SEARCH

SEARCH BY CITATION

Keywords:

  • evolution;
  • isoform;
  • Porifera;
  • primmorph;
  • T-box gene

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Background information. T-box transcription factors are a large family of transcriptional regulators involved in many aspects of embryonic development. In a previous report, we described the isolation and genomic characterization of two T-box genes from the siliceous sponge Suberites domuncula: a Brachyury homologue, Sd-Bra, and a Tbx2 homologue, Sd-Tbx2. Elucidation of the genomic structure of Sd-Bra allowed us to demonstrate the existence of two different isoforms, resulting from alternative splicing. Moreover, we demonstrated that the shorter isoform exists in two different glycosylation states.

Results. In the present study, we demonstrate a differential subcellular localization of the three Sd-Bra isoforms, suggesting that its differential nuclear import could be an important mechanism for its functional regulation. Furthermore, we demonstrate that Sd-Tbx2 exists only in one isoform, which is mainly localized in the nucleus. The pattern of expression of Sd-Bra and Sd-Tbx2 genes is analysed in sponge tissue, in gemmules and in cultured cells.

Conclusion. These results suggest a conserved role for Sd-Bra in the control of morphogenetic movements through the regulation of cell-adhesion properties and the involvement of Sd-Tbx2 in the determination of cell identity in the early stages of differentiation, reminiscent of the function of Tbx2-3-4-5 in vertebrates during limb specification. Also, the fact that a Brachyury and a Tbx2 homologue exist in S. domuncula suggests that the first divergence from the ancestral Brachyury-like gene might be a Tbx2-like gene and not a Tbrain-like gene as had been previously suggested [Adell, Grebenjuk, Wiens and Müller (2003) Dev. Genes Evol. 213, 421–434].


Abbreviations used:
DAPI

4,6-diamidino-2-phenylindole

NLS

nuclear localization signal

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

T-box transcription factors are a large family of transcriptional regulators involved in developmental processes such as specification of the mesoderm, anterior—posterior axis patterning or specification and outgrowth of muscle and limbs (reviewed in Smith, 1997; Papaioannou, 2001). More than 50 members of this family have been identified from metazoans (Herrmann and Kispert, 1994; Papaioannou, 1997; Papaioannou and Silver, 1998; Smith, 1999; Tada and Smith, 2001). Interestingly, in contrast with other transcription factors, e.g. members of the homeobox gene family, T-box genes have not been found in plants or fungi. The characteristic DNA-binding (T-box) domain is composed of a highly conserved stretch of 180–190 amino acids (Kispert and Herrmann, 1993), which have been taken as the basis to group T-box genes into five different subfamilies: T or Brachyury, Tbrain, Tbx2, Tbx1 and Tbx6. In general, the different subfamilies are also distinguished by their pattern of expression and their function (Papaioannou and Silver, 1998). In a previous report, we described the isolation and genomic characterization of two T-box genes from the siliceous sponge Suberites domuncula (Adell et al., 2003a). The phylogenetic analysis classified one into the subfamily of Brachyury, Sd-Bra, and the other into the Tbx2 subfamily, Sd-Tbx2.

Brachyury, the founding member of the family, is characteristically expressed in the notochord and tail bud of chordates (Herrmann et al., 1990). Its function in mesoderm and notochord morphogenesis in chordates was elucidated from mouse and Xenopus mutant embryos lacking the notochord and tail (Wilkinson et al., 1990; Conlon et al., 1996). The finding of Brachyury homologues in all metazoan phyla studied and its phylogenetic position within the T-box genes suggest that Brachyury genes are the closest relatives of the common ancestral T-box (Wattler et al., 1998; Adell et al., 2003a). Despite its different pattern of expression in the embryos of different animals (Wilkinson et al., 1990; Smith et al., 1991; Kusch and Reuter, 1999; Arendt et al., 2001; Gross and McClay, 2001; Spring et al., 2002), it is accepted that the dominant expression of Brachyury in all metazoans occurs in the blastopore, which develops further into the fore- and hindgut (Arendt et al., 2001; Technau, 2001).

The vertebrate Tbx2 subfamily comprises four closely related paralogues (Tbx2-3-4-5), which evolved from a single ancestral locus by a tandem-duplication event followed by a duplication of the derived gene pair (Agulnik et al., 1996). They are involved in the patterning of several structures as notochord, somites, eyes, heart and limbs (Hayata et al., 1999; Yamada et al., 2000; Takabatake et al., 2002). Their roles in the developmental specification and growth of limbs have been extensively studied (Gibson-Brown et al., 1998; Rodriguez-Esteban et al., 1999; Simon, 1999; Takeuchi et al., 1999; Ahn et al., 2002). Besides the chordate genes and the two Tbx2 members from echinoderms (Croce et al., 2003; Gross et al., 2003), only two Tbx2 genes from protostomians and one from Placozoa have been published (Martinelli and Spring, 2003).

The elucidation of the genomic structure of Sd-Bra has allowed us to demonstrate the existence of two different isoforms corresponding to alternatively spliced isoforms (Adell et al., 2003a). Furthermore, analysis with a polyclonal antibody raised against the Sd-Bra protein showed that the shorter isoform exists in two different glycosylation states. In the present study, we demonstrate that differential splicing is important for the functional regulation of Sd-Bra, since it affects its transport to the nucleus. Moreover, using an antibody raised against the Sd-Tbx2 protein, we demonstrate that only one isoform, Sd-Tbx2, exists and that it is localized in the nucleus of a few cells scattered throughout the sponge tissue.

Sponges are diploblastic animals, which constitute the most basal animal phylum. They are composed of an epithelial-like layer, the pinacoderm, surrounding a mesohyl, which contains the specialized cells (spherulous cells, sclerocytes and choanocytes) and the archaeocytes, totipotent cells capable of differentiating into all the other cell types (Simpson, 1984). Sponges can reproduce sexually and asexually, by releasing embryos or by budding respectively.

The expression and functional analysis of the developmental genes, such as the T-box, should be performed on the corresponding embryos. However, because it is not yet possible to obtain sponge embryos, we have analysed the expression of Sd-Bra and Sd-Tbx2 in sponge gemmules and under different sponge cell-culture conditions. Gemmules are a peculiar type of buds formed during asexual reproduction in many sponge species, such as in S. domuncula. They are composed of undifferentiated and quiescent cells, the amoebocytes, covered by a hard coat of collagen reinforced by spicules. When conditions are favourable for germination, the amoebocytes escape and differentiate into the various cell types needed to build a complete young sponge.

Regarding the cell cultures, two types were compared: primmorphs and adherent aggregates. Primmorphs are 1–5 mm round-shaped aggregates, formed in suspension from dissociated single sponge cells after approx. 5 days of incubation on a moving platform. They have an organized tissue-like structure, with a central zone composed primarily of spherulous cells, surrounded by a pinacoderm-like epithelium (Müller et al., 1999; Le Pennec et al., 2003). In contrast, the dissociated sponge cell cultures on coated plastic dishes without movement first produce small aggregates in suspension, which, after a few hours, adhere to the substrate and grow as adhered aggregates during subsequent days. Although the cell—cell contacts in primmorphs are established mostly during the first hours of culture, reaching a balance around day 5, the interactions among cells and between cells and the matrix continuously remodel the growth and shape of the adherent aggregates.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Expression patterns of Sd-Bra and Sd-Tbx2 in sponge tissues

In a previous report, we described the isolation and genomic characterization of two T-box genes, Sd-Bra and Sd-Tbx2, from the siliceous sponge S. domuncula (Adell et al., 2003a). Two different isoforms of Sd-Bra (Sd-Bra1 and Sd-Bra2), corresponding to products of alternative splicing, were demonstrated. In the present study, Northern- and Western-blot analyses of Sd-Bra and Sd-Tbx2 expression were performed. Northern-blot analysis of Sd-Bra in the sponge tissue shows a double band around 1300 bp that would correspond to the two alternative splicing isoforms (Figure 1A). The two different isoforms are detected by Western blotting using a polyclonal antibody raised against Sd-Bra; however, a third band is also present that has been previously demonstrated to be a glycosylated form of Sd-Bra1 (Figure 1B) (Adell et al., 2003a).

thumbnail image

Figure 1. Analysis of Sd-Bra and Sd-Tbx2 expression in adult sponge

(A) Northern-blot analysis of total RNA extracts from the sponge S. domuncula with the corresponding riboprobes of Sd-Bra and Sd-Tbx2 cDNAs. The positions of the 28 and 18 S ribosomal RNAs are indicated. (B) Western-blot analysis of protein extracts from the sponge tissue, using polyclonal antibodies raised against Sd-Bra and Sd-Tbx2. The double band corresponding to the different glycosylated states of Sd-Bra1 and Sd-Bra2 are indicated. The positions of the molecular-mass standards (prestained) are indicated on the left. (C) Subcellular localization analysis of Sd-Bra and Sd-Tbx2 proteins in the sponge tissue by immunohistochemistry. The nucleus stained with DAPI, corresponding to cells positively stained with the specific antibody, are indicated with an arrow. (ad) Anti-Sd-Bra antibody. (a) The two cells on the left show both nuclear and cytoplasmic localization; the cell on the right shows predominantly nuclear staining. (c) A cell with exclusively cytoplasmic staining is shown. (eh) Anti-Sd-Tbx2 antibody. (e) Nuclear staining of an epithelial cell, which conforms the surface of the canal (ca). (f) Nuclear staining of several cells from the sponge mesohyl (me).

Download figure to PowerPoint

Northern-blot analysis of Sd-Tbx2 RNA expression shows a single band of the expected size (2373 bp) (Figure 1A). Western-blot analysis of protein extracts from sponge tissue with a polyclonal antibody raised against the Sd-Tbx2 protein shows a single protein product of approx. 45 kDa (Figure 1B).

The predicted molecular masses of Sd-Bra1, Sd-Bra2 and Sd-Tbx2 proteins are 35.2, 37.2 and 55.7 kDa respectively. The lower apparent molecular mass of the three proteins in the Western-blot analysis might be due to the use of a prestained molecular-mass standard.

The subcellular localization of Sd-Bra and Sd-Tbx2 in the sponge tissue was analysed by immunohistochemistry using the respective polyclonal antibodies. As expected, because of its function as a transcription factor, Sd-Tbx2 was localized in the nucleus (Figure 1C). Its expression was restricted to a few cells dispersed over the sponge tissue. Some of them could be identified as epithelial-like cells, as they were forming the epithelial-like layer of the sponge canals, but were mostly located within the mesohyl (Figure 1C).

In contrast, the subcellular localization of Sd-Bra protein was not restricted to the nucleus. Most of the Sd-Bra antibody signals showed a granular pattern in the cytoplasm of some cells dispersed in the sponge tissue. Only in a few cells was the signal seen both in the cytoplasm and the nucleus (Figure 1C).

Patterns of expression of Sd-Bra and Sd-Tbx2 in sponge cell cultures and gemmules

Dissociated sponge cells were maintained in culture for 10 days either as adhered aggregates or as primmorphs (see the Materials and methods section; Figure 2A). Adhered cells after 24 h of incubation and 9-day-old primmorphs were analysed by immunohistochemistry with the anti-Sd-Bra and anti-Sd-Tbx2 antibodies. As seen in Figure 2(B), Sd-Bra is located in the cytoplasm of cultured sponge cells. Small amounts of Sd-Bra were found in almost all the cells that adhered to the plastic after 24 h of culture; later, the expression is restricted to a few cells of the already formed primmorphs. The Sd-Tbx2 pattern of expression in adherent cells and primmorphs is similar to that seen in the sponge tissue, being localized in the nucleus of a few dispersed cells (Figure 2B). Likewise, the expression of Sd-Bra and Sd-Tbx2 was analysed in gemmules, where neither of the sponge T-box proteins were found to be expressed (Figure 2B).

thumbnail image

Figure 2. Analysis of Sd-Bra and Sd-Tbx2 expression in cultured sponge cells and in gemmules

(A) (a) Specimens of the sponge S. domuncula, maintained in the aquarium for several months; (b) sponge gemmules; (c, d) sponge aggregates adhered to the substrate for 1 and 5 days respectively; (e) sponge primmorph (×5 magnification). (B) Analysis of subcellular localization by immunohistochemistry of Sd-Bra and Sd-Tbx2 proteins in cells adhering to the poly-lysine-coated coverslips for 12 h, in primmorphs (9-day-old) and in gemmules. The positive cells with the antibody are also indicated with an arrow in the DAPI staining, which is shown next to each section. The coat of the gemmule is marked ‘co’. The scale bar shown is common to all micrographs in (B).

Download figure to PowerPoint

Sd-Bra is differentially located in the cytoplasm and in the nucleus of the sponge cells

Brachyury proteins are the prototypic members of the T-box class of transcription factors, which exert their function in the nucleus. The cytoplasmic localization of Sd-Bra protein demonstrated by immunohistochemistry was not as expected, since no report of a possible cytoplasmic localization has been published. To investigate biochemically the subcellular distribution of Sd-Bra in sponge cells, protein extracts from the cytoplasm and nucleus of sponge cells cultured for 12 h on plastic dishes were analysed by Western blotting as described in the Materials and methods section.

The antibody raised against Sd-Tbx2 and an antibody raised against Sd-myotrophin were used as controls for correct separation of the nuclear and cytoplasmic fractions respectively. The nuclear localization of Sd-Tbx2 has been demonstrated above. Myotrophin is a cytosolic regulatory protein that modulates the levels of activated NFκB (nuclear factor κB) dimers (Knuefermann et al., 2002). Cloning and expression of Sd-myotrophin has been reported previously by our group (Schröder et al., 2000). As seen in Figure 3, when using the Sd-Bra polyclonal antibody, the two isoforms corresponding to the different glycosylation states of Sd-Bra1 are present both in the cytoplasmic and the nuclear fractions at similar levels. In contrast, Sd-Bra2 is mainly found in the cytoplasmic fraction. A small amount of Sd-Bra2 can also be seen in the nuclear fraction. This could be due to contamination, since Sd-Bra2 is also present in the second cytoplasmic fraction. However, the results with the controls demonstrate that the cellular fractionation is very clean. This suggests that a small fraction of Sd-Bra2 is transported to the nucleus.

thumbnail image

Figure 3. Western-blot analysis of two sequential cytoplasmic protein extracts (C1 and C2) and the nuclear extract (N) with the anti-Sd-Bra antibody

The anti-Sd-Tbx2 (reaction with the nuclear fraction) and anti-myotrophin antibodies (reaction with the cytoplasmic fraction) are used as controls for the correct separation of the compartments.

Download figure to PowerPoint

Regulation of the expression of Sd-Bra and Sd-Tbx2 in cultured sponge cells

To test whether the level of expression of Sd-Bra and Sd-Tbx2 genes is altered during different culture conditions of the sponge cells, protein extracts from sponge tissue, cells cultured for 1, 4 or 9 days in plastic dishes, and 4- and 9-day-old primmorphs were analysed by Western blotting with the corresponding antibodies. As seen in Figure 4, Sd-Bra expression is down-regulated in primmorphs compared with the sponge tissue. However, its expression is up-regulated when the cells adhere to the substrate, even from the first day of culture. Although we have not analysed quantitatively the intensity of the bands corresponding to the three isoforms of Sd-Bra, there are no apparent changes in their relative levels between the different samples. Sd-Tbx2 expression is also up-regulated compared with the levels in the sponge tissue during the first day of culture, and it decreases later both in adherent cells and in primmorphs (Figure 4).

thumbnail image

Figure 4. Western-blot analysis of total protein extracts from (T) sponge tissue, (A) adherent cells 1, 4 and 9 days after plating and (P) primmorphs 4 and 9 days after the formation of the three-dimensional aggregates

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Subcellular localization of Sd-Bra

The existence of three isoforms of Sd-Bra has been reported by us previously (Adell et al., 2003a). An alternative splicing mechanism produces two different Sd-Bra proteins that differ by 17 amino acids within the T-box domain. This produces a shorter splice isoform (Sd-Bra1) that is susceptible to glycosylation. In the present study, we demonstrate that the longer isoform (Sd-Bra2) is found predominantly in the cytoplasmic fraction, whereas the two differentially glycosylated forms of Sd-Bra1 can be found both in the cytoplasmic and in the nuclear extracts. Therefore the subcellular localization of Sd-Bra is differentially regulated depending on a given spliced isoform, but not on the glycosylation state.

Basically, Sd-Bra could pass the nuclear pore, being a molecule of less than 40 kDa; however, for most of the low-molecular-mass molecules, an active transport has also been demonstrated (Breeuwer and Goldfarb, 1990). A putative NLS (nuclear localization signal) has been identified in T-box sequences within the Tbrain and Tbx2 subfamilies (Collavoli et al., 2003; Fan et al., 2003). Although no canonical NLS has been identified in Brachyury proteins, previous studies suggest the presence of several complex NLS between and C-terminal to the T-box domain of Brachyury, which are required in combination for its exclusive localization in the nucleus (Kispert et al., 1995). Our results suggest that the addition of 17 amino acids in the T-box domain of Sd-Bra2 might prevent the transport of Sd-Bra into the nucleus. To assess the molecular mechanism behind the behaviour of Sd-Bra isoforms, we tried to direct the translocation of Sd-Bra2 into the nucleus in cultured cells by adding the medium of leptomycin B and also by adding the morphogen retinoic acid. No effect was observed under the conditions used (T. Adell, unpublished data).

Sd-Bra is the first Brachyury reported that displays different products of alternative splicing and post-transcriptional modifications and, moreover, shows a differential subcellular localization depending on the isoform. Within the Tbrain and Tbx2 subfamilies of T-box proteins, cytoplasmic localization (Hsueh et al., 2000, 2002; Fuchikami et al., 2002; Bruce et al., 2003; Collavoli et al., 2003) and the existence of different isoforms from alternative splicing have also been described (Chieffo et al., 1997; Campbell et al., 1998; Bamshad et al., 1999); however, until now, no functional analyses have been performed. Almost no studies on Brachyury subfamily proteins have been performed, only the RNA have been studied by in situ hybridization. On the basis of our results, we expect that Brachyury proteins from other organisms also undergo post-transcriptional and post-translational regulation.

Possible function of Sd-Bra

Phylogenetic analyses suggest that Brachyury could be the more ancestral group of T-box proteins (Wattler et al., 1998; Adell et al., 2003a). In contrast with other evolutionarily conserved genes, the Brachyury gene is expressed during development in different regions within the different phyla: it is expressed in the early mesoderm of vertebrate embryos (Wilkinson et al., 1990; Smith et al., 1991), only in the notochord of ascidians (Yasuo and Satoh, 1993; Corbo et al., 1997), the presumptive endodermal gut of echinoderms (Gross and McClay, 2001), the ectodermal hindgut and caudal visceral endoderm in Drosophila (Kispert and Herrmann, 1994; Kusch and Reuter, 1999) and the gastrulating posterior pole of cnidarians (Spring et al., 2002). It has been proposed that Brachyury genes have been co-opted for various functions during evolution, and that the Brachyury gene in all metazoan embryos is expressed in the blastopore, which develops further into the fore- and hindgut (Arendt et al., 2001; Technau, 2001). An even more common role for Brachyury, and other T-box proteins, has been proposed to be the control of morphogenetic movements, involving changes in cell adhesion and migration properties (Heisenberg et al., 2000; Tada and Smith, 2000; Gross and McClay, 2001).

The expression of Sd-Bra in sponge embryos should be studied in an attempt to compare its pattern of expression and function with that documented for other metazoans. These results would contribute invaluable information regarding the original function of T-box genes in diploblastic animals, about the process of embryogenesis of the sponges and, even more, about the origin of the three germ layers of bilaterians.

Our results from the adult sponge and from the cultured cells give us some clues. High levels of Sd-Bra are detected during the first 24 h of culture, when the cell—cell and cell—matrix interactions have been established. Also, in adherent aggregates, compared with the primmorphs, an up-regulation of Sd-Bra is detected. In the adherent cultures, high rates of tissue reorganization occur, during which new cell contacts are formed through the flattening and migration of cells. Therefore we propose a possible role for Sd-Bra in morphogenetic movements through the regulation of cell adhesion and migration properties.

It has been reported that all bilaterian homologues of Brachyury proteins, from Hydra to humans, induce mesoderm formation in animal caps assays (Marcellini et al., 2003), thus supporting a common ancestral role for Brachyury proteins. It is not generally accepted that Porifera have a true ectoderm and endoderm and, even less, a mesoderm layer; however, our phylogenetic and functional studies suggest that Sd-Bra would be able to induce such structures in this kind of assay.

Possible function of Sd-Tbx2

In vertebrates, members of all T-box protein subfamilies have been found. However, in echinoderms and hemichordates, only T-box proteins from the Brachyury and Tbrain subfamilies had been reported (Tagawa et al., 1998, 2001; Shoguchi et al., 2000; Croce et al., 2001), and in Cnidarians, only the Brachyury homologues had been identified (Technau and Bode, 1999). The finding of Brachyury and Tbx2 homologues in S. domuncula suggests that the first divergence from the ancestral Brachyury-like gene might be a Tbx2-like and not a Tbrain-like gene (Adell et al., 2003a). In agreement with this hypothesis, two Tbx2 genes from echinoderms (Croce et al., 2003; Gross et al., 2003), and also one Brachyury and one Tbx2/3 gene from Trichoplax adhaerens (Martinelli and Spring, 2003), the only species from the Placozoa phylum, have recently been reported. It is accepted that the four chordate Tbx2-3-4-5 genes originated from a single ancestral locus by an initial tandem-duplication event that produced the Tbx2/3 and Tbx4/5 precursors, followed by a duplication (Agulnik et al., 1996). Interestingly, the phylogenetic tree of the Tbx2 subfamily, including this new sequence, demonstrates that echinoderm, placozoan and protostomian Tbx2 proteins fall at the base of the Tbx2-3 subgroup, but Sd-Tbx2 falls at the base of the Tbx4-5 subgroup, which has only members from chordates (T. Adell, unpublished data). This differential divergence between Porifera and the other phyla for the first ancestral T-box could be a decisive fact in their subsequent body plan organization.

Vertebrate Tbx2 genes are involved in the patterning of several structures such as notochord, eyes, heart and, especially, the limbs. Tbx2 and 3 are expressed in similar spatial—temporal patterns in both limbs (Gibson-Brown et al., 1998), whereas Tbx4 and Tbx5 function in the specification of the vertebrate hind- and forelimb respectively (Rodriguez-Esteban et al., 1999; Takeuchi et al., 1999). Also, in D. melanogaster, the Tbx2 homologue, Omb, participates in the development of wing structures (Kopp and Duncan, 1997). In the sponge, Sd-Tbx2 is up-regulated during the first day of cell culture. However, in contrast with Sd-Bra, thereafter, its expression is not increased in adherent aggregates with respect to primmorphs. This result supports its specific involvement in the regulation of the establishment of the first cell—cell contacts. Sd-Tbx2 labelling has also been found in isolated cells within the mesohyl of the sponge, which could correspond to archaeocytes or to an intermediate state of differentiation from archaeocytes to choanocytes or sclerocytes. Together, these findings suggest a possible role for Sd-Tbx2 as a cell-fate specification factor. It is proposed that Sd-Tbx2 is involved in the determination of cell identity during the early stages of differentiation, reminiscent of the function of Tbx2-3-4-5 in vertebrates during limb specification.

More than 20 T-box genes exist in vertebrates and they function as an interacting network during development. Interactions among them or their interactions with their targets are still poorly understood. The different patterns and timing of expression of the sponge T-box genes during reorganization of the cultured aggregates, and their absence in the gemmules, which are composed of undifferentiated and quiescent cells, suggest that, although the eventual discovery of a third T-box gene cannot be excluded, Sd-Bra and Sd-Tbx2 could constitute the T-box network in Porifera. It has been reported that Wnt homologues are the target genes of vertebrate Brachyury (Tada and Smith, 2000; Takeuchi et al., 2003). The Wnt receptor, Frizzled, from S. domuncula has recently been characterized (Adell et al., 2003b). Interestingly, it is up-regulated also during the first days of cell culture, suggesting that, in sponges, this signalling pathway is conserved. Hence, the sponges, the phylogenetically most basal metazoans, are an extremely useful model to understand the evolution of the animal body plan and are also a simplified model to resolve molecular interactions.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Chemicals and enzymes

Unless otherwise indicated, the sources of the chemicals and enzymes used were the same as those used in a previous study (Adell et al., 2003a).

Animals and cell cultures of S. domuncula

Live specimens of S. domuncula (Porifera, Demospongiae, Hadromerida) were collected by SCUBA diving near Rovinj (Croatia) from depths between 15 and 35 m. The sponges were brought to Mainz (Germany) and kept in 103 litre tanks at 17°C before use. The procedure for the formation of primmorphs from single cells was applied as described previously (Müller et al., 1999). Briefly, single cells obtained by dissociation in Ca2+- and Mg2+-free artificial seawater were cultured in natural seawater (Sigma, Deisenhofen, Germany), supplemented with 1$ RPMI 1640 medium with silicate (60 μM) and Fe3+ (30 μM) (Krasko et al., 2002). In the present study, the dissociated cells were seeded in coated plastic dishes, to which they adhered after a few hours. In some dishes, the cell aggregates were left to grow on the plastic for more than 10 days (termed adherent aggregates). Other dishes were placed on a moving platform after 24 h of culture. Under this condition, aggregated cells detached, forming the three-dimensional cell aggregates (primmorphs) after four more days in culture. The adherent aggregates or the primmorphs were analysed after 4 or 9 days in culture. To study the adherent sponge cells after 24 h of culture by immunohistochemistry, they were seeded directly on poly-lysine-coated coverslips (BD Biosciences, Bedford, MA, U.S.A.).

Gemmules were collected from the surface of shells in which hermit crabs reside as we have described earlier (Le Pennec et al., 2003).

Northern blotting

A sample of a sponge specimen living in the aquarium was immediately frozen, and pulverized in liquid nitrogen; RNA was extracted using the TRIzol® Reagent (Gibco BRL, Grand Island, NY, U.S.A.). Total RNA (8 μg) was fractionated by electrophoresis in 1$ agarose and transferred to a Hybond-N+ nylon membrane (Amersham Biosciences, Freiburg, Germany). Specific RNA probes corresponding to the Sd-Bra and Sd-Tbx2 cDNA sequences were synthesized with the DIG RNA labelling kit (Roche, Mannheim, Germany) and used for hybridization of the nylon membranes using DIGhyb solution (Roche). The membranes were washed at 65°C with 30 mM sodium citrate/300 mM NaCl (pH 7.0; supplemented with 0.1$ SDS) under very stringent conditions.

Antibody production

The procedure to generate Sd-Bra-specific polyclonal antibodies has been described previously (Adell et al., 2003a). Both pre- and post-immune samples were collected. The IgG fraction of the antiserum was purified by affinity chromatography using a Protein A antibody purification kit (Sigma). To generate the Sd-Tbx2-specific polyclonal antibody, the cDNA corresponding to the 501 amino acids of the predicted protein open reading frame was cloned into the pQE vector and used to transform the BL21 bacterial strain. The subsequent procedure was the same as described by Adell et al. (2003a). Similarly, polyclonal antibodies were previously raised against recombinant myotrophin as described previously (Schröder et al., 2000).

Western blotting

Tissue from a sponge kept in the aquarium was quickly frozen in liquid nitrogen, pulverized with a mortar and lysed in Tris buffer (25 mM Tris, pH 7.5, 1 mM EGTA, 1 mM EDTA and 1$ SDS). Lysates were boiled for 15 min, cleared by centrifugation (10000 g, 10 min and 4°C), and the protein concentration was determined (Micro BCA Protein Assay; Pierce, Rockford, IL, U.S.A.); 4- and 9-day-old primmorphs and 1-, 4- and 9-day-old adherent sponge cells were directly lysed, boiled, cleared and quantified as described above. Protein samples (50 μg) were size-fractionated by SDS/PAGE (10$ polyacrylamide), transferred on to nitrocellulose filters and incubated with blocking buffer (Tris-buffered saline, containing 5$ skimmed milk and 0.1$ Tween 20). After incubation with a 1:5000 dilution of the polyclonal antibody raised against Sd-Bra or Sd-Tbx2 proteins in blotting buffer (Tris-buffered saline containing 1$ skimmed milk and 0.1$ Tween 20), filters were washed and incubated with alkaline phosphatase-labelled goat anti-rabbit Ig (Dako, Glostrup, Denmark), and reactions were developed with the enhanced chemiluminescence technique (CDP-Star; Roche).

Immunohistochemistry

Fresh sponge tissue was washed with CMFSW buffer (Ca2+- and Mg2+-free artificial seawater) and fixed for 10 min in 4$ (w/v) paraformaldehyde. Siliceous spicules were removed by treating the tissue for 2 h with a 10 mM NaHF and 4$ HF solution. After washing with PBS, tissue samples were frozen in liquid N2 and embedded in tissue-teck; 9-day-old primmorphs and gemmules were also directly frozen in liquid N2 and embedded; 5 μm thick sections of sponge tissue, primmorphs and gemmules were fixed and permeabilized by incubating with methanol for 1 h at −20°C. The adhered sponge cells that had been cultured for 24 h on poly-lysine-coated coverslips were directly fixed and permeabilized by incubating with methanol for 5 min at 4°C. After blocking with 1$ BSA in PBS for 30 min, sections were incubated with the anti-Sd-Bra or anti-Sd-Tbx2 antibody overnight at 4°C. The antibody was used in a 1:500 dilution in 0.5$ BSA in PBS. Rhodamine-conjugated goat anti-rabbit Ig (Dako) was used as a secondary antibody. To visualize the DNA in the nucleus, the sections were stained with DAPI (4,6-diamidino-2-phenylindole) after incubation with the secondary antibody. The preimmune rabbit serum was used as a control.

Preparation and analysis of nuclear and cytoplasmic extracts

Sponge single cells were obtained by dissociation in Ca2+- and Mg2+-free artificial seawater, and cultured in cell-culture plastic dishes in natural seawater (Sigma). After 12 h, cells were collected. The cytoplasmic and nuclear protein extracts were obtained by using the NucBuster Protein Extraction kit (Novagen, Darmstadt, Germany). The procedures were conducted according to the manufacturer's instructions, with the exception that two extraction rounds were performed instead of one, to reduce the level of cytoplasmic proteins in the nuclear extracts. The two cytoplasmic fractions and the nuclear fraction were analysed by Western blotting, as described above, using the polyclonal antibodies raised against Sd-Bra, Sd-Tbx2 and Sd-myotrophin.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

This work was supported by grants from the Deutsche Forschungsgemeinschaft (Mü 348/14), the Bundesministerium für Bildung und Forschung (project: Center of Excellence BIOTECmarin) and the International Human Frontier Science Program (RG-333/96-M).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References