© 2001 by The Society for Integrative and Comparative Biology
Homeobox Genes, Retinoic Acid and the Development and Evolution of Dual Body Plans in the Ascidian Herdmania curvata1
1 Department of Zoology and Entomology, University of Queensland, Brisbane, Queensland 4072, Australia
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Ascidians, along with other urochordates, are the most evolutionary distant group from vertebrates to display definitive chordate-specific characters, such as a notochord, dorsal hollow nerve cord, pharynx and endostyle. Most solitary ascidians have a biphasic life history that has partitioned the development of these characters between a planktonic microscopic tadpole larva (notochord and dorsal nerve cord) and a larger sessile adult (pharynx and endostyle). Very little is known of the molecular axial patterning processes operating during ascidian postlarval development. Two axial patterning homeobox genes Otx and Cdx are expressed in a spatially restricted manner along the ascidian anteroposterior axis during embryogenesis and postlarval development (i.e., metamorphosis). Comparisons of these patterns with those of homologous cephalochordate and vertebrate genes suggest that the novel ascidian biphasic body plan was not accompanied by a deployment of these genes into new pathways but by a heterochronic shift in tissue-specific expression. Studies examining the role of all-trans retinoic acid (RA) in axial patterning in chordates also contribute to our understanding of the role of homeobox genes in the development of larval and adult ascidian body plans. Our studies demonstrate that RA does not regulate axial patterning in the developing ascidian larval neuroaxis in a manner homologous to that found in vertebrates. Although RA may regulate the expression of some ascidian homeobox genes, ectopic application of RA does not appear to alter the morphology of the larval CNS. However, treatment with similar or lower concentrations of RA, have a profound effect on postlarval development and the juvenile body plan. These changes are correlated to a dramatic reduction of Otx expression. Through these RA-induced effects we infer that while RA may regulate the expression of some homeobox genes during embryogenesis it has a far more dramatic impact on postlarval development where regulative processes predominate.
|
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Comparisons of the roles of homeobox genes in metazoan development are playing a pivotal role in the current resynthesis of developmental and evolutionary biology (e.g., Hall, 1992
; Raff, 1996; Gerhart and Kirschner, 1997). A great challenge is to understand how these conserved genes regulate the development of morphologically dissimilar taxa. Comparative developmental studies of chordates have demonstrated that these regulatory genes are expressed in a conserved manner across the spectrum of body plans displayed by urochordates, cephalochordates and vertebrates (e.g., Holland, 1991; Katsuyama et al., 1995; Wada et al., 1995, 1996b, 1998, 1999a; Holland and Holland, 1996; Gionti et al., 1998; Glardon et al., 1997; Yasuo and Satoh, 1998; Wada and Saiga, 1999; Hinman and Degnan, 2000; Hinman et al., 2000). This is particularly striking given that the modes of development of these taxa are markedly different. For example, most solitary ascidians (Urochordata: Ascideacea) rely heavily on cell-lineage based mechanisms to form a tadpole larva with only a small number of cells and tissues. Conversely, vertebrates rely on more conditional specification to pattern large numbers of migrating cells (e.g., see Davidson, 1990, 1991). The ascidian tadpole, while possessing a notochord and dorsal hollow nerve cord which are both chordate-specific characters, is morphologically simple compared to vertebrates and has little regional specialisation.
In addition to a variety of developmental modes, the chordate subphlya also exhibit a variety of life histories. Ascidians and thaliaceans (which, along with the larvaceans, comprise the subphylum Urochordata) display a biphasic life-history with morphologically dissimilar larval and adult body plans (e.g., Berrill, 1950). Most solitary ascidians develop microscopic, free swimming larvae with axial notochord, nerve cord and musculature. During metamorphosis, the axial structures are broken down and the larval endoderm and mesenchyme rudiments proliferate and undergo morphogenesis to form the adult body wall musculature, haemopoietic system and functional gut, which includes a pharyngeal basket (homologue of gill slits in amphioxus and vertebrates) and endostyle (homologue of the thyroid gland) (e.g., Cloney, 1982; Hirano and Nishida, 1997, 2000). Thus in ascidians and salps, unlike all other chordates, the development of phylotypic characters (i.e., notochord, dorsal hollow nerve cord, pharynx and endostyle) are temporally dissociated. While biphasic life histories are the most common form of development among the metazoans (e.g., Davidson, 1991; Peterson et al., 2000), this aspect of development has been largely overlooked in recent molecular developmental analyses because most model organisms are direct developers. This poses the important and largely unaddressed question of how developmental regulatory genes, such as homeobox genes, are used to pattern two distinct body plans—larval and adult.
The life history diversity within the urochordates has resulted in considerable debate as to the nature of the ancestral chordate and whether or not it developed indirectly (e.g., Garstang, 1928; Berrill, 1955; Romer, 1959, 1962; Jefferies, 1986). Recent phylogenetic analyses, suggest that the chordate ancestor was free-living and directly developing, i.e., that it developed axial locomotory and feeding structures concomitantly (Wada and Satoh, 1994; Satoh and Jeffery, 1995; Wada, 1998; Swalla et al., 2000). This supposition is largely based on data which suggests that, within the urochordates, the directly developing larvaceans are most primitive and the other two chordate subphyla are exclusively directly developing. That larvaceans are most primitive within the urochordates is, however, still open to debate as the data are not inconsistent with this taxa being placed as a divergent outgroup. If we do accept that the chordate ancestor was directly developing, then evolution along the ascidian and salp lineage is characterised by an increased dissociation of the development of locomotory from feeding structures. Furthermore then, comparisons of developmental processes in urochordates, cephalochordates and vertebrates present an opportunity to examine evolutionary processes resulting in dissociation and to understand the modular nature of the ancestral chordate body plan.
|
ASCIDIANS: DEVELOPMENT OF TWO BODY PLANS |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Embryogenesis of solitary ascidians is characterized by invariant cell cleavage and development with only a small number of cells (e.g., gastrulation commences with 110 cells and the fully developed tadpole larva has only around 2,500 cells in total; Satoh, 1994
). The fate of most tissues is determined prior to gastrulation, and induction, when it occurs, is only across small distances (e.g., reviewed Meedel, 1992; Satoh, 1994; Whittaker, 1997). Consequently, the tadpole larva has little tissue complexity and minimal axial regionalisation. The anterior neural tube is not morphologically subdivided into fore, mid or hindbrain and the axial mesoderm has essentially no morphological specialization (Nicol and Meinertzhagen, 1991). After hatching, ascidian tadpoles typically do not undergo much further morphogenesis until they receive an appropriate environmental cue to commence metamorphosis and the development of the adult body plan (e.g., Degnan et al., 1996, 1997; Eri et al., 1999). Little is known of cell patterning mechanisms during this phase of ascidian development and how metamorphic processes are influenced by embryogenesis. Ascidians do not show an extreme form of indirect development as some tissues of the larvae are resculptured during the formation of the adult and primordia are specified during embryogenesis. Unlike many indirectly developing invertebrates, the ascidian biphasic life cycle appears to be derived from a directly developing ancestor. This means that the mode of evolution of the larval and adult body plans of the ascidian is likely to be fundamentally different from these other invertebrate taxa. In ascidians, temporal dissociation of morphogenetic modules is likely to be a central mechanism underlying their origin and evolution.
Cloney (1982) has provided a clear description of the types of organs in ascidians. Transitory larval organs (the notochord, axial musculature and the dorsal nerve cord) are fully differentiated and function only in the larva and are lost during metamorphosis. Prospective juvenile organs (the branchial and atrial siphons, all endodermally derived structures of the gut and most of the adult nervous system) exist as rudiments in the larva and undergo morphogenesis and differentiation during postlarval development, and function in the juvenile. Only a few structures are differentiated and function in both the larval and juvenile body plans. These include the epidermis and inner tunic and some mesenchyme derivatives which have differentiated into blood cells in the larva. Hirano and Nishida (2000) have demonstrated through cell labelling experiments that the fate of cells from the larval endodermal rudiment (i.e., pharynx and gut primordia) is variable, suggesting some conditional specification. Also, incomplete tadpoles (formed when one of the cells in a 2-cell embryo is ablated) give rise to normal juveniles (Nakauchi and Takeshita, 1983). Together, these data suggest that regulative processes are important during ascidian metamorphosis and that the development and patterning of transitory larval organs and adult organs are not interdependent.
|
HOMEOBOXES AND DISSOCIATIONS ACROSS BODY PLANS |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Many of the gene products thought to pattern the ascidian neural tube during embryogenesis are also expressed (as shown by RT-PCR) during postlarval development and in the juvenile (e.g., ascidian homologues of Otx, Hox5, Evx, Pax2/5/8, Cdx, Hox1; Katsuyama et al., 1995
; Hinman and Degnan, 2000; Hinman, 2000; Hinman et al., 2000). These therefore may function in establishing anteroposterior identity in both body plans.
The expression of Otx (Hec-Otx) and Cdx (Hec-Cdx) homeobox genes has been examined during embryogenesis and metamorphosis in the ascidian Herdmania curvata (Hinman and Degnan, 2000; Hinman et al., 2000). Otx genes have been well characterised in a range of disparate metazoan taxa, and are almost ubiquitously expressed in the anterior neuroectoderm and ectoderm (Finkelstein and Boncinelli, 1994; Bruce and Shankland, 1998; Wada et al., 1998; Williams and Holland, 1998; Rhinn et al., 1998; Umesono et al., 1999). Expression is also detected variously in visceral endoderm, and anterior mesoderm and endoderm of the gut and pharynx in some vertebrates (Simeone et al., 1993, 1995; Ang et al., 1994; Kablar et al., 1996; Rhinn et al., 1998; Tomsa and Langeland, 1999) and amphioxus (Williams and Holland, 1998). During embryogenesis, the ascidian homologue of Otx is localised to the anterior CNS, the ectodermal stomodaeal primordium and to the anterior ectoderm (Fig. 1A; Wada et al., 1996b; Hinman and Degnan, 2000). In the juvenile, Hec-Otx is localised predominantly to anterior stomodaeal and endodermal pharyngeal structures; including the branchial tentacles, prebranchial groove, anterior third of the endostyle and dorsal edges of five of the forming branchial arches (Fig. 1B, C; Hinman and Degnan, 2000).
|
In a variety of taxa, Cdx genes are involved in general posterior patterning and, in particular, are expressed in the posterior endoderm when gut development is initiated at gastrulation (Macdonald and Struhl, 1986
; Mlodzik and Gehring, 1987a, b; Gamer and Wright, 1993; Morales et al., 1996; Marom et al., 1997; Isaacs et al., 1998). In vertebrates and amphioxus, Cdx genes are also expressed in a gradient in the posterior of the developing neural tube (Gamer and Wright, 1993; Marom et al., 1997; Brooke et al., 1998; Pillemer et al., 1998). The ascidian homologue of Cdx is localised to the posterior neural tube and posterior ectoderm of the neurula and tailbud embryo (Fig. 1D; Katsuyama et al., 1999; Hinman et al., 2000). In the juvenile, Hec-Cdx is localised to two regions of the intestine, with expression highest in the most posterior intestinal cells, just anterior of the anus and more anterior intestinal cells, close to the stomach-intestine junction (Fig. 1E, F; Hinman et al., 2000).
In vertebrates and amphioxus, homologues of Otx are expressed concomitantly in anterior neuroectoderm and in the foregut, pharynx and endoderm of the anterior branchial pouches (e.g., Simeone et al., 1995; Rhinn et al., 1998; Williams and Holland, 1998). Vertebrate and amphioxus homologues of Cdx are expressed concomitantly in the posterior gut, posterior neural tube and often in posterior mesoderm (e.g., Gamer and Wright, 1993; Beck et al., 1995; Brooke et al., 1998). The combined expression patterns of Hec-Otx and Hec-Cdx during embryogenesis and metamorphosis of the ascidian Herdmania curvata appears to be homologous to these vertebrate and cephalochordate patterns except that the neural and endodermal expression is temporally separated. Also, unlike the vertebrate and cephalochordate genes, the ascidian homologues are not detected in mesodermal tissues. There is no expression of either Hec-Otx or Hec-Cdx in the embryonic endoderm or the juvenile nervous tissue. Thus in ascidians, neuroectodermal/ectodermal and endodermal expression of these developmental regulatory genes is separated by metamorphosis. They are presumably involved in the development of these respective tissues in the two body plans. The expression data for Hec-Cdx and Hec-Otx suggest therefore, that the generation of the novel ascidian biphasic body plan was not accompanied by a deployment of these genes into new pathways but by a heterochronic shift in tissue-specific expression.
|
ANCESTRAL MODULARITY |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
If we assume that chordates were primitively direct developers, such that the ascidian biphasic development is derived, then the ancestral chordate developed a fully differentiated gut and axial locomotory system concurrently. Presumably then, developmental regulatory genes such as Otx and Cdx were expressed in these tissues at the same time. Such expressions are found in extant vertebrates and amphioxus. The dissociation of locomotory from feeding structures in ascidians therefore probably coincided with a "silencing" of regulatory genes such as Hec-Otx and Hec-Cdx in the endoderm during embryogenesis with a maintenance of their expression in the neuroectoderm/ectoderm.
The ability to temporally separate the tissue-specific expression of patterning genes implies a modularity of endodermal and neuroectodermal patterning in the ancestor. Such notions of modularity were noticed on an anatomical level by Romer (1959, 1962) who observed a distinctiveness in the development of somatic systems (i.e., notochord, nerve cord) and visceral systems (i.e., the gut) in vertebrate embryos which he interpreted as a reflection of an ancestral temporal separation. Regardless of whether the chordate ancestor was biphasic or developed directly, evolution among chordates was likely to have been facilitated by the modular nature of development in the ancestor as this would have allowed heterochronic shifts between somatic and visceral structures and their underlying molecular regulatory pathways.
It is of interest to consider whether changes in the timing of the tissue-specific expression of developmental regulatory genes such as Otx and Cdx caused the establishment of a biphasic lifestyle or was simply a result of this. In either scenario it is likely that evolutionary changes were either to the cis-regulatory regions of these genes or to the transcription factors acting upstream of them. If the cis-regulatory elements of Hec-Otx and Hec-Cdx rearranged to respond differently to the same upstream trans-factors this would suggest that these genes were the target of evolutionary change. Conversely, if evolutionary changes were to the upstream trans-acting factors this would imply that heterochronic shifts in Hec-Otx and Hec-Cdx expression were preceded by other, upstream changes. Comparisons of the expression and regulation of genes such as Otx and Cdx in extant ascidian species may provide a means to test these alternatives. In extant ascidians, there is a range of heterochronic shifts in development in which some adult structures (particularly the pharynx, parts of the gut and siphons) form in the tadpole. There is no apparent phylogenetic pattern to these developmental changes, as they have arisen or been lost independently in several lineages, with compound (colonial) ascidians that brood embryos usually having larvae with more developed endodermal structures (Wada et al., 1992). The varying degrees of temporal shifts in pharynx, gut and siphon development in extant ascidians suggest that these structures have remained essentially autonomous modules in this group of chordates. Comparisons of the timing of expression of regulatory genes such as Otx and Cdx in the neuroectoderm and endoderm in species of ascidian which vary their relative rates of development of these tissues and an examination of whether any heterochronies correlate with either cis- or trans-factors will contribute to the hypotheses discussed above.
Regardless of the extent of endodermal development in the ascidian larvae, this morphogenetic program is not completed until the larva receives an appropriate environmental cue (e.g., Degnan et al., 1997). Development of the endoderm and other adult systems is therefore ultimately under the regulation of environmental signalling. This does not necessarily imply that the evolutionary origin of such tissue dissociation was through regulation by environmental triggers, but at some point in ascidian evolution, signalling systems that were originally regulated by intrinsic factors became reliant on environmental cues.
|
HOMEOBOX GENES AND EMBRYONIC AXIAL PATTERNING |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Almost all comparative molecular developmental biology using ascidians has focused on the development of the larval body plan, in particular the neural tube and notochord. The majority of these studies have demonstrated a conservation of patterning gene expressions between ascidians, amphioxus and vertebrates (e.g., Katsuyama et al., 1995
; Wada et al., 1995, 1996b, 1998, 1999a; Corbo et al., 1997; Gionti et al., 1998; Yasuo and Satoh, 1998). For example, the ascidian homologues of Dlx, Otx, Pax2/5/8, Lim1, Hox1, Hox3, Hox5, and Cdx are expressed in anterior to posterior domains in the ascidian neural tube during neurulation and/or tailbud formation reminiscent of the expression of the vertebrate homologues of these genes (Katsuyama et al., 1995; Wada et al., 1995, 1996b, 1998; Gionti et al., 1998; Locascio et al., 1999; Hinman and Degnan, 2000; Hinman et al., 2000). These data suggest that a similar genetic program is used to pattern chordate neural tubes and hence may be assumed to have been present in the ancestral chordate. It is of interest to consider how such conservation was maintained in concert with the evolution of the diverse developmental modes exhibited by extant chordates. In ascidians and other urochordates there is a heavy reliance on intracellular determinants and stereotypic cleavage patterns during embryogenesis to create a morphologically simple larvae. Conversely, vertebrates rely on a comprehensive cell-cell signalling system to produce morphologically complex embryos exhibiting extensive regionalisation.
In vertebrates, the spatial domains of expression of genes such as Otx2, Pax2, Lim1, engrailed2, and Hox1 define distinct regions of the brain and, in particular, the midbrain/hindbrain (MBHB) region (e.g., reviewed Joyner, 1996). This region is thought to act as an organiser for anterior brain patterning and to be a vertebrate specific innovation (McMahon and Bradley, 1990; Rowitch and McMahon, 1995; Joyner, 1996; Bally-Cuif and Boncinelli, 1997). The particular expression domains of Otx, Pax2/5/8, Lim1, and Hox1 in the anterior of the ascidian neural tube have also been used to suggest that, at a molecular level, the ascidian cerebral vesicle may be tripartite (Wada et al., 1998) although morphologically there is no regionalisation into fore, mid and hind brain (Nicol and Meinertzhagen, 1991). Interestingly, the amphioxus engrailed and Pax2/5/8 gene homologues are not expressed in an equivalent MBHB region (Holland et al., 1997; Kozmik et al., 1999). Wada et al. (1996a) have also suggested that the reiterated expression pattern of the ascidian Pax 3/7 gene in the nerve cord of the tadpole larva, despite the lack of overt morphological segmentation, is a remnant from a metameric ancestral chordate. While assessments of homology based on molecular expression are still areas of debate (e.g., Dickinson, 1995; Holland et al., 1996, 1999; Abouheif et al., 1997) these combined molecular data from ascidians suggest two possible scenarios. Either, that a molecular distinction in the absence of morphological regionalisation represents maintenance of a genetic regulatory system that was used to pattern a morphologically more complex ancestor. Alternately, these data suggest that a genetic regionalistion was present in the common ancestor and was only utilised for overt morphological specialisation during the evolution of vertebrates.
|
RETINOIC ACID AND AXIAL PATTERNING PROCESSES IN VERTEBRATES AND THE ASCIDIAN LARVA |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Studies examining the role of all-trans retinoic acid (RA) in axial patterning in chordates also contribute to our understanding of the role of homeobox genes in the development of larval and adult ascidian body plans. It has been well documented, that during vertebrate gastrulation and neurulation, RA acts as, or mimics the effects of a morphogen and regulates the expression of a number of axial patterning homeobox genes (Morriss-Kay et al., 1991
; Kessel and Gruss, 1991; Chen et al., 1992; Hogan et al., 1992; Kessel, 1992; Ang et al., 1994; Marshall et al., 1994; Conlon, 1995; Simeone et al., 1995; Avantaggiato et al., 1996). It is suggested that RA emanates from the gastrulation organiser (e.g., Hensen's node) and provides positional identity to cells as they migrate through the blastopore (Chen et al., 1992; Hogan et al., 1992; Marshall et al., 1994).
In vertebrates, ectopic application of RA results in the misexpression of a number of homeobox genes which leads to defects including the loss of anterior structures, such as the forebrain and midbrain (Durston et al., 1989; Morriss-Kay et al., 1991; Kessel, 1992). For example, in the developing neural tube of the mouse, the anterior boundary of Hox1 (e.g., Hoxb-1) and the posterior boundary of Otx2 expression are shifted anteriorly, extending Hox1 into and Otx2 out of the MBHB region (Ang et al., 1994; Conlon, 1995 and references therein; Simeone et al., 1995; Avantaggiato et al., 1996). RA-treatment at slightly later stages of development leads to a loss of Otx2 expression only in the foregut and prechordal plate and a less severe phenotype (Simeone et al., 1995). This is an important observation as it demonstrates that neuroectodermal tissue is respecified although Otx2 expression is only affected in the underlying tissues, not in the neuroectoderm itself. Also, Pax2 expression in the hindbrain shifts rostrally to fuse with its anterior domain in the midbrain (Avantaggiato et al., 1996). These data suggest that RA regulation of anteroposterior patterning in vertebrates is intimately associated with features specific to vertebrate development. That is, RA regulates anteroposterior patterning through linking a temporal progression of an anterior to posterior placement of cells with a temporal increase in concentration. Furthermore, RA may affect vertical transactivating signals; RA levels in the mesoderm and endoderm may induce fates in neuroectoderm so that RA-induced loss of anterior neuroectoderm may be a consequence of underlying inductive cues. Correspondingly, RA does not appear to have a role regulating axial patterning homeobox genes in non-chordates (Müller, 1984; Oro et al., 1990; Créton et al., 1993; Imsiecke et al., 1994; Sciarrino and Matranga, 1995).
Initial observations indicated that, during ascidian embryogenesis, application of ectopic RA does lead to an apparent loss of anterior structures in ascidian larvae (DeBernardi et al., 1994; Katsuyama et al., 1995). This may be related to an RA-induced expansion of Hox1 (HrHox-1) gene expression (Katsuyama et al., 1995). However, more recent analyses have shown that the anterior CNS of the ascidian larva is not lost or respecified by RA treatment (Hinman and Degnan, 1998, 2000). Also, the expression of the ascidian Herdmania curvata Otx (Hec-Otx) and Pax2/5/8 (Hec-Pax2/5/8) genes in the CNS are not disrupted by ectopic RA in a manner similar to the perturbation of their vertebrate homologues (Hinman and Degnan, 2000). Instead, Hec-Otx expression is lost from the very anterior limit of its normal expression, which in the ascidian represents the primordium of the adult stomodaeum. Therefore, while there does appear to be some regulation of Hec-Otx by RA (whether direct or indirect), this appears to be in primordia of adult tissues rather than in the larval CNS. There are several other lines of evidence that RA has an endogenous role during ascidian embryogenesis. An ascidian homologue of an RA receptor, Hec-RAR, is expressed during embryogenesis of H. curvata (Devine et al., unpublished) and a retinoic acid responsive element drives reporter gene expression during embryogenesis of the ascidian Ciona intestinalis (Hisata et al., 1998).
While the interpretation of the effect of RA on ascidian homeobox gene expression and larval morphology is not straightforward, it appears that RA does not regulate axial patterning in the developing ascidian neuroaxis in a manner homologous to that found in vertebrates. Regulation may exist for some genes (e.g., HrHox1, Katsuyama et al., 1995 and non-neural expression of Hec-Otx, Hinman and Degnan, 2000) but this does not appear to result in homologous morphological alterations to the larval CNS. The ascidian data when combined with that from vertebrates suggest one of two evolutionary hypotheses. First, RA regulated the expression of some homeobox genes during development of the neural tube in the chordate ancestor and that during the evolution of ascidians some regulatory connections were rendered superfluous. This may have correlated with a decreased reliance on intercellular communication during embryogenesis concomitant with an increased reliance on cytoplasmic determinants. Alternately, the partial regulation exhibited presently in ascidians may represent the ancestral state which during the development of vertebrates, and as a result of an increased reliance on intercellular communication, became more extensive.
|
AN ANCESTRAL ROLE FOR RETINOIC ACID PATTERNING IN CHORDATES |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
While ectopic RA results in little morphological perturbation during embryogenesis, similar or up to 10 fold reduced concentrations of RA have a profound effect on postlarval development and the juvenile body plan (Hinman and Degnan, 1998, 2000
). There is a dramatic reduction of anterior, endodermally derived structures of the pharynx and an expansion of the oesophagus (Fig. 2A, B). During RA-perturbed postlarval development there is also a dramatic down regulation of Hec-Otx transcript abundance which is normally expressed in some of the anterior endodermal structures lost upon RA treatment (Fig. 2C). Therefore, while there is evidence of some regulation of Hec-Otx abundance during embryogenesis, this regulation is far more dramatic during postlarval development. Also, regulation of Hec-Otx expression by RA during embryogenesis is restricted to adult stomodeal primordium. Whether increased RA-regulation of Hec-Otx is related to developmental stage (i.e., postlarval rather than embryonic) or tissue specificity (i.e., endodermal rather than neuroectodermal) is unknown and further work is required to resolve this issue. Regardless, increased RA regulation of Hec-Otx in the ascidian postlarva may be related to the more regulative processes involved in patterning of the endoderm. Thus, upstream regulation of some of the homeobox genes involved in anteroposterior patterning by RA may be a feature of ascidian development. Its effect is most dramatic during postlarval development, when regulative development of a large number of migrating, proliferating cells appears to occur. Comparison of the role of RA in vertebrate development with its role in embryonic and postlarval ascidian development suggests that the RA signalling system was involved in regulating homeobox gene expression along the AP axis in the chordate ancestor. If RA was primitively linked to patterning migrating, proliferating cells in chordates, this would imply that the ancestral chordate was macroscopic (at least during part of its development) and therefore unlike the microscopic ascidian tadpole larvae observed today.
|
|
EVOLUTION OF ASCIDIANS: A PROPOSAL |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
We propose that during the evolution of the biphasic life history of ascidians from a macroscopic directly-developing ancestor the following major events occurred. First, Hox, and associated homeobox and Pax genes were involved in patterning both axial (notochord, dorsal hollow nerve cord and somatic tail muscle) and visceral (pharynx, endostyle and gut) components in the chordate ancestor. The development of axial and visceral modules became temporally dissociated, resulting in a dispersive tadpole-like larva and an adult that either remained pelagic or settled onto a substratum. The ascidian ancestor would have developed the ability to detect appropriate environmental signals. These signals became key morphogenetic regulators in the transition from larval to adult body plans, engaging intrinsic signalling systems that were inherited from the directly-developing ancestor (see Eri et al., 1999
). Selective constraints placed on reproduction and the development of the planktonic larva (Strathmann, 1978; Pechenik, 1999) resulted in (i) the ascidian tadpole becoming microscopic, and (ii) an increased reliance on maternal and early cell lineage-based processes in embryonic development. Signalling systems, such as the RA system, became largely redundant or restricted to very short range cell-cell interactions during embryonic development. These systems were maintained in postlarval development of the larger adult. Patterning mechanisms akin to those observed in extant vertebrates may no longer be a cardinal component of ascidian embryonic development but are necessary for the formation of the adult body plan. The requirement of axial patterning homeobox genes and their regulation by RA to establish the adult body plan may place regulatory constraints that inhibit the loss of their coordinated expression in the embryonic CNS. Experimental analyses of the function of RA, its receptor and target genes during ascidian embryonic and postlarval development can be used to test this proposal.
|
ACKNOWLEDGMENTS |
|---|
The authors thank the two anonymous reviewers for their helpful comments and suggestions.
|
FOOTNOTES |
|---|
1 From the Symposium HOX Clusters and the Evolution of Morphology presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 4–8 January 2000, at Atlanta, Georgia.
2 Present address of VF Hinman is Division of Biology, California Institute of Technology, Pasadena, California 91125, USA.
|
References |
|---|
|
TOP
SYNOPSISINTRODUCTIONASCIDIANS: DEVELOPMENT OF TWO...HOMEOBOXES AND DISSOCIATIONS...ANCESTRAL MODULARITYHOMEOBOX GENES AND EMBRYONIC...RETINOIC ACID AND AXIAL...AN ANCESTRAL ROLE FOR...EVOLUTION OF ASCIDIANS: A...References |
|---|
Abouheif, E., M. Akam, W. J. Dickinson, P. W. H. Holland, A. Meyer, N. H. Patel, R. A. Raff, V. L. Roth, and G. A. Wray. 1997. Homology and developmental genes. Trends Genet, 13:432-433.[CrossRef][ISI][Medline]
Ang, S. L., R. A. Conlon, O. Jin, and J. Rossant. 1994. Positive and negative signals from the mesoderm regulate the expression of mouse Otx2 in ectoderm explants. Development, 120:2979-2989.[Abstract]
Avantaggiato, V., D. Acampora, F. Tuorto, and A. Simeone. 1996. Retinoic acid induces stage-specific repatterning of the rostral central nervous system. Dev. Biol, 175:347-257.[CrossRef][ISI][Medline]
Bally-Cuif, L., and E. Boncinelli. 1997. Transcription factors and head formation in vertebrates. BioEssays, 19:127-135.[CrossRef][ISI][Medline]
Beck, F., T. Erler, A. Russell, and R. James. 1995. Expression of Cdx-2 in the mouse embryo and placenta, possible role in patterning of the extra-embryonic membranes. Dev. Dyn, 204:219-227.[ISI][Medline]
Berrill, N. J. 1950. The Tunicata. The Ray Society, London.
Berrill, N. J. 1955. The origin of vertebrates. Oxford University Press, London.
Brooke, N. M., J. Garcia-Fernàndez, and P. W. H. Holland. 1998. The ParaHox gene cluster is and evolutionary sister of the Hox gene cluster. Nature, 392:920-922.[CrossRef][Medline]
Bruce, A. E. E., and M. Shankland. 1998. Expression of the head gene Lox22-Otx in the leech Helobdella and the origin of the bilaterian body plan. Dev. Biol, 201:101-112.[CrossRef][ISI][Medline]
Chen, Y., L. Huang, A. F. Russo, and M. Solursh. 1992. Retinoic acid is enriched in Hensen's node and is developmentally regulated in the early chicken embryo. Proc. Natl. Acad. Sci. U.S.A, 89:10056-10059.
Cloney, R. A. 1982. Ascidian larvae and the events of metamorphosis. Amer. Zool, 35:391-398.
Conlon, R. A. 1995. Retinoic acid and pattern formation in vertebrates. Trends Genet, 11:314-319.[CrossRef][ISI][Medline]
Corbo, J. C., A. Erives, A. Di Gregorio, A. Chang, and M. Levine. 1997. Dorsoventral patterning of the vertebrate neural tube is conserved in a protochordate. Development, 124:2335-2344.[Abstract]
Créton, R., G. Zwaan, and R. Dohmen. 1993. Specific developmental defects in molluscs treated with retinoic acid during gastrulation. Dev. Growth & Differ, 25:357-364.[CrossRef]
Davidson, E. H. 1990. How embryos work: A comparative view of diverse modes of cell fate specification. Development, 108:365-389.[Abstract]
Davidson, E. H. 1991. Spatial mechanisms of gene regulation in metazoan embryos. Development, 113:1-26.[Abstract]
DeBernardi, F., C. Sotgia, and G. Ortolani. 1994. Retinoic acid treatment of ascidian embryos: Effects on larvae and metamorphosis. Anim. Biol, 3:75-81.
Degnan, B. M., P. R. Rohde, and M. F. Lavin. 1996. Normal development and embryonic gene activity of the ascidian Herdmania momus. Mar. Freshwater Res, 47:543-551.[CrossRef]
Degnan, B. M., D. Souter, S. M. Degnan, and S. C. Long. 1997. Induction of metamorphosis with potassium ions requires development of competence and an anterior signalling centre in the ascidian Herdmania momus. Dev. Genes Evol, 206:370-376.[CrossRef]
Dickinson, W. J. 1995. Molecules and morphology: Where's the homology? Trends Genet. , 11:119-121.
Durston, A. J., J. P. M. Timmermans, W. J. Hage, H. F. J. Hendriks, N. J. de Vries, M. Heideveld, and P. D. Nieuwkoop. 1989. Retinoic acid causes an anteroposterior transformation in the developing central nervous system. Nature, 340:140-144.[CrossRef][Medline]
Eri, R., J. M. Arnold, V. F. Hinman, K. M. Green, M. K. Jones, B. M. Degnan, and M. F. Lavin. 1999. Hemps, a novel EGF-like protein, plays a central role in ascidian metamorphosis. Development, 126:5809-5818.[Abstract]
Finkelstein, R., and E. Boncinelli. 1994. From fly head to mammalian forebrain: the story of otd and Otx. Trends Genet, 10:310-315.[CrossRef][ISI][Medline]
Gamer, L. W., and C. V. E. Wright. 1993. Murine Cdx-4 bears striking similarities to the Drosophila caudal gene in its homeodomain sequence and early expression patterns. Mech. Dev, 43:71-81.[CrossRef][ISI][Medline]
Garstang, W. 1928. The morphology of the Tunicata and its bearings on the phylogeny of the Chordata. Q. J. Microsc. Sci, 72:51-187.
Gerhart, J., and K. Kirschner. 1997. Cells, embryos and evolution. Blackwell Science, Malden MA.
Gionti, M., F. Ristoratore, A. Di Gregorio, F. Aniello, M. Branno, and R. Di Lauro. 1998. Cihox5, a new Ciona intestinalis Hox-related gene, is involved in regionalization of the spinal cord. Dev. Genes Evol, 207:515-523.[CrossRef][ISI][Medline]
Glardon, S., P. Callaerts, G. Halder, and W. J. Gehring. 1997. Conservation of Pax-6 in a lower chordate, the ascidian Phallusia mammillata. Development, 124:817-825.[Abstract]
Hall, B. K. 1992. Evolutionary developmental biology. Chapman & Hall, London.
Hinman, V. F. 2000. Homeobox genes and retinoic acid in the development of the ascidian Herdmania curvata: Evolution of body plan transitions. Ph. D. Thesis, Univer. of Queensland.
Hinman, V. F., and B. M. Degnan. 1998. Retinoic acid disrupts anterior ectodermal and endodermal development in ascidian larvae and postlarvae. Dev. Genes Evol, 208:336-345.[CrossRef][ISI][Medline]
Hinman, V. F., and B. M. Degnan. 2000. Retinoic acid perturbs Otx gene expression in the ascidian pharynx. Dev. Genes Evol, 210:129-139.[CrossRef][ISI][Medline]
Hinman, V. F., E. Becker, and B. M. Degnan. 2000. Neurectodermal and endodermal expression of the ascidian Cdx gene is separated by metamorphosis. Dev. Genes Evol, 210:212-216.[CrossRef][ISI][Medline]
Hirano, T., and H. Nishida. 1997. Developmental fates of larval tissues after metamorphosis in ascidian Halocynthia roretzi: I. Origin of mesodermal tissues of the juvenile. Dev. Biol, 192:199-210.
Hirano, T., and H. Nishida. 2000. Developmental fates of larval tissues after metamorphosis in the ascidian, Halocynthia roretzi—II. Origin of endodermal tissues of the juvenile. Dev. Genes Evol, 210:55-63.
Hisata, K., S. Fujiwara, Y. Tsuchida, M. Ohashi, and K. Kawamura. 1998. Expression and function of a retinoic acid receptor in budding ascidians. Dev. Genes Evol, 208:537-546.[CrossRef][ISI][Medline]
Hogan, B. L. M., C. Thaller, and E. Eichele. 1992. Evidence that Hensen's node is a site of retinoic acid synthesis. Nature, 359:237-241.[CrossRef][Medline]
Holland, L. Z., and N. D. Holland. 1996. Expression of AmphiHox-1 and AmphiPax-1 in amphioxus embryos treated with retinoic acid, insights into evolution and patterning of the chordate nerve cord and pharynx. Development, 122:1829-1838.[Abstract]
Holland, L. Z., and N. D. Holland. 1999. Amphioxus and the utility of molecular genetic data for hypothesising body part homologies between distantly related animals. Amer. Zool, 39:630-640.
Holland, L. Z., P. W. H. Holland, and N. D. Holland. 1996. Revealing homologies between body parts of distantly related animals by in situ hybridization to developmental genes: Amphioxus vs. vertebrates. In J. D. Ferraris and S. Palumbi (eds.), Molecular zoology. pp. 267–282, 473–483. Wiley-Liss, New York.
Holland, L. Z., M. Kene, N. A. Williams, and N. D. Holland. 1997. Sequence and embryonic expression of the amphioxus engrailed gene (AmphiEn): The metameric pattern of transcription resembles that of its segment-polarity homolog in Drosophila. Development, 124:1723-1732.[Abstract]
Holland, P. W. H. 1991. Cloning and evolutionary analysis of msh-like homeobox genes from mouse, zebrafish and ascidian. Gene, 98:253-258.[CrossRef][ISI][Medline]
Imsiecke, G., R. Borojevic, M. Custodio, and W. Müller. 1994. Retinoic acid acts as a morphogen in freshwater sponges. Invertebr. Reprod. Dev, 26:89-98.
Isaacs, H. V., M. E. Pownall, and J. M. W. Slack. 1998. Regulation of Hox gene expression and posterior development by the Xenopus homologue Xcad3. EMBO J, 17:3413-3427.[CrossRef][ISI][Medline]
Jefferies, R. P. S. 1986. The ancestry of the vertebrates. British Museum (Natural History), London.
Joyner, A. L. 1996. Enrailed, Wnt and Pax genes regulate midbrain-hindbrain boundary. Trends Genet, 12:15-19.[CrossRef][ISI][Medline]
Kablar, B., R. Vignali, L. Menotti, M. Pannese, M. Andreazzoli, C. Polo, G. M. Giribaldi, E. Boncinelli, and G. Barasacchi. 1996. Xotx genes in the developing brain of Xenopus laevis. Mech. Dev, 55:145-158.[CrossRef][ISI][Medline]
Katsuyama, Y., Y. Sato, S. Wada, and H. Saiga. 1999. Ascidian tail formation requires caudal function. Dev. Biol, 213:257-268.[CrossRef][ISI][Medline]
Katsuyama, Y., S. Wada, S. Yasugi, and H. Saiga. 1995. Expression of the labial group Hox gene HrHox-1 and its alteration induced by retinoic acid in development of the ascidian Halocynthia roretzi. Development, 121:3197-3205.[Abstract]
Kessel, M. 1992. Respecification of vertebral identities by retinoic acid. Development, 115:487-501.[Abstract]
Kessel, M., and P. Gruss. 1991. Homeotic transformations of murine vertebrae and concomitant alteration of Hox codes induced by retinoic acid. Cell, 67:89-104.[CrossRef][ISI][Medline]
Kozmik, Z., N. D. Holland, A. Kalousova, J. Paces, M. Schubert, and L. Z. Holland. 1999. Characterization of an amphioxus paired box gene, AmphiPax2/5/8: Developmental expression patterns in optic support cells, nephridium, thyroid-like structures and pharyngeal gill slits, but not in the midbrain-hindbrain boundary region. Development, 126:1295-1304.[Abstract]
Locascio, A., F. Aniello, A. Amoroso, M. Manzanares, R. Krumlauf, and M. Branno. 1999. Patterning the ascidian nervous system: Structure, expression and transgenic analysis of the CiHox3 gene. Development, 126:4737-4748.[Abstract]
Macdonald, P. M., and G. Struhl. 1986. A molecular gradient in early Drosophila embryos and its role in specifying the body pattern. Nature, 324:537-545.[CrossRef][Medline]
Marom, K., E. Shapira, and A. Fainsod. 1997. The chicken caudal genes establish an anterior-posterior gradient by partially overlapping temporal and spatial patterns of expression. Mech. Dev, 64:41-52.[CrossRef][ISI][Medline]
Marshall, H., M. Studer, H. Pöpperl, S. Aparico, A. Kuroiwa, S. Brenner, and R. Krumlauf. 1994. A conserved retinoic acid response element required for early expression of the homeobox gene Hoxb-1. Nature, 370:567-571.[CrossRef][Medline]
McMahon, A. P., and A. Bradley. 1990. The wnt-1 (int-1) proto-oncogene is required for development of a large region of the mouse brain. Cell, 69:581-595.
Meedel, T. H. 1992. Development of the ascidian embryo: Cell fate specification by autonomous and inductive processes. In E. F. Rossomando and S. Alexander (eds.), Morphogenesis. An analysis of the development of biological form. Marcel Dekker Inc., New York.
Mlodzik, M., and W. J. Gehring. 1987a. Hierarchy of the genetic interactions that specify the anteroposterior segmentation pattern of the Drosophila embryo as monitored by caudal protein expression. Development, 101:421-435.[Abstract]
Mlodzik, M., and W. J. Gehring. 1987b. Expression of the caudal gene in the germ line of Drosophila: Formation of an RNA and protein gradient during early embryogenesis. Cell, 48:465-478.[CrossRef][ISI][Medline]
Morales, A. V., E. J. de la Rosa, and F. de Pablo. 1996. Expression of the cCdx-B homeobox gene in chick embryo suggests its participation in rostrocaudal axial patterning. Dev. Dyn, 206:343-353.[CrossRef][ISI][Medline]
Morriss-Kay, G. M., P. Murphy, R. E. Hill, and D. R. Davidson. 1991. Effects of retinoic acid excess on expression of Hox-2.9 and Krox-20 and on morphological segmentation in the hindbrain of mouse embryos. EMBO J, 10:2985-2995.[ISI][Medline]
Müller, W. A. 1984. Retinoids and pattern formation in a hydroid. J. Embryol. Exp. Morph, 81:253-271.[ISI][Medline]
Nakauchi, M., and T. Takeshita. 1983. Ascidian one-half embryos can develop into functional adult ascidians. J. Exp. Zool, 227:155-158.[CrossRef][ISI][Medline]
Nicol, D., and I. A. Meinertzhagen. 1991. Cells counts and maps in the larval central nervous system on the ascidian Ciona intestinalis (L.). J. Comp. Neurol, 309:415-429.[CrossRef][ISI][Medline]
Oro, A. E., M. McKeown, and R. M. Evans. 1990. Relationship between the product of the Drosophila ultraspiracle locus and the vertebrate retinoid X receptor. Nature, 347:298-301.[CrossRef][Medline]
Pechenik, J. A. 1999. On the advantages and disadvantages of larval stages in benthic marine invertebrate life cycles. Mar. Ecol. Prog. Ser, 177:269-297.
Peterson, K. J., R. A. Cameron, and E. H. Davidson. 2000. Bilaterian origins: Significance of new experimental observations. Dev. Biol, 219:1-17.[CrossRef][ISI][Medline]
Pillemer, G., M. Epstein, B. Blumberg, J. K. Yisraeli, E. D. DeRobertis, H. Steinbeisser, and A. Fainsod. 1998. Nested expression and sequential down regulation of the Xenopus caudal genes along the anterior-posterior axis. Mech. Dev, 71:193-196.[CrossRef][ISI][Medline]
Raff, R. A. 1996. The shape of life: Genes, development, and the evolution of animal form. University of Chicago Press, Chicago.
Rhinn, M., A. Dierich, W. Shawlot, R. R. Behringer, M. Le Meur, and S.-L. Ang. 1998. Sequential roles for Otx2 in visceral endoderm and neuroectoderm for forebrain and midbrain induction and specification. Development, 125:845-856.[Abstract]
Romer, A. S. 1959. The vertebrate story. University of Chicago Press, Chicago.
Romer, A. S. 1962. The Vertebrate Body. 3rd ed. Saunders Company, London.
Rowitch, D. H., and A. P. McMahon. 1995. Pax-2 expression in the murine neural plate precedes and encompasses the expression domains of Wnt-1 and En-1. Mech. Dev, 52:3-8.[CrossRef][ISI][Medline]
Satoh, N. 1994. Developmental biology of ascidians. Cambridge University Press, Cambridge.
Satoh, N., and W. R. Jeffery. 1995. Chasing tails in ascidians; developmental insights into the origin and evolution of chordates. Trends Genet, 11:354-359.