Molecular characterization of a novel nuclear transglutaminase
that is expressed during starfish embryogenesis
Hiroyuki Sugino*, Yudai Terakawa, Akiko Yamasaki, Kazuhiro Nakamura, Yoshiaki Higuchi,
Juro Matsubara, Hisato Kuniyoshi and Susumu Ikegami
Department of Applied Biochemistry, Hiroshima University, Japan
We report the constitution and molecular characterization
of a novel tran sglutaminase (EC 2.3.2.13) that starts to
accumulate specifically in the nucleus in the starfish (Asterina
pectinifera) embryo after progression through the early
blastula stage. The cDNA for the nuclear transglutaminase
was cloned and the cDNA-deduced sequence defines a single
open reading frame encoding a protein with 737 amino acids
and a predicted molecular mass of 83 kDa. A comparison of
this transglutaminase with other members of the gene family
revealed an overall sequence identity of 33–41%. A special
sequence feature of this transglutaminase, which is not found
in other transglutaminases, is t he presence of nuclear local-
ization signal-like sequences in the N-terminal region.
Microinjection of hybrid constructs that encode the N-ter-
minal segment fused to reporter proteins into the germinal
vesicle of an oocyte produced chimeric proteins by
transcription-coupled translation. It was foun d that the
N-terminal segmen t alone was sufficient t o effect nuclear
accumulation of an otherwise cytoplasmic protein. These
results suggest that the nuclear accumulation of the trans-
glutaminase may play an important role in nuclear remod-
eling during early starfish embryogenesis.
Keywords: transglutaminase; nucleus; starfish; e mbryo;
cloning.
The class of enzymes that are commonly referred to as
transglutaminases (TG) (EC 2.3.2.13) are known mostly for

that is localized exclusively in the nucleus of starfish
(Asterina pectinifera) embryonic cells and is designated
nuclear TG (nTG). The amino-acid sequence derived from
the cDNA sequence contains putative nuclear localization
signals [15] in the N -terminal region. We demonstrate here
that the N-terminal region promotes the nuclear accumu-
lation of an otherwise cytoplasmic protein, namely pyruvate
kinase (PK), in t he A. pectinifera oocyte system. This
finding suggests that nuclear localization signals in the
N-terminal region of nTG are functional in the starfish
embryonic cells. Northern b lot analyses carried out in this
study demonstrate that nTG mRNA appears at the early
blastula stage a nd increases thereafter. The nTG protein
level inc reases in parallel w ith m RNA levels. These results
suggest that nTG is, directly or indirectly, involved in the
modification of the nuclear structure or intranuclear
signaling pathways during starfish embryogenesis [16–18].
MATERIALS AND METHODS
Cultivation of embryos
Specimens of the starfish, A. pec tinifera, were collected from
coastal waters off Japan during their breeding season and
maintained in artificial sea water in laboratory aquaria at
Correspondence to S. Ikegami, Department of Applied Biochemistry,
Hiroshima University, 1-4-4 Kagamiyama, Higashi-hiroshima,
Hiroshima 739-8528, Japan.
Fax: + 81 824 22 7059, Tel.: + 81 824 24 7948,
E-mail: [email protected]
Abbreviations: nTG, nuclear transglutaminase; GFP, g reen fluorescent
protein; PK, pyruvate kinase; TG, transglutaminase.
*Present address: Department of Applied Life Science, Faculty of

M
Tris/HCl (pH 8.0), 10 m
M
KCl, 6 m
M
(NH
4
)
2
SO
4
, 0.1% Triton X-100, 0.001% BSA,
1m
M
MgCl
2
,0.2m
M
each of four deoxyribonucleoside
5¢-triphosphates, and 4 l
M
each of the TG-specific degen-
erate oligonucleotide primers, TG5 (5¢-TAYGGNCARTG
YTGGGT-3¢;N¼ A, C, G or T; Y ¼ CorT;R¼ Aor
G) and TG 3V (5¢-CCANACRTGRAARTTCCA-3¢). The
PCR cycles were 15 s at 98 °C, 2 s at 55 °C, and 10 s at
74 °C. A total of 25 cycles were run, with the first cycle
containing an extended denaturation p eriod (2 min). The
195-bp PCR product was gel-purified and sequenced by
means of the dideoxy chain termination method using the

carried out using TG3-3 (5¢-ATCGTGTCGCTGACCAA
C-3¢)andTG3-4(5¢-CC ATTGCCGTACCCGCTG-3¢), the
sequences of which were d erived from the determined
internal region, and the adaptor-specific primers AP1 and
AP2. TG-specific primers designed from the 5¢ and 3 ¢ ends
of the obtained products, 5¢GSP1 (5¢-CGATTACAGTCG
TGGTCAGAGCTG-3¢), 5¢GSP2 (5¢-TCGTGGTCAGAG
CTGTTGTTTGTG-3¢), 3¢GSP1 (5¢-CAAGGACTGACC
TTCACTGAGATG-3¢)and3¢GSP2 (5¢-GTGGCGTTGG
GATGCAACATTGTG-3¢), were used to amplify the full-
length cDNA, and the BamHI (5¢-GCGGATCCATGGTT
CGTCGATCCACTCGC-3¢)andNotI primers (5¢-CT
GCGGCCGCTTAAGCACTCTTGACATTGAG-3¢)to
amplify the coding sequence (Fig. 1).
RNA isolation and Northern blot hybridization
Samples of poly(A)
+
RNA (0.5 lg) were prepared from
staged embryos as described previously [22]. They were
denatured and separated by formaldehyde gel electropho-
resis and transferred to nylon filters (Amersham Pharmacia
Biotech). The blots were hybridized overnight at 42 °Cin
hybridization buffer with a probe and washed according to
the manufacture’s recommended protocol. Digoxygenin-
labeled antisense RNA p robes were p repared from a
linearized plasmid DNA template, which contained a
0.27-kbp StuI–NotI restriction fragment of nTG cDNA or
0.15-kbp BamHI–EcoRI restrictio n fragmen t of A. pecti-
nifera ubiquitin cDNA (H. Sugino, unpublished data) using
the digoxygenin-RNA labeling kit (Roche Molecular Bio-

isopropyl thio-b-
D
-galactoside. Bacteria were lysed in 1%
Triton X-100 in NaCl/P
i
, sonicated with six bursts of 10 s,
and incubated at 4 °C for 1 h. Insoluble materials were
removed by centrifugation at 13 000 g for 10 min. Gluta-
thione S-transferase-conjugated nTG w as purified from th e
supernatant using glutathione–Sepharose 4B beads (Amer-
sham Pharmacia Biotech) essentially following the protocol
provided by the manufacturer.
Biochemical fractionation of embryos
Embryos were washed with ice-cold solution 1 [0.25
M
sucrose, 10 m
M
Tris/HCl (pH 8.0), and 0 .1 m
M
EDTA].
They were then resuspended in the same volume of solution
1, to which had been added 0.15 m
M
spermine and 0.5 m
M
spermidine. The suspension was homogenized by 10 strokes
with a Dounce homogenizer. To the homogenate was added
1.3 v ol. of 2.0
M
sucrose, 65 m

(Worthington Biochemical) and 2 lgÆmL
)1
of ribonuc-
lease A (Sigma Chemicals) for 15 min at 22 °C, followed by
centrifugation for 1 0 m in at 4 °C (20 000 g). The superna-
tant was collected and designated as Sup1. The pellet
obtained after this step was treated with 1% Triton X-100
and recentrifuged. T he supernatant was separated and
designated as Sup2. The pellet was resuspended in 25 m
M
Tris/HCl (pH 7.5), 1% Triton X-100, and 0.5
M
NaCl. This
suspension was incubated for 30 min at 4 °Candthen
centrifuged f or 10 min a t 20 000 g. T he supernatant was
separated and designated as Sup3. The pellet was resus-
pended in 10S buffer [50 m
M
Hepes/HCl (pH 7.2), 10 m
M
sodium phosphate, 250 m
M
NaCl, 0.3% Nonidet P-40,
0.1% Triton X-100, 0.005% SDS, 1 m
M
NaF, 0.5 m
M
dithiothreitol, a nd 0.1 m
M
phenylmethanesulfonyl fluoride]

solubilization in 4 mL of 50 m
M
Tris/HCl (pH 7.5), 8
M
urea, and 0.5% (w/v) SDS. The amount of incorporated
monodansylcadaverine was determined by measuring the
fluorescence o f the solubilized protein using a Shimazu
RF-540 fluorescence spectrophotometer with an excitation
wavelength of 340 nm, emission wavelength of 525 nm, and
a 5-nm slit. The instrument was calibrated with m ono-
dansylcadaverine in 50 m
M
Tris/HCl (pH 7.5), 8
M
urea,
and 0.5% (w/v) SDS prior to each run. One unit of enzyme
activity defined as AIU (amine incorporation unit per min)
was calculated as described previously [24].
Preparation of nTG-specific antibodies
Two portions of the putative amino acid sequence of nTG,
Leu-Asp-Tyr-His-Tyr-Asp-Glu-Asn-Ser-Glu-Pro-Leu-Asp-
Asp and Arg-Arg-Ser-Thr-Arg-Thr-Arg-Ser-Thr-Pro-Thr-
Arg-Phe-Gly-Tyr-Thr-Asp-Arg, were used to produce
nTG-specific polyclonal antibodies, anti-(nTG-M) Ig and
anti-(nTG-N) Ig, respectively. The peptides were synthe-
sized such that each of them contained an artificial Cys
residue at the N- or C-terminus, respectively, for coupling
purposes. E ach s ynthesized peptide was conjugated to
maleimide-activated keyhole limpet hemocyanin (Amer-
sham Pharmacia B iotech) a ccording to manufacturer’s

of affinity-purified Ig).
Immunoprecipitation
Concentrated Sup4 (10 lL) was incubated with the affinity-
purified anti-(nTG-N) Ig (3 lg) for 3 h at 4 °C in 400 lLof
IP buffer [50 m
M
Tris/HCl (pH 7.5), 150 m
M
NaCl, 0 .5%
Triton X-100, and 0.1% SDS]. After the incubation, 100 lL
of protein A–Sepharo se that had be en equilibrated in IP
buffer was added, and t he mixture w as then rotated
moderately for 1 h at 4 °C. Following centrifugation and
removal of the supernatant, the pellets were washed twice
with IP buffer, and resuspended with 4 00 lL of IP buffer
(total volume, 500 lL). C ontrol experiments were per-
formed using the affinity-purified anti-ANOC Ig, which was
raised against the C-terminal portion of ANO 39, a starfish
protein unrelated to nTG [22].
Ó FEBS 2002 Nuclear transglutaminase in starfish embryos (Eur. J. Biochem. 269) 1959
CGATTACAGTCGTGGTCAGAGCTGTTGTTTGTGTTCCTTGTAAATCGTAATCATCCAAA 59
ATGGTTCGTCGATCCACTCGCACCCGCAGCACCCCTACCCGCTTCGGCTACACCGACCGG 119
M V R R S T R T R S T P T R F G Y T D R
TTTGAGCCGTATGCCCGCAAGCCTAAACGGGAAACGACGCGCACAGAGGGGCGACGCTAC 179
F E P Y A R K P K R E T T R T E G R R Y
GTACCCGCCACACCACTGACTCTGCCTACGCTGAAAGAAAAAAAGACGCAACTCAAGGTG 239
V P A T P L T L P T L K E K K T Q L K V
GTGTCAGTTGATCTATGTGTGGAGCGAAACCAGCAGGAGCATAAGACCAGCAAGTACAAG 299
V S V D L C V E R N Q Q E H K T S K Y K
GTTGACAATCTGGTCCTGCGTCGTGGTCAACCGTTCCACCTCAATGTCAAGTTTGACCGA 359

F H V W N D C W M A R P D L E E G Y G G
TGGCAGGCCGTGGACGCAACCCCTCAGGAAACAAGCAACGGTGTGTACTGCATGGGACCT 1319
W Q A V D A T P Q E T S N G V Y C M G P
ACCTCTCTGCGCGCCATCAAGCAGGGTCACGTGTACATGCAGTATGACACCAAGTTTGCC 1379
T S L R A I K Q G H V Y M Q Y D T K F A
TTTGCTGAGGTCAACGCTGAAAAGGTCTACTGGAAGGTCTTCACGAAATCTAGAAAGGCC 1439
F A E V N A E K V Y W K V F T K S R K A
CCGGAGGTCATAGACATTGACTCCGATGATGTCGGATGCAAGATCAGCACCAAAGCCGTC 1499
P E V I D I D S D D V G C K I S T K A V
GGCAAATTTGAGCGTGAGGACATCACTGAGCAGTACAAGTACAAGGAAGGAACGGAGTTG 1559
G K F E R E D I T E Q Y K Y K E G T E L
GAGCGCATCGCCGTCAGAGAAGCCAGCCGTCATGTACGCAAAGCAAAGAGAATTCTCAAG 1619
E R I A V R E A S R H V R K A K R I L K
AACCTTGTCCGCGACGTGGACTTTGACGTGGACATGGCGGAGGAGTTCCCCATTGGGAAA 1679
N L V R D V D F D V D M A E E F P I G K
GATATCAAGTTCACTATCACTATGGTGAATAAGTCACAACAGACACGTAATGTCTTTCTG 1739
D I K F T I T M V N K S Q Q T R N V F L
GGTGTGACAGGAAGCACCGTGTACTACACAGGTGTTAAGAAGGCCAAGGTGTCATCCTAC 1799
G V T G S T V Y Y T G V K K A K V S S Y
AATGGCACCCTGCCACTGAAGGCAAAGGAAACGCGAGTGATTCCTGTGACTGTACCTGCG 1859
N G T L P L K A K E T R V I P V T V P A
TCTGACTACCTGCCGCAGCTCACTGACTATGCTGGCGTAACGTTCTTCATCATGGCTTCC 1919
S D Y L P Q L T D Y A G V T F F I M A S
GTCAAGGAGACCAAGCAACCATTCAGCAGGCAGTATGACGCCGTGCTTGATAAGCCTGAC 1979
V K E T K Q P F S R Q Y D A V L D K P D
CTGGAGGTCAAGACGGAGGGGCCCATTGTGCGTGGCAAGCCGTTCACAGCTATCGTGTCG 2039
L E V K T E G P I V R G K P F T A I V S
CTGACCAACCCATTGCCGTACCCGCTGACTGACTGCAGCCTACTTATGGAGGGGTCCATC 2099
L T N P L P Y P L T D C S L L M E G S I
ATTGAGGGCGCCAAACGGGTCAAAGCTCCACATGTTCCAGTGAACGGTAAGATGGCCCAG 2159

X-100 in NaCl/P
i
,theninNaCl/P
i
alone, then in acetone
()20 °C), and finally in NaCl/P
i
again. The samples were
blocked with 3% BSA in NaCl/P
i
for 30 min at 37 °C.
Incubations wi th primary and secondary antibodies were
carried out for 2 h at 37 °C. Monospecific anti-(nTG-M)
Ig (1.9 lgÆmL
)1
), which had been preincubated with the
antigenic peptide (1.1 lgÆmL
)1
of affinity-purified Ig), was
used as the negative control. The secondary antibody was
cEry MGGP 4
lHem MYGFGRGNMFRNRSTRYRRRPRYRAENYHSYMLDLLENMNEEFGRNWWGTPESHQPDS 58
nTG MVRRSTRTRSTPTRFGYTDRFEPYARKPKRETTRTEGRRYVPATPLTL 48
hKer MMDGPRSDVGRWGGNPLQPPTTPSPEPEPEPDGRSRRGGGRSFWARCCGCCSCRNAADDDWGPEPSDSRGRGSSSGTRRPGSRGSDSRRPVSRGSGVNAA 100

gpLiv MAEDLILERCDLQLEV NGRDHRTADLCRERLVLRRGQPFWLTLHFEGRGYEAGVDTLTFNAVTGPDPSEEAGTMARFSLSSAV EGGTW 88
cEry GPDGTMAEELVLETCDLQCER NGREHRTEEMGSQQLVVRRGQPFTITLNFAGRGYEEGVDKLAFDVETGPCPVETSGTRSHFTLTDCP EEGTW 97
lHem GPSSLQVESVELYTRDNAREH NTFMYDLVDGTKPVLILRRGQPFSIAIRFK-RNYNPQQDRLKLEIGFGQQPLITKGTLIMLPVSGSDTFTKDKTQW 154
nTG PTLKEKKTQLKVVSVDLCVER NQQEHKTSKYKVDNLVLRRGQPFHLNVKFD-RDFKPSTDELVLELRMGSRANVTKGTRCVAPVVTSAP DHDDW 141
hKer GDGTIREGMLVVNGVDLLSSRSDQNRREHHTDEYEYDELIVRRGQPFHMLLLLS RTYESSDRITLELLIGNNPEVGKGTHVIIPVGKGG SGGW 193

* . .* * *.*** . * .* . ** .*** ** **. ** . *

gpLiv -EAQEETGVAMRIRVGQNMTMGSDFDIFAYITNGTAESHECQLLLCARIVSYNGVLGPVCSTNDLLNLTLDPFSENSIPLH-ILYEKYGDYLTESNLIKV 566
cEry EQEEGLHMRIKLSEGANNGSDFDVFAFISNDTDKERECRLRLCARTASYNGEVGPQCGFKDLLNLSLQPHMEQSVPLR-ILYEQYGPNLTQDNMIKV 572
lHem LPSPEKEDVYFNLLDIEKIKIGQPFHVTVNIENQSSETRRVSAVLSASSIYYTGITGRKIKRENGN-FSLQPHQKEVLSIE-VTPDEYLEKLVDYAMIKL 643
nTG ILKNLVRDVDFDVDMAEEFPIGKDIKFTITMVNKSQQTRNVFLGVTGSTVYYTGVKKAKVSSYNGT-LPLKAKETRVIPVT-VPASDYLPQLTDYAGVTF 615
hKer N-RGSAEDVAMQVEAQDAVMG-QDLMVSVMLINHSSSRRTVKLHLYLSVTFYTGVSG-TIFKETKKEVELAPGASDRVTMP-VAYKEYRPHLVDQGAMLL 667
hPro HRRPVKENFLHMSVQSDDVLLGNSVNFTVILKRKTAALQNVNILGSFELQLYTGKKMAKLCDLNKTSQIQGQVSEVTLTLDSKTYINSLAILDDEPVIRG 563
. . . . . . . *.* . * . .
gpLiv RGLLIEPAANSYVLAERDIYLENPEIKIRVLGEPKQNRKLIAEVSLKNPLPVPLLGCIFTVEGAGLTKDQKSVEVPDPVEAGEQAKVRVDLLPTEVGLHK 666
cEry VALLTEYETGDSVVAIRDVYIQNPEIKIRILGEPMQERKLVAEIRLVNPLAEPLNNCIFVVEGAGLTEGQRIEELEDPVEPQAEAKFRMEFVPRQAGLHK 672
lHem YAIATVKETQQTWSEEDDFMVEKPNLELEIRGNLQVGTAFVLAISLTNPLKRVLDNCFFTIEAPGVTGAFR VTNRDIQPEEVAVHTVRLIPQKPGPRK 741
nTG FIMASVKETKQPFSRQYDAVLDKPDLEVKTEGPIVRGKPFTAIVSLTNPLPYPLTDCSLLMEGSIIEGAKR VKAPHVPVNGKMAQRVQLTPKTAGSCD 713
hKer NVSGHVKESGQVLAKQHTFRLRTPDLSLTLLGAAVVGQECEVQIVFKNPLPVTLTNVVFRLEGSGLQRPKI LNVGDIGGNETVTLRQSFVPVRPGPRQ 765
hPro FIIAEIVESKEIMASEVFTSFQYPEFSIELPNTGRIGQLLVCNCIFKNTLAIPLTDVKFSLESLGISSLQT SDHGTVQPGETIQSQIKCTPIKTGPKK 661
. . * . . *.* * . . .*. . . * * .
gpLiv LVVNFECDKLKAVKGYRNVIIGPA 690
cEry LMVDFESDKLTGVKGYRNVIIAPLPK 698
lHem IVATFSSRQLIQVVGSKQVEVLD 764
nTG LIVSFSSPQLSGVKAHVTLNVKSA 737
hKer LIASLDSPQLSQVHGVIQVDVAPAPGDGGFFSDAGGDSHLGETIPMASRGGA 817
hPro FIVKLSSKQVKEINAQKIVLITK 684
. . . .
B
Fig. 2. Nucleotide and deduced amino acid sequences of nTG. (A) The nucleotide sequence of the cDNA clone which encodes nTG and the amino
acid sequence deduced therefrom. (B) D educed amino-acid sequences for guinea pig liver TG (gpLiv) [ 5], chicken erythrocyte TG (cEry) [30],
Limulus hemocyte TG (lHem) [31], human keratinocyte TG (hKer) [3], human prostate TG (hPro) [6], and nTG are shown using the single letter
amino acid codes. Gaps have been inserted to achieve maximum similarity. Asterisks and dots at the bottom of the aligned sequences indicate
positions that are occupied by identical or chemically similar amino acids in all TG. The arrowhead indicates the active site Cys residue [31]. The
arrows indicate the positions of the H is and Asp residues of the c atalytic triad [35]. Putative nuclear localization signals [11] are underlined.

generatedviaPCRwithPK-N(5¢-GCCGGATCCGGC
CTCGAGATGCCCAAGCCAGACAGC-3¢)andPK-C
(5¢-GAGCGGCCGCTCATCAGCCGAGCTCTGGTAC
AGGCACTAC-3¢) primers. The PK fragment was digested
with XhoI and ligated with N57 fragment to give the
N57PK fragment. The nTG, nTGDN57, PK, and N57PK
fragments were separately ligated with the GFP fragment
and the vector fragment derived from KpnI/NotI-digestion
of pHEG to produce pHE-TG, pHE-TGDN57, pHE-PK,
and pHE-N57PK, respectively.
Expression of cloned cDNAs
in vivo
The constructs were separately dissolved in 10 m
M
Tris/HCl
(pH 8.5) to g ive a final concentration o f 200 ngÆlL
)1
.
Twenty picoliters of the solution, along with a small amount
of KF96 silicone oil (Shin-Etsu Chemical, Tokyo, Japan),
were then microinjected into the germinal vesicle of an
oocyte as described previously [22]. T hree to four hours
later, the injected oocytes we re examined for localization
of fluorescent proteins under a fluorescence microscope
equipped with differential interphase and epifluorescence
optics.
RESULTS
Molecular cloning of
A. pectinifera
transglutaminase

645
A
Fig. 3. Western blot analysis of nTG during embryogenesis. (A) Cyto-
solic fractions (lanes 1–3 and 7–9) and nuclear fractions (lanes 4–6 and
10–12) we re prepared from 8-h-old early b lastulae (lanes 1, 4, 7, and
10), 12-h-old mid-blastulae (lanes 2, 5, 8, and 11), and 24-h-old mid-
gastrulae (lanes 3, 6, 9, and 12). An a liquot of each fraction (60 lg
bovine serum albumin-equivalent per lane) was separated by SDS/
PAGE, and the gel was stained with Coomasssie blue (lanes 1–6) or
transferred to poly(vinylidene difluoride) membrane, followed by
staining using anti-(nTG-M) Ig as a probe (lanes 7–12). Sizes of the
molecular mass marker proteins in kDa are shown to the left. (B)
Nuclear fractions were prepared from 29- h-old midgastrulae (lan es 1
and 4), 40-h-old late gastrulae (lanes 2 and 5), and 51-h-old bipinnariae
(lanes 3 and 6). An aliquot of each fraction (3000 embryos-equivalent
per lane) was separated by SDS/PAGE, and the gel was stained with
Coomasssie blue (lanes 1–3) or transferred to poly(vinylidene difluo-
ride) membrane, followed b y s taining using an ti-(nTG-N) Ig a s a p robe
(lanes 4–6). S izes of the molecular mass marker prote ins in kD a are
shown to the left.
1962 H. Sugino et al. (Eur. J. Biochem. 269) Ó FEBS 2002
the sequence obtained by t he RACE experiments. The
cDNA contained a single open reading fram e ( ORF)
beginning with an ATG codon in an adequate context for
the initiation of translation (Fig. 2A); the sequence
CCAAAATGG s urrounding the A TG fits t he consensus
sequence CC(G/A)CCATGG f or the eukaryotic initiator
site [29]. The predicted protein consists of 737 amino acids,
with a molecular mass of 83 105 Da and an isoelectric point
of 7.9. Neither a polyadenylation signal (AATAAA) nor a

M
and 13.3 n molÆmin
)1
Æmg
)1
,
respectively, indicating that the predicted protein is a
transglutaminase.
Subcellular localization of
A. pectinifera
transglutaminase
A major characteristic feature of the A. pectinifera TG is the
presence of two putative nucle ar localization signals in the
N-terminal region, a monopartite (residues 26–30) and a
bipartite (residues 38–39 and 52–55) ones [12–14], suggest-
ing nuclear localization of this protein. To examine if the
A. pectinifera TG is a nuclear protein, we raised an
B
ab
cd
ef
gh
C
ab
cd
ef
gh
a
b
A

Ônuclear transglutaminase (nTG) Õ.
Expression pattern of nTG during embryogenesis
Early starfish development may be directed by two sources
of mRNA: (a) a pool stored in the immature oocyte
transcribed from the maternal genome during oogenesis
such as ANO39 mRNA [22], and (b) newly synthesized
mRNA transcribed from the embryonic genome. Northern
blot hybridization on poly(A)
+
RNA from blastulae and
gastrulae showed a gradual increase in the signal at 5.0-kb
during the progression of embryonic development (Fig. 5),
whereas hybridization on poly(A)
+
RNA from f ertilized
eggs showed little or no signals, s uggesting that nTG
mRNA belongs to the latter. The developmental Western
blot analysis revealed that the 90-kDa band corresponding
to the nTG protein was first detected in the mid-blastula
embryo and that the level of the band increased by t he
bipinnaria stage (Fig. 3A, lanes 10–11, Fig. 3B, lanes 4–6).
Therefore, the nTG protein s ynthesis starts at mid-blastula
stage and continues thereafter.
Immunostaining o f t he dissociated cells of embryos at
different developmental stages revealed a specific pattern of
accumulation. At the 8- to 12-h-old early blastula stages,
nTG was undetectable in the nucleus (Fig. 4C, a–d). The
nucleus of the early blastula is larger and looser than that of
embryos collected at later developmental stages (Figs 4C,
b,d,f,h). nTG starts to accumulate in the compact nucleus of

SDS/PAGE of Sup4 resulted in a prominent band with
an apparent molecular mass of 90 kDa (Fig. 6C, lane 1),
which was re cognized by the anti-(nTG-M) Ig in Western
blot analysis (Fig. 6C, lane 3). To determine if the TG
activity in Sup4 results from t he nTG protein, Sup4 was
subjected to immunoprecipitation with the antibody raised
against the N-terminal portion (Arg3–Arg20) of nTG [anti-
(nTG-N) Ig]. As a result, the TG activity was mainly
recovered in the immunoprecipitate (Fig. 7), showing t hat
the molecule, which predominantly generates the TG
activity in Sup4, is the nTG.
Identification of the segment containing nuclear
localization signals in nTG
To identify the elements(s) in nTG that determine nucleus-
specific topogenesis, we examined the localization of the
12345
10.0 -
4.0 -
3.0 -
6.0 -
Fig. 5. Expression of nTG gene. Northern blots of poly(A)
+
RNA
from fertilized eggs (lane 1), 8-h-old early blastulae (lane 2), 12-h-old
mid-blastulae (lane 3), 15-h-old late b lastulae (lane 4), and 24-h-old
midgastrulae (lane 5). The filter was hybridized with a digoxygenin-
labeled RNA probe obtained from the c DNA of nTG (upper pane l)
and of A. pectinifera ubiquitin (lower panel). Each lane was loaded
with 0.5 lgofpoly(A)
+

24-h-old midgastrulae (400 embryos-equivalent per lane) was sepa-
rated by SDS/PAGE, and the gel was stained with Coomasssie blue
(lane 1) or transferred to poly(vinylidene difluoride) membrane, fol-
lowed by staining using anti-(nTG-M) Ig as a p robe (lanes 3). As a
negative control, parallel immunoblotting was performed using anti-
(nTG-M) Ig preincubate d with the peptid e antigen (lane 2) . Sizes of the
molecular mass marker proteins in kDa are shown to the left.
100
50
0
Input
TG activity (% control)
SIPSIP
anti-nTG-N Control Ig
Fig. 7. Immunoprecipitation of nTG recovered in Sup4. Ten microliters
of concentrated Sup4 were subjected to immunoprecipitation wit h
anti-(nTG-N) Ig or control Ig (anti-ANOC Ig [22]). Total TG activity
recovered in the supernatants (S) or the immunoprecipitates (IP) was
measured, and is expressed as the percentage of the total activity in the
10 lL of concentrated Sup4 (Input). The results shown are the aver-
ages of three experiments. Error bars indicate plus one SEM.
A
nTGGFPhsp
nTG∆N57
hsp GFP
PKhsp GFP
N57hsp GFP PK
pHE-TG
pHE-TG∆N57
pHE-PK

encoding rat muscle PK [27] was engineered to include the
GFP sequence a nd the s equence of N57 that precedes the
fusion junction with PK. The construct, pHE-N57PK, was
microinjected into the germinal vesicle of an oocyte and the
subcellular localization of the expressed protein was moni-
tored. The results, as shown in Fig. 8B, clearly demonstrate
the ability of N57 to promote the nuclear accumulation of
PK (Figs 8B, d,h). Without N57, the expressed GFP–PK
fusion protein is located exclusively in the cytoplasm
(Figs 8 B, c,g).
DISCUSSION
During the early development of A. pectinifera, the embryo
undergoes e xtremely rapid cellular replication [16,18].
Slower rates of cell division characterize the embryo from
the early to mid-blastula stages. Concomitant with this rate
reduction, an increase in embryonic t ranscriptional activity
is also observed. The large swollen nuclei become smaller
and more compact, and dispersed chromatin becomes more
condensed. The present study demonstrates that nTG
initially appears in A . pectinifera embryos at t he mid-
blastula transition and that the level of the enzyme protein
increases gradually thereafter (Figs 3 and 4).
nTG is similar to t he TG of vertebrates and arthropods
[34]. Its molecular mass is within the 75–90-kDa range
known for the TG of these organisms [34]. The most unique
property of nTG, not found in other TGs, is that its
distribution is confined to the nucleus. Nuclear localization
of TG has been reported in the studies on tissue TG [8,9]. A
nuclear transport protein, importin-a3, has been shown to
be involved in the active transport of tissue TG into the

A. pectinifera embryos at the mid-blastula stage but not at
earlier stages (T. Shimizu & S. Ikemagi, unpublished
results). Although the formation of an e-(c-glutamyl)
lysine cross-link could be accounted for by several mech-
anisms such as the activation of a c-carbonyl of histone
H2B by esterification, followed by a nucleophilic attack
by an e-amino group of Lys residue of histone H4 [38], the
fact that the cross-link is formed between Gln9 of H2B
and Lys5 of H4 strongly suggests that p28 is produced
by a transamidation reaction catalyzed by TG. Although
the possibility that p28 is produced in the cytoplasm and
then translocated into the nucleus cannot be excluded, our
data show the simultaneous appearance of both n TG
and p28 in the nucleus of embryonic cells during the
progression of embryogenesis. This finding is consistent
with nTG being involved in the histone dimerization
reaction.
We have shown that the treatment of A. pectinifera
embryos with trichostatin A, a potent a nd selective
inhibitor of histone d eacetylase [39], induces hyperacetyla-
tion of histone H4 and causes developmental arrest at the
early gast rula stage [18]. Trichostatin A treatment causes
suppression of the appearance of p28 in A. pectinifera
embryos (T. Shimizu & S. Ikemagi, unpublished results).
The acetylation of Lys5 of histone H4 competes with the
TG-catalyzed histone dimerization reaction because a n
acetylated lysine re sidue is not a functional amine donor
substrate for TG. Deprivation of the amine donor for the
TG reaction to produce p28 could be the cause of
developmental arrest. However, this issue will only be

Chem. 268, 12682–12690.
5. Ikura, K., Nasu, T., Yokota, H ., Tsuchiya, Y., Sasaki, R. &
Chiba, H. (1988) Amino acid sequence of guinea pig liver trans-
glutaminase from i ts cDNA sequence. Biochemistry 27 , 2898–
2905.
6. Ho, K.C., Quarmby, V.E., French, F.S. & Wilson, E.M. (1992)
Molecular cloning of rat prostate transglutamin ase com-
plementary DNA. The major androgen-regulated protein DP1 of
rat dorsal prostate and coagul ating gland. J. Biol. Chem. 267,
12660–12667.
7. Nakaoka, H., Perez, D.M., Baek, K.J., Das, T., Husain, A.,
Misono, K., Im, M J. & Graham, R.M. (1994) Gh: a GTP-
binding protein with transglutaminase activity and receptor sig-
naling function. Science 264, 1593–1596.
8. Singh, U.S., Erickson, J.W. & Cerione, R.A. (1995) Identification
and biochemical characterization of an 80 kilodalton GTP-bind-
ing/transglutaminase from r abbit liver nuclei. B i oche mistr y 34,
15863–15871.
9. Lesort, M.A., Hanavanich, K., Zhang, J. & Johnson, G.V. (1998)
Tissue transglutaminase is increased in Huntington’s disease brain.
J. Biol. Chem. 27 3, 11991–11994.
10. Peng, X., Zhang, Y., Zhang, H., Graner, S., Williams, J.F., Levitt,
M.L. & Lokshin, A. (1999) Interaction of tissue transglutaminase
with nuclear transport protein importin-alpha3. FEBS Lett. 446,
35–39.
11. Kalderson, D., Roberts, B.L., Richardson, W.D. & Smith, A.E.
(1984) A short amino acid sequence able to specify nuclear loca-
tion. Cell 39, 499–509.
12. Dingwall, C., Sharnick, S.V. & Laskey, R.A. (1982) A polypeptide
domain that specifies migration of nucleoplasmin into the nucleus.

& Hirose, S. (1991) Cloning a nd sequence analysis of cD NA
clones for bovine aortic-endothelial-cell transglutaminase. Eur. J.
Biochem. 202, 15–21.
22. Nakajima, H., Matoba, K., Matsumoto, Y., Hongo, T., Kiritaka,
K., Sugino, H., N agamatsu, Y., Hamaguchi, Y. & Ikegami, S.
(2000) Molecular characterization of a novel nucleolar protein in
starfish oocytes which is phosphorylated before and during oocyte
maturation. Eur. J. Biochem. 267, 295–304.
23. Stoscheck, C.M. (1990) Quantitation of protein. Methods
Enzymol. 182, 50–68.
24. Lorand, L. & Gotoh, T. (1970) Fibrinoligase. Meth ods Enzymol .
19, 770–782.
25. Laemmli, U.K. (1970) Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature 227, 680–
685.
26. Kaneko, H. & Dan-Sohkawa, M. (1990) Acellularity of starfish
embryonic mesenchyme cells as shown in vitr o. Development 109,
129–138.
27. Noguchi, T., Inoue, H. & Tanaka, T. (1986) The M1- and
M2-type isozymes of rat pyruvate kinase are produced from the
same gene by alternative RNA splicing. J. Biol. Chem. 261,
13807–13812.
28. Ichinose, A., Bottenus, R.E. & Davie, E.W. (1990) Structure of
transglutaminases. J. Biol. Chem. 265, 13411–13414.
29. Kozak, M. (1977) Nucleotide sequences of 5¢-terminal ribosome-
protected initiation regions from two r eovirus messages. Nature
269, 391–294.
30. Weraarchakul-Boonmark, N., Jeong, J.M., Murthy, S.N.P.,
Engel, J.D. & Lorand, L. (1992) Cloning and expression of
chicken erythrocyte transglutaminase. Proc. Natl Acad. Sci. USA

cross-linked form of core histones present in the starfish sperm.
Biochemistry 36, 12071–12079.
39. Yoshida, M ., Kijima, M., Akita, M. & Beppu,T. (1990) Potent and
specific inhibition of mammalian histone deacetylase both in vivo
and in vitro by trichostatin A. J. Biol. Chem. 265, 17174–17179.
Ó FEBS 2002 Nuclear transglutaminase in starfish embryos (Eur. J. Biochem. 269) 1967