【作者】 牟佳佳；

【导师】 徐文方；

【作者基本信息】 山东大学 ， 药物化学， 2009， 博士

【摘要】 氨肽酶N（APN）是一类膜型锌离子依赖性的外肽酶,在很多组织如小肠、肾、肝以及中枢神经系统中都有广泛分布。该酶参与了营养素的最终水解和生物活性分子如脑啡肽和内啡肽的降解失活。另外,氨肽酶N也是冠状病毒的受体,并参与了抗原递呈的过程。更为重要的是,Saiko等人确定氨肽酶N在肿瘤细胞表面高水平表达,被称作肿瘤细胞标记物或相关抗原CD13。它在肿瘤细胞的生长、侵袭、转移和新生血管的生成等过程中发挥着关键作用。因此氨肽酶N成为抗肿瘤药物研究的一个重要靶点。到目前为止人源氨肽酶N的晶体结构未见报道。2006年Kiyoshi等人报道了大肠杆菌氨肽酶N（eAPN）以及eAPN与其特异性抑制剂Bestatin共结晶复合物的晶体结构。eAPN的晶体结构表明氨肽酶N活性区域至少包含有一个阴离子结合位点和三个疏水性口袋。三个疏水性口袋分别是位于锌离子左侧的S₁口袋和右侧的S₁’和S₂’口袋。这为氨肽酶N抑制剂的合理药物设计奠定了基础。本研究以氨肽酶N为靶点,基于Bestatin与eAPN作用模式,通过计算机辅助药物设计,构建出抑制剂的药效团模型,然后结合文献调研和本课题组已有的研究成果确定以L-精氨酸为母体骨架。在构建的L-精氨酸母核的结构基础上,根据eAPN的三维结构要求,将其侧链进行结构衍生化。在合成目标化合物之前,首先利用SYBYL7.0中的FlexX软件对所设计的化合物进行了对接打分,结果显示大部分化合物分值都接近于甚至超过阳性对照Bestatin。这说明了我们设计思路的合理性,为我们的研究提供了理论依据。本研究的合成工作以L-精氨酸为起始原料,首先将L-精氨酸的胍基用硝基保护,再将其羧基进行甲酯化保护得到关键中间体,然后将中间体与芳酰氯,芳磺酰氯,芳酸,Boc保护的氨基酸或者各种类二肽缩合,最后将羧酸甲酯转化成异羟肟酸,脱Boc保护基得到两个系列的目标化合物,即类二肽化合物（A系列）和类三肽化合物（B系列）。本文共设计、合成了107个目标化合物,并通过红外光谱、核磁共振氢谱、电喷雾质谱,高分辨质谱等方法对化合物结构进行了确证。经文献检索证明,所合成的目标化合物均为新型化合物,未见报道。对目标化合物进行了体外抑酶实验、体外抑制HL-60、K562、ES-2、A549、PLC及H7402细胞增殖实验,体内动物抑瘤实验等,以期筛选出高活性的氨肽酶N抑制剂。体外抑酶实验结果显示,B系列化合物相对于A系列化合物活性有所提高,且两个系列的化合物对APN的抑制活性都优于MMP-2。A系列中的A18和A21活性较好,IC₅₀达到了5.3和5.1μM(Bestatin,IC₅₀=3.1μM)。B系列化合物中B47,B50和B52的IC₅₀分别达到了64.8,53.4和59.0μM,接近于阳性对照Bestatin(IC₅₀=51.8μM)。MTT法体外细胞实验测定了体外抑酶活性较好的11个化合物对高表达APN/CD13的HL-60、ES-2、A549、PLC细胞以及低表达APN的K562、H7402细胞的增殖抑制活性,实验结果表明我们的化合物对APN高表达的细胞都有较好的抑制作用,对于APN低表达的H7402细胞抑制作用比较差,但是对另外一种APN低表达的细胞K562的抑制活性却比较好,其原因还有待于进一步的研究分析。化合物A22对于六种肿瘤细胞都显示出了很好的活性,可以对其进行进一步的体内动物实验研究。体内动物抑瘤实验选取了体外抗细胞增殖作用好的A22,B46,B50和B52四个化合物进行研究,实验结果表明化合物A22和B50能明显减少荷肝癌H₂₂小鼠的肺部结节数目,即能明显抑制癌细胞的转移,抑制率也超过了阳性对照药Bestatin,分别达到了61.50%和71.39%（Bestatin的抑制率为58.04%）。本文在所有化合物活性实验结果的基础上,利用计算机软件进行了初步定量构效关系的研究。构建了QSAR模型,对于指导以后同类化合物的结构优化有一定的指导作用。本研究的主要创新点是:①基于eAPN的晶体结构及Bestatin与eAPN的作用模式,利用计算机软件辅助设计合成了一系列未见文献报道的基于L-精氨酸骨架的衍生物,合成路线设计合理,原料经济易得。②大多数化合物在体外表现出较好的抑酶活性以及抑制各种肿瘤细胞增殖的活性。化合物A22和B50还表现出比阳性对照Bestatin更好的体内动物抑瘤活性。③基于化合物的结构和活性数据建立了有一定预测能力的定量构效关系模型,为今后氨肽酶N抑制剂的研究提供了指导和帮助。更多还原

【Abstract】 Aminopeptidases N, a membrane-bound zinc-dependent exopeptidase, is expressed in many tissues including intestine, kidney, liver as well as central nervous system. APN participates in the final hydrolysis of nutrients and the degradation of bioactive molecules such as enkaphalin and endorphin. Furthermore, APN serves as a receptor for corona viruses and it is also involved in the process of antigen presentation. More importantly, Saiko and coworkers have found that APN is highly expressed in tumor cells and it is regarded as tumor marker or fos-related antigen CD13. APN plays an important role in tumor cell growth, invasion, metastasis and neovascularity. All these findings make this enzyme an interesting target for anti-tumor drugs research.Now the three dimensional structure of human APN has not been found, but the crystal structures of E. Coli aminopeptidase N and enzyme-Bestatin complex were reported in 2006. The crystal structure of eAPN showed that the catalytic region of APN consisted of one negative ion binding site and three hydrophobic pockets. The three hydrophobic pockets are S₁ pocket, which lies in the left of the zinc ion, and S₁’ and S₂’, which lie in the right of the zinc ion. This supplied basement for our rational drug design.Our study targeted on APN for many years. Based on the binding mode of Bestatin with eAPN, the pharmacophore of APN inhibitors was bulit. And then combined with the literature retrieval and our lab’s achievements, we chose L-arginine as our scaffold. According to the requirements of the 3D-structure of eAPN, various fragments are introduced in the scaffold. Before synthesizing, we docked our compound with eAPN model via FlexX of sybyl 7.0. The result showed that almost all of the compounds had similar or higher score than Bestatin. This, to some extent, ensured the rationality of our design strategy and supplied basement for our study.In the synthesis, L-arginine was used as the starting material. The guanidinium group of L-arginine was protected by nitro group and the carboxy group was esterified to get the key intermediate. Then the intermediates were condensated with aromatic acyl chloride, aromatic sulfonyl chloride, aromatic carboxylic acid, Boc protected amino acid or Boc protected di-peptide analogues. The methyl ester groups of the former compounds were converted to hydroxymate group. The compounds contained Boc groups were treated with HCl/EtOAC to get the target compounds.In this study, 107 target compounds were obtained and all of them were identified by IR, ¹H-NMR, ESI-MS and HRMS. Literature retrieval proved that all the compounds were new and not reported. Additionally, in vitro anti-APN assay, in vitro anti-proliferation assays of HL-60, K562, ES-2, A549, PLC and H7402 cells were processed in this research. Finally, in vivo anti-metastasis assay in rat was conducted for some selected high activity compounds.The result of enzyme activity assay in vitro showed that B series of compounds had better activities than A series of compounds. And both of the two series showed better activities towards APN than towards MMP-2. A18 and A21 of A series exhibited better activities than others, which IC₅₀ came to 5.3 and 5.1μM (Bestatin, IC₅₀=3.1μM). B47, B50 and B52 of B series had potential activities, which IC₅₀ was 64.8, 53.4 and 59.0μM respectively, near to the positive control Bestatin (IC₅₀=51.8μM).11 compounds were selected and assayed for their anti-proliferation activity in vitro towards tumor cells which highly express APN/CD13, for example HL-60, ES-2, A549 and PLC cells, and tumor cells which lowly express APN, for example K562 and H7402 cells. MTT method was employed. The result showed that the designed compounds did have anti-proliferation activity towards tumor cells which highly express APN/CD13. But to tumor cells which lowly express APN, our compounds showed low activities to H7402 but high activities to K562. The reason why they showed good activities toward K562 remains to be discussed. Compound A22 showed excellent activities against all the 6 kinds of tumor cells, it should be studied more deeply.A22, B46, B50 and B52 were selected to conduct anti-metastasis activity assay against H₂₂ tumor cells in vivo. The result showed that A22 and BSO can decrease the number of nodes on the lung surface of mice bearing H₂₂ tumor cell, that is to say which can inhibit the metastasis of tumor cells. Their inhibition rates are higher than Bestatin （61.50%, 71.39% and 58.04%, respectively）.Finally, according to the structure and activity of the compounds, we conducted QSAR research via computer assisted software. The QSAR model we built may direct the structure optimization in the future.The creativity points of our research are as follows:①Based on the crystal structure of eAPN and the binding mode of Bestatin with the enzyme, utilizing the computer-assisted drug design software, we designed and synthesized the target compounds with the scaffold L-arginine. The scheme was feasible and the materials were economical.②Preliminary enzyme activity assay and anti-proliferation assays of HL-60, K562, ES-2, A549, PLC and H7402 cells showed most compounds possessed potential APN inhibitory activities. Compound A22 and B50 also showed better anti-metastasis activities against H₂₂ tumor cell in vivo than Bestatin.③We also bulit a QASR model of the target compounds, which can be used in the future for APN inhibitor design.更多还原