Selection Experiments as a Tool in Evolutionary and Comparative Physiology: Insights into Complex Traits-an Introduction to the Symposium1

Integrative and Comparative Biology, Jun 2005 by Swallow, John G, Garland, Theodore Jr

"The whole organism is so tied together that when slight variations in one part occur, and are accumulated through natural selection, other parts become modified. This is a very important subject, most imperfectly understood." (Darwin, 1859, The Origin of Species').

"Hence if man goes on selecting, and thus augmenting, any peculiarity, he will almost certainly modify unintentionally other parts of the structure, owing to the mysterious laws of correlation." (Darwin, 1859, The Origin of Species).

"Biological reality is so complex that we are very far from any reasonably mechanistic understanding of evolutionary processes." (Felsenstein, 1988, p. 468)

"Apparent complexity usually means a lack of understanding." (McCarthy and Roberts, 1989, p. 134)

This symposium focused on the use of selection experiments to probe complex traits. By "complex traits," we generally mean phenotypes that involve multiple subordinate traits and are polygenic. In addition, they will generally be affected by various environmental factors (e.g., both developmental and immediate temperature and nutrition), as well as interactions between genetic and environmental factors (Falconer and Mackay, 1996; Lynch and Walsh, 1998). Under this definition, complex traits tend to be those at higher levels of organization, including behavior, life history, and organismal physiology. The field of quantitative genetics developed to deal with complex traits, and artificial selection experiments, dating from the late 180Os, quickly became a cornerstone (Hill and Caballero, 1992).

We searched BIOSYS Previews to gauge how usage of complex traits, selection experiments, and some related terms have changed over the period 1969 through 2003. Figure 1 indicates some interesting trends. Complex traits were not found in the period 1969-1978. They were found in four articles during 1979-1983, and the usages there were consistent with our general definition. Relative to the other terms shown in Figure 1, usage of "complex trait" has skyrocketed in recent years. A substantial fraction of this increase is found in the biomedically oriented literature, which has followed from the recognition that such diseases and conditions as human schizophrenia are complex traits (e.g., Sullivan et al., 2003; see also Koch and Britton, 2005). In addition, major research efforts are being directed at the genetic dissection of complex traits in such model organisms as house mice (e.g., www.complextrait.org; www.jax.org/phenome; www.webqtl.org). The agricultural world is also adopting various approaches in genomics and bioinformatics in attempts to understand complex traits (e.g., Eggen, 2003).

Classical comparative physiology sought to understand "how animals work" (Schmidt-Nielsen, 1972), and was highly successful in this enterprise (e.g., see brief reviews in Mangum and Hochachka, 1998; Hochachka, 2000). An outstanding goal of modern comparative and evolutionary physiology is understanding how complex traits have evolved, including identification of factors that shape and constrain the evolution of physiological capacities (Feder et al., 1987). By explicitly incorporating an evolutionary perspective and modern tools from evolutionary biology, comparative physiologists are making great strides towards understanding the evolution of animal form and function. Several publications in the last decade have identified selection experiments (of various types) as one of the most important tools in the expanding field of evolutionary physiology (e.g., Garland and Carter, 1994; Rose et al., 1996; Bradley and Zamer, 1999; Gibbs, 1999; Feder et al., 2000; Bennett, 2003; Garland, 2003; Bradley and Folk, 2004). Our symposium attempted to illustrate the utility of selection experiments for understanding traits that are commonly studied by comparative, ecological, and evolutionary physiologists, including metabolism, locomotion, stress resistance, and aspects of life history.

A growing body of evidence indicates that many traits of traditional interest to comparative/evolutionary physiologists as measured in standard ways (e.g., locomotor performance, aerobic capacity, thermal tolerance) are heritable in the narrow (additive-genetic) sense and thus capable of responding to selection. As a consequence, selection experiments are increasingly being recognized as a powerful tool for understanding the genetic and mechanistic basis of complex physiological and morphological traits (see also Scheiner, 2002). Selection experiments with behavioral traits have a longer history and are more numerous, but have often been undertaken from a more psychological or biomedical perspective (but see Lynch [1980] for what may be considered one of the first selection experiments in ecological/evolutionary physiology).

Selection experiments allow physiologists to effect cross-generational changes and directly observe micro-evolutionary processes. Thus, they allow one to test hypotheses concerning general principles of physiological evolution. For example, selection for a single trait often results in correlated changes in other traits. Such correlated responses are commonly determined by pleiotropic gene action in which one gene affects more than one trait. Shared biochemical, physiological or developmental pathways are the caused by pleiotropic gene action; mechanistic inferences can thus be derived from monitoring correlated changes in key physiological characters. Another question that can be approached is whether adaptive evolutionary changes (i.e., cross-generational changes in allele frequencies of populations in response to selection) tend to be in the same direction as plastic changes (i.e., changes that occur during development within individual organisms, such as during acclimation, acclimatization or physical conditioning) in response to a given "stress" (see Callahan, 2005; Swallow et al., 2005).