Toward the experimental codon reassignment in vivo: protein building with an expanded amino acid repertoire

The high precision and fidelity of the genetic message transmission are ens ured by numerous proofreading steps, from DNA replication and transcription to protein translation. The key event for translational fidelity is the pr oper codon assignment for 20 canonical amino acids. An experimental codon r eassignment is possible for noncanonical amino acids in vivo using artifici ally constructed expression hosts under efficient selective pressure. Howev er, such amino acids may interfere with the cellular metabolism and thus do not belong to the 'first' or restricted' part of the universal code, but r ather to a second or 'relaxed' part, which is limited mainly by the downstr eam proofreading in the natural translational machinery. Correspondingly, n ot all possible a-amino acids can be introduced into proteins. The aim of t his study is to discuss biological and evolutionary constraints on possible candidates for this second coding level of the universal code. Engineering of such a 'second' code is expected to have great academic as well as prac tical impact, ranging from protein folding studies to biomedicine.