The expression of the Specl gene of Strongylocentrotus purpuratus and its Lytechinus pictus homologue LpSl was analyzed in reciprocal hybrid embryos of these two species of sea urchin. While the time course of accumulation of Specl mRNA was nearly normal in hybrid embryo populations, the accumulation of LpSl mRNA was not. This was particularly evident in plutei, where the level of LpSl mRNA was less than 5% that in normal L. pictus plutei. In situ hybridization analysis of serial sections indicated that LpSl mRNA was detectable in only about 2% of hybrid plutei in either cross, whereas Specl mRNA was present in nearly all hybrid plutei; expression of either homologue was appropriately restricted to the aboral ectoderm. In crosses of L. pictus eggs with S. purpuratus sperm (LpSp), about 1% of hybrid plutei expressed LpSl RNA in most or all aboral ectoderm cells at normal levels, and did not express Specl RNA; in another 1% of the LpSp hybrid plutei the Specl and LpSl transcripts were present at normal levels in complementary, non-overlapping patches of contiguous aboral ectoderm cells. In the reciprocal SpLp cross, each hybrid pluteus expressed either only the LpSl gene (about 2%) or only the Specl gene through-out the aboral ectoderm. In SpLp hybrid gastrulae the level of LpSl mRNA was less restricted; about 2% of the embryos contained only LpSl RNA, and about half expressed only Specl transcripts, but in the remaining embryos Specl and LpSl transcripts were coexpressed in the same aboral ectoderm cells. Thus, the predominant expression of the Specl gene in hybrid embryos results from the developmentally progressive restriction of transcription of the LpSl gene to aboral ectoderm cells not expressing the Specl gene in a few hybrid embryos, while most hybrid plutei express the Specl gene exclusively.

A classical approach used to study early embryonic development in sea urchins is the construction of hybrid embryos by fertilizing the eggs of one species with the sperm from another. In early experiments the analysis of hybrid embryos was instrumental in distinguishing the effects of the maternal cytoplasm and the embryonic genome on the events of embryogenesis, and thus was crucial in establishing a relationship between gene activity and embryonic development (reviewed by Morgan, 1927, and Davidson, 1976). Sea urchin embryo hybrids have also been exploited more recently, using biochemical and molecular biological markers, to address the question of how and when maternal and zygotic genetic information is utilized in the developing embryo (Harding et al. 1954; Fedeca-Bruner et al. 1971; McClay and Hausman, 1975; Maxson and Egrie, 1980; Tufaro and Brandhorst, 1982; Crain and Bushman, 1983; Conlon et al. 1987; Bullock et al. 1988; Nisson et al. 1989a; Brandhorst et al. 1991). The usual interpretation of the sum of the classical analyses of the morphological features of hybrid embryos is that the maternal influence is predominant during early development, and that the influence of the embryonic genome becomes evident only at postgastrular stages (Davidson, 1976). However, expression of specific paternal genes can be detected at all stages in embryo hybrids, even as early as the 2-cell stage (Maxson and Egrie, 1980).

An interesting and potentially informative aspect of gene expression in hybrid embryos is the predominance of expression of some S, purpuratus genes in reciprocal hybrid embryos of S. purpuratus and L. pictus. The patterns of protein synthesis of hybrid embryos, analyzed by gel electrophoresis, closely resemble those of S. purpuratus embryos, with reduced or undetectable synthesis of several proteins specific to L.pictus (Tufaro et al. 1982; Brandhorst et al. 1991). Several cloned cDNAs have been indentified corresponding to L. pictus mRNAs whose accumulation is restricted in hybrid embryos (Conlon et al., 1987). Conversely, several 5. purpuratus genes, such as those encoding different actins (Bullock et al. 1988; Nisson et al. 1989a), show normal temporal and spatial patterns of expression in reciprocal hybrid embryos. It is thus likely that there is normal expression of S. purpuratus homologues of L. pictus genes whose expression is restricted in hybrid embryos. This predominant expression of S. purpuratus genes is not a parental effect nor a maternal cytoplasmic effect since it occurs regardless of whether the L. pictus genome is contributed by the egg or the sperm.

The Spc1 gene of S. purpuratus encodes a calcium binding protein that is expressed only in aboral ectoderm cells (Bruskin et al. 1982; Carpenter et al. 1984). The LpSl gene of L. pictus shows the same pattern of expression and encodes a protein similar to the Spc1 protein; LpSl and Spc1 proteins are 7 immunologically cross reactive (Xiang et al. 1988, 1991). Based on these observation it is likely that these two genes are homologous. Previously, we showed that in reciprocal hybrid embryos Spc1 transcripts accumulate normally (taking into account gene dosage), while LpSl transcripts are considerably reduced compared to Spc1 transcripts in hybrid plutei or LpSl transcripts in L. pictus plutei, and that this restriction can be accounted for by reduced transcription of the LpSl gene (Brandhorst et al. 1991). The investigations reported here further characterize the predominant expression of Spc1 genes over LpSl genes in hybrid embryos by using in situ hybridization to ask where the two mRNAs are expressed in individual hybrid embryos. These studies indicate that both genes are expressed only in aboral ectoderm cells of hybrid plutei, but that most embryos express only the Spc1 gene, while 1–2% express only the LpSl gene, and a few LpSp hybrid plutei (from crosses of L. pictus eggs with S. purpuratus sperm) express both genes but in mutually exclusive patches of aboral ectoderm cells.

Embryo culture

The fertilization and culture of normal S. purpuratus and L. pictus embryos was carried out at 15°C according to standard procedures. Fertilization of the eggs of each species with sperm from the other, to produce interspecies hybrid embryos, sometimes required special steps such as fertilization with highly concentrated sperm suspensions or pretreatment of eggs with light trypsin digestion. The procedures that were used have been described previously by Conlon et al. (1987) and Nisson et al. (1989a). The hybrid embryos formed healthy plutei having predominantly maternal morphological features.

Synthesis of riboprobes and RNase protection assays

Riboprobes that specifically detect the LpSl or Spc1 mRNAs were synthesized from vectors containing either 440 nucleotides of LpSl mRNA sequence (from W. Klein) or 400 nucleotides of Spc1 mRNA sequence (from R. and L. Angerer) as described by Brandhorst et al. (1991). For RNase protection assays the probes were labelled with [32P]UTP to a specific activity of 2.7 x 108 d.p.m./μg; for in situ hybridization they were labelled with 35S-UTP to a specific activity of 1.3 × 109d.p.m./μg as described by Bullock et al. (1988) and Nisson et al. (1989a and 1989b). Total RNA was isolated from all embryos as described by Bullock et al. (1988). Five μ g of RNA was hybridized with 5 x 105 d.p.m. of either probe in 80% formamide, 40 mM PIPES (pH 6.7), 0.4 M NaCl, 1 mM EDTA at 45°C overnight and RNase digestion was with 2.5 ; μg/ml RNase A and 0.14 μg/ml RNase T1 in 0.9 M NaCl, 10 mM Tris (pH 7.5), 5 mM EDTA at 20°C for 1 hour. The resistant fragments were separated by electrophoresis on a 5% polyacrylamide, 8 M urea gel, which was then dried and autoradiographed.

In situ hybridization

In situ hybridization was performed essentially as described by Angerer and Angerer, (1981) and Cox et al. (1983) with minor modifications described by Nisson et al. (1989a). Hybridization of the riboprobes was at 50°C in 0.6 M NaCl, 10 mM Tris (pH 7.4), 0.5 mM EDTA, 1 mg/ml BSA, 0.02% PVT, 0.02% Ficoll, 150 g/ml yeast tRNA, 10% dextran sulfate, 20 mM DTT and 50% formamide. RNase digestion was with 0.25 μg/ml of RNase A in 0.9 M NaCl, 10 mM DTT at 37°C for 30 minutes. Alternate serial sections of the same sets of embryos were separated by hand from intact ribbons of paraffin and every other section was placed on one slide. The number of sections of a single embryo that could be identified after in situ hybridization ranged from only 2 or 3 up to 20. For each of the embryos shown in Figs 4 and 5 at least six serial sections were identified and used for interpretation; for those in Figs 6 and 7 at least four sections were examined.

Preferential expression of Spc1 mRNA over LpSl mRNA in hybrid embryos

In populations of advanced hybrid embryos of S. purpuratus and L. pictus expression of the S. purpuratus Spc1 gene is predominant over the expression of its presumed L. pictus homologue LpSl, as reflected in the relative rates of synthesis and amounts of the respective proteins and the levels of the mRNAs (Brandhorst et al. 1991). To examine this phenomenon further we asked whether this predominant expression of Spc1 genes occurs throughout development, and whether it is uniformly observed in all hybrid embryos produced by a cross. There is little maternal Spc1 or LpSl mRNA stored in eggs. The transcripts begin to accumulate in presumptive ectoderm cells of early blastulae (Bruskin et al. 1982; Tomlinson et al. 1990). When the relative amounts of the Spc1 and LpSl mRNAs were measured throughout development of normal and hybrid embryos, using an RNase protection assay, several points were evident (Fig. 1 and Table 1). First, the levels of Spc1 mRNA were higher than LpSl mRNA in hybrid embryos at all stages, regardless of direction of the cross. Spc1 mRNA accumulated in hybrid embryos with the same time course and to approximately the same levels as in normal S. purpuratus embryos. In contrast, expression of the LpSl mRNA in hybrids was not normal. In both sets of hybrids (SpLp and LpSp) the level of LpSl mRNA increased to the gastrula stage (up to 46.6 and 21.6% of normal, respectively), but then declined sharply to 4.1 and 3.2% of the levels found in normal plutei of similar stage (Table 1).

Table 1.

Levels of expression of Spc1 and LpSl RNAs in hybrid embryos*

Levels of expression of Spc1 and LpSl RNAs in hybrid embryos*
Levels of expression of Spc1 and LpSl RNAs in hybrid embryos*
Fig. 1.

Detection of LpSl and Spc1 mRNA during development of normal and hybrid embryos. 32P-labeled riboprobes specific for either LpSl (panel A) or Spc1 (panel B) transcripts were hybridized to 5 μg of total RNA from eggs (e), early blatulae (eb), gastrulae (g) and plutei (p) from normal and hybrid embryos. After digestion with RNase A the resistant material was electrophoresed on a 5% polyacrylamide gel and autoradiographed for four hours (panel A) or one hour (panel B) at —80°C with an intensifying screen. The arrows point to the protected fragments of 440 nucleotides (panel A) and 400 nucleotides (panel B). The probes are shown in the lanes designated pr.

Fig. 1.

Detection of LpSl and Spc1 mRNA during development of normal and hybrid embryos. 32P-labeled riboprobes specific for either LpSl (panel A) or Spc1 (panel B) transcripts were hybridized to 5 μg of total RNA from eggs (e), early blatulae (eb), gastrulae (g) and plutei (p) from normal and hybrid embryos. After digestion with RNase A the resistant material was electrophoresed on a 5% polyacrylamide gel and autoradiographed for four hours (panel A) or one hour (panel B) at —80°C with an intensifying screen. The arrows point to the protected fragments of 440 nucleotides (panel A) and 400 nucleotides (panel B). The probes are shown in the lanes designated pr.

The demonstration that the level of Spc1 mRNA is higher than LpSl mRNA in populations of hybrid embryos indicated that expression of the Spc1 gene is predominant over the LpSl gene, but did not show the levels of expression of these two mRNAS or their distribution within individual embryos. To address this question we first carried out in situ hybridization on sections of plutei of LpSp embryos with riboprobes that are specific for LpSl or Spc1 mRNAs. As seen in Fig. 2, hybridization with the Specl-specific probe showed that Spc1 mRNA was present in most of the hybrid plutei, and that it was localized in the aboral ectoderm cells as it is in normal embryos. This correct spatial distribution of the Spc1 mRNA in hybrid plutei, together with the observation that the timing of its accumulation is normal during hybrid embryo development, indicates that the expression of the Spc1 gene is regulated normally in these hybrid embryos. However, when in situ hybridization was used to detect the LpSl mRNA, which is present at less than 5% of the normal level in LpSp plutei, a very different result was seen (Fig. 3). In this case a small fraction of the embryos (1.8% of 800 embryos that were counted) were expressing the LpSl mRNA at approximately normal levels in aboral ectoderm cells, while the remaining approximately 98% showed no detectable LpSl mRNA.

Fig. 2.

Aboral ectoderm-specific expression of the S. purpuratus Specl gene in normal and LpSp hybrid plutei. A single strand riboprobe specific for Specl mRNA was hybridized to sections of normal S. purpuratus plutei (panel A), and LpSp plutei (panel B). Panel C contains an enlargement of a single LpSp pluteus showing Spc1 mRNA restricted to the aboral ectoderm. In both cases virtually all of the embryos were seen to express the Specl mRNA and it was localized to the aboral ectoderm. The bright embryo with unrecognizable morphology in panel B is atypical, and we conclude that it is an artifact.

Fig. 2.

Aboral ectoderm-specific expression of the S. purpuratus Specl gene in normal and LpSp hybrid plutei. A single strand riboprobe specific for Specl mRNA was hybridized to sections of normal S. purpuratus plutei (panel A), and LpSp plutei (panel B). Panel C contains an enlargement of a single LpSp pluteus showing Spc1 mRNA restricted to the aboral ectoderm. In both cases virtually all of the embryos were seen to express the Specl mRNA and it was localized to the aboral ectoderm. The bright embryo with unrecognizable morphology in panel B is atypical, and we conclude that it is an artifact.

Fig. 3.

Expression of LpSl mRNA in LpSp hybrid plutei. A single strand riboprobe specific for LpSl mRNA was hybridized to sections of normal S. purpuratus plutei (panel A) and LpSp plutei (panel B). Panel C shows four examples of LpSl mRNA expression in different LpSp plutei. The field shown in panel B does not accurately reflect the fraction of hybrids that express the LpSl mRNA, since only 1.8% of over 800 embryos that were scored contained detectable levels of this mRNA.

Fig. 3.

Expression of LpSl mRNA in LpSp hybrid plutei. A single strand riboprobe specific for LpSl mRNA was hybridized to sections of normal S. purpuratus plutei (panel A) and LpSp plutei (panel B). Panel C shows four examples of LpSl mRNA expression in different LpSp plutei. The field shown in panel B does not accurately reflect the fraction of hybrids that express the LpSl mRNA, since only 1.8% of over 800 embryos that were scored contained detectable levels of this mRNA.

Fig. 4.

Three examples of LpSp plutei that are expressing only the LpSl gene. Adjacent serial sections of the same embryo were hybridized either with the LpSl-specific riboprobe (top) or with the Specl-specific probe (bottom). The phase-contrast image of each embryo is on the left and the dark-field image of the same section is on the right.

Fig. 4.

Three examples of LpSp plutei that are expressing only the LpSl gene. Adjacent serial sections of the same embryo were hybridized either with the LpSl-specific riboprobe (top) or with the Specl-specific probe (bottom). The phase-contrast image of each embryo is on the left and the dark-field image of the same section is on the right.

Fig. 5.

Expression of both Spc1 homologue mRNAs in individual LpSp plutei. Adjacent serial sections of four different embryos are shown; for each embryo the section on the top was hybridized with the LpSl probe while the section on the bottom was hybridized with the Spc1 probe. The phase-contrast image of each section is on the left and the dark-field image is on the right.

Fig. 5.

Expression of both Spc1 homologue mRNAs in individual LpSp plutei. Adjacent serial sections of four different embryos are shown; for each embryo the section on the top was hybridized with the LpSl probe while the section on the bottom was hybridized with the Spc1 probe. The phase-contrast image of each section is on the left and the dark-field image is on the right.

Fig. 6.

Expression of the LpSl gene in SpLp plutei. Adjacent sections of three different embryos that are expressing the LpSl mRNA are shown, demonstrating the localization of this mRNA in the aboral ectoderm and the lack of expression of Spc1 mRNA. These embryos represent 2% of the population, and the remaining embryos express only Spc1 mRNA, which is localized to the aboral ectoderm. In each panel the phase-contrast image of the embryo is on the left and the dark-field image of the same embryo is on the right; the upper dark-field panel shows a section hybridized with the LpSl probe and the lower panel shows the adjacent section hybridized with the Spc1 probe.

Fig. 6.

Expression of the LpSl gene in SpLp plutei. Adjacent sections of three different embryos that are expressing the LpSl mRNA are shown, demonstrating the localization of this mRNA in the aboral ectoderm and the lack of expression of Spc1 mRNA. These embryos represent 2% of the population, and the remaining embryos express only Spc1 mRNA, which is localized to the aboral ectoderm. In each panel the phase-contrast image of the embryo is on the left and the dark-field image of the same embryo is on the right; the upper dark-field panel shows a section hybridized with the LpSl probe and the lower panel shows the adjacent section hybridized with the Spc1 probe.

Fig. 7.

Expression of the LpSl and Specl genes in SpLp gastrulae. Adjacent sections of two embryos are shown representing the two categories of these embryos that are expressing LpSl mRNA. The phase-contrast images are on the left, dark-field image on the right, the LpSl probe is on top and the Specl probe is on the bottom in both panels. Panel A displays an example of an embryo that is expressing LpSl RNA in the presumptive aboral ectoderm and is not expressing Specl mRNA (1.7% of these embryos were in this category). Panel B contains an example of an embryo that is expressing both LpSl and the Specl mRNA in overlapping cells of the presumptive aboral ectoderm. Approximately 50% of the hybrid gastrulae fall into this category. In the remaining embryos (about 50%, not shown) localized expression of Specl mRNA was seen in the presumptive aboral ectoderm and no LpSl mRNA was detected.

Fig. 7.

Expression of the LpSl and Specl genes in SpLp gastrulae. Adjacent sections of two embryos are shown representing the two categories of these embryos that are expressing LpSl mRNA. The phase-contrast images are on the left, dark-field image on the right, the LpSl probe is on top and the Specl probe is on the bottom in both panels. Panel A displays an example of an embryo that is expressing LpSl RNA in the presumptive aboral ectoderm and is not expressing Specl mRNA (1.7% of these embryos were in this category). Panel B contains an example of an embryo that is expressing both LpSl and the Specl mRNA in overlapping cells of the presumptive aboral ectoderm. Approximately 50% of the hybrid gastrulae fall into this category. In the remaining embryos (about 50%, not shown) localized expression of Specl mRNA was seen in the presumptive aboral ectoderm and no LpSl mRNA was detected.

This surprising result demonstrates that the higher level of expression of the Spc1 gene compared to the LpSl gene reflects the exclusive accumulation of the Spc1 mRNA in most LpSp hybrid plutei. Because previous studies have demonstrated that there is no significant loss of LpSl DNA in hybrid embryos (Brandhorst et al. 1991), this result cannot be explained by the absence of the LpSl gene. Interestingly, within those embryos that did express the LpSl mRNA, it was found only in cells of the aboral ectoderm, in sections where the aboral ectoderm could be identified un-equivocally. Because there are examples where the orientation or the aberrant morphology of the hybrid embryos make it impossible to definitively distinguish oral and aboral ectoderm we cannot rule out completely the possibility that there sometimes is expression in the oral ectoderm. However, there were no examples of LpSl mRNA expression in endoderm.

Non-overlapping, aboral ectoderm-specific expression of Specl and LpSl in LpSp plutei

Examination of many of the LpSp plutei that express LpSl mRNA revealed that, although LpSl mRNA was properly restricted to aboral ectoderm cells, it was often present in only a subset of these cells. Some examples of complete and incomplete aboral ectoderm-specific expression are shown in Fig. 3C. The incomplete expression occurs in a variety of patterns; in some cases the mRNA was present in single large or small patches of contiguous cells in a part of the aboral ectoderm (e.g. in only one wall or in one wall and at the vertex of the embryos); in other cases the LpSl mRNA was present in more than one separate patch of aboral ectoderm cells. To ask whether the S. purpuratus homologue, Sped, is also expressed in the same hybrid embryos that are expressing the LpSl mRNA we analyzed expression of the two mRNAs in adjacent sections of the same embryos by in situ hybridization. Pairs of slides containing alternating serial sections of the same embryos were assembled and hybridized with either the LpSl riboprobe or with the Spc1 riboprobe. Of 43 different LpSp plutei that contained the LpSl mRNA, 21 were also expressing the Specl mRNA, and 22 were not. Examples of three different embryos expressing only the LpSl mRNA are shown in Fig. 4. In each of these cases we conclude that the LpSl mRNA was present in all cells of the aboral ectoderm as is found in normal L. pictus embryos. This is based on our interpretation of the sections shown here and on additional serial sections of the same embryos that are not shown.

In the embryo shown in Fig. 4C the LpSl mRNA was present on the anal wall of the ectoderm only up to approximately the position where the anus intersects the ectoderm; the cells in the remaining portion of this wall (i.e., the supranal ectoderm) did not contain either the LpSl or the Spc1 mRNA. Although gene expression in the supranal ectoderm has not previously been distinguished from the remainder of the aboral ectoderm in studies of S. purpuratus embryos, we have found that the LpSl mRNA is not present in this region in similar sections of normal L. pictus plutei (data not shown). In these cases the signal reappears in the adjacent aboral ectoderm as the sections pass out of the plane of the anus, indicating that the region of supranal ectoderm that is defined by lack of LpSl expression is fairly narrow.

Four examples of hybrid plutei that are expressing both the Spc1 and LpSl mRNA are shown in Fig. 5. The most striking characteristic of the expression of these two mRNAs is that they are expressed in non-overlapping regions of the ectoderm. In each case shown the patterns of complementary expression are different; this reflects, to some extent, the diversity of expression patterns that were seen. In the examples where the morphology of the embryo and the angle of the section made it possible to definitively distinguish the oral and aboral ectoderm, the expression of these two genes was detected only in cells of the aboral ectoderm. In cases where it was difficult to distinguish oral from aboral ectoderm by morphology there was often a region of the ectoderm that was devoid of either transcript that was probably oral ectoderm (e.g., Fig. 5C); in some cases both probes together labelled the entire ectoderm, but we concluded that the oral ectoderm was not in the plane of the section (e.g., Fig.5B). This interpretation was based on the analysis of additional sections of the embryo (a total of eight sections for the embryo in Fig. 5B). Sometimes the set of sections was strongly suggestive but not absolutely definitive, so it is possible that there is occasional expression of one of these genes in oral ectoderm. Examination of many different LpSp plutei allow us to draw the following conclusions: (1) the predominant expression of the Spc1 gene occurs because the LpSl gene is detectably expressed only in a small fraction of the hybrid plutei; (2) all cells of the aboral ectoderm appear to express either Spc1 or LpSl genes in almost all hybrid plutei; (3) expression of both homologues appears to be restricted to aboral ectoderm; i.e., there was no definitive expression in oral ectoderm, and none detected in endoderm; (4) in most embryos that express both homologues, the two mRNAs are present in complementary sets of cells that constitute the entire aboral ectoderm.

Expression of Spc1 and LpSl mRNAs in SpLp hybrid embryos

The detection of the Spc1 and LpSl RNAs in embryos by RNase protection shown in Fig. 1 indicates that the timing and level of expression of each homologue is similar in LpSp and SpLp hybrids. Furthermore, detection of each mRNA by in situ hybridization has shown that the predominant expression of the Spc1 gene in LpSp embryos occurs because only about 2% of the embryos express the LpSl mRNA. To ask whether this is true in SpLp embryos and whether the spatial restriction of these mRNAs is also the same in these hybrids, in situ hybridization was performed on alternate serial sections of SpLp hybrid plutei. Approximately 98% of these embryos were found to be expressing the Spc1 mRNA in the aboral ectoderm, as seen for LpSp plutei. Also, as seen for LpSp hybrids, only about 2% (10/490) were expressing the LpSl mRNA at easily detectable levels and serial sections of these embryos were negative for Spc1 mRNA. The LpSl RNA was properly restricted to the aboral ectoderm of these embryos; three examples are shown in Fig. 6. In contrast to the LpSp plutei, no examples of SpLp plutei were found where the LpSl mRNA was expressed in only a portion of the aboral ectoderm. However, in those SpLp hybrid plutei where LpSl mRNA was easily detected no Spc1 mRNA was found, indicating that the mutually exclusive expression of the two genes is the rule in both sets of hybrid plutei. It should be mentioned that a higher percentage of SpLp than LpSp plutei also showed a low and non-uniform distribution of silver grains that was slightly higher in the aboral ectoderm, with the LpSl riboprobe. This may represent authentic LpSl expression at much lower levels in these embryos, but because the signal was not clearly above background we have not considered it important for our analysis of preferential expression of Spc1 genes in hybrids.

In hybrid embryo populations the Spc1 mRNA accumulates normally but the LpSl mRNA drops sharply between the gastrula and pluteus stages (Fig. 1 and Table 1). To understand better the basis for the difference between these two stages in SpLp hybrids, in situ hybridization was used to detect each mRNA in sections of gastrula stage embryos. The results for gastrulae are similar to those in plutei in two ways. First, almost all hybrid gastrulae were expressing the Spc1 mRNA at normal levels and it was localized to the cells of the presumptive aboral ectoderm. Second, a small fraction of the hybrid gastrulae (11/636=1.7%) were expressing the LpSl mRNA in the presumptive aboral ectoderm and were not expressing the Spc1 mRNA. However, in contrast to plutei, about 1/2 (124 out of 252) of the gastrulae that express Spc1 mRNA were also expressing the LpSl mRNA at a similar level and in overlapping regions of the presumptive aboral ectoderm. Fig. 7 shows one example each of LpSl positive/Specl negative and LpSl positive/Specl positive embryos. It appears, then, that the higher level of expression of LpSl genes in SpLp gastrulae compared to plutei results from coexpression of Spc1 and LpS1 genes in aboral ectoderm cells of many gastrulae which is eliminated as they develop to the pluteus stage. The small fraction of embryos that are expressing only the LpSl RNA is essentially the same at both stages (1.7% in gastrula and 2.0% in pluteus), suggesting that exclusive expression of the LpSl gene was established in these embryos by the gastrula stage and then maintained in plutei.

Previous studies have revealed that the mRNAs from several 5. purpuratus actin genes are expressed at the correct time (Cyl, Cyllla, Cylllb and M) and in the correct cell types (Cyl, Cyllla and M) in hybrid embryos of S. purpuratus and L. pictus, and that this accurate regulation is independent of the gamete of origin of the genes (Bullock et al. 1988; Nisson et al. 1989a). Futhermore, experiments with hybrid embryos of S. purpuratus and Lytechinus variegatus also have shown that the S. purpuratus Cyllla actin gene is expressed only in the correct aboral ectoderm cell lineage (B.R. Hough-Evans, R.J. Britten and E.H. Davidson, personal communication). These observations demonstrated that all of the factors necessary for the proper regulation of these genes are present in these hybrid embryos at the right time and place.

However, when a chimeric gene containing the Cyllla actin gene promoter attached to a CAT reporter gene was injected into L. variegatus embryos its mRNA was found in several other cell types, indicating that its spatial expression is aberrant, though temporally correct (Hough-Evans et al. 1987; Franks et al. 1988). This suggests that zygotically derived factors are required for the restriction of Cyllla actin gene expression to the cells of the aboral ectoderm, and that these factors are synthesized in the hybrid embryos. The use of hybrid embryos in the study of expression of the S. purpuratus actin genes has thus been useful in defining the zygotic origin of at least some mzns-acting regulatory factors. The studies on actin genes were limited, however, to analysis of only one of the two homologues of each gene present in the hybrids because gene- and species-specific probes for the L. pictus actin genes were not available. To ask further whether the homologues of the same gene from each species are regulated correctly in hybrids of S. purpuratus and L. pictus we have studied the expression of the 5. purpuratus Spec! gene and its L. pictus homologue, LpSl. In our initial analysis we found that the levels of the protein and mRNA encoded by the Specl gene are much higher than those encoded by the LpSl gene in hybrid plutei, and that this reflects differences in their transcriptional activity (Brandhorst et al. 1991). In the experiments reported here we have found also that the predominant expression of the Spc1 gene in populations of hybrid plutei is due to the restriction of detectable expression of the LpSl gene to a small fraction of plutei, and the exclusive expression of Spc1 genes in most plutei (except in cells expressing the LpSl gene).

As a result of the experiments presented here it is evident that the phenomenon of preferred transcription of the Spc1 gene over the LpSl gene in hybrid embryos is complicated. Comparison of the kinetics of accumulation of the two transcripts indicates that both transcripts begin to rise almost normally during early development, but that the amount of LpSl RNA is restricted to less than 5% of its normal level by the pluteus stage. The predominant expression of the Spc1 gene is therefore most pronounced at the later stages of embryonic development. At the pluteus stage only about 2% of the LpSp or the SpLp hybrid embryos contained detectable amounts of the LpSl mRNA while the remaining approximately 98% were expressing only the Spc1 mRNA at nearly normal levels. Among those embryos that were expressing LpSl mRNA the situation is slightly different in the two crosses. In the SpLp plutei, no examples were found where both mRNAs were expressed in the same embryo. On the other hand, in LpSp plutei about half of those expressing LpSl were also expressing Spc1 and about half were not. Furthermore, in those LpSp plutei that contained both mRNAs they were present in non-overlapping, complementary portions of the aboral ectoderm. In SpLp gastrulae, when the level of LpSl RNA was still fairly high, about 2% of the embryos are LpSl RNA positive and Spc1 RNA negative, but about 50% of the embryos contain both transcripts in overlapping regions of the presumptive aboral ectoderm. Since a similar fraction of hybrid gastrulae and plutei express only the LpSl gene, it is likely that establishment of this pattern of expression sometimes occurs before the gastrula stage and remains fixed. On the other hand, since some hybrid gastrulae show coexpression of both Spc1 and LpSl genes, establishment of exclusive Spc1 gene expression must occur over a range of developmental stages and involve the inactivation of transcription of LpSl genes in cells previously actively transcribing them. This inactivation probably does not depend on cell division since there is little or no proliferation of aboral ectoderm cells during this period (Cameron et al. 1990).

There is no significant loss or detected rearrangement of either LpSl or Spc1 DNA in these hybrid embryos (Brandhorst et al. 1991). Thus, the exclusive expression of Spc1 genes in most aboral ectoderm cells must occur in the presence of both genes. However, the expression of the LpSl gene in about 2% of hybrid plutei might be the result of random loss (or epigenetic inactivation) of the Spc1 gene in founders of clones of cells expressing the LpSl gene in these hybrids; such loss from a small fraction of aboral ectoderm cells would not have been detected by Brandhorst et al. (1991).

It is evident that, when both genes are present, the transcription of the Spc1 gene is usually favored, and that this effect becomes more pronounced as development proceeds. Perhaps the Spc1 gene (and, possibly, the related, coincidentally expressed Spec2 genes [Hardin et al. 1988; Gan et al. 1990a,b]) effectively removes limiting factors required for LpSl transcription, unless the Spc1 gene is missing or somehow randomly inactivated; alternatively, many S. purpuratus genes might compete more effectively than their L. pictus counterparts for a limiting general transcription factor, possibly accounting for the general predominance of expression of S. purpuratus genes in these hybrid embryos. Such an explanation is reminiscent of the phenomenon of nucleolar dominance observed in hybrid embryos and cells of many plants and animals whereby ribosomal RNA genes of one species are predominantly expressed regardless of the direction of the cross or the origin of the cytoplasm (Reeder, 1985). In some instances, nucleolar dominance appears to result from an imbalance in enhancer activity between ribosomal RNA genes of the two species, while in other instances there appears to be species-dominant promoter selectivity of positive transcription factors (Reeder, 1985).

The extensive coexpression of LpSl and Spc1 genes in cells of SpLp gastrulae, but not plutei, indicates that LpSl gene expression can actively occur in the presence of Spc1 DNA, while the eventual inactivation of LpSl gene activity might result from a reduction in essential transcription factors as development proceeds; the transcriptional activity of both Spc1 and LpSl genes declines after gastrulation in normal embryos (Tomlinson and Klein, 1990), consistent with this possibility. Since both genes show the correct temporal and spatial patterns of expression when active, the factors necessary for their qualitatively accurate regulation are present in hybrid embryos, at least at some level.

The aboral ectoderm of S. purpuratus plutei is derived from 11 founder cells established during early cleavage (Cameron et al. 1990). While in some instances the boundaries of patches of cells exclusively expressing Spc1 or LpSl genes appear to correspond to the boundaries of cell lineages derived from these founder cells, there is not a strict correspondence. In most cases, expression of the LpSl gene occurs in patches of contiguous cells which must be derived from several founder cells, suggesting a commitment to exclusive LpSl expression very early in cleavage. However, in some instances (such as shown in Fig. 3C, lower right), the patches of cells expressing LpSl mRNA are smaller than expected for a single founder cell lineage, indicating a relatively late initiation (or reversal) of this clonal pattern of expression.

We have argued that the Spc1 and LpSl genes are homologues, and thus might be considered alleles in hybrid embryos (Brandhorst et al. 1991). However, there are two very similar LpSl genes (Xiang et al. 1991) which are not distinguished by our hybridization probes, and the Spc1 gene is closely related to the Spec2 genes which show coincident patterns of expression (Hardin et al. 1988). There are considerable differences in the 5’-flanking sequences of the Spc1 and LpSl genes, including sites implicated by deletion analysis in transcriptional regulation (Gan et al. 1990a,b; Xiang et al. 1991). Moreover, in L. pictus embryos expression of the LpSl gene is selectively inhibited by BAPN treatment, while in 5. purpuratus embryos the Spc1 gene does not respond to this treatment (Wessel et al. 1989), indicating differences in regulation. Thus, it is not clear which members of the Spc1 and LpSl gene families are homologues and whether they function as alleles. Nevertheless, the absolute complementarity of expression of Spc1 and LpSl genes in the aboral ectoderm of hybrid plutei suggests a close regulatory relationship. It is possible that only one of the two alleles of these genes is normally expressed in a particular cell in normal embryos. There is considerable evidence for wide-spread allelic exclusivity of gene expression in somatic cells of a variety of plant and animal species (for review, Monk, 1990; Holliday, 1990). It is likely that an understanding of the molecular basis of the predominant but variegated expression of Spc1 genes in hybrid embryos will result in a better understanding of gene regulation in embryos.

We thank Laura Dike and Becky Dolan for supplying some of the fixed and embedded embryos that were used in these experiments, and R.C. and L.M. Angerer and W.H. Klein for providing recombinant Specl and LpSl plasmids. P.E.N. and M.F.G. were trainees on an institutional NRSA award, HD 07312. This research was supported by NSF Grant DCB8709865 to W.R.C., and by NIH Grant HD18332 and NSERC and BCHCRF grants to B.P.B.

Angerer
,
L. M.
and
Angerer
,
R. C.
(
1981
).
Detection of polyA+ RNA in sea urchin eggs and embryos by quantitative in situ hybridization
.
Nucl. Acids Res
.
9
,
2819
2840
.
Brandhorst
,
B. P.
,
Fillon
,
M.
,
Nisson
,
P. E.
and
Crain
,
W. R.
(
1991
).
Restricted expression of the Lytechinus pictus Specl gene homologue in reciprocal hybrid embryos with Strongylocentrotus purpuratus
.
Dev. Biol
.
144
,
405
411
.
Bruskin
,
A. M.
,
Bédard
,
P. A.
,
Tyner
,
A. L.
,
Showman
,
R. M.
,
Brandhorst
,
B. P.
and
Klein
,
W. H.
(
1982
).
A family of proteins accumulating in ectoderm of sea urchin embryos specified by two related cDNA clones
.
Dev. Biol
.
91
,
317
324
.
Bollock
,
B. P.
,
Nisson
,
P. E.
and
Crain
,
W. R.
(
1988
).
The timing of expression of four actin genes and an RNA polymerase Ill-transcribed repeated sequence is correct in hybrid embryos of the sea urchin species Strongylocentrotus purpuratus and Lytechinus pictus
.
Dev. Biol
.
130
,
335
347
.
Cameron
,
R. A.
,
Fraser
,
S. E.
,
Britten
,
R. J.
and
Davidson
,
E. H.
(
1990
).
Segregation of oral from aboral ectoderm precursors is completed at fifth cleavage in embryogenesis of Strongylocentrotus purpuratus
.
Dev. Biol
.
137
,
77
85
.
Carpenter
,
C. D.
,
Bruskin
,
A. M.
,
Hardin
,
P. E.
,
Keast
,
M. J.
,
Anstrom
,
J.
,
Tyner
,
A. L.
,
Brandhorst
,
B. P.
and
Klein
,
W. H.
(
1984
).
Novel proteins belonging to the troponin-C superfamily are encoded by a set of mRNAs in sea urchin embryos
.
Cell
36
,
663
671
.
Conlon
,
R. A.
,
Tufaro
,
F.
and
Brandhorst
,
B. P.
(
1987
).
Post-transcriptional restriction of gene expression in sea urchin interspecies hybrid embryos
.
Genes and Dev
.
1
,
337
346
.
Cox
,
K. H.
,
Deleon
,
D. V.
,
Angerer
,
L. M.
and
Angerer
,
R. C.
(
1983
).
Detection of mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA probes
.
Dev. Biol
.
100
,
197
206
.
Crain
,
W. R.
and
Bushman
,
F. D.
(
1983
).
Transcripts of paternal and maternal actin gene alleles are present in interspecific sea urchin embryo hybrids
.
Dev. Biol
.
100
,
190
196
.
Davidson
,
E. H.
(
1976
).
Gene Activity in Early Development
. Second edition.
Academic Press
,
New York
.
Fedecka-Bruner
,
B.
,
Anderson
,
M.
and
Epel
,
D.
(
1971
).
Control of enzyme synthesis in early sea urchin development: aryl sulfatase activity in normal and hybrid embryos
.
Dev. Biol
.
25
,
655
671
.
Franks
,
R. R.
,
Hough-Evans
,
B. R.
,
Britten
,
R. J.
and
Davidson
,
E. H.
(
1988
).
Spatially deranged though temporally correct expression of a Strongylocentrotus purpuratus actin gene fusion in transgenic embryos of a different sea urchin family
.
Genes and Dev
.
2
,
1
12
.
Gan
,
L.
,
Wessel
,
G. M.
and
Klein
,
W. H.
(
1990a
).
Regulatory elements from related Spec genes of Strongylocentrotus purpuratus yield different spatial patterns with lacZ reporter gene
.
Dev. Biol
.
142
,
346
359
.
Gan
,
L.
,
Zhang
,
W.
and
Klein
,
W. H.
(
1990b
).
Repetitive DNA sequences finked to sea urchin spec genes contain transcriptional enhancer-like elements
.
Dev. Biol
.
139
,
186
196
.
Hardin
,
P. E.
,
Angerer
,
L. M.
,
Hardin
,
S. H.
,
Angerer
,
R. C.
and
Klein
,
W. H.
(
1988
).
The Spec2 genes of Strongylocentrotus purpuratus’. structure and differential expression in embryonic aboral ectoderm cells
.
J. Mol. Biol
.
202
,
417
431
.
Harding
,
C. V.
,
Harding
,
D.
and
Perlman
,
P.
(
1954
).
Antigens in sea urchin hybrid embryos
.
Exp. Cell Res
.
6
,
202
210
.
Holliday
,
R.
(
1990
).
Genomic imprinting and allelic exclusion
.
Development Supplement
,
125
129
.
Hough-Evans
,
B. R.
,
Franks
,
R. R.
,
Cameron
,
R. A.
,
Britten
,
R. J.
and
Davidson
,
E. H.
(
1987
).
Correct cell-type-specific expression of a fusion gene injected into sea urchin eggs
.
Dev. Biol
.
121
,
576
579
.
Maxson
,
R. E.
and
Egrie
,
J. C.
(
1980
).
Expression of maternal and paternal histone genes during early cleavage stages of the echinoderm hybrid Strongylocentrotus purpuratus X Lytechinus pictus
.
Dev. Biol
.
74
,
335
342
.
Mcclay
,
D. R.
and
Hausman
,
R. E.
(
1975
).
Specificity of cell adhesion: differences between normal and hybrid sea urchin cells
.
Dev. Biol
.
47
,
454
460
.
Mcgowan
,
R.
,
Campbell
,
R.
,
Peterson
,
A.
and
Sapienza
,
C.
(
1989
).
Cellular mosaicism in the methylation and expression of hemizygous loci in the mouse
.
Genes and Dev
,
3
,
1669
1676
.
Monk
,
M.
(
1990
).
Variation in epigenetic inheritance
.
Trends Genet
.
6
,
110
114
.
Morgan
,
T. H.
(
1927
).
The development of species hybrids
. In
Experimental Embryology
, pp.
608
-
657
. Columbia University Press: New York.
Nisson
,
P. E.
,
Dike
,
L.
and
Crain
,
W. R.
(
1989a
).
Three Strongylocentrotus purpuratus actin genes show cell-specific expression in hybrid embryos of S. purpuratus and Lytechinus pictus
.
Development
105
,
407
413
.
Nisson
,
P. E.
,
Francis
,
S.
and
Crain
,
W. R.
(
1989b
).
Spatial patterns of gene expression in preimplantation mouse embryos
.
Mol. Reprod. and Dev
.
1
,
254
263
.
Reeder
,
R. H.
(
1985
).
Mechanisms of nucleolar dominance in animals and plants
.
J. Cell Biol
.
101
,
2013
2016
.
Tomlinson
,
C.
and
Klein
,
W. H.
(
1990
).
Temporal and spatial transcriptional regulation of the aboral ectoderm-specific spec genes during sea urchin embryogenesis
.
Mol. Reprod. Dev
.
25
,
328
338
.
Tufaro
,
F.
and
Brandhorst
,
B. P.
(
1982
).
Restricted expression of paternal genes in sea urchin interspecies hybrids
.
Dev. Biol
.
92
,
209
220
.
Wessel
,
G. M.
,
Zhang
,
W.
,
Tomlinson
,
C. R.
,
Lennarz
,
W. J.
and
Klein
,
W. H.
(
1989
).
Transcription of the Specl-like gene of Lytechinus is selectively inhibited in response to disruption of the extracellular matrix
.
Development
106
,
355
365
.
Xiang
,
M.
,
Bedard
,
P. A.
,
Wessel
,
G.
,
Filien
,
M.
,
Brandhorst
,
B. P.
and
Klein
,
W. H.
(
1988
).
Tandem duplication and divergence of a sea urchin protein belonging to the troponin C super family
.
J. Biol. Chem
.
263
,
17173
17180
.
Xiang
,
M.
,
Ge
,
T.
,
Tomlinson
,
C. R.
and
Klein
,
W. H.
(
1991
).
Structure and promoter activity of the LpSl genes of Lytechinus pictus: Duplicated exons account for LpSl proteins with eight calcium-binding domains
.
J. Biol. Chem
.
226
,
10524
10533
.