protein kinase ck2 is required for dorsal axis formation in xenopus embryos

www.elsevier.com/locate/ydbio

Developmental Biology

Protein kinase CK2 is required for dorsal axis formation in

Xenopus embryos

Isabel Domingueza,*, Junko Mizunoa, Hao Wua, Diane H. Songb, Karen Symesc,

David C. Seldina

aSection of Hematology-Oncology Department of Medicine, Boston University School of Medicine, Boston, MA 02118, USAbSection of Gastroenterology, Department of Medicine, Boston University School of Medicine, Boston, MA 02118, USA

cDepartment of Biochemistry, Boston University School of Medicine, Boston, MA 02118, USA

Received for publication 9 February 2004, revised 8 June 2004, accepted 9 June 2004

Available online 31 July 2004

Abstract

Dorsal axis formation in Xenopus embryos is dependent upon asymmetrical localization of h-catenin, a transducer of the canonical Wnt

signaling pathway. Recent biochemical experiments have implicated protein kinase CK2 as a regulator of members of the Wnt pathway

including h-catenin. Here, we have examined the role of CK2 in dorsal axis formation. CK2 was present in the developing embryo at an

appropriate time and place to participate in dorsal axis formation. Overexpression of mRNA encoding CK2 in ventral blastomeres was

sufficient to induce a complete ectopic axis, mimicking Wnt signaling. A kinase-inactive mutant of CK2a was able to block ectopic axis

formation induced by XWnt8 and h-catenin and was capable of suppressing endogenous axis formation when overexpressed dorsally. Taken

together, these studies demonstrate that CK2 is a bona fide member of the Wnt pathway and has a critical role in the establishment of the

dorsal embryonic axis.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Protein kinase CK2; h-catenin; Wnt; Xenopus; Ectopic axis induction; Dorsal axis formation

Introduction

The establishment of the embryonic axes is a funda-

mental process that occurs after fertilization and dictates

the subsequent development of the body plan. This process

is best understood in frogs, where substances that are

essential for the development of dorsal structures, called

dorsal determinants, are localized to the vegetal cortex of

the egg during oogenesis (Elinson and Holowacz, 1995;

Fujisue et al., 1993; Holowacz and Elinson, 1993). Shortly

after fertilization, a rotation of the egg cortex relative to

the underlying cytoplasm moves the dorsal determinants to

0012-1606/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.ydbio.2004.06.021

* Corresponding author. Section of Hematology-Oncology Department

of Medicine, Boston University School of Medicine, 650 Albany Street,

Boston, MA 02118. Fax: +1 617 638 7528.

E-mail address: [email protected] (I. Dominguez).

the prospective dorsal side of the embryo (Elinson and

Holowacz, 1995), specifying the dorsoventral axis. This

cortical rotation is essential for embryonic development

because if inhibited, the embryo does not develop any

differentiated structures or embryonic axes (Elinson and

Holowacz, 1995). Although the nature of the dorsal

determinants is unknown, their mechanism of action

involves the activation of canonical Wnt signaling

(Marikawa et al., 1997). Evidence for canonical Wnt

signaling in dorsal patterning comes from studies of h-catenin, a protein with a dual cellular function, utilized

both in cell adhesion and as a transcriptional co-activator

of canonical Wnt signaling. h-catenin is found maternally

(DeMarais and Moon, 1992) and it is enriched dorsally

after fertilization (Larabell et al., 1997). Overexpression of

h-catenin is sufficient to specify an ectopic axis

(Funayama et al., 1995; Guger and Gumbiner, 1995),

274 (2004) 110–124

I. Dominguez et al. / Developmental Biology 274 (2004) 110–124 111

and it is required for endogenous dorsal development

(Heasman et al., 1994).

In resting cells in culture, h-catenin is rapidly targeted forproteasomal degradation. This occurs through sequential

phosphorylation of its N-terminus by casein kinase 1 (CK1)

and glycogen synthase kinase-h (GSK3h) within a multi-

protein bdestruction complexQ in which adenomatous poly-

posis coli (APC) and axin act as scaffolds (He et al., 2004).

Stabilization of h-catenin occurs uponWnt signaling, as Wnt

activates dishevelled (dsh) through an unknown mechanism

to inhibit the action of the bdestruction complexQ (He et al.,2004). As a result, h-catenin accumulates in the cytoplasm

and translocates to the nucleus where it relieves the repression

of T cell factor (TCF) family transcription factors by groucho

and provides the transactivation domain for TCF (Giles et al.,

2003). During Xenopus development, h catenin or TCF

regulates target genes such as Xsiamois and Xenopus nodal-

related gene 3 (Xnr3) (Sokol, 1999).

Dorsal enrichment of h-catenin in the early stages of

Xenopus development is indicative of Wnt activation. In

fact, h-catenin up-regulation correlates with dorsal depletion

of GSK3h protein (Dominguez and Green, 2000). The

mechanism of depletion may involve GSK3-binding protein

(GBP), a negative regulator of GSK3h function that, like h-catenin, is sufficient and necessary for endogenous axis

formation (Yost et al., 1998) and has recently been proposed

to translocate to the dorsal side during cortical rotation

(Weaver et al., 2003). In addition, dsh is enriched dorsally

after cortical rotation (Miller et al., 1999). Several Wnts

(Xwnt2b, Xwnt5a, Xwnt7b, Xwnt8b, and Xwnt11) are

present in the early embryo (Nusse, 2001), and one

(Xwnt11) has been found to be preferentially translated in

the dorsal–vegetal region after cortical rotation (Schroeder

et al., 1999), suggesting a Wnt ligand-dependent signal may

be involved as well. However, expression of dominant-

negative forms of dsh and Wnt do not suppress endogenous

axis formation, casting doubts about the involvement of dsh

and Wnt in the establishment of the dorsal axis (Hoppler et

al., 1996; Sokol, 1996, 1999). Thus, the description of Wnt

signaling in dorsal axis development remains incomplete

(Dominguez and Green, 2001).

Recently, protein kinase CK2 was found to be associated

with elements of the canonical Wnt pathway. CK2 is a highly

conserved serine or threonine kinase that is present in

organisms from protozoans to mammals (Faust and Mon-

tenarh, 2000). CK2 is expressed almost ubiquitously in the

mammalian adult and embryo and the protein is present in

multiple cell compartments (Faust and Montenarh, 2000).

CK2 holoenzyme is a tetramer composed of two catalytic

subunits and two regulatory subunits (Allende and Allende,

1995). Mammals have two alternative catalytic subunits, a

and a’, and a single h regulatory subunit, while Xenopus

appears to have a single catalytic and regulatory isoform

(Allende and Allende, 1995). Gene targeting of the single hsubunit in mice (Buchou et al., 2003) and inactivation of all

subunits in yeast are lethal (Chen-Wu et al., 1988; Padma-

nabha et al., 1990), suggesting that the CK2 holoenzyme is

essential for cell survival. Targeting a single catalytic subunit

in mice has a more subtle phenotype; male CK2a’ null mice

are infertile due to abnormal spermatogenesis (Xu et al.,

1999). CK2 is highly expressed in the mouse embryo

(Schneider et al., 1986), but its function there is unknown.

In the Wnt pathway, CK2 is capable of phosphorylating

several proteins in vitro, but its importance in Wnt signaling

in vivo has not been determined. The first Wnt transducer

identified as a substrate was dsh (Willert et al., 1997). Since

then, other Wnt transducers have been found to be

phosphorylated by CK2. These include h-catenin (Song et

al., 2000, 2003), APC (Homma et al., 2002), groucho

(Nuthall et al., 2002), TCF-4 (Miravet et al., 2002), and

engrailed (Maizel et al., 2002). Phosphorylation of these

different substrates by CK2 appears to elicit contradictory

effects. For example, CK2 appears to positively regulate dsh,

h-catenin, and engrailed. CK2 is a major dsh kinase in insect

cells (Lee et al., 1999; Willert et al., 1997), and CK2

inhibition reduces dsh levels in mammalian cells (Song et al.,

2000). We have shown that CK2 also binds and phosphor-

ylates h-catenin in mammary epithelial cells (Song et al.,

2000) and that CK2 inhibition leads to decreased h-cateninstability and decreased h-catenin/TCF transactivation (Song

et al., 2003). CK2 phosphorylation of engrailed protein

increases its DNA binding and also inhibits its intercellular

transfer, restricting its distribution (Maizel et al., 2002). In

contrast, other reports suggest that CK2 might act as a

negative regulator of Wnt signaling, as it promotes groucho

transcriptional repression (Nuthall et al., 2002), and it

phosphorylates UBC3B, an E2 ubiquitin conjugating

enzyme, thereby enhancing h-catenin degradation (Semplici

et al., 2002). CK2 has also been reported to affect the

interaction of h-catenin and E-cadherin (Bek and Kemler,

2002), a process that seemingly would recruit h-catenin to

the plasma membrane and oppose Wnt signaling (Lickert et

al., 2000; Serres et al., 2000). In other cases, CK2

phosphorylation may play a housekeeping role, as in the

case of TCF-4, whose phosphorylation has no apparent

effect on h-catenin binding (Miravet et al., 2002). In

addition, CK2 does not bind axin and does not affect axin

mediated JNK activation (Zhang et al., 2002).

Given these conflicting observations, we decided to make

use of the well-characterized Xenopus system to test how

CK2 acts in the canonical Wnt pathway and whether it is

capable of regulating dorsal axis development. Here we show

that CK2 distribution in the early embryo was consistent with

a role in dorsal axis formation. Indeed, CK2 is a positive

regulator of Wnt signaling since CK2 is able to mimic the

ability of Wnt to promote ectopic axis formation. Further-

more, a kinase-inactive (KI) mutant of CK2 inhibited ectopic

axis formation byWnt and inhibition of CK2 activity dorsally

led to inhibition of endogenous dorsal axis formation. CK2

appears to regulate Wnt signaling at the level of h-catenin,since CK2 was able to stabilize wild-type h-catenin in vivo

and the KI-CK2 antagonized axis duplication mediated by h-

I. Dominguez et al. / Developmental Biology 274 (2004) 110–124112

catenin, but not h-catenin with a mutation in the major CK2

phosphorylation site. These results indicate that CK2 is a

bona fideWnt transducer and suggest that CK2 is involved in

dorsal axis establishment in vertebrates.

Materials and methods

Cloning CK2a, CK2b, and KI-CK2a and in vitro mRNA

synthesis

Published cDNA sequences for Xenopus CK2a and

CK2h (Jedlicki et al., 1992) were used to design primers

for PCR that also included restriction enzyme sites. RT-PCR

from stage 10 cDNA was used to obtain the cDNAs for

XCK2a and XCK2h using the following primers: (CK2a

dcha) GCT CTA GAC TCG AGATGT CAG GAC CTG TG

and (CK2a izqda) GCT CTA GAG AGC TCC TAC TGA

GTG GCT CC, and (CK2b dcha) GGA ATT CGG ATC CAT

GAG TAG CTC GGA G and (CK2b izqda) CCG CTC GAG

GAG CTC TCA ACG CAT GGT CTT. The PCR products

were digested with XhoI and XbaI and subcloned into pCS2

and pCS2 + MT (for myc-tagged proteins).

A kinase-inactive mutant of XCK2a (KI-XCK2a) was

generated by PCR with the following primers: (XCK2a KI

fwd) TTG TAA GAATTC TCA AGC CTG and (XCK2a KI

rev) CAG GCT TGA GAA TTC TTA CCA. This changes

the residue involved in phosphotransfer at Lys68 to Arg.

For generation of pCS2-myc-T393A-h-catenin, h-cate-nin (T393A) was excised with BamHI and SmaI from

pGEX-2T-h-catenin (T393A) (Song et al., 2003) and

subcloned into pJ13M, which contains a myc epitope tag

(Sells and Chernoff, 1995). The sequence including the tag

was amplified by PCR with primers including ClaI sites and

subcloned into the T-A cloning vector pCR2.1, digested

with ClaI, subcloned into pCS2, and sequenced. pCS2-myc-

h-catenin was generated by substituting wild-type h-catenininto the generated plasmid with Sac1.

Plasmids pCS2-XCK2a, pCS2-XCK2h, pCS2-KI-

XCK2a, pCS2 + MT-XCK2a, pCS2 + MT-KI-XCK2a,

pCS2-myc-T393A-h-catenin, pCS2-myc-h-catenin, pSP6-

lacZ, and pSP64T-Xwnt8 were linearized and mRNA was

transcribed in vitro using the SP6 mMessage mMachine kit

(Ambion).

Recombinant b-catenin-GST fusion protein

Bacterial expression of recombinant GST-h-catenin fusionprotein was produced as previously described (Song et al.,

2003). Recombinant protein was 98% pure as determined by

gel electrophoresis and Coomassie blue staining.

Embryo and animal cap assays

Embryos were fertilized in vitro, dejellied in 2% cysteine

(pH 7.8), and maintained in 0.1� MMR (Newport and

Kirschner, 1982). Staging was according to Nieuwkoop and

Faber (1967). Embryos were microdissected with an eye-

lash-knife at various stages for protein analysis. For micro-

injection, embryos were transferred to 1.5% Ficoll

(Pharmacia) in 0.5� MMR and injected at stage 2 or 3.

Injections were carried out equatorially, either dorsally or

ventrally, in two blastomeres. For mRNA microinjections,

in vitro-transcribed mRNAs for XCK2a (0.5–2.3 ng),

XCK2h (0.4–1.8 ng), KI-XCK2a (0.3–4 ng), myc-XCK2a

(1.5 ng), myc-KI-XCK2a (2.7 ng), Xwnt8 (0.34–0.42 pg),

and lacZ (0.5–4 ng) were used. For protein microinjections,

two blastomeres were injected with 5 nl containing 200 pg

of recombinant GST-h-catenin fusion protein with or

without 0.3 or 1.25 U of recombinant CK2a/h (NEB).

Embryos were stopped after microinjection at the stages

indicated in the text as described below.

For animal caps experiments, embryos were injected in

two animal blastomeres at stage 2 in 1.5% Ficoll (Pharma-

cia) in 0.5� MMR with in vitro-transcribed mRNAs for

XCK2a and XCK2h or Xwnt8. Animal caps were dissected

with forceps at stage 8 in 0.8� MMR and cultured until

sibling control embryos reached stage 10. Embryos were

also injected ventroequatorially with the same mRNAs and

analyzed for axis duplication (see below).

Embryo development was visualized under a Leica MZ6

dissecting microscope and photographed during tadpole

stages 38–41. Axis duplication and axis defects were

analyzed at tadpole stages. For axis defects, embryos were

scored according to the dorsoanterior index (DAI; Kao and

Elinson, 1988). This is a 0–10 scale, where normal embryos

have a value of 5, the most ventralized embryos have a value

of 0, and the most dorsalized embryos have a value of 10.

Histology

Injected embryos were fixed with MEMFA for 1 h,

dehydrated in ethanol, incubated in xylene, and embedded

in paraffin. Eight-micrometer serial sections were cut and

stained with hematoxylin/eosin as previously described

(Dominguez et al., 1995).

In situ hybridization

In situ hybridization was performed according to the

protocol of Harland (Harland, 1991) with some modifica-

tions. Only RNase A was used after hybridization, antibody

incubation was performed in Ptx (1� PBS, 0.1% Triton X-

100), and 10% goat serum in Ptx was used for blocking.

Digoxygenin-labeled probes were transcribed from the

linearized plasmids pCS2-XCK2a and pCS2-XCK2h using

a DIG RNA labeling kit (Roche). As a control, a sense

probe was obtained from pCS2-XCK2h. Staining solution

contained NBT/BCIP (4-nitro blue tetrazolium chloride/5-

bromo-4-chloro-3-indolyl-phosphate) and afterwards

embryos were fixed in Bouin’s, bleached, and cleared in

BB/BA (Benzyl benzoate/benzyl alcohol) when necessary.


RT-PCR

RT-PCR was performed as described (Schneider and

Mercola, 1999). Briefly, RNA was isolated with Trizol

according to the manufacturer’s instructions (Invitrogen).

Reactions with and without RT were carried out. The PCR

reactions included 32P-g-dATP (0.15 ACi) and the products

were separated in 5% acrylamide gels and visualized and

quantitated in a Molecular Imager System GS-525 (Bio-

Rad). The following primers, cycles, and annealing temper-

atures were used for PCR:

XCK2a (product 206 bp) fwd: 5V-cgg ggc aaa tac agt gaa

gt-3V rev: 5V-tgt tcg aag aca agt gct gg-3V (25 cycles, 608Cannealing)

XCK2h (product 123 bp) fwd: 5V-ttt tgg cac tgg att tcc tc-3V rev: 5V-agc tgg tat gcc ata ggg tg-3V (25 cycles, 608Cannealing)

Xnr3 (product 367 bp) fwd: 5V-gtg cag ttc cac aga atg ag-

3V rev: 5V-cca tgg atc ggc aca aca ga-3V (25 cycles, 608Cannealing)

Xsiamois (product 205 bp) fwd: 5V-aag ata act ggc att cct

gag c-3V rev: 5V-ggt agg gct gtg tat ttg aag g-3V (28 cycles,608C annealing)

Xtwin (product 344 bp) fwd: 5V-tcc tgt gtt ctg ccc acc a-3Vrev; 5V-ctg ttg ggt gcc gat ggt a-3V (33 cycles, 608Cannealing)

Xvent1 (product 147 bp) fwd: 5V-ttc cct tca gca tgg ttc

aac-3Vrev: 5V-gca tct cct tgg cat att tgg-3V (26 cycles, 608Cannealing)

Ornithine decarboxylase (ODC) (product 228 bp) fwd:

5V-cag cta gct gtg gtg tgg-3V rev: 5V-caa cat gga aac tca

cac c-3V (22–30 cycles, 55–608C annealing).

Embryo extractions, immunoblotting, and

immunoprecipitation analysis

For endogenous and exogenous protein expression, five

embryos (or 10 embryo thirds) were homogenized in SDS-

PAGE sample buffer (Boston Bioproducts), boiled, and the

equivalent of 0.15 embryos was loaded directly on gels. For

PKC~ , immunoblotting embryos were lysed in homogeni-

zation buffer A [0.1� MMR with inhibitors: 10 mM NaCl,

0.2 mM KCl, 0.2 mM CaCl2, 0.1 mM MgCl2, 0.5 mM

HEPES (pH 7.4), 1 mM DTT, 2 mM PMSF, 1 mM Na3VO4,

and 4 Ag/ml of leupeptin, aprotinin, and pepstatin] and

cleared lysates obtained by centrifugation at 14,000 rpm for

15 min at 48C, 2� sample loading buffer was added to the

cleared lysates, boiled, and the equivalent of 0.15 embryos

was loaded.

For immunoprecipitation of CK2a and CK2h, 10

embryos were lysed in homogenization buffer B [50 mM

Tris–HCl (pH 7.5), 50 mM NaCl, 1 mM EDTA, 1 mM

EGTA, 0.2% Tx-100, 1 mM Na3VO4, 1 mM DTT, 10 mM

sodium pyrophosphate, 1 mM NaF, 5 Ag/ml of leupeptin,

aprotinin, antipain, and pepstatin, and 2 mM PMSF] and

cleared lysates obtained by centrifugation at 14,000 rpm for

15 min at 48C were subjected to immunoprecipitation with

anti-CK2a antibody (Santa Cruz) or myc antibody (Roche)

followed by incubation with protein G sepharose (Amer-

sham). Beads were washed three times with 50 mM Tris–

HCl (pH 7.5), 50 mM NaCl, and twice with 50 mM Tris–

HCl (pH 7.5) before in vitro kinase assays.

For GST-h-catenin detection, 5–10 embryos were homo-

genized in buffer A and cleared by centrifugation at 14,000

rpm for 15 min, 2� sample loading buffer was added to the

cleared lysates, boiled, and the equivalent of 2.5 embryos

for h-catenin and 0.15 for tubulin was electrophoresed on

8% SDS-PAGE gels.

Gels were electroblotted onto polyvinylidene difluoride

(PVDF) membranes (Millipore) and quantitative immuno-

blotting analysis was performed. The membrane was blocked

with PBS with 0.05% Tween (PBST) and 10% goat serum

(Sigma), followed by incubation with primary antibodies

diluted in blocking buffer: anti-CK2a (Stressgen, BS

Canada), anti-CK2h (BD Transduction), anti-myc (Roche),

anti-CK1q (BD Transduction), anti-h-catenin (BD Trans-

duction) or anti-PKC~ (Santa Cruz), and followed by

horseradish peroxidase-conjugated goat anti-mouse IgG

antibodies (Jackson ImmunoResearch). Competition assays

were performed incubating the primary antibody with 0.34–

2.5 Ag/ml of recombinant CK2 (NEB). Incubations were 1 h

each, separated by three 10-min washes with PBST.

Peroxidase activity was visualized with an ECL system

(Amersham). Amonoclonal antibody against tubulin (Sigma)

was used to confirm equal loading. When using the same

membrane for different antibody incubations, the membrane

was not stripped. For quantitative analysis of each band,

integrated pixel density minus background density was

determined using a Fluor-S MultiImager (BioRad).

Kinase assays

For activity measurements, five embryos were homo-

genized in CK2 homogenization buffer C [50 mM Tris–HCl

(pH 7.9), 1 mM EDTA, 10% glycerol, 1 mM Na3VO4, 1

mM DTT, 10 mM sodium pyrophosphate, 1 mM PMSF, 1

mM NaF, 4 Ag/ml of leupeptin, aprotinin, and pepstatin, and

2 mM PMSF] and cleared by centrifugation at 14,000 � g

for 15 min. Cleared lysates (0.25 embryo equivalents) or

immunoprecipitates were used for CK2 kinase assays. CK2

kinase activity was performed in kinase buffer [50 mM

Hepes (pH 7.8), 150 mM KCl, 10 mM MgCl2, 10 AM DTT,

100 AM Na3VO4, 50 AM g32-P-GTP (1000 cpm/pmol), and

250 AM peptide substrate (RRREEETEEE) in a total

volume of 25 Al]. Incubations were performed at 228C for

30 min. Samples incubated without the substrate peptide

were used as controls. The reaction was stopped by the

addition of 2� Laemli sample buffer. One fourth of each

reaction mixture was subjected to SDS/16% PAGE followed

by autoradiography as described previously (Schagger and

von Jagow, 1987). Radioactive phosphate incorporation was


measured in a Molecular Imager System GS-525 (BioRad).

Cleared lysates were also subjected to 8% SDS-PAGE

followed by immunoblotting and densitometry to determine

the amount of tubulin in each sample.

Fig. 1. CK2a and CK2h mRNA and protein expression during early

development. (A) RNA from embryos at maternal stages 1 through 8 was

isolated and RT-PCR was performed with specific primers for XCK2a and

h. RT-PCR for ornithine decarboxylase (ODC) is shown as a control. This

experiment was repeated twice with identical results. (B) Immunoblotting

confirms protein expression. Anti-human CK2a (left) recognizes an

appropriate 45-kDa band and anti-human CK2h (right) recognizes an

appropriate 25-kDa band in stage 2 embryo lysates. (C) Identity of the

immunoreactive bands was confirmed by competition with recombinant

human CK2a/h. Stage 14 extract alone (�) or in the presence (+) of 0.34

Ag/ml of CK2 to compete with the CK2a (upper panel) or 2.5 Ag/ml of

CK2 to compete with the CK2h (lower panel). The higher molecular weight

band detected with the CK2a antibody is a nonspecific cross-reactive

protein, as it is not competed by recombinant CK2 and its level is variable

between antibody lots compared with the authentic CK2a band (compare B

and C).

Results

CK2a and CK2b are expressed in the early embryo

For CK2 to function in dorsal axis formation, it must be

present and active in the embryo at the appropriate time, that

is, provided maternally and functional in the early embryo.

The expression of CK2a and CK2h mRNA was examined

before the onset of zygotic transcription, at embryonic

stages 1–8. RT-PCR analysis showed that mRNAs for

CK2a and CK2h are present at all these stages of

development (Fig. 1A). We used immunoblotting with

specific antibodies against CK2a and CK2h to ensure that

the CK2a and CK2h proteins are expressed in the early

embryo. Although these antibodies were raised against

human CK2 subunits, since CK2 protein is highly con-

served, they recognized bands of the appropriate size in

Xenopus embryos (approximately 45 and 25 kDa for CK2a

and CK2h, respectively, stage 2 is shown as a representa-

tive; Fig. 1B). Immunoreactive CK2a and CK2h were

present throughout all stages of embryonic development,

consistent with the RT-PCR results (data not shown) and

were competed when the antibody was preabsorbed with

recombinant CK2a/h (Fig. 1C).

To determine the spatial distribution of the maternal

transcripts of CK2a and CK2h in early Xenopus develop-

ment, embryos were fixed and analyzed for localization of

mRNA by in situ hybridization. Digoxigenin-labeled anti-

sense probes against Xenopus CK2a and CK2h subunits

were used. At stages 3 and 8, CK2a and CK2h transcripts

were detected in the animal region of the embryo (Fig. 2A,

right two panels). A CK2h sense probe control showed no

staining (Fig. 2A, left panel).

To determine whether the distribution of mRNA and

protein correspond, stage 7 embryos were divided into

animal (A), medial (M), and vegetal (V) sections. The

obtained thirds were homogenized and subjected to immu-

noblotting. We found that CK2a and CK2h were present in

the upper two thirds of the embryo (A and M sections) and

nearly absent from the vegetal third (Fig. 2B). The

expression of each kinase was quantitated using the Fluor-

S and expressed as a ratio of the amount of each protein

relative to the amount in the medial portion (Fig. 2C). While

all the enzymes were depleted in the vegetal portion of the

embryo, perhaps due to exclusion by yolk proteins, CK2a

and CK2h were reduced the most. CK2h and GSK3h were

both enriched in the medial portion, compared with the

animal portion. This important region is where the dorsal

determinants act (Kageura, 1997), and has been proposed to

contain a h-catenin-inducing activity that will activate h-

catenin during mesoderm induction (Schohl and Fagotto,

2003). Taken together, these studies show that CK2a and

CK2h mRNA are provided maternally, encoding proteins

that are expressed early enough to participate in dorsal axis

formation. Moreover, these data show that CK2a and CK2hare present in the region of the embryo where h-catenin is

up-regulated during early development.

CK2 promotes ectopic axis formation and Wnt target gene

expression

To determine whether CK2 is a regulator of dorsal axis

formation, we took advantage of the fact that overexpression


of canonical Wnts and Wnt transducers, for example, h-catenin, dishevelled, GBP, or a dominant-negative form of

GSK3h, in the prospective ventral side of the embryo

induce a secondary ectopic dorsal axis including anterior

structures such as the eyes and cement gland (Dominguez et

al., 1995; Funayama et al., 1995; Guger and Gumbiner,

1995; He et al., 1995; Pierce and Kimelman, 1995; Smith

and Harland, 1991; Sokol et al., 1991, 1995; Wodarz and

Nusse, 1998; Yost et al., 1998). mRNA encoding CK2a

and/or CK2h were microinjected at stage 2 or 3 (2- or 4-cell

stage embryos, respectively) into the equatorial regions of

both ventral blastomeres. Microinjection of either subunit

alone failed to produce a secondary axis (Table 1). Micro-

injection of both CK2a and CK2h mRNAs resulted in

duplication of the neural tube at stages 16–19 (not shown)

and duplicated axes at stages 38–41 (Fig. 3A). These effects

are similar to those of Xwnt8 mRNA, injected as a positive

Fig. 2. CK2a and CK2h are localized in the upper two thirds of the embryo. (A)

CK2h sense RNA probe (left) or antisense RNA probes to CK2a (middle) an

respectively. Staining is shown in blue; brown staining is due to embryo pigmen

Immunoblotting with CK2a, CK2h, CK1q, GSK3h, and PKC~ and control tubu

depicted in the scheme underneath: A (animal), M (medial) and V (vegetal). O

autoradiography exposure times to allow comparison of the bands. (C) Immuno

normalized. In this panel, for each antibody, the amount of immunoreactive prot

(medium third, gray bar; animal third, black bar; vegetal third, light gray bar). CK2

other enzymes.

control (Table 1 and Fig. 6). The effects of CK2a plus

CK2h were dose dependent (Table 1). At the highest dose

shown (1.1 ng of CK2a + 0.9 ng of CK2h mRNAs), the

ectopic axes were classified as complete in 56% of the

embryos, that is, they contained anterior structures such as

eyes and cement gland, and 7% of the embryos developed a

partial duplication of the axis, with no head structures

visible. At higher doses of mRNA, the embryos were

hyperdorsalized with phenotypes ranking from 7 to 10 in the

dorsoanterior index (DAI) scale (Kao and Elinson, 1988)

(data not shown). Control embryos injected with water or

lacZ mRNA did not develop a secondary axis (Table 1).

Histological sections of CK2a plus CK2h-induced ectopic

axes revealed that they contained properly organized

notochords, somites, and neural tubes (Fig. 3B).

To verify that injection of CK2a and CK2h mRNAs

produced an increase in functional CK2a and CK2h

For in situ hybridization, embryos at stages 3 and 8 were incubated with a

d CK2h (right). Numbers 3 and 8 in the figure indicate stages 3 and 8,

tation. This experiment was repeated three times with identical results. (B)

lin antibodies on extracts from embryos at stage 7 divided into thirds as

ne representative immunoblot out of four is shown, with adjustment of

blots were quantitated from the actual immunoblot using the Fluor-S and

ein in each third of the embryo is compared to the medial third, set at 1.0

a and CK2h were markedly depleted from the vegetal third compared to the

Table 1

Effect of ventral overexpression of mRNAs on axial development

mRNAs mRNA in ng

(except were indicated)

Total number of

injected embryos

Complete axis

duplication

Partial axis

duplication

Single axis Other

phenotypes

CK2a/CK2h 1.1/0.9 96 54 (56%) 7 (7%) 29 (30%) 6 (6%)a

0.8/0.58 87 18 (21%) 32 (37%) 37 (42%) 0 (0%)

0.6/0.4 54 1 (2%) 35 (65%) 16 (30%) 2 (3%)a

0.5/0.4 37 0 (0%) 7 (23%) 30 (76%) 0 (0%)

0.5/0.4 34 0 (0%) 19 (29%) 24 (71%) 0 (0%)

CK2ab 2.3 150 0 (0%) 0 (0%) 145 (97%)c 5 (3%)a

CK2hb 0.9 65 0 (0%) 0 (0%) 65 (100%) 0 (0%)

KI-CK2ab 1.6 24 0 (0%) 0 (0%) 24 (100%) 0 (0%)

KI-CK2a/CK2hb 1.6/0.58 44 0 (0%) 0 (0%) 44 (62%) 17 (28%)d

Xwnt8b 0.34 pg 26 7 (27%) 13 (50%) 6 (23%) 0 (0%)

0.17 pg 28 2 (8%) 13 (46%) 13 (46%) 0 (0%)

LacZb 2 64 0 (0%) 2 (3%) 38 (95%) 1 (2%)a

Waterb – 26 0 (0%) 0 (0%) 26 (100%) 0 (0%)

Uninjectedb – 127 0 (0%) 0 (0%) 125 (98%) 2 (2%)a

Other phenotypes:a Gastrulation defects.b Embryos from different experiments were pooled.c Single axes, but many of these embryos were curved.d Slightly shorter embryos due to posterior defects.


proteins, embryos microinjected with different amounts

of mRNA were harvested at stage 10, and kinase assays

and immunoblotting were performed. As the dose of

injected mRNA was increased, the level of CK2a and

CK2h proteins and kinase activity increased (Fig. 3C).

To determine whether increased CK2 expression and

activity of CK2 leads to up-regulation of Wnt target

genes such as Xsiamois, Xtwin, and Xnr3, expression of

these genes was assessed using semiquantitative RT-PCR.

Ectopic expression of CK2a and CK2h protein in the

ventral side of the embryo led to increases in Xsiamois,

Xtwin, and Xnr3 expression in whole embryos at stage

10 (Fig. 3D). At the two doses of CK2 studied,

increases in Xsiamois, Xtwin, and Xnr3 correlated with

induction of secondary axes. Concomitantly with this,

CK2 overexpression led to down-regulation of genes

preferentially expressed in the ventral embryo, such as

Xvent1 (Fig. 3D).

To confirm that CK2 induces dorsal gene expression, an

animal cap explant assay was carried out, with Xwnt8 as a

positive control. Two doses of CK2a/h or Xwnt8 were

injected at stage 2 in animal blastomeres. In parallel

experiments, the same doses were injected ventroequato-

rially to assess their ability to induce ectopic axis. Semi-

quantitative RT-PCR analysis shows that CK2a/h and

Xwnt8 cause a dramatic induction of Wnt target genes,

Xsiamois, Xtwin, and Xnr3 (Fig. 4), correlating with their

ability to induce ectopic axes (Table 1). Taken together,

these experiments demonstrate that overexpression of CK2

is sufficient to promote the development of an ectopic dorsal

axis. The tetrameric form of CK2 but not the catalytic

subunit alone is functional in Wnt signaling. Indeed,

expression of CK2a/h is able to up-regulate h-catenintarget genes at a level comparable to Xwnt8. These results

support the hypothesis that CK2 is a positive regulator of the

Wnt pathway since it is able to mimic the biological

function of canonical Wnts.

CK2 stabilizes b-catenin in Xenopus embryos in vivo

The canonicalWnt signaling pathway produces alterations

in gene expression by regulating the nuclear availability of h-catenin, the required cofactor for TCF/LEF family tran-

scription factors. We have previously shown that CK2

inhibition destabilizes h-catenin in mammalian cells in

culture, and that h-catenin mutated at the major CK2

phosphorylation site at T393 has reduced stability when

injected into Xenopus embryos (Song et al., 2000, 2003).

Here we examine whether CK2 is able to stabilize h-cateninin vivo. For this purpose, we injected recombinant GST-h-catenin protein alone or together with CK2 protein into

Xenopus embryos. Microinjection of purified protein has the

advantage of avoiding differences in translation efficiency of

injected mRNAs. It is impossible to carry out such experi-

ments with endogenous h-catenin because it is masked in

immunoblots by the presence of yolk protein. First, we tested

whether or not injected recombinant h-catenin protein

behaves as h-catenin does in vivo that is, whether it is more

stable in the dorsal side of the embryo than in the ventral side

(Larabell et al., 1997). Wild-type GST-h-catenin fusion

protein was injected into the equatorial region of both

blastomeres at stage 3, either dorsally or ventrally. Embryos

were harvested every 10 min and lysates were prepared for

immunoblotting. Like endogenous h-catenin, recombinant

GST-h-catenin was less stable when injected in the ventral

side than in the dorsal side of the embryo (Fig. 5A). In three

independent experiments, the half-life for wild-type GST-h-catenin was approximately 30 min in the ventral side and 80

min in the dorsal side. To determine whether CK2 affects h-catenin stability, GST-h-catenin was coinjected with two

Fig. 3. CK2 induces dorsal axis duplication. (A) Embryos were injected ventrovegetally at stage 2 or 3. Control embryos injected with water or 2 ng of mRNA

encoding h-gal are indistinguishable from uninjected embryos (left); embryos injected with 1.1 ng of CK2a and 0.9 ng of CK2h mRNAs are shown at stage 38

(right). Embryos injected with CK2a and CK2h exhibit duplicated body axes with heads that include the most anterior structures; four eyes are indicated in one

embryo with lines (N) and duplicated cement glands (brown spot) are shown in one embryo with arrowheads ( ). (B) Histological analysis of embryos injected

with CK2a and CK2h mRNA confirms duplication of internal dorsal embryonic structures. Transverse sections of a control embryo (a) or an embryo injected

with CK2a and CK2h mRNAs (b). In the latter, note the presence of two neural tubes (nt) and notochords (nc) and two pairs of somites (s) surrounding the

notochord. (C) Ectopic axis formation correlates with expression and activity of CK2. Extracts from control embryos (lane 1), embryos injected with 1.1 ng of

CK2a plus 0.9 ng of CK2h (lane 2), and embryos injected with 0.8 ng of CK2a plus 0.58 ng of CK2h (lane 3) harvested at stage 10. For immunoblotting (IB),

0.15 embryo equivalents were used and for an in vitro kinase assay (KINASE), 0.25 embryo equivalents were used. (D) Ectopic axis formation by CK2

correlates with the induction of h-catenin/TCF target genes. Embryos were injected with mRNA coding for h gal (lane 1) or CK2a and CK2h (lane 2, 1.1 and

0.9 ng; and lane 3, 0.8 and 0.58 ng, respectively) at stages 2–3. Embryos were harvested at stage 10 for RT-PCR (upper panels) and immunoblotting with

specific CK2a and h antibodies (IB, lower 2 panels). Semiquantitative RT-PCR analysis confirmed induction of dorsal markers Xsiamois, Xtwin, and Xnr3 in

whole embryos, correlating with the expression of CK2 subunits. Xvent1 is a ventrolateral marker, and ODC serves as a normalization control. Numbers in the

figure represent the quantitation of the obtained bands. One representative experiment out of eight is shown.


different amounts of recombinant CK2a/h (0.3 or 1.25 U).

The embryos were injected equatorially in the ventral side of

the embryo. At the high dose of CK2, the half-life of ventrally

injected GST-h-catenin was prolonged, compared to that of

dorsally injected GST-h-catenin (Figs. 5B and C). Tubulin is

shown as a loading control in these experiments. Together,

these results indicate that, like Wnt, CK2 positively regulates

h-catenin stability in vivo and induces overexpression of

TCF/h-catenin target genes and dorsal axis formation.

A kinase-inactive mutant of CK2 antagonizes Wnt signaling

To verify that CK2 mediates Wnt signaling in the early

embryo, it is necessary to demonstrate that CK2 is both

Fig. 4. CK2a/h, like Xwnt8, induces h-catenin/TCF target genes in animal

cap explants. Two animal blastomeres of stage 2 embryos were injected

with varying amounts of mRNA coding for CK2a and CK2h (lane 3, 1.1

and 0.9 ng; and lane 4, 0.5 and 0.4 ng, respectively), mRNA for Xwnt8

(lane 5, 0.34 pg; and lane 6, 0.17 pg) or left uninjected (lane 2). At stage 10,

RNA from animal caps (dissected at stage 8) was analyzed for Xsiamois,

Xtwin, Xnr3 and control ODC expression by RT-PCR. PCR samples from

whole embryos (lane 1) and minus RT (lane 7) were also analyzed. CK2 is

as effective as Xwnt8 in inducing the expression of Wnt target genes. This

induction correlates with the expression of CK2 subunits (not shown) and

with the ability to induce ectopic axes in sibling embryos injected ventrally

with the same doses of CK2a/h and Xwnt8 (Table 1). Xtwin expression in

control animal caps may reflect maternal expression of the gene (Laurent et

al., 1997). One representative experiment out of two is shown.


sufficient to induce a Wnt signal and required to transmit it.

For this purpose, we generated a kinase-inactive (KI) mutant

of the catalytic subunit of CK2a. This mutant has an

alteration in the ATP binding lysine (Lys68Arg) that is

critical for catalytic activity and that has been shown to

behave as a dominant-negative mutant in other systems

(Vilk et al., 1999). To confirm that this mutant lacks kinase

activity, embryos injected with myc-tagged wild-type CK2a

or myc-KI-CK2a were homogenized at stage 12, and CK2a

was immunoprecipitated using myc antibodies. Immuno-

blotting and an in vitro kinase assay were then performed on

the immunoprecipitates. Fig. 6A shows that while immu-

noprecipitated wild-type CK2a is active, the KI-CK2a has

no detectable kinase activity.

We next tested whether the kinase-inactive mutant could

block the ability of injected Wnt to induce an ectopic axis.

Xwnt8 mRNA was injected at a dose sufficient to induce

double axes in approximately 50% of the embryos (Figs.

6B and C). Coinjection of KI-CK2a mRNA and Xwnt8

mRNA led to a marked reduction in the number of

embryos with a duplicated axis (Figs. 6B and C). While

50% of the embryos injected with Xwnt8 mRNA had

complete axis duplication, only 4% of the embryos

injected with Xwnt8 and KI-CK2a mRNA had complete

double axes (Fig. 6C), and the ectopic axis was very short

in those with partial axis duplication (Fig. 6B, right panel).

No phenotype was observed when mRNA for KI-CK2a

was introduced into the ventral side of the embryo (Table

1). Moreover, injection of KI-CK2a and CK2h mRNA did

not induce ectopic axis formation, indicating that the

kinase activity of the enzyme is required for this effect

(Table 1). Immunoblotting to verify the expression of KI-

CK2a was performed in all experiments (not shown).

These results demonstrate that catalytically active CK2 is

not only capable of inducing an ectopic axis, but its

activity is required for ectopic axis induction by a

canonical Wnt.

Given the observations about phosphorylation of h-catenin by CK2, we hypothesized that the effect of the KI-

CK2a mutant was mediated at least in part by inhibition of

h-catenin phosphorylation, consistent with our previous

observations on CK2 phosphorylation promoting h-cateninstability. To test this, we compared the ability of the KI-

CK2a to antagonize axis duplication by wild-type h-cateninand T393A h-catenin, a h-catenin with a mutation of the

major CK2 phosphorylation site. As seen in Fig. 6C, KI-

CK2a reduced the number of embryos with duplicated axes

when injected along with wild-type h-catenin but had no

effect on axis duplication by the T393A mutant, even when

injected at higher doses of mRNA. Immunoblotting to

confirm the expression of KI-CK2a and T393A h-cateninwas performed in all experiments (not shown). Thus,

phosphorylation of h-catenin at T393 is a mechanism by

which CK2 modulates Wnt signaling in Xenopus embryos.

CK2 activity is required for endogenous dorsal axis

formation

To determine whether CK2 plays a role in endogenous

axis formation, we expressed the kinase-inactive mutant

in the dorsal side of the developing embryo. Increasing

doses of the KI-CK2a mRNA were injected into the

dorsal equatorial region at stages 2–3. Fig. 7A shows

representative embryos obtained in these experiments.

Injected embryos, shown at stage 40, had shortened axes,

frequently lacking all of the most anterior structures (Fig.

7A). Although there was some variability in the extent of

the ventralization between different batches of embryos,

on average more than 70% of the embryos showed

ventralized phenotypes. In contrast, dorsal injection of

water or mRNA encoding lacZ mRNA had no effect on

endogenous axis formation. The embryos obtained in four

experiments were pooled and their phenotypes were

ranked according to the DAI scale (Fig. 7B). The

embryonic phenotypes obtained ranged from DAI 0 to

DAI 4 with an average DAI of 3.6. Immunoblotting to

confirm the expression of KI-CK2a was performed in all

experiments (not shown).

We next correlated ventralization with CK2 activity.

Embryos injected with KI-CK2a mRNA were harvested

at stage 12 and CK2 was immunoprecipitated with anti-

CK2 antibody. Fig. 7C shows that CK2 protein was

immunoprecipitated from both control embryos (C) and

embryos overexpressing KI-CK2a (KI). Embryos that had

been injected with KI-CK2a mRNA had much lower

CK2 kinase activity than control embryos, consistent with

a dominant-negative effect of the KI mutant (Fig. 7C).

Moreover, we found that that CK2h co-immunoprecipi-

tates with both endogenous CK2a and KI-CK2a,

Fig. 5. CK2 stabilizes h-catenin in vivo. (A) Embryos were injected in two ventrovegetal (left) or dorsovegetal (right) blastomeres at stage 3 with 208 pg of

purified GST-h-catenin as depicted in the scheme shown at the top; the dorsal region of h-catenin up-regulation is shaded. Embryos were collected at the

indicated times (in minutes) after injection for immunoblot analysis. GST-h-catenin is quickly degraded ventrally while it is more stable dorsally. This

experiment was repeated three times with similar results. (B) Here, embryos were injected with purified GST-h-catenin alone or with 0.3 (+) or 1.25 (++) U of

purified CK2. Embryos were collected at the indicated times after injection for immunoblot analysis. The addition of CK2 to the ventral side of the embryo

stabilizes h-catenin to the same level as in the dorsal side. Tubulin is shown as a loading control. The figure shows one representative experiment out of three.

(C) Quantitation of the experiment in panel B using the Fluor-S. The amount of protein at time 0 was set as 1.0, and the change in normalized protein levels

over time is displayed (D = Dorsal, V = Ventral, An = Animal, Vg = Vegetal).


indicating that the KI mutant, like the wild-type enzyme,

is able to form tetrameric complexes (Fig. 7C). It is

likely that the residual kinase activity in the KI-CK2a

immunoprecipitates is due to endogenous CK2a mole-

cules that form tetrameric complexes with the overex-

pressed KI-CK2a and with endogenous CK2h. These

experiments suggest that CK2 activity is necessary for

endogenous dorsal axis formation in Xenopus embryos.

Discussion

In diverse organisms, the formation of the dorsal axis of

the embryo involves up-regulation of h-catenin, the tran-

scriptional coactivator of canonical Wnt signaling (Holland,

2002), but the mechanism of h-catenin up-regulation is not

clearly understood. Recent biochemical experiments have

shown that protein kinase CK2 is associated with and

Fig. 6. KI-CK2a inhibits ectopic axis formation promoted by Xwnt8 and WT but not mutant h-catenin. (A) To demonstrate that KI-CK2a has no enzymatic

activity, control embryos (C) and embryos injected with 3 ng of myc-CK2a mRNA (WT) or myc-KI-CK2a mRNA (KI) were homogenized at stage 12. Ten

embryos per treatment were subjected to immunoprecipitation with anti-myc antibody and used in an in vitro kinase assay. Immunoblots of myc-CK2a and

myc-KI-CK2a proteins (IB, upper panel) and in vitro CK2 peptide phosphorylation (KINASE, lower panel) are shown. A control immunoprecipitation without

extract (�) was also performed. (B) Two ventrovegetal blastomeres at stages 2–3 were injected with lacZ mRNA, 0.34 pg of Xwnt8 mRNA, or 0.34 pg of

Xwnt8 mRNA together with 1.6 ng of KI-CK2a mRNA. Representative morphology of the XWnt8 injected embryos is presented at stage 41. Control embryos

were indistinguishable from uninjected embryos (not shown). KI-CK2a was able to suppress ectopic axis formation by Xwnt8. Similar phenotypes were

obtained with h-catenin, as quantitated in the table in panel C. (C) Data from multiple axis duplication experiments were pooled to quantitate proportions of

single, partial, and double axes. Phenotypes were assessed at stages 38–41; numbers shown at the top of each bar indicate the number of embryos in each

category. Black bar, percentage of embryos with complete axis duplication (secondary axis with head, cement gland, and eyes); gray bar, percentage of embryos

with incomplete axis duplication (secondary axis lacking head structures); white bar, percentage of normal embryos. The quantities of mRNA injected were

0.34 pg XWnt8, 0.25 ng T393A h-catenin (T393A), 0.38 ng wild-type h-catenin (WT). The KI-CK2a mRNAwas used at 1.6 ng in the injections with XWnt8

and T393A h-catenin, and at 0.4 ng with wild-type h-catenin.


phosphorylates h-catenin in mammalian cells (Song et al.,

2000). Moreover, inhibition of CK2 results in decreased

stability of h-catenin and reduced TCF reporter trans-

activation in transient transfection assays (Song et al., 2000,

2003). We hypothesized that CK2 might be involved in

regulating h-catenin during dorsal axis formation in

Xenopus embryos. Consistent with this hypothesis, we

found that CK2 protein was expressed maternally in the

appropriate location in the embryo. Its overexpression was

sufficient to induce a complete ectopic axis. In doing so,

CK2 mimicked Wnt effects by stabilizing h-catenin and

inducing h-catenin/TCF target genes. Furthermore, a

kinase-inactive mutant of CK2a inhibit ectopic axis

formation promoted by exogenous Xwnt8 and it also

inhibited endogenous dorsal axis formation. Thus, CK2

meets criteria for being a positive regulator of Wnt

signaling and dorsal axis development in Xenopus embryos.

It is possible that CK2 plays other developmental roles in

embryogenesis, in addition to dorsal axis specification. CK2

phosphorylates many developmentally expressed gene

Fig. 7. Inhibition of normal dorsoanterior development by KI-CK2a. (A) Two ventrovegetal blastomeres at stages 2–3 were injected with 2 ng lacZ mRNA

(upper embryo) or 2 ng of KI-CK2a mRNA (other embryos). Morphology of the injected embryos is presented at stage 38. LacZ-injected embryos are

indistinguishable from uninjected embryos (not shown). Inhibition of the endogenous dorsal axis by KI-CK2a mRNA expression is demonstrated by the

absence of the most anterior structures (eyes and cement gland) and the trunk. Embryos are shown in increasing degree of ventralization. (B) Embryonic

phenotypes were scored at stage 38 and categorized according to the DAI scale of Kao and Elinson (1988), DAI = 5 being a normal embryo and DAI = 0 a

completely ventralized one. Numbers shown at the top of each bar indicate the number of embryos in each category. (C) Expression of KI-CK2a mRNA

inhibits endogenous CK2 activity in vivo. Ten embryos were injected with 3 ng of KI-CK2a (KI) or nothing (C) and subjected to immunoprecipitation with

anti-CK2a followed by immunoblotting (IB, upper two panels) or in vitro kinase assay (KINASE, lower panel). A control immunoprecipitation without extract

(�) was also performed. Endogenous CK2h (middle panel) co-immunoprecipitates with both endogenous CK2a (first lane) and with KI-CK2a (second lane).

Expression of KI-CK2a reduces CK2 kinase activity in the immunoprecipitates (lower panel).


products, including ANTP, cactus, Hsix-1, mef2c, xfg5.1,

nkx2.5, even-skipped, cut, and hoxb-6 (Meggio and Pinna,

2003). The regulatory h subunit is essential for early

embryonic viability, as has been demonstrated in multiple

systems. Gene targeting of the single h subunit in mice and

complete disruption of the CK2h gene in Drosophila are

early embryonic lethal (Buchou et al., 2003; Jauch et al.,

2002), although a viable hypomorphic allele of Drosophila

CK2h was found in a screen for adult fly brain defects

(Jauch et al., 2002). In sea urchins, CK2 activity peaks

during blastula stages and remains high until the pluteus

stage. In these embryos, inhibitors of CK2 block develop-

ment at the gastrula stages, suggesting that CK2 activity is

necessary for proper gastrulation and hatching of the

embryo (Delalande et al., 1999). Thus, CK2 may play a

role in multiple stages of embryogenesis or there may be

species-specific differences in how isozymes of CK2

function.

In Xenopus, two highly homologous genes coding for

CK2a (Jedlicki et al., 1992; Wilhelm et al., 2001) and one

for CK2h (Jedlicki et al., 1992) have been described.

Xenopus CK2a genes 1 and 2 code for proteins that differ

by three amino acids. Xenopus laevis is a tetraploid species;

thus, genes isolated in Xenopus often have two paralogs. No

gene encoding an alternative CK2a’ isoenzyme has been

found to date (Wilhelm et al., 2001), and we found no

evidence of a smaller CK2a’ subunit in our immunoblots;

thus, our experiments focused exclusively on the known

Xenopus CK2a and CK2h subunits. We showed that

ectopic expression of both XCK2a and XCK2h mRNAs

in the ventral side of the embryo was sufficient to induce an

ectopic axis and to increase expression of the Wnt target

genes Xsiamois, Xtwin, and Xnr3. This induction was

powerful enough to generate ectopic axes that were

complete in more than half of embryos, comparable to the

effects shown with ectopic expression of canonical Wnt

signaling pathway members including Xwnt8 (Smith and

Harland, 1991; Sokol et al., 1991), dsh (Sokol et al., 1995),

KI-GSK3h (Dominguez et al., 1995; He et al., 1995; Pierce

and Kimelman, 1995), and h-catenin (Guger and Gumbiner,


1995). At higher doses of mRNA, we obtained hyper-

dorsalized embryos, similar to those obtained with Xwnt8

(Christian et al., 1992), KI-GSK3h (Pierce and Kimelman,

1995), or by lithium chloride treatment (Kao and Elinson,

1988), further supporting the role of CK2 role in canonical

Wnt signaling (data not shown).

Due to the high degree of conservation between CK2

from different organisms (Jedlicki et al., 1992), we also

injected mouse and human forms of CK2a and CK2hsubcloned in mammalian expression vectors. Even though

we could synthesize mRNA in vitro, mammalian mRNA

was not efficiently translated in the Xenopus embryo and no

phenotype was obtained.

We find that balanced expression of both catalytic and

regulatory subunits of XCK2 was required for axis

duplication. This suggests that a key target substrate of

CK2 is recruited via binding to the h subunit, as occurs for a

subset of CK2 substrates such as p53 (Appel et al., 1995).

The fact that both subunits are required to transduce a Wnt

signal may contribute to explain why screens in Drosophila

based upon interruption of single genes have not uncovered

wingless phenotypes associated with CK2 subunit muta-

tions. Furthermore, there are multiple catalytic and regu-

latory subunit genes in flies that may be redundant in many

tissues. In mice, there is redundancy of the two catalytic

subunits in all organs except the testis, as CK2a’ null mice

are normal except for male germ cell development (Xu et

al., 1999).

To demonstrate that CK2 activity is required for

endogenous axis establishment, we generated a kinase-

inactive mutant of the catalytic subunit (KI-CK2a). A

similar KI-CK2a acts as a dominant negative in other

systems (Vilk et al., 1999). This mutant did not induce an

axis by itself or in combination with the CK2h subunit,

indicating that a functional enzyme is required for ectopic

axis induction. The KI-CK2a mutant did act as a

dominant negative in the embryos, as it was able to

suppress Xwnt8 axis effects, demonstrating that CK2

activity is required for canonical Wnt signaling. More-

over, ectopic expression of this mutant in the dorsal side

of the embryo resulted in ventralized embryos, indicating

that CK2 activity is necessary for normal dorsal axis

Fig. 8. Proposed model of CK2 mechanism of action during dorsoventral axis form

egg (left) and will move to the future dorsal side of the embryo during cortical rotat

of the embryo where CK2 is present and constitutively active. There, down-regul

phosphorylation of the dorsal determinants or their protein targets in the Wnt pathw

program (right). Region of CK2 expression in the upper two-thirds of the embry

formation. The ventralized phenotypes ranged between 0

and 4 in the DAI scale, similar to the results obtained

when GBP or h-catenin function is ablated, and a much

greater DAI effect than the phenotypes obtained by dorsal

expression of GSK3h, dominant-negative dsh, or domi-

nant-negative Wnt (Dominguez et al., 1995; Heasman et

al., 1994, 2000; Hoppler et al., 1996; Sokol, 1996; Yost

et al., 1998). Our data show that the KI-CK2a mutant

binds the endogenous regulatory subunit CK2h and

displaces the endogenous catalytically active CK2a from

the tetrameric form required for Wnt signaling (Fig. 7).

CK2 was able to stabilize wild-type h-catenin in vivo and

KI-CK2a inhibited axis duplication promoted by wild-

type h-catenin but not by h-catenin mutated in its major

CK2 phosphorylation site (T393A), indicating that CK2

regulates the Wnt pathway at least, but not necessarily

exclusively, through its action upon h-catenin.In conclusion, experiments here demonstrate that CK2

plays an essential role in the complex process of dorsal axis

determination. The current model of dorsal axis formation

proposes that in the fertilized embryo, h-catenin degrada-

tion is promoted by GSK3h phosphorylation. When the

dorsal determinants move dorsally during cortical rotation,

they promote signaling that stabilizes h-catenin. In cells in

culture, h-catenin is regulated by a destruction complex

that involves kinases, phosphatases, and scaffold proteins.

In the Xenopus embryo, we do not know how h-catenin is

dorsally up-regulated, as the nature of the dorsal determi-

nants is still unknown. However, recent experiments have

proposed that the dorsal determinants might include dsh

(Miller et al., 1999) and GBP (Weaver et al., 2003), two

Wnt transducers implicated in dorsal axis formation. These

translocate to the dorsal side of the embryo where GSK3his down-regulated (Dominguez and Green, 2000), h-cateninaccumulates (Larabell et al., 1997), and dsh is phosphory-

lated (Miller et al., 1999; Rothbacher et al., 2000). CK2,

expressed most highly in the medial part of the embryo

where h-catenin up-regulation occurs, may contribute to

Wnt signaling by phosphorylating and stabilizing h-cateninitself, perhaps through the major CK2 phosphorylation site

at T393 identified in mammalian cells (Song et al., 2003).

Alternatively, CK2 may act upon dsh, GBP, or other as yet

ation. Dorsal determinants are located in the vegetal cortex of the fertilized

ion (middle). This movement will bring the dorsal determinants to the region

ation of GSK3h by GBP, inhibition of the destruction complex by dsh, and

ay by CK2 (e.g., h-catenin, dsh, GBP) will initiate the dorsal developmental

o is shaded.


uncharacterized dorsal determinants. Indeed, previously,

our laboratory has found that CK2 inhibition in mammalian

cells leads to dsh instability (Song et al., 2000).

The lack of CK2 expression in the vegetal part of the

embryo is consistent with a model in which the constitu-

tively active CK2 kinase does not act until the physical

process of cortical rotation brings the dorsal determinants to

the waiting enzyme in the medial portion of the embryo

(Fig. 8). This model is consistent with vegetal cortex

transplantation studies that suggest that it is the combination

of the dorsal determinants plus uncharacterized factors in

the cytoplasm of the equatorial (medial) part of the embryo

that are required for dorsal determination (Kageura, 1997).

Future experiments will validate this hypothesis through a

more detailed analysis of the signaling pathway and of

biochemical studies of functional protein complexes and

their localization in the developing embryo.

Acknowledgments

We thank the members of Karen Symes laboratory for

their help with the Xenopus experiments. We thank Kerwin

Tang for help with immunoblots and data analysis. We

thank Sergei Sokol for providing the Xwnt8 and pJ3M

plasmids used in this study. We thank Tamara Holowacz,

Talat Malik, and Keiji Itoh for suggestions with the in situ

protocol. We thank Mary Williams and Anne Hinds for

assistance with embryo sectioning. We thank members of

the Seldin laboratory, Sergei Sokol, Tamara Holowacz, and

Michael Stanbrough for helpful discussions and critical

reading of the manuscript. Supported by ACS grant IRG-72-

001-28 (I.D.), a pilot research grant from the Department of

Medicine at Boston University School of Medicine (I.D.),

NIH R01 CA87375 (K.S.), NIH R01 CA71796 (D.S.), and

P01 ES11624 (D.S.). D.S. is a Scholar of the Leukemia and

Lymphoma Society.

References

Allende, J.E., Allende, C.C., 1995. Protein kinases. 4. Protein kinase CK2:

an enzyme with multiple substrates and a puzzling regulation. FASEB J.

9, 313–323.

Appel, K., Wagner, P., Boldyreff, B., Issinger, O.G., Montenarh, M., 1995.

Mapping of the interaction sites of the growth suppressor protein p53

with the regulatory beta-subunit of protein kinase CK2. Oncogene 11,

1971–1978.

Bek, S., Kemler, R., 2002. Protein kinase CKII regulates the interaction of

beta-catenin with alpha-catenin and its protein stability. J. Cell Sci. 115,

4743–4753.

Buchou, T., Vernet, M., Blond, O., Jensen, H.H., Pointu, H., Olsen, B.B.,

Cochet, C., Issinger, O.G., Boldyreff, B., 2003. Disruption of the

regulatory beta subunit of protein kinase CK2 in mice leads to a cell-

autonomous defect and early embryonic lethality. Mol. Cell. Biol. 23,

908–915.

Chen-Wu, J.L., Padmanabha, R., Glover, C.V., 1988. Isolation, sequencing,

and disruption of the CKA1 gene encoding the alpha subunit of yeast

casein kinase II. Mol. Cell. Biol. 8, 4981–4990.

Christian, J.L., Olson, D.J., Moon, R.T., 1992. Xwnt-8 modifies the

character of mesoderm induced by bFGF in isolated Xenopus ectoderm.

EMBO J. 11, 33–41.

Delalande, C., Belle, R., Cormier, P., Mulner-Lorillon, O., 1999. Transient

increase of a protein kinase activity identified to CK2 during sea urchin

development. Biochem. Biophys. Res. Commun. 266, 425–431.

DeMarais, A.A., Moon, R.T., 1992. The armadillo homologs beta-catenin

and plakoglobin are differentially expressed during early development

of Xenopus laevis. Dev. Biol. 153, 337–346.

Dominguez, I., Green, J.B.A., 2000. Dorsal downregulation of GSK3(beta)

by a non-Wnt-like mechanism is an early molecular consequence of

cortical rotation in early Xenopus embryos. Development 127, 861–868.

Dominguez, I., Green, J.B., 2001. Missing links in GSK3 regulation. Dev.

Biol. 235, 303–313.

Dominguez, I., Itoh, K., Sokol, S.Y., 1995. Role of glycogen synthase

kinase 3 beta as a negative regulator of dorsoventral axis formation in

Xenopus embryos. Proc. Natl. Acad. Sci. U. S. A. 92, 8498–8502.

Elinson, R.P., Holowacz, T., 1995. Specifying the dorsoanterior axis in

frogs: 70 years since Spemann and Mangold. In: Pedersen, R.A.,

Schatten, G.P. (Eds), Current Topics in Developmental Biology, vol. 30,

pp. 253–285.

Faust, M., Montenarh, M., 2000. Subcellular localization of protein kinase

CK2. A key to its function? Cell Tissue Res. 301, 329–340.

Fujisue, M., Kobayakawa, Y., Yamana, K., 1993. Occurrence of dorsal axis-

inducing activity around the vegetal pole of an uncleaved Xenopus egg

and displacement to the equatorial region by cortical rotation. Develop-

ment 118, 163–170.

Funayama, N., Fagotto, F., McCrea, P., Gumbiner, B.M., 1995. Embryonic

axis induction by the armadillo repeat domain of beta-catenin: evidence

for intracellular signaling. J. Cell Biol. 128, 959–968.

Giles, R.H., van Es, J.H., Clevers, H., 2003. Caught up in a Wnt storm: Wnt

signaling in cancer. Biochim. Biophys. Acta 1653, 1–24.

Guger, K.A., Gumbiner, B.M., 1995. beta-Catenin has Wnt-like activity and

mimics the Nieuwkoop signaling center in Xenopus dorsal–ventral

patterning. Dev. Biol. 172, 115–125.

Harland, R.M., 1991. In situ hybridization: an improved whole-mount

method for Xenopus embryos. Methods Cell Biol. 36, 685–695.

He, X., Saint-Jeannet, J.P., Woodgett, J.R., Varmus, H.E., Dawid, I.B.,

1995. Glycogen synthase kinase-3 and dorsoventral patterning in

Xenopus embryos. Nature 374, 617–622.

He, X., Semenov, M., Tamai, K., Zeng, X., 2004. LDL receptor-related

proteins 5 and 6 in Wnt/beta-catenin signaling: arrows point the way.

Development 131, 1663–1677.

Heasman, J., Crawford, A., Goldstone, K., Garner-Hamrick, P., Gumbiner,

B., McCrea, P., Kintner, C., Noro, C.Y., Wylie, C., 1994. Overexpression

of cadherins and underexpression of beta-catenin inhibit dorsal meso-

derm induction in early Xenopus embryos. Cell 79, 791–803.

Heasman, J., Kofron, M., Wylie, C., 2000. Beta-catenin signaling activity

dissected in the early Xenopus embryo: a novel antisense approach.

Dev. Biol. 222, 124–134.

Holland, L.Z., 2002. Heads or tails? Amphioxus and the evolution of

anterior-posterior patterning in deuterostomes. Dev. Biol. 241, 209–228.

Holowacz, T., Elinson, R.P., 1993. Cortical cytoplasm, which induces

dorsal axis formation in Xenopus, is inactivated by UV irradiation of the

oocyte. Development 119, 277–285.

Homma, M.K., Li, D., Krebs, E.G., Yuasa, Y., Homma, Y., 2002.

Association and regulation of casein kinase 2 activity by adenomatous

polyposis coli protein. Proc. Natl. Acad. Sci. U. S. A. 99, 5959–5964.

Hoppler, S., Brown, J.D., Moon, R.T., 1996. Expression of a dominant-

negative Wnt blocks induction of MyoD in Xenopus embryos. Genes

Dev. 10, 2805–2817.

Jauch, E., Melzig, J., Brkulj, M., Raabe, T., 2002. In vivo functional

analysis of Drosophila protein kinase casein kinase 2 (CK2) beta-

subunit. Gene 298, 29–39.

Jedlicki, A., Hinrichs, M.V., Allende, C.C., Allende, J.E., 1992. The

cDNAs coding for the alpha- and beta-subunits of Xenopus laevis

casein kinase II. FEBS Lett. 297, 280–284.


Kageura, H., 1997. Activation of dorsal development by contact between

the cortical dorsal determinant and the equatorial core cytoplasm in

eggs of Xenopus laevis. Development 124, 1543–1551.

Kao, K.R., Elinson, R.P., 1988. The entire mesodermal mantle behaves as

Spemann’s organizer in dorsoanterior enhanced Xenopus laevis

embryos. Dev. Biol. 127, 64–77.

Larabell, C.A., Torres, M., Rowning, B.A., Yost, C., Miller, J.R., Wu, M.,

Kimelman, D., Moon, R.T., 1997. Establishment of the dorso-ventral

axis in Xenopus embryos is presaged by early asymmetries in beta-

catenin that are modulated by the Wnt signaling pathway. J. Cell Biol.

136, 1123–1136.

Laurent, M.N., Blitz, I.L., Hashimoto, C., Rothbacher, U., Cho, K.W.,

1997. The Xenopus homeobox gene twin mediates Wnt induction of

goosecoid in establishment of Spemann’s organizer. Development 124,

4905–4916.

Lee, J.S., Ishimoto, A., Yanagawa, S., 1999. Characterization of mouse

dishevelled (Dvl) proteins in Wnt/Wingless signaling pathway. J. Biol.

Chem. 274, 21464–21470.

Lickert, H., Bauer, A., Kemler, R., Stappert, J., 2000. Casein kinase II

phosphorylation of E-cadherin increases E-cadherin/beta-catenin

interaction and strengthens cell–cell adhesion. J. Biol. Chem. 275,

5090–5095.

Maizel, A., Tassetto, M., Filhol, O., Cochet, C., Prochiantz, A., Joliot, A.,

2002. Engrailed homeoprotein secretion is a regulated process.


Marikawa, Y., Li, Y., Elinson, R.P., 1997. Dorsal determinants in the

Xenopus egg are firmly associated with the vegetal cortex and behave

like activators of the Wnt pathway. Dev. Biol. 191, 69–79.

Meggio, F., Pinna, L.A., 2003. One-thousand-and-one substrates of protein

kinase CK2? FASEB J. 17, 349–368.

Miller, J.R., Rowning, B.A., Larabell, C.A., Yang-Snyder, J.A., Bates, R.L.,

Moon, R.T., 1999. Establishment of the dorsal–ventral axis in Xenopus

embryos coincides with the dorsal enrichment of dishevelled that is

dependent on cortical rotation. J. Cell Biol. 146, 427–437.

Miravet, S., Piedra, J., Miro, F., Itarte, E., Garcia de Herreros, A., Dunach,

M., 2002. The transcriptional factor Tcf-4 contains different binding

sites for beta-catenin and plakoglobin. J. Biol. Chem. 277, 1884–1891.

Newport, J., Kirschner, M., 1982. A major developmental transition in early

Xenopus embryos: I. characterization and timing of cellular changes at

the midblastula stage. Cell 30, 675–686.

Nieuwkoop, P.D., Faber, J., 1967. Normal Table of Xenopus laevis. North-

Holland, Amsterdam.

Nusse, R., 2001. The Wnt gene Homepage. In http://www.stanford.edu/

~rnusse/wntwindow.html.

Nuthall, H.N., Husain, J., McLarren, K.W., Stifani, S., 2002. Role for Hes1-

induced phosphorylation in Groucho-mediated transcriptional repres-

sion. Mol. Cell. Biol. 22, 389–399.

Padmanabha, R., Chen-Wu, J.L., Hanna, D.E., Glover, C.V., 1990.

Isolation, sequencing, and disruption of the yeast CKA2 gene: casein

kinase II is essential for viability in Saccharomyces cerevisiae. Mol.

Cell. Biol. 10, 4089–4099.

Pierce, S.B., Kimelman, D., 1995. Regulation of Spemann organizer

formation by the intracellular kinase Xgsk-3. Development 121,

755–765.

Rothbacher, U., Laurent, M.N., Deardorff, M.A., Klein, P.S., Cho, K.W.,

Fraser, S.E., 2000. Dishevelled phosphorylation, subcellular local-

ization and multimerization regulate its role in early embryogenesis.

EMBO J. 19, 1010–1022.

Schagger, H., von Jagow, G., 1987. Tricine-sodium dodecyl sulfate-

polyacrylamide gel electrophoresis for the separation of proteins in

the range from 1 to 100 kDa. Anal. Biochem. 166, 368–379.

Schneider, V.A., Mercola, M., 1999. Spatially distinct head and heart

inducers within the Xenopus organizer region. Curr. Biol. 9, 800–809.

Schneider, H.R., Reichert, G.H., Issinger, O.G., 1986. Enhanced casein

kinase II activity during mouse embryogenesis. Identification of a 110-

kDa phosphoprotein as the major phosphorylation product in mouse

embryos and Krebs II mouse ascites tumor cells. Eur. J. Biochem. 161,

733–738.

Schohl, A., Fagotto, F., 2003. A role for maternal beta-catenin in early

mesoderm induction in Xenopus. EMBO J. 22, 3303–3313.

Schroeder, K.E., Condic, M.L., Eisenberg, L.M., Yost, H.J., 1999.

Spatially regulated translation in embryos: asymmetric expression of

maternal Wnt-11 along the dorsal-ventral axis in Xenopus. Dev. Biol.

214, 288–297.

Sells, M.A., Chernoff, J., 1995. Epitope-tag vectors for eukaryotic protein

production. Gene 152, 187–189.

Semplici, F., Meggio, F., Pinna, L.A., Oliviero, S., 2002. CK2-dependent

phosphorylation of the E2 ubiquitin conjugating enzyme UBC3B

induces its interaction with beta-TrCP and enhances beta-catenin

degradation. Oncogene 21, 3978–3987.

Serres, M., Filhol, O., Lickert, H., Grangeasse, C., Chambaz, E.M.,

Stappert, J., Vincent, C., Schmitt, D., 2000. The disruption of adherens

junctions is associated with a decrease of E-cadherin phosphorylation

by protein kinase CK2. Exp. Cell Res. 257, 255–264.

Smith, W.C., Harland, R.M., 1991. Injected Xwnt-8 RNA acts early in

Xenopus embryos to promote formation of a vegetal dorsalizing center.

Cell 67, 753–765.

Sokol, S.Y., 1996. Analysis of Dishevelled signalling pathways during

Xenopus development. Curr. Biol. 6, 1456–1467.

Sokol, S.Y., 1999. Wnt signaling and dorso-ventral axis specification in

vertebrates. Curr. Opin. Genet. Dev. 9, 405–410.

Sokol, S., Christian, J.L., Moon, R.T., Melton, D.A., 1991. Injected Wnt

RNA induces a complete body axis in Xenopus embryos. Cell 67,

741–752.

Sokol, S.Y., Klingensmith, J., Perrimon, N., Itoh, K., 1995. Dorsalizing and

neuralizing properties of Xdsh, a maternally expressed Xenopus

homolog of dishevelled. Development 121, 3487.

Song, D.H., Sussman, D.J., Seldin, D.C., 2000. Endogenous protein kinase

CK2 participates in Wnt signaling in mammary epithelial cells. J. Biol.

Chem. 275, 23790–23797.

Song, D.H., Dominguez, I., Mizuno, J., Kaut, M., Mohr, S.C.,

Seldin, D.C., 2003. CK2 phosphorylation of the armadillo repeat

region of beta-catenin potentiates Wnt signaling. J. Biol. Chem.

278, 24018–24025.

Vilk, G., Saulnier, R.B., St Pierre, R., Litchfield, D.W., 1999. Inducible

expression of protein kinase CK2 in mammalian cells. Evidence for

functional specialization of CK2 isoforms. J. Biol. Chem. 274,

14406–14414.

Weaver, C., Farr III, G.H., Pan, W., Rowning, B.A., Wang, J., Mao, J., Wu,

D., Li, L., Larabell, C.A., Kimelman, D., 2003. GBP binds kinesin light

chain and translocates during cortical rotation in Xenopus eggs.


Wilhelm, V., Neckelman, G., Allende, J.E., Allende, C.C., 2001. The

genomic structure of two protein kinase CK2alpha genes of Xenopus

laevis and features of the putative promoter region. Mol. Cell. Biochem.

227, 175–183.

Willert, K., Brink, M., Wodarz, A., Varmus, H., Nusse, R., 1997. Casein

kinase 2 associates with and phosphorylates dishevelled. EMBO J. 16,

3089–3096.

Wodarz, A., Nusse, R., 1998. Mechanisms of Wnt signaling in develop-

ment. Annu. Rev. Cell Dev. Biol. 14, 59–88.

Xu, X., Toselli, P.A., Russell, L.D., Seldin, D.C., 1999. Globozoospermia

in mice lacking the casein kinase II alpha’ catalytic subunit. Nat. Genet.

23, 118–121.

Yost, C., Farr III, G.H., Pierce, S.B., Ferkey, D.M., Chen, M.M., Kimelman,

D., 1998. GBP, an inhibitor of GSK-3, is implicated in Xenopus

development and oncogenesis. Cell 93, 1031–1041.

Zhang, Y., Qiu, W.J., Chan, S.C., Han, J., He, X., Lin, S.C., 2002. Casein

kinase I and casein kinase II differentially regulate axin function in Wnt

and JNK pathways. J. Biol. Chem. 277, 17706–17712.

http:\\www.stanford.edu\~rnusse\wntwindow.html

protein kinase ck2 is required for dorsal axis formation in xenopus embryos

Documents