Comparative Biochemistry and Physiology Part B 119 (1998) 593 – 597

Purification and characterization of Microcystis aeruginosa (freshwater

cyanobacterium) lectin

Masato Yamaguchi a, Mituru Jimbo a, Ryuichi Sakai a, Koji Muramoto b, Hisao Kamiya a,*
a
Department of Marine Biochemistry, School of Fisheries Sciences, Kitasato Uni6ersity, Sanriku, Iwate 022 -0101, Japan
b
Department of Biological Resource Science, Graduate School of Agriculture, Tohoku Uni6ersity, Sendai, Miyagi 981 -8555, Japan

Received 3 July 1997; received in revised form 21 November 1997; accepted 22 December 1997

Abstract

Microcystis aeruginosa, strain M228, a laboratory culture of freshwater cyanobacterium, showed hemagglutinating activity
against rabbit, horse and human ABO erthrocytes. Crossed absorption tests revealed the presence of a single type of lectin in the
extract of M228 strain cells. The lectin, termed MAL, was purified in combination with the affinity chromatography on
acid-treated agarose gel and the gel permeation chromatography in an electrophoretically pure form. MAL was a glycoprotein
containing 7.8% neutral sugars and was composed of a single polypeptide having a molecular weight of 57 kDa. Isoelectric point
was estimated to be pH 6.4. Hemagglutinating activity of the lectin was inhibited effectively by N-acetyl-D-galactosamine and by
glycoproteins. D-galactose and lactose also showed moderate inhibitory activity. The destruction of the hemagglutinating activity
by a 2-mercaptoethanol treatment suggests the presence of intra-chain disulfide bond(s) essential for the activity in the molecule.
The sequence of the amino-terminal region of MAL was determined as Val-Lys-Ala-Ser-Lys-Val-Ser-Thr-Ser-Gln-Ala-Gly-Ser-
Lys-Glu-Lys-Lys-Ala. © 1998 Elsevier Science Inc. All rights reserved.

Keywords: Agglutinin; Blue-green algae; Cyanobacteria; Hemagglutinin; Lectin; Microcystis aeruginosa

1. Introduction the other species which has been examined for the
hemagglutinating activity [3]. The hemagglutinating
Lectins or carbohydrate-binding proteins are proteins factor in M. aeruginosa possessed proteinous nature
of non-immune origin which agglutinate various types and the activity was common in the extract from both
of cells or precipitate glycoconjugates [7]. Lectins have natural blooms and laboratory cultures [22]. The
been identified in various types of cells and tissues of physicochemical properties of the responsible lectins,
almost all taxa from virus to higher vertebrates [6]. A however, is remained unknown. Toward the under-
wide distribution of lectin molecules suggests some standings of biological roles of lectins in cyanobacteria,
association of lectins in the physiological processes at we have recently examined the hemagglutinating activ-
these sites [20]. In cyanobacteria or blue-green algae, ity of M. aeruginosa M228 strain and found that the
the occurrence of lectins has been reported for only few lectin in M228 cells was calcium ion-dependent and was
species to date. Lyngbya majuscula, a marine cyanobac- inhibitable by N-acetyl-D-galactosamine, lactose, and
terium, is the first species reported for its hemaggluti- also various glycoproteins [19]. In the present paper,
nating activity [1]. Microcystis aeruginosa, a freshwater isolation and detailed properties of the lectin purified
bloom-forming species in eutrophic lakes and ponds, is from cultured specimens of M. aeruginosa, strain M228
are described. This paper is the first description for the
* Corresponding author. Tel.: +81 192 442121; fax: + 81 192 purification and the properties of the lectin isolated
443932; e-mail: hisao-ka@nnettown.ne.jp from the cyanobacteria.

0305-0491/98/$19.00 © 1998 Elsevier Science Inc. All rights reserved.

PII S0305-0491(98)00033-9
594 M. Yamaguchi et al. / Comparati6e Biochemistry and Physiology, Part B 119 (1998) 593–597

2. Materials and methods 2.3. Crossed absorption tests

2.1. Unialgal culture of M. aeruginosa M228 strain Crossed absorption test was carried out using the
extract prepared from M228 cells as follows. The ex-
M. aeruginosa M228 strain was obtained from tract was reacted with an equal volume of packed horse
Tokyo Metropolitan Research Laboratory. It was or rabbit erythrocytes and kept for 16 h at 4°C. After
cultured in a glass bottle containing MA medium centrifugation, the supernatant was tested against the
[10] with aeration (120 ml min − 1) for 14 – 20 days. same erythrocyte type used for absorption and against
The temperature and light intensity was adjusted at the other erythrocyte type. The hemagglutinating activ-
23°C and 45 mE m − 2 per s with photoperiod of ity of supernatant was also examined against human
15L:9D for whole culture period. The cultured cells ABO erythrocytes.
were concentrated by a centrifugation at 5000×g for
15 min, lyophilized and extracted twice under sonica- 2.4. Purification of the lectins from M. aeruginosa
tion in 10 volumes of 0.15 M NaCl containing 1 M228 Cells
mM phenylmethylsulfonyl fluoride (PMSF) for 5
min. After centrifugation at 12000×g for 10 min at To the extract of M228 cells solid ammonium sulfate
4°C, the supernatant was kept at −20°C until use. was added to a final concentration of 20% saturation.
The precipitate formed was removed by centrifugation
2.2. Agglutination test at 12000×g for 20 min at 4°C. Additional amount of
ammonium sulfate was added to the supernatant to be
Serial two-fold dilutions of the test solution (50 m l) a final concentration of 45% saturation. The precipitate
were made in multiwell microtiter plates using 0.15 obtained was dissolved into small volume of 20 mM
M NaCl containing 10 mM CaCl2 as the diluent. Tris–HCl buffer (pH 7.5) containing 0.15 M NaCl, 10
Agglutinating activity against horse, rabbit and hu- mM CaCl2, and 1 mM PMSF (THB) and dialyzed
man ABO erythrocytes was assayed using a 4% ery- overnight against THB. The retentate was applied to a
column of Sepharose 4B (2.2×20 cm, Pharmacia Bio-
throcyte suspension (50 ml). After allowing the
tech., Tokyo) which had been acid treated [5], and
microtiter plates to stand at room temperature for 1
subsequently equilibrated with THB. After washing the
h, the agglutination titer of the maximum dilution
column with THB, the adsorbed lectin was eluted with
giving positive agglutination was recorded. The effect
0.2 M D-galactose in THB. Each 4 ml of fraction was
of treatment with a reducing agent on the hemagglu-
collected, and the absorption at 280 nm and hemagglu-
tinating activity was examined by incubating partially
tinating activity against rabbit erythrocytes were mea-
purified lectin with 2-mercaptoethanol at a final con-
sured. The active fractions were further purified by
centration of 5% for 1 h at 37°C. A remaining
HPLC on a column of TSK G3000SW (7.5 × 600 mm,
hemagglutinating activity was estimated using rabbit Tosoh, Tokyo) at 0.5 ml min − 1 using 20 mM Tris–ac-
erythrocytes suspension. etate buffer (pH 7.0) containing 10 mM CaCl2. Each
An effect of sugars and glycoproteins on hemag- eluate was monitored by absorption at 280 nm and
glutinating activity was examined as follows. The hemagglutinating activity against rabbit erythrocytes.
purified lectin sample (25 m l) was allowed to react The peak which showed the hemagglutinating activity
with various concentrations of inhibitors (25 m l) for was purified on the same conditions to give pure lectin
1 h at 37°C. A total of 50 m l of 4% rabbit erythro- (280 mg).
cyte suspension was then added to the mixture and Anti-purified lectin sera was raised in a male rabbit
the agglutination was measured after standing for 1 using Freund’s incomplete adjuvant (Wako Chemicals,
h at 37°C. The results were expressed by the mini- Tokyo). IgG was isolated from serum by affinity chro-
mum concentration of the inhibitors that completely matography on a HiTrap Protein A column (1 ml,
inhibit the hemagglutination. The following sugars Pharmacia Biotech., Tokyo) according to the manufac-
(200 mM) and glycoproteins (0.25%) were used: D-ri- turer’s protocol. Neutralization of the hemagglutinating
bose, deoxy-D-ribose, D-glucose, D-galactose, L-ara- activity in the extract of M228 cells by anti-lectin
binose, L-rhamnose, L-fucose, D-xylose, D-mannose, antibody was examined using 4% rabbit or horse ery-
D-glucosamine, D-galactosamine, D-mannosamine, N- throcytes suspension.
acetyl-D-glucosamine, N-acetyl-D-galactosamine, N-
acetylneuraminic acid, D-glucuronic acid, lactose, 2.5. Analytical method
sucrose, maltose, mucin Type I from bovine submax-
illary gland (BSM), mucin Type II from porcine Protein concentrations were quantitated [2] using
stomach (PSM), fetuin, asialoBSM, asialoPSM and bovine serum albumin as a standard. The molecular
asialofetuin. weight of the intact lectin was determined by HPLC on
 M. Yamaguchi et al. / Comparati6e Biochemistry and Physiology, Part B 119 (1998) 593–597 595

a TSK G3000SW using 0.2 M phosphate buffer con-

taining 10% ethylene glycol (pH 7.4) to remove the
interaction of the lectin with gel matrix. Sodium dode-
cyl sulfate-polyacrylamide gel electrophoresis (SDS-
PAGE) was performed in a 12% polyacrylamide gel.
Electrofocusing on a 5% polyacrylamide gel (5× 50
mm) was carried out at 200 V for 5 h. Ampholine
3.5/10 (Pharmacia Biotech., Tokyo) was used as carrier
ampholite at a concentration of 2%. The gel was
stained with 0.1% Coomassie brilliant blue R-250.
Amino acid analysis was carried out using the
purified lectin by our previous methods [14,15]. Neutral
sugar content was determined by phenol-sulfuric acid
method using D-glucose as a standard [4].
The amino-terminal sequence of the purified lectin
was analyzed using a Shimadzu gas-phase PSQ-1 Fig. 1. Affinity chromatography of MAL on acid-treated Sepharose
sequencer. 4B. The sample was applied to a column of HCl-treated Sepharose
4B (2.2×20 cm) previously equilibrated with 20 mM Tris–HCl
buffer (pH 7.5) containing 10 mM CaCl2, 0.15 M NaCl and 1 mM
PMSF (THB). After washing the column with the same buffer the
3. Results adsorbed lectin was eluted with 0.2 M D-galactose in THB. Fractions
of 4 ml were collected and measured for the absorption at 280 nm
3.1. Properties of M. aeruginosa M228 lectin ( – ) and the hemagglutinating activity against rabbit erythrocytes
(“ – “). The active fractions (indicated by a horizontal bar) were
combined.
The extract of M228 cells reacted with rabbit (titer
128) and horse erythrocytes (titer 64) as well as human
ABO erythrocytes (titer 16 – 64). In the crossed absorp- cyte was inhibited both by simple sugars and
tion tests with M228 extract, a packed suspension of glycoproteins. As indicated (Table 1), N-acetyl-D-galac-
rabbit erythrocytes absorbed the hemagglutinating ac- tosamine was the most effective inhibitor among simple
tivity against rabbit, horse and human ABO erythro- sugars examined. Lactose and D-galactose also inhib-
cytes. Contrary, horse erythrocytes absorbed the ited the hemagglutinating activity. Among glyco-
hemagglutinating activity against rabbit, horse and hu- proteins examined, asialoBSM, PSM, asialoPSM and
man ABO erythrocytes. The hemagglutinating activity asialofetuin showed strong inhibitory activity in this
was demolished after the treatment with 2- order.
mercaptoethanol. MAL migrated as a single band corresponding to 57
kDa in SDS-PAGE, irrespective of a treatment with
3.2. Purification and properties of M. aeruginosa M228 2-mercaptoethanol (Fig. 2). In gel permeation chro-
Lectin matography on a TSK G-3000SW column using Tris–
acetate buffer, MAL appeared in the fraction where the
The hemagglutinating activity was recovered effec- molecular weight was calculated to be 13 kDa from the
tively in the precipitate formed by the addition of elution time of the lectin. However, the molecular
ammonium sulfate to 45% saturation. In the affinity weight of MAL was estimated to be 72 kDa when
chromatography on acid-treated Sepharose 4B, the
lectin was eluted as a single peak from the column with Table 1
D-galactose (Fig. 1). No hemagglutinating activity was The minimum inhibitory concentration of saccharides to MAL*
detected in the unadsorbed fractions. Succeeding HPLC
Saccharides Minimum inhibitory
on a TSK G3000SW gave one active protein peak
concentration
associated with a few small peaks of contaminants.
Repeated HPLC on the same column gave a protein in N-acetyl-D-galactosamine 0.80
an electrophoretically pure form. The purified lectin, (mM)
D-galactose (mM) 25.0
termed MAL, agglutinated rabbit erythrocytes as low
lactose 12.5
as 0.73 mg ml − 1. In a typical run, 1.6 g of dried M.
aeruginosa cells (1.1 g of protein) gave 280 mg of MAL. PSM (%) 0.002
asialoPSM (%) 0.002
The recovery in terms of the hemagglutinating activity
asialoBSM (%) 0.002
was 20%. asialofetuin (%) 0.063
Sugar inhibition tests revealed that the hemaggluti-
nating activity of the purified lectin on rabbit erythro- * The purified lectin diluted to a titre of 64 were used for tests.
596 M. Yamaguchi et al. / Comparati6e Biochemistry and Physiology, Part B 119 (1998) 593–597

4. Discussion

MAL agglutinated rabbit, horse and human ABO

erythrocytes. In crossed absorption tests with M228
extract, rabbit erythrocytes absorbed the hemaggluti-
nating activity against horse and rabbit erythrocytes.
Similarly, horse erythrocytes absorbed the activity
against rabbit and horse erythrocytes. The agglutinat-
ing activity against human ABO erythrocytes was lost
after the treatment of the extract with a packed rabbit
or horse erythrocytes. These results suggested that the
hemagglutinating activity of M. aeruginosa M228 cells
was attributed to a single type of lectin. This is also
supported by the result that the antibody of MAL
Fig. 2. SDS-Polyacrylamide gel electrophoresis of MAL. SDS-PAGE
neutralized the hemagglutinating activity of M228 ex-
was carried out on 12% gel with and without 2-mercaptoethanol. 1. tract against both rabbit and horse erythrocyte.
Standard proteins; phosphorlylase b (94 kDa), bovine serum albumin The hemagglutinating activity of MAL was inhibited
(67 kDa), ovalbumin (43 kDa), carbonic anhydrase (30 kDa), soy- effectively by glycoproteins. N-acetyl-D-galactosamine
bean trypsin inhibitor (20.1 kDa), a-lactoalbumin (14.4 kDa); 2. was the most effective inhibitor among the simple sug-
MAL with 2-mercaptoethanol; 3. MAL without 2-mercaptoethanol.
ars tested. D-galactose and lactose also inhibited the
hemagglutinating activity but required much higher
concentrations than N-acetyl-D-galactosamine.
MAL was purified effectively in combination with
determined on the same column using the phosphate affinity chromatography on acid-treated Sepharose 4B
buffer containing 10% ethylene glycol. The isoelectric and HPLC on TSK G3000SW. The addition of PMSF
point was determined to be pH 6.4. Neutral sugar was practical to avoid the partial hydrolysis of MAL
content determined by phenol-sulfuric acid method was during the separation. The molecular weight of the
7.8%. Mannose was detected as the major constituent subunit of MAL was determined to be 57 kDa from the
of the neutral sugar component of MAL. Unidentified mobility of the lectin in SDS-PAGE with or without a
sugar component was, however, also detected. In amino reducing agent. In analytical gel permeation chro-
acid composition analysis, high content of Asx (15.1 matography on TSK G3000SW in the presence of 10%
mol.%) and Arg (9.2 mol.%) was observed as shown in ethylene glycol, the apparent molecular weight of intact
Table 2. It also contained Cys (3.5 mol.%). No Met was MAL was estimated to be 72 kDa. The ratio of the
detected in the lectin molecule. The sequence of the subunit and the apparent molecular weight of the intact
amino-terminal region was determined as Val-Lys-Ala- MAL was 1.26. These results suggest that the MAL
Ser-Lys-Val-Ser-Thr-Ser-Gln-Ala-Gly-Ser-Lys-Glu- molecule is composed of a single peptide. At this point
Lys-Lys-Ala. MAL is unique, since most of the lectin for N-acetyl-D-
Anti-MAL antibody neutralized the hemagglutinat- galactosamine or D-galactose found in various plants
ing activity against rabbit and horse erythrocyte in the and animals are composed of multiple subunits
extract of M228 cells. [8,9,12,13,18]. The difference in the molecular weight of
MAL observed for gel permeation chromatography and
the subunit determined from SDS-PAGE could be be-
cause of the interaction of the lectin with gel matrix. It
Table 2
is also noteworthy that MAL is a glycoprotein contain-
Amino acid composition of MAL (mol.%) ing 7.8% neutral sugar in spite of most of the lectins
originated from prokaryotic organisms are simple
Amino acid Mol.% Amino acid Mol.% proteins or lowly glycosylated [11,21]. A neutral sugar
composition of MAL, however, is remains to be solved.
Asx 15.1 Met 0.0
Glx 5.4 Ile 7.3
Some of the lectins specific for N-acetyl-D-galac-
Ser 1.0 Leu 9.6 tosamine or D-galactose such as the soybean lectin [16]
Thr 6.6 Trp 0.4 and peanut lectin [17] are used as valuable tools be-
Gly 9.3 Phe 3.4 cause of their mitogenic activity. MAL, however, did
Ala 9.1 Lys 2.5
not show any mitogenic activity when checked with
Pro 9.0 His 0.2
Val 7.5 Tyr 0.9
mouse spleen lymphocytes (data not shown). The treat-
Arg 9.2 Cys (half) 3.5 ment of MAL with 2-mercaptoethanol destroyed the
hemagglutinating activity completely. This observation
 M. Yamaguchi et al. / Comparati6e Biochemistry and Physiology, Part B 119 (1998) 593–597 597

together with the occurrence of Cys (half) in MAL tinin from the snail Helix pomatia. Ann NY Acad Sci
1974;234:183 – 97.
suggests the presence of intra-molecular disulfide
[9] Hatakeyama T, Kohzaki H, Nagatomi H, Yamasaki N. Purifica-
bond(s) in the molecule essential for the hemagglutinat- tion and characterization of four Ca2 + -dependent lectins from
ing activity. the marine invertebrate, Cucccumaria echinata. J Biochem
Also determined is the amino acid sequence of the 1994;116:209 – 14.
[10] Ichimura T. Isolation and cultivation of microalgae. In:
amino-terminal region of MAL, but any homology in
Nishizawa K, Chihara M, editors. Methods in Algalogical Stud-
the sequence from other plant and animal lectins has ies. Tokyo: Kyoouritsu Shuppan, 1979:294 – 305.
not been detected so far. [11] Joshi B, Khire JM, SivaRaman H, Khan MI. Purification and
characterization of a lectin from Xanthomonas campestris
NCIM5028. Can J Microbiol. 1996;42:609 – 12.
[12] Lotan R, Siegelman HW, Lis H, Sharon N. Subunit structure of
Acknowledgements soybean agglutinin. J Biol Chem 1974;249:1219– 24.
[13] Lotan R, Skutelsky E, Danon D, Sharon N. The purification,
The authors are indebted to Dr M.F. Watanabe, composition, and specificity of the anti-T lectin from peanut
Tokyo Metropolitan Research Laboratory for M. (Arachis hypogaea). J Biol Chem 1974;250:8518 – 23.
[14] Muramoto K, Kamiya H. Recovery of tryptophan in peptides
aeruginosa M228 strain used in this study.
and proteins by high-temperature and short-term acid hydrolysis
in the phenol. Anal Biochem 1990;89:223 – 30.
[15] Muramoto K, Sunahara S, Kamiya H. Measurement of tryp-
References tophan in peptides by acid hydrolysis in the presence of phenol
and its application to the amino acid sequence of a sea anemone
[1] Boyd WC, Almodovar LA, Boyd LG. Agglutinins in marine toxin. Agric Biol Chem 1987;51:1607 – 16.
algae for human erythrocytes. Transfusion (Phil) 1966;6:82 – 3. [16] Novogrodsky A, Katchalski E. Transformation of neu-
[2] Bradford M. A rapid method for the quantitation of microgram raminidase — treated lymphocytes by soybean agglutinin. Proc
Nat Acad Sci USA 1973;70:2515 – 8.
quantities of protein utilizing the principle of protein-dye bind-
[17] Novogrodsky A, Lotan R, Ravid A, Sharon N. Peanut agglu-
ing. Anal Biochem 1976;72:248–54.
tinin, a new mitogen that binds to galactosyl sites exposed after
[3] Carmichael WW, Bent PE. Hemagglutination method for detec-
neuraminidase treatment. J Immunol 1975;115:1243– 8.
tion of freshwater Cyanobacteria (blue-green algae) toxins. Appl
[18] Rogers DJ, Hori K. Marine algal lectins: New development
Environ Microbiol 1981;41:1383–8.
Hydrobiologica 1993;260/261:589– 93.
[4] Dubois M, Gills KA, Hamilton JK, Reberts PA, Smith F.
[19] Sakamoto S, Yamaguchi M, Watanabe MF, Watanabe M,
Colorimetric method for determination of sugars and related Kamiya H. Distribution and characterization of lectins from
substances. Anal Chem 1956;28:350–6. natural and cultured Microcystis spp. In: Yasumoto T, Oshima
[5] Ersson B, Aspberg K, Porath J. The phytohemagglutinin from Y, Fukuyo Y, editors. Harmful and Toxic Algal Blooms. Inter-
sunn hemp seeds (Crotalaria juncea). Purification by biospecific govern. Oceanographic Commision of UNESCO. 1996:569–72.
affinity chromatography. Biochim Biophys Acta 1973;310:446 – [20] Sharon N, Lis H. Lectins. London: Chapman and Hall, 1989.
52. [21] Wadstrom T, Trust TJ, Brooks DE. Bacterial surface lectins. In:
[6] Gold ER, Balding P. Receptor-Specific Proteins. Amsterdam: Bøg-Hasen TC, Spengler GA, editors. Lectins. Biology, Bio-
Excerpta Medica, 1975. chemistry, Clinical Biochemistry, vol. 3. Berlin: Walter de
[7] Goldstein IJ, Hughes RC, Monsigny M, Osawa T, Sharon N. Gruyter, 1983:479 – 94.
What should be called a lectin? Nature 1980;285:66. [22] Watanabe MF, Ozawa Y, Oishi S. Hemagglutinating activity of
[8] Hammarstrom H. Structure, specificity, binding properties, and Microcystis aeruginosa (Cyanobacterium). Nippon Suisan Gak-
some biological activity of a blood group. A reactive hemagglu- kaishi 1987;53:1643 – 6.

. .