gain of function. Loss of function of any enzyme in the core ABP would cause a loss of anthocyanin production and loss of function of F3′H or F3′5′H would cause a shift from blue to red. Transitions in the opposite directions would necessitate the gain of enzymatic function. Furthermore, after a loss-of-function transition, additional loss-of-function mutations may accumulate, making reversals even less likely (Zufall and Rausher, 2004). In the case of blue to red shifts, substrate specificity for the product of F3H may evolve, making reversals less likely as well (Zufall and Rausher, 2003).
Changes in flower color could also occur because of altered metabolic flux in side branches of the flavonoid pathway. At each of the intermediate steps in the ABP, side branches lead to the production of other important compounds such as flavones, flavonols, and proanthocyanidins (tannins) (Fig. 2.3). In fact, the entire flavonoid pathway is likely to be more complicated than that depicted in Fig. 2.3. For instance, in Arabidopsis recent profiling of both flavonoid production and gene expression in flavonoid mutants resulted in the identification of 15 new compounds and the functional characterization of 2 new genes in the pathway (Yonekura-Sakakibara et al., 2008). Changes in the abundance or activity of side-branch enzymes could result in flux away from or toward anthocyanin production and affect changes in flower color. For instance, deep-pink flowers of Nicotiana tobacum can be converted to white by overexpressing a side-branch enzyme [anthocyanidin reductase (ANR)] (Xie et al., 2003). Similarly, conversion of white flowers to those producing anthocyanins has been achieved in Petunia by silencing the side-branch enzyme flavonol synthase (FLS) and activating DFR (Davies et al., 2003). Given that the side-branch pathways result in compounds that are important for a number of other plant functions such as UV protection and herbivore resistance (Winkel-Shirley, 2002; Treutter, 2006), selection for these compounds could result in pleiotropic changes in flower color (Strauss and Whittall, 2006). However, changes in flower color caused by altered biosynthetic flux requires increases in expression levels or enzyme activity and are thus gain-of-function mutations, which, although certainly possible, are likely rarer than loss-of-function mutations (Orr, 2005).
A few studies have begun to determine the genetic basis of natural flower color variation. In Petunia axillaris, white flowers can be made pink by the introduction, either through introgression or transgenetics, of a functional copy of AN2, the R2R3-myb transcription factor that controls expression of genes late in the core ABP (Fig. 2.3). Sequence analysis of AN2 alleles from a range of collections indicated 5 independent loss-of-function mutations, suggesting that loss of color arose multiple times (Hoballah et al., 2007). However, most of these alleles also displayed strong down-regulation and thus it is possible that the loss-of-function