Field Homology A Useful Concept in Studies of Brain Evolution

The importance of morphogenetic fields as major units of development and evolution, and as major natural comparison characters for homology considerations makes the field homology concept very useful in evolution studies, including the brain evolution studies (Puelles and Medina, 2002). Each morphogenetic field gives rise to a specific set of derivatives found in the adult, and each organ, region, or subdivision found in the adult is the result of a specific field and its interactions with other fields, and can be formed with derivatives of several morphogenetic fields. In the brain, most cellular derivatives of each morphogenetic field remain in close association within specific radial histogenetic divisions, or in specific nuclei, areas, or cell condensations (such as patches or islands) within radial divisions. This is so because of the predominant radial glial-guided migration followed by immature neurons during development, in their way from the ventricular zone to the mantle (Rakic, 1972, 1995; Nieuwenhuys, 1998c; see 00116), although some subpopulations of cells undergo long-distance tangential migration (Anderson et al., 1997, 2001; Marin and Rubenstein, 2001). Another reason for the close association of cells derived from a particular morphogenetic field is that they tend to express similar combinations of cell adhesion molecules (such as cadherins) allowing their aggregation by homotypic binding, but distinct from those of adjacent fields (Redies et al., 2000). The prevalence of radial migration and the existence of radial histoge-netic domains in the brain has been observed using fate map studies of different brain fields (Marin and Puelles, 1995; Cobos et al., 2001a, 2001b; GarciaLopez et al., 2004). This is also observed after labeling of discrete cellular clones in the ventricular zone of the embryonic telencephalon, which produces mature cells that are aligned along a narrow band extending from the ventricle to the pial surface (Striedter et al., 1998; Noctor et al., 2001). The existence of radial histogenetic domains in the adult brain, whose cellular components are mostly derived from specific morphogenetic fields, allows the use of such radial divisions as the natural comparison units for homology considerations in the adult brain. Thus, specific radial histogenetic divisions found in the adult in different animals, and having the same morphogenetic origin and topolo-gical location, such as the dorsal thalamus, the striatal domain, or the derivatives of a particular rhombomere, can be considered homologous as a field (Puelles and Medina, 2002).

However, for further homology considerations, careful analysis of the different cell groups found in the radial histogenetic division must be done, since a division may contain both homologous and nonhomologous cell groups when compared to the cell groups in the same radial division of another vertebrate (Puelles and Medina, 2002). This can be due to the fact that, as noted above, homologous morphogenetic fields may undergo partially divergent evolution during late development, producing novel cell groups and features with no counterpart in other animals or in the ancestor. This is the case of the dorsal thalamus, which becomes very large in birds and mammals and contains many cell groups with no apparent counterpart in the dorsal thalamus of extant reptiles, or possibly in the common ancestor of extant birds, mammals, and reptiles (Butler, 1994a, 1994b). In addition, adult radial histogenetic divisions may contain immigrant cells that originate in different morphogenetic fields. These immigrant cells in a specific histogenetic region are not homologous to 'native' cells of the region, in the same animal or in other animals. However, they can be homologized to similar immigrant cells of the same region in other animals if they have an identical origin (and other similar features, such as neurotransmitter content) and can be traced back to the common ancestor (using cladistic analysis). An example of this is found in the striatum (Figure 1), a part of the basal ganglia found in the basal telencephalon of adult reptiles, birds and mammals, and derived from a homologous subpallial morphogenetic field (LGE in mammals), expressing Dlx family genes (Reiner et al., 1998; Marin et al., 1998; Smith-Fernandez et al., 1998; Puelles et al., 2000). The striatum in these animals contains homologous GABAergic projection neurons that originate in the LGE or LGE-like field, having specific neuro-peptide contents and connections (Reiner et al., 1998). In addition, the striatum in these animals contains similar sets of interneurons, such as the cholinergic interneurons (Reiner et al., 1998). In mouse, the striatal cholinergic interneurons immigrate from a different morphogenetic field of the subpallium, characterized by expression of the homeobox gene Nkx2.1 (Marin et al., 2000), and with the involvement of a Lhx7/8, a LIM-homeobox gene dowstream of Nkx2.1 (Zhao et al., 2003). This is possibly true for the striatal cholinergic interneur-ons of birds and reptiles as well. The cholinergic interneurons of the striatum are neither homologous to the striatal projection neurons nor to other groups of striatal noncholinergic interneurons, but - if their identical origin is confirmed - they could be homologous to the striatal cholinergic neurons of different reptilian, avian, and mammalian species.

In a few cases, homologous cell groups occupy very different locations in the adult, in different histogenetic radial divisions, although they originate in the same morphogenetic field. This is the case of the facial branchial motoneurons, which in reptiles, birds, and mammals originate in rhombo-meres 4 and 5, and their axons exit the brain in the 7th cranial nerve (facial nerve), through rhombo-mere 4 (Lumsden and Keynes, 1989; Medina et al., 1993; Medina and Reiner, 1994; McKay et al., 1997). However, after being born and starting to grow their axon, these motoneurons migrate caud-ally in reptiles and mammals (forming the facial nerve genu), to finally occupy a position at the level of rhombomeres 6-7 (Medina et al., 1993; Auclair et al., 1996; McKay et al., 1997). In birds, however, facial motoneurons stay within rhombomeres 4-5 and are seen at this same location in the adult (Medina and Reiner, 1994). In this case, the general segmental division derived from each rhom-bomere can still be homologized as a field between species, and also occupy comparable topological positions and relation to neighbors. When analyzing the specific cell groups, it is clear that, independent of their distinct position in the adult, the facial motoneurons of birds, reptiles, and mammals are homologous. Thus, for homology considerations of specific cell groups in the adult, it is extremely important to know the exact morphogenetic origin of the group (i.e., origin with respect to the morphogenetic fields found during development), and the origin is a prevalent criterion independent of the final location of the cells in the adult. Once an identical origin from homologous morphogenetic fields has been confirmed, other attributes can be considered for investigating the putative homology of specific cells found in the adult, such as neurotransmitter and/or neuropeptide content, connections, neurotransmitter receptor expression, etc. Connectivity and receptor features need to be considered with caution since these may have changed during the evolution of homologous cells (Striedter, 1998; see A Tale of Two CPGs: Phylogenetically Polymorphic Networks). In Search of the Brain Archetype in Vertebrates: Developmental Regulatory Genes as Useful Tools for Deciphering the Archetype and Identifying Homologous Fields

As noted above, morphogenetic fields exist in specific locations within the animal or organ/region, and this also applies to the brain. The brain of vertebrates shares a common basic organization plan (or archetype; also called 'Bauplan'; see Basic Nervous System Types: One or Many?, Evolution of the Deuterostome Central Nervous System: An Intercalation of Developmental Patterning Processes with Cellular Specification Processes) where the morphogenetic fields and their derivatives are located. This basic organization plan is better appreciated during development, and developmental regulatory genes, combined with other data such as fate mapping results, are very useful tools for trying to unravel it (Puelles and Rubenstein, 1993, 2003;' Puelles and Medina, 2002; see Evolution of the Deuterostome Central Nervous System: An Intercalation of Developmental Patterning Processes with Cellular Specification Processes, Origins of the Chordate Central Nervous System: Insights from Hemichordates). As noted above, these master control genes show highly evolutionarily conserved sequences, and encode transcription factors or signaling proteins that regulate expression of other genes, thus controlling patterning and morphogenesis of specific body parts (Carroll et al., 2001). They also show generally conserved expression patterns in the embryonic brain, and restricted expression domains comparable across species (Smith-Fernandez et al., 1998; Puelles et al., 2000; Murakami et al., 2001; Bachy et al., 2002; Brox et al., 2003, 2004). These genes help in understanding important organization features of the developing brain, such as its curved longitudinal axis, the existence of longitudinal and transverse (segmental) divisions, and the existence of smaller subdivisions, each one in a specific location within the general plan. Expression of these genes indicates that, during development, the brains of different vertebrates pass through a highly similar stage (a kind of phylotypic stage), in which the embryonic brain shows clearly comparable molecularly distinct divisions and subdivisions, and a common archetype. These divisions and subdivisions represent distinct morphogenetic fields, and each one is characterized by the expression of a specific combination of master control genes. Thus, these genes are very valuable tools for locating these morphogenetic fields within the general brain archetype, and for finding homologous morphogenetic fields in the brain of different vertebrates (Puelles and Medina, 2002). Understanding this is very important for later studying the derivatives of each field in the adult, for searching cases of field homology and cell group homology in the adult, and for understanding how morphological divergence occurs in evolution. Evolution of Homologous Fields in the Brain: The Case of the Pallium

In the brain, homologous morphogenetic fields can give rise to adult homologous structures showing a high degree of evolutionary conservation, such as the striatum (Figure 1; Reiner et al., 1998; Marin et al., 1998; see The Evolution of the Basal Ganglia in Mammals and Other Vertebrates) or, often, give rise to structures that are homologous as a field in different vertebrates, but that show a high degree of variation between species though keeping a few basic common features (e.g., the neocortex of mammals, the hyperpallium of birds, and the dorsal cortex of reptiles, all of which derive from the dorsal pallium; Medina and Reiner, 2000; see Do Birds and Reptiles Possess Homologues of Mammalian Visual, Somatosensory, and Motor Cortices?, The Origin of Neocortex: Lessons from Comparative Embryology). Within the pallium, there are striking differences between animal groups. For example, the dorsomedial pallium shows an extraordinary development in mammals (in particular, its dorsal pallial derivative, the neocortex) but, in contrast, the ventrolateral pallium shows the greatest development in reptiles and especially in birds, giving rise to a large structure called dorsal ventricular ridge (Butler, 1994b; Striedter, 1997; see Evolution of the Nervous System in Reptiles, Do Birds and Reptiles Possess Homologues of Mammalian Visual, Somatosensory, and Motor Cortices?, The Evolution of Vocal Learning Systems in Birds, Evolution of the Amygdala in Vertebrates, The Origin of Neocortex: Lessons from Comparative Embryology). One of the main factors that may be involved in such variations is the possible existence of changes in the networks of developmental regulatory genes that operate in fore-brain development (Carroll et al., 2001; Gilbert and Burian, 2003). Indeed, numerous developmental regulatory genes play key roles in patterning, specification, cell proliferation, and/or cell differentiation of specific parts in the central nervous system (from the spinal cord to the telencephalon), and changes in their expression (e.g., by knockout mutation producing a lack of function) lead to morphological alterations that include changes in the relative size of regions and lack of formation of particular sets of neurons, among other changes (Ericson et al., 1997; Briscoe et al., 1999; Sussel et al., 1999; Yun et al., 2001, 2003 Bishop et al., 2002, 2003; Muzio et al., 2002). An interesting observation is that the relative size of one particular region (radial histogenetic division) can be affected by both genes expressed in its own primordium during development, or genes expressed in adjacent morphogenetic fields, and it appears that several patterning genes play key opposing roles in establishing the relative size of regions. This occurs in the telence-phalon, where the relative size of the pallium is enlarged by knockout mutations affecting homeobox genes expressed in the subpallium (such as Gsh2), and is reduced by knockout mutations affecting homeo-box genes expressed in the pallium (such as Pax6) (Stoykova et al., 1996, 2000; Toresson et al., 2000; Yun et al., 2001). Any alteration in the equilibrium between genes having opposing roles in specifying subpallial versus pallial territories (either by overexpression or by lower expression of one of the two groups) may have led to the very large pallium observed in mammals and birds.

Within the pallium, the opposing roles of Emx1/2 versus Pax6 (expressed in opposing gradients in the pallium) can affect the relative size of caudomedial versus rostrolateral pallium and the specific areas in them (Bishop et al., 2002, 2003). This, in fact, is affecting the relative size of the medial pallium

(displaced caudally in mammals) and adjacent parts of dorsal pallium versus the rest of the pallium, including the ventrolateral pallium or part of it (i.e., piriform cortex and claustrum; Stoykova et al., 1996). In addition, some of the genes involved in internal pallial patterning and specification, such as Emxl and Emx2, also have a role in cell proliferation, thus affecting growth and size of the pallial areas where they show higher expression (Bishop et al., 2003), and the level of expression of these genes can be affected by signaling proteins that diffuse from organizer centers, such the cortical hem located at the edge of the medial pallium (Ragsdale and Grove, 2001; Garda et al., 2002). Other developmental regulatory genes (such as the LIM-homeobox genes and several LIM cofactor genes) are expressed in combinatorial patterns in the pallium and other parts of the forebrain, and - as in the spinal cord - may play key roles in the differentiation of specific sets of neurons and in the establishment of specific connections (Retaux et al., 1999; Bulchand et al., 2001, 2003; Moreno et al., 2004). In addition, one of these genes (Lhx2) plays a role in the formation of the dorsomedial pallium and its boundary with the cortical hem, and lack of its function produces severe malformation of the dorsomedial pallium (affecting specially the medial pallium), without affecting any part of the ventrolateral pallium, which expresses the antagonist LIM protein Lmo3 (Bulchand et al., 2001; Vyas et al., 2003).

This provides an idea of only some of the complex and important networks of regulatory genes operating in pallial development (as in the development of other brain regions). Any alteration in the expression of these regulatory genes within or near the pallium may have led to the differences observed within the pallium between mammals and birds. Further, the pallium has important reciprocal connections with the thalamus, and a recent study indicates that developing thalamocortical axons releases a mitogenic factor that increases cell proliferation of pallial cells, indicating that epigenetic influences from distant regions can also modulate pallial development (Dehay et al., 2001). Thus, alterations affecting the number or the pathfinding of thalamocortical axons may influence the final size of cortical areas as well. Nevertheless, this does not affect the initial specification and area formation of the pallium, which is due to intrinsic factors, mediated by locally expressed developmental regulatory genes (Miyashita-Lin et al., 1999). Until now, developmental evolutionary neurobiologists have focused on analyzing similarities in expression pattern of a number of these regulatory genes, and this has been extremely valuable for understanding the brain archetype and finding homologous fields. In the future, we will have to center on investigating what differs in brain development of different animals, by searching for differences in expression patterns of some master control genes and by searching changes at the level of promoter/enhancer, activators, repressors, and cofactors regulating the expression of master control genes (Medina et al., 2005). This will help us understand what exact developmental mechanisms led to the morphological variations found in the brain of different vertebrates. Evolution of New Fields in the Vertebrate Brain: Analysis of the Lamprey

Although most vertebrates share a basic brain archetype, it appears that early vertebrates (see Evolution of the Deuterostome Central Nervous System: An Intercalation of Developmental Patterning Processes with Cellular Specification Processes) lacked at least some of the whole set of morphogenetic fields found in the forebrain of tetrapods, suggesting that some new fields evolved after vertebrates first appeared (Medina et al., 2005). This is based on analysis of the lamprey, a jawless fish close to the origin of vertebrates. During development, the forebrain of these animals show some of the same molecularly distinct divisions found in tetrapods, including comparable diencephalic divisions/subdivisions, and comparable pallial and subpallial divisions in the telencephalon (Murakami et al., 2001). In the tele-ncephalon of developing lamprey, the subpallium is characterized by Dlx expression, whereas the pallium expresses Pax6 and Emxl genes, showing patterns comparable to those observed in other vertebrates. However, the subpallium does not show any expression of the pallidal marker gene Nkx2.1, suggesting that these animals lack the morphoge-netic field giving rise to the pallidum (Murakami et al., 2001; this gene is also called TTF-1), which agrees with the apparent absence of this structure in adult lampreys (Pombal et al., 1997a, 1997b). In zebra fish, a teleost jawed fish, the subpallium includes both striatal and pallidal subdivisions, expressing either only Dlx or both Dlx and Nkx2.1 orthologue genes (Rohr et al., 2001), suggesting that the pallidal subdivision appeared in the telencephalon as a novel morphogenetic field during the transition from jawless to jawed vertebrates (Medina et al., 2005). This example indicates that novel morphogenetic fields can evolve by appearance of novel expression domains of developmental regulatory genes (also involving a cascade a downstream genes), and these novel fields can lead to the production of novel histogenetic divisions and cell groups in the adult. Understanding what exact mechanisms lead to the appearance of novel expression domains of master control genes during development will be essential for understanding evolution of novel structures in the brain.

For homology and evolutionary considerations, another way of looking at this is by analyzing the hierarchy of the fields, and how addition of a new field can affect the rest. The whole embryo is considered to be the primary field, being later subdivided into secondary and then tertiary fields, and so on (Gilbert et al., 1996; Carroll et al, 2001). If we arbitrarily consider the telencephalic vesicle as a 'primary' field (to simplify counts), then pallium and subpallium are 'secondary' fields (Figure 1). Within the subpallium, the appearance of a Nkx2.1 expression domain divides it into two 'tertiary' fields (striatal and pallidal). (This is an oversimplification, since - atleast in tetrapods - the Nkx2.1 expressing domain of the subpallium is further subdivided into at least two fields: a 'true' pallidal field expressing Nkx2.1 and Dlx genes, and an anterior entopedun-cular field expressing Nkx2.1, Dlx plus sonic hedgehog (Puelles et al., 2000; Marin and Rubenstein, 2002).) At a certain point of development, the lamprey subpallium can be considered homologous to that of jawed vertebrates as 'secondary' morphogenetic fields. They give rise to subpallial divisions in the adult that can be considered homologous as a field, but have the following major differences. After subpallial subdivision into 'tertiary' fields, the Nkx2.1 expressing pallidal field of jawed vertebrates is not homologous to any field in the lamprey. The striatal 'tertiary' field (expressing only Dlx genes) of jawed vertebrates remains homologous to the single subpallial field of lamprey, and both give rise to a homologous striatum (Figure 1). Nevertheless, the striatum produced in jawed vertebrates is the result of both similar plus newly evolved interactions of its primordium with other modules, such as the novel Nkx2.1 expressing field. For this reason, the resultant striatum of jawed vertebrates has some new features with no counterpart in the lamprey striatum. The new features of the striatum can even go further if immigrant cells coming from the newly evolved field arrive in the striatum and integrate establishing connections with native cells, as it occurs with striatal interneurons in amniotes (see above) and in some amphibians (Figure 1). This involves an evolutionary increase in both morphological and functional complexity of the striatum. Thus, novel fields can have consequences that go further than just producing a new histogenetic territory, since they can influence, and trigger, important morphological and functional changes in adjacent and/or distant fields.

Was this article helpful?

0 0

Post a comment