specification of endoderm in the sea urchin embryo

m D Mechanisms of Development 67 (1997) 35-47

Specification of endoderm in the sea urchin embryo

Robert E. Godin’, David C. Klinzing’, William A. Porcaro, Susan G. Ernst*

Department of Biology, Tufts Vniversify, Medford, MA 02155. USA

Received 27 February 1997; received in revised form 10 June 1997; accepted 18 June 1997

Abstract

Expression of the endoderm specific gene Endol6, was used to monitor endodenn specification in developmentally arrested and in dissociated embryos. P-APN treatment halts gastrulation, however, in two species of sea urchins End016 mRNA is still expressed, suggesting that specification of the endodermal lineage has taken place in these developmentally arrested embryos. End016 mRNA is not expressed in embryos dissociated at the 4-&cell stage unless they are reassociated shortly after the 16-cell stage. Interestingly, dissociation after the 16-cell stage also results in a lack of End016 expression. Lithium is unable to rescue End016 expression in these dissociated embryos. These results indicate that early signaling events mediated by cell-cell contact are required for the initial specification and maintenance of the endoderm in the developing embryo. 0 1997 Elsevier Science Ireland Ltd.

Keywords: Sea Urchin; Endoderm; Endol6; Lithium; Cell dissociation; /3-APN

1. Introduction

In eggs of many animals spatially restricted maternal factors determine cell fate (reviewed in Davidson, 1989,

1990, 1991; Ding and Lipshitz, 1993; Kessler and Melton, 1994). Subsequent cleavages and later morphogenic movements create new developmental contexts and cell-cell

interactions which result in the emergence of specific cell types. The sea urchin embryo combines partitioning of maternal factors and inductive cell-signaling events to

establish five founder territories by, or just after, the 64- cell stage (Davidson, 1989). Isolation of molecular markers has provided us with the tools to study these early events of territory specification (reviewed in Ernst, 1997a).

The vegetal plate is the most complex of the five founder territories of the sea urchin embryo, giving rise to the sec-

ondary mesenchyme and endoderm. The first visible sign of endoderm formation is the elongation of cells to form the thickened vegetal plate. During gastrulation, the vegetal plate invaginates to form the archenteron. Treatment of sea urchin embryos with inhibitors of glycosylation, sulfa-

* Corresponding author. Tel.: +l 617 6273541; fax: +l 617 6273805.

’ These authors have contributed equally to this work.

tion, inhibitors of collagen maturation and antibodies against specific ECM components have all been shown to

disrupt cell-ECM interactions and block the visible formation of endoderm and the morphogenic movements of gas-

trulation (Solrush and Katow, 1982; Wessel and McClay,

1987; Benson et al., 1991; Burdsal et al., 1991; Burke et al., 1991). Of these inhibitors /3-aminoproprionitrile (/I-APN), an inhibitor of the enzyme lysl-oxidase which is responsible

for the triple helix formation of collagen, has been used extensively to arrest development at the mesenchyme blastula stage (Butler et al., 1987; Wessel and McClay, 1987;

Wessel et al., 1989; Benson et al., 1991). Morphological observations of these embryos has shown that gastrulation and formation of the skeletal system are blocked. Further-

more, the mRNA for the endodermal marker, LvN1.2, and the Endol protein, and the gut-specific enzyme alkaline phosphatase were not expressed in embryos treated with

@-APN, suggesting that endoderm formation does not take place in embryos in which cell-ECM interactions have been disrupted (Wessel and McClay, 1987; Wessel et al., 1989; Benson et al., 1991; Wessel, 1993).

Similarly, disruption of cell-cell interactions by removal of blastomeres, tiers of cells or complete dissociation of embryos, has been shown to inhibit gastrulation and cause

0925-4773/97/$17.00 0 1997 Elsevier Science Ireland Ltd. All rights reserved

PII SO925-4773(97)00095-6

36 RE. Godin et al. /Mechanisms of Development 67 (1997) 35-47

changes in territory-specific gene expression (Hiirstadius, 1973; Hurley et al., 1989; Stephens et al., 1989). The impor- tance of cell-cell contact in sea urchin development was

demonstrated by early work of Hijrstadius (1973). Culturing

of the animal half of a sea urchin embryo resulted in the formation of a ciliated ball of cells completely lacking visi-

ble endoderm. However, by culturing the animal half in the presence of the micromeres, the vegetal most cells of the

embryo, normal development was rescued (reviewed in Hiirstadius, 1973). Recent molecular studies (Khaner and

Wilt, 1990; Khaner and Wilt, 1991; Ransick and Davidson, 1995; Amemiya, 1996) support the role of early inductive interactions between the micromeres and the overlying

macromeres in vegetal plate specification and formation of the endoderm.

Endol6, an endoderm specific gene (Nocente-McGrath et

al., 1989; Ransick et al., 1993; Soltysik-Espailola et al., 1994) has proven to be particularly useful in analyzing endoderm specific gene expression in the sea urchin embryo

(Nocente-McGrath et al., 1989, 1991; Livingston and Wilt,

1992; Ransick and Davidson, 1993, 1995; Ransick et al., 1993; Yuh et al., 1994; Klein et al., 1995; Mao et al., 1996;

McCoon et al., 1996; Yuh and Davidson, 1996). Initial characterization indicated that End016 expression begins just prior to gastrulation (Nocente-McGrath et al., 1989). However, recent results have demonstrated that there exists

a maternal pool of End016 mRNA and that zygotic transcription begins by about 10 h during the cleavage stage, producing a single 6.6 kb mRNA (Godin et al., 1996). By

the late blastula stage, the splicing pattern of the End016 transcript changes resulting in the production of the 6.6 kb

mRNA and two 8.6 kb mRNAs. Ransick and Davidson (1995) have shown that removal of

the micromeres at the 16-cell stage (fourth cleavage) results in a significant reduction in the level of End016 expression

later in development. However, some End016 expression was observed in three out of four experiments that had no micromeres present. The End016 expression seen in these experiments could have been due to a timing factor in which

the inductive effect of the micromeres had already taken place prior to their removal. Alternatively, it could have been due to tissue autonomous expression of End016 in

the vegetal plate territory. In addition, work by Henry et al. (1989) and Khaner and Wilt (1990), using dissociated

mesomeres, suggested that cells in the animal half of the embryo, normally fated to give rise to ectoderm, may have the ability to form endodermal structures under certain conditions. We have employed End016 as an early marker to study the events that control specification and differentiation of the vegetal plate territory and future endoderm lineage, The use of this early marker for the vegetal plate has resulted in an alternative interpretation of earlier work examining endoderm formation in /3-APN arrested embryos and has led us to reinvestigate the specification of the endoderm and factors which are thought to play a role in this

process.

2. Results

2.1. Expression of End016 mRNA in /3-APN treated embryos

To determine if the endoderm specific gene, Endo16, is

expressed in developmentally arrested embryos, batches of embryos were treated with 0.5 mM &APN just after fertilization to block ECM assembly and arrest development at the mesenchyme blastula stage. RNA was isolated from /3-

APN treated embryos and control embryos from the same

batch and analyzed by RNase protection assays (Fig. 1). RNA samples were probed for Endo16, an early marker of vegetal plate specification, and Specl, an aboral ectoderm specific gene, as a control whose expression is not affected by &APN treatment (Wessel et al., 1989). As we have previously reported (Godin et al., 1996) this longer message-

specific probe protected two End016 mRNA fragments (520 and 420 bp) that were first observed approximately 18 h post-fertilization and increased in abundance throughout

gastrulation (Fig. 1, lanes l-4). Specl mRNA (Fig. 1, lanes l-4) protected a major band of 280 bp, which was

present at 15 h and steadily increased throughout development (Wessel et al., 1989). Other bands in these lanes represent undigested probe. In fl-APN treated embryos (lanes 5- 8), Specl mRNA was expressed although at lower levels

relative to the control samples, as previously reported (Wes- se1 et al., 1989). Surprisingly, we also detected End016 mRNA in these /3-APN treated embryos although the appearance of the message was delayed and neither band

was detected until 36 h of development (lanes l-4 versus lanes 5-8). The delay in expression correlated with an observed morphological delay in the development of these

&APN treated embryos (data not shown). This delay in development was observed as slower cleavage and a later hatching time relative to controls and has also been reported by other investigators (Wessel and McClay, 1987; Wessel et al., 1989). Therefore, unlike other endoderm specific genes that have previously been studied in S. purpuratus, End016 mRNA is expressed in fl-APN treated embryos.

2.2. End016 homologue from L. variegatus

In S. purpurutus expression of Specl was only slightly lowered in fi-APN treated embryos (Wessel et al., 1989; also see Fig. l), while expression of the L. variegatus homologue of Specl (LvSl) was severely disrupted when embryos of this species were treated with @-APN (Wessel et al., 1989). Therefore, it has been suggested that aboral ectoderm specification may vary between these species in

response to /3-APN treatment (Wessel et al., 1989). Recently it has been shown that in fact specification of the aboral ectoderm does vary between these species (Wikramanayake et al., 1995). To determine if specification of the vegetal plate also shows this species variation in response to &APN treatment, we analyzed the expression of End016 in L. var-

RE. Godin et al. /Mechanisms of Development 67 (1997) 35-47 31

iegatus. First, a L. variegafus cDNA library was screened

with a probe for S. purpuratus End016 and a 1.8 kb clone

was isolated, sequenced and identified as the End016 homologue (Fig. 2A). We named this gene LvEndol6. This 1.8 kb clone is 78% similar to and 6 1% identical to the corresponding region of S. purpuratus Endol6. Based on sequence data of the 1.8 kb partial cDNA fragment, it was determined that

LvEndo16 codes for a protein with domains similar to those established for S. purpurutus Endo16, including a large

cysteine-rich region, two putative calcium-binding domains and an RGD cell-binding site (Fig. 2A). Identity between the conserved domains is greater than seen in the entire protein. The protein showed interesting variations between

the two species in the number and arrangement of these

domains (Fig. 2B). Surprisingly, we observed three type B potential calcium-binding repeats (Soltysik-Espaiiola et al.,

1994), which were interrupted by a type A repeat. Experiments were performed to determine if the temporal

and spatial expression patterns of the L. variegatus homo-

logue of End016 are the same as those in S. purpuratus. Northern blot analysis using LvEndo16 as a probe detected a major high molecular weight band of -6.5 kb that fol-

lowed the same temporal expression pattern as End016 in S. purpuratus (Fig. 2C). Longer exposure of the Northern

blot revealed a second band of -8 kb (data not shown),

suggesting that the LvEndo16 transcript is alternatively

>

>

3

CONTROL

spliced in this species as it is in S. purpuratus (Godin et

al., 1996). Initially we also had difficulty detecting the

longer End016 mRNAs of S. purpuratus due to incomplete transfer from the gel and were only able to characterize the

precise temporal expression pattern of the three End016 mRNAs by RNase protection analysis using message specific probes (Godin et al., 1996). Whole mount in situ hybridization shows that LvEndo16 mRNA is found exclusively

in the invaginating archenteron during gastrulation (Fig. 2D). Therefore, the general temporal and spatial pattern of

End016 expression appears to be the same in S. purpuratus and L. variegatus.

2.3. Expression of LvEndo16 mRNA in P-APN treated embryos

This newly identified clone was used to examine the expression of @zdo16 in L. variegatus embryos treated

from the time of fertilization with 0.5 mM /3-APN to disrupt ECM assembly and arrest development. At various times after treatment, RNA was isolated from control and p-

APN treated embryos and analyzed by RNase protection assays (Fig. 3). Both LvEndo16 and LvSl, the L. variegatus Specl homologue, showed the expected patterns of expres-

sion in control batches of embryos. Based on the S. purpuratus End016 sequence, the probe used to detect End016 in

R-APN

15 18 24 36 15 18 24 36 + -

1 2 3 4 5 6 7 8 9 10

Fig. 1. Analysis of Endold and Specl expression in p-APN treated embryos. Embryos were treated with 0.5 mM fl-APN just after fertilization and RNA was

isolated and analyzed by RNase protection assays. Control embryos were from me same batch of embryos but were not treated with P-APN. RNA samples

were analyzed for both End016 and Specl in the same reaction. Stage of development in hours post fertilization is indicated above each lane. In untreated

samples we observed the expected expression pattern for both genes (lanes 14). End016 was detected as two bands (520 and 420 bp, indicated by closed

arrows) and was first observed at 18 h with an increase in abundance during gastrulation. Specl was detected as a major band of 280 bp (indicated by an open

arrow) and was present at 15 h with a steady increase throughout development. Other bands in these lanes represent the two undigested probes. Full length

probes are indicated by an asterisk (lane 10). Some full length probe was detected in the yeast control (lane 9). In /3-APN treated embryos (lanes 5-8), Specl was expressed although at lower levels. End016 was also expressed in these embryos but at a lower level and also showed a delay in expression relative to the

controls (lanes 14 versus S-8). This delay in expression correlated with an observed morphological delay in the development of fl-APN treated embryos.


A . 1

61

121

181

241

301

361

421

481

541

601

661

721

781

841

901

961

1021

1081

1141

1201

1261

1321

1381

1441

1501

1561

1621

1681

1741

1801

GAATTCCGTTAGTTTGATAGACGAGCAGGTAGTCCGTTGCTGCGAGG~GACGGGG~GG NSVSLIDEQVVR@@EEDGEG

CCAAAGAGAATGTTTAGCTGCGCTCCGTAGGAACCACTTCCACTGA QREeLAALRRNHFTAVaNTD

TGAGCTTTTCCAGATAGACATTTTTAATACCATTGTGCGATGC ELFQIDIFNTIVRTEEPINA

AGAGCACAGGTGCTGCATGAGACAGAATGCCGACATCAGACATCAGAGCCGCCTGCTTCTCTCGAGT

EHR@+MRQNADIRAA@FSRV AGATAGGCAGGTTGCTATGCTCTCATTCAGTCCCCCGATCATTG~GTCC~CAT DRQVAMLSFSPPIIEKVQNI

AAGGGAAGATCTTCTCGATGTTGCAGAGGTTAACAGACTCT REDLLDVAEVNRLTSVODEM

GAAAACGGCGCCATCGTCAGAACTTAGGGGGGGCGGTGGCCTGTG~GTGTTGTCGATT KTAPSSELRGRWPVKK@@RF

CAAGAACGACCGGAAGTTTCCTTGTATCAGGGATACCGCG~C~CTAGCAGATC~TC KN DR K F P@I RDTAKQLADQS

ATGCGATGGCCCTCCTCCTAGAGTGTACTTCCCGTGGATTCTACGAG~TTTATTCCTGA

0 CDGPPPRVYFPWILREFIPD TGATTTCGAAGGTCCTTTCCTTGATCCTTCACACCCATGTTGCTCTCTGACCGGTGACAT DFEGPFLDPSHP@@SLTGDM

GCGACATGAATGCATCAAGGAAAATAGCATGGCATGGCTGTTCTTGGATTGAG~TACC~TCAC

R H E@I K EN SMAVLGLRIPIT CCAGGCCATGCCTATGGTGGAGTTTTTCAGAGAAATGCCACCTGAGATGCGTTTTGATAT QAMPMVEFFREMPPEMRFDI CGATCACCAAAATGCCATATGCCAGGCTGCAGAAGATTTCTCCTCTATGCTGATCC D H Q N A I@Q AAEDFKILYADP

AGAAGCCAATCTTGATGTAGATCTCAACCAACTCACGGGAGA EANLDVDLNQLTGIEAELQE

ACTTCAAGACCTCTTGGCAGAAGAAATGAAATGATGGATGAAGAAGAGGATGAAGACGA I LQDLLAEEMMDEEEDEDDEH

TGGAGAAATGATGCAGGAATTGATGGAACAGAATGAAGUGAAGAAGAGGATGAGGAAGA GEMMQELMEQNEEEEEDEEE

GGAAGAAGACGATCGGAGACGGAAAAGAAGGCAAGCTGAAGTATAGGAGGCCTACGGC EEDDRRRKRRQAEKYRRPTA

CAAGAAGGTGCGCCAGTCTTATGGGAGAACTATTTATTTATGAGGTCGACAGTCTGGTTCATTG KKVRQSYGRTIYEVDSLVH@

CTGCGAAGAAAATTCTGTAACACAAGCAAGTTCTTCTTGCATGACC~CGCGTGGTTGC~CT

0 CEENSVTQAS S@MT NAWLQL GCTGAATGATGCATGTTTTGCCCTGCCTGATAATTCCCC~GATTCCTCTGCGGGATCCGT LNDA@F ALPDNSQDSSAGSV

TGGTCTAGATTCTGCTTCAGAGAGTGCACAACCTAGTAATTGTGCGCGCCCCAGCACC GLDSASESAQPSNNVRAPA.P

AATCAAATTTGGCGGGCGAGGAGACATTGAAACAAGTATGT I K F G G_R_E_D=I E T S M S T T D S T M

GGAATCAAATACTGAT~~~~CTTCAGAAAGTGTTGCGGTCTCTATGGGGATGCTCA ESNTDEENFRKWGLYGDAQ

ACTCGAATGCTTTCAAATTCACTTTAGTCCTGACGCCCCAGTGA LE@FQIHFSPDAmSSQIESE

GGGATCTTCAAATCCATCAACAGGGGATAGTACATCAG~GGGAGTTTATCAGAGAGTGA GSSNPSTGDSTSEGSL'SOSE

AGGGTCTGAAAGTGMGGTTCGGAGGGAG;WG GSESEGSEGEESEGEEDEEE

AGAATTTGAATCAGAAGGGGAAGATAGCGATAGCGACAAGA

EFESEGEDSD.SLQEAEEP[c ACAGACAGGAAGTGAAATGATGGAGACCGATTTTGAAGAAGA QTGSEHME TDFEEEEEEEEE

GGAAGAAGAAGAAGACATGGAAGAATCGGCACCACAGGCACCACAGGGGGTGATGG~CGAGGGGGA E[DMEESAaQGVMENEGE GCAATCTGAATCAAACGTCCCAGAATTTTG~G~GGCGACGGCTCTG~TCTTCACAGGG QSESNVPEFEEGDGSESSQG

AGAAAGAAGAAGAAGAAGAAGAAGAGGAAGAAGAGG 1836 ERRRRRRRGRRG

60

120

180

240

300

360

420

480

540

600

660

720

780

840

900

960

1020

1080

1140

1200

1260

1320

1380

1440

1500

1560

1620

1680

1740

1800

RE. Godin et al. /Mechanisms of Development 67 (1997) 35-47

I SignId sequencs

ml Cystelns Rsglon

# Acklk Rapeats

EF Hand loop Ilke Repeats

39

this species was predicted to be common to the sequence

found in both the long and short forms of the message and as

expected only one band was detected (Fig. 3A; also see Fig. 1). As previously reported in j3-APN-treated embryos, the

expression of LvSl was disrupted (Fig. 3B; Wessel et al., 1989) resulting in barely detectable levels of LvSl mRNA. In contrast, LvEndol6 expression was clearly detected in these &APN treated embryos (Fig. 3A). However, the level was lower and delayed relative to the controls, similar

to that seen in S. purpuratus. The delay in LvEndol6 expression is likely due to the delayed development of P-APN

treated embryos. These results showed that the different

effect of matrix disruption by /3-APN on the Spec gene family (SpecllLvSl) is specific to the Spec genes and does not reflect a general difference in response between the

species to @-APN treatment or in regard to specification of the vegetal plate territory.

2.4. Expression of End016 mRNA in dissociated embryos

Manipulation of blastomeres and disruption of cell-cell interactions have been shown to affect normal development as well as territory specific gene expression (Hurley et al., 1989; Stephens et al., 1989). Removal of the micromeres

from 16-cell stage embryos resulted in a reduction of endoderm formation, presumably because of the loss or reduction of a signal from the micromeres to the macromeres

(reviewed in Horstadius, 1973; Davidson, 1990). However, molecular analysis revealed that several of these embryos

Fig. 2. Nucleotide and nucleotide-derived amino acid sequence of the Endol6 homologue from L. variegatus. Genbank Accession No. U89340. (A) Based on

partial cDNA sequence data (1.8 kb), LvEndo16 codes for domains similar to that established in S. purpuratus Endol6. Cysteines are enclosed in diamonds,

type A repeats are boxed, the first amino acid of each type B repeat is boxed and the RGD sequence is indicated by a double underline. (B) The protein varies

in the number and arrangement of the domains identified in S. purpuratus Endol6. (C) Northern blot analysis using LvEndol6 as a probe detected a major

band of -6.5 kb (arrow) that followed the same temporal expression pattern as End016 in S. purpuratus. Times shown in hours of development correspond to

blastula (9), gastrula (15, 17) and prism (25) stages. (D) Whole-mount in situ hybridization of late gastrula (LG) stage embryo showed that LvEndoZ6 expression is specific to the endoderm. Bar, 10 pm.


still expressed End016 (Ransick and Davidson, 1995), suggesting that derivatives of the macromeres may have the

innate potential to become endoderm and activate End016 expression. Therefore, we analyzed the expression of End016 in dissociated embryos to remove all influence of cell-cell interactions. S. purpurutus eggs were fertilized in

PABA to aid in the removal of the fertilization envelopes (Hall, 1978) and cultured until the 4-8cell stage. Embryos

were then dissociated by shaking in several washes in calcium-free sea water to disrupt cell-cell contacts prior to the 16-cell stage before the division of the four vegetal blastomeres into micromeres and macromeres. By placing the

culture on a rotary shaker and culturing under calcium- free conditions, the cells continued to dissociate from

each other at each cleavage and a culture of single cells was maintained. A separate portion of these embryos was

cultured with gentle stirring so that cells cleaved but did not

completely dissociate. A sample of the completely dissociated cells was removed after 1 h and allowed to reassociate in normal sea water. At 24 and 50 h post-dissociation,

RNA was isolated from each of the cultures and analyzed by RNase protection assays. In Fig. 4B, and as previously reported (Stephens et al., 1989), Specl was expressed in dissociated embryos but at a lower level relative to intact

embryos. Similar results were seen in both the 24 h (Fig. 4B, lanes l-4) and 50 h samples (Fig. 4B, lanes 5-8), although

the level of Specl was much lower in the dissociated samples at 50 h. In control non-dissociated embryos, we

observed normal levels of End016 expression (Fig. 4A, lane 1). However, in the completely dissociated embryos

A. CONTROL 8-APN

15 20 23 25 15 20 23 25

1 2 3 4 5 6 7 8

12345678

Fig. 3. Expression of LvEndoZ6 and LvSl mRNA in P-APN treated

embryos. Embryos were treated with 0.5 mh4 &%F’N to disrupt gastrula-

tion and RNA was isolated and analyzed by RNase protection assays.

Stage of development in hours post fertilization is indicated above each

lane. Duplicate samples were probed for LvEndol6 (A) or LvSl (B). Con-

trol RNA samples showed expression patterns similar to those seen in S.

purpuratus for both LvEndo16 and LvSl (lanes 14, also see Fig. 1). As previously reported for P-APN treated embryos, the expression of LvSl is

dramatically reduced and barely at the level of detection. In contrast

LvEndol6 is expressed and, similar to S. purpuratus expression, is delayed

relative to the controls (lanes 5-8).

A. 24H 50H

C DGRCDGR

w

/#_ I iT i >”

12345678

B.

-_ ._.~

i 2 3 4 5 6 7 8

Fig. 4. Expression of End016 mRNA in dissociated S. pwpuratus embryos.

RNA was isolated at 24 and 50 h and duplicate samples were analyzed by

RNase protection assays. C, control; D, dissociated; G, gentle dissociation;

R, reassociated. (A) In control non-dissociated embryos, normal levels of

End016 expression were observed. However, in the completely dissociated

embryos no End016 mRNA was detected. In embryos that were gently

dissociated, some cell contacts were maintained and End016 was

expressed, although at a lower level than in controls. In embryos that

were allowed to reassociate, expression of End016 was observed at a

level similar to that of the control. The same results were observed and

are more dramatic in the 50 h samples where Endold is normally more

abundant (lanes 5-8). (B) Specl was still expressed in dissociated cells as

previously reported, but a decrease in the level of mRNA was seen in later

stage dissociated embryos.

no End016 mRNA was detected (Fig. 4A, lane 2), while

in the embryos that were gently dissociated and maintained some cell contact End016 was expressed slightly, although at a lower level than the controls (Fig. 4A, lane 3). In embryos that were allowed to reassociate after 1 h of dissociation, expression of End016 was observed at levels similar to the control (Fig. 4A,.lane 4). The same results

were observed and were more dramatic in the 50 h samples where End016 is normally more abundant (lanes 5-8). These results clearly demonstrated that cell-cell contact

after the 8-cell stage is necessary to express End016 and to specify endoderm in the sea urchin embryo.

2.5. Expression of End016 in late reassociated embryos

Previous investigations have indicated that under prolonged culture conditions, animal half embryos can re-specify cells to form vegetal structures (Horstadius, 1973; Khaner and Wilt, 1990). However, the above results showed that after 50 h of dissociation, isolated cells still did not express End016 and suggested that cells must be in contact for End016 expression to take place. To investigate further the necessity of cell-cell contact for EndoZ6 expression and to determine how long the presumed signal and/or ability to


12345676

Fig. 5. Endold expression in reassociated embryos. Cells were dissociated

at the 4-8-cell stage and reassociated after 1, 2.5, or 8.5 h of dissociation.

Samples were analyzed for the expression of End016 (closed arrow) and

Specl (open arrow) by RNase protection assays. Full length probes are

indicated by an asterisk in lane 8 and protected fragments are indicated by

arrows. Lane 1 is a control RNA sample isolated at 24 h post-fertilization,

while lane 2 is an RNA sample from embryos that were dissociated at the

4-8-cell stage (approximately 34 h) and remained dissociated until har-

vest at 24 h post-fertilization. RNA samples in lanes 3-5 are from embryos

dissociated at the 46-cell stage and reassociated after 1, 2.5 and 8.5 h.

RNAs were isolated at 24 h post-fertilization which was approximately 12

h after reassociation of the 8.5 h dissociated cells. Lane 6 RNA is also from

cells that were reassociated at 8.5 h after dissociation, however, these cells

were cultured for 24 h post-reassociation, or approximately 36 h post-

fertilization. Lane 7 is the yeast control with a small amount of full length

probe remaining in this sample, while lane 8 is the undigested yeast control

showing the two probes.

respond to that signal is maintained, cells were kept dissociated for varying lengths of time prior to reassociation.

Embryos were dissociated at the 4-g-cell stage as described above, allowed to reassociate after 1, 2.5 or 8.5 h of disso-

ciation and cultured for an additional 12 h to allow for cell- contact mediated events to take place before RNA was isolated. Embryos were all from the same batch. The 1 and 2.5 h time points represent reassociation prior to the 64-cell

stage, while the 8.5 h time point represents reassociation after the 64-cell stage based on morphology of control embryos. Samples of each culture were observed carefully

under a microscope for the presence of clumping as an indication that cell-cell reassociation had occurred. Cell counts confirmed that dissociated cells divided at a rate similar to controls as has been reported by others (Hurley et al., 1989; Stephens et al., 1989) (data not shown). RNA was isolated and samples were probed for both Endold and Specl in RNase protection assays (Fig. 5). Control embryos

showed normal levels of End016 expression at 24 h (lane 1) and, as previously observed, dissociated cells did not express End016 (lane 2). Cells dissociated for 1 h and then reassociated expressed End016 mRNA when the RNA was isolated from embryos harvested 24 h post-fertilization (Fig. 5, lane 3; also see Fig. 4). Similarly, cells dissociated for 2.5 h and then reassociated and RNA isolated at 24 h post-fertilization also expressed Endol6, but at a

reduced level as was seen in the 1 h reassociated sample.

In contrast, after cultures were maintained as single cell

suspensions for 8.5 h before reassociation, we found that

End016 was not ex-pressed in cells harvested 12 h later which corresponded to 24 h post-fertilization (Fig. 5, lane 5). However, when these reassociated cells were maintained in culture for an additional 12 h (36 h post-fertilization) we

found that End016 was expressed (lane 6), although not to the level seen in the earlier reassociated embryos. Since

these reassociated cells that were maintained in culture for

a prolonged period express Endol6, they presumably regained the capacity to establish endoderm. The appropri-

ate pattern of expression of Specl was obtained in these RNA samples, confirming the viability of the cells and the

A.

B. 1 2 3 4 5

Fig. 6. Expression of End016 in embryos dissociated after the 16-cell

stage. Embryos were dissociated at the 32%64cell stage and RNA was

isolated at 4 and 16 h after dissociation. Control RNA was isolated from

intact embryos prior to dissociation at the 3264-cell stage and then at 12

and 24 h post-fertilization. Samples were probed for the long and short

forms of End016 mRNA (closed arrow) as well as Specl (open arrow). (A)

RNA samples probed for the longer form of Endol6 and Specl. Lane 1 is

control RNA from the 32-64-cell stage; lane 2 is RNA from intact control

embryos at 12 h; lane 3 RNA is from embryos dissociated at the 32-H-cell

stage and RNA isolated at 12 h; lane 4 is RNA from intact control embryos

at 24 h; and lane 5 is RNA from embryos dissociated at the 3264cell

stage and RNA isolated at 24 h. Arrows indicate protected fragments. (B)

Duplicate RNA samples probed for the smaller form of EndoI6; lanes as in

(A).


integrity of these RNA samples. These data suggested that in normal development an important interaction is lost somewhere after the 64-cell stage and that the signal and/

or competency required to specify the vegetal plate occurs only in a brief developmental window that extends at least

2.5 h after the 16-cell stage. These data demonstrated

further that early cell contact is essential for End016 expression to occur.

2.6. Expression of End016 in late dissociated embryos

To further analyze the window of signaling involved in

endoderm specification, the expression of End016 was examined in embryos dissociated after the 16-cell stage

and the formation of the micromeres. Embryos were dissociated at the 32-64-cell stage, after the proposed signaling from the micromeres had taken place (Ransick and David-

son, 1995). Control embryos were from the same batch and

were maintained intact. RNA was isolated 4 or 16 h after dissociation, corresponding to 12 or 24 h post-fertilization (Fig. 6). As expected, the longer forms of End016 mRNA

were not detected at 12 h but were seen at 24 h in control

intact embryos (Fig. 6A). However, in the dissociated embryos no End016 expression was observed. These results suggest that cell interactions beyond the 16-cell stage are needed for the maintenance of End016 expression. Specl expression was analyzed to insure the viability of the cells

and the integrity of the RNA samples and was found to follow the correct pattern of expression in all experimental

cases, Because the longer forms of End016 mRNA are not

normally detected until 18 h of development in intact embryos, it was possible that we might have missed an ear-

lier effect of cell dissociation on End016 expression. There- fore, we analyzed the expression of the smaller form of the message which is transcribed at around 10 h of development in normal intact embryos (Godin et al., 1996). In dissociated embryos we observed a lower level of the smaller message

relative to the controls in both the 12 and 24 h time points (Fig. 6B). Since the amount of End016 mRNA levels did not

increase beyond the amount of maternal n-RNA seen at dissociation at the 32-64-cell-stage (lane l), the most likely explanation is that no new transcription of End016 had taken place. However, if this expression does represent new transcription, then these data indicate that the change in splicing pattern did not occur in dissociated embryos as it

does in intact embryos since no longer transcripts were

detected (Fig. 6A).

2.7. Effect of lithium on End016 expression in dissociated embryos

Lithium is a well known vegetalizing agent in intact sea urchin embryos and has been shown to significantly increase the number of End016 expressing cells (Nocente-McGrath et al., 1991; Ransick et al., 1993). Furthermore, in embryos derived from isolated mesomere pairs treated with lithium

some cells will change fate and become endoderm (Living- ston and Wilt, 1989). These data suggest that lithium can replace the signal required for endoderm specification. To see if lithium could rescue endoderm specification and End016 expression in dissociated embryos, embryos were

dissociated at the 4-g-cell stage and immediately treated with 40 mM LiCl. Cells were cultured in the presence of

lithium for 12 h and then washed and placed into CFSW. A

sample of the intact control embryos was treated with 40 mM LiCl at the same time as the dissociated cells and the lithium was washed out 12 h later. Cells and embryos were cultured for an additional 24 h and RNA was isolated and

analyzed by RNase protection assays (Fig. 7). Intact embryos treated with lithium (Fig. 7, lane 2) showed an increase in the level of End016 n-RNA at 48 h of development relative to control embryos (Fig. 7, lane l), demon-

strating that these embryos were responding to lithium in the predicted manner. As we have previously shown, disso-

ciated cells did not express End016 (Fig. 7, lane 3), however, surprisingly, no End016 mRNA was detected in dissociated-lithium-treated cells (Fig. 7, lane 4). Specl expression was not affected by lithium treatment, but was

reduced in the dissociated cells as expected. These data demonstrate that lithium alone cannot replace the required signal in individual cells to specify vegetal plate and activate End016 expression and that cell-cell interactions are necessary for the previously observed effect of lithium.

C C D D

LiCI: - + - +

i

>

3

Fig. 7. Expression of End016 in dissociated-lithium treated embryos.

Embryos were dissociated at the 4-cell stage and cultured either in the

presence or absence of 40 mM LiCl for 12 h. RNA was isolated at 48 h

post-fertilization and analyzed for the expression of End016 (closed arrow)

and Specl (open arrow). C, control; D, dissociated. Lane 1 is RNA from

intact control embryos; lane 2 RNA is from intact control embryos treated

with lithium; lane 3 RNA is from dissociated cells; and lane 4 RNA is from

dissociated cells treated with LiCI. Arrows indicate protected fragments.

RE. Godin et al. /Mechanisms of Development 67 (1997) 3.5-47 43

3. Discussion

3.1. Endoderm speci’cation in developmentally arrested embryos

We have used expression of the End016 gene from two different species of sea urchins to examine vegetal plate specification in developmentally arrested embryos and in dissociated embryonic cells. We have shown that in p-

APN treated embryos, vegetal plate specification in both S. purpuratus and L. variegatus, as measured by End016 expression, does take place despite an obvious inhibition of gastrulation and lack of visible endoderm formation.

These results are in contrast to previous findings of /3- APN treated embryos, in which it was shown that other

endoderm-specific molecules, End01 and LvIVl.2, were not expressed in arrested embryos, which has been inter-

preted as a lack of presumptive endoderm formation (Ben- son et al., 1991; Wessel, 1993). The apparent discrepancy between these results can be reconciled by considering that End016 is expressed much earlier than the other markers examined and is a marker for specification of the endoderm, whereas Endol and LvN1.2, which are expressed later in development relative to Endol6, most likely represent mar-

kers for commitment and differentiation of the endoderm (Chen and Wessel, 1996). Since Endol expression was

determined at the protein level it is possible that the End01 mRNA was expressed in fi-APN treated embryos. The presence of the longer alternatively spliced forms of End016 mRNA in /3-APN treated embryos indicates that

not only transcription, but also correct processing of the transcript occurred. This finding suggests that developmen-

tal events that take place up until 18 h in normal embryos also occurred in these developmentally arrested embryos.

It has previously been shown that the two species of sea urchins used in this study display different responses to ECM disruption with P-APN in regards to ectoderm specific

gene expression as indicated by the Spec genes (Wessel et al., 1989). The fact that the disruption of collagen-maturation and inhibition of gastrulation did not affect End016 expression in either S. purpuratus or L. variegatus suggests that the variation seen within the Spec gene family is gene- specific and not a general phenomenon. A putative ECM response element (ERE) within the 5’ promoter region of the L. variegatus Spec gene (Seid et al., 1995) may account for the variation seen within the Spec gene family. Specifi- cation of the aboral ectoderm has been shown to vary between these two species (Wikramanayake et al., 199.5), however, our findings suggest that specification of the vegetal plate is the same.

3.2. Endoderm speci$cation in dissociated embryos

We found that endodenn specification, as indicated by End016 expression, did not take place in cultures of cells in which cell-cell contacts were completely disrupted at the

4-8 and at the 32-64 cell stage. In recent experiments,

Ransick and Davidson (1995) found that removal of micro-

meres resulted in a decrease in the number of End016 expressing cells, while the presence of just one micromere was sufficient to induce End016 expression to levels com-

parable to that of intact controls. However, complete removal of the micromeres from the intact embryo resulted in the expression of End016 in three out of four experiments. There are two possible interpretations of this result.

One is that this represents tissue autonomous expression of Endo16. The other, favored by Ransick and Davidson

(1995), is that an inductive signaling event specifying the endoderm must have taken place prior to removal of the

micromeres. Our results, demonstrating a lack of End016 expression in cells derived from blastomeres separated at

the 4-8-cell stage prior to the formation of the micromeres at the fourth cleavage (16cell stage) are in agreement with

the interpretation of Ransick and Davidson and support an inductive role for the micromeres in endoderm specification as originally proposed by Theodor Boveri in 1901 (reviewed in Ernst, 1997b). We observed relatively normal End016 expression in embryos that were allowed to reassociate after only 1 h of dissociation. Partially dissociated embryos, which had reduced cell contact, showed reduced levels of

End016 expression. Although the reassociated cells did not form recognizable embryos, the most likely explanation for

this result is that micromeres or micromere descendants were responsible for specifying End016 expression since micromeres have been shown to induce an ectopic vegetal

plate when transplanted throughout the embryo (HBrstadius, 1973; Khaner and Wilt, 1991; Amemiya, 1996).

3.3. Maintenance of endodemz specijication relies on cell- cell interactions

The exact timing and nature of the cell interactions needed to specify the endoderm are not known. Data from

this study, in which embryos dissociated at the 4-8-cell stage and reassociated after the @-cell stage did not activate

End016 expression, suggest that the interaction/signal that is required for endoderm specification is maintained for a finite time, perhaps equivalent to the period termed condi- tional specification (Davidson, 1989). Cell counts per-

formed on these late reassociated embryos showed that the cells had undergone a total of six or seven cleavages at the time of reassociation. Following the seventh cleavage

the micromeres are only in contact with one half of the daughter cells derived from the overlying macromeres (veg2 tier) which suggests that interactions needed to spe-

cify the endoderm most likely take place prior to the seventh cleavage.

Ransick and Davidson (1995) showed that intact embryos that had their micromeres removed at progressively later stages still expressed Endol6 at a significant level. How- ever, our data showed that End016 was not expressed in cells derived from embryos dissociated at the 32-64-cell


stage, a time after the presumed signaling event has taken place at the 16-cell stage. This observation suggests that after the 64-cell stage complete specification of the vegetal plate requires interaction between the presumptive vegetal plate cells themselves as previously proposed by Davidson (1989). Our data support the idea that once cells are con-

ditionally specified by the micromeres, the vegetal plate cells undergo intra-territorial signaling that is also cell-

cell contact dependent to maintain the specification and

continued development of the endoderm. Although our experiments do not differentiate between signaling from

the micromeres or lateral signaling between the presumptive vegetal plate cells, they demonstrate a need for cells to be in contact with each other for End016 expression to take place.

No End016 mRNA was detected when embryos were dissociated at the 4-8-cell stage, allowed to reassociate 8.5 h later and RNA isolated 12 h after reassociation. How-

ever, a sample of these cells left reassociated for an additional 12 h (24 h reassociated, 36 h post-fertilization) transcribed and accumulated clearly detectable levels of

End016 mRNA. Previous investigators have observed that if two or more pairs of mesomeres isolated from 16-ceil

stage embryos are maintained in association in culture for

approximately 3 days, the resulting embryoids will start to display evidence of gut differentiation (Khaner and Wilt, 1990). Additionally, when McClay and Logan (1996) sur- gically removed the entire gut from mid-gastrula stage embryos, these embryos were able to organize a completely new digestive system. It is likely that expression of End016 observed in embryos reassociated for 24 h represents similar regulative events and does not represent the normal endoderm specification that occurs early in sea urchin embryo-

genesis.

3.4. Lithium and endodenn specification in dissociated embryos

The nature of the signal that mediates endoderm specification is still unknown. Lithium is known to increase the number of endoderm expressing cells in intact embryos (Nocente-McGrath et al., 1991; Ransick et al., 1993) and lithium sensitive modules have been identified in the promoter region of End016 (Yuh and Davidson, 1996). How- ever, results presented here demonstrate that cell contact is necessary for lithium’s effect since lithium can not activate

End016 expression in dissociated cells. Recently, Ghiglione et al. (1996) have shown that sea urchin embryos have a graded response to the effect of lithium along the A-V axis. In addition, their study demonstrated that the effect on the spatial expression of the hatching enzyme seen with lithium treatment was different from that seen with ectopically transplanted micromeres. The inositol depletion model of lithium’s action suggests that a signal from the micromeres would function through a G-protein linked receptor activated through cell-cell contact (Berridge et al., 1989). Pre-

sumably receptors in dissociated cells would not be activated due to a lack of cell-cell contact, while in the intact embryo, cells are in direct contact and receive posi- tional information from each other and can either respond to the lithium signal (vegetal cells) or not (animal cells).

Based on the inositol depletion model mentioned above (Berridge et al., 1989), vegetal plate specification is thought

to require the activation of a-G-protein coupled receptor and a subsequent decrease in protein kinase C (PKC) activity

through the IPs second messenger pathway. However, treat-

ment with phorbol esters, such as TPA, which increase the levels of PKC, gave the same phenotype as lithium treatment (Livingston and Wilt, 1992). A new model for the mode of action of lithium has recently been proposed.

Klein and Melton (1996) demonstrated that in Xenopus lithium directly decreased the activity of GSK-3P, a kinase involved in the Wnt signaling pathway. It has also been shown that increased levels of PKC resulted in decreased activity of GSK-3/3 (Cook et al., 1996). Taken together, these results suggest that treatment with lithium, phorbol esters or the activation of the Wnt signaling pathway will

result in decreased activity of GSK-3@ and consequently an increase in the amount of endoderm. We suggest that this

decrease in GSK-3/3 activity may be involved in endoderm

specification in the sea urchin embryo.

4. Experimental procedures

4.1. Animals

Adult Strongylocentrotus purpuratus were purchased from Marinus (Long Beach, CA) and Lytechinus variegatus were purchased from Susan Decker (Hollywood, FL). Spawning, fertilization and culturing of embryos were as

previously described (Pittman and Ernst, 1984), with the exception that L. variegatus were grown at 25°C.

4.2. Embryo treatments

4.2.1. p-APN The inhibitor of lysl-oxidase, /3aminoproprionitrile (p-

APN), was used to inhibit collagen assembly. Eggs were fertilized by standard procedures and when greater than 95% fertilization was achieved a stock solution of P-APN was added to a final concentration of 0.5 mM. Embryos were cultured in the &APN until they were harvested.

4.2.2. Lithium chloride Intact embryos or dissociated cells were cultured in 40

mM LiCl as previously described (Nocente-McGrath et al.,

1991).

4.3, RNA isolation

Total RNA was isolated from embryos at selected stages


of development from egg to pluteus using the RNazol pro-

cedure (Tel-test). RNA was quantified by spectrophotome-

try and checked for integrity by ethidium bromide staining

on formaldehyde agarose gels.

4.4. Probes

Gary Wessel). Screening and sequencing were carried out

as previously described (Soltysik-Espailola et al., 1994).

Clone 7, a 1.8 kb cDNA clone was identified and used for

all subsequent experiments. Sequence analysis was performed using the GCG program from the University of

Wisconsin.

4.4.1. RNase protection 4.7. Northern blots

To make RNA antisense probes specific for the longer End016 messages in S. purpuratus, clone 29B (Godin et al.,

1996) was linearized with Xmnl which cuts within the insert to make an in vitro transcription product of 600 nt. This gives protected fragments of 520 and 420 bp corresponding to the longer messages (Godin et al., 1996). To detect the

short form of End016 mRNA, clone 4A was linearized with Xbal to produce a transcription product of 330 nt which

gives a protected fragment of 250 bp. LvEndo16 was linearized with HincZZ to make an in vitro transcription product

of 610 nt giving a protected fragment of 550 bp. Plasmid

containing the Speclb clone (gift of Drs Gary Wessel and William Klein) was linearized with Xbal to make an in vitro

transcription product of 350 nt and a protected fragment of

280 bp. LvSl (gift of Dr Gary Wessel) linearized with XbaZ, resulted in an in vitro transcription product of 590 nt and a

protected fragment of 500 bp. Template DNA was checked on a gel to confirm complete digestion. The DNA was run

through a Wizard miniprep column (Promega) and eluted with 25 ~1 of TE. In vitro transcription was carried out using

the Promega Riboprobe Kit with 50 Z.&i of 32P CTP (New

England Nuclear). Probes were transcribed at 37°C for 1.5

h. All other procedures were carried out following the manufacturer’s protocol.

RNA (10 pg) was denatured in one volume of denatura-

tion buffer (30% formamide, 8% formaldehyde, 0.6x

MOPS) and separated by electrophoresis on a 1% formaldehyde agarose gel for 5 h at 80 V (modified from Sambrook et al., 1989). RNA was transferred overnight to a nitrocellulose

membrane in 20x SSC (3 M NaCI, 3 M Na Citrate). The RNA was crosslinked to the membrane using a Stratalinker

UV crosslinker (1.2 x lo5 pJ). The membrane was prehy-

bridized in 0.4 M NaP04 (pH 6.5), 0.1% SDS and 5x Den-

hardt’s solution followed by overnight hybridization at 65°C to a 32P-labeled cDNA probe. After hybridization,

the membrane was washed three times at 65°C for 30 min

in 0.4 M NaP04 (pH 6.5), 0.1% SDS.

4.8. In situ hybridization

4.4.2. In situ hybridization The same probe that was used to detect LvEndol6 in

RNase protection assays was used for the in situ hybridization procedure.

Embryos were fixed overnight in 2.5% glutaraldehyde at

4°C following the protocol of Harkey et al. (1992) with a

slight modification (Godin et al., 1996). In situ hybridiza-

tions were carried out as described by Ransick et al. (1993). Briefly, antisense RNA probes were made using digoxy-

genin-labeled UTP nucleotides (Boehringer-Mannheim) in

the Ambion Sp6 Maxiscript kit. Probes were quantified and used at 0.02-0.2 ng/pl in the in situ hybridization procedure

and developed with NBT (nitro blue tetrazolium) and BCIP (5-bromo-chloro-3-indolyl-phosphate) (Promega).

4.9. Cell dissociation experiments

4.4.3. Northern blots L. variegatus clone 7 (see below) was used as a template

to make radioactive probes according to the manufacturer’s directions using the Random Prime kit from BRL.

4.5. RNase protection assays

RNase protection assays were performed following the

specifications of the Ambion RPAII kit. Only RNase Tl (1:25) was used for the digestions. Experiments were carried out in probe excess using 6 x lo4 cts/reaction and 10 pg of

RNA was used in all experiments. Samples were routinely probed for 18s as an internal loading control.

4.6. Cloning and sequencing

L. variegatus End016 clones were isolated from an L. variegatus Zap prism stage cDNA library (gift of Dr

Eggs were fertilized in sea water in the presence of 10 mM Tris (pH 7 6) and 5 mM PABA (p-aminobenzoic acid)

to soften the fertilization membrane (Hall, 1978). Fertiliza-

tion membranes were removed by two passages of the fer-

tilized eggs through 47 PM Nitex mesh. The fertilized eggs were settled and washed with fresh, filtered sea water and cultured at 14°C until the 4-g-cell stage (3-4 h) or 32-64- cell stage (8 h). Blastomeres were dissociated by gently pelleting embryos and by resuspension in calcium-magne-

sium free sea water containing 1 mM EGTA, twice. Cells were passed twice through 47 PM Nitex mesh to facilitate complete dissociation of the blastomeres. Dissociated blastomeres were gently pelleted and resuspended in calcium- free sea water containing 1 mM EGTA to a final concentration of l-5 x lo4 cells/ml. To produce complete dissociation, blastomeres were cultured in a flask at 14°C and

shaken at 100 rev./min. Blastomeres that were cultured under gentle conditions which resulted in partial cell-cell


adhesion were stirred with a paddle at 60 rev./mm at 14°C. To allow cells to reassociate, samples were taken from the flask of completely dissociated cells, pelleted gently and washed once with MPFSW and then resuspended in MPFSW and cultured with no stirring at 14°C. Reassocia- tion was monitored by microscopy. RNA was isolated from each culture as described above.

Acknowledgements

We would like to thank Dr Gary Wessel and Dr William Klein for kindly providing us with the Speclb and LvSl clones, respectively. We also thank Dr Gary Wessel for providing us with the L. variegatus cDNA library and Lavi- nia Gangemi for screening the library and identification of the original clone. The authors are also extremely grateful to our colleague Ana Egafia for critical reading of the manu- script and insightful thoughts and discussions during the progress of this work. This work was funded by a Mellon Research Award to S.G.E. W.A.P. was supported by a Hughes summer research fellowship for undergraduates.

References

Amemiya, S. (1996) Complete regulation of development throughout

Butler, E., Hardin, 3. and Benson, S. (1987) The role of lysl oxidase and

metamorphosis of sea urchin embryos devoid of macromeres. Dev.

Growth Diff. 38, 465-476.

collagen crosslinking during sea urchin development. Exp. Cell Res.

Benson, S., Rawson, R., Killian, C. and Wilt, F. (1991) The role of the

extracellular matrix and tissue specific gene expression in the sea

173, 174-182.

urchin. Mol. Rep. Dev. 29, 220-226.

Berridge, M.J., Downes, C.P. and Hanley, M.R. (1989) Neural and devel-

opmental actions of lithium: a unifying hypothesis. Cell 59,411-419.

Burdsal, C.A., Alliegro, M.C. and McClay, D.R. (1991) Tissue-specific,

temporal changes in cell adhesion to echinonectin in the sea urchin

embryo. Dev. Biol. 144, 327-334.

Burke, R.D., Myers, R.L., Sexton, T.L. and Jackson, C. (1991) Cell move-

ments during the initial chase of gastrulation in the sea urchin embryo.

Dev. Biol. 146, 542-557.

Chen, S.W. and Wessel, G.M. (1996) Endoderm differentiation in vitro identifies a transitional period for endoderm ontogeny in the sea urchin

embryo. Dev. Biol. 175, 57-65.

Cook, D., Fry, M.J., Hughes, K., Sumathipala, R., Woodgett, J.R. and

Dale, T.C. (1996) Wingless inactivates glycogen synthase kinase-3

via an intracellular signaling pathway which involves a protein kinase

C. EMBO J. 15.4526-4536.

Davidson, E.H. (1989) Lineage-specific gene expression and the regulative

capacities of the sea urchin embryo: a proposed mechanism. Develop-

ment 105,421~445.

Davidson, E.H. (1990) How embryos work: a comparative view of diverse

modes of cell fate specification. Development 108, 365-389.

Davidson, E.H. (1991) Spatial mechanisms of gene regulation in metazoan

embryos. Development 113, 1-26. Ding, D. and Lipshitz, H.D. (1993) Localized RNAs and their function.

Bioessays 15, 651-658.

Ernst, S.G. (1997a) Recent advances in early development of the sea

urchin embryo. In Collier, J.P. led.), Reproductive Biology of the Inver-

tebrates; Progress in Developmental Biology, Oxford and IBH Publish-

ing, New Delhi, in press.

Ernst, S.G. (1997b) A century of sea urchin development. Am. Zool. 37,

250-259.

Ghiglione, C., Emily-Fenouil, F., Chang, P. and Gache, C. (1996) Early

gene expression along the animal-vegetal axis in sea urchin embryoids

and grafted embryos. Development 122, 3067-3074.

Godin, R.E., Urry, L.A. and Ernst, S.G. (1996) Alternative splicing of the

End016 transcript produces differentially expressed mRNAs during sea

urchin development. Dev. Biol. 179, 148-159.

Hall, H.G. (1978) Hardening of the sea urchin fertilization envelope by

peroxidase-catalyzed phenolic coupling of tyrosine. Cell 15, 343-355.

Harkey, M.A., Whiteley, H.R. and Wbiteley, A.H. (1992) Differential

expression of the mspl30 gene among skeletal lineage cells in the sea

urchin embryo: a three dimensional in situ hybridization analysis. Mech.

Dev. 37, 173-184.

Henry, J.J., Amemiya, S., Wray, G.A. and Raff, R.A. (1989) Early induc-

tive interactions are involved in restricting cell fates of mesomeres in

sea urchin embryos. Dev. Biol. 136, 140-153.

Hurley, D.L., Angerer, L.M. and Angerer, R.C. (1989) Altered expression

of spatially regulated embryonic genes in the progeny of separated sea

urchin blastomeres. Development 106, 567-579.

Horstadius, S. (1973) Experimental Embryology of Echinoderms, Claren-

don Press, Oxford.

Kessler, D.S. and Melton, D.A. (1994) Vertebrate embryonic induction:

mesodetmal and neural patterning. Science 226, 596-604.

Khaner, 0. and Wilt, F. (1990) The influence of cell interactions and tissue

mass on differentiation of sea urchin mesomeres. Development 109,

625-634.

Khaner, 0. and Wilt, F. (1991) Interactions of different vegetal cells with

mesomeres during early stages of sea urchin development. Development

112, 881-890.

McClay, D.R. and Logan, C.Y. (1996) Regulative capacity of the arch-

entcron during gastrulation in the sea urchin. Development 122, 607-

616.

Klein, P.S. and Melton, D.A. (1996) A molecular mechanism for the effect

of lithium on development. Proc. Natl. Acad. Sci. USA 93, 8455-8459.

Klein, W.H., Mao, C.-A., Gan, L., Chuang, C.-K. and Wikramanayake,

A.H. (1995) Manipulating cell fates in the sea urchin embryo. In Inver-

tebrate Reproduction and Development, 7th ICIR Proceedings.

Livingston, B.T. and Wilt, F.H. (1989) Lithium evokes expression of

vegetal-specific molecules in the animal blastomeres of sea urchin

embryos. Proc. Natl. Acad Sci. USA 86, 3669-3673.

Livingston, B.T. and Wilt, F.H. (1992) Phorbol esters alter cell fate during

development of sea urchin embryos. J. Cell Biol. 119, 1641-1648.

Mao, C.A., Wikramanayake, A.H., Gan, L., Chuang, C.K., Summers, R.G.

and Klein, W.H. (1996) Altering cell fates in sea urchin embryos by

overexpressing SpOtx, an ortbodenticle-related protein. Development

122, 1489-1498.

McCoon, P.E., Angerer, R.C. and Angerer, L.M. (1996) SpFGFR, a new

member of the fibroblast growth factor receptor family, is developmen-

tally regulated during early sea urchin development. J. Biol. Chem. 27 1,

20119-20125.

Nocente-McGrath, C., Brenner, C.A. and Ernst, S.G. (1989) Endo16, a

lineage-specific protein of the sea urchin embryo, is first expressed

just prior to gastrulation. Dev. Biol. 136, 264-272.

Nocente-McGratb, C., McIsaac, R. and Ernst, S.G. (1991) Altered cell fate

in LiCl-treated sea urchin embryos. Dev. Biol. 136, 264-273.

Pittman, D. and Ernst, S.G. (1984) Developmental time, cell lineage, and

environment regulate the newly synthesized proteins in sea urchin

embryos. Dev. Biol. 106, 236-242.

Ransick, A. and Davidson, E.H. (1993) A complete second gut induced by

transplanted micromeres in the sea urchin embryo. Science 259, 1134-

1138

Ransick, A. and Davidson, E.H. (1995) Micromeres are required for nor-

mal vegetal plate specification in sea urchin embryos. Development 12 1,

32 15-3222.


Ransick, A., Ernst, S.G., B&ten, R. and Davidson, E. (1993) Whole mount

in situ hybridization shows End026 to be a marker for vegetal plate

territory in sea urchin embryos. Mech. Dev. 42, 117-124.

Sambrook, .I., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A

Laboratory Manual, 2nd edn., Cold Spring Harbor Laboratory Press,

Cold Spring Harbor, NY.

Seid, C.A., Flyntzanis, C.N., George, J.M. and Tomlinson, CR. (1995) An

extracellular matrix response element in the promoter of the LpSl of the

sea urchin Lytechinus binds a G-string protein. Mol. Cell. Biol. 6

(Suppl.), 4A.

Solrush, M. and Katow, H. (1982) Initial characterization of sulfated

macromolecules in the blastocoels of mesenchyme blastula of Strongy-

locentrotus purpuratus and Lytechinus pictus. Dev. Biol. 94, 326-

336.

Soltysik-Esptiola, M., Klinzing, D.C., Pfarr, K., Burke, R.D. and Ernst,

S.G. (1994) Endol6, a large multidomain protein found on the surface

and ECM of endodermal cells during sea urchin gastrulation, binds

calcium. Dev. Biol. 165, 73-85.

Stephens, L., Kitajima, T. and Wilt, F. (1989) Autonomous expression of

tissue-specific genes in dissociated sea urchin embryos. Development

107, 299-307.

Wessel, G.M. (1993) Differentiation of the endodermal epithelium during

gastrulation in the sea urchin embryo. In Molecular Basis of Morpho-

genesis; 51st Symposium of the Society of Developmental Biology,

Wiley-Liss, New York.

Wessel, G.M. and McClay, D.R. (1987) Gastrulation in the sea urchin

embryo requires the deposition of collagen in the extracellular matrix.

Dev. Biol. 121, 149-165.

Wessel, G.M., Zhang, W., Tomlinson, C.R., Lennarz, W.J. and Klein,

W.H. (1989) Transcription of the Speck-like gene of Lytechinus is selec-

tively inhibited in response to disruption of the extracellular matrix.

Development 106, 355-365.

Wikramanayake, A.H., Brandhorst, B.P. and Klein, W.H. (1995) Autono-

mous and non-autonomous differentiation of ectoderm in different sea

urchin species. Development 121, 1497- 1505.

Yuh, C.-H. and Davidson, E.H. (1996) Modular cis-regulatory organiza-

tion of Endol6, a gut-specific gene of the sea urchin embryo. Develop-

ment 122, 1069-1082.

Yuh, C.-H., Ransick, A.. Martinez, P., B&ten, R.J. and Davidson, E.H.

(1994) Complexity and organization of DNA-protein interactions in the

5’ regulatory region of an endoderm-specific marker gene in the sea

urchin embryo. Mech. Dev. 47, 165-186.

specification of endoderm in the sea urchin embryo

Documents