Enzyme evolution. I. The importance of untranslatable intermediates.

The proportion of two allelic mutant forms of a haploid species can be calculated from six quantities, the original proportion, the two effective growth rate constants, the two mutation rate constants for the transition from one to the other, and the elapsed time. With a program for carrying out this calculation in a mini-computer, we set about investigating the rates of exploration of multiple intragene changes in populations of enormous numbers over a large number of cell generations. It seemed evident from the outset that unless in the past very different proportions of the different types of mutational changes occurred, then neither the numbers nor the times could have been of sufficient magnitude to account for enzyme evolution by multiple simultaneous changes within a gene to yield a superior protein. Rather, we propose that the more rapid path of enzyme evolution consisted of the sequence: 1) gene duplication, 2) inactivation of a gene copy by a chain-terminating mutation or other mutational change which suppresses transcription and/or translation, 3) multiple base-pair changes and/or small deletions and additions within the non-functioning gene, and finally, 4) reversion of the mutation which was suppressing translation. It is shown that this path will lead to faster advance, if the selective advantages of different forms are dependent on the function of the gene products at certain times in history, while at other times, although this function may be necessary, it is not limiting for growth.