Experimental Approach Reveals the Role of alx1 in the Evolution of the Echinoderm Larval Skeleton.

Over the course of evolution, the acquisition of novel structures has ultimately led to wide variation in morphology among extant multicellular organisms. Thus, the origins of genetic systems for new morphological structures are a subject of great interest in evolutionary biology. The larval skeleton is a novel structure acquired in some echinoderm lineages via the activation of the adult skeletogenic machinery. Previously, VEGF signaling was suggested to have played an important role in the acquisition of the larval skeleton. In the present study, we compared expression patterns of Alx genes among echinoderm classes to further explore the factors involved in the acquisition of a larval skeleton. We found that the alx1 gene, originally described as crucial for sea urchin skeletogenesis, may have also played an essential role in the evolution of the larval skeleton. Unlike those echinoderms that have a larval skeleton, we found that alx1 of starfish was barely expressed in early larvae that have no skeleton. When alx1 overexpression was induced via injection of alx1 mRNA into starfish eggs, the expression patterns of certain genes, including those possibly involved in skeletogenesis, were altered. This suggested that a portion of the skeletogenic program was induced solely by alx1. However, we observed no obvious external phenotype or skeleton. We concluded that alx1 was necessary but not sufficient for the acquisition of the larval skeleton, which, in fact, requires several genetic events. Based on these results, we discuss how the larval expression of alx1 contributed to the acquisition of the larval skeleton in the putative ancestral lineage of echinoderms.

Development of ciliary bands in larvae of the living isocrinid sea lily Metacrinus rotundus.

Embryos and larvae of an isocrinid sea lily, Metacrinus rotundus, are described by scanning electron microscopy. Around hatching (35 h after fertilization), the outer surface of the gastrula becomes ubiquitously covered with short cilia. At 40 h, the hatched swimming embryo develops a cilia-free zone of ectoderm on the ventral side. By 3 days, the very early dipleurula larva develops a cilia-free zone ventrally, densely ciliated regions laterally, and a sparsely ciliated region dorsally. At this stage, the posterior and anterior ciliary bands first appear: the former runs along a low ridge separating the densely from the sparsely ciliated epidermal regions, while the latter is visible, at first discontinuously, along the boundary between the densely ciliated lateral regions and the cilia-free ventral zone. In the late dipleurula larva (5 days after fertilization), the anterior and posterior loops of ciliary bands are well defined. The transition from the dipleurula to the semidoliolaria larva occurs at 6 days as the posterior loop becomes rearranged to form incompletely circumferential ciliary bands. The larva becomes competent to settle at this stage. The arrangement of the ciliary bands on the semidoliolaria is maintained during the second week of development, while the larva retains its competence to settle. The larval ciliary patterns described here are compared with those of stalkless crinoids and eleutherozoan echinoderms. The closest morphological similarities are between M. rotundus and the basal eleutherozoan class Asteroidea.

Sea lily muscle lacks a troponin-regulatory system, while it contains paramyosin.

Troponin, a Ca(2+)-dependent regulator of striated muscle contraction, has been characterized in vertebrates, protochordates (amphioxus and ascidian), and many invertebrate animals that are categorized in protostomes, but it has not been detected in echinoderms, such as sea urchin and sea cucumber, members of subphylum Eleutherozoa. In this study, we examined the muscle of a species of isocrinid sea lilies, a member of subphylum Pelmatozoa, that constitute the most basal group of extant echinoderms to clarify whether troponin is lacking from the early evolution of echinoderms. Native thin filaments were released from the muscle homogenates in a relaxing buffer containing ATP and EGTA, a Ca(2+)-chelator, and were collected by ultra-centrifugation. Actin and tropomyosin, but not a troponin-like protein, were detected in the filament preparation. The filaments increased Mg(2+)-ATPase activity of rabbit skeletal muscle myosin irrespective of the presence or absence of Ca(2+). The results indicate that Ca(2+)-sensitive factor, troponin, is lacking in the thin filaments of sea lily muscle as in those of the other echinoderms, sea urchin and sea cucumber. On the other hand, a paramyosin-like protein that is absent from chordates was detected in sea lily muscle as in the muscles of the other echinoderms and invertebrate animals of protostomes.

Gene expression analysis of Six3, Pax6, and Otx in the early development of the stalked crinoid Metacrinus rotundus.

The stalked crinoid, Metacrinus rotundus, is one of the most basal extant echinoderms. Here, we show the expression patterns of Six3, Pax6, and Otx in the early development of M. rotundus. All three genes are highly expressed in stages from the gastrula to the auricularia larval stage. Ectodermal expression of MrOtx appears to be correlated with development of the ciliary band. These three genes are expressed sequentially along the embryonic body axis in the anterior and middle walls of the archenteron in the order of MrPax6, MrSix3, and MrOtx. The anterior, middle, and posterior parts of the archenteron in the late gastrula differentiate into the axo-hydrocoel, the enteric sac, and somatocoels at later stages, respectively. The three genes are expressed sequentially from the tip of the axo-hydrocoel to the bottom of enteric sac in the order of MrSix3, MrPax6, and MrOtx at the later stages. This suggests that these genes are involved in patterning of the larval endo-mesoderm in stalked crinoids. The present results suggest that radical alterations have occurred in the expression and function of homeobox genes in basal echinoderms.

Evolutionary modification of specification for the endomesoderm in the direct developing echinoid Peronella japonica: loss of the endomesoderm-inducing signal originating from micromeres.

We investigated the inductive signals originating from the vegetal blastomeres of embryos of the sand dollar Peronella japonica, which is the only direct developing echinoid species that forms micromeres. To investigate the inductive signals, three different kinds of experimental embryos were produced: micromere-less embryos, in which all micromeres were removed at the 16-cell stage; chimeric embryos produced by an animal cap (eight mesomeres) recombined with a micromere quartet isolated from a 16-cell stage embryo; and chimeric embryos produced by an animal cap recombined with a macromere-derived layer, the veg1 or veg2 layer, isolated from a 64-cell stage embryo. Novel findings obtained from this study of the development of these embryos are as follows. Micromeres lack signals for endomesoderm specification, but are the origin of a signal establishing the oral-aboral (O-Ab) axis. Some non-micromere blastomeres, as well as micromeres, have the potential to form larval skeletons. Macromere descendants have endomesoderm-inducing potential. Based on these results, we propose the following scenario for the first step in the evolution of direct development in echinoids: micromeres lost the ability to send a signal endomesoderm induction so that the archenteron was formed autonomously by macromere descendants. The micromeres retained the ability to form larval spicules and to establish the O-Ab axis.

Nervous system development of two crinoid species, the sea lily Metacrinus rotundus and the feather star Oxycomanthus japonicus.

Nervous system development in echinoderms has been well documented, especially for sea urchins and starfish. However, that of crinoids, the most basal group of extant echinoderms, has been poorly studied due to difficulties in obtaining their larvae. In this paper, we report nervous system development from two species of crinoids, from hatching to late doliolaria larvae in the sea lily Metacrinus rotundus and from hatching to cystidean stages after settlement in the feather star Oxycomanthus japonicus. The two species showed a similar larval nervous system pattern with an extensive anterior larval ganglion. The ganglion was similar to that in sea urchins which is generally regarded as derived. In contrast with other echinoderm and hemichordate larvae, synaptotagmin antibody 1E11 failed to reveal ciliary band nerve tracts. Basiepithelial nerve cells formed a net-like structure in the M. rotundus doliolaria larvae. In O. japonicus, the larval ganglion was still present 1 day after settlement when the adult nervous system began to appear inside the crown. Stalk nerves originated from the crown and extended down the stalk, but had no connections with the remaining larval ganglion at the base of the stalk. The larval nervous system was not incorporated into the adult nervous system, and the larval ganglion later disappeared. The aboral nerve center, the dominant nervous system in adult crinoids, was formed at the early cystidean stage, considerably earlier than previously suggested. Through comparisons with nervous system development in other ambulacraria, we suggest the possible nervous system development pattern of the echinoderm ancestor and provide new implications on the evolutionary history of echinoderm life cycles.

Krüppel-like is required for nonskeletogenic mesoderm specification in the sea urchin embryo.

The canonical Wnt pathway plays a central role in specifying vegetal cell fate in sea urchin embryos. SpKrl has been cloned as a direct target of nuclear beta-catenin. Using Hemicentrotus pulcherrimus embryos, here we show that HpKrl controls the specification of secondary mesenchyme cells (SMCs) through both cell-autonomous and non-autonomous means. Like SpKrl, HpKrl was activated in both micromere and macromere progenies. To examine the functions of HpKrl in each blastomere, we constructed chimeric embryos composed of blastomeres from control and morpholino-mediated HpKrl-knockdown embryos and analyzed the phenotypes of the chimeras. Micromere-swapping experiments showed that HpKrl is not involved in micromere specification, while micromere-deprivation assays indicated that macromeres require HpKrl for cell-autonomous specification. Transplantation of normal micromeres into a micromere-less host with morpholino revealed that macromeres are able to receive at least some micromere signals regardless of HpKrl function. From these observations, we propose that two distinct pathways of endomesoderm formation exist in macromeres, a Krl-dependent pathway and a Krl-independent pathway. The Krl-independent pathway may correspond to the Delta/Notch signaling pathway via GataE and Gcm. We suggest that Krl may be a downstream component of nuclear beta-catenin required by macromeres for formation of more vegetal tissues, not as a member of the Delta/Notch pathway, but as a parallel effector of the signaling (Krl-dependent pathway).

Micromere-derived signal regulates larval left-right polarity during sea urchin development.

The micromeres (Mics) lineage functions as a morphogenetic signaling center in early embryos of sea urchins. The Mics lineage releases signals that regulate the specification of cell fates along the animal-vegetal and oral-aboral axes. We tested whether the Mics lineage might also be responsible for differentiation of the left-right (LR) axis by observing of the placement of the adult rudiment, which normally forms only on the left side of the larvae, after removal of the Mics lineage. When all of the Mics lineage were removed from embryos of the regular sea urchin Hemicentrotus pulcherrimus between the 16- and 64-cell stages, the LR placement of the rudiment became randomized. However, the immediate retransplantation of the Mics rescued the normal LR placement of the rudiment, indicating that the Mics lineage releases a signal that specifies LR polarity. Additionally, we investigated whether the specification of LR polarity of whole embryos in the indirect-developing sea urchin H. pulcherrimus is affected by LiCl exposure, which disturbs the establishment of LR asymmetry in a direct-developing sea urchin. Larvae derived from normal animal caps combined with LiCl-exposed Mics descendants were defective in normal LR placement of the rudiment, suggesting that LiCl interferes with the Mics-derived signal. In contrast, embryos of two sand dollar species (Scaphechinus mirabilis and Astriclypeus manni) were resistant to alteration of LR placement of the rudiment by either removal of the Mics lineage or LiCl exposure. These results indicate that the Mics lineage is involved in specification of LR polarity in the regular sea urchin H. pulcherrimus, and suggest that LiCl impairs the normal LR patterning by affecting Mics-derived signaling.

Phylogenetic correspondence of the body axes in bilaterians is revealed by the right-sided expression of Pitx genes in echinoderm larvae.

Chordates and echinoderms are two of the three major deuterostome phyla and show conspicuous left-right (LR) asymmetry. The establishment of LR asymmetry has been explored in vertebrates, but is largely unknown in echinoderms. Here, we report the expression pattern of genes that are orthologous to the chordate left-side specific gene Pitx, cloned from the sea urchin Hemicentrotus pulcherrimus (HpPitx) and the starfish Asterina pectinifera (ApPitx). HpPitx transcripts were first detected bilaterally in one cell of the ventrolateral primary mesenchyme-cell aggregate of early prism larvae. New expression was detected asymmetrically in the right counterpart of a bilateral pair of mesodermal coelomic pouches and in the right ectoderm. In starfish bipinnaria larvae, the ApPitx signal was detected in the right coelomic pouch and in the right half of the ectoderm along the ciliary bands. These results suggest that the function of Pitx in establishing LR asymmetry was introduced in the last common ancestor of echinoderms and chordates. However, the right-side specific expression in echinoderm larvae is inverted compared to chordate embryos. This indicates that the LR axis is inversely represented between echinoderms and chordates, which supports the scenario that dorsoventral axis inversion was introduced into the chordate lineage by turning upside down.

Expression patterns of Hox genes in larvae of the sea lily Metacrinus rotundus.

We cloned eight Hox genes (MrHox1, MrHox2, MrHox4, MrHox5, MrHox7, MrHox8, MrHox9/10, and MrHox11/13c) from the sea lily Metacrinus rotundus, a member of the most basal group of the extant echinoderms. At the auricularia stage, before the formation of the pentaradial rudiment, four MrHox genes were expressed sequentially along the anteroposterior (AP) axis in the straightened mesodermal somatocoels in the order MrHox5, MrHox7, MrHox8, and MrHox9/10. The expression of MrHox7 and MrHox8 was detected as early as the hatching stage in the presumptive somatocoel region of the archenteral sac. MrHox5 was expressed in the anteriormost region of the somatocoels, where a stalk-related structure (the chambered organ) forms later. In addition to the mesodermal somatocoels, MrHox7 was expressed in the oral hood ectoderm, which gives rise to the adhesive pit. The expression of four other MrHox genes (MrHox1, MrHox2, MrHox4, and MrHox11/13c) was not detected in any of the larval stages we examined. In comparison with the mesodermal sea urchin Hox genes, the MrHox genes are expressed more posteriorly along the AP (oral-anal) axis than the sea urchin orthologs, implying that the evolution of the eleutherozoans was accompanied by a posteriorization of the larval body. Our study illuminates the possible body plan and Hox expression patterns of the ancestral echinoderm and sheds light on the larval body plan of the last common ancestor of the echinoderms and chordates.

Nervous system development of the sea cucumber Stichopus japonicus.

The nervous system development of the sea cucumber Stichopus japonicus was investigated to explore the development of the bilateral larval nervous system into the pentaradial adult form typical of echinoderms. The first nerve cells were detected in the apical region of epidermis in the late gastrula. In the auricularia larvae, nerve tracts were seen along the ciliary band. There was a pair of bilateral apical ganglia consisted of serotonergic nerve cells lined along the ciliary bands. During the transition to the doliolaria larvae, the nerve tracts rearranged together with the ciliary bands, but they were not segmented and remained continuous. The doliolaria larvae possessed nerves along the ciliary rings but strongly retained the features of auricularia larvae nerve pattern. The adult nervous system began to develop inside the doliolaria larvae before the larval nervous system disappears. None of the larval nervous system was observed to be incorporated into the adult nervous system with immunohistochemistry. Since S. japonicus are known to possess an ancestral mode of development for echinoderms, these results suggest that the larval nervous system and the adult nervous system were probably formed independently in the last common ancestor of echinoderms.

Developmental potential of small micromeres in sea urchin embryos.

The large micromeres (lMics) of echinoid embryos are reported to have distinct potentials with regard to inducing endo-mesoderm and autonomous differentiation into skeletogenic cells. However, the developmental potential of small micromeres (sMics), the sibling of lMics, has not been clearly demonstrated. In this study we produced chimeric embryos from an animal cap recombined with various numbers of sMics, in order to investigate the developmental potential of sMics in the sea urchin Hemicentrotus pulcherrimus and the sand dollar Scaphechinus mirabilis. We found that sMics of H. pulcherrimus had weak potential for inducing presumptive ectoderm cells to form endo-mesoderm structures. The inducing potential of ten sMics was almost equivalent to that of one lMic. The sMics also had the potential to differentiate autonomously into skeletogenic cells. Conversely, the sMics of S. mirabilis did not show either inductive or skeletogenic differentiation potential. The sMics of both species had the potential to induce oral-aboral axis establishment. These results suggest that the potential for sMics to differentiate into skeletogenic cells and for inducing the presumptive ectoderm to differentiate into endomesoderm differs across species, while the potential of sMics to induce the oral-aboral axis is conserved among species.

The micro1 gene is necessary and sufficient for micromere differentiation and mid/hindgut-inducing activity in the sea urchin embryo.

In the sea urchin embryo, micromeres have two distinct functions: they differentiate cell autonomously into the skeletogenic mesenchyme cells and act as an organizing center that induces endomesoderm formation. We demonstrated that micro1 controls micromere specification as a transcriptional repressor. Because micro1 is a multicopy gene with at least six polymorphic loci, it has been difficult to consistently block micro1 function by morpholino-mediated knockdown. Here, to block micro1 function, we used an active activator of micro1 consisting of a fusion protein of the VP16 activation domain and the micro1 homeodomain. Embryos injected with mRNA encoding the fusion protein exhibited a phenotype similar to that of micromere-less embryos. To evaluate micro1 function in the micromere, we constructed chimeric embryos composed of animal cap mesomeres and a micromere quartet from embryos injected with the fusion protein mRNA. The chimeras developed into dauerblastulae with no vegetal structures, in which the micromere progeny constituted the blastula wall. We also analyzed the phenotype of chimeras composed of an animal cap and a mesomere expressing micro1. These chimeras developed into pluteus larvae, in which the mesomere descendants ingressed as primary mesenchyme cells and formed a complete set of skeletal rods. The hindgut and a part of the midgut were also generated from host mesomeres. However, the foregut and nonskeletogenic mesoderm were not formed in the larvae. From these observations, we conclude that micro1 is necessary and sufficient for both micromere differentiation and mid/hindgut-inducing activity, and we also suggest that micro1 may not fulfill all micromere functions.

LiCl inhibits the establishment of left-right asymmetry in larvae of the direct-developing echinoid Peronella japonica.

The effect of LiCl on the establishment of left-right (LR) asymmetry in larvae of the direct-developing echinoid Peronella japonica was investigated with special attention to the location of the amniotic opening and ciliary band pattern. The larvae of echinoids are LR symmetric, but shortly before metamorphosis the larval LR symmetry is lost as a result of the formation of an amniotic cavity (vestibule), part of the adult rudiment, on the left side of the body. P. japonica has been considered to be the only exception among the echinoids, because the amniotic cavity forms at the midline of the larval body. In the present study we discovered the following two different LR asymmetric traits in larvae of P. japonica: the opening of the amniotic cavity initially forms at the midline of the larval body but shifts to the left dorsal side, and a looped ciliary band that initially forms with LR symmetry becomes LR asymmetric as a result of the formation of a bulge on left dorsal side. The establishment of LR asymmetry in both the location of the amniotic opening and the change in the shape of the ciliary band was influenced by exposing embryos to LiCl. Quantitative analysis of the shift in amniotic opening showed that exposure of embryos to LiCl causes repression of leftward shifting of the amniotic opening in earlier stage larvae, and leftward or rightward shifting in later stage larvae. These findings suggest that LiCl is an effective means of impairing the establishment of LR asymmetry in sea urchin embryos.

Molecular evidence of the existence of two sibling species within the echinothurioid echinoid Asthenosoma ijimai from Japanese waters.

The echinoid, Asthenosoma ijimai belonging to the order Echinothurioida from Japanese waters shows the geographical variation in morphological and ecological characters. The echinothurioid from Ryukyu Islands in southern Japan is cleary different from that of Sagami Bay and Suruga Bay in the middle part of Japan at non-molecular level. Their phylogenetic and taxonomic relationships were studied at the molecular level by allozyme analysis. The results demonstrated that the two echinothurioids from Ryukyu Islands and Sagami Bay do not share gene pools with each other, and they were fixed for different alleles at five genetic loci (Mdh, G6pd, Po, Alk-3 and Est-7) in a total of 23 enzyme genetic loci scored. This indicates no gene flow between the two echinothurioids, and is a molecular evidence for that they are reproductively isolated and genetically distinct species. The Nei's genetic distance (D=0.181) between the two were significantly higher than those between conspecific local populations, and comparable to those between closely related species in many other animals containing echinoderms. The present molecular data are well consistent with the non-molecular evidence from morphology, developmental biology and ecology. Putting these data together, we propose that the two echinothurioids should be classified as two sibling species of the genus Asthenosoma and would like to give the following scientific names: the echinothurioid species from Sagami Bay is Asthenosoma ijimai and that from Ryukyu Islands is A. ijimai R.

Molecular heterotopy in the expression of Brachyury orthologs in order Clypeasteroida (irregular sea urchins) and order Echinoida (regular sea urchins).

The expression patterns of Brachyury (Bra) orthologs in the development of four species of sand dollars (order: Clypeasteroida), including a direct-developing species, and of a sea urchin species (order: Echinoida) were investigated during the period from blastula to the pluteus stage, with special attention paid to the relationship between the expression pattern and the mode of gastrulation. The sand dollar species shared two expression domains of the Bra orthologs with the Echinoida species, in the vegetal ring (the first domain) and the oral ectoderm (the second domain). The following heterotopic changes in the expression of the Bra genes were found among the sand dollar species and between the sand dollars and the Echinoida species. (1) The vegetal ring expressing Bra in the sand dollars was much wider and was located at a higher position along the AV axis, compared with that in the Echinoida species. The characteristic Bra expression in the vegetal ring of the sand dollar embryos was thought to be involved in the mode of gastrulation, in which involution continues from the beginning of invagination until the end of gastrulation. (2) Two of the three indirect-developing sand dollar species that were examined exhibited a third domain, in which Bra was expressed on the oral side of the archenteron. (3) In the direct-developing sand dollar embryos, Bra was expressed with an oral-aboral asymmetry in the vegetal ring and with a left-right asymmetry in the oral ectoderm. In the Echinoida species, Bra was expressed in the vestibule at the six-armed pluteus stage.

Regrowth of the stalk of the sea lily, Metacrinus rotundus (Echinodermata: Crinoidea).

Sea lilies are critical to understanding the evolution of the echinoderm body plan, because they are the only extant group whose adults possess a stalk, a prevalent feature in the radiation of a number of primitive echinoderm lineages. Extensive crown regeneration ability has been reported in Metacrinus rotundus, but the regenerative potential of the stalk has never been determined in any species of sea lilies. In this study, we show that M. rotundus whose stalks have been completely excised are capable of stalk regeneration. The process is similar to the growth of the original stalk, but much slower, and the regenerated stalks are not morphologically identical to the original stalk. Since stalk regeneration, in contrast to well-studied regeneration events, probably requires little additional activation of morphogenetic programs, we refer to the stalk regeneration phenomenon as "stalk regrowth" to distinguish it as a special form of regeneration. Since specimens whose entire stalk below the basal plates had been removed were able to regrow, the basal plates, and probably the aboral nerve center within them, are essential for stalk regrowth. Sea lily stalk regrowth is described in detail, and the evolution of feather stars is discussed in light of the growth pattern of the sea lily stalk.

Larval stages of a living sea lily (stalked crinoid echinoderm).

The embryos and larvae of stalked crinoids, which are considered the most basal group of extant echinoderms, have not previously been described. In contrast, much is known about the development of the more accessible stalkless crinoids (feather stars), which are phylogenetically derived from stalked forms. Here we describe the development of a sea lily from fertilization to larval settlement. There are two successive larval stages: the first is a non-feeding auricularia stage with partly longitudinal ciliary bands (similar to the auricularia and bipinnaria larvae of holothurian and asteroid echinoderms, respectively); the second is a doliolaria larva with circumferential ciliary bands (similar to the earliest larval stage of stalkless crinoids). We suggest that a dipleurula-type larva is primitive for echinoderms and is the starting point for the evolution of additional larval forms within the phylum. From a wider evolutionary viewpoint, the demonstration that the most basal kind of echinoderm larva is a dipleurula is consistent with Garstang's auricularia theory for the phylogenetic origin of the chordate neural tube.

T-brain homologue (HpTb) is involved in the archenteron induction signals of micromere descendant cells in the sea urchin embryo.

Signals from micromere descendants play a crucial role in sea urchin development. In this study, we demonstrate that these micromere descendants express HpTb, a T-brain homolog of Hemicentrotus pulcherrimus. HpTb is expressed transiently from the hatched blastula stage through the mesenchyme blastula stage to the gastrula stage. By a combination of embryo microsurgery and antisense morpholino experiments, we show that HpTb is involved in the production of archenteron induction signals. However, HpTb is not involved in the production of signals responsible for the specification of secondary mesenchyme cells, the initial specification of primary mesenchyme cells, or the specification of endoderm. HpTb expression is controlled by nuclear localization of beta-catenin, suggesting that HpTb is in a downstream component of the Wnt signaling cascade. We also propose the possibility that HpTb is involved in the cascade responsible for the production of signals required for the spicule formation as well as signals from the vegetal hemisphere required for the differentiation of aboral ectoderm.