ISSN 1068-1620, Russian Journal of Bioorganic Chemistry, 2007, Vol. 33, No. 1, pp. 170–174. © Pleiades Publishing, Inc., 2007. Original Russian Text © V.E. Piskarev, T.L. Bushueva, I.A. Yamskov, 2007, published in Bioorganicheskaya Khimiya, 2007, Vol. 33, No. 1, pp. 182–186.

BIOCHEMICAL STUDIES

Interaction of the Laburnum anagyroides Lectin with Fucoantigens V. E. Piskareva, 1, T. L. Bushuevab, and I. A. Yamskova a Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, ul. Vavilova 28, Moscow, 117813 Russia b

Institute of Experimental Cardiology, Russian Cardiological Center, Moscow, ul. Tret’ya Cherepkovskaya 15a, Moscow, 121552 Russia Received July 18, 2006; in final form, September 4, 2006

Abstract—We studied interaction of the lectin from the bark of Golden Rain shrub (Laburnum anagyroides, LABA) with a number of basic fucose-containing carbohydrate antigens by changes in its tryptophan fluorescence. The strongest LABA binding was observed for the trisaccharide H of type 6 [α-L-Fucp-(1–2)-β-D-Galp-(1–4)-D-Glc, Ka = 4.2 × 103 å–1]. The following antigens were bound with a weaker affinity: H-disaccharide α-L-Fucp-(1–2)-D-Gal, a glucoanalogue of tetrasaccharide Ley α-L-Fucp-(1–2)-β-D-Galp-(1–4)-[α-L-Fucp-(1–3)]-D-Glc, and 6-fucosyl-Nacetylglucosamine, a fragment of core of the N-glycans family (Ka 1.1–1.7 × 103 M–1). The lowest binding was observed for L-fucose (Ka = 2.7 × 102 M–1) and trisaccharide Lea, (β-Galp-(1–3)-[α-L-Fucp-(1–4)]-GlcNAc (Ka = 6.4 × 102 M–1). The Led, Lea, and Lex pentasaccharides and Leb hexasaccharide were not bound to LABA. Key words: Laburnum anagyroides lectin, carbohydrate specificity; fucoantigens DOI: 10.1134/S1068162007010207

INTRODUCTION L-Fucose is widely distributed in animal and plant word and among the lower organisms.2 It occurs in free state, within glycosides and oligosaccharides, and more often is involved into various hetero- and polysaccharides and glycoconjugates [1]. L-Fucose in an important component of many carbohydrate antigens, in particular, the group-specific blood antigens ABH(O) and the Lewis antigens Lea, Lex, Leb, and Ley. They are recognized by the fucose-specific lectins (fucolectins). Nowadays, the fine carbohydrate specificity of several fucolectins, including the following plant, fungal, fish, and bacterial lectins: the Ulex europaeus lectin of the ordinary furze [2], the Tetragonolobus purpureus (Lotus tetragonolobus) lectin of asparagus [3], the Aleuria aurantia fungal lectin [4], lectin of the Anguilla anguilla plasma of eel, lectin of hard-roe of perch Perca fluviatilis [6], and the bacterial lectin of Pseudomonas aeruginosa [7]. The search for new fucolectins and their study are of interest, because fucoantigens participate in many pathological processes [8]. Previously [9], we have studied the specificity of the LABA fucolectin from the Golden Rain (Laburnum 1

Corresponding author; phone/fax: +7 (495) 135-9375; e-mail: [email protected]. 2 Abbreviations: LABA, the Laburnum anagyroides bark agglutinin; LTA, Lotus tetragonolobus agglutinin; and UEA I, the Ulex europaeus agglutinin I.

anagyroides) bark and determined relative inhibiting activities of a large group of fucooligosaccharides upon the interaction of LABA with the BSA-(2'-FL)20-Au15 complex of colloid gold and the neoglycoprotein with the H-type 2/6 activity. The presence of Trp residues in LABA follows from its fluorescence spectrum (data not given). Therefore, changes in the own fluorescence can be used for the study of LABA interaction with carbohydrates. In this paper, we report the determination of binding constants of LABA to a series of fucoantigens and L-fucose by the method of measurements of the own lectin fluorescence. RESULTS AND DISCUSSION A group of natural fucoantigens on the basis of lactose from human milk and synthetic fucoantigens (Table 1) were used for investigation of the carbohydrate specificity. The Ka values of the LABA interaction with fucooligosaccharides and L-fucose are given in Table 2. The studied carbohydrates can be divided into three groups according to their strength of binding to LABA. L-Fucose and Lea-3 are most weakly bound (Ka 2.7 × 102 å–1 and 6.4 × 102 å–1, respectively). 2-FG, DFL, and 6-FN belong to the second group according to the middle binding fastness to LABA (Ka 1.1–1.7 × 103 M−1). The strongest binding was observed for 2'-FL

170

INTERACTION OF THE LABURNUM ANAGYROIDES LECTIN

171

Table 1. Antigenic oligosaccharides Structure

Name

Abbreviation

Antigen

Galβ1–4Glc

Lactose

Lac

Fucα1

6-L-Fucosyl-N-acetyl-D-glucosamine

6-FN

2-L-Fucosyl-D-galactose

2-FG

H

2'-Fucosyllactose

2'-FL

H (type 6)

N-Acetyl-2'-fucosyl-β-lactosamine

2'-FL-C2

H (type 6)

N-Valeryl-2'-fucosyl-β-lactosamine

2'-FL-C5

H (type 6)

Lacto-N-fucopentaose I

LNFP I

H (type 1)

Lea -trisaccharide

Lea-3

Lea

Lacto-N-fucopentaose II

LNFP II

Lea

Lacto-N-fucopentaose III

LNFP III

Lex

Lacto-N-difucohexaose I

LND I

Leb

Type 6

6 GlcNAc Gal 2 Fucα1 Galβ1–4Glc 2 Fucα1 Galβ1–4Glcβ1-NH-CO-CH3 2 Fucα1 Galβ1–4Glcβ1-NH-CO-(CH2)3CH3 2 Fucα1 Galβ1–3GlcNAcβ1–3Galβ1–4Glc 2 Fucα1 Galβ1–3GlcNAc 4 Fucα1 Galβ1–3GlcNAcβ1–3Galβ1–4Glc 4 Fucα1 Galβ1–4GlcNAcβ1–3Galβ1–4Glc 3 Fucα1 Galβ1–3GlcNAcβ1–3Galβ1–4Glc 2 4 Fucα1

Fucα1

Galβ1–4Glc 2 3

Difucosyllactose

DFL

Fucα1 Fucα1 RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY

Vol. 33

No. 1

2007

Ley glucoanal

172

PISKAREV et al.

Table 2. Effect of the addition of various carbohydrates on the maximum change in the own fluorescence of LABA (dFmax) and the Ka binding constants dFmax

Ka × 10–3, M–1

Lac

–

–

LNFP I

–

–

LNFP II

–

–

LNFP III

–

–

LND I

–

–

L-Fuc

1.2

0.27 + –0.006

Lea-3

1.5

0.64 + –0.02

2-FG

1.7

1.29 + –0.02

DFL

1.7

1.13 + –0.02

6-FN

2.0

1.73 + –0.04

2'-FL-C2

2.5

2.84 + –0.06

2'-FL-C5

2.6

3.14 + –0.06

2'-FL

3.2

4.24 + –0.06

Carbohydrate

(Ka 4.2 × 103 å–1). Thus, according to its carbohydrate specificity, LABA belongs to the group of fucolectins that recognize the antigen of H type 2/6 and relative antigens: Ley (α-L-Fucp-(1–2)-β-D-Galp-(1–4)-[α-LFucp-(1–3)]-D-GlcNAc) and the glucoanalogue of Ley (DFL). The UEA I lectin is the most thoroughly studied member of this group [2]. Both UEA I and LABA also have the affinity for 6-FN; in addition, LABA is weakly bound to Le‡-3. The interaction strength of LABA with the fucoantigens can be described by the following sequence: H typ 2 > H type 6 > (Ley) > Ley glucoanalogue = 6-FN > L-Fuc = Le‡-3. The same regularity is observed for UEA-I; however, there is no information on Le‡-3. The interaction of UEA I with LNFP II is also absent. The Ka value for the binding of L-Fuc to UEA I is 3 × 103 å–1 [10], which is one order of magnitude higher than that for LABA (see Table 2). A rather strong binding of free Lfucose comparable with the interaction with fucoantigens is characteristic of other fucolectins (Ka = (1–3) × 103 å–1) [11]. We had previously found that Ka of the LABA interaction with p-nitrophenyl α-L-fucopyranoside is one

and a half orders of magnitude higher than that with L-fucose [9]. This fact pointed out to either the presence of a hydrophobic binding site near the carbohydrate site in LABA, which is characteristic of some plant lectins, or to the involving of a hydrophobic amino acid residue in the fucose-binding site, similar to Tyr in A. aurantia [13]. N-Acetyl and N-valeryl derivatives of β-glycosamine of disaccharide 2'-FL were synthesized to investigate the nature of this site. One can see from the picture that the binding of Nacyl-2'-FL-glycosamine to LABA is practically the same an even slightly decreased in comparison with the unmodified 2'-FL. We conclude that the hydrophobic binding site or the carbohydrate binding site of LABA involves an aromatic amino acid residue rather than an aliphatic residue. The dramatic increase in the binding of p-nitrophenyl α-L-fucopyranoside is associated with the π–π interaction of its p-nitrophenyl residue with a Tyr or Phe residue in LABA. EXPERIMENTAL We used in this study distilled solvents and reagents of reagent grade and analytical grade quality. LABA (Lectinotest, Ukraine) was purified by affinity chromatography. The gradient HPLC system involved two Gilson 305 pumps (France), a Rheodyne 7125 injector (United States), a Waters 405 refractometric detector (United States), an Uvicord S2 UV-detector (Pharmacia, Sweden), and a potentiometric Kipp and Zonen recorder (Netherlands). The reversed phase HPLC was carried out on a tandem of columns Separon RP 18 (5 µm, 8 × 250 mm). A Toyopearl HW-40F column (Japan, 5 × 95 cm) was used for the preparative gel chromatography. Salts, monosaccharides, and lactose were removed on a column (7.5 × 95 cm) packed with Sephadex G-15 (Pharmacia, Sweden). 1H NMR spectra were recorded on a Bruker WM500 spectrometer (Germany) with the working frequency of 500 MHz at 300°K using acetone as an internal standard (δ 2.225 ppm). The samples were prepared by double lyophilization from 99.9% D2O. Fluorescence spectra were recorded on a Shimadzu 5000 fluorimeter (Japan) with excitation at 295 nm and slots of 5 nm on both monochromators. The absorption spectra were measured on a Beckman DU-8 spectrometer (United States). Oligosaccharides used in this study (Table 1) were isolated from the human milk according to procedure [14] with insignificant changes. The milk was defatted, proteins were precipitated with acetone (1 : 1), and the precipitate was removed by centrifugation (3000 g, 30 min). The supernatant was tenfold concentrated and applied onto a column (7.5 × 90 cm) packed with Sephadex G-25 in water. The carbohydrate-containing fractions eluted before lactose and salts were lyophilized. Neutral oligosaccharides were separated from

RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY

Vol. 33

No. 1

2007

INTERACTION OF THE LABURNUM ANAGYROIDES LECTIN

with the ligand, all the sites have an equal affinity to the ligand, and the binding in one site exerts no influence on the binding in the other sites. Then, the number of existing sites α can be determined by the equation:

F, relative units 2.75 11–13

2.50

α = (F – F0)/(Fmax – F0),

2.25 2.00

where F0 is the fluorescence in the starting point of titration, Fmax is the fluorescence of protein completely saturated by the protein ligands and determined by extrapolation of the plot of F dependence on [S] to the area of infinite concentrations of ligand (carbohydrate). The concentration of free carbohydrate can be determined as:

7–10

1.75 1.50 6

1.25

1–5

1.00 0.75

173

C = [S] – [α][P], 0

10

20

30

40

where [S] is the concentration of added carbohydrate, and [P] is the total lectin concentration. The Ka was determined from the Scatchard plot (α/C – α).

50 [S], mM

Changes in the fluorescence of LABA after the addition of the following carbohydrates, points: (1–5) Lac, LNFP I–III, and LND; (6) L-Fuc; (7–10) 6-FN, 2-FG, Lea, and DFL; (11–13) 2'-FL-C2, 2'-FL-C5, and 2'-FL.

REFERENCES

the sialated oligosaccharides on a Dowex 1x8 ionexchanger (200–400 mesh, acetate form) eluted with water, concentrated, and fractionated by the gel chromatography on a Toyopearl HW-40(s) column (5 × 90 cm) in water as described in [15, 16]. The resulting fractions (L-F, L-2F, LNT-F, and LNT-2F) were repeatedly purified on the same column. Further, each fraction was fractionated by rpHPLC in water [17–20]. Fractions 2'-FL, DFL, and LND I were obtained from the fractions L-F, L-2F, and LNT-2F, respectively. The LNFP I fraction was separated into three fractions: LNFP I, LNFP II, and LNFP III. Each oligosaccharide was rechromatographed. Structures of the oligosaccharides were determined by 1H NMR [17, 18, 20]. The 2-FG and Le‡-3 oligosaccharides were synthesized according to the procedures in [21] and [22], respectively. N-Acetyl and N-valeryl derivatives of 2'-FL were prepared from β-glycosamine [23] by acylation with the corresponding anhydrides [24]. Interaction of LABA with the carbohydrates was studied according to changes in protein fluorescence after the addition of a carbohydrate solution to them. Small aliquots of a carbohydrate solution (200– 500 mM) were successively added to the LABA solution (0.02 mg/ml) in 0.01 M phosphate buffer (pH 7.2). The fluorescence spectrum was recorded after 30 min, and an integral intensity (F) was calculated with the correction of fluorescent background of the carbohydrate and the sample dilution. Quantum yield of the protein fluorescence increased when a lectin interacted with a carbohydrate (the figure). This fact can be used for the calculation of binding constants [25, 26]. We believed that the changes in fluorescence are directly proportional to the portion of existing sites occupied RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY

1. Vanhooren, P.T. and Vandamme, E.J., J. Chem. Technol. Biotechnol., 1999, vol. 74, pp. 479–497. 2. Matsumoto, I. and Osawa, T., Biochim. Biophys. Acta, 1969, vol. 194, pp. 180–189. 3. Kalb, A.J., Biochim. Biophys. Acta, 1968, vol. 168, pp. 532–536. 4. Kochibe, N. and Furukawa, K., Biochemistry, 1980, vol. 19, pp. 2841–2846. 5. Springer, G.F., Desai, P.R., and Kolecki, B.J., Biochemistry, 1964, vol. 3, pp. 1076–1085. 6. Piskarev, V.E., Bushueva, T.L., and Yamskov, I.A., Prikl. Biokhim. Mikrobiol., 2004, vol. 40, pp. 105–109. 7. Perret, S., Sabin, C., Dumon, C., Pokorna, M., Gautier, C., Galanina, O., Ilia, S., Bovin, N., Nicaise, M., Desmadril, M., Gilboa-Garber, N., Wimmerova, M., Mitchell, E.P., and Imberty, A., Biochem. J., 2005, vol. 389, pp. 325–332. 8. Listinsky, J., Siegal, G., and Listinsky, C., Am. J. Clin. Pathol., 1998, vol. 110, pp. 425–440. 9. Piskarev, V.E., Lutsik-Kordovskii, M.D., Piskareva, E.L., and Yamskov, I.A., Prikl. Biokhim. Mikrobiol., 2003, vol. 39, pp. 581–587. 10. Horejsi, V., Ticha, M., and Kocourek, J., Biochim. Biophys. Acta, 1977, vol. 499, pp. 290–300. 11. Pereira, M.E.A. and Kabat, E.A., Biochemistry, 1974, vol. 13, pp. 3184–3192. 12. Goldstein, I.J and Poretz, R.D, in The Lectins, Liener, I.E.., Sharon, N.., and Goldstein, I.J., Eds., Orlando: Academic, 1986, p. 47. 13. Amano, K., Fujihashi, M., Ando, A., Miki, K., and Nagata, Y., Boisci. Biotechnol. Biochem., 2004, vol. 68, pp. 841–847. 14. Kobata, A., Methods Enzymol., 1972, vol. 28, pp. 262– 268. 15. Thurl, S., Ottermanns, J., Muller-Werner, B., and Sawatzki, G., J. Chromatogr., 1991, vol. 568, pp. 291– 300.

Vol. 33

No. 1

2007

174

PISKAREV et al.

16. Stahl, B., Thurl, S., Zeng, J., Karas, M., Hillenkamp, F., Steup, M., and Sawatzki, G., Anal. Biochem., 1994, vol. 223, pp. 218–226. 17. Dua, V. and Bush, C., Anal. Biochem., 1983, vol. 133, pp. 1–8. 18. Cheetham, N.W.H. and Dube, V.E., J. Chromatogr., 1983, vol. 262, pp. 426–430. 19. Dua, V.K. and Bush, C.A., J. Chromatogr., 1985, vol. 328, pp. 259–269. 20. Sabharval, H., Nilsson, B., Lindh, F., Gronberg, G., Sjoblad, S., and Lundblad, A., Carbohydr. Res., 1988, vol. 178, pp. 145–154.

21. Bahl, O., Methods Enzymol., 1972, vol. 28, pp. 738–743. 22. Rana, S.S. and Matta, K.L., Carbohydr. Res., 1983, vol. 117, pp. 101–111. 23. Likhosherstov, L.M., Novikova, O.S., Derevitskaya, V.A., and Kochetkov, N.K., Carbohydr. Res., 1986, vol. 146. 24. Likhosherstov, L.M., Novikova, O.S., Shibaev, V.N., and Kochetkov, N.K., Izv. Ross. Akad. Nauk, Ser. Khim., 1996, no. 7, pp. 1848–1851. 25. Lee, Y.C., J. Biochem., 1997, vol. 121, pp. 818–825. 26. Jolley, M.E. and Glaudemans, C.P.J., Carbohydr. Res., 1974, vol. 33, pp. 377–382.

RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY

Vol. 33

No. 1

2007