Maternal information and localized maternal mRNAs in eggs and early embryos of the ascidian Halocynthia roretzi

Summary

The mosaic behavior of blastomeres isolated from ascidian embryos has been taken as evidence that localized ooplasmic factors (cytoplasmic determinants) specify tissue precursor cells during embryogenesis. Experiments involving the transfer of egg cytoplasm have revealed the presence and localization of various kinds of cytoplasmic determinants in eggs of Halocynthia roretzi. Three cell fates, epidermis, muscle and endoderm, are fixed by cytoplasmic determinants. The three kinds of tissue determinants move in different directions during ooplasmic segregation. Prior to the onset of the first cleavage the three kinds of determinants reside in egg regions that correspond to the future fate map of the embryo and then they are differentially partitioned into specific blastomeres. In addition to tissue-specific determinants, there is evidence suggesting that ascidian eggs contain localized cytoplasmic factors that are responsible for controlling the cleavage pattern and morphogenetic movements. Transplantation of posterior-vegetal egg cytoplasm to an anterior-vegetal position causes a reversal of the anterior-posterior polarity of the cleavage pattern. Localized cytoplasmic factors in the posterior-vegetal region are involved in the generation of a unique cleavage pattern. When vegetal pole cytoplasm is transplanted to the animal pole or equatorial position of the egg, ectopic gastrulation occurs at the site of transplantation. This finding supports the idea that vegetal pole cytoplasm specifies the site of gastrulation. Recently, we started a cDNA project to analyze maternal mRNAs. An arrayed cDNA library of fertilized eggs of H. roretzi was constructed, and more than 2000 clones have been partially sequenced so far. To estimate the proportion of the maternal mRNAs that are localized in the egg and embryo, 150 randomly selected clones were examined by in situ hybridization. We found eight mRNAs that are localized in the eight-cell embryo, of which three were localized to the myoplasm (a specific region of the egg cytoplasm that is partitioned into muscle-lineage blastomeres) of the egg, and then to the postplasm of cleavage-stage embryos. These results indicate that the proportion of localized messages is much higher than we expected. These localized maternal messages may be involved in the regulation of various developmental processes.     

Development of a muscle actin specified by maternal and zygotic mRNA in ascidian embryos

In this investigation, we characterize the embryonic and adult actins and describe the embryonic expression of a muscle actin in the ascidian Styela. Two-dimensional polyacrylamide gel electrophoresis showed that embryos, tadpole larvae, and adult organs contain three major and two minor isoforms of actin. Two of the major isoforms, which are present in the mantle, branchial sac, alimentary tract, and gonads of adults and in eggs, embryos, and heads and tails of tadpoles, are likely to be cytoplasmic actins. The third major isoform, which was enriched in the mantle and branchial sac of adults and localized primarily in the tails of tadpoles, is a muscle actin. The muscle actin isoform was not detected in eggs and early embryos. Radioactivity incorporation studies showed that the cytoplasmic actins were synthesized throughout early development, but muscle actin synthesis was first detected between the 16- and 64-cell stages, 2-3 hr after fertilization. Two lines of evidence indicate that embryonic muscle actin synthesis is directed in part by maternal mRNA. First, poly(A)+ RNA isolated from unfertilized eggs directed the synthesis of muscle actin in an mRNA-dependent reticulocyte lysate. Second, muscle actin was synthesized in anucleate egg fragments. Arguments are also presented that muscle actin synthesis is not directed exclusively by maternal mRNA. It is concluded that embryonic and adult Styela exhibit actin heterogeneity, that one of the actin isoforms is a muscle actin, and that the muscle actin is synthesized during embryogenesis under the direction of maternal and zygotic mRNA.     

Two pathways of maternal RNA localization at the posterior-vegetal cytoplasm in early ascidian embryos

A cDNA library prepared from fertilized eggs of the ascidian Halocynthia roretzi was screened for prelocalized mRNAs in the early embryo by means of whole-mount in situ hybridization using a digoxigenin-labeled antisense RNA of each clone. Random mass screening of 150 cDNAs in a fertilized egg yielded six different clones which showed mRNA localization in the posterior-vegetal cytoplasm of the 8-cell embryo. An in situ hybridization study of the detailed spatial distribution of each mRNA in embryos of various stages revealed that there are, in contrast to the identical localization in embryos after the 16-cell stage, two distinct patterns of RNA distribution at earlier stages. One is colocalization with the myoplasm from the prefertilization stage to the 8-cell stage (type I postplasmic RNAs). The other is delayed accumulation of RNA at the posterior-vegetal cytoplasm after fertilization (type II postplasmic RNAs). We found that both types of RNAs associate with the cytoskeleton, but that they show different sensitivities to inhibitors of the cytoskeleton; translocation of the type I RNAs is dependent upon microfilaments during the first phase of ooplasmic segregation and dependent upon microtubules during the second phase of segregation, whereas translocation of the type II RNAs is dependent upon microfilaments throughout ooplasmic segregation. These results show that there are two pathways for the localization of the RNAs at the posterior-vegetal cytoplasm in the 8-cell embryo of the ascidian H. roretzi.     

Gap-junctional permeability in early and cleavage-arrested ascidian embryos.

Using the whole-cell voltage clamp technique, we have studied junctional conductance (Gj), and Lucifer Yellow (LY) coupling in 2-cell and 32-cell ascidian embryos. Gj ranges from 17.5 to 35.3 nS in the 2-cell embryo where there is no passage of LY, and from 3.5 to 12.2 nS in the later embryo where LY dye spread is extensive. In both cases, Gj is independent of the transjunctional potential (Vj). Manually apposed 2-cell or 32-cell embryos established a junctional conductance of up to 10 nS within 30 min of contact. Furthermore, since we did not observe any significant number of cytoplasmic bridges at the EM and Gj is sensitive to octanol, it is probable that blastomeres in the 2-cell and 32-cell embryos are in communication by gap junctions. In order to compare Gj in the two stages and to circumvent problems of cell size, movement and spatial location, we used cytochalasin B to arrest cleavage. Gj in cleavage-arrested 2-cell embryos ranged from 25.0 to 38.0 nS and remained constant over a period of 2.5 h. LY injected into a blastomere of these arrested embryos did not spread to the neighbour cell until they attained the developmental age of a 32- to 64-cell control embryo. Our experiments indicate a change in selectivity of gap junctions at the 32-cell stage that is not reflected by a macroscopic change in ionic permeability.     

The location of maternal mRNA in eggs and embryos

Recent studies have shown that some maternal mRNAs are localized in specific cytoplasmic regions of eggs and embryos and are rearranged in concert with the cytoplasmic movements that fix the embryonic axes. The localization and ooplasmic segregation of mRNA molecules may be mediated by their association with specific egg cytoskeletal domains.     

An early role of maternal mRNA in establishing the dorsoventral pattern in pelle mutant Drosophila embryos

Mutations of the maternal effect locus pelle (pll) cause dorsalized Drosophila embryos. In extreme mutants, the embryo develops into a long hollow tube of dorsal cuticular structures with no sign of ventral pattern elements. Injection of wild-type cytoplasm or poly(A)+RNA into mutant pll embryos partially restores the normal pattern. Rescuing activity is present in the wild-type cytoplasm until the late blastoderm stage, but is already absent from the poly(A)+RNA fraction by the time of pole cell formation. At the same time, pll embryos fail to respond to injected biologically active poly(A)+RNA. This indicates that pll+ mRNA is lost early from the pool of maternal RNA and that there is a non-RNA component of rescue. This component, most likely the pll+ protein, appears to be unequally distributed in wild-type embryos.     

Patterning of an ascidian embryo along the anterior-posterior axis through spatial regulation of competence and induction ability by maternally localized PEM

In ascidian embryos, anterior-posterior (A-P) patterning of the vegetal cells is regulated by posteriorizing activities of a localized egg region known as posterior-vegetal cortex/cytoplasm (PVC). PEM is an essential component of the PVC and is involved in the posterior-specific cell cleavage pattern. Here we report a novel function of PEM independently of its function in cleavage regulation; it controls cell fate by excluding competence to respond to the FGF signal for notochord induction from posterior-vegetal cells. PEM was found to regulate the nuclear accumulation of β-catenin, an upstream activator of the competence factor. PEM also influences A-P patterning in the animal hemisphere. It was found to regulate FGF signal expression and restrict the occurrence of brain induction only in the anterior region. Our results suggest a model in which PEM patterns the embryo along the A-P axis through regulation of the spatial distribution of competence and induction ability.     

Differentiation of histospecific ultrastructural features in cells of cleavage-arrested early ascidian embryos

Summary Ultrastructural features of histospecific differentiation were found in early cleavage stage ascidian embryos treated with cytochalasin B and held thereby in cleavagearrest until hatching time. Markers characteristic of tissue differentiation during normal embryonic and larval stages ofCiona intestinalis were expressed in muscle and two brain cell lineages of cleavage-arrested whole embryos and in epidermal and notochordal cell lineages of cleavage-arrested partial embryos. These features were muscle myofilaments and myofibrils, melanosomes of the brain pigment cells, cilium-derived structures present in a proprioceptive brain cell, extracellular test material of epidermal cell origin, and the sheath filaments, membrane leaflets, and vacuolar colloid associated with notochord cells. All of these ultrastructural markers of differentiation were blocked in their development by treatment of gastrula stage embryos with actinomycin D, an inhibitor of RNA synthesis, and presumably result from the expression of new gene activity. At the time of cleavage-arrest the five cell lineages studies still contained two or more unsegregated lineage pathways. Subsequent developmental autonomy within the lineages is consistent with the hypothesis of segregation during early development of functionally independent gene regulatory factors.     

Isolation and characterization of beta-catenin downstream genes in early embryos of the ascidian Ciona savignyi

Nuclear localization of beta-catenin is most likely the first step of embryonic axis formation or embryonic cell specification in a wide variety of animal groups. Therefore, the elucidation of beta-catenin target genes is a key research subject in understanding the molecular mechanisms of the early embryogenesis of animals. In Ciona savignyi embryos, nuclear accumulation of beta-catenin is the first step of endodermal cell specification. Previous subtractive hybridization screens of mRNAs between beta-catenin-overexpressed embryos and nuclear beta-catenin-depleted embryos have resulted in the identification of beta-catenin downstream genes in Ciona embryos. In the present study, I characterize seven additional beta-catenin downstream genes, Cs-cadherinII, Cs-protocadherin, Cs-Eph, Cs-betaCD1, Cs-netrin, Cs-frizzled3/6, and Cs-lefty/antivin. All of these genes were expressed in vegetal blastomeres between the 16-cell and 110-cell stages, although their spatial and temporal expression patterns were different from one another. In situ hybridizations and real-time PCR revealed that the expression of all of these genes was up-regulated in beta-catenin-overexpressed embryos, and down-regulated in beta-catenin-suppressed embryos. Therefore, the accumulation of beta-catenin in the nuclei of vegetal blastomeres activates various vegetally expressed genes with potentially important functions in the specification of these cells.     

Functions of maternal mRNA in early development

In this review, the types of mRNAs found in oocytes and eggs of several animal species, particularly Drosophila, marine invertebrates, frogs, and mice, are described. The roles that proteins derived from these mRNAs play in early development are discussed, and connections between maternally inherited information and embryonic pattern are sought. Comparisons between genetically identified maternally expressed genes in Drosophila and maternal mRNAs biochemically characterized in other species are made when possible. Regulation of the meiotic and early embryonic cell cycles is reviewed, and translational control of maternal mRNA following maturation and/or fertilization is discussed with regard to specific mRNAs.     

Antisense oligodeoxyribonucleotide-directed cleavage of maternal mRNA in Xenopus oocytes and embryos

We have investigated the effect of specific antisense oligodeoxynucleotides (oligos) on endogenous histone H4 mRNA in Xenopus oocytes, eggs and embryos. In unfertilised eggs and non-matured oocytes, one 20-mer oligo (H4-1) mediated the RNAse H-like cleavage of up to 95% of H4 mRNA (which included polysomal mRNA), and cleavage was still obtained when the size of the oligo was reduced to a 10-mer; no cleavage was observed with 6- and 8-mers. The residual uncleaved mRNA appeared to be completely inaccessible to H4-1 since a second injection caused no further cleavage. A second 20-mer (H4-2) directed against a different region of H4 mRNA was much less effective (less than 5% cleavage). In fertilised embryos, injections of H4-1 and an oligo directed against the localised Vg1 mRNA caused less cleavage than in oocytes and also showed signs of inducing localised, non-specific mRNA cleavage. However we have been able to prepare fertilised embryos devoid of Vg1 mRNA by maturing and fertilising oligo-injected oocytes in vitro.