Evolution of a New Enzymatic Activity in Nature

Enzyme evolution in natural systems typically involves several steps: (1) gene duplication, (2) change in functionality, and (3) selection for activity/specificity (see Fig. 2.2). Duplications that occur at the individual gene level provide the starting point for enzyme evolution.

FIGURE 2.2 General scheme for natural evolution of enzyme activity. A, Parental gene; A/B gene encoding protein with dual activity that can perform activity B poorly; A/B, gene that encodes protein with dual activity where B is the major activity; B gene encoding activity B that is unable to perform activity A; A* represents a gene pseudogene that becomes excised.

Mutations constantly arise in genes, but their accumulation depends on stringency of the selection pressure for the function of the gene product. There are three common fates that befall duplicated genes (Fig. 2.2): (1) retention of function, (2) change of function (either change in activity or change in expression pattern), or (3) loss of function followed eventually by excision.

Changes in enzyme function typically follow one of the three mechanisms (Gerlt and Babbitt, 2001). The first mechanism is one in which a partial reaction or a strategy for stabilization of energetically unfavorable transition state is maintained, while the substrate specificity changes. In a second mechanism, substrate specificity is maintained, but the chemistry changes during evolution. A third mechanisminvolves retaining only the active site architecture, without maintaining either substrate specificity or chemical mechanism.

Whichever of the mechanisms predominate, several features are likely to be common. An initial gene duplication event is followed by the accumulation of multiple mutations in one of the copies. A prerequisite for alteration of specificity is that the original tight active site substrate specificity should relax allowing a number of potential substrates to bind, or the same substrate to bind in alternate conformations. Once an alternate substrate is capable of binding (or the same substrate in a different binding conformation), an altered enzymatic transformation may occur, resulting in the accumulation of a novel product. If the new product conveys a selective advantage, over successive generations the accumulation of further mutation/selection can lead to an increase in the new activity. This ‘‘tuning’’ to the new substrate often occurs at the cost of catalytic efficiency with respect to the original transformation. Thus, a characteristic of newly evolved enzymes, or enzymes caught in transition, would be the observation of relaxed specificity. Examples of this can be found in the fatty acid desaturases (Broun et al., 1998; Dyer et al., 2002), where enzymes that exhibit ‘‘unusual’’ specificity with respect to the parental enzymes are often bifunctional in that they are capable of performing the archetypal reaction, often with lower catalytic rates than the parental enzyme (Shanklin and Cahoon, 1998). Amino acid substitutions that change the geometry of the binding pocket can be either direct, that is, when the amino acid side chains directly line the binding pocket, or alternatively can be at sites remote from the binding pocket and mediate their effects via subtle changes in the relative organization of secondary structural elements. In this context, amino acid side chains have been referred to as ‘‘molecular shims’’(Whittle et al., 2001) that orient the substrate with respect to the active site in a very precise manner similar to the way carpentry shims are used to level furniture. The stronger the selection pressure for the improvement in activity, the faster it will progress.

Similarly, in the case of changes to the chemistry occurring on the same substrate, it is envisaged that the enzyme became bifunctional with respect to reaction outcome either by acquiring two or more alternate binding modes or by alterations in the amino acid side chains that participate in catalysis. This has been observed for the Fad2 family of fatty acid modification enzymes (Broadwater et al., 2002; Broun et al., 1998). Once the new reaction occurs even at low levels, selection can favor mutations that increase the new activity and lead to improved fitness at the organismal level and thus provide selective advantage. In either case where substrate specificity changes, or chemistry on the same substrate alters, the ability of an enzyme to perform alternate reactions shows it has the potential to acquire a new dominant activity.

Duplicated genes that do not provide a selective advantage are rapidly excised by unequal crossover at meiosis. Evidence for this includes studies in which subfunctionalization is shown to occur rapidly upon polyploidization in cotton (Adams et al., 2003) and the observation of lower than expected occurrence of pseudogenes (Force et al., 1999).