nai

Project Progress

In order to generate laboratory models of protocellular protein catalysts, we have designed a novel, partially random protein library based on the DNA-binding domain of the human retinoid-X-receptor. We chose this domain because of its small size, stable fold, and two closely juxtaposed recognition loops. We replaced the two loops with segments of random amino acids, and used mRNA-display to isolate new proteins with different functional properties.

In the first selection, variants that specifically recognize adenosine triphosphate (ATP) were isolated. This work demonstrated that it is possible to significantly alter the function of a natural protein from DNA binding to ATP recognition while retaining the protein fold. In a second selection, we used the same library to isolate a novel protein that catalyzes a template-directed oligonucleotide ligation reaction. The chemistry of the reaction is the same as that catalyzed by protein polymerases, i.e. the attack of a 3’-hydroxyl on a 5’-triphosphate. We are currently characterizing the properties of this new catalyst.

Using experimental and computer simulation methods, we continue to characterize our previously isolated ATP binding protein, which was obtained from a random sequence protein library. This protein exhibits a folding pattern, not previously seen in biological proteins. Using NMR spectroscopy, we showed that a new variant of this protein, which was selected for higher thermodynamic stability, retains the same overall structure. We expect to interpret the role of each mutation that arose during the evolutionary optimization of this protein.

{{ 1 }}

In order to explain the emergence of proteins that mediate transfer of ions, nutrients waste products and environmental signals across protocellular walls we performed extensive computer simulation studies of model membrane peptides. These studies rely on an observation that, although contemporary membrane proteins are large and complex, their structural motifs are simple, with α-helices being most common. This suggests that membrane proteins might have evolved from simple building blocks. We demonstrated that the insertion of α-helices into a membrane is unfavorable, but stability can be regained through their specific recognition and association into larger assemblies. We further showed that, despite their simple structure, the emerging transmembrane structures could possess properties that appear to require markedly larger complexity. These properties can be subtly modulated by local modifications to the sequence rather than global changes in molecular architecture. This is a convenient evolutionary solution because it does not require imposing conditions on the whole amino acid sequence.

{{ 2 }}

{{ 3 }}