- Email: [email protected]
The positions of the disulfide bonds and the glycosylation site in a lectin of the acorn barnacle Megabalanus rosa
52
Biochimica et Biophysica Acta, 1039 (1990) 52-60 Elsevier
BBAPRO 33643
The positions of the disulfide bonds and the glycosylation site in a lectin of the acorn barnacle Megabalanus rosa Koji Muramoto
and Hisao Kamiya
School of Fisheries Sciences, Kitasato University, Sanriku, lwate (Japan) (Received 20 November 1989)
Key words: Lectin; Agglutinin; Invertebrate lectin; Disulfide bond; Glycoprotein; (Acorn barnacle)
The positions of the interchain and intrachain disulfide bonds and the glycosylation site in a lectin of the acorn barnacle Megabalanus rosa were determined. The lectin (M~ 140000) is composed of the same subunit (Mr 22000) which is
cross-linked by disulfide bonds to form a dimer. Intact lectin yielded two fragments, CB1 and CB2, by cleavage with cyanogen bromide. One intrachain and two interchain disulfide bonds were identified as Cys-53-Cys-61, Cys-14-Cys-50' and Cys-50-Cys-14', respectively, by enzymatic digestion and Edman degradation of CBI. Two intrachain disulfide bonds were determined as Cys-78-Cys-168 and Cys-144-Cys-160 by enzymatic digestion of CB2. The two intrachain disulfide bonds are well conserved through all invertebrate lectins and calcium-dependent animal lectins. S-Carboxamidomethylated lectin was digested with Staphylococcus aureus V8 proteinase and separated by reversed-phase HPLC. Glycopeptides were detected by the 4-N,N-dimethylamino-4'-azobenzene sulfonyl hyrazide method. Sequence analyses of the glycopeptides showed that a carbohydrate chain attached to Asn-39.
Introduction Lectins are a group of sugar-binding proteins which recognize specific carbohydrate structures and agglutinate a variety of animal cells by binding to cellsurface glycoproteins and glycolipids. The present state of knowledge permits us to organize the known animal lectins into several categories [1]. One group consists of fl-galactose-binding proteins isolated from various tissues and the serum of many vertebrates, such as the electric eel [2], chicken [3] and human [4]. They are designated as S-type lectins, as they require thiol-reducing agents to maintain their activity. A second group consists of the calcium-dependent (C-type) animal lectins, including a mannose-binding protein isolated from mammalian liver [5] and membranous glycoprotein receptors isolated from mammalian and avian livers [6,7]. Each group exhibits its own homologous amino-acid sequences which are believed to be carbo-
hydrate-recognition domains. Invertebrate lectins isolated from the flesh fly (Sarcophaga peregrina) [8], the acorn barnacle (Megabalanus rosa) [9] and the sea urchin (Anthocidaric crassispina) [10] have now been
CBI
CB2
(b ¢.J
c
\ \
Abbreviations: HPLC, high-performance liquid chromatography; DABS, 4-N,N-dimethylamino-4'-azobenzenesulfonyl; TFA, trifluoroacetic acid; PITC, phenylisothiocyanate; FITC, fluorescein isothiocyanate; BRA, M. rosa lectin; SDS, sodium dodecyl sulfate; CB, cyanogen bromide peptides. Correspondence: K. Muramoto, School of Fisheries Sciences, Kitasato University, Sanriku, lwate 022-01, Japan.
O-O-O-O-/ 3O
l
4O
Q'~Q't• ID-e 5O
Fraction Number
Fig. 1. Gel filtration of a cyanogen bromide digest of BRA-2. The digest was applied to a column (2.2;,<80 cm) of Sephadex G-50 equilibrated with 50 mM NH4HCO3. 3-ml fractions were collected.
0167-4838/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
53
B
A
CEI
16K
11K
1
2
3
1
2
3
Fig. 2. SDS-15% polyacrylamide gel electrophoresis of cyanogen bromide fragments of BRA-2. 1, CB1; 2, CB1 digested with trypsin after modification with PITC; and 3, CB2. (A) without fl-mercaptoethanol; (B) with fl-mercaptoethanol. K represents molecular mass in kDa.
g
i=
c5
0
.t6° I I
0
1U
21J
Time(min) Fig. 4. Reversed-phase HPLC separation of disulfide-bond-containing peptides resulting from Edman degradation of fraction C for 3 cycles.
O
2O
Time(rain)
40
Fig. 3. The reversed-phase HPLC separation of disulfide-bond-containing peptides resulting from chymotryptic digestion of CB1. C represents the disulfide bond containing peptide by amino-acid analysis after performic oxidation.
found to contain a C-type carbohydrate-recognition domain. The lectins isolated from the coelomic fluid of the acorn barnacle M . rosa are multisubunit proteins with an M r value of 140000 or 64000, which we refer to as BRA-2 and BRA-3, respectively [11]. The molecular weights of the two subunits were estimated to be 13 000 and 22000, respectively, by SDS-polyacrylamide gel electrophoresis, and they are cross-linked by disulfide bonds to form dimers. The subunit of BRA-3 consists of 138 amino-acid residues and has no carbohydrate chain [9]. In the preceding paper [12], we described the amino-acid sequence of BRA-2. Its subunit is a glycoprotein of 173 amino acids. The sequence showed homologies with those of other invertebrate lectins, as well as with C-type lectins.
54 TABLE I
Amino-acid compositions of BRA-2 and its CNBr fragments (as a subuniO //
BRA-2 Asx Glx Cys b Ser Thr Gly Ala Pro Val Arg Met Ile Leu Trp Vhe Lys His Tyr
(23) a (21) (8) (13) (8) (6) (16) (9) (15) (7) (2) (4) (11) (6) (4) (5) (9) (6)
Total
173
CB1
CB2
5.6 (6) 9.9 (11) 4.0 (4) 2.4 (4) 5.4 (7) 3.4 (4) 3.8 (4) 7.6 (7) 7.1 (7) 1.4 (1) (1) 4.1 1.8 3.0 3.6 4.1 1.7
/
15.9 (17) 9.7 (10) 3.6 (4) 6.9 (10) 2.5 (1) 2.1 (2) 10.1 (12) 2.4 (2) 7.1 (8) 7.0 (6) (1) 3.9 (4) 7.2 (7) 4.4 (4) 1.5 (1) 2.9 (2) 6.1 (5) 4.2 (5)
(4) (2) (3) (3) (4) (1)
72
/ / / /
/
/
/
/
/
/
/
4O
/ ¢M
// o //
U
// //
//
1
TC2
2O
101
a The numbers in parentheses are from the sequence analysis. b Determined as S-carboxymethylated cysteine. I
0 TABLE II
Amino-acid compositions of disulfide-containing peptides obtained from enzymatic digestion and Edman degradation of CNBr fragments of BRA -2 CB1 C ASX GIx Cys Ser Thr Gly Ala Pro Val Arg Met lle Leu Trp Phe Lys His Tyr
3.9 (4) a 4.1 (5) 1.7 (4) 0.9 (1) 2.6 (2) 3.2 (4) 0.7 (1) 6.0 (6) 4.1 (4) 0.6 (1)
CB2 CE1 1.6 (1) 3.7 (4) (2) 1.2 (1) 1.4 (1) 0.8 (1) 3.6 (4) 1.6 (2) 0.6 (1)
CE2 2.2 (3) 1.2 (1) 2.0 (2) 0.7 (1) 0.9 (1) 1.0 (1) 1.0 (1) 0.9 (1)
TC1
TC2
3.1 (3) 1.7 (2)
1.0 (1) 1.3 (2)
1.2 (1) 1.5 (2) 1.0 (1) 1.0 (1)
3.1 (4) 1.1 (1)
1.1 (1)
0.7 (1) 1.2 (1)
20
Ti me (mi n)
40
0
Fig. 5. Reversed-phase HPLC separation of disulfide-bond-containing peptides resulting from the enzymatic digestion of CB2. CB2 was digested successivelywith trypsin and chymotrypsin.
I n this paper, we describe the positions of i n t e r c h a i n a n d i n t r a c h a i n disulfide b o n d s a n d the glycosylation site in BRA-2. G l y c o p e p t i d e s derived from the enzymatic digest of B R A - 2 were detected b y the 4-N, Ndimethylamino-4'-azobenzene sulfonyl (DABS)-hydrazide method. This m e t h o d involves reversed-phase H P L C analysis of D A B S - h y d r a z o n e s of glycopeptide hydrolysates at high sensitivity [13].
Materials and M e t h o d s
Cyanogen bromide cleavage o f B R A - 2 N.D. (1) 1.3 (1) 1.3 (1) 1.6 (2)
0.7 (1)
N.D. (1) 0.8 (1) 0.5 (1) 1.3 (1)
1.1 (1) 0.6 (1)
Total
37
16
15
Positions
2-21 45-62
5-21
48-62
9 144-146 159-164
14 73-80 167-172
a The numbers in parentheses are from the sequence analysis. C, chymotryptic peptide of CB1; CE, chymotryptic peptides after 3 cycles of Edman degradation; TC, chymotryptic and tryptic peptides of CB2. N.D., not determined.
B R A - 2 (10 mg) was cleaved with c y a n o g e n b r o m i d e in 70% formic acid, as previously described (see Ref. 14). T h e isolation of c y a n o g e n - b r o m i d e peptides was carried out o n a Sephadex G-50 c o l u m n (2.2 × 80 cm) equilibrated with 50 m M N H a H C O 3. Yields were 3.2 mg for CB1 a n d 3.7 m g for CB2. CB1 was purified by reversed-phase H P L C o n a T S K gel p h e n y l - 5 P W R P (4.6 × 75 mm, Toso) using a linear g r a d i e n t of acetonitrile in 0.1% trifluoroacetic acid (TFA).
Isolation o f disulfide-bond-containing peptides from CB1 CB1 (500 /~g) was digested with c h y m o t r y p s i n (P-L Biochemicals) ( 5 0 : 1 , w / w ) i n 0.5 ml of 0.1 M amm o n i u m acetate ( p H 7.0) at 37 ° C for 12 h. The digest
55 was separated by reversed-phase HPLC on a TSK ODS 120T column (10 #m, 4.6 x 250 mm, Toso) using a linear gradient of acetonitrile in 0.1% TFA at a flow rate of 1.0 ml/min. One-tenth of each fraction was submitted for amino-acid analysis after performic acid oxidation. Fraction C, showing disulfide bonds by amino-acid analysis, was subjected to Edman degradation for three cycles. The peptide was dissolved in 100 /~1 of 60% pyridine and 2 #1 of phenylisothiocyanate (PITC). The coupling reaction was performed for 30 min at 50 ° C under a nitrogen atmosphere. After drying under vacuum, the residue was treated with TFA for 7 min at 50 ° C. The acid was removed by evaporation with a stream of nitrogen. The residue was dissolved in 50 /~1 of distilled water, and the terminal amino-acid derivative was extracted with 100 #I of ethyl acetate. The aqueous phase was dried under vacuum and used in the next cycle. The resulting peptides were separated by reversed-phase H P L C . Disulfide-bond-containing peptides were subjected to amino-acid sequence analysis by the fluorescein isothiocyanate (FITC) method [15] to locate the peptides in the sequence.
A
C
B
D
Isolation of disulfide-bond-containing peptides from CB2 CB2 (500 #g) was digested successively with trypsin (Worthington) and chymotrypsin (substrate/enzyme = 50:1) in 0.5 ml of 0.1 M ammonium acetate (pH 7.0). Each digestion was carried out for 16 h at 37 o C. The digest was separated by reversed-phase HPLC, as described above.
Assignment of interchain disulfide bonds CB1 (200 pg) was reacted with 5 btl of PITC in 100 #1 of 60% aqueous pyridine at 50 ° C for 40 min. After drying under vacuum, the residue was washed with 500 #1 of b e n z e n e / e t h y l acetate (1:1) to remove excess reagents, and then dried again. The modified peptide was digested with T P C K / t r y p s i n (25:1) in 100 /~1 of 0.1 M ammonium acetate (pH 7.0) containing 2 M urea at 37 ° C for 12 h. SDS-polyacrylamide gel electrophoresis was performed on 15% slab gel, according to the method of Laemmli [16].
Neutral sugar analysis Samples were hydrolyzed with 50/~1 of 2 M T F A in evacuated, sealed glass tubes (3 × 60 mm) at 105 ° C for 4 h and dried under vacuum. The samples were dissolved in 5 #1 of distilled water and ddrivatized with 40 #1 of 0.1% DABS-hydrazide ethanol solution containing 0.25% trichloroacetic acid at 5 0 ° C for 2 h [13]. DABSsugars were separated by reversed-phase H P L C on an O D S Hypersil (3-/~m particles, 4.6 X 50 mm, Shandon Southern Products) column and detected at 485 nm. The column was developed by an isocratic solvent consisting of 28.5% acetone/0.08 M acetic acid at a flow rate of 1.0 m l / m i n at 50°C, and washed with 80% acetone/0.08 M acetic acid before the next run.
Aminosugar analysis Glycoprotein was hydrolyzed with 50 #1 of 4 M HC1 in an evacuated, sealed glass tube (3 × 60 mm) at 110 ° C for 4 h and dried under vacuum. The hydrolyzate was derivatized with DABS-chloride by the method of Knecht and Chang [17]. Aminosugar derivatives were separated by reversed-phase HPLC, as described above. Amino-acid analysis and sequence analysis were performed as described in the preceding article [12]. I
0
15 0
|
II
15 0 Time(min)
I
150
15
Fig. 6. HPLC chromatograms of DABS derivatives obtained from glycopeptide hydrolyzates. S-carboxamidomethylated BRA-2 (1 nag) was digested with V8 proteinas¢ and separated by reversed-phase HPLC. A fiftieth of each fraction was hydrolyzed in 2 M TFA at 105 ° C for 4 h and derivatized with DABS-hydrazide as in the text. 2 #1 of the derivatized sample was subjected to chromatography. (A) 5 #1 of 0.125 mM sugar mixture was derivatized. 2 #1 (28 pmol) of the derivatized sample was subjected to chromatography. 1, gentiobiose; 2, galactose; 3, glucose; 4, mannose; 5, xylose; 6, fucose; 7, rhamnose; and 8, deoxyribose. (B) V3, (C) V4 and (D) V5.
Results
BRA-2 is composed of six identical subunits ( M r 22000), which form dimers cross-linked by disulfide bonds [11]. Upon treatment with cyanogen bromide, BRA-2 yielded two fragments, CB1 and CB2, by gel filtration on a Sephadex G-50 column equilibrated with 50 mM N H 4 H C O 3 (Fig. 1). The amino-acid compositions of the two fragments, together with that of the whole chain, are shown in Table I. The molecular weights of CB1 and CB2 were estimated to be 16 000
++ ..j
~.~ E]
II
~o
r-
!~.~)~"
.++}
F
L~4" 1--
D !5 ,e4"
~'m
Z
X
"r
~~ ,~.
:
25.94";
"o "o
~,q4"~ C
~~
~. °~
2
•~~ ~.~ ..j
!.~
f-
[
(-
" ~ ~
9~
57 and 11000, respectively, by SDS-polyacrylamide gel electrophoresis (Fig. 2). Upon reduction with fl-mercaptoethanol, the fragments split in half. The aminoterminal sequence of CB1 proved to be identical to that of BRA-2. CB2 has valine and tryptophan as the amino-terminal amino acids. Consequently, it was assumed that the interchain disulfide bonds are included in CB1 located at the amino-terminal region. By digestion of CB1 with chymotrypsin, fraction C containing two disulfide bonds was isolated as shown in Fig. 3. The peptide was separated into two fragments by reversed-phase HPLC after Edman degradation with phenylisothiocyanate for three cycles (Fig. 4). From the amino-acid compositions of CE1 and CE2 (Table II), one intrachain and two interchain disulfide bonds were located as follows: Cys-50-Cys-61, Cys-14-Cys-47' and Cys-47-Cys-14'. Chymotrypsin could not hydrolyze the peptide bond between Trp-54 and Val-55. Cystine was not recovered from CE1 due to the modification with PITC. Because arginine was present only at position 21 in CB1, the positions of interchain disulfide bonds were confirmed by SDS-polyacrylamide gel electrophoresis of tryptic digest of CB1, whose amino groups had been blocked with PITC to prevent the hydrolysis of lysine bonds (Fig. 2). CB1 was converted to the half-size of the molecule, indicating that Cys-14 and Cys-47 were cross-linked with the counterpart of the molecule by disulfide bonds. The peptide bond between Arg-21 and Val-22 was not hydrolyzed by arginine endopeptidase (mouse submandibular proteinase, Takara). Successive enzymatic digestion with trypsin and chymotrypsin of CB2 yielded TC1 and TC2 (Fig. 5). From the analyses of these two peptides, two intrachain disulfide bonds were located as Cys-78-Cys-168 and Cys-144-Cys-160. BRA-2 is a glycoprotein which contains 0.6% mannose, 0.2% fucose and 2.3% glucosamine by weight. To assign the position of the glycosylation site, Scarboxamidomethylated BRA-2 was digested with V8 proteinase and separated by reversed-phase HPLC (see Fig. 2 of Ref. 12). A fiftieth of each fraction was hydrolyzed in 2 M TFA at 105°C for 4 h and derivatized with DABS-hydrazide. The DABS derivatives were analyzed by reversed-phase HPLC. V4 and V5 gave DABS-mannose and DABS-fucose at the same ratios as the S-CAM-BRA-2 (Fig. 6). No DABS-hydrazone was detected for the remaining fractions. The amino-acid sequences of both V4 and V5 were determined to be Leu-Gln-Glu-X-Val-Thr-Asn-Thr-Phe-His-Gly-CysAsn-His-Cys-Pro- by the FITC method as shown in Fig. 7. The fourth cycle of Edman degradation gave unknown derivatives between glutamic acid and the asparagine derivative. Aspartic acid was identified after the acid hydrolysis of the derivatives with 4 M HC1 for 3 h at l l 0 ° C (Fig. 8). Thus, it was concluded that a carbohydrate chain attached to Asn-39. The amino-acid
4.A
4.11
E 0 A y HV
H
V I
~d Fig. 8. HPLC chromatograms of FITC derivatives resulting from the fourth cycle of Edman degradation of V4. Left panel is a chromatogram of authentic FITC derivatives of amino acids. 4-A, before acid hydrolysis; 4-B, after acid hydrolysis in 4 M HCI for 3 h at 110°C.
compositions of V4 and V5 were in good agreement with the expected values (see Table II in Ref. 12). The structural difference between V4 and V5 was not detected in this experiment. The complete amino-acid sequence of the subunit of BRA-2 is shown in Fig. 9 together with that of BRA-3 for comparison. Discussion
There are two interchain and three intrachain disulfide bonds in the subunit of BRA-2. Two of the intrachain disulfide bonds are well conserved throughout invertebrate lectins and C-type animal lectins in spite of low sequence.homology among them; the BRA-2 and BRA-3 show 25% identity in the amino-acid sequence. These disulfide bonds form characteristic loops which are tentative carbohydrate-recognition domains. Another intrachain disulfide bond forming a small loop is also well conserved, though not always. On the other hand, there is no similarity in interchain disulfide bonds. Two interchain disulfide bonds cross-link to form dimers at the carboxyl-terminal segment of the subunit in BRA-3. However, two interchain disulfide bonds are located at the amino-terminal segment in BRA-2. This finding suggests that the positions of interchain disulfide bonds are not critical as long as they do not interfere with the carbohydrate-recognition domain. This is a useful finding if one wants to construct a new artificial multifunctional protein by combining the carbohydrate recognition domain and another functional protein.
58
.,..,~
30
UreA- 2 Fig. 9. Schematic representations of the positions of the disulfide bonds and the glycosylationsite of BRA-2 and BRA-3 (see opposite page). The numbering starts from the amino-terminus.
There is only one Asn-X-Ser/Thr sequence which is a possible glycosylation site in BRA-2. In fact, the site was glycosylated as shown in the result. There is no consistency for the glycosylation in animal lectins. For example, BRA-3 has no carbohydrate chain. Although there is room for discussion of the biological impor-
tance of the carbohydrate chain, the carbohydrate chain seems to be unnecessary for the carbohydrate binding activity of lectins. The DABS-hydrazide method for the detection of glycopeptides from H P L C has the advantage of being capable of providing the sugar composition with only a small amount of sample. Therefore,
59 08
mIRA-3 80 Fig. 9 (continued).
the glycopeptide can be easily assigned as to whether it represents the carbohydrate chain of the parent glycoprotein or not.
Science and Culture of Japan, and a Grant-in-Aid from the Fisheries Agency, Japan. References
Acknowledgements This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education,
1 Drickamer, K. (1988) J. Biol. Chem. 263, 9557-9560. 2 Parountaud, P., Levi, G., Teichberg, V.I. and Strosberg, A.D. (1987) Proc. Natl. Acad. Sci. USA 84, 6345-6348. 3 0 h y a m a , Y., Hirabayashi, J., Oda, Y., Ohno, S., Kawasaki, H.,
60
4 5 6 7 8
9
Suzuki, K. and Kasai, K. (1986) Biochem. Biophys. Res. Commun. 34, 51-56. Hirabayashi, J. and Kasai, K. (1988) J. Biochem. 104, 1-4. Drickamer, K., Dordal, M.S. and Reynolds, L. (1986) J. Biol. Chem. 261, 6878-6887. Drickamer, K., Mamon, J.F., Binns, G. and Leung, J.O. (1984) J. Biol. Chem. 259, 770-778. Spiess, M. and Lodish, H.F. (1985) Proc. Natl. Acad. Sci. USA 82, 6465-6469. Takahashi, H., Komano, H., Kawauchi, N., Kitamura, N., Nakanishi, S. and Natori, S. (1985) J. Biol. Chem. 260, 1222812233. Muramoto, K. and Kamiya, H. (1986) Biochim. Biophys. Acta 874, 285-295.
10 Giga, Y., Ikai, A. and Takahashi, K. (1987) J. Biol. Chem. 262, 6197-6203. 11 Muramoto, K., Ogata, K. and Kamiya, H. (1985) Agric. Biol. Chem. 49, 85-93. 12 Muramoto, K. and Kamiya, H. (1990) Biochim. Biophys. Acta 1039, 42-51. 13 Muramoto, K., Goto, R. and Kamiya, H. (1987) Anal. Biochem. 162, 435-442. 14 Grosse, E. (1967) Methods Enzymol. 11,238-255. 15 Muramoto, K., Kawauchi, H. and Tuzimura, K. (1978) Agric. Biol. Chem. 42, 1559-1563. 16 Laemmli, U.K. (1970) Nature 227, 680-685. 17 Knecht, R. and Chang, J.-Y. (1986) Anal. Chem. 58, 2375-2379.
Biochimica et Biophysica Acta, 1039 (1990) 52-60 Elsevier
BBAPRO 33643
The positions of the disulfide bonds and the glycosylation site in a lectin of the acorn barnacle Megabalanus rosa Koji Muramoto
and Hisao Kamiya
School of Fisheries Sciences, Kitasato University, Sanriku, lwate (Japan) (Received 20 November 1989)
Key words: Lectin; Agglutinin; Invertebrate lectin; Disulfide bond; Glycoprotein; (Acorn barnacle)
The positions of the interchain and intrachain disulfide bonds and the glycosylation site in a lectin of the acorn barnacle Megabalanus rosa were determined. The lectin (M~ 140000) is composed of the same subunit (Mr 22000) which is
cross-linked by disulfide bonds to form a dimer. Intact lectin yielded two fragments, CB1 and CB2, by cleavage with cyanogen bromide. One intrachain and two interchain disulfide bonds were identified as Cys-53-Cys-61, Cys-14-Cys-50' and Cys-50-Cys-14', respectively, by enzymatic digestion and Edman degradation of CBI. Two intrachain disulfide bonds were determined as Cys-78-Cys-168 and Cys-144-Cys-160 by enzymatic digestion of CB2. The two intrachain disulfide bonds are well conserved through all invertebrate lectins and calcium-dependent animal lectins. S-Carboxamidomethylated lectin was digested with Staphylococcus aureus V8 proteinase and separated by reversed-phase HPLC. Glycopeptides were detected by the 4-N,N-dimethylamino-4'-azobenzene sulfonyl hyrazide method. Sequence analyses of the glycopeptides showed that a carbohydrate chain attached to Asn-39.
Introduction Lectins are a group of sugar-binding proteins which recognize specific carbohydrate structures and agglutinate a variety of animal cells by binding to cellsurface glycoproteins and glycolipids. The present state of knowledge permits us to organize the known animal lectins into several categories [1]. One group consists of fl-galactose-binding proteins isolated from various tissues and the serum of many vertebrates, such as the electric eel [2], chicken [3] and human [4]. They are designated as S-type lectins, as they require thiol-reducing agents to maintain their activity. A second group consists of the calcium-dependent (C-type) animal lectins, including a mannose-binding protein isolated from mammalian liver [5] and membranous glycoprotein receptors isolated from mammalian and avian livers [6,7]. Each group exhibits its own homologous amino-acid sequences which are believed to be carbo-
hydrate-recognition domains. Invertebrate lectins isolated from the flesh fly (Sarcophaga peregrina) [8], the acorn barnacle (Megabalanus rosa) [9] and the sea urchin (Anthocidaric crassispina) [10] have now been
CBI
CB2
(b ¢.J
c
\ \
Abbreviations: HPLC, high-performance liquid chromatography; DABS, 4-N,N-dimethylamino-4'-azobenzenesulfonyl; TFA, trifluoroacetic acid; PITC, phenylisothiocyanate; FITC, fluorescein isothiocyanate; BRA, M. rosa lectin; SDS, sodium dodecyl sulfate; CB, cyanogen bromide peptides. Correspondence: K. Muramoto, School of Fisheries Sciences, Kitasato University, Sanriku, lwate 022-01, Japan.
O-O-O-O-/ 3O
l
4O
Q'~Q't• ID-e 5O
Fraction Number
Fig. 1. Gel filtration of a cyanogen bromide digest of BRA-2. The digest was applied to a column (2.2;,<80 cm) of Sephadex G-50 equilibrated with 50 mM NH4HCO3. 3-ml fractions were collected.
0167-4838/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
53
B
A
CEI
16K
11K
1
2
3
1
2
3
Fig. 2. SDS-15% polyacrylamide gel electrophoresis of cyanogen bromide fragments of BRA-2. 1, CB1; 2, CB1 digested with trypsin after modification with PITC; and 3, CB2. (A) without fl-mercaptoethanol; (B) with fl-mercaptoethanol. K represents molecular mass in kDa.
g
i=
c5
0
.t6° I I
0
1U
21J
Time(min) Fig. 4. Reversed-phase HPLC separation of disulfide-bond-containing peptides resulting from Edman degradation of fraction C for 3 cycles.
O
2O
Time(rain)
40
Fig. 3. The reversed-phase HPLC separation of disulfide-bond-containing peptides resulting from chymotryptic digestion of CB1. C represents the disulfide bond containing peptide by amino-acid analysis after performic oxidation.
found to contain a C-type carbohydrate-recognition domain. The lectins isolated from the coelomic fluid of the acorn barnacle M . rosa are multisubunit proteins with an M r value of 140000 or 64000, which we refer to as BRA-2 and BRA-3, respectively [11]. The molecular weights of the two subunits were estimated to be 13 000 and 22000, respectively, by SDS-polyacrylamide gel electrophoresis, and they are cross-linked by disulfide bonds to form dimers. The subunit of BRA-3 consists of 138 amino-acid residues and has no carbohydrate chain [9]. In the preceding paper [12], we described the amino-acid sequence of BRA-2. Its subunit is a glycoprotein of 173 amino acids. The sequence showed homologies with those of other invertebrate lectins, as well as with C-type lectins.
54 TABLE I
Amino-acid compositions of BRA-2 and its CNBr fragments (as a subuniO //
BRA-2 Asx Glx Cys b Ser Thr Gly Ala Pro Val Arg Met Ile Leu Trp Vhe Lys His Tyr
(23) a (21) (8) (13) (8) (6) (16) (9) (15) (7) (2) (4) (11) (6) (4) (5) (9) (6)
Total
173
CB1
CB2
5.6 (6) 9.9 (11) 4.0 (4) 2.4 (4) 5.4 (7) 3.4 (4) 3.8 (4) 7.6 (7) 7.1 (7) 1.4 (1) (1) 4.1 1.8 3.0 3.6 4.1 1.7
/
15.9 (17) 9.7 (10) 3.6 (4) 6.9 (10) 2.5 (1) 2.1 (2) 10.1 (12) 2.4 (2) 7.1 (8) 7.0 (6) (1) 3.9 (4) 7.2 (7) 4.4 (4) 1.5 (1) 2.9 (2) 6.1 (5) 4.2 (5)
(4) (2) (3) (3) (4) (1)
72
/ / / /
/
/
/
/
/
/
/
4O
/ ¢M
// o //
U
// //
//
1
TC2
2O
101
a The numbers in parentheses are from the sequence analysis. b Determined as S-carboxymethylated cysteine. I
0 TABLE II
Amino-acid compositions of disulfide-containing peptides obtained from enzymatic digestion and Edman degradation of CNBr fragments of BRA -2 CB1 C ASX GIx Cys Ser Thr Gly Ala Pro Val Arg Met lle Leu Trp Phe Lys His Tyr
3.9 (4) a 4.1 (5) 1.7 (4) 0.9 (1) 2.6 (2) 3.2 (4) 0.7 (1) 6.0 (6) 4.1 (4) 0.6 (1)
CB2 CE1 1.6 (1) 3.7 (4) (2) 1.2 (1) 1.4 (1) 0.8 (1) 3.6 (4) 1.6 (2) 0.6 (1)
CE2 2.2 (3) 1.2 (1) 2.0 (2) 0.7 (1) 0.9 (1) 1.0 (1) 1.0 (1) 0.9 (1)
TC1
TC2
3.1 (3) 1.7 (2)
1.0 (1) 1.3 (2)
1.2 (1) 1.5 (2) 1.0 (1) 1.0 (1)
3.1 (4) 1.1 (1)
1.1 (1)
0.7 (1) 1.2 (1)
20
Ti me (mi n)
40
0
Fig. 5. Reversed-phase HPLC separation of disulfide-bond-containing peptides resulting from the enzymatic digestion of CB2. CB2 was digested successivelywith trypsin and chymotrypsin.
I n this paper, we describe the positions of i n t e r c h a i n a n d i n t r a c h a i n disulfide b o n d s a n d the glycosylation site in BRA-2. G l y c o p e p t i d e s derived from the enzymatic digest of B R A - 2 were detected b y the 4-N, Ndimethylamino-4'-azobenzene sulfonyl (DABS)-hydrazide method. This m e t h o d involves reversed-phase H P L C analysis of D A B S - h y d r a z o n e s of glycopeptide hydrolysates at high sensitivity [13].
Materials and M e t h o d s
Cyanogen bromide cleavage o f B R A - 2 N.D. (1) 1.3 (1) 1.3 (1) 1.6 (2)
0.7 (1)
N.D. (1) 0.8 (1) 0.5 (1) 1.3 (1)
1.1 (1) 0.6 (1)
Total
37
16
15
Positions
2-21 45-62
5-21
48-62
9 144-146 159-164
14 73-80 167-172
a The numbers in parentheses are from the sequence analysis. C, chymotryptic peptide of CB1; CE, chymotryptic peptides after 3 cycles of Edman degradation; TC, chymotryptic and tryptic peptides of CB2. N.D., not determined.
B R A - 2 (10 mg) was cleaved with c y a n o g e n b r o m i d e in 70% formic acid, as previously described (see Ref. 14). T h e isolation of c y a n o g e n - b r o m i d e peptides was carried out o n a Sephadex G-50 c o l u m n (2.2 × 80 cm) equilibrated with 50 m M N H a H C O 3. Yields were 3.2 mg for CB1 a n d 3.7 m g for CB2. CB1 was purified by reversed-phase H P L C o n a T S K gel p h e n y l - 5 P W R P (4.6 × 75 mm, Toso) using a linear g r a d i e n t of acetonitrile in 0.1% trifluoroacetic acid (TFA).
Isolation o f disulfide-bond-containing peptides from CB1 CB1 (500 /~g) was digested with c h y m o t r y p s i n (P-L Biochemicals) ( 5 0 : 1 , w / w ) i n 0.5 ml of 0.1 M amm o n i u m acetate ( p H 7.0) at 37 ° C for 12 h. The digest
55 was separated by reversed-phase HPLC on a TSK ODS 120T column (10 #m, 4.6 x 250 mm, Toso) using a linear gradient of acetonitrile in 0.1% TFA at a flow rate of 1.0 ml/min. One-tenth of each fraction was submitted for amino-acid analysis after performic acid oxidation. Fraction C, showing disulfide bonds by amino-acid analysis, was subjected to Edman degradation for three cycles. The peptide was dissolved in 100 /~1 of 60% pyridine and 2 #1 of phenylisothiocyanate (PITC). The coupling reaction was performed for 30 min at 50 ° C under a nitrogen atmosphere. After drying under vacuum, the residue was treated with TFA for 7 min at 50 ° C. The acid was removed by evaporation with a stream of nitrogen. The residue was dissolved in 50 /~1 of distilled water, and the terminal amino-acid derivative was extracted with 100 #I of ethyl acetate. The aqueous phase was dried under vacuum and used in the next cycle. The resulting peptides were separated by reversed-phase H P L C . Disulfide-bond-containing peptides were subjected to amino-acid sequence analysis by the fluorescein isothiocyanate (FITC) method [15] to locate the peptides in the sequence.
A
C
B
D
Isolation of disulfide-bond-containing peptides from CB2 CB2 (500 #g) was digested successively with trypsin (Worthington) and chymotrypsin (substrate/enzyme = 50:1) in 0.5 ml of 0.1 M ammonium acetate (pH 7.0). Each digestion was carried out for 16 h at 37 o C. The digest was separated by reversed-phase HPLC, as described above.
Assignment of interchain disulfide bonds CB1 (200 pg) was reacted with 5 btl of PITC in 100 #1 of 60% aqueous pyridine at 50 ° C for 40 min. After drying under vacuum, the residue was washed with 500 #1 of b e n z e n e / e t h y l acetate (1:1) to remove excess reagents, and then dried again. The modified peptide was digested with T P C K / t r y p s i n (25:1) in 100 /~1 of 0.1 M ammonium acetate (pH 7.0) containing 2 M urea at 37 ° C for 12 h. SDS-polyacrylamide gel electrophoresis was performed on 15% slab gel, according to the method of Laemmli [16].
Neutral sugar analysis Samples were hydrolyzed with 50/~1 of 2 M T F A in evacuated, sealed glass tubes (3 × 60 mm) at 105 ° C for 4 h and dried under vacuum. The samples were dissolved in 5 #1 of distilled water and ddrivatized with 40 #1 of 0.1% DABS-hydrazide ethanol solution containing 0.25% trichloroacetic acid at 5 0 ° C for 2 h [13]. DABSsugars were separated by reversed-phase H P L C on an O D S Hypersil (3-/~m particles, 4.6 X 50 mm, Shandon Southern Products) column and detected at 485 nm. The column was developed by an isocratic solvent consisting of 28.5% acetone/0.08 M acetic acid at a flow rate of 1.0 m l / m i n at 50°C, and washed with 80% acetone/0.08 M acetic acid before the next run.
Aminosugar analysis Glycoprotein was hydrolyzed with 50 #1 of 4 M HC1 in an evacuated, sealed glass tube (3 × 60 mm) at 110 ° C for 4 h and dried under vacuum. The hydrolyzate was derivatized with DABS-chloride by the method of Knecht and Chang [17]. Aminosugar derivatives were separated by reversed-phase HPLC, as described above. Amino-acid analysis and sequence analysis were performed as described in the preceding article [12]. I
0
15 0
|
II
15 0 Time(min)
I
150
15
Fig. 6. HPLC chromatograms of DABS derivatives obtained from glycopeptide hydrolyzates. S-carboxamidomethylated BRA-2 (1 nag) was digested with V8 proteinas¢ and separated by reversed-phase HPLC. A fiftieth of each fraction was hydrolyzed in 2 M TFA at 105 ° C for 4 h and derivatized with DABS-hydrazide as in the text. 2 #1 of the derivatized sample was subjected to chromatography. (A) 5 #1 of 0.125 mM sugar mixture was derivatized. 2 #1 (28 pmol) of the derivatized sample was subjected to chromatography. 1, gentiobiose; 2, galactose; 3, glucose; 4, mannose; 5, xylose; 6, fucose; 7, rhamnose; and 8, deoxyribose. (B) V3, (C) V4 and (D) V5.
Results
BRA-2 is composed of six identical subunits ( M r 22000), which form dimers cross-linked by disulfide bonds [11]. Upon treatment with cyanogen bromide, BRA-2 yielded two fragments, CB1 and CB2, by gel filtration on a Sephadex G-50 column equilibrated with 50 mM N H 4 H C O 3 (Fig. 1). The amino-acid compositions of the two fragments, together with that of the whole chain, are shown in Table I. The molecular weights of CB1 and CB2 were estimated to be 16 000
++ ..j
~.~ E]
II
~o
r-
!~.~)~"
.++}
F
L~4" 1--
D !5 ,e4"
~'m
Z
X
"r
~~ ,~.
:
25.94";
"o "o
~,q4"~ C
~~
~. °~
2
•~~ ~.~ ..j
!.~
f-
[
(-
" ~ ~
9~
57 and 11000, respectively, by SDS-polyacrylamide gel electrophoresis (Fig. 2). Upon reduction with fl-mercaptoethanol, the fragments split in half. The aminoterminal sequence of CB1 proved to be identical to that of BRA-2. CB2 has valine and tryptophan as the amino-terminal amino acids. Consequently, it was assumed that the interchain disulfide bonds are included in CB1 located at the amino-terminal region. By digestion of CB1 with chymotrypsin, fraction C containing two disulfide bonds was isolated as shown in Fig. 3. The peptide was separated into two fragments by reversed-phase HPLC after Edman degradation with phenylisothiocyanate for three cycles (Fig. 4). From the amino-acid compositions of CE1 and CE2 (Table II), one intrachain and two interchain disulfide bonds were located as follows: Cys-50-Cys-61, Cys-14-Cys-47' and Cys-47-Cys-14'. Chymotrypsin could not hydrolyze the peptide bond between Trp-54 and Val-55. Cystine was not recovered from CE1 due to the modification with PITC. Because arginine was present only at position 21 in CB1, the positions of interchain disulfide bonds were confirmed by SDS-polyacrylamide gel electrophoresis of tryptic digest of CB1, whose amino groups had been blocked with PITC to prevent the hydrolysis of lysine bonds (Fig. 2). CB1 was converted to the half-size of the molecule, indicating that Cys-14 and Cys-47 were cross-linked with the counterpart of the molecule by disulfide bonds. The peptide bond between Arg-21 and Val-22 was not hydrolyzed by arginine endopeptidase (mouse submandibular proteinase, Takara). Successive enzymatic digestion with trypsin and chymotrypsin of CB2 yielded TC1 and TC2 (Fig. 5). From the analyses of these two peptides, two intrachain disulfide bonds were located as Cys-78-Cys-168 and Cys-144-Cys-160. BRA-2 is a glycoprotein which contains 0.6% mannose, 0.2% fucose and 2.3% glucosamine by weight. To assign the position of the glycosylation site, Scarboxamidomethylated BRA-2 was digested with V8 proteinase and separated by reversed-phase HPLC (see Fig. 2 of Ref. 12). A fiftieth of each fraction was hydrolyzed in 2 M TFA at 105°C for 4 h and derivatized with DABS-hydrazide. The DABS derivatives were analyzed by reversed-phase HPLC. V4 and V5 gave DABS-mannose and DABS-fucose at the same ratios as the S-CAM-BRA-2 (Fig. 6). No DABS-hydrazone was detected for the remaining fractions. The amino-acid sequences of both V4 and V5 were determined to be Leu-Gln-Glu-X-Val-Thr-Asn-Thr-Phe-His-Gly-CysAsn-His-Cys-Pro- by the FITC method as shown in Fig. 7. The fourth cycle of Edman degradation gave unknown derivatives between glutamic acid and the asparagine derivative. Aspartic acid was identified after the acid hydrolysis of the derivatives with 4 M HC1 for 3 h at l l 0 ° C (Fig. 8). Thus, it was concluded that a carbohydrate chain attached to Asn-39. The amino-acid
4.A
4.11
E 0 A y HV
H
V I
~d Fig. 8. HPLC chromatograms of FITC derivatives resulting from the fourth cycle of Edman degradation of V4. Left panel is a chromatogram of authentic FITC derivatives of amino acids. 4-A, before acid hydrolysis; 4-B, after acid hydrolysis in 4 M HCI for 3 h at 110°C.
compositions of V4 and V5 were in good agreement with the expected values (see Table II in Ref. 12). The structural difference between V4 and V5 was not detected in this experiment. The complete amino-acid sequence of the subunit of BRA-2 is shown in Fig. 9 together with that of BRA-3 for comparison. Discussion
There are two interchain and three intrachain disulfide bonds in the subunit of BRA-2. Two of the intrachain disulfide bonds are well conserved throughout invertebrate lectins and C-type animal lectins in spite of low sequence.homology among them; the BRA-2 and BRA-3 show 25% identity in the amino-acid sequence. These disulfide bonds form characteristic loops which are tentative carbohydrate-recognition domains. Another intrachain disulfide bond forming a small loop is also well conserved, though not always. On the other hand, there is no similarity in interchain disulfide bonds. Two interchain disulfide bonds cross-link to form dimers at the carboxyl-terminal segment of the subunit in BRA-3. However, two interchain disulfide bonds are located at the amino-terminal segment in BRA-2. This finding suggests that the positions of interchain disulfide bonds are not critical as long as they do not interfere with the carbohydrate-recognition domain. This is a useful finding if one wants to construct a new artificial multifunctional protein by combining the carbohydrate recognition domain and another functional protein.
58
.,..,~
30
UreA- 2 Fig. 9. Schematic representations of the positions of the disulfide bonds and the glycosylationsite of BRA-2 and BRA-3 (see opposite page). The numbering starts from the amino-terminus.
There is only one Asn-X-Ser/Thr sequence which is a possible glycosylation site in BRA-2. In fact, the site was glycosylated as shown in the result. There is no consistency for the glycosylation in animal lectins. For example, BRA-3 has no carbohydrate chain. Although there is room for discussion of the biological impor-
tance of the carbohydrate chain, the carbohydrate chain seems to be unnecessary for the carbohydrate binding activity of lectins. The DABS-hydrazide method for the detection of glycopeptides from H P L C has the advantage of being capable of providing the sugar composition with only a small amount of sample. Therefore,
59 08
mIRA-3 80 Fig. 9 (continued).
the glycopeptide can be easily assigned as to whether it represents the carbohydrate chain of the parent glycoprotein or not.
Science and Culture of Japan, and a Grant-in-Aid from the Fisheries Agency, Japan. References
Acknowledgements This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education,
1 Drickamer, K. (1988) J. Biol. Chem. 263, 9557-9560. 2 Parountaud, P., Levi, G., Teichberg, V.I. and Strosberg, A.D. (1987) Proc. Natl. Acad. Sci. USA 84, 6345-6348. 3 0 h y a m a , Y., Hirabayashi, J., Oda, Y., Ohno, S., Kawasaki, H.,
60
4 5 6 7 8
9
Suzuki, K. and Kasai, K. (1986) Biochem. Biophys. Res. Commun. 34, 51-56. Hirabayashi, J. and Kasai, K. (1988) J. Biochem. 104, 1-4. Drickamer, K., Dordal, M.S. and Reynolds, L. (1986) J. Biol. Chem. 261, 6878-6887. Drickamer, K., Mamon, J.F., Binns, G. and Leung, J.O. (1984) J. Biol. Chem. 259, 770-778. Spiess, M. and Lodish, H.F. (1985) Proc. Natl. Acad. Sci. USA 82, 6465-6469. Takahashi, H., Komano, H., Kawauchi, N., Kitamura, N., Nakanishi, S. and Natori, S. (1985) J. Biol. Chem. 260, 1222812233. Muramoto, K. and Kamiya, H. (1986) Biochim. Biophys. Acta 874, 285-295.
10 Giga, Y., Ikai, A. and Takahashi, K. (1987) J. Biol. Chem. 262, 6197-6203. 11 Muramoto, K., Ogata, K. and Kamiya, H. (1985) Agric. Biol. Chem. 49, 85-93. 12 Muramoto, K. and Kamiya, H. (1990) Biochim. Biophys. Acta 1039, 42-51. 13 Muramoto, K., Goto, R. and Kamiya, H. (1987) Anal. Biochem. 162, 435-442. 14 Grosse, E. (1967) Methods Enzymol. 11,238-255. 15 Muramoto, K., Kawauchi, H. and Tuzimura, K. (1978) Agric. Biol. Chem. 42, 1559-1563. 16 Laemmli, U.K. (1970) Nature 227, 680-685. 17 Knecht, R. and Chang, J.-Y. (1986) Anal. Chem. 58, 2375-2379.