ガレクチン9・分枝糖鎖複合体およびシアル化糖鎖 …xraylab/report/poster/...crd,...

3
W68 +1 +2 +3 +4 +5 S4 S5 loop H52 G53 +1 +2 +3 はじめに Introduction ガレクチンとは,細胞表面上の糖鎖を認識し結合する動物レクチンタンパク質で,近年,生 体内で様々な興味深い機能を持つことが明らかとなり,注目されている。これまでにヒトガレク チンは14種類発見されており,その構造からプロトタイプ,キメラタイプ,タンデムリピートタイ プに分けられる。ガレクチン9は,タンデムリピートタイプに属し,免疫応答において,好酸球遊 走因子や活性化T細胞のアポトーシスを誘導因子等として働いていることが報告されている。 ガレクチン9は,枝分かれ糖鎖(分枝糖鎖)およびシアル化糖鎖に対して親和性が高い。ガレ クチン9の糖鎖認識機構と,それが持つ機能との関係について新たな知見を得るため,ガレク チン9・糖鎖複合体の立体構造をX線結晶解析により決定した。 The galectins are a family of -galactoside-specific animal lectins which contain conserved elements for carbohydrate recognition, and have attracted much attention as novel regulators of physiological systems. Currently, there are 14 members of the mammalian galectin family, classified into three subtypes on the basis of structure; the prototype , the chimera-type, and the tandem-repeat-type galectins. Human galectin-9, having high affinity for N-glycan-type oligosaccharides with branches and sialylated oligosaccharides, is involved in eosinophil chemoattraction and apoptosis of T helper type 1 cells, in the immune sysytem. To elucidate this unique feature, X-ray structures of human galectin-9 C-terminal domain in complexes with the bianntenary pyridylaminated oligosaccharide and 2-3 sialyllactose were determined. ガレクチン9・糖鎖複合体の立体構造 Overall structure of galectin-9 / oligosaccharide complex O H R N H H OH H H H COOH CH3CO O O CH2OH H H H OH OH H H OH O CH2OH H OH H H OH H H O OH H OH H CH2OH R = N O CH2OH H H H OH NHCOCH3 H H O CH2OH H OH H H OH OH H H O O O CH2OH H H OH H OH H H O O CH2OH H H H OH NHCOCH3 H OH O H H OH H H OH H O CH2 O CH2OH H H H OH NHCOCH3 H H O CH2OH H OH H H OH OH H H O O O CH2OH H H OH H OH H H O C H2 OH CH2OH H H H OH NHCOCH3 H O N H SiaLac BIPA Gal +1 Glc +2 Sia 1 Gal +1 GlcNAc +2 GlcNAc +5 GlcNAc +6 Man +3 Man +4 14) 23) 14) 12) 14) 14) 16) 13) Prototype Tandem-repeat-type Chimera-type Galectin1, 2, 5, 7, 10, 11, 13, 14 & 15 galectin4, 6, 8, 9, & 12 Galectin3 1ガレクチンの3つのタイプの構造(左)と枝分かれ糖鎖(BIPA)およびシアリルラクトース SiaLac)の化学構造式(右)。単糖ユニットの名称とグルコシド結合も示す。 Fig. 1. Three subtypes of galectin family (left), and chemical structures of the bianntenary pyridylaminated oligosaccharide (BIPA) and 2-3 sialyllactose (SiaLac). The names of sugar units and glucoside bonds are indicated (right). H223 N225 N237 R239 E258 E242 W255 H235 N248 R221 D241 H223 N225 N237 R239 E258 E242 W255 H235 N248 R221 Gal +1 GlcNAc +2 Man +3 Gal +1 Glc +2 Sia 1 B Galectin-9 C-CRD / SiaLac A Galectin-9 C-CRD / BIPA W255 S4 S5 loop D241 E242 +1 +2 +3 +4 +5 +1 +2 +3 Branched oligosaccharide (N-glycans) are generally thought to be in dynamic motion in vitro and in vivo. Galectin9 Galectin1 Conf. 1 Conf. 2 Conf. 3 Conf. 4 Unusual short contacts! NO! YES! YES! YES! YES! NO! NO! NO! Molecules are loosely packed. <<< 枝分かれ糖鎖認識機構 Recognition mechanism of branched oligosaccharide 2ヒトガレクチン9N末側糖鎖認識ドメイン(N-CRD,左,シアン色)とC末側糖鎖認識ドメイン(C-CRD右,ピンク色)の立体構造をそれぞれ独立に決定した。 結晶中の分子パッキング状態により,C-CRD の方が糖鎖複合体を作成するのに有利であることがわかった。 Fig. 2. The X-ray structures of human galectin-9 N-terminal carbohydrate recognition domain (N- CRD, left, cyan) and C-terminal carbohydrate recognition domain (C-CRD, right, pink) are independently determined. In a crystal of C-CRD, molecules are loosely packed, being favorable to prepare the complexes with oligosaccharides. Molecules are tightly packed. N-CRD C-CRD 図3.ヒトガレクチン9C-CRDの糖鎖結合部位に,枝分かれ糖鎖(BIPAの枝の先端部分)(A),およびシ アリルラクトース(SiaLac)(B)が結合した構造を示す。ガレクチン9は,多くの水素結合により糖鎖を認 識している。SiaLacを認識するためにArg221がコンフォメーションを変えているのがわかる。 Fig. 3. Carbohydrate-binding sites of human galectin-9 C-CRD with bound BIPA (A) and SiaLac (B). Gal +1 , GlcNAc +2 and Man +3 of BIPA are visible in the electron density. Many hydrogen bond interactions are between the protein and the oligosaccharide. Depending on the binding of SiaLac, Arg221 changes its conformation to recognize a Sia -1 moiety. 4結晶中の分子パッキングからヒトガレクチン9C-CRDは枝分かれ糖鎖によって架橋されていること がわかった。枝分かれ部分(青色)をモデリングにより補った構造を示す(A)。ウシガレクチン1において も,よく似た構造が報告されているおり(Bourne, Y. et al., (1994) Nat. Struct. Biol. 1, 863–870),糖 鎖との相互作用がガレクチン9よりも多くなっている(B)。 Fig. 4. Molecular packing in a crystal indicates that galectin-9 C-CRDs are cross-linked by BIPA. Also bovine galectin-1 was reported to be cross-linked by the biantennary oligosaccharide (Bourne, Y. et al., (1994) Nat. Struct. Biol. 1, 863–870). The crosslinking structures of human galectin-9 C-CRD (A) and bovine galectin-1 (B) are shown. In human galectin-9 C-CRD, the modelled sugar are shown in blue. There are many protein-oligosaccharide interactions in bovine galectin-1 than human galectin-9 C-CRD. A Galectin-9 C-CRD / BIPA B Galectin-1 / bianntenary oligosaccharide (PDB code: 1SLA) 5ガレクチン9が枝分かれ糖鎖に対して高い親和性を持つ機構を模式的に示す。生体内では,枝分か れ糖鎖は激しく動いていて,様々なコンフォメーションをとっている。ガレクチン1(緑色)は,タンパク質と 多くの相互作用を形成するため,Conf.3にしかに結合できない。ガレクチン9(ピンク色)は,タンパク質 との相互作用は少ないが,あらゆる構造の分子糖鎖に結合できる(Conf.1, Conf.2, Conf.3すべて)。 ガレクチン9の様々な構造の分枝糖鎖と結合できる能力は,糖鎖との相互作用の少なさを補って余りあ るので,ガレクチン9の方がガレクチン1より枝分かれ糖鎖に対する親和性が高いと考えられる。 Fig. 5. A schematic view of the recognition of a branched oligosaccharide.Galectin-1 having a deep carbohydrate-binding site can form a stable protein-ligand complex of low structural energy. However, the conformation of the bound branched oligosaccharide may be restricted by strong protein-ligand interactions. In the case of galectin-9, protein-ligand interactions are limited to Gal +1 and GlcNAc +2 , and other sugar units are free from the protein, meaning that galectin-9 can recognize the antennae of a branched oligosaccharide in various conformations. In galectin-9, the less structural energy in the formation of a protein-ligand complex is probably compensated for by the ability to accept branched oligosaccharide in various conformations. Yoshida, H., Teraoka, M., Nishi, N., Nakakita, S., Nakamura, T., Hirashima, M. & Kamitori, S. (2010). Xray structures of human galectin9Cterminal domain in complexes with a biantennary oligosaccharide and sialyllactose. J. Biol. Chem., 285, 3696936976. ガレクチン9・分枝糖鎖複合体およびシアル化糖鎖複合体のX線結晶解析 吉田裕美 1 , 寺岡美沙 1 , 西 1 , 中北愼一 2 , 中村隆範 2 , 平島光臣 3 , 神鳥成弘 1 1 香川大学総合生命科学研究センター, 2 医学部分子細胞機能学, 3 医学部免疫病理学 X-RAY STRUCTURES OF HUMAN GALECTIN-9 C-TERMINAL DOMAIN IN COMPLEXES WITH A BIANTENNARY OLIGOSACCHARIDE AND SIALYLLACTOSE Hiromi Yoshida 1 , Misa Teraoka 1 , Nozomu Nishi 1 , Shin-ichi Nakakita 1 , Takanori Nakamura 2 , Mitsuomi Hirashima 3 and Shigehiro Kamitori 1 Life Science Research Center 1 , Department of Endocrinology 2 , and Departments of Immunology and Immunopathology 3 , Faculty of Medicine, Kagawa University

Upload: vudieu

Post on 22-Apr-2018

229 views

Category:

Documents


4 download

TRANSCRIPT

Page 1: ガレクチン9・分枝糖鎖複合体およびシアル化糖鎖 …xraylab/report/poster/...CRD, left, cyan) and C-terminal carbohydrate recognition domain (C-CRD, right, pink)

W68

+1 +2+3

+4+5

S4 ‐ S5 loopH52 ‐ G53

+1+2+3

はじめに Introduction

ガレクチンとは,細胞表面上の糖鎖を認識し結合する動物レクチンタンパク質で,近年,生体内で様々な興味深い機能を持つことが明らかとなり,注目されている。これまでにヒトガレクチンは14種類発見されており,その構造からプロトタイプ,キメラタイプ,タンデムリピートタイプに分けられる。ガレクチン9は,タンデムリピートタイプに属し,免疫応答において,好酸球遊走因子や活性化T細胞のアポトーシスを誘導因子等として働いていることが報告されている。ガレクチン9は,枝分かれ糖鎖(分枝糖鎖)およびシアル化糖鎖に対して親和性が高い。ガレクチン9の糖鎖認識機構と,それが持つ機能との関係について新たな知見を得るため,ガレクチン9・糖鎖複合体の立体構造をX線結晶解析により決定した。

The galectins are a family of -galactoside-specific animal lectins which containconserved elements for carbohydrate recognition, and have attracted much attention asnovel regulators of physiological systems. Currently, there are 14 members of themammalian galectin family, classified into three subtypes on the basis of structure; theprototype , the chimera-type, and the tandem-repeat-type galectins. Human galectin-9,having high affinity for N-glycan-type oligosaccharides with branches and sialylatedoligosaccharides, is involved in eosinophil chemoattraction and apoptosis of T helpertype 1 cells, in the immune sysytem. To elucidate this unique feature, X-ray structuresof human galectin-9 C-terminal domain in complexes with the bianntenarypyridylaminated oligosaccharide and 2-3 sialyllactose were determined.

ガレクチン9・糖鎖複合体の立体構造 Overall structure of galectin-9 / oligosaccharide complex

O

H

RNH

H

OH

H

H

H

COOH

CH3CO

O

O

CH2OH

HH

H

OH

OH

HH

OHO

CH2OH

HOH

HH OH

HH

OOHH

OHH

CH2OH

R =

N

O

CH2OH

HH

H

OH

NHCOCH3

HH

O

CH2OH

HOH

H

H

OH

OH

HH

O

O O

CH2OH

HH

OH

H

OH

H

H

O

O

CH2OH

HH

H

OH

NHCOCH3

HOH

OH

H

OHH H

OH

H

O

CH2

O

CH2OH

HH

H

OH

NHCOCH3

HH

O

CH2OH

HOH

H

H

OH

OH

HH

O

O O

CH2OH

HH

OH

H

OH

H

H

O

CH

2

OH

CH2OH

HH

H

OH

NHCOCH3

HO N

H

SiaLac

BIPA

Gal+1 Glc+2Sia‐1

Gal+1 GlcNAc+2

GlcNAc+5 GlcNAc+6

Man+3

Man+4

1‐4)2‐3)

1‐4) 1‐2)

1‐4) 1‐4)

1‐6)

1‐3)

Prototype

Tandem-repeat-type

Chimera-type

Galectin‐1, ‐2, ‐5, ‐7, ‐10, ‐11, ‐13, ‐14 & ‐15

galectin‐4, ‐6, ‐8, ‐9, & ‐12

Galectin‐3

図1.ガレクチンの3つのタイプの構造(左)と枝分かれ糖鎖(BIPA)およびシアリルラクトース(SiaLac)の化学構造式(右)。単糖ユニットの名称とグルコシド結合も示す。

Fig. 1. Three subtypes of galectin family (left), and chemical structures of thebianntenary pyridylaminated oligosaccharide (BIPA) and 2-3 sialyllactose (SiaLac).The names of sugar units and glucoside bonds are indicated (right).

H223

N225

N237

R239

E258

E242

W255

H235

N248

R221

D241

H223

N225

N237

R239

E258

E242

W255

H235N248

R221

Gal+1

GlcNAc+2 Man+3

Gal+1

Glc+2

Sia‐1

B Galectin-9 C-CRD / SiaLacA Galectin-9 C-CRD / BIPA

W255

S4 ‐ S5 loopD241 ‐ E242

+1 +2+3

+4

+5

+1

+2+3

Branched oligosaccharide (N-glycans) are generally thought to be in dynamic motion in vitro and in vivo.

Galectin‐9 Galectin‐1

Conf. 1 Conf. 2 Conf. 3 Conf. 4Unusual short contacts!

NO! YES!

YES! YES! YES!

NO!NO!

NO!

Molecules are loosely packed.

<<<

枝分かれ糖鎖認識機構 Recognition mechanism of branched oligosaccharide

図2.ヒトガレクチン9のN末側糖鎖認識ドメイン(N-CRD,左,シアン色)とC末側糖鎖認識ドメイン(C-CRD,右,ピンク色)の立体構造をそれぞれ独立に決定した。 結晶中の分子パッキング状態により,C-CRDの方が糖鎖複合体を作成するのに有利であることがわかった。

Fig. 2. The X-ray structures of human galectin-9 N-terminal carbohydrate recognition domain (N-CRD, left, cyan) and C-terminal carbohydrate recognition domain (C-CRD, right, pink) areindependently determined. In a crystal of C-CRD, molecules are loosely packed, being favorableto prepare the complexes with oligosaccharides.

Molecules are tightly packed.

N-CRD C-CRD

図3.ヒトガレクチン9のC-CRDの糖鎖結合部位に,枝分かれ糖鎖(BIPAの枝の先端部分)(A),およびシアリルラクトース(SiaLac)(B)が結合した構造を示す。ガレクチン9は,多くの水素結合により糖鎖を認識している。SiaLacを認識するためにArg221がコンフォメーションを変えているのがわかる。

Fig. 3. Carbohydrate-binding sites of human galectin-9 C-CRD with bound BIPA (A) and SiaLac (B).Gal+1, GlcNAc+2 and Man+3 of BIPA are visible in the electron density. Many hydrogen bondinteractions are between the protein and the oligosaccharide. Depending on the binding ofSiaLac, Arg221 changes its conformation to recognize a Sia-1 moiety.

図4.結晶中の分子パッキングからヒトガレクチン9のC-CRDは枝分かれ糖鎖によって架橋されていることがわかった。枝分かれ部分(青色)をモデリングにより補った構造を示す(A)。ウシガレクチン1においても,よく似た構造が報告されているおり(Bourne, Y. et al., (1994) Nat. Struct. Biol. 1, 863–870),糖鎖との相互作用がガレクチン9よりも多くなっている(B)。

Fig. 4. Molecular packing in a crystal indicates that galectin-9 C-CRDs are cross-linked by BIPA.Also bovine galectin-1 was reported to be cross-linked by the biantennary oligosaccharide(Bourne, Y. et al., (1994) Nat. Struct. Biol. 1, 863–870). The crosslinking structures of humangalectin-9 C-CRD (A) and bovine galectin-1 (B) are shown. In human galectin-9 C-CRD, themodelled sugar are shown in blue. There are many protein-oligosaccharide interactions in bovinegalectin-1 than human galectin-9 C-CRD.

A Galectin-9 C-CRD / BIPA

B Galectin-1 / bianntenary oligosaccharide(PDB code: 1SLA)

図5.ガレクチン9が枝分かれ糖鎖に対して高い親和性を持つ機構を模式的に示す。生体内では,枝分かれ糖鎖は激しく動いていて,様々なコンフォメーションをとっている。ガレクチン1(緑色)は,タンパク質と多くの相互作用を形成するため,Conf.3にしかに結合できない。ガレクチン9(ピンク色)は,タンパク質との相互作用は少ないが,あらゆる構造の分子糖鎖に結合できる(Conf.1, Conf.2, Conf.3すべて)。ガレクチン9の様々な構造の分枝糖鎖と結合できる能力は,糖鎖との相互作用の少なさを補って余りあるので,ガレクチン9の方がガレクチン1より枝分かれ糖鎖に対する親和性が高いと考えられる。

Fig. 5. A schematic view of the recognition of a branched oligosaccharide.Galectin-1 having a deepcarbohydrate-binding site can form a stable protein-ligand complex of low structural energy.However, the conformation of the bound branched oligosaccharide may be restricted by strongprotein-ligand interactions. In the case of galectin-9, protein-ligand interactions are limited toGal+1 and GlcNAc+2, and other sugar units are free from the protein, meaning that galectin-9 canrecognize the antennae of a branched oligosaccharide in various conformations. In galectin-9,the less structural energy in the formation of a protein-ligand complex is probably compensatedfor by the ability to accept branched oligosaccharide in various conformations.

Yoshida, H., Teraoka, M., Nishi, N., Nakakita, S., Nakamura, T., Hirashima, M. & Kamitori, S. (2010).   X‐ray structures of human galectin‐9 C‐terminal domain in complexes with a biantennary oligosaccharide and sialyllactose. J. Biol. Chem., 285, 36969‐36976.

ガレクチン9・分枝糖鎖複合体およびシアル化糖鎖複合体のX線結晶解析

吉田裕美1, 寺岡美沙1, 西 望1, 中北愼一2, 中村隆範2, 平島光臣3, 神鳥成弘1

1 香川大学総合生命科学研究センター, 2 医学部分子細胞機能学, 3医学部免疫病理学

X-RAY STRUCTURES OF HUMAN GALECTIN-9 C-TERMINAL DOMAIN IN COMPLEXES WITH A BIANTENNARY OLIGOSACCHARIDE AND SIALYLLACTOSE

Hiromi Yoshida1, Misa Teraoka1, Nozomu Nishi1, Shin-ichi Nakakita1, Takanori Nakamura2, Mitsuomi Hirashima3 and Shigehiro Kamitori1

Life Science Research Center1, Department of Endocrinology2, and Departments of Immunology and Immunopathology3, Faculty of Medicine, Kagawa University

Page 2: ガレクチン9・分枝糖鎖複合体およびシアル化糖鎖 …xraylab/report/poster/...CRD, left, cyan) and C-terminal carbohydrate recognition domain (C-CRD, right, pink)

2つの糖鎖認識ドメインを持つ変異型ガレクチン8のX線結晶解析吉田裕美,山下 哲,寺岡美沙,伊藤愛子,中北愼一,西 望,神鳥成弘

香川大学総合生命科学研究センター

X-RAY STRUCTURE OF A PROTEASE-RESISTANCE MUTANT FORM OF HUMAN GALECTIN-8 WITH TWO CARBOHYDRATE RECOGNITION DOMAINS

Hiromi Yoshida, Satoshi Yamashita, Misa Teraoka, Aiko Itoh, Shin-ichi Nakakita, Nozomu Nishi, Shigehiro KamitoriLife Science Research Center and Faculty of Medicine, Kagawa University

要旨 Summaryガレクチン8は,2つの異なる糖鎖認識ドメインをN末側(N-CRD),C末側(C-CRD)に持つタンデムリピートタイプのガレクチ

ンで,細胞マトリックス相互作用や細胞接着に関与している。N-CRDは2-3シアル化糖鎖に強い親和性を持ち,C-CRDはN-CRDより糖鎖親和性が低い。これまでに多くのガレクチンのX線結晶解析が報告されているが,それらはすべて1つのCRDを対象としたもので,2つのCRDを含むX線構造はまだない。これは2つのCRDをつなぐリンカー部分のプロテアーゼ感受性が高いためである。今回,2つのCRDを持つプロテアーゼ耐性変異型ガレクチン8(G8Null)と糖鎖との複合体のX線解析に成功した。また,疑似2量体を形成したガレクチン8のN-CRD(G8N)についてもX線結晶解析を行った。これらの結果より,2つのCRDの糖鎖認識の違いを明らかにし,さらにガレクチン8のCRDおよび分子の集合状態についての新たな知見を得た。

Galectin-8 is a tandem-repeat-type -galactoside-specific animal lectin having N- and C-terminal carbohydraterecognition domains (N-CRD and C-CRD, respectively) with a difference in carbohydrate-binding specificity, involvedin cell-matrix interaction and cell adhesion. N-CRD exhibits strong affinity for 2-3 sialylated oligosaccharides, andC-CRD usually exhibits relatively lower affinity for oligosaccharides than does N-CRD. There have been manystructural studies on galectins with a single CRD, but no X-ray structure of a galectin containing both CRDs has beenreported, because tandem-repeat-type galectins are susceptible to proteolysis due to the presence of a linkerpeptide between two CRDs. Here, the X-ray structure of a protease-resistant mutant form of human galectin-8(G8Null) having both CRDs and the novel pseudo-dimer structure of galectin-8 N-CRD (G8N) in complexes with 2-3sialylated oligosaccharide ligands were determined. The results revealed a difference in specificity between the N-and C-CRDs, and provided new insights into the association of CRDs and/or molecules of galectin-8.

糖鎖認識機構と分子集合状態 Carbohydrate recognition mechanism and multiple association modes

図2.(A)G8Nullの折れ曲がり構造。N-CRD(青色)とC-CRD(緑色)の軸はほぼ平行であるが面は直交している。 (B)プロペラ(2枚羽)型疑似2量体の構造。 (C)G8NullにおけるCRD間(S0とS1間)の相互作用。(D)G8N疑似2量体における分子間(F0間)の相互作用。

Fig. 2. (A) Bend-shaped structure of G8Null with the bound SiaLac and Lac. N-CRD, C-CRD, and thelinker (His-Met) are indicated by blue, green, and red, respectively. (B) Propeller-shaped structure ofthe dimer of G8N. (C) Interactions between S0 and S1 in G8Null with hydrogen bonds shown bydotted lines. (D) Interactions between F0 -strands (Mol-A and Mol-B) in G8N with hydrogen bonds.

図3.(A)結晶中で2分子のG8Nullによって形成される2量体。 (B)2量体を2回対称軸方向から見る。結晶中では,さらにN-CRD同士が相互作用している(赤点線の楕円内)。 (C)2量体における分子間の塩橋および水素結合。 (D)N-CRD間(F0とS6b’間)の水素結合。

Fig. 3. (A) The dimer formed by two molecules of G8Null. (B) Dimer viewed down the 2-fold symmetryaxis with symmetry operated N-CRDs. For clarity, symmetry operated C-CRDs are not shown. Theinterface between N-CRDs is indicated by red dotted ellipses. (C) Salt-bridges and hydrogen bondsat the interface of the dimer. (D) Hydrogen bonds between the F0 and S6b’ of N-CRDs.

図5.予想されるガレクチン8のCRDおよび分子の集合状態。N-CRD(グレイ色),C-CRD(白色)を正方柱で表す。楕円は糖鎖結合部位である。溶液中において,ガレクチン8の2つのCRDは動いており,近づいたり離れたりしていると考えられる(i および ii)。2つのCRDが十分近づいた時,N-CRD(S0)とC-CRD(S1)との間で水素結合が形成され,折れ曲がった構造になる(iii)。この構造は,部分的に2量体を形成することができ(iv),さらに高分子状の格子構造を形成するかもしれない(v)。一方,もし,十分にC-CRDから離れたN-CRD同士が近づくとプロペラ型疑似2量体を部分的に形成できるのかもしれない(vi)。

Fig. 5. The putative multiple association modes of domains and molecules of galectin-8. N-CRD and C-CRD are shown as gray and white tetragonal prisms, respectively, with an ellipse as thecarbohydrate-binding site. Galectin-8 should be in a dynamic state in solution, the two domainsmoving toward or away from each other (i and ii). When the two domains are close enough, thebend-shaped structure likely forms via specific hydrogen bonds between N-CRD (S0) and C-CRD(S1) (iii). A bend-shaped galectin-8 could partially form a dimer in solution (iv). Moreover, G8Nullcould builds a polymeric lattice through inter-dimeric interactions between F0 and S6b’ (v). If N-CRDsisolated from a C-CRD approach each other, the propeller-shaped pseudo-dimer of N-CRD mightpartially form (vi).

F1F2

F3

F4F5S1 S3 S4

S5S2

S6a

S6b

F0

F1

F2F3F4F5

S5S4

S3S2S6a

S1

S0S6b

LacSiaLac

N-CRD axis

mol-y

mol-x

C-CRD axis

W86

R59

Q47

R45N67

R69

E89

N79

K71

GlcNAc+2

Gal+1

Sia-1

H65

R72

W249

N215

S213

N231 R233

E252N242

N235

H229

I 236

Rotation by 90°

(i)

(ii)(vi)

(iii)

(iv)

(v)

(iii)’

(F1-F5) b-sheetN-CRD C-CRD

Carbohydrate-binding site Bend-shaped structurePropeller-shaped dimer

Dimer

Polymeric lattice

F0

F1 F2

F3 F4 F5

S1 S2 S3 S4 S5

S6aF0

F1

F2F3F4F5

S1S2

S3S4S5

S6a

S6b

SiaLacNAc

SiaLacNAc

S6b

mol-y

mol-x

Mol-B axis

O

CH2OH

HH

H

OH

OH

HOH

HO

CH2OH

HOH

H

H

OH

OH

HH

O

O

H

RNH

H

OH

H

H

H

COOH

CH3CO

O

O

CH2OH

HH

H

OH

OH (NHCOCH3)

HH

OHO

CH2OH

HOH

H

H OH

HH

O

S1 F2 S3 S4 S5

S6a S6b F3 F4 F5 H1 S2 F1

F0

Lac SiaLac (SiaLacNAc)

Gal +1 GlcNAc+2 Gal +1 Glc+2 (GlcNAc+2)Sia‒1

OHH

OHH

CH2OH

R =

G8M(N-CRD) 1 MMLSLNNLQNIIYNPVIPFVGTIPDQLDPGTLIVICGHVPS-DADRFQVDLQNGSSMKPRADVAFHFNPRFKRAGCIVCNTLINEG8M(C-CRD) 184 ---------------RLPFAARLNTPMGPGRTVVVKGEVNA-NAKSFNVDLLAGKS----KDIALHLNPRLNIK-AFVRNSFLQEG9S(N-CRD) 1 --MAFSGSQAPYLSPAVPFSGTIQGGLQDGLQITVNGTVLSSSGTRFAVNFQTGF---SGNDIAFHFNPRFEDGGYVVCNTRQNGG9S(C-CRD) 188 -----------HPAYPMPFITTILGGLYPSKSILLSGTVLPS-AQRFHINLCSG------NHIAFHLNPRFDEN-AVVRNTQIDN

G8M(N-CRD) 85 KWGREEITY--DTPFKREKSFEIVIMVLKDKFQVAVNGKHTLLYGHRIG-PEKIDTLGIYGKVNIHSIGFSFSS 155G8M(C-CRD) 248 SWGEEERNI-TSFPFSPGMYFEMIIYCDVREFKVAVNGVHSLEYKHRFKELSSIDTLEINGDIHLLEVRSW--- 317G9S(N-CRD) 81 SWGPEERKT--HMPFQKGMPFDLCFLVQSSDFKVMVNGILFVQYFHRV-PFHRVDTISVNGSVQLSYISFQ--- 149 G9S(N-CRD) 254 SWGSEERSLPRKMPFVRGQSFSVWILCEAHCLKVAVDGQHLFEYYHRLRNLPTINRLEVGGDIQLTHVQT---- 323

(Linker)G8M 156 DLQSTQASSLELTEISRENVPKSGTPQL 183G9S 150 PPGVWPANPAPITQTVIHTVQSAPGQMFSTPAIPPMMYP 187G8Null 156 HM

G8M 1 156 184 317N- N-CRD Linker C-CRD -C

G9S 1 150 188 323

図1.(A)ラクトース(Lac),2-3シアルラクトース( SiaLac ) , 2-3 シ ア ル ラ ク ト サ ミ ン(SiaLacNAc)の化学構造式を示す。(B)ガレクチン8,ガレクチン9のN-CRD,C-CRDのアミノ酸配列アライメントを示す。

Fig. 1. (A) Chemical structure of lactose (Lac),2-3 sialyllactose (SiaLac), and 2-3sialyllactosamine (SiaLacNAc). (B) Aminoacid alignment of N- and C-CRD of thehuman galectin-8 and -9 with secondaryelements.

L3 S4

L5

N7

N6

A188

A189

R190

L191

N192

S0

S1

V16

L8’

Q9’

N10’

I11’

I12’

Y13’

Y13I12

I11

N14

P15

Mol-B Mol-A

Mol-A axisC-CRD axis

S6b‘

Mol-A

Mol-B

Q115’

Q115

Y266 K204 R315E313E268

R315’E313’K204’

Y266’

E268’

N14

I11

I12

Y13

Y93’

E90’

I91’

T92’

F0

S6b‘

F0

A

BC

D

A

B

C

D

A B

Galectin-8・糖鎖複合体の立体構造 X-ray structure of galectin-8/oligosaccharide complex

A

B

図4.(A)G8Nに結合したSiaLacNac。ガレクチン8特有の長いS3-S4ループ(赤色)内にあるArg59がシアル酸部分のカルボキシル基と塩橋を形成している。Arg45とGln47 もシアル酸部分の認識に関わっている。 (B)G8NullのC-CRDに結合したLac。C-CRDには,シアル酸部分が入るスペースはあるが,認識するアミノ酸が存在しない。

Fig. 4. (A) The bound SiaLacNAc in G8N of G8N-SiaLacNAc with hydrogen bonds. Arg59 in the S3-S4 loop (red) efficiently forms salt-bredge interactions with a carboxyl group of Sia-1, and Arg45and Gln47 also contribute to the recognition of Sia-1. (B) The bound Lac in C-CRD of G8Null.Although C-CRD has a large space for Sia-1, there is no amino acid residue for recognizing Sia-1.

Glc+2Gal+1

Page 3: ガレクチン9・分枝糖鎖複合体およびシアル化糖鎖 …xraylab/report/poster/...CRD, left, cyan) and C-terminal carbohydrate recognition domain (C-CRD, right, pink)

立体構造解析に基づくボツリヌス菌ヘマグルチニンHA70の糖鎖認識反応機構

山下 哲1,吉田裕美1, 内山 昇2, 中北ゆかり1, 中北愼一1,  殿塚隆史3,小熊惠二4,西河 淳3,神鳥成弘1

1香川大学総合生命科学研究センター, 2㈱レクザム香川工場, 3東京農工大学農学部,4岡山大学医学部

CARBOHYDROTE RECOGNITION MECHANISM OF HA70 FROM CLOSTRIDIUM BOTULINUM DEDUCED FROM X-RAY STRUCTURES IN COMPLEXES WITH SIALYLATED OLIGOSACCHARIDES

Satoshi Yamashita1, Hiromi Yoshida1, Noboru Uchiyama2, Yukari Nakakita1, Shin-ichi Nakakita1, Takashi Tonozuka3, Keiji Oguma4, Atsushi Nishikawa3, Shigehiro Kamitori 1

1Life Science Research Center and Faculty of Medicine, Kagawa University, 2Rexxam Co., Ltd., Kagawa Factory3Tokyo University of Agriculture and Technology, 4Okayama University Graduate of Medicine, Dentistry and Pharmaceutical Science

要旨 Summaryボツリヌス菌が産生する神経毒素は,非毒素成分および3つのヘマグルチニン成分(HA33,HA17,HA70)とともにプロジェニター毒素複合体を形成している。ヘマ

グルチニン成分は,上皮細胞上の糖鎖に結合することにより毒素の侵入を助けている。ヘマグリチニン成分の1つであるHA70の糖鎖認識機構について新たな知見を得るために,タイプC毒素のHA70(HA70/C)とシアル化糖鎖との複合体のX線結晶解析および糖鎖マイクロアレイによる結合アッセイを行った。その結果,HA70/Cは,α2-3-,α2-6-シアル化糖鎖の両方を認識し結合することができるが,その親和性は,α2-3-シアル化糖鎖の方が,α2-6-シアル化糖鎖よりも高いことがわかった。

Clostridium botulinum produces the botulinum neurotoxin, forming a large complex as progenitor toxins in association with nontoxic nonhemagglutininand/or several different hemagglutinin (HA) subcomponents, HA33, HA17 and HA70, which bind to carbohydrate of glycoproteins from epithelial cells in theinfection process. To elucidate the carbohydrate recognition mechanism of HA70, X-ray structures of HA70 from type C toxin (HA70/C) in complexes withsialylated oligosaccharides were determined, and a binding assay by the glycoconjugate microarray was performed. These results suggested that HA70/Ccan recognize both 2-3- and 2-6-sialylated oligosaccharides, and that it has a higher affinity for 2-3-sialylated oligosaccharides.

HA70・糖鎖複合体の立体構造 X-ray structure of HA70 / oligosaccharide complex

シアル化糖鎖認識機構と親和性 Recognition mechanism of sialylated oligosaccharide

図1.(A)予想されているプロジェニター毒素複合体の模式図。14の分子からなると考えられている。 (B)α2-3-シアルラクトース(3SiaLac), α2-3-シアルラクトサミン(3SiaLacNAc)α2-6-シアルラクトース(6SiaLac) α2-6-シアルラクトサミン(6SiaLacNAc)の化学構造式。

Fig. 1. Schematic diagram of progenitor toxin complexand the chemical structure of ligands. (A) A hypothetical14-mer model of the progenitor toxin complex. (B)Chemical structure of 2-3-sialyllactose (3SiaLac), 2-3-sialyllactosamine (3SiaLacNAc), 2-6-sialyllactose(6SiaLac), and 2-6-sialyllactosamine (6SiaLacNAc)with the names of sugar units.

図2.(A)HA70/Cの3量体構造を示す。結合している3SiaLac(グレー色),6SiaLac(シアン色)を空間充填モデルで描いている。結晶化中に,HA70/Cは2つのフラグメント,HA22-23/C(青色)とHA55/Cに分解されていた。HA55/Cは,ドメインⅠ(マジェンタ色),Ⅱ(黄色),Ⅲ(緑色)からなる。 (B)糖鎖結合部位の拡大図を示す。ここでは,結合糖鎖は棒状モデルで描いている。ドメインⅡの2-7シートとドメインⅢの1-6シートは連続的に大きなシートを形成している。ドメインⅡの1-6シートとドメインⅢの2-7シートは,大きなシートと平行に位置するが,それらの間にはギャップがあり,ここに糖鎖が結合する。糖鎖と相互作用している部分を赤色で示す。

Fig. 2. Structure of HA70/C with the bound ligands. (A) Homo-trimer of HA70/C with the bound ligand in the space-filling mode with graycarbon (3SiaLac) and cyan carbon (6SiaLac). The proteolytical digestion during the crystallization process gives HA22-23/C and HA55/Cfragments, and HA55/C can be divided into three domains, domain-I, -II and -III.HA22-23/C, and domain-I, -II, and -III of HA55/C are in blue,magenta, yellow and green, respectively. Symmetry-operated molecules are in light colors. (B) A close-up view of the carbohydrate-bindingsite with the bound ligand shown as a stick model. The 2-7 sheet of domain-II and 1-6 sheet of domain-III are forming a large -sheet ofnine -strands by two domains. The 1-6 sheet of domain-II and 2-7 sheet of domain-III are located on the same side of the large -sheetwith a gap between them. In this gap, the concave surface for a ligand to bind is formed, as shown in red.

図3.(A)結合した3SiaLacとHA70/Cとの間の水素結合を点線で示す。3SiaLacは伸長したコンフォメーションをとっている。 (B)6SiaLacは折れ曲がったコンフォメーションで結合している。6SiaLacNAcの占有率は低く,HA70/Cの糖鎖結合親和性は3SiaLacの方が6SiaLacNAcよりも高いと考えられる。

Fig. 3. Interactions between the protein and ligand. (A) The bound 3SiaLac between domain-II(yellow) and -III (green) is shown with hydrogen bonds by dotted lines. (B) The bound 6SiaLac isshown. The refined occupancy factors of the ligands suggest that the complexes with 3SiaLacare stable relative to the complexes with 6SiaLac.

図4.(A)糖鎖マイクロアレイによる結合アッセイの結果を示す。SiaLacNAc修飾したBSAを濃度を変えて(65-100g/ml),エポキシ活性化したガラス板にスポットした。Cy3標識したHA70/C(10g/ml)を用いてリガンドカップリングを行った。バックグラウンド補正を行った後,3回測定したそれぞれのスポットのシグナルを平均した。 (B)エラーバーとともにグラフに表す。3SiaLacNAcの方が結合親和性が高いことがわかる。

Fig. 4. Binding assay of HA70/C against SiaLacNAc-BSA. (A) Varying concentrations ofSiaLacNAc-BSA (65–1000 μg/ml) were spotted on epoxyactivated glass slides. Ligandcoupling was measured using 10 μg/ml of Cy3-labeled HA70/C. The net intensity value foreach spot was determined as the signal intensity minus the background value, and Data arethe average of three columns. The average values of net intensities are also listed withestimated standard deviations. (B) Data are plotted with error bars, suggesting that HA70/Chas a higher affinity for 3SiaLacNAc.

3SiaLacNAc 6SiaLacNAc(mg/ml)

1000

500

250

125

65

9052±11779799±1364

10615± 72113185±1339

14052±1553

1466±1072224±1372267±1292536± 402549± 95 1000 (mg/ml)

500 250 125 65

3SiaLacNAc 6SiaLacNAc0

2000

4000

6000

8000

10000

12000

14000

16000

18000

0 65 125 250 500 1000

3SiaLacNAc

6SiaLacNAc

Conc. (mg/ml)

Net

inte

nsity

HA70

Neurotoxin

Non toxin Non HA

HA33

HA17

Sia Gal Glc (GlcNAc)

3SiaLac (NAc)

6SiaLac (NAc)

O

H

RNH

H

OH

H

H

H

COOH

CH3CO

O

O

CH2OH

HH

H

OH

OH (NHCOCH3)

HH

OHO

CH2OH

HOH

H

H OH

HH

O

OHH

OHH

CH2OH

O

H

RNH

H

OH

H

H

H

COOH

CH3CO O

OH

O

CH2OH

HH

H

OH

OH (NHCOCH3)

HH

OHO

CH2 .

HOH

H

H OH

HH

O

R =

A

B

N514

D513

R525

R460D462

T524

Y522

N320

Sia

Gal

Glc

N514

D513

R525

R460D462

T524

Y522

N320

Sia

Gal

Glc

A B

Domain-III

Domain-II

Domain-I

HA22-23/C

A B

Domain-II

Domain-III

A B