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

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

chromato-graphic methods. The hemagglutination-inhibition

pro-file 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

9257-Da 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

calcu-lated 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;

carbohydrate-binding 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

inter-actions.

1)

However, the physiological functions of most

lectins remain to be investigated in more detail,

especially for those of plant origin.

1,2)

The

carbohy-drate-binding specificity and molecular structure of

lectins are diverse and generally depend on the

organ-isms 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

con-served.

3,4)

The diversity of lectins in their

carbohydrate-binding specificity enables them to be used as

conven-ient 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

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

specific-ity and molecular structures, including those specific

for high-mannose

N

-glycans

6–14)

and core (

1,6)

fuco-sylated

N

-glycans.

15,16)

They have also shown some

interesting biological activities, including

anticarcino-genic

17)

and antiviral effects.

6–9,11–13)

Their antiviral

activities are especially remarkable because they inhibit

in vitro

infection by the immunodeficiency virus

(HIV-1) and influenza virus with a half-maximal effective

concentration (EC

50

) in the low nanomolar to picomolar

range through their binding with viral envelope

glyco-proteins.

18)

Many algal lectins, especially from red

algae, share some common characteristics of low

mo-lecular weight, monomeric form, thermostability, and

divalent cation-independent hemagglutination, and

hav-ing affinity only for glycoproteins and not for

mono-saccharides. 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. giraffa

24)

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

prefer-ential 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

y _{To whom correspondence should be addressed. Tel/Fax: +81-82-424-7931; E-mail: kanhori@hiroshima-u.ac.jp}

Abbreviations: CBA,Codium barbatumlectin; BLAST, basic local alignment search tool; BSM, bovine submaxillary mucin; GNA,Galanthus

nivalislectin; DTT, dithiothreitol; ESI-MS, electrospray ionization-mass spectrometry; HIV, human immunodeficiency virus; Lys-C,

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.TheC. barbatumgreen alga was collected from

Satsuma-Iwojima Island of Kagoshima in Japan, and was kept at30_{C until}

being used. A portion of the C. barbatum tissue collected from Tanegashima Island of Kagoshima in Japan was also stored at30_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 phosphorylcho-line (PC 300S (N)) column was generously provided by Dr. K. Muramoto (Tohoku University, Japan). D-Glucose, D-galactose, N-acetyl-D-galactosamine, transferrin, fetuin, porcine thyroglobulin (PTG), bovine submaxillary mucin (BSM), and porcine stomach mucin (PSM) were purchased from Sigma-Aldrich Co. (MO, USA). D-Mannose,L-fucose,N-acetyl-D-glucosamine,N-acetylneuraminic acid, lactose, and yeast mannan were purchased from Nacalai Tesque Co. (Kyoto, Japan).D-Xylose andL-rhamnose were purchased from Wako Chemical Co. (Osaka, Japan). The desialylated (asialo-) derivatives of glycoproteins were prepared by hydrolyzing the parent sialoglycopro-tein with 0.05MHCl 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), chymotrypsi-nogen (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 theC. barbatumlectin.A frozen sample (500 g) of

C. barbatumwas 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 mMTris–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}

;500g. The precipitate was dissolved and thoroughly dialyzed against TBS. The internal fraction was further centrifuged for 30 min at10;000g, and the supernatant was recovered as a salting-out fraction. This salting-out fraction was dialyzed against a 20 mMTris–HCl buffer at pH 7.5 (TB) containing 1Mammonium sulfate. Five mL each of the dialyzate was applied to the TSKgel Phenyl-5PW column (7:575_{mm) that had been}

equilibrated with TB containing 1Mammonium 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– 0M) 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 (7:8300mm) that had been equilibrated with a 50 mMphosphate buffer at pH 7.0 containing 0.15M NaCl (PBS), according to the method reported by Watanabeet 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 trypsin-treated rabbit erythrocytes (TRBC).29) _{Serially two-fold dilutions}

(25mL each) of a test solution were first prepared in saline, and to each well, 25mL 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

hemag-glutination activity. The effects of pH, temperature, and divalent

cations on the hemagglutination activity were next examined. To determine the effect of pH, 500mL 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, 500mL each of a test solution was incubated for 30 min at various temperatures from 30_{C to 100C and then subjected to the hemagglutination assay}

after cooling. To determine the effect of divalent cations, 500mL of a test solution was dialyzed overnight at 4_{C against 50 mMEDTA in}

PBS, and the internal fraction was measured for its hemagglutination activity. Additionally, to this internal fraction, an equal volume of 20 mMCaCl2or 20 mMMgCl2in 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 (25mL 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 hemaggluti-nation titer of 4 was added, and the plate was gently shaken and allowed to stand at room temperature for 1 h. Finally, 25mL 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 macro-scopically 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, L-fucose, 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)}

2-mercaptoethanol. The gel was stained with Commasie brilliant blue R-250 after electrophoresis.

Preparation ofS-pyridylethylated (PE-) CBA. S-Pyridylethylation

lectin solution, and the mixture incubated for another 2 h. The resulting PE-CBA, reduced andS-pyridylethylated (rPE)-CBA or non-reduced

and S-pyridylethylated (nrPE)-CBA was purified by reverse-phase

HPLC in the YMC Protein-RP column (6:0250_{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

(100mg) 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 mMTB (pH 8.5) containing 4Murea. The peptide fragments were separated by reverse-phase HPLC in the YMC Protein-RP column (6:0250_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 electro-spray 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

N-terminal 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 5}0_{cDNA ends (3}0_{and 5}0_{RACE) of}

CBA.Total RNA ofC. barbatumwas extracted from the RNAlater-treated algal tissues by using the Plant RNA Isolation reagent (Invitrogen). Full-length cDNAs were synthesized from 5mg 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 0.2mL each of}

a 10-fold dilution of the synthesized cDNA as a template to 8 aliquots of a 9.8-mL solution containing 1mL of a 10Blend 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 rPE-CBA (Table 1)), and 0.25 units of Blend Taq DNA polymerase (Toyobo). The reaction was performed with a T Gradient Thermo-cycler (Biometra, Go¨ttingen, Germany) under the following condi-tions: 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 64C (2C increments) for 30 s and extension at 72C 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 0.2mL 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 intoEscherichia coliDH5competent 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 3}0_{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 64C (2C}

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 10A-attachment mix (Toyobo), before subcloning into the pGEM-T Easy vector (Promega) and transforming intoE. coliDH5competent cells. Plasmid purification and DNA sequencing were performed as already described.

To verify the accuracy of the CBA sequences obtained by the 50_and

30_{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 5}0_{and 3}0

terminal sequences of CBA cDNA obtained by 50 _{and 3}0_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 position-specific 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

Phenyl-5PW 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

Table 1. Primer Sequences Used in the cDNA Cloning of CBA

Primer Sequence (from 50_{to 3}0_{) Position}

CBA d F1 ACICAYGGIATHAARAAYGAb

101–120

CBA d F2 AAYGAYTGYGGIGTICCIGTb _116–135

CBA R1 TCCAAGCAGCATACGAACAC 198–217

CBA R2 TCATCAGTCCCAGTCCAACA 124–143

CBA 50_{end F AATATAAATTAAACGTAAGGTATCGC 1–26}

CBA 30_{end R AGTTCTGCATATGGATGTAC 457–476}

GeneRacer 30_primera

GCTGTCAACGATACGCTACGTAACG —

GeneRacer 30_{nested primer}a

CGCTACGTAACGGCATGACAGTG —

GeneRacer 50_primera

CGACTGGAGCACGAGGACACTGA —

GeneRacer50_{nested primer}a

GGACACTGACATGGACTGAAGGAGTA —

a

These primers were inferred from the GeneRacer Kit (Invitrogen).

b

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

SS-linked 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 Ca

2þ

and Mg

2_þ

(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

any of the monosaccharides and one disaccharide

examined. However, it was inhibited by such

glycopro-teins as PTG, asialo-PTG, BSM and asialo-BSM,

although not by the other glycoproteins examined,

including yeast mannan (Table 3). The strongest

inhib-ition was observed with asialo-PTG. The

hapten-inhib-ition 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

20.1 14.3 6.5

18.0 9.0

C 1 2 3 (kDa)

1.0

0.10

Absorbance at 280 nm Hemagglutination titer

Hemagglutination titer

Fraction

B

A

0 5 10 15 20 25

0 5 10 15

0.08

0.06

0.04

0.02

0 1

100 150

0 50

Absorbance at 280 nm

Fraction

(NH

4

)2

SO

4

(

M

[image:4.595.58.280.64.329.2]

)

Fig. 1. Purification of theC. barbatumLectin.

[image:4.595.307.537.75.187.2]

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 chromatogra-phy. 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 fromC. barbatum(CBA): lane 1, molecular weight marker; lane 2, CBA with 2-mercaptoethanol; lane 3, CBA without 2-mercaptoethanol.

Table 2. Purification Procedure for the Lectin fromC. barbatum

Purification procedure Protein

a _THAb _MACc

(mg) (yield, %) (mg/mL)

Extraction 450.4 179200 2.51

(100.0) Ammonium sulfate-precipitation 184.6 323584 0.57

(75%-(NH4)2SO4) (180.6) Hydrophobic chromatography 8.0 10240 0.78

(5.7)

Gel filtration 2.1 3328 0.63

(1.9)

a

Protein contents were measured by the Folin-Lowry method (1951).30)

b_{THA, total hemagglutination titer (volumehemagglutination titer, the}

reciprocal of the highest two-fold dilution exhibiting positive hemaggluti-nation).

c_{MAC, minimum agglutination concentration, the protein concentration of}

the highest dilution exhibiting positive hemagglutination.

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

Sugar or glycoprotein

Minimum inhibitory concentration

Salting-out Purified fraction lectin

(CBA)

Monosaccharides and disaccharidea

(mM) >100 >100

Glycoproteins (g/mL) N-Glycan

Complex type

Transferrin >2000 >2000

Asialo-transferrin >2000 >2000

High-mannose type

Yeast mannan >2000 >2000

Complex type and high-mannose type

Porcine thyroglobulin (PTG) 500 125

Asialo-PTG 250 31

N/O-Glycan

Fetuin >2000 >2000

Asialo-fetuin >2000 >2000

O-Glycan

Porcine stomach mucin >2000 >2000 Bovine submaxillary mucin (BSM) 500 500

Asialo-BSM 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

TheD-glucose,D-mannose,D-galactose,N-acetyl-D-glucosamine,N

[image:4.595.310.535.296.541.2]

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)

be-tween rPE-CBA (9992 Da) and a CBA subunit

polypep-tide (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

deter-mined 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

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 3

0

and

5

0

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 5

0

-untranslated region

(UTR), 112 bp of 3

0

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

de-duced 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)

deter-mined by the ESI-MS analyses of intact CBA,

nrPE-CBA, and rPE-CBA. This difference indicates that the

A

B

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 ESI-MS, and values in parentheses indicate those calculated from the sequences. These values are shown together with the S-pyridylethy-lated 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 degrada-tion. The signal peptide region is underlined.

Absorbance at 280 nm

Absorbance at 220 nm

B

A

0 10 20 30 40 50 60

0 10 20 30 40 50 60

Acetonitrile (%)

Retention time (min)

L5

L4 L3

L2 L1 0.05

0

0.2

0 0.4

Retention time (min)

Acetonitrile (%)

Fig. 2. Reverse-Phase HPLC of rPE-CBA and Its Peptide Fragments Generated by Enzymatic Digestion.

four

C-terminal

amino

acids

(-Asn-Met-Asp-Thr,

462 Da) of the subunit polypeptide were

post-transla-tionally 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

N

-acetylgalactosamine-specific lectins previously isolated

from other species belonging to this genus,

22–24)

as the

activity of CBA was not inhibited by

N

-acetylgalactos-amine 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

recog-nized the complex type of

N

-glycans. However,

trans-ferrin 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)

fucosy-lated

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

oligosac-charide-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

combina-tion of ESI-MS, Edman degradacombina-tion, and cDNA cloning.

CBA consisted of an SS-linked homodimer of a

9257-Da subunit and contained a total of 14 half-cystines

involved in disulfide linkages. A potential

N

-glycosyla-tion site (Asn

83

-Gly

84

-Ser

85

) 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 Asn

83

-Gly

84

-Ser

85

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

degrada-tion. 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,

includ-ing lectins, durinclud-ing

in silico

searches for the amino acid

sequences. The amino acid sequence of CBA is

there-fore distinct from that of land plant lectins, and also

from the monosaccharide-binding lectins of other green

algae like

Enteromorpha prolifera

19)

and

Ulva pertusa

21)

that are respectively specific for

L

-fucose/

D

-mannose

and

N

-acetylglucosamine. Although there is no

infor-mation 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

composi-tion, 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-translation-ally 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

synthe-sized 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

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) Sharon N and Lis H, ‘‘Lectins’’ 2nd ed., Kluwer Academic Publisher, Dordrecht, pp. 1–454 (2003).

2) van Damme EJM, Peumans WJ, Barre A, and Rouge P,Crit.

Rev. Plant Sci.,17, 575–692 (1998).

3) Lis H and Sharon N,Chem. Rev.,98, 637–674 (1998). 4) Dodd RG and Drickmer K,Glycobiology,11, 71–79 (2001). 5) Ohno Y, Naganuma T, Ogawa T, and Muramoto K,J. Agric.

Food Chem.,54, 548–553 (2006).

6) 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).

7) 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). 8) Bewley CA, Cai M, Ray S, Ghirlando R, Yamaguchi M, and

Muramoto K,J. Mol. Biol.,339, 901–914 (2004).

9) 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). 10) Hori K, Sato Y, Ito K, Fujiwara Y, Iwamoto Y, Makino H, and

Kawakubo A,Glycobiology,17, 479–491 (2007).

11) Sato Y, Okuyama S, and Hori K,J. Biol. Chem.,282, 11021– 11029 (2007).

12) Sato Y, Hirayama M, Morimoto K, and Hori K, Biochem.

Biophys. Res. Commun.,405, 291–296 (2011).

13) Sato Y, Hirayama M, Morimoto K, Yamamoto N, Okuyama S, and Hori K,J. Biol. Chem.,286, 19446–19458 (2011). 14) Hung LD, Sato Y, and Hori K,Phytochemistry,72, 855–861

(2011).

15) Hori K, Matsubara K, and Miyazawa K, Biochim. Biophys.

Acta,1474, 226–236 (2000).

16) Okuyama S, Nakamura-Tsuruta S, Tateno H, Hirabayashi J, Matsubara K, and Hori K,Biosci. Biotechnol. Biochem., 73, 912–920 (2009).

17) 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).

18) Ziolkowska NE and Wlodawer A,Acta Biochim. Pol.,53, 617– 626 (2006).

19) Ambrosio AL, Sanz L, Sanchez EI, Todel CW, and Calvete JJ,

Arch. Biochem. Biophys.,415, 245–250 (2003).

20) Sampaio AH, Rogers DJ, and Barwell CJ,Bot. Mar.,41, 427– 433 (1998).

21) Wang S, Zhong FD, Zhang YJ, Wu ZJ, Lin QY, and Xie LH,

Acta Biochim. Biophys. Sin.,36, 111–117 (2004).

22) Rogers DJ, Loveless EW, and Balding P, ‘‘Lectins: Biology, Biochemistry, Clinical Biochem.,’’ Walter de Gruyter, Berlin, pp. 155–160 (1986).

23) Fabregas J, Mun˜oz A, Llovo J, and Carracedo A,J. Exp. Mar.

Biol. Ecol.,124, 21–30 (1988).

24) 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).

25) Wu AM, Song SC, Chang SC, Wu JH, Chang KS, and Kabat

EA,Glycobiology,7, 1061–1066 (1997).

26) ‘‘Seaweed Resources of the World,’’ eds. Critchley AT and Ohno M, Japan International Cooperation Agency, Yokosuka, pp. 1–431 (1998).

27) Hori K, Miyazawa K, Fusetani N, Hashimoto K, and Ito K,

Biochim. Biophys. Acta,873, 228–236 (1986).

28) 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).

29) Hori K, Miyazawa K, and Ito K,Bull. Jpn. Soc. Sci. Fish.,52, 323–331 (1986).

30) Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ,J. Biol.

Chem.,193, 265–275 (1951).

31) Scha¨gger H and von Jogow G,Anal. Biochem.,166, 368–379 (1987).

32) Bendtsen JD, Nielsen H, von Heijne G, and Brunak S,J. Mol.

Biol.,340, 783–795 (2004).

33) Tsuji T, Yamamoto K, Irimura T, and Osawa T,Biochem. J.,

195, 691–699 (1981).

34) Yamamoto K, Tsuji T, Irimura T, and Osawa T,Biochem. J.,

195, 701–713 (1981).

35) de Waard P, Koorevaar A, Kamerling J, and Vliegenthart JFG,

J. Biol. Chem.,266, 4237–4243 (1991).

36) Spik G, Bayard B, Fournet B, Strecker G, Bouquelet S, and Montreuil J,FEBS Lett.,50, 296–299 (1975).

37) Gavel Y and von Heijne G,Protein Eng.,3, 433–442 (1990). 38) 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).

39) Fouquaert E, Hanton SL, Brandizzi F, Peumans WJ, and van Damme EJM, Plant Cell Physiol., 48, 1010–1021 (2007).

40) 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).

41) Mansfield MA, Peumans WJ, and Raikhel NV,Planta, 173, 482–489 (1988).

42) Bednarek SY, Wilkins TA, Dombrowski JE, and Raikhel NV,

Plant Cell,2, 1145–1155 (1990).