Structure and evolution of the genetic code viewed from the perspective of the experimentally expanded amino acid repertoire in vivo

Much effort has been devoted recently to expanding the amino acid repertoire in protein biosynthesis in vivo. From such experimental work it has emerged that some of the non-canonical amino acids are accepted by the cellular translational machinery while others are not, i.e. we have learned that some determinants must exist and that they can even be anticipated. Here, we propose a conceptual framework by which it should be possible to assess deeper levels of the structure of the genetic code, and based on this experiment to understand its evolution and establishment. First, we propose a standardised repertoire of 20 amino acids as a basic set of conserved building blocks in protein biosynthesis in living cells to be the main criteria for genetic code structure and evolutionary considerations. Second, based on such argumentation, we postulate the structure and evolution of the genetic code in the form of three general statements: (i) the nature of the genetic code is deterministic; (ii) the genetic code is conserved and universal; (iii) the genetic code is the oldest known level of complexity in the evolution of living organisms that is accessible to our direct observation and experimental manipulations. Such statements are discussed as our working hypotheses that are experimentally tested by recent findings in the field of expanded amino acid repertoire in vivo. Received 30 June 1999; accepted 9 July 1999