Translational Incorporation and Metabolic Effects of Non-canonical Amino Acids

Abstract

One branch of protein engineering uses the incorporation of non-canonical amino acids into an otherwise unaltered protein polyamide backbone in order to influence its structure and functionality beyond that typically found in nature. Previous studies have established that non-canonical amino acids that are sufficiently similar to canonical amino acids may be misrecognised as the canonical amino acid in natural protein translational systems, resulting in their residue-specific incorporation. This thesis is concerned with the interactions of such non-canonical amino acids with other cellular metabolic systems, the methodology of translational incorporation of non-canonical amino acids, and expands the range of non-canonical amino acids known to translationally incorporate into proteins. Threonine dehydratase/deaminase (TDH), an important regulatory protein in the branched-chain amino acid biosynthesis pathway, was found to be allosterically inhibited by structural analogues of isoleucine, its natural inhibitor. Substituted non-proteinogenic phenylalanine derivatives were investigated for their ability to inhibit cell growth and protein expression. Of the nine compounds tested, two compounds - 2-methylphenylalanine (2-Me-Phe) and 2-chlorophenylalanine (2-Cl-Phe) - were found to be capable of inhibiting cell growth and protein expression in a dose-dependent manner in E. coli. These inhibitory effects were reversed or partially reversed by supplementation of phenylalanine. This was originally interpreted as resulting from an interplay between inhibition of phenylalanine biosynthesis by 2-Cl-Phe and 2-Me-Phe with inhibition of cell membrane amino acid transport proteins; depending on the relative strength of these effects, the competition between the fluorinated phenylalanine analogue at the phenylalanyl-tRNA synthetase might be biased in either direction. However, further experiments to directly measure cellular phenylalanine levels by HPLC indicated that cellular phenylalanine levels are not strongly affected by either 2-Me-Phe or 2-Cl-Phe. These effects as well as growth inhibition effects were reexamined and concluded to be due to other factors, potentially including competitive inhibition of 2-Me-Phe and 2-Cl-Phe at the phenylalanine tRNA synthetase. Methodology for incorporation of non-canonical amino acids using a standard non-auxotrophic cell strain was optimised for high incorporation rates and minimal native protein expression. Following this, the relationship between residue-specific incorporation rate and non-canonical amino acid concentration in the growth medium was determined for a range of isotopically labelled and fluorinated non-canonical amino acids, demonstrating that good incorporation of non-canonical amino acids can be achieved under the correct conditions using non-auxotrophic cell strains. For the fluorinated analogues, the determined trends did not necessarily reflect what was expected based on the ease of incorporation of the same analogues in cell-free protein synthesis (CFPS) experiments, which was related to differences in ability to be transported across cell membranes. Three non-canonical amino acids substituted with fluorine at the side-chain beta-position, 3-fluoroalanine, 3,3,3-trifluoroalanine, and 3-fluorovaline, were found to be capable of replacing their corresponding canonical amino acids during translational protein biosynthesis using CFPS. These results suggest that beta-fluoro analogues of other canonical amino acids may also be capable of translational incorporation using CFPS. The utility of CFPS as a tool to study misincorporation among the 20 canonical amino acids was investigated. Experiments conducted to establish the cause of this anomalously high rate of misincorporation indicated that codon matching errors between tRNA and mRNA at the ribosome was likely to be the principal mechanism. Show more