On the evolution of the genetic code.

Abstract

The TE (translation error) model for evolution of the genetic code starts with a primitive cell possessing random, ambiguous codon assignments, and a very error-ridden translation process, and shows how such cells, from the "necessity" of minimizing the effects of translation errors, can evolve the highly ordered codon catalogue observed today. The real value of this model is in the recognition that at sufficiently early stages in evolution the fundamental information -transferring processes, i.e., translation, replication, and transcription, must have been errorridden, and in the derivation of some of the possible consequences of this fact. This concept and its general consequences are not restricted to codes derived in basically stochastic ways. All of evolution, with the possible exception of the more recent evolution [long dash] e.e., the last billion years or thereabouts[long dash]must be viewed as being limited and therefore defined by the accuracy with which information transfer can take place in the cell. Very probably the evolution of accurate translation mechanisms, etc., occurred in a series of more or less discrete stages, each beginning with an improvement in a given information transfer process (through mutation), which then lead to the gradual evolutionary working out of its ramifications, which, in turn, set the stage for[long dash]made possible[long dash]a further improvement in information transfer, etc. Evolution during any of these hypothetical stages could be qualitatively different from that occurring in any other stage, for the basic cell type would probably differ from one stage to another more drastically than do any of the cell types now extant. If we picture an early cell as possessing solely "statistical proteins," from which it can fashion only a very few low-specificity, slow-acting, inaccurate "ur-enzymes," it is quite clear that such a cell is not capable of evolving the great variety of cell types found on earth today[long dash]an evolution which requires a cell to possess many precisely defined enzyme functions, capable of subtle modifications, and a metabolic pattern tightly interrelated through various feedback controls of its enzymes. Thus to date, we may have witnessed only the final "divergent" stage in evolution. All the previous stages[long dash]which might be considered "convergent" by virtue of their having as the sole or main "goal" the improvement of some feature of information transfer[long dash]have perhaps gone undetected so far.