A twist in sea urchin gastrulation and mesoderm specification

WeiWeng*.Jan Cheethamt.Jeff Hardint and Judith M.Venuti"':t: *Department of Anatomy and Cell Biology, College of Physicians & Surgeons of Columbia University, 630 W. I 68th Street, New York, NY 10032, U.S.A. and

tDepartment of Zoology, University of Wisconsin, Madison, WI 53706, U.S.A.

Abstract

The bHLH (basic helix-loop-helix) transcription factor, sea urchin myogenic factor-1 (SUM-1), plays an important role in myogenic determination during sea urchin embryogenesis. SUM-1-mediated transactivation is restricted to the mesenchyme lineages in transgenic sea urchin embryos, suggesting that other factors, either positive or negative, influence the activity of SUM-1 in different embryonic cell types. While post-translational regulation of vertebrate myogenic factors has been suggested from in vitro studies, it has never been demonstrated in vivo. The most compelling in vitro experiments have shown that the mesodermal bHLH, twist, negatively regulates myogenic bHLHs. However, in the vertebrate embryo, twist and myogenic bHLHs are not expressed coincidentally, and different concentrations of twist playa role in the differentiation of different muscle lineages (somatic versus visceral) in Drosophila embryos. The gene expression studies in vertebrates and the genetic experiments in Drosophila suggest disparate roles for twist in these organisms. To gain a better understanding of the role of twist in mesodermal and myogenic specification, we cloned a sea urchin twist homologue and characterized its role in gastrulation and myogenesis in this simple embryo. Our data suggest that twist from Lytechinus variegatus functions after gastrulation and initial specification of the embryonic mesoderm of the sea urchin.

Introduction

The experiments of Horstadius in the first half of this century highlighted the necessity for cell-cell interactions in the development of different cell types of the sea urchin embryo [1). His experiments established a role for cell--cell signalling in the specification of cell fates in the sea urchin . Horstadius showed that the micromeres influenced the differentiation of other embryonic cells, setting the stage for present day experiments which have begun to unravel the molecular pathways involved in micromere signalling and the specification of other embryonic lineages. It is now thought that the derivatives of the micromeres (the skeletogenic or primary mesenchyme cells) are autonomously specified during the

tTo whom correspondence should be addressed. 153

I

I S4 W.Weng et al.

early cleavage stages. However, it is still not clear when and how the different

secondary mesenchymal lineages are specified during sea urchin embryonic

development [2]. Cell-lineage analysis of the vegetal plate of the mesenchyme blastula­

stage embryo suggested that lineages that arise from this region, the endoderm and

secondary mesenchyme cells, are specified prior to the initiation of gastrulation

[3]. Cell dissociation experiments have substantiated this for the endoderm [4J,

but when and how the different secondary mesenchyme lineages are specified is

uncertam.

The only secondary mesenchyme lineage for which we have sufficient

molecular information concerning its specification is the myogenic lineage, which

is specified during early to mid-gastrulation through a programme involving a sea

urchin homologue of the vertebrate MyoD family, SUM-l [5,6]. This

transcription factor was cloned from Lytechinus va riegatus and shown to be

expressed in the presumptive muscle cells before overt myogenic differentiation

takes place towards the end of gastrulation [5,6]. When we examined the activity

of SUM-l in different embryonic lineages we found that the transactivational

activity of SUM-l was restricted to the mesenchymal lineages (results not shown) .

Further analysis of SUM -1 and its role in the specification of the myogenic lineage

has led us to examine other factors which might playa role in mesoderm specifi­

cation in the sea urchin embryo.

Studies of mesodermal development in Drosophila have identified a

gene, twist, which is essential for the earliest formation of the mesoderm in the fly

[7]. Twist, like SUM-I , is a member of the HLH family of transcrip tional

regulators which are important in a variety of determinative events during

development. Twist homologues have now been identified in the human [8], mouse [9], frog [10], lancelet [11J, Drosophila virilis [12], Tribolium [13] leech [14]

and Caenorhabditis elegans [15] . Drosophila twist homozygous-null embryos fail to make mesoderm [16],

suggesting a role in the earliest specification of this tissue. However, initial analysis

of mouse embryo twist mutants suggests that murine twist functions after gastru­

lation [17]. Whereas twist from Drosophiu. was shown in vivo to act as a transcrip­

tional activator for early mesoderm-specific genes, in vitro analysis of murine twist

demonstrated that it can act as a repressor of myogenic genes [18]. Gene expression

studies in vertebrates [19] and genetic experiments in Drosophila [20] suggest

disparate roles for twist in these different organisms. We therefore wish to use our

transactivation assay as a functional test to better understand the role of twist in

mesodermal and myogenic specification. To begin, we have identified a sea urchin

twist homologue and characterized its role in gastrulation and myogenesi s.

Surprisingly, higher levels of Lv -twist (twist from L. variegatus) transcripts are

expressed following the initiation of gastrulation, suggesting that the primary

function of sea urchin twist occurs after the early specification of the mesoderm.

Materials an

Sea urchin er Lytechinus vari U.S.A.) or Beau

variegatus embl

[6,7].

Molecular cI(J Two degenerat(

(G/A/T/C)ATG

(T/C)TT(G/AfT used to PCR-am

stage endomeso(

University, Pro"

of each primer v

KCI, 100 jJ.g/ml

(Fisher Scientific

(30 s at 94°C and

extension at 72°1 itated with ethar

with SalI, gel pur

U.S .A.). Clones 1

and were analyse

Reverse tram RNA L. variegatus e~ Millipore-filterel

using Tri-reagem

30 min of DNa:

(25 :24:1, by vol.

RNA was anneal

TTGATCTTCT

carried out using

(Gibco, BRL). ,

amplification wi!

CCACAATATI

with 30 cycles at

Immunocytol Embryos were s

minimum of 20 r

temperature WI

phosphate buffe

Embryos fixed i

(PBST); subseql

'I!lI.................

,w- the different . n embryonic

yme blastula­endoderm and

of gastrulation eendoderm [4J, are specified is

' have sufficient lineage, which

einvolving a sea -1 [5,6]. This

td shown to be I differentiation

r ed the activity ransactivational ults not shown).

0I . 1Iyogemc meage soderm specifi­

lve identified a [)derm in the fl y transcriptional events during

the human [8J , 'J [13] leech [14]

mesoderm [16J, " initial analysis ,ns after gas tru ­:t as a transcrip­of murine twist ;ene expression

14 [20J suggest wish to use our role of twist in ied a sea urchin d myogenesiso transcrtpts are at the primary ~ mesoderm.

Sea urchin gastrulation and mesoderm specification 155

Materials and methods

Sea urchin embryo culture Lytechinus variegatus adults were obtained from Susan Decker (Davie, FL, U.S.A.) or Beaufort Biologicals (Duke Marine Lab, Beaufort, NC, U.S.A.). L. variegatus embryos were cultured by stirring at 17°C as described previously [ 6,7].

Molecular cloning of the bHLH domain of the Lv-twist Two degenerate primers [primer 1: 5'-CCCTCGAG(C/A)G(G/A/T/ C)GT (G /A/T/C)ATGGC (GIAlT/C)AA(T/C)GT-3' and primer 2: 5'-CCGTCGAC (TIC)TI(GIAffIC)A(GIA)(GIAffIC)GT(T/C)TG(GIAff)AT(T/C)TI-3'J were used to PCR-amplify a twist-specific bHLH domain from an L. variegatus prism stage endomesoderm enriched eDNA library (provided by Gary Wessel, Brown University, Providence, RI, U.S.A.). A 1 f.LI portion of eDNA library and 0.5 f.Lg of each primer were mixed in 20 mM Tris-HCI (pH 8.3), 2 mM MgCI2, 25 mM KCI, 100 f.Lg/ml gelatin, 50 mM dNTP and 5 units of Taq DNA Polymerase (Fisher Scientific, Pittsburgh, PA, U.s.A.) in a 50 f.L1 reaction volume. Forty cycles (30 s at 94°C and 3 min at 50°C) of PCR reaction was followed by a single 8 min extension at 72°e. PCR reaction products were extracted with phenol, precip­itated with ethanol and resuspended in deionized H 20 . The DNA was digested with Sail, gel purified and ligated into pBluescriptIIKS+ (Strata gene, La Jolla, CA, U.S.A.). Clones were sequenced using the dideoxy chain-termination method [21J and were analysed with the BLAST search engine (NCB I).

Reverse transcriptase-PCR (RT-PCR) analyses of embryonic RNA L. variegatus embryos were cultured to appropriate developmental stages in Millipore-filtered artificial sea water (MFASW) at 17°e. Total RNA was prepared using Tri-reagent (Molecular Research Center, Inc., Cincinnati, OH), followed by 30 min of DNase I treatment at 37°C, phenollchloroform/iso-amyl alcohol (25:24:1, by vol.) extraction and ethanol precipitation. A 10 f.Lg portion of total RNA was annealed with 5 pg of a sequence-specific oligonucleotide (5' -ATCTI­TTGATCTICTICGTCACTC-3 ' ) at 65°C for 2 min. Reverse transcription was carried out using SuperScript II RTase, following the manufacturer's instructions (Gibco, BRL). About 2 f.LI of the 20 f.LI total RT reaction was used for PCR amplification with the reverse primer mentioned above and a forward primer (5'­CCACAATATITCGAAGAACTICAG-3'). PCR amplification was performed with 30 cycles at 94°C for 2 min, 55°C at 1 min and 72°C at 1.5 min.

Immunocytochemistry Embryos were stained as whole mounts after fixation in - 20°C methanol for a minimum of 20 min, or in 3.7% paraformaldehyde in MFASW for 30 min at room temperature with similar results. Methanol-fixed embryos were w;lshed in phosphate buffered saline (PBS) and incubated with antibodies diluted in PBS. Embryos fixed in paraformaldehyde were washed in PBS with 0.1 % Tween 20 (PBST); subsequent incubations and washes were performed in PBST. Primary

I S6 W.Weng et al.

antibodies were incubated for 1 h to overnight, followed by 3 X 5 min washes. Secondary antibodies were incubated for a minimum of 30 min, followed by 3 X 5 min washes . Embryos were mounted in PBS/glycerol (50:50, v/v) and viewed with a Zeiss Axiovert 100 Epifluorescence microscope. Embryos were photographed with an Olympus OM-1 35 mm camera using Kodak 400 ASA Gold print film , or a Dage video camera attached to a Power Computing computer equipped with Scion imaging software.

Antibodies Anti-(Drosophila twist) polyclonal antibodies were obtained from Bruce Patterson (NCI, NIH, Bethesda, MD, U .S.A.) and Han Nyugen (Albert Einstein College of Medicine, Bronx, NY, U .S.A.) and used at 1:200 dilution . Goat anti­mouse FITC secondary antibody was diluted 1:1 00 in PBS (Cappel, ICN Pharmaceuticals, Inc., Costa Mesa, CA, U .S.A.).

Results and discussion

Isolation of a twist homologue in L variegatus To isolate a twist homologue from L. variegatus, degenerate primers were designed to two conserved regions of the Drosophila twist bHLH domain (Figure 1 A)

Figure 1 A. 5' -<X:C'l.'CGIIGC~-3 '

A C--. Prbnet ... J

Dm-Twist QRVMANVIIBRQRTQSLIIDAFKS LQQ lIP- -- -'rLP--- --sDXLS:.t QTLKLII!l'R YID FL

3 ' -=AAG'l"l"l'GIIGAIII'rl".l'CAGC'I.'OC C­5'

T G C A C

B. -.. Primcr-l ....1c "'liz I LOOP B.JJJr u

Lv-Twist QRVLARVRBRQRTQSLRDAFARLRKIIP----'rLP-----sDKLSKIQTLK

M-'l'IfiBt QRVMAHVRERQR~SLREAFAaLRKIIP----TLP-----SDXL8KIQTLK 92\ R-'l'Ifist QRVMAll\lRERQ~SLll£All'AaLRKI IP- --- TLP --- --SDXLSItI QTLK 92\ X-'l'IfiBt QRVMAHVRERQ~SLREAPSSLRKIIP----TLP-----SDKLSKIQTLK 92\ Rro-Twist QRVLARVRBRQRTQSLRDAFPQLRKI VP--- -'rLP-- ---SDKLSKIQTLK 92\ Bb-Twist QRVLARVRBR -RTQSIMEAFSSLRKI IP- ---TU' --- --SDKLSKIQTLK 90\ Dv-Twist QRVMAll\lRERQ~SLRDAFKAI.QQIIP- --- TLP --- --SDXLSKIQTLK 88\ Dm-Twist QRVMAll\lRERQ~SLIIDAFKS IIP----TLP-----SDKLSKIQTLK 88\ Ce-twi8t QRACARRRBRQRTKELRDAFTLLRKLIP----SMP-----SDKMSKIHTLR 65\.-­

Primer-2

PCR cloning ofthe Lv-twist bHLH domain

(A) Degenerate oligonucleatide primers to conserved regions o( the Drosophila twist bHlH were used to

ampli(y lv-twist (rom a prism-stage sea urchin mesendoderm-enriched eDNA library. These

oligonucleotide primers recognize residues in the bHlH that are specific to twist (B) Comparison o( the

lv-twist bHLH sequence with those o(the murine (M-Twist), human (H-twist), Xenopus laevis (X-twist),

leech (Hro-Twist),lancelet (Bb-Twist), D. virilis (Dv-twist), D. melanogaster (Om-twist), and C. elegans (Ce­

twist) twist sequences shows the twist bHlH is highly conserved amongst these different species.

RT-PC

L. variegah embryon

stage

RT-PCR analysis RNA isolated from dij

primers (rom those I

gastrulation has cea

mesoderm. Abbreviati(

M-Gast, early and mi

control.

and used to am mesoderm-enric 123 bp fragment species (Figure 1 observed for the' to be most similal addition, high sin bHLH proteins, eHAND [25].

Temporal exp RNA from diffe when Lv-twist t Lv-twist are pre: reappear during: The zygotic expr the completion subsequent to the role in later devel,

Spatial expres Immunocytoche antibodies gener; localization of tw is spatially restrict potential role in t!

Summary

The sea urchin tv.

of other species, dimerization mo

i l r ll....~~........~

--- -

5 min washes. owed by 3 X 5

od viewed with •photographed d print film, or ~quipped with

d from Bruce ~lbert Einstein ion. Goat anti­,(Cappel, ICN

5 were designed in (Figure 1 A)

~YIDFL

~_r-2

~c-

!LK

!Lit 92' !LIt 92\ !LIt 92\ ~ 92' !LIt 90' !Lit 88\ !LK 88' lUI. 65'

liLH were used to

IIA library. These

Comporison of the

us laevis (X-twist),

WId C. elegans (Ce­

• species.

Sea urchin gastrulation and mesoderm specification 157

Figure 2 RT-PCR

L. variegatus HB E l E M l l' Pr +­embryonic MB Gast Coni

stages

RT-PCR analysis of Lv-twist expression RNA isolated from different-stage L variegatus embryos was analysed by RT-PCR using a different set of primers from those used to clone the bHLH. The highest levels of twist transcripts are found after

gastrulation has ceased. This suggests that Lv-twist functions after the initial specification of the

mesoderm. Abbreviations:HB, hatching blastula; £- and L-MB, early and late mesenchyme blastula; £- and

M-Gast, earfy and mid-stage gastrula; L-Gast and L' -Gast, two separate late gastrula; Pr, prism; Cont,

control.

and used to amplify a -123 bp DNA fragment from a late gastrula-stage mesoderm-enriched cDNA library. The deduced amino acid sequence of the 123 bp fragment was compared with the bHLH regions of twists from other species (Figure 1B). An average of 87% identity and over 90% similarity was observed for the bHLH regions of twists from other species . Lv-twist was found to be most similar to the bHLH regions of Xenopus, mouse and ascidian twists. In addition, high similarities were shared with other vertebrate mesodermal-specific bHLH proteins, including Dermo-1 [22], Paraxis/Meso1 [23], Scleraxis [24] and eHAND [25].

Temporal expression pattern of Lv-twist RNA from different embryonic stages was analysed by RT-PCR to determine when Lv-twist transcripts are expressed during development. Transcripts of Lv-twist are present at low levels at the blastula stage but then disappear and reappear during gastrulation, with highest levels at the pluteus stage (Figure 2). The zygotic expression pattern of Lv-twist transcripts reveals highest levels after the completion of gastrulation. These data suggest that Lv-twist functions subsequent to the establishment of the different germ-layers and has an important role in later developmental events.

SpatiaJ expression of Lv-twist Immunocytochemical analysis of Lv-twist expression employed polyclonal antibodies generated to the Drosophila twist protein. The immunocytochemical localization of twist in developing sea urchin embryos suggests that most Lv-twist is spatially restricted to the mesendoderm of the gastrula stage embryo, indicating a potential role in the specification of lineages that arise from this region (Figure 3C).

Summary

The sea urchin twist homologue, Lv-twist, has high sequence similarity to twists of other species, suggesting conserved function through its DNA binding and dimerization motifs. Expression of Lv-twist suggests that its primary function

~

IS.8 W.Wengetal.

Figure 3

Indirect immunofluorescent localization

Polyclonal antibodies to !he full length Drosophila twist protein was used to examine !he spatial pattern

of Lv-twist expression at different embryonic stages. No expreSSion is observed in !he blastula stage (A)

and only weak expreSSion is observed in the vegetal plate at the early gastrula (B). Highest levels of

protein were deteaed in !he mesendoderm at !he gastrula stage (q.

occurs after the initial specification of the embryonic mesoderm of the sea urchin. While our data do not preclude the possibility that low levels of maternal or zygotic Lv-twist protein may function to specify the embryonic mesoderm, high levels of expression after the initial events of gastrulation suggest that Lv-twist also serves a later function. It remains to be determined if Lv-twist influences the activity of the sea urchin myogenic bHLH SUM-I.

References 1. Hiirstadius, S. (1973) Experimental Embryology of Echinoderms, Clarendon Press, Oxford 2. Davidson, E., Cameron, R.A. and Ransick, A. (1998) Development 125,3269-3290 3. Ruffins, S.W. and Enensohn, C.A. (1996) D evelopment 122, 253-263 4. Chen, S.w. and Wessel, G.M. (J 996) Dev. BioI. 175, 57-{'5 5. Venuti, ].M., Goldberg, L., Chakraborty, T., Olsen, E.N. and Klein, W.H. (1991) Proc. Nat!.

Acad . Sci. U.S.A. 88, 6219-{'223 6. Venuti,].M., Gall, L., Kozlowski, MT and Klein,W.H. (1993) Mech. Dev. 41, 3-14 7. Thisse, B., Stoetzel, c., Gorostiza-Thisse , C. and Periin-Schmitt, F (1988) EMBO]. 7,

2175-2183 8. Wang, S.M., Coljee, V.W., Pignolo, R.J., Rotenberg, M.O., Criswfalo, V.]. and Sierra E (1997)

Gene 187, 83-92 9. Wolf, c., Thisse, C ., Swetzel, 'c., Thisse, B., Gerlinger, P. and Perrin-Schmitt, E (1991) Dev.

BioI. 143,363-373 10. Hopwood, N.D., Pluck, A. and Gurdqn,].B. (1989) Ceil 59, 893-903 1\. Yasui, K., Zhang, S.c., Uemura, M., Aizawa, S. and Ueki, T. (1998) Dev. BioI. 195,49-59 12. Pan, D., Valentine, S.A. and Courey, A.J. (1994) Mech. Dev. 46, 41-53 13. Sommer, R.J. and Tautz, D. (1994) Dev. Genet. 15,32-37 14. SotO,].G., Nelson, B.H. and Weisblat, D.A. (1997) Gene 199, 31-37 15. Harfe, B.D., Gomes, A.V., Kenyon, c., Liu,]. , Krause, M. and Fire, A. (1998) Genes Dev. 12,

2623-2635 16. Leptin, M, Casal, ] ., Grunewald, B. and Reuter, R. (1992) Development Suppl . 23-31 17. Chen, Z.E and Behringer, R.R. (1995) Genes Dev. 9, 686-<>99 18. Spicer, D., Rhee,]., Cheung, W. and Lassar, A. (1997) Science 272,14761-14780 19. Baylies, M.K. and Bate, M. (1996) Science 272,14811-14814 20. Wessel, G.M., Zhang, W. and Klein, W.H. (1990) Dev. BioI. 140,447-454 21. Sanger, E, Niken, S. and Coulson, A.R. (1977) Proc. Nad. Acad. Sci. U.S.A. 74,5463-5467

. 22. Li, L., Cserjesi, P. and Olsen, E.N. (1995) Dev. BioI. 172,280-292 23. Burgess, R., Cserjesi, P., Lignon, K.L. and Olsen, E.N. (1995) Dev. BioI. 168, 296-306 24. Cserjesi, P., Brown, D. , Lignon, K.L. , Lyons, G.E., Copeland, N.G ., Gilbert, D.G ., ]enkins,

N.J. and Olsen, E.N. (1995) Development 121,1099-1110 25. Cserjesi, P., Brown, D., Lyons, G.E. and Olsen, E.N. (1995) Dev. BioI. 170,664-<>78

Neural cranial in the t mexico.

Lennart Olssol1

Evolutionary Biol4 Sweden

Introduction

The neural crest an important rol, features, such a~ neural crest is ah [3-5].

Surpris Ambystoma me; Sellman and HOI for a long time s. the development to gain deeper in not been used. V(

the role of the n~ during axolotl he

Theim amphibian crani; perturbing neur development, as this should be tt been thought no effects on muscle the connective ti! [8]. Maybe this is

In this formation of the extirpation and / indirect method tissues are derive neu ral crest cell

':'To whom correspOl

! 1~ 1I""""""""·