resulting molecules.

The study of NRPS, as well as other multi-domain proteins, led to the idea that scientists can engineer multi-domain proteins to perform a custom biochemical process (Cane et al., 1998). This science, often called combinatorial protein bio­chemistry or protein engineering, exploits the independent stability and modularity of protein domains to generate novel unnatural proteins by domain swapping.

This engineering step has applications throughout many scientific fields including, but not limited to novel natural product biosynthesis, at­tenuation or perturbation of signaling cascades, and poten­tially the creation of proteins of novel function.