Purification, Characterization, and cDNA Cloning of a Novel Lectin from the Green Alga, Codium barbatum.

Biosci. Biotechnol. Biochem., 76 (4), 805–811, 2012

Purification, Characterization, and cDNA Cloning of a Novel Lectin
from the Green Alga, Codium barbatum
Danar P RASEPTIANGGA, Makoto H IRAYAMA, and Kanji H ORIy
Food Science and Biofunctions Division, Graduate School of Biosphere Science, Hiroshima University,
1-4-4 Kagamiyama, Higashi-Hiroshima 739-8528, Japan
Received December 8, 2011; Accepted December 28, 2011; Online Publication, April 7, 2012
[doi:10.1271/bbb.110944]

A novel lectin (CBA) was isolated from the green
alga, Codium barbatum, by conventional chromatographic methods. The hemagglutination-inhibition profile with sugars and glycoproteins indicated that CBA
had preferential affinity for complex type N-glycans but
not for monosaccharides, unlike the other known
Codium lectins specific for N-acetylgalactosamine.
CBA consisted of an SS-linked homodimer of a 9257Da polypeptide containing seven cysteine residues, all of
which were involved in disulfide linkages. The cDNA of
the CBA subunit coded a polypeptide (105 amino acids)
including the signal peptide of 17 residues. The calculated molecular mass from the deduced sequence was
9705 Da, implying that the four C-terminal amino acids
of the CBA proprotein subunit were post-translationally

truncated to afford the mature subunit (84 amino acids).
No significantly similar sequences were found during an
in silico search, indicating CBA to be a novel protein.
CBA is the first Codium lectin whose primary structure
has been elucidated.
Key words:

Codium barbatum; lectin; carbohydratebinding specificity; cDNA cloning; primary
structure

Lectins, or carbohydrate-binding proteins, are widely
distributed in nature and play some important roles as
recognition molecules in cell-cell or cell-matrix interactions.1) However, the physiological functions of most
lectins remain to be investigated in more detail,
especially for those of plant origin.1,2) The carbohydrate-binding specificity and molecular structure of
lectins are diverse and generally depend on the organisms from which they originate, although lectins have
been classified into several families based on the
amino acid sequences of their carbohydrate-recognition
domains, some of which are also evolutionarily conserved.3,4) The diversity of lectins in their carbohydratebinding specificity enables them to be used as convenient tools to decode the carbohydrate structures on a cell
surface and in a body fluid. Recent studies have been

directed to the nutritional aspects of lectins, because
many edible plants, including cereals, vegetables, and
fruits, contain lectin proteins, and some plant lectins
y

have been reported to affect the transport of other food
gradients in an intestinal tract model.1,5)
Lectins have been isolated and characterized from
macroalgae and microalgae. These ‘algae-specific’
lectins have shown novel carbohydrate-binding specificity and molecular structures, including those specific
for high-mannose N-glycans6–14) and core (1,6) fucosylated N-glycans.15,16) They have also shown some
interesting biological activities, including anticarcinogenic17) and antiviral effects.6–9,11–13) Their antiviral
activities are especially remarkable because they inhibit
in vitro infection by the immunodeficiency virus (HIV1) and influenza virus with a half-maximal effective
concentration (EC50 ) in the low nanomolar to picomolar
range through their binding with viral envelope glycoproteins.18) Many algal lectins, especially from red
algae, share some common characteristics of low molecular weight, monomeric form, thermostability, and
divalent cation-independent hemagglutination, and having affinity only for glycoproteins and not for monosaccharides. These properties of algal lectins are
dissimilar to most land plant lectins that have affinity
for monosaccharides and consist of oligomeric forms.2)

Monosaccharide-binding lectins have recently been
reported from several species of green algae belonging to
the genera Enteromorpha,19) Ulva,20,21) and Codium.22–24)
Among these, the lectins from the Codium, C. fragile
ssp. tomentosoides,22) C. fragile ssp. atlanticum,22) C.
tomentosum,23) and C. giraffa24) are commonly specific
for N-acetylgalactosamine and/or N-acetylglucosamine
and consist of oligomeric forms, except for a monomeric
lectin from C. giraffa. The resemblance of the lectin
properties may be derived from the close evolutionary
distance between the green algae of the genus Codium
and the land plant, although there is no information
on the primary structures of Codium lectins. The lectin
from C. fragile ssp. tomentosoides has shown preferential affinity for the Forssman antigen sugar chain
that is a marker of cancer.25) Thus, the lectins of the
genus Codium are interesting targets for their application
as biochemical and clinical reagents, as well as to
provide insight into the molecular evolution of lectins in
the plant kingdom. The characterization of Codium
lectins may also be significant in evaluating their

To whom correspondence should be addressed. Tel/Fax: +81-82-424-7931; E-mail: kanhori@hiroshima-u.ac.jp
Abbreviations: CBA, Codium barbatum lectin; BLAST, basic local alignment search tool; BSM, bovine submaxillary mucin; GNA, Galanthus
nivalis lectin; DTT, dithiothreitol; ESI-MS, electrospray ionization-mass spectrometry; HIV, human immunodeficiency virus; Lys-C, lysylendopeptidase-C; PB, phosphate buffer; PBS, phosphate-buffered saline; PSM, porcine stomach mucin; PTG, porcine thyroglobulin; RACE, rapid
amplification of cDNA ends; PE, S-pyridylethylated; SDS–PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TB, Tris buffer; TBS,
Tris-buffered saline; TRBC, trypsin-treated rabbit erythrocyte; WGA, wheat germ agglutinin

806

D. PRASEPTIANGGA et al.

nutritional aspects, because Codium has been reported
as edible.26)
Our recent screening of lectins in marine algae
showed the extract from C. barbatum to have strong
hemagglutination activity. However, this activity was
not inhibited by any of the monosaccharides examined,
including N-acetylgalactosamine, unlike the other
known Codium lectins. This present study deals with a
novel lectin isolated from C. barbatum.

Materials and Methods
Materials. The C. barbatum green alga was collected from SatsumaIwojima Island of Kagoshima in Japan, and was kept at
30  C until
being used. A portion of the C. barbatum tissue collected from
Tanegashima Island of Kagoshima in Japan was also stored at
30  C
in RNAlater (Invitrogen) for subsequent RNA extraction. TSKgel
Phenyl-5PW and TSKgel ODS 80TM columns were purchased from
Tosoh Corporation (Tokyo, Japan), and the YMC Protein-RP column
from YMC (Kyoto, Japan). The silica gel-immobilized phosphorylcholine (PC 300S (N)) column was generously provided by Dr. K.
Muramoto (Tohoku University, Japan). D-Glucose, D-galactose, Nacetyl-D-galactosamine, transferrin, fetuin, porcine thyroglobulin
(PTG), bovine submaxillary mucin (BSM), and porcine stomach mucin
(PSM) were purchased from Sigma-Aldrich Co. (MO, USA). DMannose, L-fucose, N-acetyl-D-glucosamine, N-acetylneuraminic acid,
lactose, and yeast mannan were purchased from Nacalai Tesque Co.
(Kyoto, Japan). D-Xylose and L-rhamnose were purchased from Wako
Chemical Co. (Osaka, Japan). The desialylated (asialo-) derivatives of
glycoproteins were prepared by hydrolyzing the parent sialoglycoprotein with 0.05 M HCl for 1 h at 80  C, with subsequent dialysis overnight
against saline.27) A marker kit for molecular weight determination by
SDS–PAGE was purchased from Tefco (Tokyo, Japan). As reference
proteins in gel filtration, bovine serum albumin (67 kDa), chymotrypsinogen (25 kDa), and ribonuclease A (13.7 kDa) were obtained from
Sigma-Aldrich Co., and ovalbumin (44 kDa) from Wako Chemical Co.

Lysylendopeptidase (Lys-C) was purchased from Sigma-Aldrich Co.,
all other chemicals used in this study being of the highest purity
available.
Purification of the C. barbatum lectin. A frozen sample (500 g) of
C. barbatum was thawed, cut into small pieces, and ground in liquid
nitrogen into powder. The powdered alga was stirred overnight at 4  C
with 500 mL of a 20 mM Tris–HCl buffer (pH 7.5) containing 150 mM
NaCl (TBS). The mixture was centrifuged at 13,500 g for 30 min, and
to the supernatant, solid ammonium sulfate was added to attain a final
concentration with 75% saturation. The mixture was kept overnight at
4  C and centrifuged for 30 min at 13;500 g. The precipitate was
dissolved and thoroughly dialyzed against TBS. The internal fraction
was further centrifuged for 30 min at 10;000 g, and the supernatant was
recovered as a salting-out fraction. This salting-out fraction was
dialyzed against a 20 mM Tris–HCl buffer at pH 7.5 (TB) containing
1 M ammonium sulfate. Five mL each of the dialyzate was applied to
the TSKgel Phenyl-5PW column (7:5  75 mm) that had been
equilibrated with TB containing 1 M ammonium sulfate. The column
was washed for 20 min with the starting buffer and then eluted for
40 min with a linear gradient of ammonium sulfate concentration (1.0–

0 M) in TB and then with TB alone for 30 min. The flow rate was
0.5 mL/min. Fractions of 2 mL were collected and measured for their
absorbance at 280 nm and for hemagglutination activity. The active
fractions showing hemagglutination activity were pooled, dialyzed
against TBS, and subjected to gel filtration in the PC 300S (N) column
(7:8  300 mm) that had been equilibrated with a 50 mM phosphate
buffer at pH 7.0 containing 0.15 M NaCl (PBS), according to the
method reported by Watanabe et al.28) The column was eluted at a flow
rate of 0.5 mL/min with PBS, and fractions of 2 mL were collected and
measured for their absorbance at 280 nm and for hemagglutination
activity.
Hemagglutination assay. Hemagglutination activity was determined
in a 96-well microtiter V-plate, using a 2% suspension (v/v) of trypsintreated rabbit erythrocytes (TRBC).29) Serially two-fold dilutions

(25 mL each) of a test solution were first prepared in saline, and to each
well, 25 mL of TRBC was added. The plate was gently shaken and
allowed to stand at room temperature for 1 h. Hemagglutination was
macroscopically observed and judged as positive when more than 50%
of TRBC in the well had agglutinated. The hemagglutination activity is
expressed as a titer, the reciprocal of the highest two-fold dilution

exhibiting positive hemagglutination.
TRBC was prepared from rabbit blood purchased from the
Laboratory of Animal Experiments (Hiroshima, Japan). The blood
cells were washed three times with 50 volumes of saline and suspended
in saline to give a 2% (v/v) suspension of native erythrocytes. A tenth
volume of 0.5% (w/v) trypsin in saline was added to a 2% suspension
of native erythrocytes, and the mixture was incubated for 60 min at
37  C. The incubated erythrocytes were washed four times with saline,
and a 2% suspension (v/v) of TRBC was prepared in saline.
Determination of the protein contents. The protein contents were
determined by the method of Lowry et al.,30) using bovine serum
albumin as a standard.
Effects of pH, temperature, and divalent cations on the hemagglutination activity. The effects of pH, temperature, and divalent
cations on the hemagglutination activity were next examined. To
determine the effect of pH, 500 mL each of a test solution was dialyzed
overnight against 100 mL of a 50 mM buffer of various pH values
between 3 and 10 at 4  C. The pH-treated solution was further dialyzed
against PBS (pH 7.0) and subjected to the hemagglutination assay. The
buffers used were glycine-HCl for pH 3.0, acetate for pH 4.0 and 5.0,
phosphate for pH 6.0 and 7.0, Tris–HCl for pH 8.0, and carbonate for

pH 9.0 and 10.0. To determine the effect of temperature, 500 mL each
of a test solution was incubated for 30 min at various temperatures
from 30  C to 100  C and then subjected to the hemagglutination assay
after cooling. To determine the effect of divalent cations, 500 mL of a
test solution was dialyzed overnight at 4  C against 50 mM EDTA in
PBS, and the internal fraction was measured for its hemagglutination
activity. Additionally, to this internal fraction, an equal volume of
20 mM CaCl2 or 20 mM MgCl2 in saline was added, the mixture being
kept for 2 h at room temperature and then measured for its
hemagglutination activity.
Hemagglutination-inhibition assay. The inhibitory effects of sugars
and glycoproteins on the hemagglutination of the lectin solution were
next examined. Serially two-fold dilutions (25 mL each) of a sugar or a
glycoprotein were first prepared in saline in a microtiter V-plate. To
each well, an equal volume of the lectin solution with a hemagglutination titer of 4 was added, and the plate was gently shaken and
allowed to stand at room temperature for 1 h. Finally, 25 mL of TRBC
was added to each well, and the plate was gently shaken and allowed to
stand for another 1 h. The inhibition of hemagglutination was macroscopically observed, the inhibition activity being expressed as the
minimum inhibitory concentration (mM or mg/mL), the lowest
concentration of a sugar or glycoprotein at which complete inhibition

of hemagglutination was achieved.
The sugars and glycoproteins used were D-glucose, D-mannose,
D-galactose, N-acetyl-D-glucosamine, N-acetyl-D-galactosamine, Lfucose, N-acetylneuraminic acid, D-xylose, L-rhamnose, and lactose
as sugars, and transferrin, asialo-transferrin, fetuin, asialo-fetuin,
yeast mannan, PTG, asialo-PTG, BSM, asialo-BSM, and PSM as
glycoproteins.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–
PAGE). SDS–PAGE was performed by using 15% gel.31) The sample
was boiled at 100  C for 5 min with or without 2% (v/v) 2mercaptoethanol. The gel was stained with Commasie brilliant blue
R-250 after electrophoresis.
Preparation of S-pyridylethylated (PE-) CBA. S-Pyridylethylation
was performed according to the method previously described.10)
Briefly, the purified lectin (200 mg each of CBA) was dissolved in
400 mL of 500 mM TB (pH 8.5) containing 6 M guanidine-HCl and
10 mM EDTA. The lectin solution was allowed to stand in the presence
or absence of DTT (8 mL) at room temperature for 2 h in a nitrogen gas
atmosphere. Subsequently, 2 mL of 4-vinyl pyridine was added to the

Novel Lectin from the Green Alga, Codium barbatum

807

Table 1. Primer Sequences Used in the cDNA Cloning of CBA
Sequence (from 50 to 30 )

Primer
CBA d F1
CBA d F2
CBA R1
CBA R2
CBA 50 end F
CBA 30 end R
GeneRacer 30 primera
GeneRacer 30 nested primera
GeneRacer 50 primera
GeneRacer50 nested primera
a
b

b

ACICAYGGIATHAARAAYGA
AAYGAYTGYGGIGTICCIGTb
TCCAAGCAGCATACGAACAC
TCATCAGTCCCAGTCCAACA
AATATAAATTAAACGTAAGGTATCGC
AGTTCTGCATATGGATGTAC
GCTGTCAACGATACGCTACGTAACG
CGCTACGTAACGGCATGACAGTG
CGACTGGAGCACGAGGACACTGA
GGACACTGACATGGACTGAAGGAGTA

Position
101–120
116–135
198–217
124–143
1–26
457–476
—
—
—
—

These primers were inferred from the GeneRacer Kit (Invitrogen).
Key to symbols of the degenerated nucleotides: I represents inosine, Y represents C and T, H represents A, T and C, R represents A and G.

lectin solution, and the mixture incubated for another 2 h. The resulting
PE-CBA, reduced and S-pyridylethylated (rPE)-CBA or non-reduced
and S-pyridylethylated (nrPE)-CBA was purified by reverse-phase
HPLC in the YMC Protein-RP column (6:0  250 mm), eluting with a
linear gradient between 10% and 70% acetonitrile in 0.1% TFA. The
eluate was monitored by its absorption at 280 nm, and the peak
containing PE-CBA was manually collected.
Enzymatic cleavage and separation of the peptides. rPE-CBA
(100 mg) was digested at 37  C for 18 h with lysylendopeptidase-C
(Lys-C) at an enzyme/substrate ratio of 1/100 (w/w) in 100 mM TB
(pH 8.5) containing 4 M urea. The peptide fragments were separated by
reverse-phase HPLC in the YMC Protein-RP column (6:0  250 mm),
eluting with a linear gradient between 2% and 70% acetonitrile in 0.1%
TFA. The eluate was monitored by its absorption at 220 nm, and the
peaks were manually collected.
Molecular mass determination of the proteins and peptides. The
molecular masses of intact CBA, nrPE-CBA, rPE-CBA, and the
peptides generated by enzymic digestion were determined by electrospray ionization-mass spectrometry (ESI-MS), using LTQ Orbitrap XL
(Thermo Fisher Scientific, MA, USA) and LCQ instruments (Finnigan,
Bremen, Germany).
Determination of the N-terminal amino acid sequences. The Nterminal amino acid sequences of rPE-CBA and its peptide fragments
were determined by using a Procise 492 HT protein sequencing system
(Applied Biosystems, CA, USA).
Rapid amplification of the 30 and 50 cDNA ends (30 and 50 RACE) of
CBA. Total RNA of C. barbatum was extracted from the RNAlatertreated algal tissues by using the Plant RNA Isolation reagent
(Invitrogen). Full-length cDNAs were synthesized from 5 mg of total
RNA by using a GeneRacer kit (Invitrogen) according to the
manufacturer’s instructions.
The first PCR for 30 RACE was initiated by adding 0.2 mL each of
a 10-fold dilution of the synthesized cDNA as a template to 8 aliquots
of a 9.8-mL solution containing 1 mL of a 10  Blend Taq buffer
(Toyobo, Osaka, Japan), 2 nmol each of dNTP, 6 pmol of the
GeneRacer 30 primer, 50 pmol of the degenerated CBA d F1 primer
(designed according to the N-terminal amino acid sequence of rPECBA (Table 1)), and 0.25 units of Blend Taq DNA polymerase
(Toyobo). The reaction was performed with a T Gradient Thermocycler (Biometra, Go¨ttingen, Germany) under the following conditions: denaturation at 94  C for 5 min, followed by 35 cycles consisting
of denaturation at 94  C for 30 s, annealing at a gradient temperature of
50  C to 64  C (2  C increments) for 30 s and extension at 72  C for
90 s, and the final extension step at 72  C for 5 min. The PCR products
in 8 aliquots were pooled, diluted 100-fold, and used as a template for
nested PCR. Nested PCR was performed by the same method, except
that 0.2 mL of the dilution was used as a template, and 2 pmol of the
GeneRacer 30 nested primer and 50 pmol of the degenerated CBA d F2
primer (designed according to the sequence of the peptide fragment
generated by Lys-C digestion) as a primer pair (Table 1), and an
annealing temperature of 54  C were used. The nested PCR products
were subcloned into the pGEM-T Easy vector (Promega, WI, USA)

and transformed into Escherichia coli DH5 competent cells. Plasmids
from the transformants were purified with a HiYield Plasmid Mini kit
(RBC Bioscience, New Taipei City, Taiwan) according to the
manufacturer’s instructions. DNA sequencing was performed by using
BigDye Terminator Cycle Sequencing kit ver. 3.1 with an ABI 3130xl
genetic analyzer (Applied Biosystems).
50 RACE was performed in a similar way to 30 RACE just described,
except that the primer pair of GeneRacer 50 and CBA R1 designed
from the sequence obtained by 30 RACE (Table 1) were used. The PCR
reaction was performed by using a high-fidelity DNA polymerase
KOD Plus Neo kit (Toyobo) as follows: denaturation at 94  C for
2 min, followed by 35 cycles consisting of denaturation at 98  C for
10 s, annealing at gradient temperatures of 50  C to 64  C (2  C
increments) for 30 s and then extension at 68  C for 30 s, and the final
extension step at 68  C for 5 min. Nested PCR was then performed by
the same method, except for using a 100-fold dilution of the first PCR
products as a template, and the GeneRacer 50 nested primer and CBA
R2 as the primer pair (Table 1). The PCR products were treated with a
10  A-attachment mix (Toyobo), before subcloning into the pGEM-T
Easy vector (Promega) and transforming into E. coli DH5 competent
cells. Plasmid purification and DNA sequencing were performed as
already described.
To verify the accuracy of the CBA sequences obtained by the 50 and
0
3 RACE operations, full-length cDNA of CBA was further amplified
by using KOD Plus Neo (Toyobo) and a primer pair of CBA 50 end F
and CBA 30 end R (Table 1) which were designed from the 50 and 30
terminal sequences of CBA cDNA obtained by 50 and 30 RACE.
Subcloning and DNA sequencing were then performed as already
described.
Sequence data processing. Similarity searches for CBA were
performed by using the basic local alignment search tool (BLAST)
with algorithms of protein-protein BLAST (BLASTP) and positionspecific iterated (PSI) BLAST. The signal peptide region was predicted
with SignalP 3.0.32)
Nucleotide sequence accession number. The sequence data for CBA
have been submitted to the DDBJ/EMBL/GenBank databases under
accession no. AB675415.

Results
Purification of the C. barbatum lectin
The lectin from C. barbatum was extracted with TBS
and efficiently recovered as a precipitate with 75%saturated ammonium sulfate (the salting-out fraction).
Hydrophobic chromatography in the TSKgel Phenyl5PW column for the salting-out fraction gave a single
peak of hemagglutination activity which was eluted as a
shoulder peak of protein (Fig. 1A). The active fractions
were pooled, dialyzed, and further purified by gel
filtration in the PC-300S (N) column to afford a single
peak (Fig. 1B). The purified lectin thus obtained gave a
single protein peak by reverse-phase HPLC in the YMC

808

Table 2. Purification Procedure for the Lectin from C. barbatum
1.0

Purification procedure

0

5

10

15

20

1

Extraction

450.4

Ammonium sulfate-precipitation
(75%-(NH4 )2 SO4 )
Hydrophobic chromatography

184.6

0

Gel filtration

8.0
2.1

25

THAb
MACc
(yield, %) (mg/mL)
179200
(100.0)
323584
(180.6)
10240
(5.7)
3328
(1.9)

2.51
0.57
0.78
0.63

a

Fraction

Protein contents were measured by the Folin-Lowry method (1951).30)
THA, total hemagglutination titer (volume  hemagglutination titer, the
reciprocal of the highest two-fold dilution exhibiting positive hemagglutination).
c
MAC, minimum agglutination concentration, the protein concentration of
the highest dilution exhibiting positive hemagglutination.

b

1

C

150

2 3

(kDa)

0.08
100
0.06

20.1
14.3
6.5

0.04

18.0
9.0

50

0.02

0

5

10

15

Hemagglutination titer

B 0.10
Absorbance at 280 nm

Proteina
(mg)

(NH4)2SO4 (M)

Hemagglutination titer

Absorbance at 280 nm

A

D. PRASEPTIANGGA et al.

Table 3. Hemagglutination-Inhibition Assay of the Salting-Out
Fraction and a Purified Lectin (CBA) from C. barbatum with
Monosaccharides, a Disaccharide, and Glycoproteins
Minimum inhibitory
concentration

0

Fraction

Fig. 1. Purification of the C. barbatum Lectin.
A, Hydrophobic chromatography in a TSKgel Phenyl-5PW
column of the precipitate obtained by salting-out with 75%-saturated
ammonium sulfate. Fractions of 2 mL were collected and measured
for their absorbance at 280 nm (filled circles) and for their
hemagglutination activity (unfilled circles). The active fractions
denoted by a bar were pooled. B, Gel filtration on a PC 300S (N)
column of the active peak obtained by hydrophobic chromatography. Fractions of 2 mL were collected and measured for their
absorbance at 280 nm (filled circles) and for their hemagglutination
activity (unfilled circles). The active fraction denoted by a bar was
recovered as the purified lectin (CBA). C, SDS–PAGE of the
purified lectin from C. barbatum (CBA): lane 1, molecular weight
marker; lane 2, CBA with 2-mercaptoethanol; lane 3, CBA without
2-mercaptoethanol.

Protein-RP column (data not shown). The purified lectin,
named CBA, respectively gave a single protein band of
about 9 kDa and 18 kDa by reducing and non-reducing
SDS–PAGE (Fig. 1C), suggesting that CBA was the SSlinked dimeric protein of a 9-kDa subunit. The relative
molecular mass of intact CBA was estimated to be about
18 kDa by gel filtration in the PC300S (N) column (data
not shown). The yield of CBA was 2.1 mg from 500 g
(wet weight) of the frozen alga. The result of purification
is summarized in Table 2.
Effects of pH, temperature, and divalent cations on
the hemagglutination activity of CBA
The hemagglutination activity of CBA was stable
under a wide range of pH values from 3 to 10 and was
not changed by incubating at 70  C for 30 min, although
it was inactivated when the incubation temperature
exceeded 80  C. The activity was not affected by either
the presence of EDTA or the addition of such divalent
cations as Ca2þ and Mg2þ (data not shown).
Carbohydrate-binding specificity of CBA
The carbohydrate-binding specificity of CBA was
evaluated by a hemagglutination-inhibition assay with
a series of sugars and glycoproteins (Table 3). The
hemagglutination activity of CBA was not inhibited by

Sugar or glycoprotein

Monosaccharides and disaccharidea (mM)
Glycoproteins (g/mL)
N-Glycan
Complex type
Transferrin
Asialo-transferrin
High-mannose type
Yeast mannan
Complex type and high-mannose type
Porcine thyroglobulin (PTG)
Asialo-PTG
N/O-Glycan
Fetuin
Asialo-fetuin
O-Glycan
Porcine stomach mucin
Bovine submaxillary mucin (BSM)
Asialo-BSM

Salting-out
fraction

Purified
lectin
(CBA)

>100

>100

>2000
>2000

>2000
>2000

>2000

>2000

500
250

125
31

>2000
>2000

>2000
>2000

>2000
500
500

>2000
500
500

Values indicate the lowest concentration of sugar (mM) and glycoprotein
(mg/mL) to completely inhibit the hemagglutination activity of a titer, 4.
a
The D-glucose, D-mannose, D-galactose, N-acetyl-D-glucosamine, N-acetylD-galactosamine, L-fucose, N-acetylneuraminic acid, D-xylose, and Lrhamnose as monosaccharides, and lactose as a disaccharide, were
examined.

any of the monosaccharides and one disaccharide
examined. However, it was inhibited by such glycoproteins as PTG, asialo-PTG, BSM and asialo-BSM,
although not by the other glycoproteins examined,
including yeast mannan (Table 3). The strongest inhibition was observed with asialo-PTG. The hapten-inhibition profile of the purified lectin was almost the same as
that of the crude lectin (the salting-out fraction) (Table 3).
Molecular mass of CBA
The molecular mass of intact CBA was determined
to be 18500 Da by ESI-MS (data not shown). This is
comparable to the relative molecular mass (about
18 kDa) estimated by both non-reducing SDS–PAGE
(Fig. 1C) and gel filtration in the PC 300S (N) column.
The molecular masses of nrPE-CBA and rPE-CBA were

Novel Lectin from the Green Alga, Codium barbatum

A

0

Absorbance at 220 nm

Acetonitrile (%)

0.05

0

10

20
30
40
Retention time (min)

50

60

B

0.4

0.2

Acetonitrile (%)

Absorbance at 280 nm

A

B

809

L5

L3
L4
L1

L2

0
0

10

20
30
40
Retention time (min)

50

60

Fig. 2. Reverse-Phase HPLC of rPE-CBA and Its Peptide Fragments
Generated by Enzymatic Digestion.
A, Reverse-phase HPLC of rPE-CBA in a YMC Protein-RP
column. rPE-CBA was purified by reverse-phase HPLC in the YMC
Protein-RP column. The peak denoted by a bar was collected for
further examination. B, Reverse-phase HPLC of the peptide fragments in a YMC Protein-RP column. The peptide fragments
generated by digesting rPE-CBA with Lys-C were subjected to
reverse-phase HPLC on the YMC Protein-RP column. The peaks
were manually collected for further examination.

respectively determined to be 18500 Da and 9992 Da by
ESI-MS. These results indicate that CBA had no free
sulfhydryl groups, the mass of an SS-linked subunit
polypeptide (half size) being 9257 Da with reduced
sulfhydryl groups. The differential mass (735 Da) between rPE-CBA (9992 Da) and a CBA subunit polypeptide (9257 Da) corresponds to the total mass of seven
S-pyridylethyl groups (105 Da  7), indicating that
CBA contained seven cysteine residues per subunit, all
of which were involved in disulfide linkages.
Primary structure of CBA
The amino acid sequence of CBA was principally
determined by cDNA cloning (Fig. 3A). The N-terminal
amino acid sequences of rPE-CBA and its peptide
fragments generated by digestion with Lys-C were first
determined before cDNA cloning. Figure 2 shows the
elution pattern of rPE-CBA and its digested products by
reverse-phase HPLC in the YMC Protein-RP column.
The digested products gave several major peaks for
peptide fragments (L1–L5) in the HPLC data. The
peptide fragments thus obtained, as well as rPE-CBA,
were next sequenced for their N-terminal amino acids by
Edman degradation, except for L1, as shown in Fig. 3B.
The molecular masses of the peptide fragments determined by ESI-MS coincided with the calculated values
from the sequences, confirming the validity of the
sequencing operation (Fig. 3B). Although a potential
N-glycosylation site (Asn-Gly-Ser) was found in the

Fig. 3. Nucleotide and Amino Acid Sequences of CBA.
A, The numbers represent the positions of nucleotides and amino
acids. The underlined regions represent the signal peptide predicted
with SignalP 3.0. The stop codon TGA is shown by an asterisk.
B, The partial amino acid sequences determined by Edman
degradation of rPE-CBA and its peptide fragments (represented by
solid lines) are compared with the complete amino acid sequence
deduced from CBA cDNA. Numbers above the sequences represent
the position of the amino acids. Numeric values below the lines
indicate the molecular masses (Da) of peptides determined by ESIMS, and values in parentheses indicate those calculated from the
sequences. These values are shown together with the S-pyridylethylated ones. The two peptide fragments (represented by dashed lines)
were not analyzed for their N-terminal sequences. The asterisk
represents the residue at position 2 of the CBA subunit, where a
different amino acid was detected between the sequence deduced
from the cDNA and the determined sequence by Edman degradation. The signal peptide region is underlined.

sequence for the L5 fragment, it was not glycosylated
because Asn was identified at the position by Edman
degradation and the determined molecular mass of L5
coincided with the calculated value from its sequence
(Fig. 3B). In addition, the ESI-MS analysis shows that
L1 contained the two peptide fragments of ITHGIK
(667.4 Da) and VSTSYGSSK (914.4 Da), although their
N-terminal sequences were not analyzed.
cDNA cloning for CBA was performed by 30 and
0
5 RACE as described in the Materials and Methods
section to elucidate the complete amino acid sequence of
CBA. Full-length cDNA of the CBA subunit consisted
of 476 bp containing 46 bp of a 50 -untranslated region
(UTR), 112 bp of 30 UTR, and 318 bp of an open reading
frame (ORF) (Fig. 3A). ORF coded a polypeptide of 105
amino acids, including the signal peptide of 17 residues.
The N-terminal amino acid sequences of rPE-CBA and
the peptide fragments, which had been determined by
Edman degradation, were found in the sequence deduced from CBA cDNA (Fig. 3A), except that serine
was deduced from cloning at position 2, whereas
threonine was determined by Edman degradation. The
calculated molecular mass of the CBA subunit (88
amino acids) with serine at position 2 was 9705 Da. The
molecular mass of the CBA subunit substituted with
threonine at position 2 was therefore estimated to be
9719 Da which differs from the mass (9257 Da) determined by the ESI-MS analyses of intact CBA, nrPECBA, and rPE-CBA. This difference indicates that the

810

D. PRASEPTIANGGA et al.

four C-terminal amino acids (-Asn-Met-Asp-Thr,
462 Da) of the subunit polypeptide were post-translationally truncated to afford the mature subunit. The
mature subunit contained seven cysteine residues
(Fig. 3B). It was concluded from these data that CBA
consisted of an SS-linked homodimer of a 9257-Da
polypeptide (84 amino acids). No significantly similar
sequences to the CBA subunit could be found during an
in silico search, indicating CBA to be a novel protein.

Discussion
We isolated in this study a novel lectin from the green
alga, C. barbatum, which we call Codium barbatum
lectin (CBA). The hemagglutination-inhibition profile of
CBA was obviously distinct from those of the Nacetylgalactosamine-specific lectins previously isolated
from other species belonging to this genus,22–24) as the
activity of CBA was not inhibited by N-acetylgalactosamine or N-acetylglucosamine. CBA is therefore the
first example among the Codium lectins having no
affinity for monosaccharides. However, the activity was
inhibited by PTG and remarkably further inhibited by
asialo-PTG (Table 3), suggesting that the branched
sugar moiety of PTG was responsible for this inhibition.
With respect to the glycan structure, PTG is known to
have two types of N-glycans in the molecule: a complex
type and high-mannose type.33–35) Yeast mannan, which
contains only high-mannose type N-glycans, was not
inhibitory, leading to the supposition that CBA recognized the complex type of N-glycans. However, transferrin and fetuin bearing the complex type of N-glycans
were also not inhibitory. This discrepancy may be
derived from a difference in the number of core (1,6)
fucose residues of the complex type of N-glycans
between PTG and transferrin or fetuin. We therefore
speculate that CBA could recognize core (1,6) fucosylated N-glycans that are predominant in PTG, but are
few or absent in transferrin and fetuin.36) Details of the
carbohydrate-recognition mode, including the oligosaccharide-binding specificity, of CBA are now under
investigation in our laboratory.
As the first example among Codium lectins, we have
elucidated the primary structure of CBA by a combination of ESI-MS, Edman degradation, and cDNA cloning.
CBA consisted of an SS-linked homodimer of a 9257Da subunit and contained a total of 14 half-cystines
involved in disulfide linkages. A potential N-glycosylation site (Asn83 -Gly84 -Ser85 ) was found in the sequence
of the CBA subunit, although the presence of the
consensus peptides (Asn-X-Thr/Ser) does not always
lead to glycosylation.37) In addition, the determined
molecular mass of the peptide fragment (L5) containing
the triplet sequence of Asn83 -Gly84 -Ser85 coincided with
the calculated mass from the sequence, suggesting that
CBA had no carbohydrate. Serine was deduced at the
second residue of the CBA subunit from cloning,
whereas threonine was determined by Edman degradation. The presence of a different amino acid at position 2
of the CBA subunit suggests that isoforms of the CBA
subunits exist, or polymorphism between the local
populations of Satsuma-Iwojima Island and Tanegashima
Island, where the algal samples were respectively
collected for lectin purification and RNA extraction

(see the Materials and Methods section). The occurrence
of isolectins has also been reported for the red alga,
Hypnea japonica.15,16)
No significantly similar proteins were found, including lectins, during in silico searches for the amino acid
sequences. The amino acid sequence of CBA is therefore distinct from that of land plant lectins, and also
from the monosaccharide-binding lectins of other green
algae like Enteromorpha prolifera19) and Ulva pertusa21)
that are respectively specific for L-fucose/D-mannose
and N-acetylglucosamine. Although there is no information on the amino acid sequences of Codium lectins,
several lectins from Codium, including those isolated
from C. fragile ssp. tomentosoides,22) C. fragile ssp.
atlanticum,22) C. tomentosum,23) and C. giraffa,24) have
been partially characterized. CBA clearly differed from
known Codium lectins with respect to such biochemical
properties as the molecular mass, amino acid composition, and carbohydrate-binding specificity, indicating
CBA to be a novel lectin. It would be of interest to
elucidate and compare the primary structures of lectins
from other species of Codium, because about 18 species
of the genus have been reported in Japan.38)
The present study has demonstrated the interesting
proposition that the four amino acids in the C-terminal
region of the CBA precursor could be post-translationally truncated. Many lectins contain propeptide regions
at the N- and/or C-termini of their precursors, and the
propeptides are truncated in the mature form.1–4) Some of
the propeptide regions have been reported to be required
for the subcellular localization and inactivation of lectins
in a temporarily inappropriate location.39) For instance,
Galanthus nivalis (snowdrop) lectin (GNA) is synthesized on the endoplasmic reticulum as a preproprotein
which contains a signal peptide and a C-terminal
propeptide,40) and both the pre- and pro-peptides of the
lectin are necessary for trafficking to the vacuole.39)
Wheat germ agglutinin (WGA) is also processed and
sorted to the vacuole in a similar fashion,41,42) although
the two lectins, GNA and WGA, are structurally and
functionally distinct from each other. A lectin (BCA),
which recognized the cluster of non-reducing terminal
(1,2) mannose residues of high-mannose N-glycans, has
recently been isolated from the green alga, Boodlea
coacta, and its precursor also contained a propeptide at
the C-terminus.13) The recombinant of the mature form of
BCA presented hemagglutination activity, whereas the
pro-form did not show this activity (details of these
results will appear elsewhere), suggesting that the
propeptide regions of algal lectin precursors could also
control their hemagglutination activity and, possibly,
their subcellular localization. The sequence of the
propeptide from CBA was quite different from the
known subcellular localization signals of both lectins and
of other proteins.43) Further investigations are required to
elucidate such functions of the propeptide as preparing
the recombinants of mature- and prepro-forms of CBA,
and to examine the hemagglutination activities and
subcellular localizations of these forms in the alga.

Acknowledgments
We are grateful to Dr. Ryuta Terada (Kagoshima
University) for his help in collecting the algal sample

Novel Lectin from the Green Alga, Codium barbatum

and to Dr. Koji Muramoto (Tohoku University) for
providing the silica gel-immobilized phosphorylcholine
(PC 300S (N)) column. We thank Dr. Tomoko Amimoto
in the Natural Science Center for Basic Research and
Development (N-BARD) at Hiroshima University for
measuring the ESI-MS data. We also thank Dr.
Lawrence M. Liao (Hiroshima University) for a critical
reading of the manuscript. This work was partially
supported by the Program for Promotion of Basic and
Applied Researches for Innovations in Bio-Oriented
Industry of Japan.

References
1)
2)
3)
4)
5)
6)

7)

8)
9)

10)
11)
12)
13)
14)
15)
16)

17)

18)

Sharon N and Lis H, ‘‘Lectins’’ 2nd ed., Kluwer Academic
Publisher, Dordrecht, pp. 1–454 (2003).
van Damme EJM, Peumans WJ, Barre A, and Rouge P, Crit.
Rev. Plant Sci., 17, 575–692 (1998).
Lis H and Sharon N, Chem. Rev., 98, 637–674 (1998).
Dodd RG and Drickmer K, Glycobiology, 11, 71–79 (2001).
Ohno Y, Naganuma T, Ogawa T, and Muramoto K, J. Agric.
Food Chem., 54, 548–553 (2006).
Boyd MR, Gustafson KR, Mcmahon JB, Shoemaker RH,
O’Keefe BR, Mori T, Gulakowski RJ, Wu L, Rivera MI,
Laurencot CM, Currens MJ, Cardellina JH II, Buckheit RWJr,
Nara PL, Pannell LK, Sowder RC II, and Henderson LE,
Antimicrob. Agents Chemother., 41, 1521–1530 (1997).
Bokesch HR, O’Keefe BR, McKee TC, Pannell LK, Patterson
ML, Gardella RS, Sowder RC II, Turpin J, Watson K, and
Buckheit RWJr, Biochemistry, 42, 2578–2584 (2003).
Bewley CA, Cai M, Ray S, Ghirlando R, Yamaguchi M, and
Muramoto K, J. Mol. Biol., 339, 901–914 (2004).
Mori T, O’Keefe BR, Sowder RC II, Bringans S, Gardella RS,
Berg S, Cochran P, Turpin JA, Buckheit RWJr, McMahon JB,
and Boyd MR, J. Biol. Chem., 280, 9345–9353 (2005).
Hori K, Sato Y, Ito K, Fujiwara Y, Iwamoto Y, Makino H, and
Kawakubo A, Glycobiology, 17, 479–491 (2007).
Sato Y, Okuyama S, and Hori K, J. Biol. Chem., 282, 11021–
11029 (2007).
Sato Y, Hirayama M, Morimoto K, and Hori K, Biochem.
Biophys. Res. Commun., 405, 291–296 (2011).
Sato Y, Hirayama M, Morimoto K, Yamamoto N, Okuyama S,
and Hori K, J. Biol. Chem., 286, 19446–19458 (2011).
Hung LD, Sato Y, and Hori K, Phytochemistry, 72, 855–861
(2011).
Hori K, Matsubara K, and Miyazawa K, Biochim. Biophys.
Acta, 1474, 226–236 (2000).
Okuyama S, Nakamura-Tsuruta S, Tateno H, Hirabayashi J,
Matsubara K, and Hori K, Biosci. Biotechnol. Biochem., 73,
912–920 (2009).
Fukuda Y, Sugawara T, Ueno M, Fukuta Y, Ochi Y, Akiyama
K, Miyazaki T, Matsuda S, Kawakubo A, and Kato K,
Anticancer Drugs, 17, 943–947 (2006).
Ziolkowska NE and Wlodawer A, Acta Biochim. Pol., 53, 617–
626 (2006).

19)
20)
21)
22)

23)
24)

25)
26)

27)
28)

29)
30)
31)
32)
33)
34)
35)
36)
37)
38)

39)

40)

41)
42)
43)

811

Ambrosio AL, Sanz L, Sanchez EI, Todel CW, and Calvete JJ,
Arch. Biochem. Biophys., 415, 245–250 (2003).
Sampaio AH, Rogers DJ, and Barwell CJ, Bot. Mar., 41, 427–
433 (1998).
Wang S, Zhong FD, Zhang YJ, Wu ZJ, Lin QY, and Xie LH,
Acta Biochim. Biophys. Sin., 36, 111–117 (2004).
Rogers DJ, Loveless EW, and Balding P, ‘‘Lectins: Biology,
Biochemistry, Clinical Biochem.,’’ Walter de Gruyter, Berlin,
pp. 155–160 (1986).
Fabregas J, Mun˜oz A, Llovo J, and Carracedo A, J. Exp. Mar.
Biol. Ecol., 124, 21–30 (1988).
Alvarez-Hernandez S, de Lara-Isassi G, Arreguin-Espinoza R,
Arreguin B, Hernandez-Santoyo A, and Rodriguez-Romero A,
Bot. Mar., 42, 573–580 (1999).
Wu AM, Song SC, Chang SC, Wu JH, Chang KS, and Kabat
EA, Glycobiology, 7, 1061–1066 (1997).
‘‘Seaweed Resources of the World,’’ eds. Critchley AT and
Ohno M, Japan International Cooperation Agency, Yokosuka,
pp. 1–431 (1998).
Hori K, Miyazawa K, Fusetani N, Hashimoto K, and Ito K,
Biochim. Biophys. Acta, 873, 228–236 (1986).
Watanabe Y, Abolhassani M, Tojo Y, Suda Y, Miyazawa K,
Igarashi Y, Sakuma K, Ogawa T, and Muramoto K,
J. Chromatogr. A, 1216, 8563–8566 (2009).
Hori K, Miyazawa K, and Ito K, Bull. Jpn. Soc. Sci. Fish., 52,
323–331 (1986).
Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ, J. Biol.
Chem., 193, 265–275 (1951).
Scha¨gger H and von Jogow G, Anal. Biochem., 166, 368–379
(1987).
Bendtsen JD, Nielsen H, von Heijne G, and Brunak S, J. Mol.
Biol., 340, 783–795 (2004).
Tsuji T, Yamamoto K, Irimura T, and Osawa T, Biochem. J.,
195, 691–699 (1981).
Yamamoto K, Tsuji T, Irimura T, and Osawa T, Biochem. J.,
195, 701–713 (1981).
de Waard P, Koorevaar A, Kamerling J, and Vliegenthart JFG,
J. Biol. Chem., 266, 4237–4243 (1991).
Spik G, Bayard B, Fournet B, Strecker G, Bouquelet S, and
Montreuil J, FEBS Lett., 50, 296–299 (1975).
Gavel Y and von Heijne G, Protein Eng., 3, 433–442 (1990).
Verbruggen H, Leliaert F, Maggs CA, Shimada S, Schils T,
Provan J, Booth D, Murphy S, Clerck O, Littler DS, Littler
MM, and Coppejans E, Mol. Phylogenet. Evol., 44, 240–254
(2007).
Fouquaert E, Hanton SL, Brandizzi F, Peumans WJ, and
van Damme EJM, Plant Cell Physiol., 48, 1010–1021
(2007).
van Damme EJM, Kaku H, Perini F, Goldstein IJ, Peeters B,
Yagi F, Decock B, and Peumans WJ, Eur. J. Biochem., 202, 23–
30 (1991).
Mansfield MA, Peumans WJ, and Raikhel NV, Planta, 173,
482–489 (1988).
Bednarek SY, Wilkins TA, Dombrowski JE, and Raikhel NV,
Plant Cell, 2, 1145–1155 (1990).
Imai K and Nakai K, Proteomics, 10, 3970–3983 (2010).