SEARCH

SEARCH BY CITATION

Keywords:

  • starfish embryo;
  • fibrous ECM;
  • mesenchyme cells;
  • monoclonal antibody;
  • scaffold;
  • contraction;
  • shape formation;
  • morphogenesis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

By using a monoclonal antibody (4H11 Mab), we have investigated morphogenetic functions of a fibrous component of the blastocoelic extracellular matrix in relation to cellular activities during early development of the starfish Asterina pectinifera. The 4H11 fibers fill the blastocoele from the late-cleavage to late-gastrula stage and contain the 370-kDa proteinaceous molecule secreted only by the epithelial cells. When 4H11 Mab is introduced into the blastocoele of blastulae, the embryos reveal three distinct morphological abnormalities after the mid-gastrula stage: (1) Distribution of mesenchyme cells confined near the tip of the archenteron, (2) swelling of the posterior ectoderm, and (3) suppressed growth of the mouth, esophagus, and coelomic pouches. These abnormalities occur together with alterations in the distribution of the 4H11 fibers. In embryos recovering from the effect of 4H11 Mab, the mesenchyme cells rearrange the 4H11 fibers. We propose that 4H11 fibers play direct roles in the morphogenesis of starfish embryos by providing a dynamic scaffold not only for the mesenchyme cells but also for the epithelial cells. Moreover, 4H11 fibers have a resist force from within, in concert with the mesenchyme cells, to counter the bulging force intrinsic to the epithelia and hold the epithelia in specific positions, once the positions have been decided. Developmental Dynamics 232:915–927, 2005. © 2005 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In certain developing multicellular animals, the extracellular matrix (ECM) is steadily synthesized by cells to organize the blastocoelic space of the embryonic body. Starfish embryos offer many advantages for studying the in vivo morphogenetic function of the blastocoelic ECM and interaction between it and the cells. Starfish embryos accommodate a large amount of ECM in their spacious blastocoele after the blastula stage. The transparency of the cellular component, which consists of two epithelial monolayers (the ectodermal and endodermal walls) and the mesenchyme cells, allows us to observe the interaction between the ECM and cells in living states during normal development and under experimental conditions.

Crawford and his colleagues have performed scanning and transmission electron microscopy experiments on the blastocoelic ECM of embryos of the starfish Pisaster ochraceus. According to their findings, a fibrous ECM, a dominant component of the blastocoelic ECM, appears in the early-gastrulae as short fibrils closely associated with the ectodermal and endodermal walls (Abed and Crawford,1986). The ECM grows into a branched meshwork by the mid-gastrula stage, which populates the posterior portion of the blastocoele more densely. The meshwork consists of longer strands that radiate from the esophagus to the ectodermal wall by the time the mesenchyme cells make their appearance. Based on the observation that mesenchyme cells actively traverse the region of the blastocoele occupied by these esophageal strands (Crawford and Chia,1982), they speculated that the fibrous ECM plays a role as a scaffold onto which the mesenchyme cells migrate and that the mesenchyme cells rearrange the fibrous ECM into a radial pattern.

In bipinnaria larvae, the fibrous ECM was manifested as 20-nm fibers laid out singly or in bundles in close association with the basal lamina of the ectodermal wall (Crawford,1989; Crawford et al.,1997). Condensation of the ECM appeared in three places, i.e., under the dorsoposterior ectoderm, lining the stomodeum in a long array, and between the esophagus and the dorsal ectodermal wall, where the “waist” is situated (Crawford,1990; Crawford et al.,1997). Based on these facts, Crawford and his colleagues speculated that the fibrous ECM functions to support the shape of the body wall, while the mesenchyme cells modify it by changing its length. However, they presented little experimental evidence to prove this attractive suggestion.

We have devised a method to introduce the antibody into the blastocoele of living starfish embryos by treating them with Ca2+-free seawater (CFSW) containing the antibody (Kaneko et al.,1995b). We have also reported preliminary results on the introduction of a monoclonal antibody (4H11 Mab) specifically recognizing a fibrous component of embryonic ECM into starfish blastulae. The Mab affected the movement of the mesenchyme cells.

In this study, we examined the dynamics, molecular nature, and functions of 4H11 fibers, a 4H11 Mab-positive fibrous ECM, during embryogenesis of the starfish Asterina pectinifera. 4H11 fibers contain a 370-kDa proteinaceous molecule synthesized by the epithelial cells not by the mesenchyme cells. 4H11 Mab caused an abnormal distribution of the fibers through its divalence. The abnormal distribution of the fibers, in turn, caused abnormal distribution of the mesenchyme cells, on one hand, and morphological abnormality and retardation in the ectoderm and the endoderm, on the other hand. Recovery from these abnormalities was eventually observed, owing partially to the activity of the mesenchyme cells. The functions of 4H11 fibers in embryogenesis are discussed in relation to the mesenchyme cell activity and to the fibrous ECM reported by Crawford and his colleagues.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

4H11 Antigen in Normal Embryos

Figure 1 shows the temporal and spatial distribution of the 4H11 antigen detected by indirect immunostaining of eggs and embryos at various stages of development. No staining was detected in immature eggs (Fig. 1A,B). The antigen was discerned first in the newly formed blastocoele as well as on the basal cell surface at the late-cleavage stage (128-cell stage; Fig. 1C,D). In the just-hatched blastula, the antigen increased in amount and became condensed along the inner surface of the blastula wall on or in close association with the basal lamina (Fig. 1E,F). At the mid-gastrula stage, the antigen began to extend into the blastocoelic cavity in a fibrous form, while the remaining bulk condensed along the inner surface of the ectodermal and endodermal wall (Fig. 1G,H). The fibrous form of the antigen was more condensed in the space sandwiched between the archenteron and the posterior ectoderm compared with the anterior portion of the blastocoele. The fibrous material filled the blastocoelic cavity by the late-gastrula stage and was more condensed in a region between the archenteron and the lateral ectodermal wall (Fig. 1I, arrow, J). At the bipinnaria stage, the condensation of the fibrous material was more prominent in areas where the two epithelia are relatively close to one another, such as at the oral hood, the dorsal ectoderm, and the posterior protrusion (Fig. 1K, arrows, L). The cytoplasm of the cells was not stained in the stages examined. In the control embryos, no immunoreactivity was detected anywhere at any of the stages examined (photo not shown).

thumbnail image

Figure 1. Distribution of the 4H11 antigen in the normal course of development of Asterina pectinifera. A–L: Paired indirect immunofluorescent and phase-contrast photomicrographs of sections were taken at the following stages: unfertilized egg (A,B), early-blastula (C,D), late-blastula (E,F), mid-gastrula (G,H), late-gastrula (I,J), and bipinnaria (K,L). The antigen was detected in the blastocoelic cavity after the early-blastula stage. The contrast of C was adjusted to be slightly weaker than other darkfield images to show the weakness of the fluorescence. G,K: Condensation of the antigen lining of the inner surface of the body wall is frequently seen to be displaced from the position at sites where the wall is expanding rapidly (arrowheads in G). I,K: The antigen is also seen to condense under the ectodermal walls where two epithelia are held relatively close to one another (arrows). See text for further explanation. Scale bar = 100 μm in B (applies to A–L).

Download figure to PowerPoint

Figure 2 shows a closer view of the features of the 4H11 antigen. In the paraformaldehyde in seawater (PFA–SW) -fixed whole gastrulae (Fig. 2A), the antigen appeared as a mixture of three different states, i.e., thin fibers, bundles of these fibers, and an amorphous state (Fig. 2B,C). The amorphous portion of the antigen was examined at a higher magnification in the anterior portion of crushed, unfixed embryos. The antigen was found to be a three-dimensional meshwork of a fibrous material adhering to one another to form thicker and longer fibers (Fig. 2D). We will call these fibers “4H11 fibers.”

thumbnail image

Figure 2. Higher magnifications of the 4H11 antigen filling the blastocoelic cavity. A–C: 4H11 antigen in late-gastrula stage is shown by indirect immunofluorescence. B,C: 4H11 antigen is seen in different forms, such as thin fibers (white arrowheads), bundles of fibers (black arrowheads), and an amorphous form (arrow). D: A late-gastrula was crushed alive and stained directly with fluorescein isothiocyanate–4H11 monoclonal antibody, and the anterior portion was photographed. Note the fibrous nature of the antigen and the meshwork made from the fibers in an unfixed state. Scale bar = 50 μm in A,D, 10 μm in B (applies to B,C).

Download figure to PowerPoint

In Figure 2B, the condensation of 4H11 fibers lining the epithelium was seen to be displaced as thick bundles from the wall surrounding the blastopore (Fig. 2B). Similar displacement was observed in embryos at different developmental stages (Fig. 1G, arrowheads; K) and CFSW-treated embryos (Fig. 6A, arrows). This displacement seems to concern regions of epithelia that are quickly growing or expanding.

Immunobiochemical Features of 4H11 Fibers

To examine the molecular features of 4H11 fibers, three immunoblotting experiments were done. Figure 3A shows the results from the first, in which isolated whole ECM derived from embryos at two different developmental stages were compared for equal protein-amount bases over a relatively prolonged exposure time for detection (Fig. 3D,F). In the ECM of the mid-gastrula stage (Fig. 3D,E), 4H11 Mab recognized a band of apparent molecular mass of 370 kDa and bands showing higher molecular mass (Fig. 3A, lane 1). No bands other than the 370-kDa band could be clarified further. The ECM of the late-gastrula stage (Fig. 3F,G) showed only a weak signal of the 370-kDa band (Fig. 3A, lane 2), suggesting that some other ECM component(s) had replaced the 4H11 fibers. In both ECM samples, the signal was found at the top of the gel (Fig. 3A).

thumbnail image

Figure 3. Western blotting analysis and the isolated whole extracellular matrix (ECM). A: Comparison of 4H11 antigen in the isolated whole ECM between mid-gastrulae (lane 1) and late-gastrulae (lane 2). Total protein to apply, 1 μg; exposure time for detection, 6 min. B: Effect of enzyme digestion on immunoreactivity of 4H11 antigen. Isolated whole ECM of mid-gastrulae was treated without enzyme (lane 1) or with trypsin (lane 2) or peptide-N4-(N-acetyl-b-D-glucosaminyl) asparagine amidase F (PNGase F; lane 3) and immunoblotted with 4H11 monoclonal antibody. The filled arrowhead, 370 kDa; the open arrowhead, 344 kDa. Total protein to apply, 1 μg; exposure time for detection, 1 min. C: Immunoblotting of various components constituting mid-gastrulae: lane 1, the isolated whole ECM; lane 2, the whole embryo; lane 3, the mesenchyme cells; lane 4, the epithelial cells. Total protein to apply, 15 μg; exposure time for detection, 1 min (except for lane 1 in which 5 μg of total protein was applied). D–G: Isolated whole ECM prepared from mid-gastrula (D,E) and late-gastrula (F,G) are shown as light and dark images under laser scanning microscopy. They show that the original shape of embryos and the configuration of 4H11 fibers are preserved in the isolated whole ECM. Scale bar = 50 μm in D (applies to D–G).

Download figure to PowerPoint

In the second series of experiments, isolated whole ECM of the mid-gastrulae stage was examined after digestion over a shorter exposure time for detection than was performed for Figure 3A (Fig. 3B, lane 1). No band was detectable after the isolated whole ECM had been treated with trypsin (Fig. 3B, lane 2). Peptide-N4-(N-acetyl-b-D-glucosaminyl) asparagine amidase F (PNGase F) treatment reduced the apparent molecular mass of the 4H11 antigen to 344 kDa (Fig. 3B, lane 3). These results indicate that 4H11 antigen is proteinaceous in nature and contains at least the 26 kDa of the N-linked oligosaccharide(s).

The third series was carried out to find out whether the 4H11 antigen was synthesized by the mesenchyme cells as well as the epithelial cells. Because trypsin completely digested the 4H11 antigen, a population of cultured mesenchyme cells and epithelial cells dissociated in the late-gastrula stage were treated with trypsin to remove the already secreted 4H11 fibers from the samples (see Experimental Procedures section). When the detection was done with the same exposure time as Figure 3B, whole embryos showed the same amount of 370-kDa signal as the isolated whole ECM (Fig. 3C, lanes 1, 2). Immunoreactivity was detected for the 370-kDa band of the epithelial cells (Fig. 3C, lane 4) but not for that of the mesenchyme cells (Fig. 3C, lane 3). This result shows that 4H11 antigen is synthesized only by the epithelial cells, despite no detection of antigen inside the epithelial cells by immunofluorescence microscopy (cf, Fig. 1).

Morphological Abnormalities Induced by 4H11 Mab

In an attempt to examine the function of 4H11 fibers in morphogenesis, the fluorescein isothiocyanate (FITC) -4H11 Mab was introduced into the blastocoele of the late-blastula stage by the CFSW method (see Experimental Procedures section). In control experiments, FITC–IgG (the second antibody for the indirect immunofluorescence microscopy) was used instead of FITC–4H11 Mab. The embryos were allowed to develop to the bipinnaria stage and were observed in whole-mounts.

The FITC–4H11 Mab-treated embryos developed normally for the first 10 hr after the treatment (Fig. 4A,C). The fluorescence of the Mab was found to homogeneously fill the blastocoelic cavity at the beginning of the observation (Fig. 4B) but gradually condensed to the periphery in the later stages (Fig. 4D,F,I). The fluorescence-positive area, on the other hand, began to become detached from the growing ectodermal and endodermal walls without changing its original size and shape (Fig. 4D). The first morphological abnormalities took place approximately 20 hr after the treatment. In contrast to the control embryos, whose mesenchyme cells were distributed throughout the blastocoelic cavity (Fig. 4G), most of the mesenchyme cells of the experimental embryos were confined to an area around the tip of the archenteron (Fig. 4E, arrowheads), which coincides with the periphery of the fluorescent-positive area (Fig. 4E,F). Here, we refer to this peculiar state as the “balloon-like condensation;” it was detached from the ectodermal and the posterior half of the endodermal wall, covering only the bulged tip of the archenteron. The second abnormality was found in the shape of the embryo, i.e., the posterior half of the lateral body wall was bulged (Fig. 4E, arrows) compared with the control embryos (Fig. 4G).

thumbnail image

Figure 4. Effects of fluorescein isothiocyanate–4H11 monoclonal antibody (FITC–4H11 Mab) on morphogenesis. A–F,H,I,K,L: Paired light (A,C,E,H,K) and direct immunofluorescence (B,D,F,I,L) photomicrographs of embryos in whole-mounts were taken at 2.5 hr (A,B), 10 hr (C,D), 20 hr (E,F), 30 hr (H,I), and 50 hr (K,L) after treatment with Ca2+-free seawater containing FITC–4H11 Mab. G,J,M: The control embryos in whole-mounts, to which were applied FITC–IgG in place of FITC–4H11 Mab, are shown only in light photomicrographs. E,H,K: The time after the treatment of these control embryos corresponds to that of E, H, and K, respectively. c, coelomic pouch; e, esophagus; m, mouth; s, stomodeum. See text for further explanations. Scale bar = 50 μm in A (applies to A–M).

Download figure to PowerPoint

By 30 hr, the balloon-like condensation had deflated and the mesenchyme cells had become dispersed throughout the blastocoelic cavity, while the posterior ectodermal wall was recovering its constricted form (Fig. 4H,I). Some of the mesenchyme cells were seen to stretch between the balloon-like condensation and the ectodermal wall where new constrictions were being formed (Fig. 4H, arrowheads). On the other hand, new abnormalities became apparent in the archenteron. Although the compartmentalization was taking place normally, the mouth, the esophagus, and the coelomic pouches were smaller (Fig. 4H) than those of the control embryos (Fig. 4J).

By 50 hr, both the mesenchyme cells and the posterior ectodermal wall had resumed their normal profile (Fig. 4K). Retardation of esophagus and coelomic pouches was still obvious while the stomodeum had acquired its normal size (Fig. 4K,M). The fluorescence of the balloon-like condensation had been translocated to the mesenchyme cells, the coelomic pouch cells, and the cells of the digestive tract (Fig. 4L). These effects were observed in most of the treated embryos. The control embryos were morphologically indistinguishable from the normal embryos at all stages examined.

When the embryos were treated at the late-blastula stage with Fab fragments of 4H11 Mab, no morphological abnormality was observed at 20 hr after the treatment (Fig. 5A; cf. Fig. 4G) or thereafter. The presence of the Fab fragments was confirmed by processing the embryos with the secondary antibody for indirect immunofluorescence. The Fab fragments were not only found in the blastocoele (Fig. 5B) but also became bound specifically to the 4H11 fibers (Fig. 5C). No staining was detected in the control embryos (photo not shown).

thumbnail image

Figure 5. Effect of Fab fragments of 4H11 monoclonal antibody on morphogenesis. A,B: Paired light (A) and indirect immunofluorescence (B) photomicrographs of embryos in a whole-mount were taken 20 hr after the treatment with Ca2+-free seawater containing the Fab fragments. A,B: No morphological abnormality is discerned (A), while the fluorescence is present throughout the blastocoelic cavity (B). C: A higher magnification of a squashed embryo treated in the same manner as in B. The fluorescence is manifested in the blastocoelic cavity in a fibrous pattern. Scale bar = 100 μm in A (applies to A,B), 50 μm in C.

Download figure to PowerPoint

Distribution of 4H11 Fibers and Mesenchyme Cells in Morphologically Abnormal Embryos

To view the distribution of the 4H11 fibers and mesenchyme cells in embryos having the balloon-like condensation, late-blastulae were treated with CFSW with or without 4H11 Mab and allowed to develop for 20 to 30 hr (cf. Fig. 4E,H), then processed for double staining with 4H11 Mab and MC5 Mab (see Experimental Procedures section).

In control embryos at 20 hr after treatment, 4H11 fibers were found to be distributed throughout the blastocoelic cavity with the mesenchyme cells scattered therein (Fig. 6A) as in the normal embryos (cf. Fig. 1I). In 4H11 Mab-treated embryos, on the other hand, 4H11 fibers were found to condense into thick bundles lining the ectodermal and endodermal walls in addition to forming the balloon-like condensation around the tip of the archenteron (Fig. 6B). Some 4H11 fibers were stretched between the balloon-like condensation and the anterior ectodermal wall (Fig. 6B,C). Most of the mesenchyme cells were associated with the balloon-like condensation (Fig. 6B,D). Another prominent feature of the 4H11 Mab-treated embryos was the lack of 4H11 fibers in the blastocoelic cavity beneath the bulged posterior portion of the ectodermal wall (Fig. 6B,C).

thumbnail image

Figure 6. Distribution of 4H11 fibers and the mesenchyme cells in 4H11 monoclonal antibody (Mab) -treated embryos 20 hr after the treatment. In the experimental embryos, 4H11 fibers (green) and the mesenchyme cells (red) are seen as merged images after double immunostaining with 4H11 Mab and MC5 Mab. A: A Ca2+-free seawater–treated embryo (control experiment). Thick bundles of 4H11 fibers can be seen dislocated at the posterior region (arrows) and a mesenchyme cell is associated with the bundle (right side of arrow). B,C: A 4H11 Mab-treated embryo. After treatment with 0.5 mg/ml of 4H11 Mab, the embryo is shown in dark (B) and brightfield images (C). The photo in B is a stacked image consisting of 11 optical sections. Other immunofluorescent photos are taken as one optical section. D: A higher magnification of the region around the tip of the archenterons of B. E,F: Two embryos treated with 4H11 Mab at the concentration of 2.5 mg/ml. E,F: Constrictions are formed on the ectodermal walls where part of the balloon-like condensations have become detached (arrowheads). In these experiments, epithelia are also stained red by MC5 Mab. Scale bars = 100 μm in A (applies to A–C,E,F), 20 μm in D.

Download figure to PowerPoint

When a higher concentration of 4H11 Mab (2.5 mg/ml) was applied, the balloon-like condensation remained attached to the ectodermal wall in various degrees (Fig. 6E,F). In these embryos, the ectodermal wall was constricted at the site where the balloon-like condensation had separated from the ectodermal wall to return to the archenteron, leaving a thick bundle of 4H11 fibers between the two epithelia (Fig. 6E,F, arrowheads).

At 30 hr, the distribution of 4H11 fibers of the control embryo was similar to that at 20 hr, except that the 4H11 fibers had increased somewhat in amount (Fig. 7A, cf. Fig. 6A). In the 4H11 Mab-treated embryos, on the other hand, the balloon-like condensation looked considerably deflated (Fig. 7B,C). The deflated condensation was connected to the ectodermal wall by bundles of 4H11 fibers in all directions. It was also found that 4H11 fibers had been newly laid in the space between the archenteron and the posterior portion of the ectodermal wall, which resumed the constricted form of normal embryos (Fig. 7B, cf. Fig. 6B,E,F). The mesenchyme cells were frequently found outside of the deflated balloon-like condensation, including the posterior half of the blastocoelic cavity (Fig. 7B). At a higher magnification, the mesenchyme cells were shown to stretch along thick bundles of the 4H11 fibers (Fig. 7D–F) and/or to associate with 4H11 fibers as if they were pulling on the 4H11 fibers (Fig. 7G–I).

thumbnail image

Figure 7. Distribution of 4H11 fibers and the mesenchyme cells in 4H11 monoclonal antibody (Mab)-treated embryos 30 hr after the treatment. Embryos were treated as in Figure 6. A: A Ca2+-free seawater–treated embryo (control). B,C: A 4H11 Mab-treated embryo. A: One optical section. B: A stacked image consisting of 11 optical sections. D–I: Higher magnifications of a part of two embryos in which the interaction between 4H11 fibers and the mesenchyme cells are seen. D–F: 4H11 fibers are formed into a thicker bundle by a mesenchyme cell. D,G: Differential interference microscopic images. E,H: 4H11 Mab staining (green). F,I: Double staining with 4H11 Mab (green) and MC5 Mab (red). These photos are on the same optical section. G–I: 4H11 fibers look as if they had been bundled and stretched tight by the mesenchyme cells. These photos are shown as one optical section (D–F) or stacked images consisting of six optical sections (G–I). In these experiments, MC5 Mab stained the epithelia red. Scale bars =100 μm in A (applies to A–C), 10 μm in D (applies to D–I).

Download figure to PowerPoint

Contraction of Isolated Whole ECM by Mesenchyme Cells Associated With 4H11 Fibers

To examine whether the mesenchyme cells generate tension in association with 4H11 fibers, two series of experiments were conducted. The first was carried out to elucidate whether the mesenchyme cells can shorten 4H11 fibers, thereby generating a pulling force. To this end, isolated whole ECM was microinjected with approximately 20 mesenchyme cells, estimated by counting the nuclei under primordium iodide (PI) staining (photo not shown, see Experimental Procedures section). The change in the volume of the ECM was measured, assuming the ECM to be a simple column (Fig. 8). The mesenchyme cells caused not only local shrinkage of the ECM (Fig. 8B, arrow) but also some 20% shrinkage of the isolated whole ECM (Fig. 8C,D). The injected mesenchyme cells were packed tightly instead of spreading away from one another as they would when placed as a cluster on a cultural substratum (photo not shown). This phenotype was observed in all four experimental trials (a total of eight individuals). No shrinkage was detected in the ECM incubated for 2 hr or more without injected mesenchyme cells or when the epithelial cells were injected (preliminary observation).

thumbnail image

Figure 8. Effect of microinjection of the mesenchyme cells into the isolated whole extracellular matrix (ECM). A–D: Light photomicrographs were taken at 0 hr (A,C) and 2 hr (B,D) after injection with the mesenchyme cells through the tract of the archenteron. A,B: Local shrinkage of ECM took place near the mass of injected mesenchyme cells (arrow in B); the m in A shows the cluster of the mesenchyme cells. C,D: The ECM shown in A and B is outlined to quantify the change of volume. C,D: When regarded as a simple columnar structure, the ECM was calculated to be approximately 12.1 μm3 and 9.7 μm3, respectively. Scale bar = 100 μm in A (applies to A–D).

Download figure to PowerPoint

The second series of experiments focused on the question of whether the mesenchyme cells actually cause the ECM to shrink. The mesenchyme cells were closely observed during the deflation process of the archenteron complex (see Experimental Procedures section; Kaneko et al.,1990). Figure 9 shows two time points of the process: immediately after (Fig. 9A,B) and 17 min after (Fig. 9C,D) removal of the ectoderm from the late-gastrulae. Quick shrinkage or condensation of the ECM made it impossible to perceive the fibrous nature of the 4H11 fibers (Fig. 9B,D). The relatively smooth contour of the anterior most portion of the ECM has turned rough as though its portions were being pulled on by the nearby mesenchyme cells (Fig. 9E,H). The fluorescence is also discernible as condensing around some mesenchyme cells (Fig. 9E,F,H,I, arrowheads). The mesenchyme cells, on the other hand, developed many spike-like projections during this short period (Fig. 9G,J).

thumbnail image

Figure 9. Deflation of the archenteron complexes. A–D: The archenteron complexes were fixed with paraformaldehyde in seawater, immediately (A,B) and 17 min (C,D) after removal of the ectodermal cells, processed for double-immunostaining with MC5 monoclonal antibody (Mab) and 4H11 Mab, and photographed in the same optical section at the same intensity of the scanning microscopy (see Experimental Procedures section). Deflation of the extracellular matrix (ECM) took place concomitantly with condensation of 4H11 fibers. E,F,H,I: Higher magnifications of the anterior portions of (A,B) and (C,D) are shown in (E,F) and (H,I), respectively. Condensation of 4H11 fibers took place around the mesenchyme cells (red) during the process of deflation. The intensity of the laser applied for (I) was less than (F) to show the local condensation of 4H11 fibers. G,J: Typical morphology of the mesenchyme cells is shown. E–J: Numerals indicate individual mesenchyme cells. J: Spike-like filopodia are shown. Scale bars = 100 μm in A (applies to A–D), 20 μm in E (applies to E,F,H,I), 14 μm in G (applies to G,J).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We have analyzed the effect of a Mab specifically recognizing a fibrous component of blastocoelic ECM, which we designated as 4H11 fibers, in the development of the starfish A. pectinifera. The distribution of 4H11 fibers under both normal and experimental conditions suggests their roles in embryonic morphogenesis in relation to the mesenchyme cells.

4H11 Mab Recognizes a Single Kind of Proteinaceous Molecule

Immunobiochemical analyses reveal that 4H11 fibers consist of a 370-kDa proteinaceous molecule containing at least 26 kDa of the N-linked oligosaccharide(s) (Fig. 3B). We believe 4H11 Mab recognizes a single kind of molecule because a single band is detected in the epithelial cells, which are responsible for synthesizing the molecule (Fig. 3C). It is not evident, at present, whether bands heavier than the 370-kDa one are due to an artifact or to the failure by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) to separate some molecule(s) bound to the 4H11 fibers (Fig. 3A). The latter is probably the case, because 4H11 molecules are organized into fibers of different thickness when present in the blastocoele (Fig. 2). It is paradoxical that 4H11 Mab was obtained from the cultured population of the mesenchyme cells, which do not synthesize the antigen (Fig. 3C). We consider that the epithelial cells synthesized the antigen before they died to leave a population of only mesenchyme cells (Kaneko et al.,1995a). Further identification of 4H11 molecules is now in progress.

Distribution of 4H11 Fibers in Normal Development

4H11 fibers appeared at the early-blastula stage (Fig. 1C) and filled the blastocoelic cavity as a thick meshwork by the late-gastrula stage (Fig. 1I). The meshwork was more condensed in places where the shape of the ectoderm departed from a sphere, such as in the posterior half of the mid-gastrula (Fig. 1G), the lateral concavity of the late-gastrula (Fig. 1I, arrow), and the regions at which the two epithelia were held close to one another, such as at the oral hood, the dorsal ectoderm, and the posterior protrusion of the larvae (Fig. 1K, arrows, L). 4H11 fibers were also condensed on or near the basal lamina throughout the developmental stages after the late-blastula stage (Fig. 1E,G,I,K). These features of 4H11 fibers are comparable in the timing of their appearance, the amount and localization in the blastocoele, and the network structure with those of the fibrous ECM reported by Crawford and his colleagues in the starfish P. ochraceus (Crawford and Chia,1982; Abed and Crawford,1986; Crawford,1989,1990; Crawford et al.,1997). We, however, have no evidence indicating that 4H11 fibers are identical to the fibers studied by Crawford and others. Figure 2D shows that the fibrous feature of the 4H11 fibers is not an artifact of fixation as was speculated by Strathmann (1989).

Behavior of 4H11 Fibers and Mesenchyme Cells Under Our Experimental Conditions

Our method to easily introduce macromolecules into the blastocoele of living embryo (Kaneko et al.,1995b) provided us with highly reproducible morphological abnormalities that concern both the 4H11 fibers and mesenchyme cells. The coincidence of the abnormal distribution of 4H11 fibers and the mesenchyme cells in embryos treated with 4H11 Mab (Figs. 4E,F, 6B–D) strongly supports the role of fibrous ECM to provide a scaffold for mesenchyme cell migration, as suggested by Crawford and Chia (1982).

The second and third of the suggestions made by Crawford and others concerning the supporting function of the fibrous ECM and the role of the mesenchyme cells to rearrange or to modify the length of the fibrous ECM, respectively, are easy to address, but hard to verify. We have obtained some experimental evidence in support of these ideas.

To verify that 4H11 fibers support the shape of the body wall by resisting its bulging force, an experiment was designed to interfere with this function. In the experiment, either 4H11 Mab or its Fab fragments was introduced into the blastocoele of living blastulae. Only the Mab affected the distribution of 4H11 fibers by fixing them in position at the time of its introduction (Fig. 4B,D), indicating that the effect is caused by the divalent nature of the Mab. The affected embryos showed an abnormally bulged body wall (Figs. 4E, 6C), from the space beneath which 4H11 fibers were absent (Fig. 6B). This area is the space that is expected to have a higher concentration of the fibers compared with other parts of the body space in normally developing embryos (Fig. 1I, 6A). It is also obvious in Figure 6E,F that the balloon-like condensation can cause concavity on the ectodermal wall (arrowheads). These findings strongly support the resisting function of the 4H11 fibers. This notion is further supported by the presence of thick bundles of 4H11 fibers stretching between the ectodermal wall and various parts of the archenteron in embryos recovering from the bulging effect of the Mab (Fig. 7B).

We have presented several experimental findings suggesting the ability of the mesenchyme cells to maneuver 4H11 fibers. First, the balloon-like condensation begins to deflate only after the mesenchyme cells appear and become attached to it (Fig. 4C–F). Mesenchyme cells are thought to rearrange the 4H11 fibers constituting the balloon-like condensation until it is completely loosened, as inferred from the unnatural distribution of mesenchyme cells in the treated embryos until later stages of development (Fig. 4H vs. J, K vs. M). Translocation of the fluorescence from the balloon-like condensation to the mesenchyme cells indicates the possibility that the mesenchyme cells consume the Mab or Mab-bound 4H11 fibers (Fig. 4I,L).

The ability of the mesenchyme cells to gather 4H11 fibers into thick bundles is demonstrated in the process of recovery of the treated embryos from the bulging effect of the Mab (Fig. 7D–I). If these bundles are actually resisting the bulging force of the ectodermal wall as mentioned earlier, and as suggested by Crawford and others, the mesenchyme cells must be shortening the bundles in one way or another. The ability of the mesenchyme cells to shorten the 4H11 fibers is indirectly shown in Figure 8. The mesenchyme cells can affect the ECM or 4H11 fibers either locally (Fig. 8B, arrow) or generally, as shown by the 20% shrinkage of the whole ECM (Fig. 8C,D). The mesenchyme cells seem to condense the adjacent 4H11 fibers close to themselves by their filopodia to bring about the shortening effect (Fig. 9F,I). The involvement of the filopodia formation in exertion of mechanical tension against the fibrous ECM has been suggested for other cell types (Hodor et al.,2000).

Role of 4H11 Fibers and Mesenchyme Cells in Normal Development

Our results indicate several aspects of the physiological significance of 4H11 fibers in the normal development of the starfish. 4H11 fibers are secreted on or near the basal lamina by the epithelial cells during the early stages of development, i.e., early-blastula to early-gastrula (Fig. 1E–H). The fibers spread out autonomously into the blastocoelic cavity or between two epithelia (Fig. 1G). One of the functions of the fiber is considered to provide a scaffold not just for mesenchyme cells but also for the epithelial cells. This function is conjectured from two pieces of evidence. One is the fact that 4H11 fibers are always more condensed on or near the basal lamina compared with the adjacent blastocoelic space throughout the developmental stages examined (Figs. 1, 6A, 7A). This peripheral condensation of 4H11 fibers is considered to be capable of providing some mechanical and molecular rigidity to the basal lamina on which the epithelial cells can attach and stabilize themselves. The other is that, when this peripheral condensation of the fibers is immobilized on or near the basal lamina by 4H11 Mab (Fig. 4D,F,I,L), the development of the corresponding epithelia is retarded (Fig. 4E vs. G, H vs. J, and K vs. M). From this finding and the frequent dislocation of the peripheral condensation from the basal lamina at sites of quick expansion of the body wall epithelium (Figs. 1G,K, 2B, arrowheads, 6A, arrows), we infer that the epithelial cells need to expand the peripheral condensation when the epithelium expands.

When the time comes to give a specific shape to the embryo, more fibers are secreted by the epithelial cells (Fig. 3C) for the mesenchyme cells to maneuver. The mesenchyme cells extend the fibers across the body wall and the archenteron in a thin meshwork when relatively weak, but a general force is required to counter the bulging tendency intrinsic to the epithelia (Honda et al.,1983) to produce the cylindrical contour of the posterior ectoderm (Fig. 6A). When more extreme shaping is necessary, the mesenchyme cells condense the fibers into thicker bundles to exert stronger resist forces (Figs. 1I,K, 7A). Once the length of the meshwork or the bundles is determined, 4H11 fibers can hold the epithelia in place. The mesenchyme cells are probably also responsible for loosening the dislocated peripheral condensation of the 4H11 fibers (Fig. 6A). To further ascertain the role of 4H11 fibers and mesenchyme cells in morphogenesis, experiments are now underway to quantify the effects of changes in the number of mesenchyme cells on embryonic development.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Embryos

Eggs and embryos of the starfish Asterina pectinifera were obtained as described previously (Dan-Sohkawa,1976). Briefly, mature eggs were collected by treating dissected ovaries with 1-methyladenine (Kanatani,1969). Spawned eggs were inseminated with diluted dry sperm and allowed to develop to various stages in artificial seawater (SW), Jamarin U (Jamarin Lab., Osaka), at 20°C.

Cultured Mesenchyme Cells and Dissociated Epithelial Cells

A population of only mesenchyme cells was prepared in culture from mid-gastrulae as described previously (Kaneko et al.,1995b). The cultured mesenchyme cells were collected by the two methods described below and used for preparation of monoclonal antibodies, Western blotting, and microinjection experiments. The epithelial cells, on the other hand, were prepared as dissociated cells of the mid-gastrulae (Dan-Sohkawa and Kaneko,1989). They were treated with SW containing trypsin (Sigma) at the concentration of 1 mg/ml (trypsin–SW) for 15 min at room temperature, and processed for Western blotting.

Isolation of Whole ECM

Packed mid-gastrulae (1 ml) were washed with 10 ml of 1.2 M glycine solution supplemented with 1% SW and 4% newborn calf serum (M.A. Bioproduct) at room temperature and washed twice with 10 ml of chilled SW. Next, they were washed twice with 10 ml of the chilled SW containing 1% Triton X-100, 2 mM phenylmethyl sulfonyl fluoride (Sigma) together with four tablets of protease inhibitor cocktail (Complete, ethylenediaminetetraacetic acid–free [Roche], ECM isolation–SW). These washing steps were carried out by hand centrifugation. The embryos were then replaced in 100 ml of ECM isolation–SW and stirred gently on ice. After 10 min, they were washed four times with SW by centrifugation at 1,700 × g for 5 min to collect the whole ECM. The same procedure was also applied to late-gastrulae.

Archenteron Complexes

Archenterons accompanied by varying portions of the ECM surrounding them (archenteron complexes) were prepared as described previously (Kaneko et al.,1990). They were incubated at room temperature in SW supplemented with 4% newborn calf serum (culture medium; Kaneko et al.,1990) on a plastic culture dish coated with 0.1% collagen solution (Koken, Tokyo) to monitor the deflation process of the ECM.

Preparation of 4H11 Mab and MC5 Mab

The cultured mesenchyme cells were prepared from 5 ml of packed mid-gastrulae. They were scraped off from the substratum using a rubber policeman, centrifuged at 400 × g for 10 min and stocked as a pellet at −80°C. A male BALB/c mouse was injected intraperitoneally with one third of the stocked material suspended in phosphate-buffered saline (PBS), which was mixed with the same volume of Freund's complete adjuvant (Gibco BRL). Second and third boosts were conducted similarly 3 and 5 weeks after the first injection, respectively, but antigens were emulsified with Freund's incomplete adjuvant (Gibco BRL). The mouse was killed 3 days after the last boost. The spleen was dissected, and dissociated spleen cells were fused to SP-1 myeloma cells according to the method described by Galfre and Milstein (1981). Hybridomas were grown in Iscove's modified Dulbecco's medium (Hazleton Biologicals) supplemented with 10% fetal calf serum (Microbial Associates). Hybridoma supernatants recognizing blastocoelic ECM were screened by indirect immunofluorescent staining of embryos sectioned as described below. One colony staining a fibrous component of the blastocoelic ECM was selected by picking up single hybridoma cells under a light microscope and designated as 4H11 Mab.

To obtain a monoclonal antibody against the mesenchyme cells, the same procedure was followed, except that the mesenchyme cells were detached from the substratum using trypsin–SW. The obtained monoclonal antibody was designated as MC5 Mab; it has a strong ability to recognize mesenchyme cells while also responding weakly to epithelial cells. A subclass of 4H11 Mab and MC5 Mab was identified as IgG2b and IgG1, respectively, using a Monoclonal Typing Kit (Birmingham Res., Birmingham, AL).

Determination of Protein Concentration

The protein concentration of antibody solutions was determined by protein–dye binding assay with bovine serum albumin as the standard (Bradford,1976).

Purification of 4H11 Mab and MC5 Mab

4H11 hybridoma cells were intraperitoneally injected at 2.5 × 106 cells per BALB/c male mouse immunologically evoked 1 week previously with 0.5 ml of 2, 6, 10, 14-tetramethylpentadecane (Wako Chemicals, Osaka, Japan). Ascites fluid was collected from the mouse on the 10th day by centrifuging at 400 × g for 10 min, and the supernatant was pooled. IgG was purified from the supernatant by affinity chromatography on protein A–Sepharose (Pharmacia, Uppsala) according to the manufacturer's instructions and was named purified 4H11 Mab. To introduce it into the blastocoele of living embryos, the concentration was adjusted to 0.5 mg/ml and in some cases to 2.5 mg/ml (Fig. 6E,F) and dialyzed against CFSW (Jamarin Lab., Osaka, Japan).

Fluorescent Labeling of 4H11 Mab

4H11 Mab was fluorescently labeled in two ways. First, the purified 4H11 Mab was labeled with fluorescein isothiocyanate (FITC, Sigma), after Clausen (1969). Briefly, 5 mg of the purified 4H11 Mab was dialyzed against 0.15 M sodium chloride solution containing 10% (v/v) carbonate buffer pH 9.5. The dialysate was concentrated to 75 mg/ml and incubated at the ratio of 20/1 with the same dialysis solution containing 15 mg/ml FITC for 1 hr at room temperature. Free FITC was removed from the protein solution by chromatography on a Sephadex G-25 column (Pharmacia, Uppsala). The eluate was adjusted to 0.5 mg/ml and dialyzed against CFSW and is called FITC–4H11 Mab. Second, the purified 4H11 Mab was labeled with IgG2b labeling kit (Zeon Alexa Fluor 488 Mouse IgG2b Labeling kit, Molecular Probes) according to the manufacturer's instructions and called “Alexa 488–4H11 Mab.”

Preparation of Fab Fragments of 4H11 Mab

Fab fragments of the IgG were purified from the ascites fluid as described (Harlow and Lane,1988). In brief, purified 4H11 Mab was dissolved in PBS containing 10 mM cysteine (Sigma) and 2 mM ethylenediaminetetraacetic acid (Wako Chemicals, Osaka, Japan) at the concentration of 30 mg/ml. It was digested with papain (25 units/mg protein, Sigma) at an enzyme/substrate ratio by weight of 1/100 at 37°C for 16 hr, after which N-ethylmaleimide (Sigma) was added at a final concentration of 10 mM. After 15 min at 37°C, the digest was loaded onto a Protein-A cellulofine column (Seikagaku Kogyo, Tokyo, Japan), which had been equilibrated with PBS. Fab fragments were concentrated by precipitation with 80% ammonium sulfate, dissolved in CFSW, adjusted to 0.6 mg/ml, and dialyzed against CFSW. They are called “Fab fragments of 4H11 Mab” in this study.

SDS-PAGE and Immunoblotting

Packed experimental materials, such as isolated whole ECM, whole mid-gastrulae, the trypsin-treated epithelial cells or the trypsin-treated mesenchyme cells, were solubilized by adding an equal volume of 4% SDS in 0.2 M Tris-HCl (pH 6.8), mixed vigorously, and boiled for 3 min. Next, the amount of total protein of each sample was determined using BCA protein assay kit (Pierce). Samples for SDS-PAGE were prepared by adding glycerol and 2-mercaptoethanol (2-ME) at the final concentrations of 10% and 5%, respectively. SDS-PAGE was carried out using a ready-made 3–10% gradient gel (Atto Co.). Prestained SDS-PAGE standards (Bio-Rad) supplemented with 1 mg/ml of laminin (Sigma) were used for calculation of the molecular weights of the samples. Proteins were transferred to polyvinylidene difluoride (PVDF) membrane. All the immunostaining steps were carried out in Tris (hydroxymethyl) aminomethane-buffered saline (TBST; 150 mM NaCl, 0.05% Tween 20, 50 mM Tris-HCl pH 7.5) containing 5% skim milk. After blocking, the PVDF membrane was treated with the purified 4H11 Mab (0.2 μg/ml) for 1 hr at room temperature followed by several washing steps and then treated with horseradish peroxidase–conjugated goat anti-mouse IgG antibody (2,000-fold dilution; Pierce). After several washings without skim milk, chemifluorescent detection was performed using ECL Western blotting detection system (ECL; Amersham Biosci.) according to the manufacturer's instructions.

Enzymatic Digestion

Isolated whole ECM of mid-gastrulae was solubilized by 2% SDS in 0.1 M Tris-HCl (pH 6.8). They were diluted with 4% 2-ME to adjust the final concentrations of Tris-HCl and SDS to 25 mM and 0.5%, respectively. The samples were boiled for 10 min. Next, NP-40 and phosphate buffer (pH 7.0) were added according to the manufacturer's instructions. The samples were then incubated with trypsin (1 mg/ml) or PNGase F (500,000 U/ml; New England Biolabs., Inc.) for 3 hr at 37°C. The reactions were stopped by adding an equal volume of double concentrated sample buffer (4% SDS, 0.2 M Tris-HCl pH 6.8, 10% glycerol, 5% 2-ME) and boiling for 3 min. The procedures for SDS-PAGE and immunoblotting were as described above.

Introduction of Antibodies or Fab Fragments Into the Blastocoele of Living Embryos (CFSW Method)

Embryos were treated at the late-blastula stage with CFSW containing either the purified 4H11 Mab or the Fab fragments of 4H11 Mab according to a method previously (Kaneko et al.,1995b) with a minor modification in the operating temperature. Briefly, 0.1 to 0.5 ml of packed embryos was washed once with 10 ml of CFSW at 0°C. They were subsequently incubated at 0°C for 15 min with 5 volumes of CFSW containing the antibody or the Fab fragments. This treatment was stopped by dilution with 10 ml of SW at room temperature. The embryos were washed twice with 10 ml of SW and allowed to develop in SW at 20°C. As controls, plain CFSW containing 0.5 mg/ml of rabbit anti-mouse FITC–IgG (Cappel, Westchester) was used for this treatment. The treated embryos are referred to as 4H11 Mab-treated embryos, FITC-4H11 Mab–treated embryos, 4H11 Fab-treated embryos, CFSW-treated embryos, or FITC-IgG–treated embryos according to the treatment.

Indirect Immunofluorescence Microscopy

To prepare sections of normal embryos, embryos of different developmental stages ranging from the unfertilized egg to the bipinnaria stage (Fig. 1) were fixed in 4% PFA–SW, dehydrated in an ethanol series, and embedded in polyester wax (BDH, London; Kusakabe et al.,1984). They were sectioned at 8 μm and applied to poly-lysine–coated slide glasses (Ettensohn and McClay,1988). After dewaxing in absolute ethanol, the sections were rehydrated in PBS and incubated with the conditioned medium of 4H11 hybridoma culture (4H11 CM) for 15 min at room temperature in a moist chamber. They were washed three times with PBS and incubated further for 20 min with the FITC–IgG diluted to 1/50 in PBS. The sections were washed with PBS and mounted with 50% glycerol in PBS. They were examined under a fluorescence microscope coupled with phase contrast or light microscope equipment (Olympus BH-2). Photomicrographs were taken with Kodak Tri-X film. As controls for the normal embryos, sections were incubated with the CM of SP-1 myeloma culture (SP-1 CM) instead of 4H11 CM.

To stain 4H11 Fab-treated embryos (Fig. 5) and their control counterparts, the embryos were fixed in ethanol at −20°C for 30 min, washed twice with PBS, incubated with rabbit anti-mouse FITC–Fab (Sigma) for 20 min at room temperature and washed three times with PBS. After pressing lightly between a slide glass and a coverslip, the embryos were examined for immunolocalization of the antigens as described above.

Double staining of 4H11 Mab-treated embryos and the control CFSW-treated embryos (Figs. 6, 7) was carried out as follows. The embryos were fixed in PFA–SW and then post-fixed in ethanol. In the first indirect immunostaining, the fixed embryos were treated with 4H11 CM and Alexa 488-conjugated anti mouse goat antibody (2,000-fold; Molecular Probes) as the primary and secondary antibody, respectively. They were then subjected to secondary indirect immunostaining, in which the CM of MC5 hybridoma culture (MC5 CM) and Texas Red-conjugated goat anti-mouse IgG1 γ1 chain-specific (2,000-fold; Southern Biotech Associates, Inc., Birmingham, AL) were chosen as the primary and secondary antibody, respectively. All of the embryo incubation procedures were performed at room temperature for 20 min. In the case of the archenteron complexes (Fig. 9), they were fixed with PFA–SW alone. The first indirect immunostaining was performed by applying MC5 CM and Texas Red-conjugated goat anti-mouse IgG1 (2,000-fold) as the primary and secondary antibody, respectively. Next, they were immunostained with the Alexa 488-labeled 4H11 Mab. The timing, duration, and frequency of washing were carried out using PBS in the same manner as immunostaining of the sectioned embryo. Immunostained embryos and archenteron complexes were examined with an Olympus laser scanning confocal microscope. The images were processed using Adobe PhotoShop.

Direct Immunofluorescence Microscopy

Packed mid-gastrulae (0.1 ml) were suspended in 0.5 ml of SW and crushed to small pieces by pipetting five times through a narrow needle (26 gauge × 0.5 inch, 0.45 × 13 mm, Terumo, Tokyo). The pieces were incubated at room temperature with 20 μg/μl of FITC-conjugated 4H11 Mab in SW for 10 min and then washed three times with SW. They were immobilized between a slide glass and a coverslip and viewed under an epifluorescence microscope (Olympus BH-2).

Isolated whole ECM was fixed with PFA–SW alone. After washing with PBS, it was immunostained with 0.3 μg/μl of the Alexa 488–4H11 Mab and visualized under an Olympus laser scanning confocal microscope as described above.

Living FITC-4H11 Mab–treated and control CFSW-treated embryos were confined to a narrow space between the slide glass and the coverslip and fixed at intervals with 4% PFA–SW gently applied from the side of the coverslip (Kaneko et al.,1990). Immediately after the embryos stopped swimming, they were examined and photographed (Olympus BH-2).

Microinjection of Mesenchyme Cells Into the Isolated Whole ECM

Isolated whole ECM of the late-gastrula stage was placed with culture medium in a microinjection chamber devised by Keihart (1982). The mesenchyme-cells were collected from the culture dishes by treatment with trypsin-SW for 10 min at room temperature. After washing twice with culture medium, the mesenchyme cells were aspirated into a microneedle and injected into the isolated whole ECM through the tract, which used to be occupied by the archenteron and its lumen. The specimen was photographed immediately and 2 hr after injection under an Olympus laser scanning confocal microscope. Next, the isolated whole ECM was transferred to PFA–SW, fixed for 15 min at room temperature, washed with PBS, and incubated in PBS containing PI (Wako, Osaka, Japan) at the concentration of 0.1 μg/μl for 15 min at room temperature to stain the nuclei of the injected mesenchyme cells. After washing with PBS, the nuclei of mesenchyme cells were counted using the optical slicing function of an Olympus laser scanning confocal microscope.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank the members of the Asamushi Marine Biological Station of Tohoku University for supplying the starfish. Thanks are also due to Mr. Noriya Miyata for technical assistance with the photography. We also thank Ms. Yuko Takahashi and Dr. Yukiko Sato for the helpful arrangements for the photography.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Abed M, Crawford BJ. 1986. Changes in structure and location of extracellular matrix of the blastocoel during development of the starfish Pisaster ochraceus. Can J Zool 64: 14361443.
  • Bradford MM. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248254.
  • Clausen J. 1969. Immunochemical techniques for the identification and estimation of macromolecules. In: WorkTS, WorkE, editors. Laboratory techniques in biochemistry and molecular biology. 1st ed. Amsterdam: North-Holland Publishers. p 539.
  • Crawford BJ. 1989. Ultrastructure of the basal lamina and its relationship to extracellular matrix of embryos of the starfish Pisaster ochraceus as revealed by anionic dyes. J Morphol 199: 349361.
  • Crawford BJ. 1990. Changes in the arrangement of the extracellular matrix, larval shape, and mesenchyme cell migration during asteroid larval development. J Morphol 206: 147161.
  • Crawford BJ, Chia FS. 1982. Genesis and movement of mesenchyme cells in embryos of the starfish Pisaster ochraceus. In: LawrenceJM, editor. International Conference, Tampa Bay. Rotterdam: AA Balkema. p 505511.
  • Crawford BJ, Campbell SS, Reimer CL. 1997. Ultrastructure and synthesis of the extracellular matrix of Pisaster ochraceus embryos preserved by freeze substitution. J Morphol 232: 133153.
  • Dan-Sohkawa M. 1976. A “normal” development of denuded eggs of the starfish Asterina pectinifera. Dev Growth Differ 18: 439445.
  • Dan-Sohkawa M, Kaneko H. 1989. Sorting out of presumptive stomach cells of the starfish embryo. Dev Growth Differ 31: 503508.
  • Ettensohn CA, McClay DR. 1988. Cell lineage conversion in the sea urchin embryo. Dev Biol 125: 396409.
  • Galfre G, Milstein C. 1981. Preparation of monoclonal antibodies: strategies and procedures. Methods Enzymol 73: 346.
  • Harlow E, Lane D. 1988. Antibodies: a laboratory manual. Cold Spring Harbor: Cold Spring Harbor Laboratory Press. 726 p.
  • Hodor PG, Illies MR, Broadley S, Ettensohn CA. 2000. Cell substrate interactions during sea urchin gastrulation: migration primary mesenchyme cells interact with and align extracellular matrix fibers that contain ECM3, a molecule with NG2-like and multiple calcium-binding domains. Dev Biol 222: 181194.
  • Honda H, Dan-Sohkawa M, Watanabe K. 1983. Geometrical analysis of cells becoming organized into a tensile sheet, the blastular wall, in the starfish. Differentiation 25: 1622.
  • Kanatani H. 1969. Induction of spawning and oocyte maturation by 1-methyladenine in starfishes. Exp Cell Res 57: 333337.
  • Kaneko H, Takaichi S, Yamamoto M, Dan-Sohkawa M. 1990. Acellularity of starfish embryonic mesenchyme cells as shown in vitro. Development 109: 129138.
  • Kaneko H, Kawahara M, Dan-Sohkawa M. 1995a. Primary culture of mesodermal and endodermal cells of the starfish embryo. Zool Sci 12: 551558.
  • Kaneko H, Okushima Y, Iizuka M, Dan-Sohkawa M. 1995b. A simple method for introducing macromolecules into the blastocoel of living starfish embryos. Zool Sci 12: 559564.
  • Keihart OP. 1982. Microinjection of echinoderm eggs: apparatus and procedures. Methods Cell Biol 25: 1331.
  • Kusakabe M, Sakakura T, Nishizuka Y, Sano M, Matsukage A. 1984. Polyester wax embedding and sectioning technique for immunohistochemistry. Stain Tech 59: 127132.
  • Strathmann RR. 1989. Existence and functions of a gel filled primary body cavity in development of echinoderms and hemichordates. Biol Bull 176: 2531.