Purification, characterization and molecular cloning of tyrosinase
from the cephalopod mollusk,
Illex argentinus
Tetsushi Naraoka
1,2
, Hidemitsu Uchisawa
1
, Haruhide Mori
2
, Hajime Matsue
3
, Seiya Chiba
2
and Atsuo Kimura
2
1
Aomori Industrial Research Center, Aomori;
2
Division of Applied Bioscience, Graduate School of Agriculture,
Hokkaido University, Sapporo;
3
Aomori University of Health and Welfare, Aomori, Japan
Tyrosinase (monophenol,
L
-DOPA:oxygen oxidoreductase)
was isolated from the ink of the squid, Illex argentinus.
Squid tyrosinase, termed ST94, was found to occur as a
covalently linked homodimeric protein with a molecular
mass of 140.2 kDa containing two copper atoms per a
subunit. The tyrosinase activity of ST94 was enhanced by
proteolysis with trypsin to form a protein, termed ST94t,

evolutional relationships of the structures have also been
elucidated on the basis of the amino acid sequences
conserved around two copper-binding sites that form an
oxygen-binding active center [3–6]. These proteins are
classified into a superfamily and are of interest from the
viewpoint of their molecular evolution [7–9].
These copper proteins are particularly important in
arthropods. Arthropod phenoloxidases [10–13], which are
given the same EC number as tyrosinase, are known to be
involved in the host defense system termed the prophenol-
oxidase cascade as a terminally active molecule in the system
[14–17]. Hemocyanins are macromolecules that function
as oxygen carriers in the hemolymph of arthropods and
mollusks [7]. Hemocyanins are also found to exhibit
phenoloxidase activity [6,18,19], which is amplified after
certain treatments such as proteolysis or exposure to
detergents, or by interactions with specific proteins
[6,8,20–22]. This activation suggests that these hemocyanins
may have roles as phenoloxidases in some important
biological events.
Among the mollusks, the emission of ink for defense
against predators is a well-known characteristic behavior of
most cephalopods, which indicates their high capacity for
melanogenesis. We reported previously that a fraction
from the ink of the squid, Illex argentinus,inwhichthe
illexin-peptidoglycan (IPG) possessing a novel mucopoly-
saccharide structure and tyrosinase were contained, showed
anti-tumor activity against Meth A fibrosarcoma in BALB/
c mice [23–26]. The anti-tumor activity was thought to be
expressed through immunostimulation because the fraction

difficulty of isolating the tyrosinases that occur in ink, which
show extremely complex polymorphism (as observed at
least in ink of I. argentinus [26]). In other mollusks, although
tyrosinases have been isolated from a bivalve [34] and a
gastropod [35], there have been no reports on their amino
acid sequences.
In a previous paper, we reported a protein that occurred
in the ink of I. argentinus with weak tyrosinase activity,
whichmigratedasa94-kDaproteinonpolyacrylamidegel
electrophoresis under native condition [26]. The protein,
termed ST94, was assumed to be a partially activated
tyrosinase, one of the abundant proteins in the ink. In this
paper, we describe the purification, proteolytic activation,
some enzymatic properties and molecular cloning of the
squid tyrosinase ST94. This is the first report on the primary
structure of a molluscan tyrosinase (see Footnote on
p. 4026), which contributes to evolutional studies on type
3 copper proteins.
Materials and methods
Materials
L
-3,4-Dihydroxyphenylalanine (
L
-DOPA),
D
-3,4-dihydroxy-
phenylalanine (
D
-DOPA), dopamine, pyrocatechol,
L

the run, the gel was stained for visualizing proteins with
Coomassie Brilliant Blue (CBB), or stained for detecting
of tyrosinase activity with 5 m
ML
-DOPA or 0.5 m
M
L
-tyrosine in 0.1
M
sodium phosphate buffer (pH 6.8) at
room temperature.
MALDI-TOF mass spectrometry experiments were per-
formed on a Voyager-DE STR (Applied Biosystems, Foster
City, CA, USA) according to the manufacturer’s instruc-
tions. Synapinic acid dissolved in a 2 : 1 mixture of 0.1% (by
volume) aqueous trifluoroacetic acid and 0.1% (by volume)
trifluoroacetic acid containing acetonitrile was used as a
matrix for the analyses. Spectrometry was performed in
positive linear mode. Bovine serum albumin (BSA) and
horse heart myoglobin were used as mass number standards.
Amino acid sequences were analyzed by the gas-phase
Edman degradation method using a protein sequencer
PPSQ-10 (Shimadzu Corp., Kyoto, Japan), according to the
manufacturer’s instructions.
Protein content was determined by the method of Lowry
et al. [36] or by measuring absorbance at 205 nm [37] using
BSA as a standard. Uronic acid, hexose and methylpentose
were determined by the carbazole-sulfuric acid method [38],
the phenol–sulfuric acid method [39] and the cysteine–
sulfuric acid method [40], respectively.

M
sodium phosphate
(pH 7.4), and dialyzed against the same buffer (fraction
AS60, 300 mL).
The fraction AS60 (100 mL) was added to 20 mL of 3
M
ammonium sulfate containing 10 m
M
sodium phosphate
(pH 7.4) and applied to a column (2.6 · 28 cm) of Phenyl-
Sepharose CL-4B equilibrated with 0.5
M
ammonium
sulfate in 10 m
M
sodium phosphate (pH 7.4). After washing
with the same buffer, the column was eluted with a linear
gradient of 0.5–0
M
ammonium sulfate in 10 m
M
sodium
phosphate (pH 7.4) in a total volume of 1.5 L, and then
further eluted with 10 m
M
sodium phosphate (pH 7.4) at a
flow rate of 1 mLÆmin
)1
. Fractions of 10 mL were collected
and analyzed for uronic acid using the carbazole–sulfuric

NaCl) were concentrated and desalted by
ultrafiltration as described above, then recovered as a
solution of 1 m
M
sodium phosphate (pH 7.4).
Trypsin-treatment of ST94
ST94 dissolved in 10 m
M
Tris/HCl buffer (pH 8.0) at a final
concentration of 250 lgÆmL
)1
was treated with TPCK-
treated trypsin (2.5 lgÆmL
)1
)for2 hat25 °C. The resulting
proteolyte of ST94, termed ST94t, was purified by gel
permeation HPLC. Gel permeation HPLC was performed
with a L7100S HPLC system (Hitachi) using a column of
G3000SW
XL
(7.8 mm · 300 mm; TOSOH, Tokyo, Japan)
equilibrated with 0.2
M
NaCl in 0.1
M
sodium phosphate
(pH 7.0) (flow rate 0.5 mLÆmin
)1
). Fractions containing
ST94t were concentrated and desalted by ultrafiltration as

M
MBTH, 2% (by volume) N,N-
dimethylformamide and the enzyme solution in 50 m
M
sodium phosphate buffer (pH 6.8). In the analyses for
monophenols, the corresponding o-diphenol at a final
concentration of 1 l
M
was added to the reaction mixture
to shorten the lag period. Formation of the MBTH-adduct
of o-quinone was followed at the isosbestic point wavelength
of each MBTH-adduct at 25 °C, and the steady-state rate
of oxidation of substrate was determined using the molar
extinction coefficient of MBTH-adduct at the isosbestic
point wavelength taken from the literature [43]. The K
m
and
V
max
values for different substrates were obtained from the
Hanes–Woolf equation.
Extraction of RNA and first strand cDNA preparation
All DNA and RNA manipulations were carried out by
standard techniques except where otherwise noted [44].
PCR experiments were performed using a GeneAmp PCR
System 9700 (Applied Biosystems). Poly(A)
+
RNA was
extractedfromaninksac(1.5g)ofI. argentinus using a
QuickPrep mRNA purification kit (Amersham Biosci-

(TYR1 and TYR2) were designed from these sequences to
lower the degeneracy at the 3¢ end regions: TYF1, 5¢-ATGG
TNGAYGTNWSNCARTCNGA-3¢;TYF2,5¢-ATGGT
NGAYGTNWSNCARAGYGA-3¢; TYR1, 5¢-TGDATR
TAYTCYTGNGGNGACA-3¢;TYR2,5¢-TGDATRTA
YTCYTGNGGRCTCA-3¢. PCR was carried out using a
TITANIUM Taq DNA polymerase (Clontech) in 25 lLof
reaction mixture composed of the cDNA (0.5 lL), 25 pmol
of sense primer (TYF1 or TYF2), 25 pmol of antisense
primer (TYR1 or TYR2), 0.2 m
M
dNTPs, 0.5 lLoftheTaq
DNA polymerase and 1 · PCR buffer under the following
conditions: after holding (94 °C, 1 min), 30 cycles of
denaturing (94 °C, 10 s), annealing (56 °C, 30 s) and
elongation (72 °C, 30 s), followed by holding (72 °C,
3 min). The amplified product (about 200 bp) that occurred
in the reaction mixture containing the primers TYF1 and
TYR1 was subcloned into pT7 Blue T-vector (Novagen,
Madison, WI, USA) following purification by 2% (m/v)
agarose gel electrophoresis using a MiniElute purification kit
(Qiagen, Tokyo, Japan). The clones were subjected to DNA
sequencing on both strands by the dideoxy chain termin-
ation method. The sequencing reaction was performed using
a Thermo Sequenase Primer Cycle Sequencing kit (Amer-
sham Biosciences) with Texas red-labeled M13 forward
primer () 21) or M13 reverse primer () 29). The samples
were analyzed with a DNA sequencer SQ-5500 (Hitachi).
For 5¢-RACE of ST94 cDNA, the specific primer,
TYR3 (5¢-CGTCTGCCGATTTCCAATTCTTCTG-3¢),

4
and1unitofKOD
Plus DNA polymerase in the PCR buffer supplied. PCR
was performed under the following conditions: holding
(94 °C, 1 min), 30 cycles of denaturing (94 °C, 5 s) and
annealing/elongation (68 °C, 3 min). The PCR products
about 2.2 kbp in length were subcloned into pT7 Blue with a
Perfectly Blunt Cloning kit (Novagen) following purifica-
tion by agarose gel electrophoresis. Clones of the amplified
fragments were subjected to DNA sequencing on both
strands by primer walking using a BigDye Terminator cycle
sequencing kit with a DNA sequencer, ABI PRISM 3100
(Applied Biosystems). The nucleotide sequence data are
available in the DDBJ/EMBL/GenBank databases under
the accession numbers AB107880 and AB107881 for the
squid tyrosinase ST94 cDNA-1 and cDNA-2, respectively.
Sequence analysis
Sequences were analyzed using a software
DNASIS
(Hitachi
Software, Tokyo, Japan). A homology search was carried
out with
NCBI
-
BLAST
2.0 program available at DDBJ web
server (). Phylogenetic tree was
deduced by a neighbor-joining analysis based on the
alignment of amino acid sequences constructed using the
CLUSTALW

530 nm; ——, absorbance at 280 nm. Fractions containing ST94,
indicated with horizontal bar, were pooled. (B) Native-PAGE of the
tyrosinase-active fractions. Samples were run on 10–15% gradient gels
and stained with CBB (left), with
L
-DOPA (center) and with
L
-tyrosine
(right). Lane M, marker proteins, thyroglobulin (669 kDa), ferritin
(440 kDa), catalase (232 kDa), lactate dehydrogenase (140 kDa) and
BSA (66.4 kDa); Lane 1, fraction no. 170; lane 2, fraction no. 180; lane
3, fraction no. 190; lane 4, fraction no. 198; lane 5, fraction no. 204.
The arrow indicates the band of ST94 (lanes 3 and 4). (C) IEF-native
2D-PAGE of the pooled fraction containing ST94. The first dimen-
sion, IEF (pH 3-9); the second dimension, native-PAGE (10–15%
gradient gel). The gels were stained with CBB (left) and with
L
-DOPA
(right). Lane M, mass marker proteins. Positions of pI marker proteins
were indicated at the top; amyloglucosidase (pI 3.50), soybean trypsin
inhibitor (pI 4.55), b-lactoglobulin A (pI 5.20), bovine carbonic
anhydrase B (pI 5.85), horse myoglobin (pI 7.35) and lentil lectin
(pI 8.65). The arrow indicates the spot of ST94.
Ó FEBS 2003 Tyrosinase from the mollusk, Illex argentinus (Eur. J. Biochem. 270) 4029
Fig. 1A. Most of the IPG was eluted in the breakthrough
fraction and separated from tyrosinase; these two compo-
nents could not be separated by anion-exchange chroma-
tography and gel permeation chromatography [26]. In
native-PAGE of the tyrosinase-active fractions (Fig. 1B),
the protein bands were not observed in the position

Materials and methods. As shown in Fig. 2A, ST94
generated a proteolyte, termed ST94t, showing slightly
higher mobility than ST94 on native-PAGE. ST94t in the
proteolysate of ST94 was purified as the enzyme prepar-
ation with a yield of 0.23 mg protein from 0.30 mg of ST94
and a specific activity of 103 U per mg protein by gel
permeation HPLC. The results indicated that ST94t was an
activated tyrosinase molecule bearing the stable catalytic
domain of ST94.
Molecular mass of the squid tyrosinase
ST94 electrophoresed to a position corresponding to that of
a protein with a molecular mass of about 70 kDa on SDS/
PAGE under reducing conditions, whereas it migrated
as a  140 kDa protein under nonreducing conditions
(Fig. 2B,C). In MALDI-TOF mass spectrometry, a parent
ion signal of ST94 was observed at m/z 140.2 kDa. These
results indicated that ST94 is a 140.2-kDa protein composed
of two 70.1-kDa subunits that are linked, probably by a
disulfide bond. These estimates of the molecular mass of
ST94 were supported by the result of gel permeation HPLC
(Fig. 3). There was a minor band between 140 and 232 kDa
in the native-PAGE of Fig. 2A (lane 1), implying that ST94
enables to form an oligomeric protein. On SDS/PAGE,
ST94t electrophoresed to a position corresponding to that
of a protein of about 65 kDa under reducing conditions,
while migrated as an  130 kDa protein under nonreducing
conditions (Fig. 2B,C). In MALDI-TOF mass spectro-
metry, the trypsin-treated reaction mixture of ST94 showed
Fig. 2. PAGE analyses of ST94 and ST94t. Purified ST94 and the trypsin treatment reaction mixture of ST94 were subjected to PAGE analyses.
Lane M, molecular mass markers; lane 1, ST94; lane 2, the trypsin treatment reaction mixture of ST94. The closed arrow and the open arrow

by the same procedure and determined to be 0.24% (by
mass) in good agreement with the value calculated from the
literature [7] (16 copper atoms per subunit): 17.5 ± 0.3 ng
copperÆmL
)1
was detected in KLH solution of 7.3 lg
proteinÆmL
)1
. The ST94 solution was also subjected to the
analysis of manganese, but no manganese was detected.
The N-terminal amino acid sequences of ST94 and ST94t
wereshowntobeNH
2
-MVDVSQSDGLQSXLDRFADD
(X represents an amino acid undetermined) and NH
2
-
ISTLATMSPQEYIQ, respectively, which indicated that the
N-terminal region of ST94 was truncated with trypsin to
generate the N-terminal of ST94t. Each analysis for ST94
and for ST94t showed a single N-terminal sequence,
allowing us to speculate that ST94 is a homodimeric protein.
Enzymatic properties of the squid tyrosinase
Effects of pH and temperature on stability and acti-
vity. As ST94 and ST94t showed the identical results on
pH- and temperature-effects, we describe the data of ST94t in
this section. ST94t retained more than 90% of its activity
after incubation at 4 °C for 24 h within a pH range of 6.5–11
(Fig. 4A). The optimum pH for o-diphenolase activity of
ST94t was determined to be pH 8.0 (Fig. 4B) with correction

buffer with various pH) was
incubated for 24 h at 4 °C, followed by measurement of residual activity by the dopachrome method. (B) The o-diphenolase activity of ST94t was
measured under various pH conditions at 25 °C; the assay mixture (3 mL) contained 5 m
ML
-DOPA and ST94t (0.36 lg) in 0.1
M
buffer (pH 3.6–
9.2). The reaction was monitored at 475 nm. The data were corrected by subtraction of the increase caused by auto-oxidation of
L
-DOPA. The
following buffers were used: d, sodium acetate buffer (pH 3.6–5.6); s, sodium phosphate buffer (pH 5.7–8.0); j, Tris/HCl buffer (pH 8.0–9.0);
h, sodium carbonate buffer (pH 9.2–10.8); m, sodium phosphate buffer (pH 11.2–11.9).
Ó FEBS 2003 Tyrosinase from the mollusk, Illex argentinus (Eur. J. Biochem. 270) 4031
incubated at pH 7.4 for 20 min (Fig. 5A). Similar stability
toward temperature has been reported for the subunit of
hemocyanin from a gastropod, Rapana thomasiana grosse
[45]. The conformational stability of hemocyanin was
influenced generally by the aggregation state; the association
of structural subunits to hemocyanin increased the stability
[45]. Covalently linked dimeric form, the characteristic
structure of ST94 possibly contributes to the stability.
The effects of temperature on the o-diphenolase activity
of ST94t were investigated in the range 4–50 °Cusing
L
-DOPA as a substrate. Tyrosinase from mushroom was
also examined for comparison with ST94t. As shown in
Fig. 5B, the o-diphenolase activities of both enzymes
correlated linearly with temperature according to the
Arrhenius equation, within the range from 4 to 40 °C.
When the reactions were carried out above 45 °C, however,

period, as reported for other tyrosinases, which was
shortened by addition of each mono-oxygenation product
(diphenol) as a cofactor [2,28]. From the comparison of
reaction efficiency (k
0
/K
m
), dopamine appeared to be
oxidized most effectively by ST94 as well as by ST94t,
whereas DHPPA, which is known to be good substrate for
several tyrosinases [42,43], was shown to be a poor substrate
for these enzymes. The K
m
value for
L
-tyrosine was higher
Fig. 5. Thermostability (A) and temperature-dependency of o-diphenolase activity (B) of ST94t. (A) ST94t solution (6.8 lgÆmL
)1
of 20 m
M
sodium
phosphate buffer, pH 7.4) was incubated at 20–75 °C for 20 min and residual activity was measured by the dopachrome method. Activity after
incubation on ice is taken as 100%. (B) The o-diphenolase activity of ST94t (d) was measured at 4–50 °C using the dopachrome method (0.30 lg
per assay). Mushroom tyrosinase (s) was also examined for comparison (3.8 lg per assay). The reaction rates, v (DA
475
Æmin
)1
per mg protein) were
plotted according to the Arrhenius equation (T, absolute temperature).
Table 1. Effects of tyrosinase inhibitors on the o-diphenolase activities of

5
>10
5
4032 T. Naraoka et al. (Eur. J. Biochem. 270) Ó FEBS 2003
than that for the
D
-isomer. The same tendency was observed
for isomers of DOPA. This catalytic stereospecificity of
I. argentinus tyrosinase was similar to that reported for
S. officinalis tyrosinase [28]. From the comparison of ST94
and ST94t, it appeared that the trypsin treatment of ST94
caused a fall in the K
m
value and a rise in the k
0
value,
resulting in approximately five to 70 times higher reaction
efficiency of ST94t for oxidation of each substrate. These
results suggest that a limited proteolysis for activation is
involved in the natural regulation system of tyrosinase in
cephalopods, as in the case of arthropods [14–17].
Molecular cloning of ST94 cDNA
The cDNA cloning of tyrosinase ST94 was carried out
by degenerate RT-PCR and RACE using the first strand
cDNA of poly(A)
+
RNA extracted from an ink sac of
I. argentinus as a template. First, in order to detect the target
cDNA, RT-PCR was carried out using degenerate primers
designed from the N-terminal amino acid sequences of ST94

cDNA-1 and cDNA-2, respectively. These two cDNA
sequences were also confirmed by other PCR experiments
performed using the first strand cDNA independently
prepared from the same poly(A)
+
RNA preparation as a
template. Both of the cDNAs covered the complete open
reading frames of 1878 bp (nucleotide 122–1999) encoding
putative 625-amino acid proteins with 121 bp of the 5¢
untranslated regions and the 3¢ untranslated regions
containing polyadenylation signals (AATAAA) at three
positions and the poly(A)-tails. The criteria for a consensus
translation initiation site were observed around the putative
initiator ATG codon (CCGAAATGG) of the largest open
reading frames [50]. The molecular mass numbers calculated
for the encoded proteins in the open reading frames of
cDNA-1 and cDNA-2 were 70 975 and 71 046, respectively.
Deduced amino acid sequences
The amino acid sequence deduced from the nucleotide
sequence of cDNA-1 was shown to contain sequences that
agreed with the N-terminal amino acid sequences of ST94
and ST94t determined by Edman degradation, whereas the
nucleotide sequence of cDNA-2 was different from that of
cDNA-1 at 15 positions, resulting in amino acid substitu-
tions at four positions (Fig. 6). In particular, the substitu-
tion from Gly27 to Glu27 caused by the single base
substitution, which did not agree with the result obtained
from the N-terminal amino acid sequence analysis of ST94,
was observed in cDNA-2. In the N-terminal sequence
analysis, a portion of the ST94 preparation isolated from

(lmolÆmin
)1
Æmg
)1
)
k
0
(s
)1
)
k
0
/K
m
(m
M
)1
Æs
)1
)
K
m
(m
M
)
V
max
(lmolÆmin
)1
Æmg

-Tyrosine 0.57 0.194 0.453 0.79 0.35 8.74 18.6 53 67
D
-Tyrosine 0.32 0.216 0.505 1.6 0.24 7.98 17.0 71 44
Tyramine 0.32 1.59 3.72 12 0.12 27.6 58.7 490 41
Ó FEBS 2003 Tyrosinase from the mollusk, Illex argentinus (Eur. J. Biochem. 270) 4033
Furthermore, the nucleotide sequences of cDNA-1 and
cDNA-2 were almost identical: the calculated homology
was 99.3% except for the poly(A)-tails. Therefore, the
cDNA-2 was thought to be of an allelic variant message for
tyrosinase ST94, expressed in the individual of I. argentinus
used for poly(A)
+
RNA extraction in this study. Both
amino acid sequences deduced from the two cDNAs were
revealed to possess two putative copper-binding sites
(critical regions for tyrosinase activity), as described below,
and no amino acid substitution was observed in these two
sites (Figs 6 and 7). Therefore, the variant of ST94 was
thought to be able to function as tyrosinase as well as ST94.
Fig. 6. Nucleotide and deduced protein sequences of ST94 cDNA-1. The nucleotides (upper) are numbered from the first base; the amino acids
(lower) are numbered from the initiating methionine. Base substitutions at 15 positions and amino acid substitutions at four positions observed in
cDNA-2 are shown in italics. The N-terminal amino acid sequences of ST94 and ST94t obtained by Edman degradation are underlined. The
putative copper ligands are circled. A potential N-linked glycosylation site is shown by an asterisk. Cysteine residues are indicated by d.The
putative polyadenylation signals are indicated by double underlining.
4034 T. Naraoka et al. (Eur. J. Biochem. 270) Ó FEBS 2003
These findings suggested the occurrence of other ST94
variants in the gene pool of I. argentinus. Sequence poly-
morphism of the tyrosinase gene has also been observed in
Neurospora crassa [51].
The N-terminal sequence of ST94 was found to start at

the mass spectrometry (70.1 kDa for ST94 and 63.8 kDa
for ST94t). The cause of the difference of about 1 kDa
except for mass of two copper atoms (127 Da) remains
unclear; however, the difference was presumed to be due to
post-translational modifications, for example, glycosylation
at the unique potential glycosylation site for N-linked
carbohydrate found at Asn333.
It was demonstrated that mature ST94 subunit could
undergo the digestion of N-terminal 51 amino acid residues
by trypsin. In prophenoloxidases in two insects, Bombyx
mori and Manduca sexta, the N-terminal region consisting
of 51 amino acid residues were also cleaved by the
prophenoloxidase-activating enzyme (PPAE) to generate
active phenoloxidases [11,13]. This similarity suggests
that the conformational change of ST94 caused by the
elimination of the N-terminal region is required principally
for its activation. Furthermore, the cleavage site of ST94
Fig. 7. Comparison of the amino acid sequences at two putative copper-binding sites, Cu(A) and Cu(B), in ST94 and other type 3 copper proteins.
Numbers indicate positions of the amino acid residues in each sequence. Gaps (–) have been introduced to optimize the alignment. The putative
copper ligands of histidine residues conserved in all proteins are labeled with d, those conserved in molluscan proteins and tyrosinases with an s,
andthoseconservedinarthropodproteinswithanh. The identical residues are shaded. IaY, I. argentinus tyrosinase ST94; OdHc and OdHe,
Octopus dofleini hemocyanin functional unit c and e, respectively (SWISS-PROT, accession No. O61363); SoHh, S. officinalis hemocyanin unit h
(SWISS-PROT, P56826); HpHg, Helix pomatia b
c
-hemocyanin unit g (SWISS-PROT, P56823); McH2c, M. crenulata hemocyanin unit 2-c
(SWISS-PROT, P81732); MmY, Mus musculus tyrosinase (SWISS-PROT, P11344); GgY, Gallus gallus tyrosinase (DDBJ, D88349); CeY,
C. elegans hypothetical protein K08E3.1 (DDBJ, Z81568); AoY, Aspergillus oryzae tyrosinase (DDBJ, D37929); SgY, Streptomyces glaucescens
tyrosinase (SWISS-PROT, P06845); BmP, B. mori prophenoloxidase subunit 1 (DDBJ, D49370); PlP, Pacifastacus leniusculus prophenoloxidase
(DDBJ, X83494); PiH, Panulirus interruptus hemocyanin subunit a (SWISS-PROT, P04254); LpH, Limulus polyphemus hemocyanin II (SWISS-
PROT, P04253).

two subunits. Some of the other cysteine residues may
be involved in intramolecular disulfide bridges, creating a
packed structure in the C-terminal region similar to the
C-terminal domain of hemocyanin [3,6]. As described
above, the I. argentinus tyrosinase was found to exhibit
temperature-dependency of o-diphenolase activity like a
psychrophilic enzyme. As there is no cysteine residue
around the copper-binding sites of ST94, we suggest that
the active center of ST94 is more flexible, which allows this
enzyme to retain higher activity at low temperatures [46].
Structural similarity of ST94 to other tyrosinases
and hemocyanins
BLAST homology search of the amino acid sequence of
ST94 deduced from cDNA-1 revealed that two putative
copper-binding sites for Cu(A) and Cu(B), characteristically
conserved regions in type 3 copper proteins, were present in
ST94 [6]. Comparison of these sites of ST94 with those of
tyrosinases, hemocyanins and phenoloxidases from other
organisms are shown in Fig. 7. These two copper-binding
sites of ST94 were similar to those of molluscan hemo-
cyanins as well as tyrosinases from microorganisms, a
nematode and vertebrates, and less similar to those of
phenoloxidases and hemocyanins from arthropods. Six
histidine residues in these sites of ST94 assumed to be
copper-binding ligands could be arranged in similar posi-
tions with those of molluscan hemocyanins and tyrosinases
[7,9]; highly conserved amino acid sequences and residues
around these copper ligands in molluscan hemocyanins and
tyrosinases were also well conserved in ST94. On the other
hand, molluscan hemocyanins revealed that one of the

Evolutional scenarios of hemocyanins and tyrosinases
have been proposed [7–9]. It has been suggested, for
example, that molluscan hemocyanins and tyrosinases
genetically diverged from their common ancestor, sepa-
rately from the evolutional line of hemocyanins and
phenoloxidases in arthropods, and that the divergence of
molluscan hemocyanins and tyrosinases occurred earlier
than that of arthropod hemocyanins and phenoloxidases, at
an extremely early evolutional stage of life similar to that
when aerobic metabolism was established and metazoans
emerged. As described above, the structural features of
molluscan tyrosinase ST94, the nearest relative of molluscan
hemocyanins, seem to support these evolutional scenarios.
Acknowledgements
The authors wish to thank Dr Isoshi Nukatsuka, Hirosaki University
and Ms Junko Murakami, Aomori Prefectural Environmental and
Health Center, for excellent technical support and valuable advice on
metal analysis.
References
1. Mason, H.S. (1965) Oxidases. Annu.Rev.Biochem.34, 595–634.
2. Rodrı
´
guez-Lo
´
pez, J.N., Tudela, J., Varo
´
n, R., Garcı
´
a-Carmona,
F. & Garcı

derha
¨
ll, K. (1995) cDNA
cloning of prophenoloxidase from the freshwater crayfish Paci-
fastacus leniusculus and its activation. Proc.NatlAcad.Sci.USA
92, 939–943.
11. Hall, M., Scott, T., Sugumaran, M., So
¨
derha
¨
ll, K. & Law, J.H.
(1995) Proenzyme of Manduca sexta phenol oxidase:
purification, activation, substrate specificity of the active
enzyme, and molecular cloning. Proc.NatlAcad.Sci.USA92,
7764–7768.
12. Fujimoto, K., Okino, N., Kawabata, S., Iwanaga, S. & Ohnishi, E.
(1995) Nucleotide sequence of the cDNA encoding the proenzyme
of phenol oxidase A
1
of Drosophila melanogaster. Proc. Natl Acad.
Sci. USA 92, 7769–7773.
13. Kawabata, T., Yasuhara, Y., Ochiai, M., Matsuura, S. & Ashida,
M. (1995) Molecular cloning of insect pro-phenol oxidase: a
copper-containing protein homologous to arthropod hemocyanin.
Proc.NatlAcad.Sci.USA92, 7774–7778.
14. Ashida, M. & Dohke, K. (1980) Activation of pro-phenoloxidase
by the activating enzyme of the silkworm, Bombyx mori. Insect
Biochem. 10, 37–47.
15. Johansson, M.W. & So
¨

peptidoglycan fraction from squid ink. Biol. Pharm. Bull. 17,
846–849.
24. Takaya, Y., Uchisawa, H., Hanamatsu, K., Narumi, F., Okuzaki,
B. & Matsue, H. (1994) Novel fucose-rich glycosaminoglycans
from squid ink bearing repeating unit of trisaccharide structure.
Biochem. Biophys. Res. Commun. 198, 560–567.
25.Takaya,Y.,Uchisawa,H.,Narumi,F.&Matsue,H.(1996)
Illexins A, B and C from squid ink should have a branched
structure. Biochem. Biophys. Res. Commun. 226, 335–338.
26. Naraoka, T., Chung, H S., Uchisawa, H., Sasaki, J. & Matsue, H.
(2000) Tyrosinase activity in antitumor compounds of squid ink.
Food Sci. Technol. Res. 6, 171–175.
27. Harris, J.R. & Markl, J. (1999) Keyhole limpet hemocyanin
(KLH): a biomedical review. Micron 30, 597–623.
28. Prota, G., Ortonne, J.P., Voulot, C., Khatchadourian, C., Nardi,
G. & Palumbo, A. (1981) Occurrence and properties of tyrosinase
in the ejected ink of cephalopods. Comp.Biochem.Physiol.68B,
415–419.
29. Palumbo, A., Misuraca, G., d’Ischia, M. & Prota, G. (1985) Effect
of metal ions on the kinetics of tyrosine oxidation catalysed by
tyrosinase. Biochem. J. 228, 647–651.
30. Palumbo, A., d’Ischia, M., Misuraca, G., De Martino, L. & Prota,
G. (1994) A new dopachrome-rearranging enzyme from the ejec-
ted ink of the cuttlefish Sepia officinalis. Biochem. J. 299, 839–844.
31. Gesualdo, I., Aniello, F., Branno, M. & Palumbo, A. (1997)
Molecular cloning of a peroxidase mRNA specifically expressed in
the ink gland of Sepia officinalis. Biochim. Biophys. Acta 1353,
111–117.
32. Palumbo,A.,DiCosmo,A.,Gesualdo,I.&Hearing,V.J.(1997)
Subcellular localization and function of melanogenic enzymes in

determination. J. Biol. Chem. 175, 595–603.
41. Fling, M., Horowitz, N.H. & Heinemann, S.F. (1963) The isola-
tion and properties of crystalline tyrosinase from Neurospora.
J. Biol. Chem. 238, 2045–2053.
42. Espı
´
n, J.C., Morales, M., Varo
´
n, R., Tudela, J. & Garcı
´
a-Ca
´
novas,
F. (1995) A continuous spectrophotometric method for
determining the monophenolase and diphenolase activities of
apple polyphenol oxidase. Anal. Biochem. 231, 237–246.
43. Espı
´
n, J.C., Morales, M., Garcı
´
a-Ruiz, P.A., Tudela, J. & Garcı
´
a-
Ca
´
novas, F. (1997) Improvement of a continuous spectro-
photometric method for determining the monophenolase and
diphenolase activities of mushroom polyphenol oxidase. J. Agric.
Food Chem. 45, 1084–1090.
44. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular

ll, K. & Lee, B.L. (2002) A new easter-type serine protease
cleaves a masquerade-like protein during prophenoloxidase acti-
vation in Holotrichia diomphalia Larvae. J. Biol. Chem. 277,
39999–40004.
53. Lerch, K. (1982) Primary structure of tyrosinase from Neurospora
crassa: II. Complete amino acid sequence and chemical structure
of a tripeptide containing an unusual thioether. J. Biol. Chem. 257,
6414–6419.
54.Lee,S.Y.,Lee,B.L.&So
¨
derha
¨
ll, K. (2003) Processing of an
antibacterial peptide from hemocyanin of the freshwater crayfish
Pacifastacus leniusculus. J. Biol. Chem. 278, 7927–7933.
55. Gielens, C., De Geest, N., Xin, X Q., Devreese, B., Van Beeumen,
J. & Pre
´
aux, G. (1997) Evidence for a cysteine-histidine thioether
bridge in functional units of molluscan haemocyanins and location
of the disulfide bridges in functional units d and g of the
b
c
-haemocyanin of Helix pomatia. Eur. J. Biochem. 248, 879–888.
56. Fujita, Y., Uraga, Y. & Ichisima, E. (1995) Molecular cloning and
nucleotide sequence of the protyrosinase gene, melO,from
Aspergillus oryzae and expression of the gene in yeast cells.
Biochim. Biophys. Acta 1261, 151–154.
4038 T. Naraoka et al. (Eur. J. Biochem. 270) Ó FEBS 2003