Programmed cell death in the ascidian embryo: modulation by FoxA5 and Manx and roles in the evolution of larval development

Programmed cell death (PCD) has been discounted in the ascidian embryo because the descendants of every embryonic cell appear to be present in the tadpole larva. Here we show that apoptotic PCD is initiated in the epidermis and central nervous system (CNS) but not in the endoderm, mesenchyme, muscle, and notochord cells during embryogenesis in molgulid ascidians. However, the affected cells do not actually die until the beginning of metamorphosis. Although specific patterns of PCD were different in distantly related ascidian species, the results suggest that removal of CNS cells by apoptosis is a urchordate feature predating the origin of the vertebrates. Certain molgulid ascidian species have evolved an anural (tailless) larva in which notochord cells fail to undergo the morphogenetic movements culminating in tail development. These anural species include Molgula occulta, the sister species of the urodele (tailed) species Molgula oculata. We show that PCD in the notochord cell lineage precedes the arrest of tail development in M. occulta and other independently evolved anural species. The notochord cells are rescued from PCD and a tail develops in hybrid embryos produced by fertilizing M. occulta eggs with M. oculata sperm, implying that apoptosis is controlled zygotically. Antisense inhibition experiments show that zygotic expression of the FoxA5 and Manx genes is required to prevent notochord PCD in urodele species and hybrids with restored tails. The results provide the first indication of PCD in the ascidian embryo and suggest that apoptosis modulated by FoxA5 and Manx is involved in notochord and tail regression during anural development. Differences in PCD that occur between ascidian species suggest that diversity in programming apoptosis may explain differences in larval form.     

Interspecific hybridization between an anural and urodele ascidian: differential expression of urodele features suggests multiple mechanisms control anural development

Anural development in the ascidian Molgula occulta was examined using tissue-specific markers and interspecific hybridization. Unlike most ascidians, which develop into a swimming tadpole larva (urodele development), M. occulta eggs develop into a tailless slug-like larva (anural development) which metamorphoses into an adult. M. occulta embryos show conventional early cleavage patterns, gastrulation, and neurulation, but then diverge from the urodele developmental mode during larval morphogenesis. M. occulta larvae do not contain a pigmented sensory cell in their brain or form a tail with differentiated notochord and muscle cells. As shown by in situ hybridization with cloned probes and analysis of in vitro translation products, M. occulta embryos do not accumulate high levels of alpha actin or myosin heavy chain mRNA. In contrast, acetylcholinesterase is expressed in muscle lineage cells, indicating that various muscle cell features are differentially suppressed. M. occulta embryos also lack tyrosinase activity, suggesting that suppression of brain pigment cell differentiation occurs at an early step in development. M. occulta eggs fertilized with sperm from Molgula oculata (a closely related urodele species) develop into hybrid larvae exhibiting some of the missing urodele features. Some hybrid embryos develop tyrosinase activity and differentiate a brain pigment cell and a short row of notochord cells, and form a short tail. These urodele features appeared together or separately in different hybrid embryos suggesting that they develop by independent mechanisms. In contrast, alpha actin and myosin heavy chain mRNA accumulation was not enhanced in hybrid embryos. These results suggest that multiple mechanisms control anural development.     

Factors necessary for restoring an evolutionary change in an anural ascidian embryo.

The ascidian Molgula oculata has a tailed (or urodele) larva, whereas Molgula occulta develops directly via a tailless (or anural) embryo. Interspecific hybrid embryos produced by fertilizing M. occulta eggs with M. oculata sperm (M. occulta x M. oculata hybrids) can develop urodele larval structures, including a brain pigment cell and a short tail containing a small notochord. Development of larval features differs in individual M. occulta clutches: some eggs develop into hybrids with both a brain pigment cell and a tail, some into hybrids with either a brain pigment cell or a tail, and others into hybrids without urodele features. The expression of a 58-kDa protein (p58), which is present in eggs and embryos of urodele ascidians but lacking in those of most anural species, also varies in expression between different clutches of M. occulta eggs. Western blot and immunofluorescence studies show that p58 expression is correlated with the ability of hybrid embryos to express urodele features. For example, clutches of M. occulta eggs containing relatively high levels of p58 produce many hybrids with both a brain pigment cell and a tail. Differential expression of p58 occurs during oogenesis in M. occulta individuals: p58 is found at similar levels in previtellogenic oocytes, but in some animals it disappears during vitellogenesis, while in others it persists throughout oogenesis and is present in mature eggs. When M. occulta eggs are extracted with Triton X-100, p58 remains in the detergent-insoluble fraction, suggesting that it is associated with the cytoskeleton. In most unfertilized M. occulta eggs, p58 is uniformly distributed, but after fertilization it is localized in the uncleaved zygote and then concentrated in embryonic ectoderm, notochord, and muscle lineage cells. Despite containing high levels of p58, gynogenetic hybrid embryos, produced by fertilizing M. occulta eggs with uv-irradiated M. oculata sperm, develop into hybrids without a brain pigment cell or a tail. The results suggest that both a functional paternal genome and p58 are necessary for restoration of larval features in M. occulta x M. oculata hybrids. The cytoskeletal complex containing p58 may mediate the localization of key maternal factors in the egg or may be involved in cellular interactions during embryogenesis which are responsible for development of urodele cell fates.     

An evolutionary change in the muscle lineage of an anural ascidian embryo is restored by interspecific hybridization with a urodele ascidian

Anural ascidians do not develop into a conventional tailed larva with differentiated muscle cells, however, embryos of some anural ascidian species retain the ability to express acetylcholinesterase (AChE) in a vestigial muscle cell lineage. This study examines the number of AChE-positive cells that develop in the anural ascidian Molgula occulta relative to that in the closely related urodele (tailed) species, Molgula oculata. Histochemical assays showed that M. oculata embryos develop 36 to 38 AChE-positive cells, consistent with the number of tail muscle cells expressed in other urodele ascidians. In contrast, M. occulta embryos develop a mean of only 20 AChE-positive cells in their vestigial muscle lineage. Cleavage-arrested embryos of the anural species express AChE only in B-line blastomeres, showing that the vestigial muscle lineage cells are derived from the primary muscle lineage. Less than the expected number of AChE-positive B-line cells develop in cleavage-arrested anural embryos, however, implying that the allocation of primary muscle lineage cells is decreased. Eggs of the anural species can be fertilized with sperm of the urodele species resulting in the development of some larvae that contain a short tail and/or a brain melanocyte, specific features of urodele larvae. The typical urodele number of AChE-positive cells is restored in some of these hybrid embryos. Both primary and secondary muscle lineages are restored because cleavage-arrested hybrid embryos develop more AChE-positive cells in the B-line blastomeres and supernumerary AChE-positive cells in the A-line blastomeres. Hybrid embryos that develop the urodele complement of AChE-positive cells also form a tail and/or a brain melanocyte showing that restoration of muscle lineage cells is coupled to the development of other urodele features. AChE expression occurred in anural embryos with disorganized or dissociated blastomeres, indicating that AChE expression is determined autonomously. It is concluded that an evolutionary change in the allocation of larval muscle lineage cells occurs during development of the anural ascidian M. occulta which can be restored by interspecific hybridization with the urodele ascidian M. oculata.     

Development of vestigial tail muscle acetylcholinesterase in embryos of an anural ascidian species.

1. The ascidian Molgula arenata produces an anural larva lacking a tail and other structural features of typical urodele larvae in the family Molglidae, yet its embryos developed a histochemically detectable acetylcholinesterase in the tail muscle rudiment. Development of the myoblasts seemed to fail during the neurula stage. 2. Larval enzyme activity occurred at a mean of 5--6% of the level found in the urodele species Molgula occidentalis and Molgula manhattensis, as measured by scanning integrating microdensitometry of the histochemical reaction product. Some anural larvae had as much as 20% of the enzyme activity in urodele larvae. 3. This example of vestigial expression in the absence of other urodele larval features further illustrates the autonomy of a histospecific enzyme development thought to be controlled by an egg cytoplasmic determinant. Partial suppression of the determinant might be the cause of this diminished expression. 4. Two other anural molgulid species, Molgula occulta and Bostrichobranchus pilularis, did not have vestigial larval enzyme and possibly have lost the determinant completely.     

Evolution and development of brain sensory organs in molgulid ascidians

The ascidian tadpole larva has two brain sensory organs containing melanocytes: the otolith, a gravity receptor, and the ocellus, part of a photoreceptor. One or both of these sensory organs are absent in molgulid ascidians. We show here that developmental changes leading to the loss of sensory pigment cells occur by different mechanisms in closely related molgulid species. Sensory pigment cells are formed through a bilateral determination pathway in which two or more precursor cells are specified as an equivalence group on each side of the embryo. The precursor cells subsequently converge at the midline after neurulation and undergo cell interactions that decide the fates of the otolith and ocellus. Molgula occidentalis and M. oculata, which exhibit a tadpole larva with an otolith but lacking an ocellus, have conserved the bilateral pigment cell determination pathway. Programmed cell death (PCD) is superimposed on this pathway late in development to eliminate the ocellus precursor and supernumerary pigment cells, which do not differentiate into either an otolith or ocellus. In contrast to molgulids with tadpole larvae, no pigment cell precursors are specified on either side of the M. occulta embryo, which forms a tailless (anural) larva lacking both sensory organs, suggesting that the bilateral pigment cell determination pathway has been lost. The bilateral pigment cell determination pathway and superimposed PCD can be restored in hybrids obtained by fertilizing M. occulta eggs with M. oculata sperm, indicating control by a zygotic process. We conclude that PCD plays an important role in the evolution and development of brain sensory organs in molgulid ascidians.     

The forkhead gene FH1 is involved in evolutionary modification of the ascidian tadpole larva.

The forkhead gene FH1 encodes a HNF-3beta protein required for gastrulation and development of chordate features in the ascidian tadpole larva. Although most ascidian species develop via a tadpole larva, the conventional larva has regressed into an anural (tailless) larva in some species. Molgula oculata (the tailed species) exhibits a tadpole larva with chordate features (a dorsal neural sensory organ or otolith, a notochord, striated muscle cells, and a tail), whereas its sister species Molgula occulta (the tailless species) has evolved an anural larva, which has lost these features. Here we examine the role of FH1 in modifying the larval body plan in the tailless species. We also examine FH1 function in tailless speciesxtailed species hybrids, in which the otolith, notochord, and tail are restored. The FH1 gene is expressed primarily in the presumptive endoderm and notochord cells during gastrulation, neurulation, and larval axis formation in both species and hybrids. In the tailless species, FH1 expression is down-regulated after neurulation in concert with arrested otolith, notochord, and tail development. The FH1 expression pattern characteristic of the tailed species is restored in hybrid embryos prior to the development of chordate larval features. Antisense oligodeoxynucleotides (ODNs) shown previously to disrupt FH1 function were used to compare the developmental roles of this gene in both species and hybrids. As described previously, antisense FH1 ODNs inhibited endoderm invagination during gastrulation, notochord extension, and larval tail formation in the tailed species. Antisense FH1 ODNs also affected gastrulation in the tailless species, although the effects were less severe than in the tailed species, and an anural larva was formed. In hybrid embryos, antisense FH1 ODNs blocked restoration of the otolith, notochord, and tail, reverting the larva back to the anural state. The results suggest that changes in FH1 expression are involved in re-organizing the tadpole larva during the evolution of anural development.     

Identification of a cytoskeletal protein localized in the myoplasm of ascidian eggs: localization is modified during anural development

The myoplasm of ascidian eggs is a localized cytoskeletal domain that is segregated to presumptive larval tail muscle cells during embryonic development. We have identified a cytoskeletal protein recognized by a vertebrate neurofilament monoclonal antibody (NN18) which is concentrated in the myoplasm in eggs and embryos of a variety of ascidian species. The NN18 antigen is localized in the periphery of unfertilized eggs, segregates with the myoplasm after fertilization, and enters the larval tail muscle cells during embryonic development. Western blots of one-dimensional and two-dimensional gels showed that the major component recognized by NN18 antibody is a 58 x 10(3) Mr protein (p58), which exists in at least three different isoforms. The enrichment of p58 in the Triton X-100-insoluble fraction of eggs and its reticular staining pattern in eggs and embryos suggests that it is a cytoskeletal protein. In subsequent experiments, p58 was used as a marker to determine whether changes in the myoplasm occur in eggs of anural ascidian species, i.e. those exhibiting a life cycle lacking tadpole larvae with differentiated muscle cells. Although p58 was localized in the myoplasm in eggs of four urodele ascidian species that develop into swimming tadpole larvae, this protein was distributed uniformly in eggs of three anural ascidian species. The eggs of two of these anural species contained the actin lamina, another component of the myoplasm, whereas the third anural species lacked the actin lamina. There was no detectible localization of p58 after fertilization or segregation into muscle lineage cells during cleavage of anural eggs. NN18 antigen was uniformly distributed in pre-vitellogenic oocytes and then localized in the perinuclear zone during vitellogenesis of urodele and anural ascidians. Subsequently, NN18 antigen was concentrated in the peripheral cytoplasm of post-vitellogenic oocytes and mature eggs of urodele, but not anural, ascidians. It is concluded that the myoplasm of ascidian eggs contains an intermediate filament-like cytoskeletal network which is missing in anural species that have modified or eliminated the tadpole larva.     

Multiple origins of anural development in ascidians inferred from rDNA sequences

Ascidians exhibit two different modes of development. A tadpole larva is formed during urodele development, whereas the larval phase is modified or absent during anural development. Anural development is restricted to a small number of species in one or possibly two ascidian families and is probably derived from ancestors with urodele development. Anural and urodele ascidians constitute a model system in which to study the evolution of development, but the phylogeny of anural development has not been resolved. Classification based on larval characters suggests that anural species are monophyletic, whereas classification according to adult morphology suggests they are polyphyletic. In the present study, we have inferred the origin of anural development using rDNA sequences. The central region of 18S rDNA and the hypervariable D2 loop of 28S rDNA were amplified from the genomic DNA of anural and urodele ascidian species by the polymerase chain reaction and sequenced. Phylogenetic trees inferred from 18S rDNA sequences of 21 species placed anural developers into two discrete groups corresponding to the Styelidae and Molgulidae, suggesting that anural development evolved independently in these families. Furthermore, the 18S rDNA trees inferred at least four independent origins of anural development in the family Molgulidae. Phylogenetic trees inferred from the D2 loop sequences of 13 molgulid species confirmed the 18S rDNA phylogeny. Anural development appears to have evolved rapidly because some anural species are placed as closely related sister groups to urodele species. The phylogeny inferred from rDNA sequences is consistent with molgulid systematics according to adult morphology and supports the polyphyletic origin of anural development in ascidians. Correspondence to: W.R. Jeffery     

Expression of a muscle determinant gene, macho-1, in the anural ascidian Molgula tectiformis

In anural (tailless) ascidian species, functional embryonic muscle is not formed. In urodele (tailed) ascidians, macho-1 functions as a maternally supplied factor for embryonic muscle formation. The failure of embryonic muscle development in anural ascidians may be due to the suppression of macho-1 expression. In this paper, however, we report the expression of macho-1 in embryos of an anural ascidian, Molgula tectiformis. Although M. tectiformis has lost the developmental potential to form functional embryonic muscle, macho-1 was expressed in a very similar manner as in urodele ascidians. This result, together with those of previous studies, strongly suggests that in M. tectiformis the upstream genetic cascade responsible for muscle formation is intact, while the downstream cascade including the expression of muscle structural genes is severely affected.Electronic Supplementary Material Supplementary material is available for this article at     

Ultrastructural and histochemical study of anural development in the ascidian Molgula pacifica (Huntsman)

Summary Tadpole development is eliminated in the life cycle of the ascidian Molgula pacifica. The elimination of a tailed larva is termed anural development, in contrast to urodele development which is exhibited by most ascidian species. In the present study, transmission electron microscopy and histochemistry were used to gain a better understanding of anural development in M. pacifica. The fine structure of M. pacifica oocytes and fertilized eggs was similar to urodele oocytes and eggs, except that a perivitelline space and test cells were absent. M. pacifica embryos exhibited the typical cleavage pattern of urodele embryos. Gastrulation was initiated at the vegetal pole, as in urodeles, and occurred at the same time as in two urodele species (Molgula manhattensis and Pyura haustor). However, changes in cell shapes and cell movements of the vegetal pole cells that participate in gastrulation were highly modified compared to commonly studied ascidians. The changes in shapes and movements of the vegetal pole cells were minimal and resulted in embryos having a very small archenteron and blastopore. The presence of large, yolky cells in the interior of the embryo likely restricted vegetal cell movements. Two ultrastructurally distinct types of epidermal cells were evident at the gastrula stage. When gastrulae were manually dechorionated from their surrounding mucous-follicular envelope layers, the embryos were already surrounded by a thin tunic. When day 1 juveniles in the process of hatching were sectioned along the anterior-posterior axis, regional differences in cell types were evident. Differentiated muscle cells in the posterior region were not evident. Day 1 M. pacifica juveniles, anural-developing M. provisionalis juveniles and tadpoles from three urodele species were tested for their abilities to express AchE activity. The highest levels of AchE activity were detected in the larval tail muscle cells of urodeles, low levels of activity were detected in the posterior region of M. provisionalis juveniles, whereas M. pacifica juveniles did not exhibit AchE activity. The results are discussed in terms of evolutionary mechanisms responsible for anural development in ascidians. Offprint requests to: W.R. Bates     

Vestigial Brain Melanocyte Development During Embryogenesis of an Anural Ascidian

Anural development was examined in the ascidian Bostrichobranchus digonas using specific markers for differentiated urodele ascidian larval cells and tissues. In this ovoviviparous anural ascidian, eggs, embryos and developing juveniles were present in the gonads, brood sacs, and atrial cavity, respectively. Morphological studies indicated that B. digonas embryos do not develop into tailed larvae with an extended notochord and differentiated muscle cells. In addition, these embryos lack detectable expression of the muscle-specific markers acetylcholinesterase, alpha actin, and myosin heavy chain. In striking contrast to other anural ascidian embryos, however, B. digonas embryos can develop tyrosinase in several melanocyte precursor cells and eventually form a brain pigment cell. The melanocyte does not become part of a definitive brain sensory organ (otolith) and subsequently disappears during metamorphosis. A period of tyrosinase expression was also observed following metamorphosis in which many tyrosinase-positive cells appear in the body of the developing juvenile. The results demonstrate that different urodele features can be uncoupled during the evolution of anural development. The development of a vestigial brain melanocyte also suggests that B. digonas evolved from a urodele ancestor rather than from another anural ascidian lacking a brain pigment cell.     

Evolution of alternate modes of development in ascidians.

Ascidians have evolved alternate modes of development in which the conventional tadpole larva is remodeled or eliminated. Adultation, the precocious development of adult features in the larval head, is caused by superimposing the larval and adult differentiation programs. Caudalization, the addition of muscle cells to the larval tail, is caused by enhancing muscle induction or increasing the number of muscle cell divisions before terminal differentiation. Adultation and caudalization are correlated with increased egg size, suggesting dependence on maternal processes. Anural development, the elimination of the larval stage, is caused by maternal and zygotic events resulting in abbreviation and deletion of larval developmental programs. An example of a maternal change in anural species is the modification of the egg cytoskeleton during oogenesis, whereas a zygotic change may involve altered cell interactions during embryogenesis. Interspecific hybridization experiments suggest that some aspects of anural development may be caused by loss-of-function mutations. The dissociation of developmental programs is a key process in changing the mode of development in ascidians.     

Evolutionary reorganizations of ontogenesis in ascidians of the genus Molgula

The data on comparative, experimental, and molecular embryology of ascidians (genus Molgula) published during the last 15 years have been reviewed. Some representatives of this genus evolved from development with a tailed larva (tadpole) to direct development associated with the loss of larval structures, such as tail, notochord, sensory organs, and differentiated muscles. The data on evolutionary reorganizations of ontogenesis in ascidians of the genus Molgula have been compared with those in sea urchins, anuran amphibians, and some other organisms.     

Evolutionary reorganizations of ontogenesis in ascidians of the genus Molgula

The data on comparative, experimental, and molecular embryology of ascidians (genus Molgula) published during the last 15 years have been reviewed. Some representatives of this genus evolved from development with a tailed larva (tadpole) to direct development associated with the loss of larval structures, such as tail, notochord, sensory organs, and differentiated muscles. The data on evolutionary reorganizations of ontogenesis in ascidians of the genus Molgula have been compared with those in sea urchins, anuran amphibians, and some other organisms.     

The evolution of anural larvae in molgulid ascidians

Ascidians are urochordates, marine invertebrates with non-feeding motile chordate tadpole larvae, except in the family Molgulidae. Urodele, or tailed, Molgulids have typical ascidian chordate tadpole larvae possessing tails with muscle cells, a notochord, and a dorsal hollow nerve cord. In contrast, anural (or tail-less) Molgulids lack a tail and defining chordate features. Molecular phylogenies generated with 18S and 28S ribosomal sequences indicate that Molgulid species fall into at least four distinct clades, three of which have multiple anural members. This refined and expanded phylogeny allows careful examination of the factors that may have influenced the evolution of tail-less ascidians.     

Cellular features of an apoptotic form of programmed cell death during the development of the ascidian, Boltenia villosa

The phylogenetic position of ascidians near the base of the chordate tree makes them ideal organisms for evolutionary developmental studies of programmed cell death (PCD). In the present study, the following key features of an apoptotic form of PCD are described in Boltenia villosa: fragmentation of DNA, increases in plasma membrane permeability, decreases in mitochondrial activity, production of reactive oxygen species (ROS), and caspase activation. First, evidence is presented for apoptosis of cells within the ovary. Later in development, during the early phase of larval tail resorption at the beginning of metamorphosis, some notochord nuclei showed DNA fragmentation and their cell corpses were rapidly eliminated from the larval body. In striking contrast to the rapid demise of notochord cells, larval muscle cells persisted for more than a week within developing juveniles. Rhodamine 123 and MTT experiments suggest that mitochondria within some of the resorbed larval tail muscle cells were metabolically active for more than a week. Furthermore, resorbed tail muscle cells contained a muscle-specific intermediate filament, termed p58, despite relatively high levels of ROS activity and the ubiquitination of their plasma membranes at day two. Corpses of larval tail muscle cells containing aggregated pigment granules survived within juveniles for more than a month, in contrast to the rapid elimination of notochord cells. Evidence consistent with the formation of larval muscle cell apoptotic bodies is presented. The most surprising result of the present study was that caspase-8, usually associated with apoptotic signaling, was activated in larval endoderm cells that develop into adult structures. When the present results were compared to features of PCD previously reported in other ascidians, significant species differences in PCD were revealed.