Gene duplication is the primary source of new genes, but the molecular evolutionary mechanisms underlying functional divergence of duplicate genes are not well understood. Des Marais and Rausher argued that data from plant dihydroflavonol-4-reductase (DFR) genes support the model that gene duplication allows the escape from adaptive conflict (EAC) among several functions of a single-copy progenitor gene. As the authors indicated, the key predictions of EAC, in comparison to other models, are that (i) adaptive changes occur in both daughter genes after duplication, and (ii) these adaptive changes must improve ancestral functions. Furthermore, EAC indicates that (iii) the improvement of several ancestral functions is constrained before duplication, although this last point was not explicitly stated. Here we show that contrary to the predictions of EAC, only one of the duplicated DFR lineages exhibited adaptive sequence changes. Owing to the lack of information on enzyme concentrations we question the accuracy of enzyme activity comparisons, and it is thus not clear that any ancestral function has been improved in either lineage.
