Negative spatial regulation of the lineage specific CyIIIa actin gene in the sea urchin embryo

The CyIIIa.CAT fusion gene was injected into Strongylocentrotus purpuratus eggs, together with excess ligated competitor sequences representing subregions of the CyIIIa regulatory domain. In this construct, the chloramphenicol acetyltransferase (CAT) reporter gene is placed under the control of the 2300 nucleotide upstream regulatory domain of the lineage-specific CyIIIa cytoskeletal actin gene. CAT mRNA was detected by in situ hybridization in serial sections of pluteus stage embryos derived from the injected eggs. When carrier DNA lacking competitor CyIIIa fragments was coinjected with CyIIIa.CAT, CAT mRNA was observed exclusively in aboral ectoderm cells, i.e. the territory in which the CyIIIa gene itself is normally expressed (as also reported by us previously). The same result was obtained when five of seven different competitor subfragments bearing sites of DNA-protein interaction were coinjected. However, coinjection of excess quantities of either of two widely separated, nonhomologous fragments of the CyIIIa regulatory domain produced a dramatic ectopic expression of CAT mRNA in the recipient embryos. CAT mRNA was observed in gut, mesenchyme cells and oral ectoderm in these embryos. We conclude that these fragments contain regulatory sites that negatively control spatial expression of the CyIIIa gene.     

Sea urchin actin gene subtypes. Gene number, linkage and evolution

The actin gene family of the sea urchin Strongylocentrotus purpuratus was analyzed by the genome blot method, using subcloned probes specific to the 3' terminal non-translated actin gene sequence, intervening sequence and coding region probes. We define an actin gene subtype as that gene or set of genes displaying homology with a given 3' terminal sequence probe, when hybridized at 55 degrees C, 0.75 M-Na+. By determining the often polymorphic restriction fragment band pattern displayed in genome blots by each probe, all, or almost all of the actin genes in this species could be classified. Our evidence shows that the S. purpuratus genome probably contains seven to eight actin genes, and these can be assigned to four subtypes. Studies of the expression of the genes (Shott et al., 1983) show that the actin genes of three of these subtypes code for cytoskeletal actins (Cy), while the fourth gives rise to a muscle-specific actin (M). We denote the array of S. purpuratus actin genes indicated by our data as follows. There is a single CyI actin gene, two or possibly three CyII genes (CyIIa, CyIIb, and possibly CyIIc), three CyIII actin genes (CyIIIa, CyIIIb, CyIIIc), and a single M actin gene. Comparative studies were carried out on the actin gene families of five other sea urchin species. At least the CyIIa and CyIIb genes are also linked in the Strongylocentrotus franciscanus genome, and this species also has a CyI gene, an M actin gene and at least two CyIII actin genes. It is not clear whether it also possesses a CyIIc actin gene, or a CyIIIc actin gene. The genome of a more closely related congener, Strongylocentrotus dr?bachiensis, includes 3' terminal sequences suggesting the presence of a CyIIc gene. In S. franciscanus and S. dr?bachiensis the first intron of the CyI gene has remained homologous with intron sequences of both the CyIIa and CyIIb genes, indicating a common origin of these three linked cytoskeletal actin genes. Of the four S. purpuratus 3' terminal subtype probe sequences only the CyI 3' terminal sequence has been conserved sufficiently during evolution to permit detection outside of the genus Strongylocentrotus. An unexpected observation was that a sequence found only in the 3' untranslated region of the CyII actin gene in the DNA of S. dr?bachiensis and S. purpuratus is represented as a large family of interspersed repeat sequences in the genome of S. franciscanus.     

Regulatory elements from the related spec genes of Strongylocentrotus purpuratus yield different spatial patterns with a lacZ reporter gene.

The Spec1 and Spec2 genes of Strongylocentrotus purpuratus are closely associated with the differentiation of aboral ectoderm. To examine cis-regulatory elements involved in the spatial expression of the Spec genes, we fused the Escherichia coli lacZ gene containing a nuclear targeting signal to 5'flanking DNA plus 5' untranslated leader sequences from Spec1, Spec2a, and Spec2c. All three genes contain 700 bp of highly conserved DNA in their upstream regions, but in Spec1 and Spec2c large insertions interrupt the conserved regions. The Spec-lacZ reporter gene plasmids were microinjected into eggs of S. purpuratus, Lytechinus variegatus, and L. pictus, and beta-galactosidase activity was determined in situ by X-gal staining. The Spec2a-lacZ fusion gene, which contained 1516 bp of 5' flanking DNA and 18 bp of 5' untranslated leader sequence, was preferentially expressed in aboral ectoderm cells in all three species. The Spec1-lacZ fusion gene was expressed in a strikingly different fashion--preferentially in primary and secondary mesenchyme cells, occasionally in aboral ectoderm cells, and less often in oral ectoderm and endoderm cells. The staining pattern was the same in either homologous or heterologous embryos. The Spec2c-lacZ fusion gene, like Spec2a-lacZ, was preferentially expressed in aboral ectoderm, but staining of other cell types was frequently observed. To further delineate sequences required for correct spatial expression, we deleted 800 bp of 5' flanking DNA from the Spec2a-lacZ fusion gene, resulting in a delta Spec2a-lacZ fusion gene that contained only the conserved DNA region. This gene fusion showed preferential expression in aboral ectoderm cells. However, the cell type specificity was not as great as with the parental Spec2a-lacZ plasmid. These experiments implied that the conserved DNA region, associated with all Spec genes examined, was insufficient for complete aboral ectoderm specificity, and suggested that a spatial repressor element existed between -1516 and -697 bp in the 5' flanking DNA of Spec2a.     

Introduction and Expression of Recombinant Genes in Ascidian Embryos

In order to examine the expression of exogenous genes introduced into ascidian eggs, two recombinant plasmids pmiwZ and pHrMA4aCAT were microinjected into the cytoplasm of fertilized eggs of Ciona savignyi and Halocynthia roretzi , respectively. The plasmid pmiwZ contains the coding sequence of bacterial β-galactosidase gene ( lac-Z ) fused with animal gene promoters, while pHrMA4aCAT was constructed by fusing about 1.4-kb long 5' flanking region of H. roretzi muscle actin gene HrMA4a with bacterial chloramphenicol acetyltransferase gene ( CAT ). Injection of approximately 160 pl of 10 μg/ml pmiwZ DNA into Ciona eggs did not affect the embryogenesis, although introduction of the same volume of 30 μg/ml pmiwZ DNA resulted in abnormal development of injected eggs. When the expression of lac-Z was examined by histochemical detection of the enzyme activity, the expression was evident in the early tailbud embryos and later stage embryos, and larvae, irrespective of linear or circular form of the plasmid. The enzyme activity appeared in various cell-types including epidermis, nervous system, endoderm, mesenchyme, notochord, and muscle. In contrast, when pHrMA4aCAT was introduced into Halocynthia eggs and the appearance of CAT protein was examined later by the anti-CAT antibody, the CAT expression was restricted to muscle cells. These results indicate that the recombinant genes introduced into ascidian eggs could express during embryogenesis and that the 1.4-kb long 5' flanking region of HrMA4a contains regulatory sequences enough for the appropriate spatial and temporal expression of the gene.     

Mosaic incorporation and regulated expression of an exogenous gene in the sea urchin embryo

A fusion gene construct in which the bacterial chloramphenicol acetyltransferase (CAT) gene is controlled by CyIIIa actin gene cis-regulatory sequences was injected into unfertilized eggs of the sea urchin Strongylocentrotus purpuratus. The distribution of CAT DNA sequences was measured directly by in situ hybridization in squashed 24-hr blastula preparations derived from these eggs. Earlier studies had shown that stable mosaic incorporation of the exogenous DNA occurs during cleavage, after which the exogenous sequences replicate at approximately the pace of the host cell genomes. The fractions of embryonic cells observed in this study to include CAT DNA sequences imply that their stable incorporation into a replicating nuclear form occurs most often in a single cell at the 3rd or 4th cleavage stages, though it may occur as early as 2nd cleavage, or as late as 7th cleavage. Corroborative measurements were carried out by the same method on squashed preparations of embryos at earlier stages, and by in situ hybridizations of CAT mRNA, both in dissociated embryos and in cytological sections of 72-hr pluteus-stage embryos. Hybridizations to CAT mRNA and to CAT DNA were carried out on alternate sections of several embryos. The results confirm unequivocally that although CAT mRNA appears only in the aboral ectoderm in embryos derived from eggs injected with the CyIIIa.CAT fusion gene, the exogenous sequences are indeed present, though silent, in the various other cell types of the late embryo.