幽門螺旋桿菌岩藻醣轉移酶結構及功能之研究

Abstract

幽門螺旋桿菌 (Helicobacter pylori)已被證實會造成人類宿主胃炎、胃或十二 指腸潰瘍，進而發展成胃癌。此致病菌的lipopolysaccharide (LPS)末端通常表現 路易士抗原 (Lewis antigen)，模擬胃上皮細胞表面上醣類分子，藉此躲避宿主的 免疫攻擊。而岩藻醣轉移酶 (fucosyltransferase)是負責最後的催化合成步驟，將 活化態岩藻醣 (fucose)轉移到受質上(如N-acetyllactosamine)，形成路易士抗原。 關於此酵素的研究，主要受限於此蛋白質低表現量或低可溶性。本論文首先有系 統地截去不同長度的C 端胺基酸序列，發現大大地改善酵素的可溶度，藉此方 便進行之後深入的研究。利用旋光光譜儀或分析型離心機，了解C 端多帶正電 區域及五個胺基酸重複序列 (heptad repeat)共80 個胺基酸是可以被截去的，並不 會影響整體結構及催化活性。若是截去更多胺基酸則會導致二級及四級(雙體變 單體)結構變化，進而關係到活性的喪失。保留越多個重複序列越會維持此酵素 雙體結構，而在高溫處理下也越能穩定酵素構形。 我們探討此酵素的基質選擇性，也利用化學修飾法和定點突變法來了解酵素 的重要特定胺基酸。在受質基質選擇性方面，α1,3-岩藻醣轉移酶只辨識β1,4-鍵 結的受質，受質上的兩個單醣種類都不可替換。在Gal 上的C2、C3 或還原端位 置，可允許有多餘醣的修飾，而C4 和C6 則否。OligoLacNAc 也是受質，每個 LacNAc 單位都可以任意接上岩藻醣。在化學修飾實驗中發現有重要的Histidine 和Arginine 參與催化反應，如Arg79、Arg89、Arg118 和Arg354 主要參與受質 的鍵結，而Arg79 和Arg195 則是與兩個基質鍵結有關。雖然Cysteine 與催化無 關，但Cys159 和Cys237 會形成分子內雙硫鍵，其餘兩個Cys168 和Cys353 則 是自由態存在。 我們也成功得到C 端截去115 個胺基酸的酵素晶體，並且解出酵素、酵素/ 基質(GDP-fucose)和酵素/產物(GDP)複合物晶體結構。此酵素結構主要是由二個 Rossmann 構形組成，是屬於典型的glycosyltransferase-B (GT-B)型。當GDP 或 GDP-fucose 進入催化區時會誘導酵素構形的些微變化。與其它GT-B 酵素結構比 較，Glu95 是扮演催化的general base 角色。Arg195、Tyr246、Glu249 和Lys250 是與GDP-fucose 有氫鍵鍵結。在反應進行時，推論帶正電的Glu249 可以穩定 oxonium cation 的過渡態。雖然此晶體結構是缺乏C 端但在晶格內仍是以雙體組 成，故可推測出C 端位置及可能的雙體結構。我們推論出此酵素反應機制及模 擬出polysaccharide 結合區，可以幫助解釋H. pylori 之LPS 的多變性，更可進而 發展出有用的抑制劑。綜合上述，本篇論文清楚地了解H. pylori α1,3-岩藻醣轉 移酶結構及功能之間的相關性。

Helicobacter pylori is well known as the primary cause of gastritis, duodenal ulcers, and gastric cancer. The lipopolysaccharide (LPS) of this pathogen contains Lewis x and Lewis y structures in the terminus to mimic the surface carbohydrates of gastric epithelial cells, which is proposed to escape the surveillance of host immune system. H. pylori α1,3-fucosyltransferase catalyzes the fucose transfer from the donor GDP-fucose to the acceptor N-acetyllactosamine (LacNAc). The research progress was previously hampered by either poor protein expression or the marginal solubility of the protein. The work in the thesis at first greatly improved the marginal solubility of the full-length protein by systematic deletion of the C terminus of H. pylori α1,3-FucT, which made it possible for further investigations. Based on the biophysical characterizations, including CD spectroscopy and analytical ultracentrifugation, up to 80 residues, including the tail rich in positive and hydrophobic residues (sequence 434-478) and half of the ten heptad repeats (399-433), can be removed without significant change in structure and catalysis. Half of the heptad repeats are required to maintain both secondary and native quaternary structures (dimeric form). Removal of more residues in the C terminus led to major structural alteration, which was correlated with the loss of enzymatic activity. In accordance with the thermal denaturation studies, the results support the idea that a higher number of tandem repeats, functioning to facilitate a dimeric structure, helps to prevent the protein from unfolding when incubated at higher temperatures. H. pylori α1,3-FucT was also subjected to biochemical characterizations, such as substrate specificity, specific chemical modifications, and site-directed mutagenesis. H. pylori α1,3-FucT is highly specific to the β1,4-linkage and does tolerate modification in the reducing end. Both sugar residues of LacNAc are essential for activity. H. pylori α1,3-FucT can sterically accommodate an additional sugar introduced at either C2 or C3 of galactose, but not C4 and C6 at the same sugar. Furthermore, when oligomeric LacNAc was subjected to the enzymatic fucosylation, one to several fucose residues observed in the mass spectra. Every LacNAc unit can be fucosylated in a random manner. H. pylori α1,3-FucT is sensitive to the modification of diethylpyrocarbonate (His-specific) and phenylglyoxal (Arg), but not N-ethylmaleimide (Cys), indicating that His and Arg play an important role in the reaction. The site-directed mutagenesis study showed that Arg79, Arg89, Arg118, and Arg354 are mainly involved in the acceptor binding, while Arg79 and Arg195 are essential for the binding with both substrates. Although cysteine residues do not participate in the catalysis, an intramolecular disulfide linkage between Cys159 and Cys237 are determined. The other residues, Cys168 and Cys353, have free thiol in their side chains. Additionally, protein crystallization was successful with the C-terminal deletion of 115 residues. Three crystal structures have been solved, including the enzyme/GDP-fucose complex and the enzyme/GDP complex. The structure is composed of two Rossmann-like fold domains, typical of the glycosyltransferase-B (GT-B) family. Specific interactions with GDP and GDP-fucose bound to the active site induce conformational changes in the C-terminal domain. Structural comparison with other GT-B members suggests Glu95 to serve as the general base in catalysis. The residues Arg195, Tyr246, Glu249 and Lys250 function to interact with the donor substrate. Asn240 is involved in the binding with the acceptor. Glu249 is proposed to stabilize the developing oxonium cation during the reaction. Although the crystallized protein lacks a substantial C terminus, these structures not only reveal subunit interactions for a dimeric structure, but also predict the location of the missing heptad repeats. We propose a catalytic mechanism and a model of polysaccharide binding to explain the observed variations in H. pylori LPS, as well as to facilitate the development of potent inhibitors. Taken together, this thesis provides clear understanding for the structure and function relationship of H. pylori α1,3-FucT.