Phenotypic plasticity of lifehistory traits

Many life-history attributes do not attain fixed values but respond to environmental conditions in an adaptive fashion. Part of the variation in life histories over species stems from the fact that different species have adopted different ways of responding to environmental factors. The functional relationship between the phenotype and the environment is called a norm of reaction (see Section 5.1). Life-history theory assumes that such a norm itself can be subject to natural selection and is optimized with respect to certain environmental conditions. We have discussed plant flowering time in the previous section as a life-history trait; however, its response to photoperiod and temperature can be seen as a reaction norm. In this section we will discuss more examples of pheno-typic plasticity and its genomic underpinning.

There is considerable confusion as to how the primarily ecological concept of a reaction norm should be interpreted on the genetic level. One view holds that reaction norms are no more than a set of life-history traits measured in different environments, and that one has to analyse each trait in each environment as a separate variable, taking into account cross-environment correlations; this approach is also known as the character-state view of phenotypic plasticity (Roff 2002). Another approach emphasizes the continuity of the reaction norm across environments. This is especially appropriate if environments form a natural gradient, for example in the case of temperature. Plasticity is then defined by the response in the average environment, plus the slope of the reaction norm if it is linear. For non-linear reaction norms there is no obvious indicator of plasticity, although the first derivative of the reaction-norm value in the average environment may be used as a local measure of plasticity when that environment changes (De Jong 1995). This approach is known as the reaction-norm view of phenotypic plasticity (Roff 2002). It has been suggested that in addition to genes determining the average response, there should exist 'genes for plasticity'; that is, genes which determine the slope of the reaction norm. Natural selection emanating from variable environments could promote such plasticity genes. In particular, plasticity is favoured (Schlichting and Smith 2002):

• if environmental change is frequent,

• if environmental cues are reliable,

• if environmental variation is fine-grained in space or time,

• in the case of coarse-grained temporal variation, if change is predictable, or

• in the case of fine-grained temporal variation, if there is a predictable sequence.

The reaction-norm perspective is very much dominated by the terminology of quantitative genetics and by statistical analysis of phenotypic data across environments. Molecular data are changing this perspective (Pigliucci 1996; Schlichting and Smith 2002). The example of flowering time discussed above, as well as the cases discussed below, illustrate that it may be difficult to reconcile the quantitative-genetics views of plasticity with modern genomic insights. The concept of a plasticity gene does not have a clear interpretation on the molecular level. Rather, plastic life-history traits are determined by networks of gene expressions, which integrate multiple cues from the environment with signals from the internal metabolism in such a way that the difference between genes for plasticity and genes for an average response cannot be seen. In this section we discuss three cases—polyphenism and body size in insects, and shade avoidance in plants—to illustrate this argument. As we will see, the genomic underpinning of life-history plasticity is less well developed than some of the other issues discussed in this chapter, so the section is phrased in terms of examples rather than a general theory.

0 0

Post a comment