Cell-lineage and clonal-contribution map of the trochophore larva of Patella vulgata (Mollusca)1

Mechanisms of Development 62 (1997) 213–226

Cell-lineage and clonal-contribution map of the trochophore larva of Patella vulgata (Mollusca)1 Wim J.A.G. Dictus*, Peter Damen Department of Experimental Zoology, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands Received 21 November 1996; revised version received 22 January 1997; accepted 22 January 1997

Abstract Molluscan development is characterised by its extremely regular cleavage pattern. In numerous molluscs the fate of various earlycleavage stage blastomeres has been determined and fate maps have been constructed. On the basis of similarities between these fate maps, a generalised molluscan cell-lineage map has been constructed. Recently, the validity of this map has been challenged. In this study, the cell-lineage of the first-, second-, and third-quartet micromeres and third-generation macromeres of the equally-cleaving gastropod mollusc Patella vulgata was studied by fluorescent cell-lineage tracer injection followed by epifluorescence microscopy and confocal laser scanning microscopy. For the first time, a complete cell-lineage map, in the form of a clonal-contribution map of the trochophore, has been constructed with the use of fluorescent cell-lineage tracers. This map both agrees and differs in a number of respects with the generalised cell-lineage map of molluscs. The most important deviation is that the micromere 2d, formerly referred to as the first somatoblast, is not the only cell that forms the foot and shell gland in Patella.  1997 Elsevier Science Ireland Ltd. Keywords: Spiral cleavage; Molluscan development; Embryo; Cell-lineage; Trochophore larva; Fate map

1. Introduction Embryos of molluscs, annelids and some other phyla have a spiral cleavage. This type of cleavage is characterised by its extremely regular pattern. Each blastomere of an early cleavage-stage embryo can be identified according to its position. In all individuals of a given species, blastomeres of corresponding positions will develop alike (Raven, 1958; Verdonk and Van den Biggelaar, 1983). Initially, it was believed that in spirally-cleaving species the fate of early cleavage-stage blastomeres was entirely autonomously specified (cf. Wilson, 1904; Costello, 1945). Therefore, spiralians were thought to have a so-called mosaic development. However, later studies have shown that inductive events do take place (Clement, 1962; Van den Biggelaar and Guerrier, 1979; Arnolds et al., 1983; Martindale et al., 1985; Martindale, 1986; Boring, 1989; Freeman and Lundelius, 1992; Damen and Dictus, 1993, 1996a,b). * Corresponding author. Tel.: +31 30 2533475; fax: +31 30 2532837. 1 Both authors contributed equally to this work.

On the basis of similarities between cell-lineage maps of a number of molluscan species, a cell-lineage map or clonal contribution map has been made for molluscs in general (for reviews, see Raven, 1958; Verdonk and Van den Biggelaar, 1983; Dohmen, 1992). According to this generalised cell-lineage map, the first-quartet micromeres give rise to the apical tuft (when present), the head region (except for the stomodaeum) and part of the prototroch (Figs. 1 and 2). The second-quartet micromeres give rise to part of the prototroch and most of the posttrochal ectoderm. Specifically, the 2d-micromere gives rise to most of the posttrochal ectoderm. Therefore, this cell is called the first somatoblast (Wistinghausen, 1891). The first somatoblast is the only cell thought to produce the foot and the shell gland. The third-quartet micromeres give rise to a small part of the posttrochal ectoderm. In gastropods, the third quartet is also involved in the formation of the ectomesoderm. The fourth-quartet micromeres and fourth-generation macromeres give rise to the endoderm. One particular micromere, the second somatoblast 4d, produces not only endoderm, but also the endomesoderm. The endomesoderm is arranged into two mesoderm bands.

0925-4773/97/$17.00  1997 Elsevier Science Ireland Ltd. All rights reserved PII S0925-4773 (97 )0 0666-7

214

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

of the archaeogastropod mollusc Patella vulgata, we constructed a complete cell-lineage map of the three micromere quartets and the macromere quartet. To this aim, the fluorescent cell-lineage tracer Lucifer Yellow-dextran was microinjected in first-, second- and third-quartet micromeres, and in third-generation macromeres. The fate of the injected cells was determined in trochophore larvae (about 24–28 h after first cleavage) with video-enhanced epifluorescence microscopy and confocal laser scanning microscopy (CLSM). The results allowed the construction of a complete clonal-contribution map of the trochophore larva. This map shows that for Patella the generalised molluscan cell-lineage map is not correct. For example, it appears that the first somatoblast 2d is not the only cell that contributes to the foot and the shell gland. These data unequivocally demonstrate that the previously challenged idea that 2d is the only cell that contributes to the foot and shell gland is incorrect.

Fig. 1. Generalised cell-lineage map of molluscs based on data from the literature (discussed in Raven, 1958; Verdonk and Van den Biggelaar, 1983; Dohmen, 1992). The structures of the trochophore larva that are formed from the first-, second- and third-quartet micromeres and the third-generation macromeres are indicated. The fates of the blastomeres of all four quadrants have been indicated in one lineage branch and differences between cells of a tier have been specified. (m, micromeres; M, macromeres. ‘m’ and ‘M’ refer to all four quadrants, i.e. all four cells of the tier in question.)

Several papers challenged the validity of the generalised molluscan cell-lineage map. Particularly, the exclusive role of the 2d-micromere in giving rise to the entire foot and the shell gland is unsubstantiated (Verdonk and Van den Biggelaar, 1983). Deletion experiments have shown that in Ilyanassa and Bithynia formation of the shell and foot is not entirely dependent on the presence of the 2dmicromere. The presence of other micromeres is essential as well (Cather, 1967; Clement, 1971; Cather et al., 1976; Verdonk and Cather, 1983). Also, in Dentalium, cell-lineage studies have shown that the foot and the shell gland are not exclusively formed by the 2d-micromere (Van Dongen and Geilenkirchen, 1974). None of the original cell-lineage maps upon which the generalised molluscan cell-lineage map is based were determined by means of the injection of cell-lineage tracers. All these maps resulted from comparing the positions of specific cells in successive cleavage stages. The disadvantage of studying cell-lineage without tracers is that it is difficult to establish with certainty the progeny of specific cells. Especially when cells migrate, e.g. during gastrulation or during patterning of the prototroch (Damen and Dictus, 1994b), or when the cell-lineage of prospective adult structures with many small cells is studied, this is very hard. Therefore, accurate cell-lineage maps derived from studies with cell-lineage tracers are urgently needed. As a necessary prelude to further studies on the specification mechanisms that operate during early development

2. Results 2.1. Early development and the formation of the mesentoblast-mother cell The early development of Patella vulgata and the formation of the stem cell of the mesoderm, the mesentoblastmother cell (3D-macromere) have been described by Van den Biggelaar (1977). Fig. 2A–D shows the formation of the first-, second- and third-quartet micromeres at the third, fourth and fifth cleavage, respectively. Between the fifth and the end of the sixth cleavage, the mesentoblast mother cell 3D is induced and thereby the embryo has acquired a dorsoventral polarity. From this moment onwards, the quadrants can be denominated A, B, C and D (Van den Biggelaar and Guerrier, 1979; Arnolds et al., 1983; Damen and Dictus, 1996a; Damen and Dictus, 1996b). Cells injected before this moment were denominated retrospectively, i.e. after injection embryos were allowed to develop until dorso-ventral symmetry was visible. Then the quadrant in which the injected cell was located was determined. 2.2. Composition of the trochophore larva Twenty-four to 28 h old trochophores (n = 360) were studied with light microscopy and SEM. These trochophores possessed a prototroch that consisted of ciliated and deciliated cells (Fig. 2E). These cells were arranged in complete and incomplete rings that divided the trochophore into a pre- and a posttrochal area. The region anterior to the prototroch, the so-called pretrochal area, gives rise to the head (cf. Raven, 1958; Van den Biggelaar and Guerrier, 1983). The region posterior to the prototroch, the so-called postrochal area, gives rise to the rest of the adult body. At the trochophore stage, this area includes the foot,

215

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

Fig. 2. Schematic drawings of several stages in the development of Patella. (A) Animal view of eight-cell stage embryo. (B) Lateral view of 16-cell stage embryo. (C) Animal view of 32-cell stage embryo. (D) Lateral view of 32-cell stage embryo. Thick lines demarcate the quadrants. The first-, second- and third-quartet micromeres, and the third-generation macromeres are indicated by different shading. The quadrants have been denominated arbitrarily, since, before induction of 3D (32-cell stage), the quadrants cannot be denominated (Van den Biggelaar and Guerrier, 1979). (E) Right-lateral view of a 28 h old trochophore larva in which various structures are denominated. The localisation of the stomodaeum, that runs medially from its opening to the outside, towards the anterior pole, is indicated in grey. The prototroch (prot.) consists of ciliated main prototroch cells and unciliated anterior and posterior supporting cells (Damen and Dictus, 1994a; Damen and Dictus, 1994b). For clarity, cilia were only drawn on the outer cells of the prototroch. pretr., pretrochal; posttr., posttrochal.

the shell gland, the mantle and the ciliated larval telotroch. The foot consists of two lobes. The shell gland comprises an annular mantle fold that actually forms the shell, and a central part, the so-called shell field.

Table 1 Contribution of primary, secondary and accessory trochoblasts of the A–D quadrants to the prototroch of Patella A

B

C

D

1m1211 1m1212 1m1221 1m1222 1m211 1m212 1m221 1m222

– – ASc ASc ASc PT PT PT

– – PT PT ASc PT PT PT

– – ASc ASc PT PT PT PSc

ASc PT PT PT PT PT PT PSc

2m111 2m112

PT PSc

PT PT

PT PSc

– –

2.3. Cell-lineage of the first-quartet micromeres At the eight-cell stage, a single first-quartet micromere was injected with Lucifer-Yellow dextran. The contribution of first-quartet micromeres to the prototroch has already been described earlier (Damen and Dictus, 1994a; Damen and Dictus, 1994b). In order to describe the additional, non-prototrochal progeny of the first-quartet micromeres, the prototrochal progeny is summarised below (see also Table 1). Since the pretrochal area, which gives rise to head structures, does not show overt differentiation of adult structures at this stage, the contribution of first-quartet micromeres to this area cannot yet be specified to particular head structures. The analysis was necessarily limited to determining in what part of the pre-

1st quartet Accessory trochoblasts

Primary trochoblasts

2nd quartet Secondary trochoblasts

Initially, all cells involved in prototroch formation become ciliated (Damen and Dictus, 1994a,b). ASc, deciliated anterior supporting cell; PT, ciliated main prototroch cell; PSc, deciliated posterior supporting cell.

216

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

Fig. 3. Fluorescence photographs and schematic drawings showing trochophores in which the progeny of a first-quartet micromere are labelled. The injected cell, as identified retrospectively after the 32-cell stage, is indicated at the right-top of each picture. (A,D,G,J) Fluorescence only. (B,E,H,K) A combination of fluorescence and low-intensity bright-field illumination. (C,F,I,L) Schematic drawings of the position of the labelled cells. (A–C) Injection of 1a. Left-lateral views. Labelling of three main prototroch cells (1a212, 1a222 and 1a221) and three anterior supporting cells (1a211, 1a1222 and 1a1221). A clone of small cells is labelled in the left side of the pretrochal area. (D–F) Injection of 1b. (D,E) Animo-ventral views. (F) Ventral view. Labelling of five main prototroch cells (1b212, 1b222, 1b221, 1b1221 and 1b1222) and one anterior supporting cell (1b211). A clone of small cells is labelled in the ventral side of the pretrochal area. (G–I) Injection of 1c. (G,H) Animo-right-lateral views. (I) Right-lateral view. Labelling of three main prototroch cells (1c212, 1c211 and 1c221), two anterior supporting cells (1c1221 and 1c1222) and one posterior supporting cell (1c222; hardly visible in this orientation). A clone of small cells is labelled in the right side of the pretrochal area. (J–L) Injection of 1d. (J,K) Animo-latero-dorsal views. (L) Dorsal view. Labelling of six main prototroch cells (1d212, 1d211, 1d221, 1d1221, 1d1222 and 1d1212), one anterior supporting cell (1d1211) and one posterior supporting cell (1d222; not visible in this orientation). In the pretrochal area a clone of small cells, anterior to 1d1211 (in the dorsal part of the trochophore), is labelled. In addition, a very small clone, clockwise to the former, is labelled. m.f., mantle fold; pretr., pretrochal area; prot., prototroch. Scale bars, 50 mm.

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

trochal area the progeny of an injected cell was located. This was done by establishing the position of a labelled clone of cells in relation to the dorsoventral axis of the embryo. Since during manipulation of the embryos the apical organ was often lost, the contribution of first-quartet micromeres to the apical organ was not determined. 2.3.1. Micromere 1a After labelling the 1a-micromere (n = 12), the resulting fluorescence pattern comprised six prototroch cells (Table 1) and many small cells of the pretrochal area (Fig. 3A–C). Three of the prototroch cells were large, ciliated, larval cells, so-called main prototroch cells (1a212, 1a221 and 1a222). The other three cells were so-called supporting cells, large, deciliated, larval cells that are associated with the ring of main prototroch cells. The small cells were ectoderm cells located in the left side of the pretrochal area, dorsally to the three main prototroch cells (Fig. 4A–D). 2.3.2. Micromere 1b The progeny of 1b (n = 18) consisted of six large cells

217

that formed part of the prototroch (Table 1) and many small ectodermal cells in the pretrochal area (Fig. 3D– F). Three of the large cell were incorporated in the ring of main prototroch cells (1b212, 1b221 and 1b222). The small cells were located at the ventral side of the pretrochal area, anterior and slightly anticlockwise to the three main prototroch cells (Fig. 4A–D). 2.3.3. Micromere 1c After injection of 1c (n = 10), fluorescence was observed in six large prototroch cells (Table 1) and in a number of small cells of the pretrochal area (Fig. 3G–I). Three of the large cells were incorporated in the ring of main prototroch cells (1c211, 1c212 and 1c221). The small cells formed part of the ectoderm of the pretrochal area and were located anterior to and somewhat ventrally of the three main prototroch cells, in the right side of the pretrochal area (Fig. 4A–D). 2.3.4. Micromere 1d In contrast to 1a, 1b and 1c, 1d (n = 17) gave rise to eight prototroch cells (Table 1). In addition, a number of

Fig. 4. Clonal-contribution map of 24–28 h old Patella trochophore. The contribution of first-, second- and third-quartet micromeres to various structures of the ectoderm is shown in (A–D). The localisation of interior clones derived from individual micromeres and macromeres is shown in (E–H). At the bottom the patterns that are used for indicating the various clones are given. Blastomeres that are not involved in the formation of ectoderm (A–D) or interior structures (E–H) are checked and dimmed. (A,E) Right side. (B,F) Left side. (C,G) Ventral side. (D,H) Dorsal side. In (G,H) a cross-section through the stomodaeum, near the opening to the outside, is drawn (cf. Fig. 2E). Scale bars, 50 mm.

218

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

small cells contributed to the ectoderm of the pretrochal area. In contrast to the progeny of the other first-quartet micromeres, the small cells derived from 1d were arranged in two fluorescent patches: one large patch directly anterior to 1d1211 (prototroch cell derived from the 1d-micromere) and a smaller patch clockwise to the larger patch (Figs. 3J–L and 4A–D). 2.4. Cell-lineage of the second-quartet micromeres At the 16-cell stage, a second-quartet micromere was injected with Lucifer-Yellow dextran. The contribution of second-quartet micromeres to the prototroch has already been described earlier (Damen and Dictus, 1994a,b). In order to describe the additional, non-prototrochal progeny of the second-quartet micromeres, the prototrochal progeny has been summarised in the text and in Table 1. 2.4.1. Micromere 2a After injection of 2a (n = 9), fluorescence was observed in two prototroch cells (Table 1) and in a patch of many small cells that was located in the posttrochal area (Fig. 5A–D). The patch of small cells was found in the ectoderm of the left half of the posttrochal area, i.e. in the left part of the foot, in the left part of the mantle fold and in the left part of the shell field (up to slightly over the mid-dorsal line) (Fig. 4A–D). The fluorescence of the cells of the shell field is rather weak because these cells are very flat and therefore do not possess much fluorescent cell-lineage tracer. Only with the use of the SIT-camera, these cells were visible so that their localisation could be determined (Fig. 5C). 2.4.2. Micromere 2b The progeny of 2b was analysed in 13 embryos, and was found to consist of two prototroch cells (Table 1) and many small cells in the posttrochal area (Fig. 5E–L). In

all embryos the 2b progeny formed a part of the shell gland and a ring of small cells bordering the prototroch at the posttrochal side. The contribution of 2b to the shell gland was located at the right-dorsal side of the trochophore and was connected with the ring of small cells (Fig. 4A–D). This contribution formed part of both the mantle fold and the shell field. Seven out of 13 trochophores in which the 2b-micromere had been injected were recorded on videotape with the SIT-camera. In these trochophores, faint labelling was observed in the pretrochal area (Fig. 5M). This faint labelling appeared to be located underneath the ectoderm. In order to further investigate this labelling, these embryos were processed for CLSM. It appeared that the labelled cells were arranged in three small, more or less parallel strips underneath the ectoderm (Fig. 6A–C). All strips ran more or less from ventral to dorsal. The two lateral strips were located in the ventral half of the pretrochal area. The middle strip was longer than the two lateral ones and, at the posterior side of the prototroch, connected with the ring of small cells that was also formed by the 2b-progeny. This middle strip formed the ventral, upper part or roof of the stomodaeum, that turns upward towards the anterior pole (Fig. 6C; cf. Figs. 2E and 8B). 2.4.3. Micromere 2c After injection of 2c (n = 10), the resulting fluorescence pattern was almost exactly a mirror-image of that of the progeny of the 2a-micromere. Tracer was present in two prototroch cells (Table 1) and in a patch of many small cells that was located in the posttrochal area (Fig. 5N–P). The patch of small cells was part of the ectoderm of the right-lateral half of the posttrochal area, i.e. of the right part of the foot, mantle fold and shell field (Fig. 4A–D). The patch of 2c-derived cells in the shell field did not reach the mid-dorsal line and was smaller than the corresponding patch derived from the 2a-micromere (Fig. 4A–D). At the

Fig. 5. Fluorescence photographs and schematic drawings showing trochophores in which the progeny of a second-quartet micromere are labelled. The injected cell is indicated at the right-top of each picture. (A,C,E,G,I,K,M,N,Q) Fluorescence only. (B,O,S) A combination of fluorescence and lowintensity bright-field illumination. (D,F,H,J,L,P,R,T) Schematic drawings of the position of the labelled cells. (A–D) Injection of 2a. (A,B) Ventro-leftlateral views. (C) Dorso-left-lateral view. (D) Left-lateral view. Labelling of one main prototroch cell (2a111) and one posterior supporting cell (2a112). In the left half of the posttrochal area a clone of cells is labelled. This clone is located in the left part of the foot, the left part of the mantle fold and the left part of the shell field. The edge of the extension of labelling of the shell field is indicated with an asterisk in (C). The cells of the shell-field are very flat and thus cannot contain much fluorescent cell-lineage tracer. Therefore, when these cells are out of focus, their fluorescence is hardly visible. Only some nuclei of shellfield cells, in the middle of the shell field (s.f.), are just visible. (E–F) Injection of 2b. Ventral views. Two main prototroch cells (2b111 and 2b112) and a ring of small cells bordering the prototroch at the posttrochal side are labelled. The position of the stomodaeum, of which the ventral upper side is labelled, is indicated with an arrow. (G–H) Right-lateral views of the same trochophore with the ring of labelled small cells. The labelled part of the shell gland consisted both of thick, heavily fluorescent mantle fold cells and flat, less fluorescent shell field cells. Labelled mantle fold cells bordered the ring of labelled small cells at the dorsal side of the trochophore. (I–J) Dorsal views of the same trochophore. The labelled part of the shell gland, bordering the ring of labelled small cells, is clearly visible. (K–L) Left-lateral views of the same trochophore. In this orientation, the labelled part of the shell gland is only visible at the edge of the trochophore. (M) Animal view after injection of 2b. Internal structures, beneath the ectoderm of the pretrochal area, are labelled. (N–P) Injection of 2c. Right-lateral views. The pattern of labelling is almost mirror-symmetrical to that after labelling of the 2amicromere (cf. A–D). In the prototroch two cells are labelled, one main prototroch cell (2c111) and one posterior supporting cell (2c112). In the right half of the posttrochal area a clone of cells is labelled. This clone is located in the right part of the foot, the right part of the mantle fold and the right part of the shell field. (Q–T) Injection of 2d, the first somatoblast. (Q,R) Left-lateral views. (S,T) Ventral views. A clone of cells is labelled that is located in the middle part of the foot, the telotroch, the ventral part of the mantle fold and a small part of the shell field. m.f., mantle fold; prot., prototroch; stom., stomodaeum; s.f., shell field; tel., telotroch. Scale bars, 50 mm.

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

219

220

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

2.4.4. Micromere 2d, the first somatoblast The cell-lineage of 2d was analysed in 20 trochophores. It appeared that 2d only contributed to posttrochal ectodermal structures: the middle part of the foot, the telotroch, the ventral part of the mantle fold and a small part of the shell field (Figs. 4A–D and 5Q–T). In three embryos a 2d1-micromere was injected at the 32-cell stage (results not shown). In these embryos the left half of the above-described structures was labelled. Apparently, 2d1 and 2d2 contribute equally to the posttrochal ectoderm and produce left (2d1) and right (2d2) mirror images of cellular patterns. 2.5. Cell-lineage of the third-quartet micromeres Third-quartet micromeres were injected with LuciferYellow dextran between the 32- and 52-cell stage.

Fig. 6. Confocal laser scanning microscopical (CLSM) images of a 24 h old trochophore after injection of 2b. (A) Z-series of animal view (steps of 2.6 mm each). Big arrow, localisation of stomodaeum; small arrow, localisation of small, more or less parallel strips of labelling underneath the ectoderm. Scale bar, 100 mm. (B) Projection of the images of (A), in which all images are superimposed on one another. Note that the position of the stomodaeum is now at the bottom of the photograph. Three parallel strips of cells underneath the pretrochal ectoderm are visible (internal). The stomodaeum (stom.) is indicated with an arrow. Scale bar, 50 mm. (C) Median optical section in which the localisation of the middle strip underneath the ectoderm is visible (*). This strip represents the ventral upper side or roof of the stomodaeum (cf. Fig. 2E). The small lumen of the stomodaeum is indicated with an arrow. Scale bar, 50 mm. m.f., mantle fold; pretr., pretrochal area; prot., prototroch; s.f., shell field; tel., telotroch.

right-dorsal side of the trochophore, the part of the shell gland that is formed by the progeny of 2b exactly fits into the part of the shell gland that is formed by the progeny of 2c.

2.5.1. Micromere 3a The 3a-derived cells were found inside the trochophore (n = 12). They formed a rod-shaped area inside the leftventral part of the trochophore. It ran from the pretrochal area, where it touched the pretrochal ectoderm, to the posttrochal area and ended in the left lobe of the foot (Fig. 7A– C). These in vivo observations were confirmed with CLSM (Fig. 7D). In addition, CLSM observation showed that the 3a-progeny consisted of numerous, relatively small cells that completely filled the interior of the left lobe of the foot. 2.5.2. Micromere 3b The progeny of micromere 3b appeared to be a mirror image of the contribution of 3a (n = 10; Fig. 7E–G). Accordingly, the 3b-progeny were located in the rightventral part of the trochophore and, hence, the rod-shaped area ended in the right lobe of the foot. CLSM observations confirmed the in vivo analysis and showed, furthermore, that the 3b-progeny, like the 3a-progeny, consisted of numerous, relatively small cells that completely filled the interior of the right lobe of the foot (Fig. 7H).

Fig. 7. Fluorescence photographs and schematic drawings showing trochophores in which the progeny of a third-quartet micromere are labelled. The injected cell is indicated at the right-top of each picture. (A,E,I,M) Fluorescence only. (B,F,J,N) A combination of fluorescence and low-intensity brightfield illumination. (C,G,K,O) Schematic drawings of the position of the labelled cells. (A–C) Injection of 3a. (A,B) Ventro-left-lateral views. (C) Leftlateral view. Label is located inside the trochophore. (D) CLSM image of another trochophore in which 3a was injected. Projection of nine optical sections (steps of 4.8 mm each) through the left half of the trochophore. Left-lateral view. Note the labelling of a ventrally located area inside the trochophore that is located between the pretrochal area and the left lobe of the foot. (E–G) Injection of 3b. (E,F) Ventro-right-lateral views. (G) Ventral view. Label is located inside the trochophore. (H) CLSM image of another trochophore in which 3b was injected. Projection of nine optical sections (steps of 4.4 mm each) through the ventral half of the trochophore. Ventral view. Note the labelling of a right-laterally located area inside the trochophore that is located between the pretrochal area and the right lobe of the foot. (I–K) Injection of 3c. Ventral views. Labelling of structures inside the trochophore and of the telotroch. (L) CLSM image of a ventral view of another trochophore in which 3c was injected. Note the labelling of the right-bottom part of the stomodaeum (*) and of a cup-shaped ectodermal invagination, laterally to the stomodaeum. This cup-shaped invagination is the anlage of the mantle cavity. Some cells of the telotroch are labelled as well. (M–O) Injection of 3d. (M,N) Ventro-left-lateral views. (O) Ventral view. Labelling of structures inside the trochophore and of a single cell of the telotroch. (P) CLSM image of a ventral view of another trochophore in which 3d was injected. Note the labelling of the left-bottom part of the stomodaeum (*) and of a cup-shaped ectodermal invagination, laterally to the stomodaeum, which is the anlage of the mantle cavity. m.c., mantle cavity; m.f., mantle fold; pretr., pretrochal area; prot., prototroch; s.f., shell field; stom., stomodaeum; tel., telotroch. Scale bars, 50 mm.

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

2.5.3. Micromere 3c After injection of the 3c-micromere (n = 19), the area between the right lobe of the foot, the right-lateral side of the mantle fold and the prototroch was labelled, as well as a rod-shaped cell or a small group of cells that formed part

221

of the telotroch (Fig. 7I–K). Since in a few embryos the telotroch was lost during the deciliation procedure, the labelled cell or cells of the telotroch were lost as well in these embryos. Using the CLSM, the labelled area between the foot, the mantle fold and the prototroch appeared to be

222

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

were part of the telotroch. Sometimes, this label was not present since in a few embryos the telotroch was lost during the deciliation procedure. 2.6. Cell-lineage of the third-generation macromeres (including the mesentoblast-mother cell 3D) Third-generation macromeres were injected with Lucifer-Yellow dextran between the 32- and 60-cell stage.

Fig. 8. (A) SEM image of the ventral side of a 26 h old trochophore larva. Besides the invagination of the stomodaeum (stom.), the invaginations of the anlage of the mantle cavity (m.c.) can be discerned. (B) Leftsagittal optical section (nearly median) of another embryo after injection of 3d. Labelling of the left-bottom part of the stomodaeum (*), that is located between the opening of the stomodaeum to the outside and the anterior pole. The small lumen of the stomodaeum is indicated with an arrow. Scale bars, 50 mm. m.c., mantle cavity; m.f., mantle fold; pretr., pretrochal area; prot., prototroch; s.f., shell field; stom., stomodaeum; tel., telotroch.

a cup-shaped invagination of the ectoderm (Fig. 7L). This invagination is the anlage of the mantle cavity (cf. Fig. 8A). It was connected with a group of 3c-derived cells that formed the right-lateral part and right-dorsal part (i.e. right-bottom) of the stomodaeum (Fig. 7L; cf. Fig. 2E). 2.5.4. Micromere 3d This experiment (n = 28) resulted in a pattern of labelling that was a mirror image of that of the 3c-progeny (Fig. 7M–O). CLSM observations demonstrated that the labelled cells also constituted a cup-shaped invagination of the ectoderm that was connected with a group of 3dderived cells that formed the left-lateral part and left-dorsal part (i.e. left-bottom) of the stomodaeum (Figs. 7P and 8B; cf. Fig. 2E). As in the previous experiment, label was present in a rod-shaped cell or a small group of cells that

2.6.1. Macromere 3A The contribution of 3A consisted of a patch of cells located in the interior left side of the trochophore, dorsally to the mantle fold and adjacent to cells of the shell field (n = 6; Figs. 4E–H and 9A–C). In two of the six trochophores an extra, smaller patch of labelled cells was present in the region of the prototroch adjacent to the inner side of the prototroch. In one of these two trochophores this patch was located near the mid-dorsal line (Fig. 9D). In the other trochophore it was located at the other side, i.e. at the right side of the trochophore. In the CLSM, the localisation of the 3A-progeny was confirmed and the 3A-progeny appeared to consist mainly of relatively large cells (Fig. 9D), indicating that these cells are endodermal (Raven, 1958; Verdonk and Van den Biggelaar, 1983). 2.6.2. Macromere 3B Injection of the 3B-macromere (n = 4) demonstrated that the progeny of this cell were located in the interior of the trochophore, in the region of the prototroch and in the pretrochal area (Figs. 4E–H and 9E–G). In one of these trochophores, the progeny of 3B were located exclusively in the pretrochal area (Fig. 9E,F). In a second trochophore, they were located in the pretrochal area and in the region of the prototroch (Fig. 9H). In a third trochophore, they were located in the right half of both the pretrochal area and the prototroch region. In a fourth trochophore, the left half of both the pretrochal area and the prototroch region were occupied by 3B-progeny. CLSM observations confirmed these observations and

Fig. 9. Fluorescence photographs and schematic drawings showing trochophores in which the progeny of a third-generation macromere are labelled. The injected cell is indicated at the right-top of each picture. (A,E,J,M) Fluorescence only. (B,F,K,N) A combination of fluorescence and low-intensity brightfield illumination. (C,G,L,O) Schematic drawings of the position of the labelled cells. (A–C) Injection of 3A. Left-lateral views. Label is located inside the posttrochal area. (D) CLSM image of another trochophore in which 3A was injected. Projection of nine optical sections (steps of 5 mm each) through the left half of the trochophore. Note the presence of labelling in two regions. The larger region appears to consist of a few large cells. The other, smaller region, seems to be occupied by a single large cell. (E–G) Injection of 3B. Right-lateral views. Label is located inside the pretrochal area. (H) CLSM image of another trochophore in which 3B was injected. Projection of nine optical sections (steps of 6.8 mm each) through the ventral part of the trochophore. Note the presence of labelled, large cells in the pretrochal area and in the region of the prototroch. (I) CLSM image of an animal view of the same trochophore as in (H). Note that the 3B-progeny is located dorsally to the ventrally located stomodaeum. (J–L) Injection of 3C. (J,K) Right-laterodorsal views. (L) Right-lateral view. Label is located inside the posttrochal area. (M–O) Injection of 3D. Ventral views. Label is arranged in a horseshoeshape inside the posttrochal area. (P) CLSM image of a ventral view of another trochophore in which 3D was injected. Note the labelling of elongated, spindle-shaped cells. (Q) CLSM image of a right-lateral view of another trochophore in which 3D was injected. Note that label is located at the region of the right-lateral part of the mantle-fold, between the posterior pole and the prototroch. (R) Nomarski image of the trochophore of (Q). (S) CLSM image of an animal view of the same trochophore as in (Q,R). Note that label is located in two, bilaterally organised regions near the circular prototroch. Small processes (arrows) are present, some of which run to the prototroch. (T) Nomarski image of the trochophore of (S). m.f., mantle fold; pretr., pretrochal area; prot., prototroch; s.f., shell field; stom., stomodaeum; tel., telotroch. Scale bars, 50 mm.

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

223

224

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

furthermore showed that the 3B-progeny consisted mainly of relatively large cells (Fig. 9H), indicating that these cells are endodermal (Raven, 1958; Verdonk and Van den Biggelaar, 1983). The localisation of the stomodaeum in relation to the 3B-progeny is visible in Fig. 9I (cf. Fig. 2E). 2.6.3. Macromere 3C The progeny of 3C was located inside the trochophore, at the right side, dorsally to the mantle fold, adjacent to cells of the shell field (n = 3; Fig. 4E–H and 9J–L). Except for being a little bit smaller, this pattern of labelling was a mirror-image of the pattern observed after labelling the macromere 3A. 2.6.4. Macromere 3D, the mesentoblast mother cell The pattern resulting from labelling 3D (n = 5) consisted of two bands of cells located in the posttrochal area of the trochophore (Fig. 4E–H and 9M– O). These two bands ran from the posterior pole to the region of the prototroch. In four of the five trochophores these bands were arranged in a horseshoeshape and touched each other at the posterior part of the trochophore. This labelling-pattern was confirmed with CLSM (Fig. 9P–T). In optical sections, the progeny of the 3D-macromere appeared to consist of elongated, spindle-shaped cells. These spindle-shaped cells might correspond to muscle cells that are involved in torsion (Smith, 1935; Crofts, 1938, 1955). In two embryos, small fluorescent processes were observed, some of which connected the anterior part of the bands with cells of the prototroch (Fig. 9S,T). In Patella and Haliotis, similar processes have been described which connect the spindle-shaped muscle cells to the prototroch (Crofts, 1938, 1955).

3. Discussion 3.1. Clonal-contribution map of the Patella trochophore On the basis of the results described in this paper, a complete clonal-contribution map of the Patella trochophore larva was constructed (Fig. 4). To our knowledge, this is the first complete clonal-contribution map of a molluscan species that has been produced with the use of celllineage tracers. The results in this paper demonstrate that embryos of Patella develop in a very rigid way. The variation between blastomere-fate in different embryos was minimal. Small size-variations were observed in the contribution of the 2a-, 2b-, 2c- and 2d-micromeres to the shell-gland. The small variations in size of the progeny of the 3A-, 3B- and 3C-macromeres appeared to be more variable. Finally, some variability was observed in the progeny of the 3cand 3d-micromeres. In some embryos, only one cell of

the telotroch was labelled, whereas in a number of other embryos, a small group of cells was labelled. The experiments in which tracer was injected show the localisation of the progeny of various blastomeres. When the position of a labelled clone in the embryo has changed compared to the position of the injected progenitor cell, it is clear that the progeny of the injected cell has migrated. Although the localisation of most clones corresponded with the localisation of their progenitor cells, this was not the case for the progeny of 2b, 3c, 3d, 3B and 3D. Apparently, the progeny of these cells migrate. To our knowledge, an extensive migration such as found for the 2b-progeny has never been described for any molluscan species. The clonal-contribution map demonstrates that the trochophore is bilaterally-symmetrically organised, since blastomeres that are mirror-symmetrically arranged along the dorsoventral axis develop into similarly organised clones of cells with a similar fate. Only minor deviations from this symmetry were observed. According to Smith (1935) and Crofts (1938, 1955), the muscle cells formed from the 3D-macromere become asymmetrically arranged in the trochophore before the beginning of torsion. Since no signs of any asymmetry, indicating the start of torsion, were observed in the 3D-progeny at 24–28 h after first cleavage, we conclude that this asymmetry does not become visible until after 28 h after first cleavage. Based of the localisation of each clone, i.e. at the exterior or in the interior of the trochophore, it is possible to discriminate between an ectodermal and an ento/mesodermal fate. Furthermore, the micromeres with an ectodermal fate were shown to contribute to the pretrochal area, the prototroch, the telotroch, the foot, the mantle cavity, the mantle fold, or the shell field. Since for Patella no reliable markers are available to discriminate between endoderm and mesoderm, it was not possible to determine which blastomeres contribute to the endoderm, the mesoderm or both. 3.2. Comparison of the clonal-contribution map of the Patella trochophore with the generalised molluscan celllineage map By comparing our results with cell-lineage data from the literature, a number of similarities and differences between the fates of specific blastomeres in Patella and that of corresponding blastomeres in other species were found. In Patella, the first-quartet micromeres form the pretrochal ectoderm and part of the prototroch. This is in accordance with the generalised molluscan clonal-contribution map. The fate of the second-quartet micromeres of Patella strongly deviates from that of their generally assumed prospective fate in molluscs. According to the generalised molluscan cell-lineage map, the 2d-micromere produces most of the posttrochal ectoderm, and is the only cell that produces the foot and the shell gland. Therefore, it

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

is considered to be the first somatoblast. This concept is based on observations in unequally cleaving molluscs and annelids (Wistinghausen, 1891; Wilson, 1892; Lillie, 1895; Conklin, 1897). Our results clearly demonstrate that this is not the case in Patella. Since, in Patella, 2a, 2c and 2d all have a significant contribution to the foot and the shell gland, it can be concluded that Patella has no single first somatoblast. This implies that the generally accepted concept that in molluscs the 2d-micromere is the only first somatoblast is, at least for Patella, incorrect. According to Verdonk and Van den Biggelaar (1983), the ectomesoderm in gastropods derives from third-quartet micromeres only. In Patella, most of the ectomesoderm is derived from third-quartet micromeres. Our data indicate that part of the 2b-progeny is located directly underneath the pretrochal ectoderm and consists of small cells. Since at present no reliable molecular markers are available to discriminate between endodermal, mesodermal and neural tissues, it remains unclear whether these cells will form endodermal, mesodermal or neural tissue. The progenies of the 3a- and 3b-micromeres obtain a position inside the trochophore. Based on their size and their position in the prospective foot, these progenies probably are ectomesoderm that will form the foot musculature. The contribution of the 3c- and 3d-micromeres to the ectoderm forming the mantle cavity is in agreement with the generalised cell-lineage map. Regarding the fate of the third-generation macromeres, the results of this paper are compatible with the generalised molluscan cell-lineage map according to which the 3A-, 3B- and 3C-macromeres produce endoderm and the 3D-macromere produces two mesoderm bands, characteristic for spiralian development. The clonal-contribution map of the Patella embryo opens new possibilities for the study of specification of cells in experimental conditions. Differences such as presented between the Patella clonal-contribution map and the generalised molluscan cell-lineage map might also be demonstrated in the analysis of cell-lineage of other species. Therefore, the generalised molluscan cell-lineage map must be used carefully when one wants to draw conclusions about the contributions of early-cleavage stage blastomeres to structures in the trochophore of any given molluscan species.

4. Experimental procedures 4.1. Embryos Embryos of the common limpet, Patella vulgata (Mollusca, Gastropoda) were obtained as described before (Van den Biggelaar, 1977). Embryos were kept in Milliporefiltered sea water (pore size 1.2 mm; MPFSW) and were dejellied somewhere between the 4- and 32-cell stage by treating them with acidified MPFSW (pH 3.9) for 2–3 min.

225

All experiments were carried out at 19–20°C. All stages of development that are referred to are in hours and minutes after first cleavage at 19–20°C. 4.2. Cell-lineage tracer injections Injections of the cell-lineage tracer Lucifer Yellow-dextran (MW 10 000 Da, D-1825, Molecular Probes, Eugene, USA) were carried out as described before (Damen and Dictus, 1994a,b). 4.3. Observation of trochophores with an epifluorescence microscope The progeny of the injected cell was determined by analysing the pattern of fluorescent cells in trochophores of 24–28 h after first cleavage. Normal trochophores of this stage possess a prototroch, shell, shell gland, foot, and a telotroch. All abnormal trochophores, i.e. those whose gross morphology differed from that of normal trochophores, were discarded. In order to immobilise the rapidly moving trochophore larvae, they were deciliated by transferring them to 0.6 M sodium acetate in MPFSW. Subsequently, the trochophores were studied in this solution with an epifluorescence microscope (Zeiss Axiovert 35M; Oberkochen, Germany) equipped with a Newvicon camera (DAGE-MTI, Michigan City, USA) or a Silicon Intensified Target (SIT) camera (DAGE-MTI). All images were recorded and photographed as described before (Damen and Dictus, 1994b). 4.4. Observation of trochophores with a confocal laser scanning microscope A number of trochophores previously analysed in vivo were processed for confocal laser scanning microscopy. Trochophores were fixed in 4% formaldehyde in 0.1 M phosphate buffer (pH 7.4) for 1 h. After rinsing in buffer, the embryos were transferred to 70% ethanol and stored at 4°C in the dark for several days up to several months. Before observation with the CLSM, trochophores were dehydrated in a graded series of ethanol and transferred into a 1:1 (v/v) mixture of ethanol and Murray’s (1:2 (v/v) mixture of benzylalcohol and benzylbenzoate) for 1 min. Then, the embryos were mounted in Murray’s in convex polished object slides and observed with a Bio-Rad MRC600 confocal laser scanning microscope (Bio-Rad Lasersharp Ltd., Oxfordshire, UK). Images from the CLSM were photographed on Kodak Tmax 100 ASA film with a film recorder (Lasergraphics, Irvine, CA, USA). 4.5. Scanning electron microscopy Trochophores of the appropriate stage were fixed, processed and observed as described before (Damen and Dictus, 1994b).

226

W.J.A.G. Dictus, P. Damen / Mechanisms of Development 62 (1997) 213–226

Acknowledgements The authors wish to thank H.A. Wagemaker for his technical assistance and Prof. Dr. J.A.M. van den Biggelaar and Dr. M.R. Dohmen for critically reading the manuscript. The staff from the aquarium are thanked for taking care of the limpets and the department of image processing is acknowledged for the photographic services rendered. Last, but not least, Mr. W.J. Hage is thanked for his assistance with the CLSM.

References Arnolds, W.J.A., Van den Biggelaar, J.A.M. and Verdonk, N.H. (1983) Spatial aspects of cell interactions involved in the determination of dorsoventral polarity in equally cleaving gastropods and regulative abilities of their embryos, as studied by micromere deletions in Lymnaea and Patella. Roux’s Arch. Dev. Biol. 192, 75–85. Boring, L. (1989) Cell-cell interactions determine the dorsoventral axis in embryos of an equally cleaving opisthobranch mollusc. Dev. Biol. 136, 239–253. Cather, J.N. (1967) Cellular interactions in the regulation of development in annelids and molluscs. Adv. Morphog. 9, 67–125. Cather, J.N., Verdonk, N.H. and Dohmen, M.R. (1976) Role of the vegetal body in the regulation of development in Bithynia tentaculata (Prosobranchia, Gastropoda). Am. Zool. 16, 455–468. Clement, A.C. (1962) Development of Ilyanassa following removal of the D macromere at successive cleavage stages. J. Exp. Zool. 149, 193–216. Clement, A.C. (1971) Ilyanassa. In Reverberi, G. (ed.), Experimental Embryology of Marine and Fresh-water Invertebrates, North-Holland, Amsterdam, pp. 188–214. Conklin, E.G. (1897) The embryology of Crepidula. J. Morphol. 13, 1– 226. Costello, D.P. (1945) Experimental studies of germinal localization in Nereis. J. Exp. Zool. 100, 19–66. Crofts, D.R. (1938) The development of Haliotis tuberculata, with special reference to organogenesis during torsion. Phil. Trans. R. Soc. Lond. B 552, 219–268. Crofts, D.R. (1955) Muscle morphogenesis in primitive gastropods and its relation to torsion. Proc. Zool. Soc. Lond. 125(3,4), 711–750. Damen, P. and Dictus, W.J.A.G. (1993) Role of gap junctions in mesoderm induction in Patella vulgata (Mollusca, Gastropoda): a reinvestigation. In Hall, J.E., Zampighi, G.A. and Davis, R.M. (eds.), Progress in Cell Research, Vol. 3, Gap Junctions, Elsevier, Amsterdam, pp. 219–223. Damen, P. and Dictus, W.J.A.G. (1994a) Cell-lineage analysis of the prototroch of the gastropod mollusc Patella vulgata shows conditional specification of some trochoblasts. Roux’s Arch. Dev. Biol. 203, 187– 198. Damen, P. and Dictus, W.J.A.G. (1994b) Cell-lineage of the Prototroch of Patella vulgata (Gastropoda, Mollusca). Dev. Biol. 162, 364–383. Damen, P. and Dictus, W.J.A.G. (1996a) Spatial and temporal coinci-

dence of induction processes and gap-junctional communication in Patella vulgata (Mollusca, Gastropoda). Roux’s Arch. Dev. Biol. 205, 401–409. Damen, P. and Dictus, W.J.A.G. (1996b) Organiser role of the stem cell of the mesoderm in prototroch patterning in Patella vulgata (Mollusca, Gastropoda). Mech. Dev. 56, 41–60. Dohmen, M.R. (1992) Cell lineage in molluscan development. Microsc. Res. Tech. 22, 75–102. Freeman, G. and Lundelius, J.W. (1992) Evolutionary implications of the mode of D quadrant specification in coelomates with spiral cleavage. J. Evol. Biol. 5, 205–247. Lillie, F.R. (1895) The embryology of the Unionidae. J. Morphol. 10, 1– 100. Martindale, M.Q. (1986) The ’organizing’ role of the D quadrant in an equal-cleaving spiralian, Lymnaea stagnalis as studied by UV laser deletion of macromeres at intervals between third and fourth quartet formation. Int. J. Inv. Reprod. Dev. 9, 229–242. Martindale, M.Q., Doe, C.Q. and Morrill, J.B. (1985) The role of animalvegetal interaction with respect to the determination of dorsoventral polarity in the equal-cleaving spiralian, Lymnaea palustris. Roux’s Arch. Dev. Biol. 194, 281–295. Raven, Chr.P. (1958) Morphogenesis: the Analysis of Molluscan Development, International Series of Monographs on Pure and Applied Biology, Vol. 2 (Harris, J.E. and Yemm, E.W. (eds.)). Pergamon Press, London. Smith, F.G.W. (1935) The development of Patella vulgata. Phil. Trans. R. Soc. Lond. B 225, 95–125. Van den Biggelaar, J.A.M. (1977) Development of dorsoventral polarity and mesentoblast determination in Patella vulgata. J. Morphol. 154, 157–186. Van den Biggelaar, J.A.M. and Guerrier, P. (1979) Dorsoventral polarity and mesentoblast determination as concomitant results of cellular interactions in the mollusk Patella vulgata. Dev. Biol. 68, 462–471. Van den Biggelaar, J.A.M. and Guerrier, P. (1983) Origin of spatial organization. In Verdonk, N.H., Van den Biggelaar, J.A.M. and Tompa, A.S. (eds.), The Mollusca, Development, Vol. 3, Academic Press, New York, pp. 179–213. Van Dongen, C.A.M. and Geilenkirchen, W.L.M. (1974) The development of Dentalium with special reference to the significance of the polar lobe. I, II and III. Division chronology and development of the cell pattern in Dentalium dentale (Scaphopoda). Proc. Kon. Ned. Akad. v. Wet. Ser. C 77, 57–100. Verdonk, N.H. and Cather, J.N. (1983) Morphogenetic determination and differentiation. In Verdonk, N.H., Van den Biggelaar, J.A.M. and Tompa, A.S. (eds.), The Mollusca, Development, Vol. 3, Academic Press, New York, pp. 215–251. Verdonk, N.H. and Van den Biggelaar, J.A.M. (1983) Early development and the formation of the germ layers. In Verdonk, N.H., Van den Biggelaar, J.A.M. and Tompa, A.S. (eds.), The Mollusca, Development, Vol. 3, Academic Press, New York, pp. 91–121. Wistinghausen, C. v (1891) Untersuchungen u¨ber die Entwicklung von Nereis Dumerilii, I. Theil. Mitth. a.d. Zool. Stat. zu Neapel. X, I. Wilson, E.B. (1892) The cell-lineage of Nereis. J. Morphol. 6, 361–481. Wilson, E.B. (1904) Experimental studies in germinal localization. II. Experiments on the cleavage-mosaic in Patella and Dentalium. J. Exp. Zool. 1, 197–268.