archenteron elongation in the sea urchin embryo is a...

DEVELOPMENTAL BIOLOGY 121,253-262 (1987)

Archenteron Elongation in the Sea Urchin Embryo Is a Microtubule-Independent Process

JEFFREY D. HARDIN

Graduate Group in Biophysics and Medical Physics and Department of Zoology, University of California, Berkeley, Califrmzia 94720

Received September 19, 1986; accepted in revised fomn December 5, 1986

Earlier studies using colchicine (L. G. Tilney and J. R. Gibbins, 1969, J. Cell Sci. 5,195-210) had suggested that intact microtubules (MTs) are necessary for archenteron elongation during the second phase of sea urchin gastrulation (secondary invagination), presumably by allowing secondary mesenchyme cells (SMCs) to extend their long filopodial processes. In light of subsequently discovered effects of colchicine on other cellular processes, the role of MTs in archenteron elongation in the sea urchin, Lytechinus pi&s, has been reexamined. Immunofluorescent staining of ectodermal fragments and isolated archenterons reveals a characteristic pattern of MTs in the ectoderm and endoderm during gastrulation. Ectodermal cells exlhibit arrays of MTs radiating away from the region of the basal body/ciliary rootlet and extending along the periphery of the cell, whereas endodermal cells exhibit a similar array of peripheral MTs emanating from the region of the apical ciliary rootlet facing the lumen of the archenteron. MTs are found primarily at the bases of the filopodia of normal SMCs. fl-Lumicolchicine (0.1 mM), an analog of colchicine which does not bind tubulin, inhibits secondary invagination, indicating that the effects previously ascribed to the disruption of MTs are probably due to the effects of colchicine on other cellular processes. The MT inhibitor nocodazole (5-10 pg/ml) added prior to secondary invagination does not prevent gastrulation or spontaneous exogastrulation, even though indirect immunofluorescence indicates that cytoplasmic MTs are completely disrupted in drug-treated embryos. Transverse tissue sections indicate that a comparable amount of cell rearrangement occurs in nocodazole-treated and control embryos. Significantly, SMCs in nocodazole-treated embryos often detach prematurely from the tip of the gut rudiment and extend abnormally large broad lamellipodial protrusions but are also capable of extending long slender filopodia comparable in length to those of control embryos. These results indicate that cytoplasmic MTs are not essential for either filopodial extension by SMCs or for the active epithelial cell rearrangement which accompanies archenteron elongation during sea urchin gastrulation. 0 1987 Academic Press, IX.

The sea urchin gastrula is a convenient model system for studying the roles of cell motility and changes in cell shape during embryonic development. Sea urchin gastrulation occurs in two stages (Dan and Okazaki, 1956). Primary invaginatim involves the inward bending of the vegetal plate to forrn a “table-topped” archenteron (Okazaki, 1975). During secondary invagination this invagination elongates to form a tube that is longer but smaller in diameter. Secondary invagination is accompanied by the rearrangement of epithelial cells in the wall of the archenteron (Ettensohn, 1985); this rearrangement actively lengthens the gut rudiment (Hardin and Cheng, 1986).

Although the basic morphogenetic movements of gastrulation have been described in increasing detail, it has proven more difficult to ,unravel exactly how the cytoskeleton functions in these movements. In particular, several studies have focused on the role of microtubules (MTs)’ in later stages of sea urchin development. In a

pioneering study, Gibbins et al. (1969) used transmission electron microscopy to follow changes in MT structure during primary mesenchyme cell ingression, migration, and association to form the calcareous spicules of the pluteus larva. Tilney and Goddard (1970) analyzed the polymerization of MTs following their disruption in the ciliated ectoderm of Arbacia pun&data blastulae and described a radiating array of MTs originating near the basal body and extending along the periphery of these cells. Most important in the context of the present study, Tilney and Gibbins (1969) found that colchicine and hydrostatic pressure could at least partially inhibit the second phase of gastrulation, and they therefore con- cluded that MTs were necessary for secondary invagination to occur. MTs were found at the bases of filopodia extended by secondary mesenchyme cells in untreated embryos, and they were presumed to stabilize these filopodia as they extended across the blastocoel and attached to the blastocoel roof. It was believed that the filopodia would then contract to pull the archenteron across the blastocoel.

’ Abbreviations used: MT, microtubule; SMC, secondary mesenchyme However, recent evidence suggests that a reexami- cell; MFSW, Millipore-filtered seawater; DIC, differential interference nation of the role of MTs during secondary invagination contrast; DMSO, dimethyl sulfoxide; PBS, phosphate-buffered saline. is in order. First, as Tilney and Gibbins (1969) were quick

253 0012-1606/87 $3.00 Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

254 DEVELOPMENTAL BIOLOGY VOLUME 121, 1987

to acknowledge, their results using colchicine were vari- able, and the use of hydrostatic pressure is problematic, since high pressure also disrupts microfilaments. Second, since the time of this early study, colchicine has been found to exert effects on other cellular processes, including disruption of membrane ion transport (Bos and Emmelot, 1974), osmoregulation (Beebe et aZ., 1979), and nucleoside uptake (Mizel and Wilson, 1972). The avail- ability of lumicolchicine, a derivative of colchicine that does not bind tubulin but presumably exhibits the MT- independent effects of colchicine (Wilson and Friedkin, 1967), has now made it possible to learn whether the effect of colchicine on gastrulation is due to its action on MTs or some other cellular process. Indeed, several early experiments utilizing colchicine have been rein- terpreted as a result, including the effects of the drug on melanophore movements in teleosts (Lambert and Fingerman, 1976) and on the control of the timing of the cell cycle in echinoderm eggs (Sluder et aZ., 1986). Finally, filopodial contraction during secondary invagination is probably less important in lengthening the gut rudiment than active rearrangement of epithelial cells in the wall of the archenteron (Ettensohn, 1985; Hardin and Cheng, 1986).

In light of these findings, the role of MTs during secondary invagination has been reexamined here, using the more specific inhibitor of tubulin polymerization, nocodazole. Although nocodazole has some long-term effects on cellular processes apparently unrelated to MTs (De Brabander et aZ., 1975), it is not toxic to sea urchin embryos (Karp and Solursh, 1985). Experiments using nocodazole and /3-lumicolchicine, combined with indirect immunofluorescent staining of MTs, indicate that intact cytoplasmic MTs are not necessary for the extension of filopodia by secondary mesenchyme cells or for the active epithelial cell rearrangement and cell shape changes that occur during secondary invagination.

MATERIALS AND METHODS

Procurement of embryos. Gametes of Lytechinus pictus (Marinus Biologicals, Winchester, CA) were collected “dry” by intracoelomic injection of isotonic KCl, fertilized, and cultured in Millipore-filtered seawater (MFSW) as described previously (Hardin and Cheng, 1986). Cul- tures were maintained either in large beakers with stir- ring paddles at 16°C or in 6- or 20-ml aliquots in tissue culture dishes; embryos were collected for experimental manipulation by gentle hand centrifugation.

Microswpy. Living embryos were attached to coverslips coated with a solution of 1 mg/ml poly-L-lysine HBr (Sigma). The coverslips were then attached to microscope slides at the edges with silicone sealant (Dow Corning) and viewed using an Olympus differential in-

terference contrast (DIC) microscope and either a 20X or 40x Olympus DIC objective.

Tissue sectioning. Embryos were fixed and collected as described previously (Hardin and Cheng, 1986); fixed embryos were dehydrated in increasing concentrations of ethanol, cleared in Histosol (National Diagnostics), and embedded in Paraplast embedding medium (Mo- noject Scientific). Embedded embryos were sectioned at a thickness of 8 pm, stained with hematoxylin and eosin, mounted, and viewed using DIC optics.

Drug treatments. For lumicolchicine experiments, embryos were collected at the early gastrula stage and placed in 0.1 mM @-lumicolchicine (Sigma) dissolved in MFSW. For nocodazole experiments, a nocodazole stock solution was added to early gastrulae cultured in tissue culture dishes. The stock solution consisted of 2 mg/ml nocodazole (Sigma) in pure dimethyl sulfoxide (DMSO); a sufficient amount of the stock solution was added to achieve a final concentration of 5-10 /*g/ml nocodazole. Parallel control experiments using an equivalent amount of DMSO without nocodazole were also performed.

Indirect immunajZuorescence. Two protocols were used. For whole-mount preparations, a procedure modified from Harris et al. (1980) was used. Embryos were attached to coverslips coated with poly+lysine; embryos were then quickly transferred to 100% ice-cold methanol for 6-10 set and immediately placed in phosphate-buffered saline (PBS) at 22°C. Alternatively, for viewing isolated archenterons, embryos were placed in 0.5-l ml ice-cold methanol for 6-10 set in a 15-ml conical centri- fuge tube. PBS (8-10 ml) at 22°C was then quickly added, and the embryos were fragmented by 2-3 passes through a loose Dounce homogenizer (15 ml, A pestle; Kontes Glass Co., Vineland, NJ). The fragments were then attached to poly-L-lysine-coated coverslips. In both cases, specimens on coverslips were rinsed three times in PBS, and primary antibody (a monoclonal antibody against sea urchin flagellar tubulin; Asai et al, 1982) was applied to the coverslips for 1 hr at 37°C in a humid chamber. The preparation and characterization of this antibody have been described previously (Asai et ab, 1982). Spec- imens were rinsed three times in PBS, and secondary antibody (FITC-conjugated goat anti-mouse, Cappel Laboratories, Cochranville, PA) was applied for 1 hr at 37°C. The specimens were then rinsed three times in PBS and mounted under coverslips in 90% glycerol, 10% PBS, 100 mg/ml 1,4-diazobicyclo-(2,2,2)octane (Aldrich), an antibleaching agent (Wordeman et ab, 1986). Mounted specimens were viewed using a Nikon Diaphot fluorescence microscope equipped with either a Nikon 40X fluorescence or a Zeiss 63X Neofluar water-immersion objective and a Nikon 2.5X or 5X camera ocular. Specimens were photographed with a Nikon FX-35A camera using Technical Pan 2415 35mm film (Kodak).

JEFFREY D. HARDIN Microtubules and Sea Urchin Gastrulatim 255

Methanol Fixation

RE:SULTS ments at the gastrula stage, a pattern very similar to that found by Tilney and Goddard (1970) for the blastula stage is observed. MTs can be seen radiating away from

The procedures presented here show that methanol is an acceptable and convement fixative for use in indirect immunofluorescence studies of sea urchin embryos. Glutaraldehyde, the fixative of choice for preserving MT structure, results in unacceptable autofluorescence even following reduction with sodium borohydride (Hollen- beck and Cande, 1985), and as a result it was not used in this study. Methanol has been used extensively to study the cleavage divisions of the sea urchin embryo (Harris et al., 1980; Balczon and Schatten, 1983), and in the one study that compared MT fixation in methanol- and glutaraldehyde-fixed preparations, no appreciable difference was found in MT morphology (Balczon and Schatten, 1983). In addition, cellular morphology of methanol-fixed embryos compares favorably with glutaraldehyde-fixed embryos at the gastrula stage (J. Hardin, unpublished observations). In at least one case, in the ectoderm, a direct comparison can be made be- tween the two fixation procedures (cf. Tilney and God- dard, 1970, and Results), and in this case both procedures yield identical results. Methanol thus seems to be a re- liable fixative for use in studying the distribution of MTs during sea urchin gastrulation.

Microtubule Morphology in the Ectoderm

When indirect immunofluorescent staining is performed on whole-mount preparations or ectoderm frag-

- the region of the ciliary rootlet (i.e., the point of insertion of the cilium into the cell body) on the apical surfaces of ectoderm cells in a characteristic pattern (Fig. 1A). In profile, MTs can be seen extending along the periphery of each cell toward the inner basal surface (Fig. 1B). The distribution of MTs in the ectoderm does not appear to change during gastrulation.

Microtubule Morphology in Isolated Archenterons

Although it is possible to obtain good images of MTs in whole mounts of gastrulae, fluorescence of ectodermal MTs obscures the MT morphology of endoderm cells in the archenteron. To circumvent this problem, fixed and extracted embryos were homogenized in a loose Dounce homogenizer and the resulting tissue fragments were attached to poly-L-lysine-coated coverslips (see Mate- rials and Methods). The embryos often shear apart pref- erentially, yielding archenteron fragments, whole archenterons, or vegetal plates with attached archenterons (Fig. 2). Such fragments allow improved antibody la- beling and resolution of the MT morphology of individual cells in the archenteron. The overall morphology of such fragments is indistinguishable from that of their coun- terparts in intact embryos, and the results obtained from these fragments have always been checked, as far as possible, with corresponding whole-mount preparations.

The distribution of MTs in the cells of the archenteron

FIG. 1. Microtubule morphology in the ectoderm. (A) Apical staining pattern. Note the array of microtubules radiating away from the region of the basal body/ciliary rootlet. ci, cilium. (B) Profile view of ectodermal fragments. Note the microtubules extending away from the apical region along the periphery of each cell (arrow). Scale bar = 5 Wm.


FIG. 2. Microtubule morphology in the archenteron. (A) Early gastrula. (B) Midgastrula. Note the “basket” of MTs in each cell radiating away from the region of the ciliary rootlet facing the lumen of the archenteron (dotted line). (C) Peripheral arrangement of MTs in a late gastrula. Cell boundaries are clearly defined. (D) Late gastrula. Note the flattening of the cells in the archenteron wall. A mitotic figure can be seen clearly (arrow). (E) Completion of gastrulation. A secondary mesenchyme cell associated with the archenteron is visible (arrow). Scale bar = 10 pm.

wall has some similarities to that in ectoderm cells (Fig. 2). MTs appear to originate from one or perhaps several MT organizing centers located at the apical (luminal) surfaces of endoderm cells, a location which corresponds to the sites of the rootlets of the short stubby cilia which characterize the archenteron at this time (Morrill and Santos, 1985). The MTs form a basketlike network extending along the periphery of each cell and thus con- veniently outline cell boundaries in the archenteron (Figs. 2B and 2C). This basic pattern does not change

appreciably during secondary invagination, although the “baskets” of MTs do change their shape somewhat to correspond to the flattened shape that the cells in the archenteron adopt by the end of gastrulation (Hardin and Cheng, 1986). The endoderm cells of the early gastrula (Fig. 2A) gradually lose their columnar organization as gastrulation proceeds, adopting first a slanted stacked appearance (Fig. 2B) and finally exhibiting the almost squamous morphology of the late gastrula (Figs. 2D and 2E). In addition, mitotic figures can occasionally

JEFFREY D. HARDIN Microtubules and Sea Urchin Gmtmlatim 257

be seen in the wall of the gut rudiment (Fig. 2D), but Lumicolchicine Inhibits Archenteron Elongation no more than one or two mitotic figures per specimen have ever been observed.. Because an earlier study used colchicine to depoly-

It is also possible to ,visualize individual secondary merize MTs (Tilney and Gibbins, 1969), control experi- mesenchyme cells at the tip of the archenteron in iso- ments using the colchicine analog ,&lumicolchicine were lated fragments. On the basis of transmission electron first performed to determine whether the results of this micrographs, Tilney and Gibbins (1969) reported that earlier study were due to the specific effects of colchicine in some secondary mesenchyme cells MTs are found at on tubulin or to other effects of the drug on the embryo. the bases of the filopodia, but that MTs are rarely present When 0.1 mM P-lumicolchicine is applied to embryos 2 at the tips of these extensions. Indirect immunofluores- hr prior to the onset of secondary invagination, dramatic cence confirms this result; MTs can be seen along the effects are seen on the further progress of invagination. periphery of SMCs and extending into the broad bases In over 90% of the embryos examined, archenteron elon- of filopodial processes, but they are usually not observed gation is markedly inhibited (Fig. 3). Many embryos ex- in the long slender portions of the filopodia which com- hibit a bent shortened gut rudiment (Fig. 3A) or an out- prise most of their length (data not shown). ward buckling of the vegetal plate with a short archen-

FIG. 3. The effects of 0.1 mM/3-lumicolchicine on archenteron elongation. The drug was added 2 hr prior to the onset of secondary invagination. (A) Many embryos exhibit a flaccid short archenteron. (B) Archenteron elongation ceases when the gut rudiment is one half to two-thirds of the way across the blastocoel, even though differentiation and spicule formation are unaffected. sp, spicule rudiment. (C) Control embryo at same time as (A) and (B). Scale bar = 25 pm. (D) Microtubules in the ectoderm of drug-treated embryos (compare with Fig. 1A). (E) Microtubules in the archenteron of a drug-treated embryo viewed from the animal pole (compare with Fig. 2A). Note the normal appearance of the MTs, which can be seen along the periphery of each cell (arrows). Scale bar = 5 pm.


teron, resembling so-called “enteroexogastrulae” (Dan and Okazaki, 1956; data not shown). Others have rela- tively normal morphology, except that the tip of the archenteron never reaches the animal pole. Instead, the tip of the gut rudiment stops one-half to two-thirds of the way across the blastocoel, remaining there through subsequent development (Fig. 3B). Moreover, embryos treated with @-lumicolchicine show normal MT morphology (Figs. 3D and 3E), so the effects of the drug on invagination are not due to any effect on MTs but instead represent an effect on some other cellular process. Since Tilney and Gibbins used concentrations of colchicine up to 50 times greater than the concentration of lumicolchicine used in this study, it is likely that many of the

effects of colchicine on invagination that they observed were not due to the specific effects of the alkaloid on MTs.

Nocodaxole Depolymerixes Microtubules but Does Not Inhibit Archenteron Elolzgation

In order to determine unambiguously whether or not MTs are necessary during the second phase of gastrulation, the synthetic MT inhibitor nocodazole (De Bra- bander et al, 1975) was used to study the effects of de- polymerization of cytoplasmic MTs on gastrulation. To ensure penetration of the drug throughout the embryo, a rather high dose of the drug (5-10 pg/ml) was applied

FIG. 4. The effects of 10 pg/ml nocodazole on archenteron elongation. The drug was added 2 hr prior to the onset of secondary invagination. (A) Midgastrula. Note the prominent fdopodia (fp) as well as the broad lamellipodial protrusions (1~). (B) Late gastrula. (C) Spontaneous exogastrula. Scale bar = 25 nm. (D) Indirect immunofluorescent staining of ectoderm cells. There is no evidence of intact cytoplasmic microtubules; nuclei (nu) appear dark, and the surrounding cytoplasm (cyt) exhibits punctate staining. (E) Staining pattern in the archenteron prior to the onset of secondary invagination, viewed from the animal pole (compare with Fig. 2A). MTs again appear to be disrupted. Scale bar = 5 pm.

JEFFREY D. HARDIN Microtubules and Sea Urchin Gastrulatim 259

to the embryos 2 hr prio:r to the onset of secondary invagination. This dose is 5- to lo-fold higher than that reported to inhibit mitosis in sea urchin embryos (Karp and Solursh, 1985).

When nocodazole is applied in the manner just described, the effects of the drug are rapid. Embryos begin showing abnormal swimlming behavior within 5 min, losing the ability to swim directionally, and instead spin rapidly in tight circles. Although the drug is apparently affecting the MT-based ciliary machinery of the ectoderm cells, and presumatbly their cytoplasmic MTs as well, these embryos gastrulate essentially normally (Figs. 4A and 4B). In particular, nocodazole-treated embryos undergo epithelial cell rearrangement to an extent comparable to DMSO controls as judged by transverse tissue sections (Fig. 5). The flattening of cells in the wall of the archenteron during secondary invagination also occurs (Fig. 6). DMSO-treated embryos and untreated controls behave identically (data not shown). Occasion- ally, untreated embryos will spontaneously exogastru- late, and embryos treated with nocodazole are also capable of spontaneous exogastrulation (Fig. 4C). Since exogastrulation is accolmpanied by active cell rearrangement (Hardin and Chen, 1986), such active rearrangement is apparently unaffected by the drug.

Since archenteron elongation occurs normally in the presence of nocodazole, it is of interest to show that the drug is actually depolymerizing cytoplasmic MTs before secondary invagination begins. When drug-treated embryos are examined by indirect immunofluorescence prior to the onset of secondary invagination, all of their cells display diffuse cytoplasmic staining surrounding a dark nucleus, with no evidence of intact cytoplasmic MTs (Figs. 4D and 4E).

Although intact cytoplasmic MTs do not seem to be crucial for the rearrangement and flattening of the cells

17.5 mlate

! 15 T

0 /

2.5

t

FIG. 5. The effect of nocodazole on cell rearrangement in the archenteron. The number of cells around the circumference of the archenteron (mean + SD) was counted from midtransverse paraffin sections for mid- (white) and late l(stippled) gastrulae treated with either 10 pg/ml nocodazole in 0.5% v/v dimethyl sulfoxide or 0.5% DMSO alone.

urnid q late

DMSO Nocodazole

FIG. 6. The effect of nocodazole on cell flattening in the archenteron wall. The thickness (mean f SD) in the apical (inner) to basal (outer) direction was measured from midtransverse paraffin sections for mid- (white) and late (stippled) gastrulae treated with either 10 fig/ml no- eodazole in 0.5% v/v DMSO or 0.5% DMSO alone.

in the archenteron during secondary invagination, there are some alterations in cellular morphology induced by nocodazole. Most notably, secondary mesenchyme cells (SMCs) often exhibit abnormally broad flat lamellipodial protrusions (Fig. 4A). In addition, some SMCs leave the tip of the archenteron prematurely, adopting this broad lamellar shape as they migrate through the blastocoel. However, many SMCs are still capable of extending long narrow protrusions comparable in length to the filopodia extended by SMCs in control embryos. In fact, many of the broad protrusions taper, ending as long slender processes (Fig. 4A). Thus, although intact cytoplasmic MTs seem to aid in the formation of filopodial protrusions by SMCs, they do not seem to be essential for this process, despite the fact that they are usually found at the bases of filopodia in normal embryos (see above).

DISCUSSION

The powerful techniques of immunocytochemistry have been used for years to study the role of the cytoskeleton in motility and cell shape changes in tissue culture cells, but it has proven more difficult to apply these same techniques to postcleavage stage embryos. Recent studies have included localization of contractile proteins during mouse and chick neurulation (Sadler et ak, 1982; Lee and Nagle, 1985), localization of actin and tubulin during early development in Drosophila (Karr and Al- berts, 1986), and an investigation of the role of MTs in the spreading of the epiblast during chick gastrulation (Mareel et al, 1984). These techniques have also been used to study the role of MTs during plant morphogenesis (Lloyd, 1986). However, no immunofluorescence studies have been made of the changes in the cytoskeleton during sea urchin gastrulation. Furthermore, the role of the cytoskeleton in epithelial cell rearrangement

260 DEVELOPMENTAL BIOLOGY VOL~JME 121, 1987

has been unexplored. This study investigates some of these questions.

Llrug Studies of the Role of Microtubules in Gastrulation

This study shows that MTs are not crucial for secondary invagination and suggests that caution should be exercised when interpreting the results of experiments employing MT inhibitors, particularly colchicine. Many of the early studies of the role of MTs in morphogenesis used colchicine, which has since been shown to have side effects, particularly at the cell membrane (see Introduction). The inhibition of secondary invagination by lumicolchicine argues that the effects of both lumicolchicine and colchicine on gastrulation are due to some shared property or properties other than the ability to bind tubulin. Therefore one cannot conclude from earlier colchicine studies (Tilney and Gibbins, 1969) that secondary invagination depends on intact cytoplasmic MTs. Furthermore, the occurrence of secondary invagination despite the demonstrated disruption of MTs with nocodazole, a MT inhibitor with a greater specificity and fewer side effects than colchicine (De Brabander et al., 1975), suggests that in Lytechinus pi&us, secondary invagination, and the cell rearrangements that accompany it, are largely MT-independent.

Studies using nocodazole also require cautious inter- pretation, especially when cells are exposed to the drug for an extended period of time (De Brabander et ah, 1975). This is particularly important when using nocodazole to study a developmental process such as sea urchin gastrulation, which occurs over several hours, as opposed to the short-term exposures commonly used to study tissue culture cells. Because nocodazole did not prevent secondary invagination and the efficacy of the drug could be confirmed by examining the integrity of cytoplasmic MTs, the unambiguous conclusion can be drawn that intact MTs are not crucial for archenteron elongation.

The Role of Microtubules in Morphogenetic Shape Changes

Because nocodazole does not appear to affect the changes in shape that cells in the archenteron undergo during secondary invagination, the general question arises as to what role MTs play in motility and changes in cell shape. The answer to this question is unclear; certain motile events in some cell types seem to depend on MTs, whereas similar processes in other cell types do not. The polarity of motile fibroblasts, but not their protrusive activity, can be disrupted with MT poisons (Vasiliev, 1982); on the other hand, movement of fish keratocytes, including directional movement, does not

depend on MTs (Euteneuer and Schliwa, 1984). Similarly, columnarization of neural plate cells during amphibian and chick neurulation may depend on MTs (Burnside, 1973; Karfunkel, 1974), but MTs are not necessary for the elongation of cells in the lens primordium (Beebe et ah, 1979). In at least one cultured cell line, MT morphology does not seem to have any direct link to cell shape (De Brabander et ah, 1986). Although MT morphology was not examined, one study using colcemid suggests that the shape changes that occur during amphibian gastrulation are largely MT-independent (Cooke, 1973). It is possible that MTs are not directly responsible for determining cell shape in many systems but rather stabilize shape changes which are produced by other intracellular machinery (Hilfer and Searls, 1986). Such a role for MTs would be consistent with the observations presented here for sea urchin gastrulation.

The Role of Microtubules in Filopodium Formation

Based on the effects of colchicine it appeared that MTs were necessary for the initiation and stabilization of filopodia extended by secondary mesenchyme cells (Til- ney and Gibbins, 1969). In contrast, this study shows that despite disruption of cytoplasmic MTs with nocodazole, filopodia in drug-treated embryos are often as long and as slender as those found in controls. This finding is consistent with what is known about the structure of long slender protrusions in other systems. Microspikes extended by growth cones in neurons contain bundled actin but MTs only insert at their bases (Letourneau, 1983). Similarly, the filopodia of transformed sea urchin coelomocytes are filled with paracrystalline arrays of actin, but no MTs (Edds, 1984). The acrosomal reaction in Thyone sperm involves the rapid extension of a long acrosomal process, accompanied by the explosive polymerization of actin (Tilney et ah, 1973), and a similar mechanism seems to be operating during the initial stages of microvillus elongation in fertilized sea urchin eggs (Begg et a.Z., 1982; Carron and Longo, 1982). In the sea urchin embryo, filopodial protrusions extended by primary mesenchyme cells contain an actin meshwork but are devoid of MTs except at their bases (Katow and Solursh, 1981). Since actin bundles are found along the length of secondary mesenchyme filopodia while MTs are found primarily at their bases (Tilney and Gibbins, 1969), it is likely that restructuring of actin within the filopodia is the most important process involved in their extension.

However, intact MTs may increase the likelihood that such slender protrusions will form, perhaps by allowing a greater degree of localization of protrusive activity. This may be the case with fibroblasts; colcemid treat-

JEFFREY D. HARDIN Microtubules and Sea Urchin Gastrulation 261

ment causes fibroblasts to form a symmetrical lamelli- podium around the entire cell periphery, instead of in a single location (Vasiliev, 1982). Since the secondary mesenchyme cells of drug-treated sea urchin embryos possess many broad lamellar protrusions in contrast to the slender localized filopodia of the controls, a mechanism similar to that operating in fibroblasts may be operating in secondary mesenchyme cells.

The Role of Microtubules in Cell Rearrangement

Epithelial cell rearrangement is an important phe- nomenon in many morphogenetic systems (Fristrom, 1976; Keller, 1978; Bode and Bode, 1984; Ettensohn, 1985; Locke, 1985; Keller and Trinkaus, 1987). However, little or no information is available on the role of MTs in these systems. Although MTs are consistently found in cells of the archenteron throughout sea urchin gastrulation, they do not appear to be necessary for the active epithelial cell rearrangement that occurs during secondary invagination (Hardin atnd Cheng, 1986). Colchicine treatment of Drosophila imaginal discs suggests that MTs are not necessary for epithelial cell rearrangement in this system as well (Fristrom and Fristrom, 1975). MTs may help to stabilize the shapes of cells in the invaginating gut rudiment and the surrounding ectoderm, but they do not appear to be indispensable for the cell rearrangements that occur during secondary invagination. It is still not known what intracellular mechanisms are responsible for epithelial cell rearrangement during sea urchin gastrulation, but the search for such mechanisms must continue at the cytoskeletal level. As more attention is given to the intracellular basis of cellular behavior during sea urchin gastrulation, our under- standing of secondary invagination, and epithelial cell rearrangement in general, should be improved.

I thank Dr. Pat Harris for suggesting this project, Linda Wordeman for technical advice and for providing the FITC-conjugated secondary antibody and mounting medium, Dr. David Asai for providing the antitubulin monoclonal antibody, and Dr. Dave Weisblat for the use of his 63X objective. Helpful comments on the manuscript were supplied by Mike Koonce and by Drs. Manfred Schliwa, J. P. Trinkaus, Ray Keller, and Fred Wilt. This work was supported by a NSF predoctoral fellowship and by NIH Grants HDl5043 to F. Wilt and HD18979 to R. Keller.

REFERENCES

ASAI, D. J., BROKAW, C. J., THOMPSON, W. C., and WILSON, L. (1982). Two different monoclonal antibodies to tubulin inhibit the bending of reactivated sea urchin spermatozoa. Cell Motil. 2, 599-614.

BALCZON, R., and SCHATTEN, G. (1983). Microtubule-containing deter- gent-extracted cytoskeletons in sea urchin eggs from fertilization through cell division: Antitubulin immunofluorescence microscopy. Cell MotiL 3, 213-226.

BEEBE, D. C., FEAGANS, D. E., BLANCHETTE-MACKIE, E. J., and NAU,

M. E. (1979). Lens epithelial cell elongation in the absence of microtubules: Evidence for a new effect of colt hicine. Science 206,836- 838.

BEGG, D. A., REBHUN, L. I., and Fr, :T, H. (1982) Structural organization of actin in sea urchi 1 egg cort,ex: Microvillar elongation in the absence of actin filament 1 ndle formation. J. Cell BioL 93,24- 32.

BODE, P. M., and BODE, H. &. ( )84). Formation of pattern in regen- erating tissue pieces o’ Hyc v attenuafa. III. The shaping of the body column. Dev. Biol 106, ‘115-325.

Bos, C. J., and EMMELOT, P. (1974). Studies on plasma membranes. XXI. Inhibition of liver plasma membr me enzymes by tumour-pro- moting phorbolester, mitotic inhibitor., and cytochalasin B. Chew.- Biol. Interact. 8, 349-361.

BURNSIDE, B. (1973). Microtuhlrle and ricrofilaments in amphibian neurulation. Amer. ZooL 13, 989 W’?

CARRON, C. P., and LONGO, F. J. (19821. Relation of cytoplasmic alka- linization to microvillar elongation and microfilament formation in the sea urchin egg. Dev. Biol. 89,128-137.

COOKE, J. (1973). Properties of the primary organization field in the embryo of Xenopus laevis. IV. Pattern formation and regulation following early inhibition of mitosis. J. EmbryoL Exp. MwphoL 30,49- 62.

DAN, K., and OKAZAKI, K. (1956). Cyto-embryological studies of sea urchins. III. Role of secondary mesenchyme cells in the formation of the primitive gut in sea urchin larvae. BioL Bull. 110,29-42.

DE BRABANDER, M., GEUENS, G., NUYDENS, R., WILLEBRORDS, R., AERTS, F., and DE MEY, J. (1986). Microtubule dynamics during the cell cycle: The effects of taxol and nocodazole on the microtubule system of Pt K2 cells at different stages of the mitotic cycle. Int. Rev. CytoL 101,215-274.

DE BRABANDER, M., VAN DE VEIRE, R., AERTS, F., GEUENS, G., BORGERS, M., DESPLENTER, L., and DE CREE, J. (1975). Oncodazole (R 17934): A new anti-cancer drug interfering with microtubules. Effects on neoplastic cells cultured in vitro and in viwo. In “Microtubules and Microtubule Inhibitors” (M. Borgers and M. De Brabander, Eds.), pp. 509-521. North-Holland, Amsterdam.

EDDS, K. T. (1984). Differential distribution and function of microtubules and microfilaments in sea urchin coelomocytes. Cell MotiL 4, 269-281.

ETTENSOHN, C. A. (1985). Gastrulation in the sea urchin embryo is accompanied by the rearrangement of invaginating epithelial cells. Dev. Biol. 112,383-390.

EUTENEUER, U., and SCHLIWA, M. (1984). Persistent, directional motility of cells and cytoplasmic fragments in the absence of microtubules. Nature (Londmz) 310, 58-61.

FRISTROM, D. (1976). The mechanism of evagination of imaginal discs of Drosophila melanoguster. III. Evidence for cell rearrangement. Dew. BioL 54,163-171.

FRISTROM, D., and FRISTROM, J. W. (1975). The mechanism of evagination of imaginal discs of Drosophila melanogaster. I. general con- siderations. Dev. Biol. 43, l-23.

GIBBINS, J. R., TILNEY, L. G., and PORTER, K. R. (1969). Microtubules in the formation and development of the primary mesenchyme in Arbacia punctulata. I. The distribution of microtubules. J. Cell BioL 41,201-226.

HARDIN, J. D., and CHENG, L. Y. (1986). The mechanisms and mechanics of archenteron elongation during sea urchin gastrulation. Dev. BioL 115,490-501.

HARRIS, P., OSBORN, M., and WEBER, K. (1980). Distribution of tubulin- containing structures in the egg of the sea urchin Strongylocentrotus purpuratus from fertilization through first cleavage. J. Cell BioL 84, 668-679.

262 DEVELOPMENTAL BIOLOC Y VOLUME 121, 1987

HILFER, S. R., and SEARLS, R. L. (1986). Cytoskeletal dynamics in animal morphogenesis. In “Developmental Biology: A Comprehensive Syn- thesis. Vol. 2: The Cellular Basis of Morphogenesis” (L. Browder, Ed.), pp. 3-29. Plenum, New York.

HOLLENBECK, P. J., and CANDE, W. Z. (1985). Microtubule distribution and reorganization in the first cell cycle of fertilized eggs of Lyte- chinus p&us. Eur. J. Cell Biol. 37, 140-148.

KARFUNKEL, P. (1974). The mechanism of neural tube formation. Int. Rev. CytoL 38,245-271.

KARP, G. C., and SOLURSH, M. (1985). Dynamic activity of the filopodia of sea urchin embryonic cells and their role in directed migration of the primary mesenchyme in vitro. Dev. BioL 112,2’76-283.

KARR, T. L., and ALBERTS, B. M. (1986). Organization of the cytoskeleton in early Drosophila embryos. J. Cell BioL 102,1494-1509.

KATOW, H., and SOLURSH, M. (1981). Ultrastructural and time-lapse studies of primary mesenchyme cell behavior in normal and sulfate- deprived sea urchin embryos. Exp. Cell Res. 136,223-245.

KELLER, R. E. (1978). Time-lapse cinematographic analysis of super- ficial cell behavior during and prior to gastrulation in Xenopus lot&s. J. Mwrphol. 157,223-248.

KELLER, R. E., and TRINKAUS, J. P. (1987). Rearrangement of enveloping layer cells without disruption of the epithelial permeability barrier as a factor in Fundulous epiboly. Dev. BioL 120,12-24.

LAMBERT, D. T., and FINGERMAN, M. (1976). Evidence for a nonmicro- tubular colchicine effect in pigment granule aggregation in mela- nophores of the fiddler crab, Uca pugilator. Camp. Biochem. Physiol. C 53,25-28.

LEE, H.-Y., and NAGLE, R. G. (1985). Studies on the mechanisms of neurulation in the chick: Interrelationship of contractile proteins, microfilaments, and the shape of neuroepithelial cells. J. Exp. ZooL 235,205-215.

LETOURNEAU, P. (1983). Differences in the organization of actin in the growth cones compared with the neurites of cultured neurons from chick embryos. .I Cell BioL 97,963-973.

LLOYD, C. W. (1986). Microtubules and the cellular morphogenesis of plants. In “Developmental Biology: A Comprehensive Synthesis” (L. Browder, Ed.), pp. 31-57. Plenum, New York.

LOCKE, M. (1985). The structure of epidermal feet during their development. Tissue Cell 17, 901-921.

MAREEL, M., BELLAIRS, R., DE BRUYNE, G., and VAN PETEGHEM, C. (1984). Effect of microtubule inhibitors on the expansion of hypoblast and

margin of overgrowth of chick blastoderms. J. Embryo1 Ezp. Mar- phoL 81,273-286.

MIZEL, S. B., and WILSON, L. (1972). Nucleoside transport in mammalian cells. Inhibition by colchicine. Biochemistry 11, 2573-25’78.

MORRILL, J. B., and SANTOS, L. L. (1985). A scanning electron micro- scopical overview of cellular and extracellular patterns during blas- tulation and gastrulation in the sea urchin, &tech&us variegatus. In “The Cellular and Molecular Biology of Invertebrate Develop- ment.” (R. H. Sawyer and R. M. Showman, Eds.), pp. 3-33. Univ. South Carolina Press, Columbia, SC.

OKAZAKI, K. (1975). Normal development to metamorphosis. In “The Sea Urchin Embryo” (G. Czihak, Ed.), pp. 177-232. Springer-Verlag, Berlin.

SADLER, T. W., GREENBERG, D., and COUGHLIN, P. (1982). Actin distribution in the mouse neural tube during neurulation. Science 215, 172-174.

SLUDER, G., MILLER, F. J., and SPANJIAN, K. (1986). The role of spindle microtubules in the timing of the cell cycle in echinoderm eggs. J. Exp. 2001. 238,325-336.

TILNEY, L. G., and GIBBINS, J. R. (1969). Microtubules and filaments in the filopodia of the secondary mesenchyme cells of Artia pun&data and Echinarachnius pawna. J. Cell Sci 5,195-210.

TILNEY, L. G., and GODDARD, J. (1970). Nucleating sites for the assembly of cytoplasmic microtubules in the ectodermal cells of blastulae of Arbacia punctulata. J. Cell. Biol. 46, 564-575.

TILNEY, L. G., HATANO, S., ISHIKAWA, H., and MOOSEKER, M. S. (1973). The polymerization of actin: Its role in the generation of the acrosomal process of certain echinoderm sperm. J. Cell BioL 59,109-126.

VASILIEV, J. M. (1982). Pseudopodial attachment reactions. In “Cell Behavior” (R. Bellairs, A. Curtis, and G. Dunn, Eds.), pp. 135-158. Cambridge Univ. Press, Cambridge.

WILSON, L., and FRIEDKIN, M. (1967). The biochemical events of mitosis. II. The in viva and in vitro binding of colchicine in grasshopper embryos and its possible relation to inhibition of mitosis. Biochem- isby 6, 3126-3135.

WORDEMAN, L., MCDONALD, K. L., and CANDE, W. Z. (1986). The distribution of cytoplasmic mictrotubules throughout the cell cycle of the centric diatom Stephanopqzis turris: Their role in nuclear migration and positioning the mitotic spindle during cytokinesis. J. Cell BioL 102, 1688-1698.

archenteron elongation in the sea urchin embryo is a...

Documents