By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Many aquatic sessile organisms (corals, hydrozoans, bryozoans, tunicates) are made up of modular units (usually called “polyps” or “zooids”) that cumulatively form structures termed “colonies,” the produced tessellation of their structural modules (Rinkevich,2002). Colony growth and development (known as astogeny; Boardman and Cheetham,1973) is characterized by iterative replication of the modules, a phenomenon that is controlled by a genetically mediated developmental program (Sánchez and Lasker,2003). However, although astogeny can be portrayed morphologically, it is unclear which genes participate or shape colony landscape (Rinkevich,2002) or which level of organization and astogenic features should be first studied to evaluate astogenic traits. One approach is to follow the participation of basic key genes that are related to RNA metabolism, and are important for embryo development such as the DEAD-box proteins.
Candidates of the DEAD-box protein family are putative RNA helicases presenting eight conserved amino acid motifs, that are named after one of the motif, DEAD (Asp-Glu-Ala-Asp; Linder et al.,1989; Wassarman and Steitz,1991; Schmid and Linder,1992). DEAD Box proteins are found in many organisms from bacteria to humans and are associated with processes related to RNA metabolism, such as transcription, splicing, translation, degradation, transport, and RNA stability (Schmid and Linder,1992; Iost and Dreyfus,1994; Py et al.,1996). They function as ATP-dependent RNA helicases, which unwind double stranded RNA. Some of them are also RNPases, remodeling RNA–protein interactions (Linder et al.,2001). DEAD box proteins are divided into several subfamilies according to their sequences and functions. The highly conserved PL10 (or Ded1p) subfamily, contains proteins that participate in translation initiation (Chuang et al.,1997; Iost et al.,1999; Linder,2003), nucleocytoplasmic transport (Askjaer et al.,1999, 2000), actively function in spermatogenesis (Foresta et al.,2000), in tissue development and virus infections (Chong et al.,2004; Yedavalli et al.,2004). However, despite the importance of PL10 in tight regulation of gene expression, our knowledge on PL10 functions in vertebrates is limited, anecdotal, and fragmented (Abdelhaleem,2005), and because only recently have they been grouped as subfamily, no consistent nomenclature exists.
The human genome contains two homologues of the PL10 family: DDX3 and DDX3Y (also known as DBX and DBY, respectively). DDX3 is a cytoplasmic protein, one of the five X-linked genes that have a homologue located in the nonrecombining region of the Y chromosome. DDX3 escapes chromosome X inactivation and is expressed at comparable levels in various male and female tissues (Lahn and Page,1999). This protein reacts with the hepatitis C virus core protein to inhibit capped RNA translation (Owsianka and Patel,1999; Mamiya and Worman,1999). When up-regulated in HIV-1 infection, DDX3 facilitates the export of long, singly spliced or unspliced HIV RNAs from the nucleus to the cytoplasm by means of the CRM1-Rev pathway (Yedavalli et al.,2004). Sequence analysis revealed that DDX3 shares 91% protein sequence identity with DDX3Y, the Y-linked homolog, which has two transcripts, differing in their 3′ end. The longer transcript is expressed in many tissues, but no protein product has been found. The shorter transcript is translated only in the testes (Foresta et al.,2000; Ditton et al.,2004), where it is predominantly found in spermatogonia and its removal leads to several testicular pathologies. A DDX3Y gene product has also been implicated in graft versus host disease (Vogt et al.,2002).
Mouse has three PL10 related genes: DDX3, DDX3Y, and PL10. DDX3 and DDX3Y are located on chromosomes X and Y, respectively, and are expressed in multiple tissues (Leroy et al.,1989; Gee and Conboy,1994; Sowden et al.,1995). The autosomal retrogene PL10 (known also as D1Pas1) is expressed in testicular tissues at the pachytene stage of male meiosis, the stage in which sex chromosomes are inactivated. These three genes exhibit 95–98% similarity at the amino acid level and nucleotide similarity of 84–90% (Gee and Conboy,1994). The zebrafish PL10 homologue is widely expressed during embryogenesis (Olsen et al.,1997). An3, the Xenopus laevis PL10 is localized in animal hemisphere of Xenopus oocytes (Gururajan and Weeks,1997). An3 binds with high affinity directly to CRM1 (Exportin 1), the nuclear export receptor, and a specific oocyte derived mRNA was proposed as an activator for its ATPase activity (Askjaer et al.,2000).
In the invertebrates (studied only in unitary organisms), PL10-related genes are important in spermatogenesis and in differentiation processes. Belle protein in Drosophila is essential for male and female fertility, for viability, and larval growth (Johnstone et al.,2005). CnPL10, the PL10 homologue in hydra, is expressed in germline cells and differentiating somatic cells of the interstitial cell lineage (Mochizuki et al.,2001). In Planaria, DjvlgA, the PL10 homologue, is expressed in ovaries and in spermatogonia, spermatocytes, and spermatids but not in sperm. In asexual planarians, the location of DjvlgA was determined to be in the mesenchymal space, possibly in neoblasts, and in regenerating tissues, in undifferentiated or differentiating tissues (Shibata et al.,1999). It is, therefore, of great interest to reveal the possible functions of this DEAD-box protein subfamily in ontogeny and colonial astogeny of ascidians, which are considered a key group in the study of chordates evolution (Cameron et al.,2000; Holland,2000).
BS-PL10: Sequence and Phylogenetic Analysis
The PL10 related gene from B. schlosseri (Fig. 1a,b, see details in the Experimental Procedures section) termed as BS-PL10 was isolated by reverse transcriptase-polymerase chain reaction (RT-PCR) and rapid amplification of cDNA ends (RACE) methods. The gene is composed of four exons and five introns, and the mRNA has 2,502 bases. The open reading frame (ORF) is of 2,130 bases, coding for a putative protein of 709 amino acids with a molecular weight of 78.4 kDa. Figure 1c depicts the nucleotide and the amino acid sequence of BS-PL10 ORF. It has 55–57% similarity to the PL10 protein from mammalians, zebrafish, and Xenopus. The N and C terminal regions of the protein contain nine RGG motifs known to be involved in RNA binding (Kiledjian and Dreyfus,1992). The highly conserved helicase core region of 352 amino acids (aa; marked with gray background; Fig. 1c) contains the eight characteristic sequence motifs of the DEAD-box family. The conserved core region protein of BS-PL10 reveals high similarity to Ciona intestinalis PL10 (85.7% similarity), and 74% similarity to PL10-related genes from mammals (human, chimpanzee, mouse, rat), Xenopus, and zebrafish. Decreased similarities were depicted for Hydra (71%), Drosophila (68%), and yeast (63%) homologues. The comparison of BS-PL10 gene to the Ciona intestinalis (a solitary ascidian whose genome was sequenced) genome database (Ciona intestinalis v1.0) identified only one homologue in the Ciona intestinalis genome, pointing on the existence of only one member of PL10 subfamily in ascidians.
We performed multiple amino acid alignments of the conserved core proteins originating from PL10 homologues with highest resemblance to BS-PL10 using CLUSTALX (1.81) software package. A phylogenetic tree by the neighbor-joining method (1,025 bootstraps) was produced (Fig. 1d). Phylogenetic and molecular evolutionary analyses were conducted using MEGA version 2.1 (Kumar et al.,2001). This phylogenetic analysis (Fig. 1d) reveals that the mammalian PL10 genes clustered into two groups according to their chromosomal location: the DDX3-related genes (located on chromosome X), and the DDX3Y related genes (located on chromosome Y). The mouse PL10 is clustered with the DDX3 group. BS-PL10 protein depicts the structure of an ancestral homologue to DDX3 and DDX3Y, with high amino acid sequence similarities to both. The appearance of, at least, two PL10 paralogues in mammals, one on chromosome X and one on chromosome Y, corresponds with the Lahn and Page (1999) idea that Y chromosome has developed shortly after the divergence of the mammalian and avian lineages.
BS-PL10 Expression in Total RNA and Protein Extracts
Thirty micrograms of total RNA was extracted from several Botryllus colonies at blastogenic stages A–D (Fig. 1b). The Northern blot analysis was performed using a probe of 1-kb fragment spanning the 3′ region of BS-PL10 mRNA. A single band of approximately 2.5 kb was depicted, fitting the size of the BS-PL10 mRNA isolated by PCR techniques (Fig. 2a).
We compared levels of BS-PL10 protein expression along blastogenic stages A–D in 10 laboratory-cultured Botryllus schlosseri colonies. Each genet was first subcloned into several ramets, and proteins were extracted when ramets reached the desired blastogenic stage. This protocol enabled us to compare results obtained from the same specific genetic background, age, and reproductive status differing only in their blastogenic stage. Samples were analyzed by Western blot analysis using anti BS-PL10 polyclonals. The control was preimmune serum derived from the same rabbit.
Figure 2b reveals the typical outcome represented by two unrelated genotypes. A major band of 78-kDa protein (a size that correlates with the predicted mass), not revealed when using the preimmune serum, is detected by the antibodies. In both cases (Fig. 2b), this protein was expressed in surprisingly high quantities and in blastogenic stage D (the takeover phase) or late stage C, whereas during blastogenic stages A and B, only traces of the protein were detected. Analysis of the other eight genotypes revealed a similar pattern of BS-PL10 protein expression in the colony: the highest expression was always at blastogenic stage D, while the lowest expression was observed at stages A and B. The expression at blastogenic stage C varied as the protein extracts were obtained from early or late stage C. Blastogenic stage D levels of BS-PL10 78-kDa protein were genotype specific. Some genotypes express more than others in repeated examinations.
As adult zooids and developing buds (that are residing side-by-side, embedded within the same colonial matrix) represent, simultaneously, two opposing colonial developmental trends (degeneration versus morphogenesis, respectively), we compared the 78-kDa protein expression patterns in isolated buds and in their corresponding adult zooids, at the four blastogenic stages (this analysis was first performed on young colonies, before commencing sexual reproduction). Western analyses results (Fig. 2c) revealed four to nine times higher expressions of BS-PL10 78-kDa protein in buds than in zooids, levels that were expressed at all blastogenic stages, whereas each zooid/bud pair represent a different genet. The age, health, and genetic origin (in gravid colonies also the reproductive state) of the colony influenced buds/zooids expression differences. In older animals, after the onset of reproduction, the difference is slightly lower (up to fourfold increase; data not shown)
In some of the samples, in addition to the 78-kDa protein, a fainter band of 71 kDa was detected. This was markedly observed in some samples coming from stage D buds (Fig. 2d). The 78-kDa protein, however, remained the major BS-PL10 protein.
Analysis of BS-PL10 protein expression in buds isolated from one colony (i.e., identical genet) at different stages of a single blastogenic cycle revealed that the quantity of BS-PL10 78-kDa protein per total bud protein extract was roughly similar in all blastogenic stages (Fig. 2d). There was no difference between sexually matured and young colonies (data not shown). The rapid increase in buds size during the end of blastogenic stage C and at stage D increased buds' contribution to total protein content, leading to a higher quantity of BS-PL10 protein per total colonial protein content (Fig. 2b).
In Situ Hybridization Assays
Two different sets of in situ hybridization assays were conducted on histological sections: (1) in situ RNA hybridization that was performed with specific antisense RNA probes (specific sense probes were used as controls); and (2) BS-PL10 protein distribution patterns that were analyzed with anti BS-PL10 polyclonal antibodies (rabbit preimmune serum was used as control). Results from some of these experiments are presented in Figure 3. In all these experiments, the controls revealed negative signals (data not shown).
BS-PL10 mRNA and protein expression patterns showed similar outcomes in all tissues and organs studied. The high levels of expression of both BS-PL10 mRNA and protein (Figs. 3a1 and 3a2, respectively) were recorded in primary and secondary buds at all blastogenic stages (regardless of sexual reproduction), in macrophage-like cell aggregates located in sinuses surrounding the endostyles, in early staged germ cells, and in certain organs of developing embryos.
In the developing buds, BS-PL10 mRNA and protein were highly expressed in all organs other than mature oocytes (see below). Certain cells, like those situated in the inner parts of the developing digestive system, expressed even higher levels. In the zooids, mRNA and protein expressions varied between different organs but were always lower than the levels in the buds (with the exception of the blood sinuses' macrophages; Fig. 3a5). The blastogenic developmental transition from a bud to functional zooid was also associated with dramatic reduction in both, BS-PL10 mRNA and protein levels. In some organs, like the branchial sac, we recorded very low expressions in blastogenic stage A, immediately after the onset of a new blastogenic cycle (up to 15 times below bud levels). In the digestive system, approximately 30–80% decrease was recorded at blastogenic stage A, followed by 2–4 times diminution in stages B and C. The decrease was not equally distributed within the same organ. Certain regions of the stomach and intestines had higher expressions than others. The endostyle also contained regions of relatively high and low expressions.
Oogenesis in B. schlosseri occurs in the developing buds. In situ hybridization, immunohistochemical and Western blot analyses revealed high expression of BS-PL10 mRNA and 78-kDa BS-PL10 protein (Fig. 3a3) in the cytoplasm of young oocytes (up to 50 μm). As oocytes mature, a sharp decrease of both BS-mRNA and proteins were recorded (Fig. 3a1 and 3a2), whereas the newly formed follicular cells around the oocyte expressed high levels. Only background levels of BS-PL10 mRNA and protein were recorded in mature and ovulating oocytes. Western blot analysis of BS-PL10 proteins expressed in ovulating eggs (Fig. 3b, lane oc) demonstrated very low quantities of the 78-kDa protein. This low expression probably comes from residual follicular cells around the oocytes.
Both BS-PL10 mRNA and the 78-kDa protein were strongly expressed in early stages of sperm cell development in the buds (Fig. 3a4). As sperm matured (in zooids of blastogenic stages A and B), a gradient of decreased expression of both BS-PL10 mRNA (Fig. 3a2) and protein was depicted. Along the development of the sperm, higher levels of expression were recorded in the periphery of the testis, the site of young spermatocytes, and both BS-PL10 mRNA and protein levels decreased toward the gonad center where mature, ready-to-be-released sperm cells are located. After sperm release, at blastogenic stage C, there was no BS-PL10 mRNA or protein expressions in the testes. The 78-kDa protein expression pattern supported these results (data not shown).
The macrophage-like cell compartment that is located in the blood sinuses near the endostyle of stage D primary buds and in zooids at all blastogenic stages, is the blood cell population with the strongest BS-PL10 protein staining (Fig. 3a2, 3a5). This cell population was classified as macrophage-like based on cell's size and other general morphology parameters visualized by hematoxylin and eosin staining (not shown) and by comparison to available literature (Hirose et al.,2003; Ballarin and Cima,2005). The macrophage-like cells in this compartment, contrary to other vasculature locations, are packed together in aggregates, which contain both positively and negatively stained cells. This is a stationary population of blood cells as revealed by observation in toto and by immunohistochemical analyses, with anti-actin antibodies that clearly distinguish circulating macrophages in peripheral regions from the sinus-residing macrophage population (Rosner et al., in preparation; see also Cima and Ballarin,2000).
Primary and Secondary Buds
The secondary buds appear as thickened discs of atrial epithelium on the primary buds of the previous generation. Secondary buds rapidly transform into closed spheres and further develop by processes of folding and local invaginations of the expanding spheres. The epidermal layer of each sphere gives rise to epidermis only, whereas the inner atrial layer develops to all other tissues and organs (Berrill,1951). High expressions of BS-PL10 mRNA (Fig. 3c1 and 3c2) and protein (Fig. 3c3-5) were detected in secondary buds from the moment a bud first appeared (Fig. 3c3), or when it further developed into the circular disc stage (Fig. 3c1). Strong expressions of BS-PL10 mRNA and protein were detected in further developmental stages of growth when secondary buds became primary buds (Fig. 3c1-5; Fig. 3a2). Stage D primary buds (ready to take over the colony from the older set of zooids), although containing fully developed tissues, still possessed high levels of both BS-PL10 mRNA and protein. These results coincided with the results obtained by Western blot analysis (Fig. 2d). Once buds opened their siphons and became functional zooids, the levels of both BS-PL10 mRNA and protein decreased in all organs.
Embryos collected at blastogenic stage A (gastrula is the final developmental stage that an embryo can reach at this stage) had very low BS-PL10 mRNA or protein (Fig. 3d1; Fig. 3d3). Western blotting further confirmed only basal levels of 78-kDa protein in embryos collected from blastogenic stage A colonies (Fig. 3b, lane e1), similar to the levels expressed in mature oocytes (Fig. 3b, lane oc). BS-PL10 protein in a developing embryo increased gradually. At tail bud stage (Fig. 3b, lane e2, e2′), embryos contained 5–10 times more BS-PL10 levels than at the gastrula stage. The maximum levels of BS-PL10 protein (1.5 times higher then in previous stage) was detected in wrapped-tail embryos that were ready-to-hatch (isolated from blastogenic stage C colonies, Fig. 3b, lane e3).
A ready-to-hatch tadpole larva contains approximately 2,000 cells, featuring the basic body plan of a chordate with several distinctly developed organs. Immunohistochemical analyses of these larvae (Fig. 3d4-6) revealed high expressions of BS-PL10 protein in the endostyle, digestive system, tail muscles, and adhesive organ (Hirano and Nishida,1997). We could not detect expressions in the notochord and in the visceral nerve. After being released, tadpole larvae settle and transform into the oozooid stage. The 78-kDa protein levels in oozoids were lower than the levels of wrapped-tail stage embryos and buds (Fig. 3b, lane oz). The 78-kDa quantities in embryos, at different developmental stages, were always lower then levels in the buds (Fig. 3b, lanes b and b′).
BS-PL10 Homologue in Botrylloides
Botrylloides leachi (Fig. 4b1) is another common genus of this tunicate family (styelidae). We used Western blot and immunohistochemical analyses with anti BS-PL10 antibodies on colonies of different Botrylloides subpopulations (SP) collected along the Israeli coast (Botrilloides SP I, II, and III; Rinkevich et al.,1994; Rinkevich,1995). The Western blot analysis revealed a 78-kDa protein similar to BS-PL10 protein in all studied Botrylloides subpopulations (Fig. 4a). Immunohistochemical analysis documented high levels of BS-PL10 homologue in buds and low expression in the zooids (Fig. 4b3 and 4b4). As in Botryllus schlosseri, aggregates of positively reacting macrophages were found in the blood sinuses of the zooids near the endostyle (Fig. 4b2).
High Levels of BS-PL10 Protein Expression Correlates With Normal Bud Development
The data presented so far on intact colonies demonstrated correlation between BS-PL10 protein level and developmental stages of the modules (buds and zooids). Because buds and zooids are interconnected (besides the common blood vasculature), zooidal possible influence on BS-PL10 protein level in the buds was examined. To address this question, we performed zooidectomy of 23 colonies. After zooidectomy, bud integrity was determined using light microscopy—buds with heartbeats were considered “healthy.” After siphon opening, buds were transformed into functioning zooids. BS-PL10 protein expressions in buds from zooidectomized colonies were compared with gene-matched controls by Western blotting. Of the 23 zooidectomized colonies, 11 showed normal bud development, similar to control colonies (Fig. 5a2 and 5a1, respectively). Western blot analysis showed (Fig. 5b) similar levels of BS-PL10 protein in control (lane 3b) and zooidectomized buds, 1 (lane 1b′), 2 (lane 2b′), and 3 (lane 3b′) days after surgery. BS-PL10 levels in control zooids (Fig. 5b, lane z) were low, as expected. Two of the colonies (zooidectomized at stage C) showed unsynchronized development of the left-buds (Fig. 5c2), 3 days after zooidectomy. Some of the buds in these colonies transformed into zooids, whereas the rest of the buds were of varying sizes (they were classified as “normal” or “small” sizes; Fig. 5c2) or were resorbed. Western blot analysis (Fig. 5d) of the colony shown on Figure 5c demonstrated a decrease in BS-PL10 expression (lane 1b′) 1 day after zooidectomy, which may represent an average between normal and dying buds. Three days after zooidectomy, this colony had buds with (lane 3nb′) BS-PL10 levels similar to those of the control buds (lane b), the small buds had lower levels (lane 3sb′), and the newly formed zooids (lane 3z′) showed basal expressions as control (lane z) zooids.
Roles of the BS-PL10 Protein
We used two protocols to deliver BS-PL10 siRNAs: submersion (used mainly on young colonies to reduce seawater volume) and direct injection into proximal ampullae. The submersed colonies (1–2 months old) were at different blastogenic stages. Because of their young age, only one colony from this group was fertile. The success of the siRNA submersion treatment was determined by quantification and comparison of the 78-kDa BS-PL10 proteins coming from siRNA, control, and recovering ramets (Fig. 6a). During the injection experiment, the ramets were injected once, and analyzed during the proceeding blastogenic cycles (7–14 days after injection). This protocol enabled us to study long-term impacts caused by short BS-PL10 protein depletion free of the stress caused by the procedure. However, protein BS-PL10 protein expression resumed partially or completely by the time of the examination. The results of the two series of siRNA experiments (submersion and injection) are summarized in Table 1.
The analyses demonstrated that the buds were more resistant to siRNA administration then zooids, probably due to higher levels of BS-PL10 protein there and were only partly affected by the siRNA treatment. From the two sets of experiments presented here, 44 experimental colonies (70%) showed various types and degrees of malformations. We divided the treatments outcomes into four groups: death, zooids/buds resorption (Fig. 6b3), impaired gametogenesis manifested by the failure to produce mature germ cell (sperm or oocytes; Fig. 6b5), and malformations related to blastogenesis. The latter group included colonies with shorter blastogenic cycles (lasting 5 instead of 7 days) or colonies which lost synchronized blastogenesis. This finding was best reflected when colonies were in transition from blastogenic stage C to D or from stage D to A (Fig. 6b4, 6b8, and 6b10). In these cases, part of the zooids were at delayed blastogenic stage, whereas the others had already passed to subsequent stages, a phenomenon observed before in stressed and old colonies (Rinkevich et al.,1992). During submersion, 12% and 14% of colonies died in BS-PL10 and control siRNA solutions, respectively. Hence, death of the whole colonies will not be considered in this study as a direct effect of the BS-PL10 siRNA treatment. The rest of the control siRNA treated colonies were normal (Fig. 6b1-2).
Ten BS-PL10 siRNA-treated colonies and four controls were further studied by immunohistochemistry, revealing additional histological changes related to BS-PL10 distribution. The control siRNA-treated colonies showed normal quantities and distributions patterns of the BS-PL10 protein (Fig. 6b2). However, many BS-PL10 siRNA-treated colonies that conserved the typical external system structure, showed very low staining of BS-PL10 protein in both buds and zooids and contained disrupted or partially disintegrated organs at various degrees of severity. Internal cavities of zooids and buds in the colonies shrunk, filled with digested material and blood cells, including signet, morula and macrophage-like cells, known to be involved in phagocytosis and inflammation processes. In healthy animals, these cell types do not normally reside within these cavities.
In other treated colonies where milder effects were recorded, blastogeneic stage A primary buds contained undeveloped tissue rudiments with high BS-PL10 protein buds (probably because siRNA has no impact on already existing protein; Fig. 6b5; normal bud of same stage is shown on Fig. 3c2 and 3c5). In these buds, organogenesis was aborted, because there was no synthesis of new BS-PL10 protein. Finally, these buds were absorbed by phagocytosis before the end the blastogenic cycle (data not shown). When older buds were treated with BS-PL10 siRNA, some continued the blastogenic cycle, turning into zooids that accumulated structural malformations. In severe malformations, the affected zooid was resorbed before other zooids of the same system (Fig. 6b6, arrow). Some malformations were not lethal as those shown on Figure 6b7. The arrow in this figure points to malformations in the endostyle morphology and abnormal tunic production.
Decreased BS-PL10 protein levels in zooids is resulted in their premature death. Colonial systems having only part of the zooids and buds with decreased BS-PL10 protein content, showed unsynchronized stage D/A transition (Fig. 6b8 and 6b9). When BS-PL10 protein decrease in the zooid was induced experimentally at early stage C, unexpected shrinkage of the zooid's internal organs, such as the branchial sac and digestive system was recorded (Fig. 6b10, the system without pigments and Fig. 6b12). This colony shrinkage was associated with 20–30% decrease in BS-PL10 protein level (analyzed by the TINA software package) compared with control zooids (Fig. 6b10-the pigmented system and Fig. 6b11). Although all buds in both systems of this colony were in similar developmental stage, the death of the zooids in one of the systems accelerated its takeover process. These siRNA experiments suggest that the resorption of zooids at the end of each blastogenic cycle, involves the participation of PL10 and is expressed as sudden decrease in BS-PL10 housekeeping levels.
Constraints posed by astogeny on colonial architecture are not only environmentally mediated parameters but also have a clear genetic trait (Rinkevich,2002). However, nothing yet is known about the genes that participate or control the complex structures of colonies and the well-controlled developmental processes of colony astogeny. In this study, we isolated the styelid colonial ascidian Botryllus schlosseri BS-PL10 gene, followed and manipulated its expression and function during the various stages of life history stages of ontogeny and colonial astogeny. The BS-PL10 gene has only one transcript, while the specific antibodies recognize two protein products: a major band of 78 kDa and a minor band of 71 kDa, represented in much lower quantities than the 78-kDa product in some of the genets. We assume that this 71-kDa protein is a degradative form of the 78-kDa protein. Alternatively, the 71-kDa protein is another member of the DEAD box protein family.
B. schlosseri colony astogeny is characterized by two unique intrinsic patterns of BS-PL10 expression: cyclical expression at the whole colony level (blastogenesis) and the sharp decline (decreased expressions in time) at the zooid/organ levels. The zooids, or the functional units, always express low levels of BS-PL10 protein, whereas the developing units, the primary and secondary buds, express high levels of the protein (Fig. 7).
As specified, BS-PL10 protein levels in the colonies fluctuate in a weekly cyclical pattern that follows the blastogenic cycle of B. schlosseri: low levels at blastogenic stages A and B, rapid increase during blastogenic stage C, and maximum levels at blastogenic stage D (Fig. 7). The expression of BS-PL10 protein in the buds is regulated independently of the parent zooids. After zooidectomy, left-over buds continue to express high levels of BS-PL10 protein during all developmental stages and then are transformed into zooids that express low levels of BS-PL10 protein. However, the existence of intact zooids is needed to synchronized BS-PL10 protein level decrease among the buds in a colony during their transformation into zooids. Thus, during the lifespan of a single module, a bud, which turns into a functional zooid (duration: 3 weeks at 18–20°C), shows BS-PL10 protein levels that fluctuate and drop twice: once when the bud matures into a filtering zooid, and then, when the mature zooid degenerates (at blastogenic stage D), before its resorption. The pattern of high BS-PL10 protein levels at differentiating organs vs. low levels at homeostatic zooid state, resemble the outcomes reported for DDX3 (Yedavalli et al.,2004) and other RNA helicases of this family (Krishnan and Zeichner,2004) during viral infections.
BS-PL10 mRNA and protein levels, however, diverge between different organs/tissues. High expressions are always recorded in primary and secondary buds' organs and in the subpopulation of macrophages aggregate near the endostyles. Differential expressions are also detected in the zooidal organs where certain organs (stomach, intestine, and regions in the endostyle) possess higher levels than other organs (i.e., branchial sac). Organs that express more BS-PL10 protein are also the organs last to be resorbed during the take over process of blastogenesis. The possible diverging roles of BS-PL10 in different Botryllus developmental pathways are also marked in the siRNA treatment (Fig. 7, enclosure) that causes unlike outcomes: degeneration in low expressing organs, malformations in highly expressing organs; on the colony level blastogenic cycle is accelerated but blastogenic synchronization is impaired. The above-mentioned results suggest that correct levels of BS-PL10 are important for normal blastogenesis where BS-PL10 levels fluctuate in harmony with this developmental process.
Two lineages of pluripotent stem cells, the somatic and the germ cell lines, reside side by side in adult Botryllus colonies (Stoner et al.,1999). Concerning the somatic stem cell lineage, we identified the earliest differentiating soma cells in the secondary buds, appearing as thickened discs on the primary bud's atrial epithelium. Early in development, these cells express high levels of BS-PL10 protein, whereas differentiated soma cells express low levels. In a similar way, the germ cells, oocytes, and sperm, also express high levels of BS-PL10 mRNA and protein at early stages of their development and differentiation. These levels drop to background levels when they mature. The follicular cells around the oocytes, however, continue to express high levels of BS-PL10 mRNA and protein until ovulation. The advantage of working with B. schlosseri as a model species is that early differentiating soma and germ cells reside in discrete tissues (buds and gonadal sac, respectively), well separated from the mature tissues. Thus, although there are no known markers for stem cells and progenitors in this species, based on B. schlosseri anatomy, we may assume that higher levels of BS-PL10 mRNA and protein are associated with multipotent cells, early differentiating soma cells, germ cells, and nursing cells.
In colony astogeny, buds are formed from tissues expressing high levels of BS-PL10 protein and, then, they continue to express these levels until maturing into functional zooids. During ontogeny, however, embryos show a different pattern. In young embryos, only background levels of BS-PL10 mRNA and proteins are detected. In the advanced developmental stages of the embryo, expression levels increase in some organs (adhesive papillae, digestive system, endostyle, and tail muscles) but not in others (notochord, visceral nerve). The BS-PL10 pattern of expression differentiates, therefore, between these two developmental processes, although both processes, at different phases of colonial ontogeny and astogeny, produce similar modules (blastozooids and oozooids, respectively) that morphologically differ only in branchial stigmata (Manni et al.,1999, 2002).
Both, normal BS-PL10 expression portraits and the experimental manipulations used (zooidectomy and the siRNA assays) demonstrate that high levels of BS-PL10 expressions are regulated along colonial blastogenesis and tissue/organ specifications, closely associated with somatic cell differentiation/organogenesis as well as with germ cell (sperm and oocytes) development (Fig. 7). In organisms that have more than one PL10 homologue, such as the mammals, we expect to find a trend for specialization in PL10 traits, distributed between the different genes. An example is DDX3Y in human and PL10 in mice, which specifically function in spermatogenesis. In organisms with only a single PL10 gene, like B. schlosseri, it is expected that this gene will participate in a wide range of biological activities, some of which represent pivotal outcomes.
One of the major methods to study BS-PL10 protein function was the knockdown by siRNA treatment. As an RNA helicase, BS-PL10 may affect expressions of other proteins. It is not possible yet to perform a rescue experiment in the Botryllus system to further confirm siRNA specificity, because suitable vectors do not exist. However, similar siRNA protocols were successfully performed on Botryllus in other laboratories (Laird et al.,2005), pointing to the specificity of this protocol.
We used laboratory-bred colonies originating from specimen collected at Monterey, Half Moon Bay and Moss Landing Marinas, California. The animals were maintained on glass slides at 20°C as described by Rinkevich and Shapira (1998).
The urochordate Botryllus schlosseri (Fig. 1a) is a very common encrusting colonial sea squirt, probably a Mediterranean species that has spread ubiquitously (Berrill,1950). Each colony is composed of few to thousands of genetically identical modules (zooids), each 1–3 mm long, embedded within a gelatinous matrix, the tunic. Zooids are arranged in star-shaped structures, called systems. A network of blood vessels connects all zooids within a colony, from which blood vessel termini (ampullae) extend toward the colony margins. In B. schlosseri, as in other ascidians, a nonfeeding, short-living pelagic larva metamorphoses into a, filter-feeding adult ascidian that does not have notochord. The larval stage develops as a chordate tadpole and has a notochord, a neural tube, a bilaterally symmetrical segmented musculature, and a tail; organs that are lost during metamorphosis. However, even the adult ascidian shares several important characteristics with the chordates, i.e., the endostyle (considered homologous to the thyroid) and the gill openings in the branchial basket, which develop from the protostigmata, the homologues of the vertebrates gill slits.
As in all botryllid ascidians, the B. schlosseri colony grows by accreting the number of zooids through a highly synchronized and cyclical development process called palleal budding or blastogenesis (Berrill,1950, 1951). Blastogenesis is divided into four major stages, A–D (Fig. 1b; sensu Mukai and Watanabe,1976b), and lasts approximately 1 week at 18–20°C. This is a highly tuned developmental cycle, in which (1) a new set of zooids is established through the concurrent development and maturation of one to four primary buds per zooid, whereas, (2) the parental set of functional zooids is simultaneously deteriorating and is morphologically eliminated. During blastogenic stages A–C, bud tissues are differentiated and internal organs are formed; the functional generation of zooids remains active. At blastogenic stage D, the last 24–36 hr of each blastogenic cycle, all zooidal tissues in the colony die, mainly by an apoptotic process, and are phagocytized by specialized blood-borne cells, the macrophages (Lauzon et al.,1992, 1993; Cima et al.,2003). Zooid apoptosis gradually advances in a wave-like manner from the anterior end of each zooid toward its posterior end (Lauzon et al.,1996). Simultaneously, the developing buds mature into the new generation of functional zooids.
In young colonies, no gonads are detected and the colony undergoes strictly blastogenesis. Then, testes develop with or without rudimentary undeveloped ova. In sexually matured colonies, testes and ova are situated side-by-side within the same bud. Ova do not mature in a single blastogenic generation; they are transported to the developing buds of the succeeding generations until ovulation and fertilization are finally attained with precise coordination between sexual reproduction and blastogenesis (Berrill,1950; Milkman,1967; Mukai and Watanabe,1976a). Embryogenesis also follows the blastogenic cycle, so that embryos mature and hatch just before adult zooids' regression.
Zooids were removed from colonies at blastogenic stages A–C (following that of Zaniolo et al.,1976; Lauzon et al.,2002). Then, we monitored bud development in the zooidectomized systems, comparing them to bud development in control intact subclones of the same genet. At two or three time points after the surgery (usually after 1, 2, and 3 days), buds were surgically removed from the system, and their proteins were extracted. After the third sampling, the control system was sacrificed and total proteins were extracted from control buds and control zooids, separately. BS-PL10 expression was analyzed by Western immunoblotting.
Isolation of BS-PL10 cDNA and Gene Sequencing
DNA was extracted as described by Graham (1978). Total RNA was isolated with RNeasy Mini or Midi kits (Qiagen, Valencia, CA). Degenerate primers were designated according to the conserved DEAD-box region of RNA helicases. Primers were synthesized by Proligo Primers & Probes (www.proligo.com, Germany) and used for PCR reactions (40 cycles at 94°C for 30 sec, 30 sec at 50°C, and 60 sec at 72°C). An initial 400-bp fragment was isolated, and further enlarged by several consecutive TAIL-PCR reactions performed as described by Liu et al. (1995). First, full-length BS-PL10 cDNA was achieved by synthesizing the total cDNA with reverse transcription kit (ImProm-II, Promega, Madison, WI). Then, the 5′ and the 3′ regions were isolated, respectively, by using 5′-RACE system kit for rapid amplification of cDNA ends (Gibco BRL Reagent Assembly version 2, manufactured by Invitrogen-Life Technologies, Gaithersburg, MD), and RLM–RACE kit (catalog no. 1700 Ambion, Austin, TX). Full-length BS-PL10 cDNA was synthesized using “Expand High Fidelity PCR system” (Roche Applied Science, Mannheim, Germany). All DNA fragments were subcloned into pDrive Cloning Vector (PCR Cloning Kit-QIAGEN, Valencia, CA) and sequenced for both strands
Labeling of the probe was done by DNA Labeling Kit (HexaLabel no. K0612, Fermentas, Hanover, MD). Hybridization was at 65°C in a solution of 5× standard saline citrate, 5 × Denhardt, 0.5% sodium dodecyl sulfate (SDS), and 10 μg/ml ssDNA.
In Situ Hybridization
Animals were fixed in 4% paraformaldehyde, dehydrated, and embedded in paraffin. Sections of 5 μm were made by hand microtome. Digoxigenin (DIG) -labeled RNA probes (sense and antisense) were synthesized using DIG RNA Labeling Kit (SP6/T7; Roche Molecular Biochemicals, Mannheim, Germany). In situ hybridization of the labeled probes to the tissue sections was performed in the method developed by Breitschopf et al. (1992) for paraffin-embedded tissue. The samples were observed with Olympus BX50 Upright microscope and photographed by Supercam camera (Applitec, Holon, Israel).
Polyclonal Antibody Production and Immunohistochemistry
A 930-bp SalI–PstI fragment containing the 5′ and part of the consensus regions of the BS-PL10 cDNA was subcloned into QIAexpress pQE30 vector (Qiagen, Valencia, CA). The plasmid was transformed into M15 bacteria and the 6×His-BS-PL10 chimeric protein was induced by IPTG. Large quantities of chimeric protein were purified by Ni-NTA Agarose (Qiagen) according to manufacturer's instructions. The purified protein was injected into two rabbits after collecting preimmunized serum (Sigma, Israel). The sera obtained from immunized rabbits were further cleaned by affinity columns. The 6×His-BS-PL10 chimeric protein was coupled to CNBr-activated Sepharose 4B (Amersham-Pharmacia, Uppsala, Sweden) and used to purify anti BS-PL10 antibodies from the immunized serum
Imunohistochemistry was performed as described by Lapidot et al. (2003) with the following modification: antigen retrieval was performed by microwaving the sections for 30 min in 1 mM ethylenediaminetetraacetic acid (EDTA), pH 8.0 solution. The affinity purified anti–BS-PL10 polyclonal antibodies were used at 1:2,000 dilutions. The secondary antibody, goat anti rabbit AP (Jackson Immunoresearch Laboratories, Inc., West Grove, PA) was used at 1:10,000 dilution.
Protein Extraction and Immunoblotting
Tissue samples were lysed in a solution containing 50 mM Tris pH 6.8, 100 mM dithiothreitol (or 250 mM β-mercaptoethanol) and 2% SDS. We sampled a single zooid or two buds per 4 μl of solution. Protein concentration was determined by dot blot using “TINA” software package version 2.07d (Raytest, Germany).
Samples, 3 μg of protein extract each, were subsequently separated on the basis of their molecular mass using 7–10% SDS-polyacrylamide gels, and transferred to nitrocellulose papers. BS-PL10 antibodies were used at 1:2,000 dilution. The secondary antibody, goat anti-rabbit horseradish peroxidase (Sigma, St. Louis, MO) was used at 1:12,000 dilutions. Signals were detected using SuperSignal West Pico Chemiluminescent Substrate kit (PIERCE, Rockford, IL).
Synthesis and Administration of siRNA
Four different siRNA from the 5′ region of BS-PL10 mRNA were synthesized and purified using Silencer siRNA Construction Kit (Ambion, Austin, TX). The specific primers used to synthesize these siRNAs were as follows: A15 5′-AACAAGAGCAATATCGCGGACCCTGTCTC-3′ and A13 5′-AAGTCCGCGATATTGCTCTTGCCTGTCTC-3′; A25 5′-AATGGGATGAAAGACGAGGAGC-CTGTCTC-3′ and A23 5′-AACTCCTCGTCTTTCATCCCACCTGTCTC-3′; A35 5′-AAGATTGGAATATAGCTCTCCCCTGTCTC-3′ and A33 5′-AAGGAGAGCTATATTCCAATCCCTGTC-TC-3′; A55 5′-AAATACGACGATATACCAGTGCCTGTCTC-3′ and A53 5′-AACACTGGTATATCGTCGTATCCT-GTCTC-3′. An additional set of primers was used to synthesize an unrelated control siRNA (Laird et al.,2005) derived from plasmid sequence: Control5 5′-AACCATCTGCTAATCTGTAAC-CCTGTCTC-3′ and Control3 5′-AAGTTACAGATTAGCAGATGGCCTGTC-TC-3′.
The siRNA (positive or control) were delivered to animals by two methods: (1) Animal submersion: young animals were soaked in seawater containing a mixture of the four BS-PL10–specific siRNAs at a concentration of 10 nM each or 10 nM control siRNA. The seawater was changed every other day. Animals were checked daily. Developing malformed colonies and controls were picked for further analyses. Specific BS-PL-10 knockdown was investigated by Western blotting. (2) Animal injection: Colonies were subcloned 1 week before injections, to get blastogenic cycle-matched genotypes for simultaneous positive and control injections. The four BS-PL10–derived siRNA oligonucleotides were mixed (final concentration of 10 μM, each) in buffer solution containing 25 mM HEPES, 10 mM cysteine, 50 mM EDTA in seawater. The control siRNA oligonucleotide final concentration was 40 μM. We microinjected 3 μl of either control or positive siRNA mixture into peripheral B. schlosseri ampullae, using Narishige (Japan) microinjector and micromanipulator. To avoid artifacts caused by the injection and related stress, observations on the colonies were started from one blastogenic generation after the injection.
We thank M. Tom, E. Moiseeva, and C. Rabinowitz for excellent technical assistance.