以結構資訊為基礎來開發半乳糖凝集素及乙型葡萄糖醛酸酶之抑制物

Structure-based Studies to Develop Inhibitors of Galectin and β-Glucuronidase

林仙雅(Hsien-Ya Lin)

指導教授 ： 林俊宏

共同指導教授 ： 方俊民(Jim-Min Fang)

國立臺灣大學/理學院/化學研究所/博士(2016年)

https://doi.org/10.6342/NTU201602091

摘要

本論文選定兩個標的蛋白，分別是人類半乳糖凝集素 (Galectin) 與腸道共生菌的乙型葡萄糖醛酸酶 (β-Glucuronidase)。這兩類蛋白已被證實在癌症治療上扮演重要的角色，例如半乳糖凝集素促使癌症細胞的生長，及加快腫瘤惡化的進程，而細菌的乙型葡萄糖醛酸酶造成藥物在腸胃道代謝，引發病患對抗癌藥物的副作用。由於這兩類蛋白家族中，各別成員的胺基酸序列和三級結構都具有高度的保留性。若想要分別針對特定的成員，開發有效且具選擇性的抑製劑，是一項極具困難與挑戰性的任務。有鑒於以蛋白質結構，為藥物設計導向的諸多成功案例，我們整合了蛋白質晶體和核磁共振光譜所得到的結構資訊，以及等溫滴定量熱法等所測定的結合參數，來開發這兩種治療標靶的選擇性抑製劑。
半乳糖凝集素，因結合細胞表面醣體的半乳糖苷而得名，其醣辨識區可分為五個子結合區 (A–E)。迄今最有效的半乳糖凝集素抑製劑 （TD139; 3,3’-deoxy-3,3’- bis-(4-[m-fluorophenyl]-1H-1,2,3-triazol-1-yl)-thio-digalactoside） ，是針對半乳糖凝集素-3 (Galectin-3) 所開發，該抑制劑已被初步證實對特發性肺纖維化 (idiopathic pulmonary fibrosis) 具有療效。然而，有關半乳糖凝集素-3 如何與TD139 作用，及 TD139 為何能對半乳糖凝集素家族成員有選擇性，尚缺乏結構上的解釋與分析。在本文的第一部分中，我們研究半乳糖凝集素-1，-3 和 -7 ，如何與三個硫代二半乳糖苷 (Thiodigalactoside; TDG) 的衍生物作用，其中包括TDG 本身， TD139 和 TAZTDG (3-deoxy-3-(4-[m-fluorophenyl]-1H-1,2,3-triazol-1-yl)-thio- digalactoside)。出人意料的，我們發現 TAZTDG 對這三個半乳糖凝集素有兩種不同的結合模式，其氟苯基 (fluorophenyl group) 會優先結合到半乳糖凝集素-3 的子結合區 B，而對半乳糖凝集素-1 和 -7 則傾向結合到子結合區 E。我們的研究不僅首次證實了子結合區 E的貢獻，也為未來其他半乳糖凝集素選擇性抑製劑的開發，提供了重要的資訊。
第二部分的研究對象為細菌的乙型葡萄糖醛酸酶。相比於人類酵素，細菌的乙型葡萄糖醛酸酶在活性區附近，多了一段人類酵素所沒有的 loop，該段區域命名為 bacterial loop。目前已知的選擇性抑製劑 （ASN 03273363） 即結合於此細菌獨有的 loop 上，並不會影響人類酵素。然而，我們發現超過半數以上的共生菌，序列中並不包含該段loop，即便包含，其高度的序列變異性，也造成 ASN 03273363 的抑制效果不彰。為了解決這些問題，我們以酵素催化的過渡態類似物–isofagomine（IFG） 作為基礎，合成一系列衍生物，嘗試尋找對細菌酵素具專一性的抑制物。結果發現，在一號碳引入取代基可使 IFG 衍生物具備選擇性。此外，我們針對雙歧桿菌 (Bifidobacterium dentium) 的乙型葡萄糖醛酸酶，利用組合式化學快速合成篩選具潛力的 IFG 衍生物。雙歧桿菌酵素因不含前述的 loop 區域，而無法被 ASN 03273363 所抑制，我們成功地篩選出一不可逆抑制物，該抑制物與非催化區域的半胱氨酸形成共價鍵，與人類酵素的選擇性可上達百倍。

關鍵字

半乳糖凝集素 ； 乙型葡萄糖醛酸酶 ； 腸道共生菌 ； 選擇性抑制物 ； 蛋白質結構

並列摘要

In this thesis, two validated drug targets are focused for structural studies, i.e., human β-galactoside-binding lectins (galectins) and bacterial β-glucuronidases (GUSs), which have been known as key players in tumor progression and drug metabolism, respectively. Because the primary sequence and tertiary structure were remarkably conserved among the galectin and GUS families, developing potent and selective inhibitors for specific members has become a longstanding challenging. We particularly applied multi-disciplinary approaches for structure-based drug development, such as synthetic chemistry, X-ray crystallography, isothermal titration calorimetry, and NMR spectroscopy.
To date, TD139 (3,3ʹ-deoxy-3,3ʹ-bis-(4-[m-fluorophenyl]-1H-1,2,3-triazol- 1-yl)-thio-digalactoside) representing the most potent inhibitor for galectin-3, one of the most prominent galectin family members involved in several pathological processes, has been approved for the clinical trial of idiopathic pulmonary fibrosis. However, the full structural information concerning subsites A–E of galectin and the interactions with TD139 are not currently available. In the first part of this thesis, we studied the binding contributions of these subsites in galectins-1, -3, and -7 with several sialylated/sulfated disaccharides and three thio-digalactoside (TDG) derivatives, including TDG, TD139, and TAZTDG (3-deoxy-3-(4-[m-fluorophenyl]-1H-1,2,3-triazol-1-yl)-thio- digalactoside). Surprisingly, we found that the fluorophenyl group of TAZTDG preferentially bound to subsite B in galectin-3, whereas the same group favored binding at subsite E in galectins-1 and -7. The characterized dual binding modes demonstrate how binding potency, reflected in decreased Kd values of the TDG-derived inhibitors from μM to nM levels, is improved. The resulting information offers insights into the development of selective inhibitors for individual galectins.
In the second part, we focused on microbial GUSs, which interfere with xenobiotic detoxification and thus impact human health. Currently several selective inhibitors (such as ASN 03273363) for bacterial GUSs are known to rely on the unique loop that is located near the enzyme active site and is found only in microbial enzymes. However, there are two crucial problems associated with ASN 03273363. One is that the bacterial loop exhibits significant sequence variation, whilst the other is the presence of more than half of bacterial GUSs lacking this loop (NL-GUSs). To solve these problems, we developed a series of uronic isofagomine (IFG) derivatives as transition-state analogues, which directly interact with the conserved catalytic residues of GUSs. The results indicated that substituents introduced at the C1-position play an important role in determining the selectivity between bacterial and mammalian GUSs. Moreover, a combinatorial method was applied to rapidly generate and screen a variety of uronic IFG derivatives. Among them, an irreversible inhibitor was identified to target the non-catalytic cysteine that is conserved in several NL-GUSs, but is absent in both the human and loop-containing bacterial GUS. The selectivity is up to 2–3 orders of magnitude greater for bacterial NL-GUSs (IC50 in the range of nM) than for HsGUS (IC50 of μM).

並列關鍵字

galectin ； β-glucuronidase ； microbiome ； selective inhibitor ； protein structure

參考文獻

2 Erickson, J.; Neidhart, D. J.; VanDrie, J.; Kempf, D. J.; Wang, X. C.; Norbeck, D. W.; Plattner, J. J.; Rittenhouse, J. W.; Turon, M.; Wideburg, N.; et al. Design, activity, and 2.8 A crystal structure of a C2 symmetric inhibitor complexed to HIV-1 protease. Science 249, 527-533 (1990).

4 Anderson, A. C. The process of structure-based drug design. Chem Biol 10, 787-797 (2003).

5 Verlinde, C. L.; Hol, W. G. Structure-based drug design: progress, results and challenges. Structure 2, 577-587 (1994).

6 Munshi, N.; Jeay, S.; Li, Y.; Chen, C. R.; France, D. S.; Ashwell, M. A.; Hill, J.; Moussa, M. M.; Leggett, D. S.; Li, C. J. ARQ 197, a novel and selective inhibitor of the human c-Met receptor tyrosine kinase with antitumor activity. Mol Cancer Ther 9, 1544-1553 (2010).

7 Eathiraj, S.; Palma, R.; Volckova, E.; Hirschi, M.; France, D. S.; Ashwell, M. A.; Chan, T. C. Discovery of a novel mode of protein kinase inhibition characterized by the mechanism of inhibition of human mesenchymal-epithelial transition factor (c-Met) protein autophosphorylation by ARQ 197. J Biol Chem 286, 20666-20676 (2011).