The Questions Of Evodevo

Evo-devo starts from the postulate that a causal-mechanistic interaction must exist between the processes of individual development and the processes of evolutionary change. Understanding these interactions and their consequences for organismal evolution represents the central research goal. Hence, the core question of evo-devo has two interrelated components: evolution's influence on development and development's influence on evolution. This reciprocal interrelationship constitutes a genuinely dialectical and systemic research agenda. The following will be a brief characterisation of the major research questions that arise from this general agenda.

How did development originate?

This question relates to the origins of multicellularity and the evolution of life cycles. John Bonner, a major influence in triggering the evo-devo revolution, early on reflected on the relations between organism size, internal complexity, reproductive success and life-cycle selection (Bonner 1965, 1988). Most of these ideas were based on the study of extant colonial or aggregating unicellular organisms such as cellular slime moulds. In early multicellular aggregates competition among cells to become the ones that propagate the next generation was possibly an important factor. The transition between the cell as the unit of selection and the multicellular individual as the unit of selection would have been the key evolutionary event at the origin ofdevelopment (Buss 1987).

A different approach targets the physical properties of cells and tissues. Single-cell organisms that existed before the emergence of multicellularity possessed liquid-like viscoelasticity, adhesiveness and chemical excitability. Consequently, protometazoan cell aggregates must have had an inherent capacity to self-organise spatial patterns. Development would have arisen at the point when certain cells achieved organisational control over other cells, e.g. by releasing diffusible chemical substances, and this capacity would have resulted in cell aggregates consisting of non-uniformly distributed cell states. In conjunction with differential adhesion (Steinberg 1963) and other generic physical mechanisms (Newman 1994) such simple systems can produce an array of 'generic forms', whose shapes and sizes are much determined by the physico-chemical conditions of the environment in which they form (Newman et al. 2006). Because of this strong environmental influence, it is assumed that in early forms of development the close correlation between genotype and phenotype observed in modern organisms would not have existed yet. Rather the genotype-phenotype relation might have been one-to-many during what has been called a 'pre-Mendelian phase' of evolution (Newman and Müller 2000). Subsequent selectional fixation and genetic routinisation would have resulted in the robust forms of development and the faithful Mendelian kind of inheritance seen in extant organisms.

How did the developmental repertoire evolve?

This question is predominantly approached at the genetic level, e. g. through the study of gene duplications, especially of the regulatory genes (McGinnis and Krumlauf 1992, Holland 1999), and the evolution of gene regulatory networks (Davidson et al. 1995, Wray and Lowe 2000). The genetic redundancy generated by such mechanisms can be exploited through the acquisition of new functions for these genes, a process referred to as recruitment (Keys et al. 1999) or cooption (True and Carroll 2002). Present summaries of the evolution of developmental pathways rely almost exclusively on genetics (Wilkins 2002, Carroll et al. 2005), but the epigenetic mechanisms controlling gene activation also evolve, including the processes of cell and tissue interaction and embryonic induction, which had been considered in earlier treatments of the evolutionary roles of epigenesis (Lovtrup 1974, Hall 1983, Edelman 1988).

Modularity constitutes a principle connecting the genetic and epi-genetic facets of evolving developmental repertoires in recognising that developmental systems are decomposable into components that operate according to their own intrinsically determined principles (Schlosser and Wagner 2004, Callebaut and Rasskin-Gutman 2005). Such modules can be characterised as integrated structural and process units that depend on input from other components and, in turn, influence other components by their outputs, represented, for example, by gene signalling pathways or inductive interaction networks. The evolutionary function of developmental modules would be their phenotypic selectability. A selectable developmental module can consist of a set of genes, their products and their developmental interactions, including the resulting character complex and the functional effect of that complex. The genes affecting the modular character complex would be characterised by a high degree of internal integration and a low degree of external connectivity: that is, pleiotropic connections would be largely within-module. Modularity could thus become one of the most productive approaches to the evolving genotype-phenotype relationship (von Dassow and Munro 1999).

How are established processes of development modified through evolution?

The empirical study of changes in developmental gene regulation occupies much of the present research effort (see below and contributions in this volume). A broader concept is heterochrony, i.e. evolutionary changes in the relative timing and rates of developmental processes. This classical idea has been revived by Gould (1977) and Raff and Kaufman (1983) and has since been elaborated into a powerful explanatory framework (McKinney and McNamara 1991, Parichy et al. 1992, McNamara 1997). Different forms and mechanisms of heterochrony are associated with different life-history strategies and produce different phenotypic results (Hall 1984, Raff and Wray 1989). Heterochrony has been documented in most groups of organisms, and its study is now taken to molecular and genetic levels (Parks et al. 1988, Wray and McClay 1989, Kim et al. 2000). Mutations that directly affect developmental timing have been demonstrated in animals (Ruvkun and Giusto 1989) and plants (Dudley and Poethig 1991). A number of genetic mechanisms affecting developmental timing have been tested experimentally (Dolle et al. 1993, Zakany et al. 1997). Without doubt heterochrony based on gene regulatory changes represents a powerful mode for altering morphological characters and body plans (Duboule 1994). But it remains difficult to distinguish between heterochronic phenomena that are simply a consequence of any change to development and those cases in which heterochrony of a particular process represents the causal mechanism for the evolutionary modification of a trait.

Does development play a role in phenotypic variation?

The extent to which the properties of developmental systems influence the variational and directional dynamics of phenotypic evolution is a question primarily addressed by the concept of developmental constraint. This was one of the themes that triggered evo-devo (Alberch

1982, Maynard Smith et al. 1985), and it is still relevant today. The empirical evidence for constraints is extensive, including data from comparative morphology (e.g. Wake 1982, Bell 1987, Vogl and Rienesel 1991, Caldwell 1994), comparative and experimental embryology (e.g. Alberch and Gale

1983, 1985, Müller 1989, Webb 1989, Streicher and Müller 1992), plant biology (e.g. Donoghue and Ree 2000) and quantitative genetics (e.g. Che-verud 1984, Rasmussen 1987, Wagner 1988). Whereas early conceptualisations of constraint concentrated on the limitations of phenotypic variation, later treatments emphasised also the heightened potential for change in particular aspects of the phenotypic character space (Arthur

2001). A taxon's capacity to generate heritable phenotypic variation or innovation will depend on mechanisms that reduce or overcome constraints, a controversial issue (Eberhard 2001, Wagner and Müller

2002). Much of the present work on plasticity and evolvability discussed below equally relates to the issue of constraint.

What is the contribution of development to the origin of phenotypic novelty?

Innovation and phenotypic novelty is one of the areas of evolutionary biology to which evo-devo could make a genuine contribution (Wagner 2000, Love 2003, Müller and Newman 2005b). While earlier conceptions concerning innovation were based on function shift (Mayr 1960), macromutation (Goldschmidt 1940) and symbiosis (Margulis and Fester 1991), evo-devo approaches concentrate on the role of development. One specific proposal is epigenetic causation, the idea that developmental systems do not merely transform genetic change into phenotypic change but also represent a generative component in phenotypic evolution (Müller 1990, Newman and Müller 2000). The starting point is the distinction between general selectional trends and the specificity of phenotypic response conferred by the developmental system. Selection acting on overall organismal features, such as shape, proportion, function or behaviour, can elicit epigenetic by-products that arise from the generic properties of developing cell and tissue systems, e.g. following changes in blastema size or mechanical load. New structural elements, skeletal parts for instance, can arise through this mode, without having been selected for, as a side-effect of the evolutionary modification of general developmental parameters (Müller 1990, Newman and Müller 2005). Thus, epigenetic mechanisms could have had a significant role in the origination of body parts and organismal form (Müller and Newman 2003, Love and Raff 2005).

A related approach is the origination of innovation through environmental induction (West-Eberhard 2003). This approach relies less on the physical properties of developmental systems, concentrating more on phenotypic plasticity and reaction norms as discussed below.

Does development affect the organisation of the phenotype?

The origin of higher-level organisational phenomena (homology, body plans) is one of the central questions of evo-devo (Raff 1996, Minelli

2003). Many of the new ideas on these topics were triggered by the discovery of the surprisingly high conservation of the gene regulatory apparatus in very diverse organisms. This has led to gene-based definitions of homology (Holland et al. 1996, Abouheif 1997), whereas others have pointed out the shortcomings of such reasoning (Bolker and Raff 1996, Minelli 1998). While the most notoriously conserved developmental control genes, the homeobox genes, exhibit non-homologous expression domains in vertebrate and invertebrate embryos, the reverse also applies: homologous structures can be specified by non-homologous genes (Wray 1999). New developmental concepts of hom-ology concentrate on commonalities of developmental pathways (Wagner 1989, 1996) or on the modularity of developmental systems (Minelli 1998, Gilbert and Bolker 2001). Other positions emphasise that the establishment of homology goes through different stages in which development has an important generative role, but eventually achieves independence from the underlying generative mechanisms (Müller 2003, Love and Raff 2005). Here the evolution of homology and body plans is viewed as a consequence of phenotypic integration that maintains the identity of building elements despite variation in their molecular, developmental and genetic makeup.

How does the environment interact with development and evolution?

Once thought of as crucial for understanding evolution, this question had been marginalised for several decades because of its seemingly Lamarckian connotations. But new data on the genetic and environmental aspects of developmental phenomena such as phenocopy, poly-phenism and plasticity have revitalised the interest in their evolutionary roles and led to proposals of an enlarged scope of evo-devo research (Gilbert 2001, Hall et al. 2003). The foundational concept in this domain is phenotypic plasticity (Pigliucci 2001, West-Eberhard 2003). It provides a unifying theoretical framework for the interpretation of quantitative genetic, developmental and morphological responses to environmental influences in evolving populations. The concept of plasticity is tightly interconnected with that of reaction norm (Schlichting and Pigliucci 1998, Sarkar 2003), i.e. the range of variation and phenotypes that can result from a single genotype as a response to different environmental conditions. Developmental plasticity, the mechanistic realisation of this responsiveness, is thought to represent in itself an adaptive trait of a taxon. But plasticity can also refer to evolutionary modifications of development that do not have a significant effect on the phenotypic outcome, a phenomenon often observed in species-level comparisons (Chipman 2002).

0 0

Post a comment