A vanadium-dependent bromoperoxidase in the marine red alga Kappaphycus alvarezii (Doty) Doty displays clear substrate specificity

Article abstract of DOI:10.1016/j.phytochem.2007.03.003

Bromoperoxidase activity was initially detected in marine macroalgae belonging to the Solieriaceae family (Gigartinales, Rhodophyta), including Solieria robusta (Greville) Kylin, Eucheuma serra J. Agardh and Kappaphycus alvarezii (Doty) Doty, which are important industrial sources of the polysaccharide carrageenan. Notably, the purification of bromoperoxidase was difficult because due to the coexistence of viscoid polysaccharides. The activity of the partially purified enzyme was dependent on the vanadate ion, and displayed a distinct substrate spectrum from that of previously reported vanadium-dependent bromoperoxidases of marine macroalgae. The enzyme was specific for Br^- and I^- ions and inactive toward F^- and Cl^-. The K_m values for Br^- and H₂O₂ were 2.5 × 10^-3 M and 8.5 × 10^-5 M, respectively. The halogenated product, dibromoacetaldehyde, that accumulated in K. alvarezii was additionally determined.

Full text of DOI:10.1016/j.phytochem.2007.03.003

PHYTOCHEMISTRY
Phytochemistry 68 (2007) 1358–1366
www.elsevier.com/locate/phytochem
A vanadium-dependent bromoperoxidase in the marine red alga
Kappaphycus alvarezii (Doty) Doty displays clear substrate specificity
a
b
b
Zornitsa Kamenarska , Tomokazu Taniguchi , Noboru Ohsawa ,
c
b,
*
Masanori Hiraoka , Nobuya Itoh
a
Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, Sofia 1113, Bulgaria
Department of Biotechnology, Faculty of Engineering, Toyama Prefectural University, Kurokawa 5180, Imizu, Toyama 939-0398, Japan
b
c
Usa Marine Biological Institute, Kochi University, Usa, Tosa, Kochi 781-1164, Japan
Received 4 August 2006; received in revised form 23 February 2007
Available online 16 April 2007
Abstract
Bromoperoxidase activity was initially detected in marine macroalgae belonging to the Solieriaceae family (Gigartinales, Rhodo-
phyta), including Solieria robusta (Greville) Kylin, Eucheuma serra J. Agardh and Kappaphycus alvarezii (Doty) Doty, which are impor-
tant industrial sources of the polysaccharide carrageenan. Notably, the purification of bromoperoxidase was difficult because due to the
coexistence of viscoid polysaccharides. The activity of the partially purified enzyme was dependent on the vanadate ion, and displayed a
distinct substrate spectrum from that of previously reported vanadium-dependent bromoperoxidases of marine macroalgae. The enzyme
was specific for Brꢀ and Iꢀ ions and inactive toward Fꢀ and Clꢀ. The Km values for Brꢀ and H2O2 were 2.5 · 10ꢀ3 M and 8.5 · 10ꢀ5 M,
respectively. The halogenated product, dibromoacetaldehyde, that accumulated in K. alvarezii was additionally determined.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Keywords: Kappaphycus alvarezii (Doty) Doty; Solieriaceae; Marine algae; Bromoperoxidase; Vanadium-dependent; Dibromoacetaldehyde
1. Introduction
While the biosynthesis of halogenated marine natural
products has intrigued scientists for decades, the mecha-
nism has remained unclear to date (Littlechild et al.,
2002). Haloperoxidases generally oxidize a halide anion
(Xꢀ) into a halonium cation (X+) with H2O2, and the X+
and its related species (XOH, X2, Xꢀ3 ) generated acts on
many organic substrates as an electrophile (Yamada
et al., 1985a; Itoh et al., 1988). The enzymes are divided
into three groups, specifically, heme types (eukaryotic heme
types and bacterial types containing protoporphylin IX as
a prosthetic group) (Morris and Hager, 1966; Manthey and
Hager, 1981), bacterial non-metal enzymes (perhydrolase)
(van Pee et al., 1997; Kawanami et al., 2002) and eukary-
otic vanadium-containing enzymes (Itoh et al., 1986; Ohs-
awa et al., 2001; Krenn et al., 1989). Bacterial non-metal
enzymes are related to the esterases and lipases because
they contain the representative catalytic triad (Itoh et al.,
2001).
Halogenated compounds are synthesized by several
marine algae with the aid of haloperoxidases. These com-
pounds are classified into indoles, terpenes, acetogenins,
phenols, and volatile halogenated hydrocarbons (Faulkner,
2000, 2001, 2002). Halogenated marine metabolites gener-
ally possess biological antifungal, antibacterial, antiviral
and anti-inflammatory activities (Cowan, 1999). Some
additionally function as allelochemicals in the marine envi-
ronment (Sakata et al., 1991; Denboh et al., 1997; Ohsawa
et al., 2001).
Abbreviations: BPO, bromoperoxidase; MCD, monochlorodimedone.
Corresponding author. Tel.: +81 766 56 7500x560; fax: +81 766 56
*
2498.
E-mail address: itoh@pu-toyama.ac.jp (N. Itoh).
0031-9422/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2007.03.003

Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
1359
Table 1
Vanadium-containing bromoperoxidases (BPOs) spe-
cific for Brꢀ and Iꢀ are widespread in marine macroalgae
such as Corallina (Itoh et al., 1987a; Sheffield et al.,
1993) and Ascophyllum (Vilter, 1984; Krenn et al.,
1989; Butler, 1998). Some of these enzymes display low
activity towards Clꢀ (Soedjak and Butler, 1990), and
modification of enzyme can change the halide ion speci-
ficity (Ohshiro et al., 2004). Although oxidation of bro-
mide ion occurs at the active site in the case of
vanadium-containing enzymes (Dau et al., 1999; Little-
child et al., 2002), BPOs including Corallina generally
do not display distinct substrate specificity or stereoselec-
tivity (Itoh et al., 1987b, 1988; Soedjak and Butler, 1990).
Recent studies show that flavin-dependent halogenases
involved in chlorination of secondary metabolites in bac-
teria catalyze the regioselective chlorination of trypto-
phan or pyrrole derivatives (Dong et al., 2005; van Pee
and Patallo, 2006).
BPO activities in marine macroalgae
Alga
Total activity
(U/g fresh
alga)
Specific
activity (U/mg
protein)
Rhodohyta
Solieriaceae
Kappaphycus alvarezii
Eucheuma serra
Solieria robusta
1.00
1.12
0.02
0.14
0.03
0.0006
Corallinaceae
Corallina pilulifera
Corallina officinalis
Amphilroa zonata
11.7a
19.3a
9.3a
0.54a
0.56a
0.83a
Phaeophyta
Ascophyllum nodosum
Laminaria saccharina
4.3b
12.75c
0.79b
0.57c
a
Data from Yamada et al. (1985b).
Data from Wever et al. (1985).
Data from De Boer et al. (1986).
b
c
During screening, it was initially observed that specific
samples of marine macroalgae belonging to the Solieria-
ceae family, including Kappaphycus alvarezi (Doty) Doty,
exhibited significant bromoperoxidase activity (Kam-
enarska et al., 2006). K. alvarezii and its related algae are
the basic source of carrageenan (a polysaccharide), which
is commonly applied in the food industry and medicine.
Owing to high commercial interest, K. alvarezii is widely
cultivated in the Philippines, Indonesia, Vietnam and Tan-
zania. Recently, tentative cultivation during the summer
season was introduced in Shikoku Island, Southern Japan
(Ohno et al., 1994). To date, most investigations on this
alga focus on polysaccharide content and its properties
(Ohno et al., 1996). Here, we report the properties of this
novel BPO displaying distinct substrate specificity and its
natural brominated compound was identified in K.
alvarezii.
U/ml (529%); 1 mM KBr, 0.08 U/ml (1142%); and 5 mM
KBr, 0.17 U/ml (2428%)), respectively. While the mecha-
nism of activation by KBr is currently unclear, it is spec-
ulated that the masked BPO activity is stimulated upon
incubation with Brꢀ ions. KCl did not activate the
enzyme. K. alvarezii was selected as the BPO source for
further experiments, since its specific activity was high
and biomass was available by aquaculture. The buffer
was supplemented with 5 mM KBr and 0.5 mM NaVO3
throughout the isolation and purification of BPO, unless
otherwise specified.
2.2. Enzyme purification
Since BPO of K. alvarezii was stable in the presence of
Brꢀ and vanadate ions, all purification procedures were
performed at room temperature. BPO was extracted from
the alga and partially purified using the procedures
described in Section 4. The elimination of polysaccharide
from crude enzyme by adding gel-forming agents to the
enzyme solution, such as CaCl2 or KCl, led to a drastic loss
in enzymatic activity, indicating that BPO was probably
bound to polysaccharides. The activities of a number of
enzymes were next examined that hydrolyze polysaccha-
rides: Labiaseꢁ from Streptomyces fulvissimus, mainly con-
taining N-acetylmuramoylhydrolase (Ohbuchi et al., 2001),
was effective in decreasing the viscosity of crude extracts of
K. alvarezii, but relatively ineffective in completely remov-
ing polysaccharides from the enzyme solution. Thus, only
partial purification of the enzyme was achieved. To purify
the enzyme to homogeneity, further steps to completely
hydrolyze the contaminating polysaccarides are essential,
but this may lead to loss of enzyme activity. For this rea-
son, complete purification of the enzyme was not
attempted, and most experiments were performed with par-
tially purified enzyme. The results of the purification are
summarized in Table 2.
2. Results and discussion
2.1. Distribution of BPO activity in marine algae
As shown in the comparative data in Table 1, variable
BPO activities were observed in a number of macroalgae
belonging to the Solieriaceae family.
To our knowledge, this is the first report on BPO activ-
ity in these red algae, except for the previously documented
Eucheuma denticulatum (Mtolera et al., 1996) and Meristi-
ella gelidium (Collen et al., 1994) in the Solieriaceae. These
produce volatile halogenated compounds when exposed to
stress conditions, such as at an elevated pH and at high
light intensities.
The addition of KBr to the crude extracts of K. alvar-
ezii during overnight dialysis led to improved stability.
Moreover, enhanced stability and elevated activity was
observed with increasing concentrations of KBr during
dialysis (initial activity, 0.007 U/ml (100%), 0.1 mM
KBr, 0.18 U/mg protein (129%); 0.5 mM KBr, 0.037

1360
Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
Table 2
Purification of BPO from K. alvarezii
Step
Total
protein
(mg)
Total
activity
(U)
Specific
activity (U/ (%)
mg)
0.14
Yield Purification
factor (fold)
Crude
122.5
16.6
100
1
extract
Labiase
treatment/
dialysis
Octyl-
Sepharose
QAE-
Toyopearl
66.7
61.5a
0.92
370.5
6.6
4.2
20.5
3.2
4.88
3.05
123.5 34.9
19.3 21.8
1.05
a
The increase of activity was probably due to the incubation with KBr.
This enzyme preparation indicated some indistinct pro-
tein bands on PAGE-gel, and the purity was roughly esti-
mated to be less than 50% from Fig. 1.
2.3. Enzyme properties
BPO of K. alvarezii catalyzed the bromination of mono-
chlorodimedone (MCD) 1 in the presence of bromide ion
(3.05 U/mg protein, 100%), i.e. the oxidation of Brꢀ to
Br+ and its related species, but not chlorination or fluorina-
tion. The enzyme also displayed strong oxidation activity
towards Iꢀ to produce I2 (2.08 U/mg protein, 68%).
Despite variable assay methods and activities, BPO activity
for Iꢀ appeared to be comparable with that for Brꢀ.
The influence of the KBr concentration on BPO activity
was determined. A rise in enzymatic activity was observed
with increasing KBr concentrations (from 0.5 to 20 mM).
However, a decrease in activity (85% at 40 mM) was evi-
dent at higher levels of KBr. Using the linear portion of
the saturation curve, the Km value for Brꢀ was calculated
from Lineweaver–Burk plot as 2.5 mM (Fig. 2a).
The influence of the H2O2 concentrations was also deter-
mined. The Km value was calculated as 8.5 · 10ꢀ5 M
(Fig. 2b). Inhibition by H2O2 was detected at concentra-
tions higher than 10 mM. The Km value for Brꢀ was rather
low, compared to that of previously studied BPOs, while
that for H2O2 was comparable (1.1 · 10ꢀ2 M for Brꢀ,
9.2 · 10ꢀ5 M for H2O2 reported for C. pililifera BPO (pH
6.0) (Itoh et al., 1986), and 9.4 · 10ꢀ3 M for Brꢀ,
1.04 · 10ꢀ4 M for H2O2 by Ascophyllum nodosum BPO
Fig. 2. Lineweaver–Burk plots for the bromination of MCD by BPO. (a)
Plot with varying amounts of KBr, (b) plot with varying amounts of H2O2.
The intersection of dashed line on abscissa indicates 1/Km.
(pH 5.9)) (Krenn et al., 1989). Catalase activity, which is
common in heme type haloperoxidases, was not observed.
BPO was active over a broad pH range, and the opti-
mum pH was around at 6.5. Higher activity was observed
with phosphate buffer at around pH 5.5 (Fig. 3).
When the analysis was performed in acetate or MES
buffer between pH 4.0 and 5.5, the activity was twice as
low. These finding imply that both pH and type of buffer
affects enzymatic activity. The presence of phosphate ions
had a positive impact on BPO activity. Although the
Fig. 1. Polyacrylamide gel electrophoresis (PAGE) of partially purified
BPO. About 10 lg of enzyme was subjected to electrophoresis. The gel was
stained for protein (a) and for enzyme activity (b). The arrow indicates the
position of BPO.
Fig. 3. pH optima of BPO. Enzyme activity was measured in 50 mM
acetate buffer (pH 4–5) (h), potassium phosphate buffer (pH 4–7) (r),
MES buffer (pH 5.5–7) (s) and Tris–HCl buffer (pH 7–8) (n).

Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
1361
vanadium-dependent BPOs are known to share similar
active sites with some acid phosphatases (Ishikawa et al.,
2000; Littlechild et al., 2002), the positive effect of phos-
phate ions not negative on K. alvarezii BPO has been
unclear.
The effects of a range of compounds for crude enzyme,
described in Section 4, were also determined. AgNO3,
p-CMB, iodoacetic acid (sulfuhydryl reagents), and
Na2MoO4, VOSO4, NaVO3, NaCl and EDTA did not
affect the enzymatic activity. On the other hand, a number
of divalent metal ions increased enzymatic activity (CoCl2
by 33%; MnCl2, 39%; BaCl2, 42%; NiCl2, 47%; MgCl2,
49%; CaCl2, 52%; CuCl2 and ZnSO4, both by 69%).
hexene (6) were all brominated to the corresponding bro-
mohydrin compounds by the enzyme, specifically, trans-
2-bromo-1-indanol (7), 2-bromo-1-phenylethanol (8),
2-bromo-1-(4-chlorophenyl)-1-ethanol (9), 2-bromo-3-phe-
nyl-1,3-propanediol (10), and 2-bromo-1-cyclohexanol
(11). The above products corresponded to those in a previ-
ously report, and were thus easily confirmed (Yamada
et al., 1985a; Itoh et al., 1988). However, 2-bromo-1-pheny-
lethanol (8) from styrene (3) and trans-2-bromo-1-indanol
(7) from indene (2) were racemic mixtures on chiral HPLC
analysis, and therefore no stereoselectivity in bromination
was detected for K. alvarezii BPO. The enzyme additionally
catalyzed the bromination of 3-methyl-2-buten-1-ol (12) to
2-bromo-3-metyl-1,3-dihydroxybutane (13), which dis-
played comparable retention time and mass-spectra as the
authentic compound. However, bromination of 3-methyl-
2-buten-1-ol (12) analogues, including 3-methyl-3-butene-
1-ol (14), 2-methyl-2-propene-1-ol (15), and 3-butene-1-ol
(16), was not catalyzed by K. alvarezii BPO. All unreactive
compounds contained an aliphatic isolated terminal double
bond (Fig. 4).
No brominated products were obtained with anisole
(17), o-hydroxybenzyl alcohol (18) and phenol red (19),
implying that the enzyme catalyzes only specific reactions
of addition, and not substitution of phenolic compounds.
No brominated compounds (lipophilic, polar or volatile)
were detected when phenylpyruvic acid (20) was used as a
substrate. In addition, no activity was observed on pyruvic
(21) and oxalacetic (22) acids, known precursors of volatile
halomethanes for BPO in C. pilulifera leading to formation
of CHBr3, CHClBr2 and CH2Br2 (Itoh and Shinya, 1994).
Therefore, it was concluded that K. alvarezii enzyme does
not utilize a-keto acids or phenolic compounds as a bromin-
ation substrate. This is the first report on such clear sub-
strate preference by BPO. Our data strongly suggest that
the reaction of K. alvarezii BPO proceeds only at the active
site, and does not depend on released active brominated
species, such as Br+OHꢀ. This presumption contradicts
the fact that there is no chiral induction of bromination
for some substrates. We speculate that the enzyme contains
no strict substrate binding site to control the stereoselectiv-
ity in a similar manner as Corallina BPO. Our finding pro-
vides new aspect of vanadium-BPO in marine algae, and
possible application of the enzyme to substrate-specific
bromination.
2.4. Reconstitution of BPO with vanadate ions
A reconstitution process disclosed that the BPO of
K. alvarezii is a vanadium-dependent haloperoxidase.
The initial activity of the enzyme was calculated as
3.06 U/mg protein (100%). After 5 h incubation at pH
2.5 in the presence of Na-EDTA, total loss of enzymatic
activity was observed (0%), and thus low pH was neces-
sary to inactivate the enzyme. Overnight incubation with
NaVO3 at neutral pH led to a reconstitution of the enzyme
and recovery in activity to 1.29 U/mg (42%), while no
activity was evident upon incubation in the absence of
NaVO3. Moreover, it was observed that addition of
0.5 mM NaVO3 in the buffer solution effectively main-
tained enzymatic activity.
2.5. Substrate specificity of BPO for organic compounds and
stereoselectivity in bromination
Substrate specificity and stereoselectivity are key issues
in characterizing haloperoxidases, particularly with regard
to its reaction mechanism. This is because the bromination
of most organic compounds appear to nonenzymatically
proceed with the electrophilic addition/substitution of
molecular bromine (Br2) or hypobromous acid (BrOH),
in spite of the fact that bromide ion is oxidized by a vana-
dium-peroxo complex at the active sites of vanadium-
dependent BPOs (Dau et al., 1999). According to the mech-
anism, many organic compounds are brominated by this
enzyme. Indeed, Itoh and colleagues demonstrated that
despite evident substrate preference with BPO from C. pilu-
lifera (marine red alga), several phenolic and heterocyclic
compounds, alkenes, as well as keto acids were substrates
of the enzyme (Itoh et al., 1987a, 1988). Butler reviewed
an interesting phenomenon of vanadium-dependent A.
nodosum and C. officinalis BPO/CPO, specifically, the
active site bromination reaction when indole, but not cyto-
sine, is used as a substrate (Butler, 1998). Accordingly, we
examined a number of organic compounds for bromina-
tion by BPO from K. alvarezii, and compared them with
the previous data. Data presented in Fig. 4 reveals clear
substrate specificity of the enzyme. Indene (2), styrene
(3), 4-chlorostyrene (4), cinnamyl alcohol (5), and cyclo-
2.6. Determination of brominated compounds in K. alvarezii
In view of the existence of BPO in K. alvarezii, the
algae were analyzed for brominated compounds. K. alvar-
ezii emitted an irritant compound, which caused itchiness
on skin when disrupted. Consequently, volatile haloge-
nated compounds were directly measured by heating the
algae in a head-space bottle. No halogenated compounds,
including CHBr3 or CH2Br2, were detected by GC–MS.
On the other hand, GC–MS analysis of the EtOAc extract
of K. alvarezii revealed the presence of dibromoacetalde-

1362
Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
Fig. 4. Summary of substrate spectra of vanadium-dependent BPOs of different origins. Compounds brominated by BPO from K. alvarezii are displayed
inside a solid circle, and those brominated by the enzyme from C. pilulifera inside a broken line. The compounds, that have not been tested by BPO of
C. pilulifera, are presented with an asterisk.
hyde (23) (C2H2OBr2) (tR 14.98 min); EI-MS m/z (isotope
ratio, rel. int.): 200/202/204 [M]+ (1:2:1, 8), 172/174/176
(50), 120/122 (8), 93(100), which indicated a 97% identity
with the authentic compound registered in NIST
(National Institute of Standards and Technology, USA)
mass spectral library as shown in Fig. 5, and benzaldehyde
(tR 18.78 min); EI-MS m/z: 106 [M]+ (100), 105(80). We
could not isolate the compounds because of their small
amounts and easy vaporization. Dibromoacetaldehyde
(23), which is toxic and irritant for many organisms,
may participate in the defense mechanism of algae as allel-
ochemical against invertebrates and microorganisms. BPO
is possibly involved in the biosynthesis of dibromoacetal-
dehyde (23), although the biosynthetic pathway is cur-
rently unclear.
4. Experimental
4.1. Cultivation of algae
Kappaphycus alvarezii (Doty) Doty ex Silva (Soleriaceae,
Gigartinales, Rhodophyta) from the Philippines was culti-
vated in the subtropical waters during the summer in
Kochi, Japan. The cultivation method employed thallus
fragments fixed to floating ropes in the sea (Fig. 6).
Voucher-specimen was deposited in the herbarium of
Kochi University. Fresh algae cultured were frozen at
ꢀ20 ꢂC for experiments in the laboratory. Other algal sam-
ples were cultured in Kochi.
4.2. Crude enzyme reparations form the algae
In Hawaii islands, K. alvarezii is harmful to corals as a
result of shadowing or smothering, and causes coral death,
due to its high growth rate (Rodgers and Cox, 1999). Dib-
romoacetaldehyde (23) identified in algae by our group
may be involved in algal survival in the marine environ-
ment by supporting the growth rate. From this view point,
they should be more careful in aquaculturing a large
amount of K. alvarezii in the sea.
Defrosted algae (1–2 g wet weight) were disrupted with
sea sand (40–80 mesh) using a mortar and pestle at room
temperature, and with the whole them extracted with
10 mM 2-(N-morpholino)ethanesulphonic acid–NaOH
(MES) buffer (pH 7.0, buffer A) at a ratio of 0.1 g algal
sample/0.5 ml buffer at room temperature. Next, the crude
extract was centrifuged at 4 ꢂC for 30 min at 10,000g to
obtain the supernatant.
3. Concluding remarks
4.3. Partial purification of the enzyme
BPO was first identified in Kappaphycus alvarezi belong-
ing to the Solieriaceae, which are important industrial
sources of carrageenan. The enzyme was dependent on
vanadium (V), and displayed a distinct substrate spectrum
from that of previously reported BPOs from marine macro-
algae. Dibromoacetaldehyde accumulated in K. alvarezi
was additionally determined.
All manipulations were performed at room temperature
using 10 mM MES buffer, pH 7.0 (buffer A), or 10 mM
MES buffer, pH 7.0, plus 5 mM KBr and 0.5 mM NaVO3
(buffer B), unless stated otherwise. Each 30 g (total 120 g
wet weight) of algae washed with distilled water were dis-
rupted using a mortar and pestle, extracted with each

Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
1363
Fig. 5. GC–MS analysis of brominated compound in K. alvarezii.
The protein concentration was estimated by measuring
the absorbance at 280 nm or by using the bicinchoninic
acid (BCA) method (Smith et al., 1985), calibrated with
bovine serum albumin as a standard (BCA Protein Assay
Kit, Pierce/Thermo Scientific, USA).
4.4. Polyacrylamide gel electrophoresis (PAGE)
Analytical PAGE was performed in 7% polyacrylamide
gel with Tris–HCl buffer (pH 8.9) according to the method
of Davis (1964). The gels were stained for protein with
Coomassie brilliant blue G-250, and staining of gels for
enzyme activity was carried out by incubation in 10 mM
MES buffer (pH 6.5) containing 0.5% pyrogallol, 10 mM
H2O2 and 10 mM KBr in the dark at 4 ꢂC.
Fig. 6. The photo of K. alvarezii cultivated in the sea.
150 ml buffer B (total 600 ml), and centrifuged at 10,000g
for 30 min to obtain the supernatant as described above.
Labiase (0.5 g: more than 5 U of N-acetylmuramoylhydro-
lase), a lysis enzyme obtained from Seikagaku Corp.
(Tokyo), was added to the recovered supernatant
(350 ml), and incubated at 37 ꢂC for 48 h with gentle shak-
ing to dissolve the polysaccharide and decrease the viscos-
ity. After dialysis against buffer B, the enzyme solution was
subjected to freeze-drying and dissolved in distilled H2O
(140 ml). The solution was applied to a Octyl-Sepharose
hydrophobic column chromatography (6 by 8 cm) previ-
ously equilibrated with buffer B. Following application of
the enzyme solution to the column, some non-adsorbed
proteins and partially degraded polysaccharides were
eluted with buffer B. The enzyme was eluted with a increase
in concentrations of sodium cholate (0.2%, 0.5%, 1% and
1.5%) in buffer B (250 ml). Factions showing the activity
(1% and 1.5% of sodium cholate) were combined and con-
centrated to 140 ml with an ultrafiltration unit HHP-62K
equiped with a Q0100 polysulfone membrane filter (Advan-
tec Toyo, Tokyo, Japan). Following dialysis against buffer
B, the enzyme solution was subjected to a QAE-Toyoperl-
550 C column (6 cm by 6 cm) previously equilibrated with
buffer B. The column was washed with the same buffer,
and the enzyme eluted with a linear gradient of 0.005–
1.5 M KBr. Enzyme fractions that displayed activity were
combined, concentrated to 5 ml and subjected to dialysis,
and used as purified enzyme.
4.5. Enzymatic activity
Enzymatic activity was determined with the MCD test.
Enzyme solution (0.01 ml) was assayed in the reaction mix-
ture containing 10 mM MES buffer, pH 6.5, 60 lM MCD,
20 mM KBr and 10 mM H2O2 in a total volume of 1 ml at
25 ꢂC. Enzymatic activity (U/ml) was spectrophotometri-
cally measured by the decrease in absorbance of MCD at
290 nm (e = 19,900 cmꢀ1 Mꢀ1), according to the bromina-
tion of MCD.
Catalase activity was assayed by measuring the decrease
in absorbance at 240 nm (e = 43.6 Mꢀ1 cmꢀ1) of H2O2 in a
total volume of 1.0 ml (Itoh et al., 2001).
One unit (U) of enzyme activity was defined as the
amount of the enzyme that catalyzed the formation of bro-
minated MCD or decomposition of H2O2 in 1 min.
4.6. Halide ion specificity
Halide ion specificity of BPO was determined for Fꢀ,
Clꢀ, Brꢀ and Iꢀ. The general MCD test was performed
by replacing KBr with KF and KCl for Fꢀ and Clꢀ. For
iodide, the oxidation rate was directly measured. The reac-
tion mixture consisted of 100 mM MES buffer, pH 6.5,
5 mM KI and 2 mM H2O2, and 0.005 ml of enzyme solu-
tion in a total volume of 1 ml. The reaction rate was

1364
Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
calculated from the increase in absorbance at 350 nm due
to triiodide complex (Iꢀ3 ) formation, according to the
method of Hosoya (1963). A control containing no enzyme
was included in the experiments, and the difference between
the two rates was used to determine the rate of enzyme-cat-
alyzed triiodide complex formation.
One unit of the enzyme activity was determined as the
amount producing 1 lmol of ð½I2ꢁ þ ½Iꢀ3 ꢁÞ per one min from
Iꢀ. The equation is expressed as follows: U/ml enzyme
H2O2 and 2 mM KBr in a total volume of 0.5 ml. As
potential substrates, 3-methyl-2-buten-1-ol (12), 3-methyl-3-
butene-1-ol (14), 2-methyl-2-propene-1-ol (15), 3-butene-
1-ol (16), indene (2), styrene (3), 4-chlorostyrene (4),
trans-cinnamyl alcohol (5), cyclohexene (6), anisole (17),
o-hydroxybenzyl alcohol (18), phenol red (19), phenylpyru-
vate (20) (Itoh et al., 1988, 2001),were used at a final con-
centration of 10 mM. Blank tests without H2O2 and KBr
were additionally included. After 24 h reaction in a Bio-
shaker (1250 rpm, Taiteck, Tokyo) at 25 ꢂC, the reaction
mixture was subjected to centrifugation for 10 min at
13,000g. The supernatant was transferred and extracted
with EtOAc (0.5 ml). The EtOAc extract was dried with
anhyd. Na2SO4 and analyzed by GC–MS. Polar products
from phenylpyruvate were also analyzed by GC–MS after
conversion to methyl esters using trimethylsilyldiazome-
thane. The reagent was added drop-wise into the dried
and evaporated product mixture until the emission of bub-
bles stopped.
A QP-2010 GC–MS (Shimadzu) equipped with a DB-
VRX (J&W Scientific, 60 m · 0.25 mm i.d., 0.25 lm film
thickness) was employed for GC–MS analysis. The column
pressure was 130 kPa, and the linear velocity of He gas was
27 cm/s. Samples were injected in a split mode of 1:40 at
180 ꢂC with the following temperature program: 60 ꢂC
for 5 min, increased by 10 ꢂC/min to 250 ꢂC and a 10 min
hold at 250 ꢂC. Mass spectra were obtained at 70 eV with
an electron-impact ion source (EI, 200 ꢂC).
solution = (DA350with enzyme/min ꢀ DA350without
enzyme/
min)/26 · (1.3 · 10ꢀ3/[Iꢀ]initial + 1) · (volume of reaction
mixture/volume of enzyme solution).
4.7. Km values for H2O2 and KBr
The general MCD test was performed by successively
altering concentrations of each compound, KBr (0.5, 1, 5,
10, 20, 30, 40 mM final concentration) and H2O2 (0.1,
0.5, 2, 5, 10, 15, 20 mM final concentration) in the assay
mixture. Km values were calculated from a Lineweaver–
Burk plot.
4.8. Inhibitors and activators of the enzymatic activity
Enzymatic reactions were performed in the presence of
1 mM CuCl2, ZnSO4, NiCl2, BaCl2, CoCl2, CaCl2, MgCl2,
MnCl2, AgNO3, Na2MoO4, VOSO4, NaVO3, p-chloromer-
curibenzoic acid (p-CMB), iodoacetic acid and EDTA,
respectively, using the crude enzyme preparation, which
was extracted with buffer A without KBr and NaVO3.
To establish the effects of specific compounds on the sta-
bility and activity of BPO, the crude enzyme was dialyzed
against 10 mM MES buffer A, in the presence of 5 mM
each of the above mentioned compounds, 1 and 5 mM of
NaCl, 5 mM of KCl, 0.1, 0.5, 1 and 5 mM KBr for 12 h,
and the activity changes were measured.
trans-2-Bromo-1-indanol (7); retention time (tR)
28.74 min in GC–MS, EI-MS m/z (isotope ratio, rel.
int.): 212/214 [M]+ (1:1, 4.2), 133 (87), 115 (100), 103/
105 (70).
2-Bromo-1-phenylethanol (8); tR 26.14, EI-MS m/z:
200/202 [M]+ (1:1, 1), 107 (100), 103/105 (6.4).
2-Bromo-1-(4-chlorophenyl)-1-ethanol (9); tR 30.44, EI-
MS m/z: 234/236/238 [M]+ (3:4:1, 4.3), 154 (4), 141/143
(100), 125 (25), 113/115 (80), 103 (32).
4.9. Reconstitution of the enzyme with vanadate ion
2-Bromo-3-phenyl-1,3-propanediol (10); tR 32.19, EI-
MS m/z: 230/232 [M]+ (1:1, 1), 150 (1) 133 (13), 107
(100).
2-Bromo-1-cyclohexanol (11); tR 22.15, EI-MS m/z:
178/180 [M]+ (10.6), 158/160 (2.1), 132/134 (42), 122/
124 (28), 108 (28), 99 (100).
To remove the prosthetic metal ions from BPO, the pH
of the enzyme solution was decreased to 2.5 using 1 M
HCl, and 1 mM EDTA/Na (final concentration) was added
to the solution. After 5 h incubation (complete loss of enzy-
matic activity), pH was increased from 2.5 to 6.85. The test
solution was divided into two fractions. NaVO3 (5 mM
final concentration) was added to reconstitute the enzyme
in the second fraction. The two test solutions were dialyzed
against 10 mM MES buffer, pH 7.0, plus 1 mM EDTA/Na
(final concentration) to remove excess NaVO3 for 24 h. The
enzymatic activity of reconstituted enzyme was measured
and the degree of reconstitution evaluated.
4.11. Halomethane analysis
The method of Itoh and Shinya (1994) was applied for
determining the volatile halomethanes from fresh algae.
Four samples of K. alvarezii (0.5, 1.0, 2.0 and 5.0 g wet
weight), washed with cool autoclaved seawater and dried
with a paper towel for 2 min, were individually transferred
into 20 ml vials, and sealed with silicon rubber septa and
aluminum caps. Each vial was maintained at 60 ꢂC for
15 min to ensure transfer of volatile halomethanes into
the gas phase. The gas (0.1 ml) was withdrawn and ana-
4.10. Brominating activity and product analysis
To determine the substrates of K. alvarezii BPO, 0.05 ml
of enzyme solution (total 0.03 U) was added to the reaction
mixture consisting of 100 mM MES buffer, pH 6.5, 2 mM

Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
1365
lyzed by GC–MS. The above GC–MS apparatus was
employed for analysis at a column pressure of 130 kPa.
Samples were injected in a splitless mode at a temperature
program of 40 ꢂC for 5 min, increased by 10 ꢂC/min to
200 ꢂC and a 5 min hold at 200 ꢂC. Mass spectra were
obtained at 70 eV with an electron-impact ion source (EI,
200 ꢂC).
In vitro formation of halomethanes by BPO from spe-
cific keto acids was measured in a 10.0 ml reaction mixture
comprising 50 mM MES buffer, pH 6.5, 10 mM KBr,
10 mM H2O2, 2.0 ml of enzyme extract (total 1.2 U),
5 mM pyruvic, phenylpyruvic or oxalacetic acid, respec-
tively, in a total volume of 20 ml within each vial as
described above. Reactions were performed for 24 h at
25 ꢂC, followed by 60 ꢂC for 15 min. A control run was
performed without the enzyme, KBr or H2O2.
with a yield of 78% (714 mg); 1H NMR (CD2Cl2) d
(ppm) 1.4 (s, 3H), 1.4 (s, 3H), 4.13 (t, 1H), 4.01 (d, 2H).
Octyl-Sepharose was prepared as follows: Sepharose
CL-4B (200 ml) was washed with distilled H2O, and sus-
pended in 0.6 M NaOH (260 ml). Epichlorohydrin
(20 ml) was added to the solution, and stirred for 2 h in a
water bath at 40 ꢂC. The mixture was filtered, and the
epoxy-activated Sepharose CL-4B was extensively washed
with distilled H2O. Activated Sepharose was resuspended
in 0.1 M NaOH (260 ml) containing 20% dioxane (v/v)
and 3.4 g of octylamine, and stirred for 20 h in a water bath
at 30 ꢂC. Next, octyl-Sepharose was filtered, and com-
pletely washed with 0.1 M NaCl containing 35% 1-PrOH
(v/v), followed by distilled H2O. All other chemicals were
of analytical grade, and available commercially.
Acknowledgements
4.12. Analysis of the stereoselectivities of bromination
The scholarship from Japan Society for the Promotion
of Science (JSPS) for the post-doctoral research of Dr.
Kamenarska is gratefully acknowledged. This work was
partly supported by a Grant-in Aid for Scientific Research
provided by the JSPS.
To determine the absolute configuration of the bro-
mohydrine products, analytical HPLC using a Shimadzu
LC-10AT system equipped with
a Chiracel OB-H
(4.6 · 250 mm, Daicel Chemical Industries, Osaka, Japan)
was performed for trans-2-bromo-1-indanol (7) and
(R,S)-2-bromo-1-phenylethanol (8). The mobile phase con-
tained n-hexane/2-propanol (9:1 w/w) at a flow rate of
0.5 ml/min. The product was spectrophotometrically
detected at 220 nm. Retention times of racemic trans-2-
bromo-1-indanol (7) were 12.3 and 16.8 min, while those
of (R,S)-2-bromo-1-phenylethanol (8) were 16.6 and
20.5 min, respectively (Kawanami et al., 2002).
References
Butler, A., 1998. Vanadium haloperoxidases. Curr. Opin. Chem. Biol. 2,
279–285.
Collen, J., Abrahamsson, K., Ekdahl, A., Pedersen, M., 1994. The
involvement of hydrogen peroxide in the production of volatile
halogenated compounds by Meristiella gelidium. Phytochemistry 36,
1197–1202.
4.13. Analysis of brominated compounds in K. alvarezii
Cowan, M.M., 1999. Plant products as antimicrobial agents. Clin.
Microbiol. Rev. 12, 564–582.
Samples of K. alvarezii (about 5 g) were extracted with
EtOAc, MeOH and CHCl3 (15 ml each), respectively, at
37 ꢂC with stirring at 115 rpm for 48 h. Each extract was
evaporated to a small amount, and a portion was subjected
to GC–MS analysis (see Section 4.10 for details). Acids
isolated with MeOH and CHCl3 were analyzed after con-
version to their methyl esters using trimethylsilyldiazome-
thane, as described above. Similar GC–MS conditions
were applied for analysis.
Dau, H., Dittmer, J., Epple, M., Hanss, J., Kiss, E., Rehder, D., Schulzke,
C., Vilter, H., 1999. Bromine K-edge EXAFS studies of bromine
binding to bromoperoxidase from Ascophyllum nodosum. FEBS Lett.
457, 237–240.
Davis, B.J., 1964. Disc electrophoresis II, method and application to
human serum proteins. Ann. New York Acad. Sci. 121, 404–427.
De Boer, E., Tromp, M.G.M., Plat, H., Krenn, B.E., Wever, R., 1986.
Vanadium(V) as a essential element for haloperoxidase activity in
marine brown algae: purification and characterization of a vana-
dium(V)-containing bromoperoxidase from Laminaria saccharina.
Biochim. Biophys. Acta 872, 104–115.
Denboh, T., Suzuki, M., Mizuno, Y., Ichimura, T., 1997. Suppression of
Laminaria sporelings by allelochemicals from coralline red algae. Bot.
Marina 40, 249–256.
4.14. Chemicals
Dong, C., Flecks, F., Unversucht, S., Haupt, C., van Pee, K.-H.,
Naismith, J.H., 2005. Tryptohan 7-halogenase (PmA) structure
suggests a mechanism for regioselective chlorination. Science 306,
2216–2219.
2-Bromo-3-metyl-1,3-dihydroxybutane (13) was synthe-
sized from 3-methyl-2-buten-1-ol (12) by bromination with
N-bromosuccinimide (Hanzlik, 1988). 3-Methyl-2-buten-1-
ol (12) (430 mg, 5 mmol), water (225 mg, 12.5 mmol) and
N-bromosuccinimide (1.78 g, 10 mmol) were dissolved in
10 ml of DMSO, and maintained at room temperature
for 20 min. To terminate the reaction, excess water was
added to the mixture. The product was extracted with
CH2Cl2, dried with anhydrous Na2SO4, and purified by sil-
ica gel chromatography using n-hexane and EtOAc (4:1).
2-Bromo-3-metyl-1,3-dihydroxybutane (13) was obtained
Faulkner, D.J., 2000. Marine natural products. Nat. Prod. Rep. 17, 7–55.
Faulkner, D.J., 2001. Marine natural products. Nat. Prod. Rep. 18, 1–49.
Faulkner, D.J., 2002. Marine natural products. Nat. Prod. Rep. 19,
1–48.
Hanzlik, R.P., 1988. Selective epoxidation of terminal double bonds:
10,11-epoxyfarnesyl acetate. Org. Syn. Coll. 6, 560.
Hosoya, T., 1963. Effect of various reagents including antithyroid
compounds upon activity of thyroid peroxidase. J. Biochem. 53,
381–388.

1366
Z. Kamenarska et al. / Phytochemistry 68 (2007) 1358–1366
Ishikawa, Y., Mihara, Y., Gondoh, K., Suzuki, E-I., Asano, Y., 2000. X-
ray structures of a novel acid phosphatase from Escherichia blatae and
its complex with the transition-state analog molybdate. EMBO J. 19,
2412–2423.
Ohno, M., Largo, D., Ikumoto, T., 1994. Growth rate, carrageenan yield
and gel properties of cultured kappa-carrageenan producing red alga
Kappaphycus alvarezii (Doty) Doty in the subtropical waters of
Shikoku, Japan. J. Appl. Phycol. 6, 1–5.
Itoh, N., Shinya, M., 1994. Seasonal evolution of bromomethanes from
coralline algae and its effect on atmospheric ozone. Marine Chem. 45,
95–103.
Ohno, M., Nang, H.Q., Hirase, S., 1996. Cultivation and carrageenan
yield and quality of Kappaphycus alvarezii in the waters of Vietnam. J.
Appl. Phycol. 8, 431–437.
Itoh, N., Izumi, Y., Yamada, H., 1986. Characterization of nonheme type
bromoperoxidase in Corallina pilulifera. J. Biol. Chem. 261, 5149–
5200.
Ohsawa, N., Ogata, Y., Okada, N., Itoh, N., 2001. Physiological function
of bromoperoxidase in the red marine algae, Corallina pilulifera:
production of bromoform as an allelochemical and the simultaneous
elimination of hydrogen peroxide. Phytochemistry 58, 683–692.
Ohshiro, T., Littlechild, J., Garcia-Rodrigues, E., Isupov, M., Iida, Y.,
Kobayashi, T., Izumi, Y., 2004. Modification of halogen specificity
of a vanadium-dependent bromoperoxidase. Protein Sci. 13, 1566–
1571.
Itoh, N., Izumi, Y., Yamada, H., 1987a. Haloperoxidase-catalyzed
halogenation of nitrogen-containing aromatic heterocycles represented
by nucleic bases. Biochemistry 26, 282–289.
Itoh, N., Izumi, Y., Yamada, H., 1987b. Characterization of nonheme
iron and reaction mechanism of bromoperoxidase in Corallina
pilulifera. J. Biol. Chem. 262, 11982–11987.
Rodgers, S.K., Cox, E.F., 1999. Rate of spread of introduced rhodophytes
Kappaphycus alvarezii, Kappaphycus striatum, and Gracilaria salicornia
and their current distributions in Kane ‘ohe Bay, O‘ahu, Hawaii.
Pacific Sci. 53, 232–241.
Itoh, N., Hasan, A.K.M.Q., Izumi, Y., Yamada, H., 1988. Substrate
specificity, regiospecificity and stereospecificity of halogenation reac-
tions catalyzed by non-heme-type bromoperoxidase of Corallina
pilulifera. Eur. J. Biochem. 172, 477–484.
Sakata, K., Iwase, Y., Ina, K., Fujita, D., 1991. Halogenated terpenes
isolated from the red alga Plocamium leptophyllum as feeding
inhibitors for marine herbivores. Nippon Suisan Gakkaishi 57, 743–
746.
Itoh, N., Kawanami, T., Liu, J.-Q., Dairi, T., Miyakoshi, M., Nitta, C.,
Kimoto, Y., 2001. Cloning and biochemical characterization of Co2+
-
activated bromoperoxidase-esterase (perhydrolase) from Pseudomonas
putida IF-3 strain. Biochim. Biophys. Acta. 1545, 53–66.
Kamenarska, Z., Hiraoka, M., Itoh, N., 2006. Preliminary study on the
bromoperoxidase activity of some macroalgae from Japan. Comp.
Rend. Acad. Bulg. Sci. 59, 71–74.
Sheffield, D.J., Smith, A.J., Harry, T.R., Rogers, L.J., 1993. Thermosta-
bility of the vanadium bromoperoxidase from Corallina officinalis.
Biochem. Soc. Trans. 21, 445.
Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H.,
Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olsen, B.J., Klenk,
D.C., 1985. Measurement of protein using bicinchoninic acid. Anal.
Biochem. 150, 76–85.
Kawanami, T., Miyakoshi, M., Dairi, T., Itoh, N., 2002. Reaction
mechanism of the Co2+-activated multifunctional bromoperoxidase-
esterase from Pseudomonas putida IF-3. Arch. Biochem. Biophys. 398,
94–100.
Soedjak, H.S., Butler, A., 1990. Chlorination catalyzed by vanadium
bromoperoxidase. Inorg. Chem. 29, 5015–5017.
Krenn, B.E., Tromp, M.G.M., Wever, R., 1989. The brown algae
Ascophyllum nodosum contains two different vanadium bromoperox-
idases. J. Biol. Chem. 264, 19287–19292.
van Pee, K.-H., Patallo, E.P., 2006. Flavin-dependent halogenases
involved in secondary metabolism in bacteria. Appl. Microbiol.
Biotechnol. 70, 631–641.
Littlechild, J., Garcia-Rodrigues, E., Dalby, A., Isupov, M., 2002.
Structural and functional comparisons between vanadium haloperox-
idase and acid phosphatase enzymes. J. Mol. Recognit. 15, 291–296.
Manthey, J.A., Hager, L.P., 1981. Purification and properties of brom-
operoxidase from Penicillus capitatus. J. Biol. Chem. 256, 11232–
11238.
van Pee, K.-H., Keller, S., Wage, T., Wynands, I., Schnerr, H., Zehner, S.,
1997. Enzymatic halogenation catalyzed via a catalytic triad and by
oxidoreductases. Biol. Chem. 381, 1–5.
Vilter, H., 1984. Peroxidases from phaeophyceae: a vanadium(V)-depen-
dent peroxidase from Ascophyllum nodosum. Phytochemistry 23, 1387–
1390.
Morris, D.R., Hager, L.P., 1966. Chloroperoxidase I. Isolation and
properties of the crystalline glycoprotein. J. Biol. Chem. 241, 1763–
1768.
Wever, R., Plat, H., De Boer, E., 1985. Isolation procedure and some
properties of the bromoperoxidase from the sea weed Ascophyllum
nodosum.. Biochim. Biophys. Acta 830, 181–186.
Mtolera, M.S.P., Collen, J., Pedersen, M., Ekdahl, A., Abrahamsson, K.,
Semest, A.K., 1996. Stress-induced production of volatile halogenated
organic compounds in Euchuma denticulatum (Rhodophyta) caused by
elevated pH and high light intensities. Eur. J. Phycol. 31, 89–95.
Ohbuchi, K., Hasegawa, K., Hamachi, M., Ozeki, K., Kumagai, C., 2001.
Isolation of a new lytic enzyme for hiochi bacteria and other lactic acid
bacteria. J. Biosci. Bioeng. 91, 487–492.
Yamada, H., Itoh, N., Izumi, Y., 1985a. Chloroperoxidase-catalyzed
halogenation of trans-cinnamic acid and its derivatives. J. Biol. Chem.
260, 11962–11969.
Yamada, H., Itoh, N., Murakami, S., Izumi, Y., 1985b. New bromop-
eroxidase from coralline algae that brominates phenol compounds.
Agric. Biol. Chem. 49, 2961–2969.