建立抗 SARS 和抗發炎作用之藥物篩選系統

Abstract

在早期藥物探索的領域上，主要的工作就是要鑑別出具有藥物治療效果的目標基因或是蛋白質，並發展出一個適當的藥物篩選平台來找出有潛力的化合物。隨著高通量的藥物篩選平台技術的發展與成熟，此項技術已成為近年來藥廠相當依賴的篩選方法；在這樣的平台中，主要為利用生化方法來建立酵素活性、受體結合或是細胞功能的測定，並將此種測定方法學應用於96孔或是384孔盤式分析，和輔以自動化機械設備來達成高速大量分子篩選的目標。此篇論文的主要目標即在於發展不同的藥物篩選平台，並藉由小量的篩選過程中找出具有潛力的化合物，並同時驗證篩選系統的穩定度和適當性。 我們分別設計一個酵素活性篩選平台與一個細胞激素釋出的細胞功能篩選平台。前者利用螢光共振能量轉移 (FRET) 原理建立以螢光訊號來反映酵素活性之藥物篩選系統，後者則為找尋可抑制細菌毒素 (LPS) 誘導周邊血液單核細胞腫瘤壞死因子α (TNF-α) 和介白質1β (IL-1β) 上升的化合物之細胞功能篩選平台。篩選平台系統的穩定性可由 Z-factor 來評估，在螢光共振能量轉移法的酵素活性篩選平台是 0.81，而細胞功能篩選平台中的 TNF-α 和 IL-1β 測定則分別是 0.72 和 0.76。由此結果看來，所發展出的這兩個篩選平台是相當穩定且可行。 酵素活性篩選平台是利用非典型肺炎冠状病毒 (SARS-CoV) 3CL蛋白脢為藥物作用標的，3CL 蛋白脢對病毒蛋白質生成、活性及修飾上扮演相當重要的角色，因此成為一個治療非典型肺炎或其他冠狀病毒引發之疾病的藥物作用目標。我們利用螢光共振能量轉移方法來篩選數百種已知的化合物，篩選出對3CL蛋白酵素活性有抑制作用的潛力化合物，其中以 hexachlorophene 是最具潛力的抑制化合物，其抑制常數Ki是4 µM；它的抑制模式是在受質結合處與受質競爭， 並且其抑制能力隨著作用時間的延長，會有增強的效果。除此之外，triclosan和nelfinavir也有抑制作用，但抑制能力卻約僅有hexachlorophene的十分之一，抑制常數 Ki分別是40 µM 和21 µM。進一步分析 hexachlorophene對蛋白脢抑制能力的專一性，在四十多種的蛋白脢活性分析中，發現 hexachlorophene 同時會對部份的cysteine和aspartic蛋白脢有抑制作用，而這些蛋白脢多與癌症調控相關。 在建立第二個藥物篩選系統上，則是利用細胞發炎的外生性刺激因子 (LPS)， 誘導人類周邊血液單核細胞產生腫瘤壞死因子α (TNF-α) 和介白質1β (IL-1β)，藉此實驗中找尋能抑制腫瘤壞死因子α和介白質1β表現，而具有抗發炎潛力的化合物。在將近五百多種的化合物中，MT4最具有抑制效果，其對於腫瘤壞死因子α和介白質1β的百分之五十抑制濃度 (IC50) 分別為22 和44 nM。接著我們利用in vitro激脢活性實驗, 探討MT4抑制訊息傳遞路徑的目標酵素，發現MT4對於p38α和p38β的抑制能力隨濃度增高而增高，對於 p38γ和p38δ 的則具中度抑制能力；MT4對於p38α、p38β、p38γ和p38δ的IC50分別是0.13 µM、0.55 µM、5.47 µM 和 8.63 µM。進一步實驗發現 MT4 為 p38α 的腺苷三磷酸 (ATP) 競爭性抑制物。此外，MT4 亦可抑制發炎相關的第二環氧化脢 (cyclooxygenase-2) 的酵素活性和降低細胞中前列腺素E2 (prostaglandin-2) 的釋出。 在整個研究中，我們設立了兩個優良的篩選平台，分別鑑定出非典型肺炎冠狀病毒 3CL 蛋白脢和 p38α 的抑制物。經過分子崁合，顯示化合物與酵素彼此間有多個可能的作用存在，這些篩選出的化合物將可做為未來進一步藥物設計的起點。

The challenge of the early drug discovery is to identify molecular targets that hold the greatest potential for therapeutic intervention and then to develop an appropriate assay model for screening libraries composed of individual organic or natural molecules. The first step in this early drug discovery pipeline is typically the establishment of high-throughput screen (HTS). And in vitro biochemical and cellular assays have long been used for HTS of large amount of molecules in 96-well and 384-well plate formats. The specific aim of this study is to develop and validate diverse methods in order to demonstrate their suitability for screening in the drug discovery, and perform small-scale screening and find potential hits or leads in an academic research. In the current work, a FRET (fluorescence resonance energy transfer) assay was developed for the evaluation of SARS-CoV 3CL protease activity. And a cell-based assay was validated for detecting the release level of TNF-α (tumor necrosis factor-α) and IL-1β (interleukin-1β) using LPS-stimulating peripheral blood mononuclear cells (PBMC) model. The assay quality parameter, Z’-factor, was employed. The Z’-factor of the FRET assay was 0.81 and the PBMC cell-based assay was 0.72 and 0.76 respectively for TNF-α and IL-1β release. In terms of the value of the Z’-factor, these screening assay qualities were classified with an excellent performance. The SARS-CoV 3CL protease is an essential enzyme for viral proteins processing and regarded as a good drug target to SARS-CoV replication. Here, hundreds of known compounds were examined by using FRET screening assay. Among these drugs, hexachlorophene was identified as the most potent inhibitor of SARS-CoV 3CL protease (Ki = 4 µM). Further characterization revealed that its binding mode was competitive with the substrate-binding site and the inhibitory effect was pre-incubation time dependent. Besides, two other known drugs, triclosan and nelfinavir, were 10-fold less potent (Ki = 40 µM and Ki = 21 µM respectively) than hexachlorophene. Furthermore, to evaluate the specificity of hexachlorophene, over forty protease assays were conducted to screen its inhibitory effects. The results showed that hexachlorophene not only inhibited SARS-CoV 3CL protease but also inhibited some of cysteine and aspartic proteases, which related to tumor progression and metastasis. After identifying hexachlorophene and its analogues as the inhibitors of SARS-CoV 3CL protease using an enzyme assay model, we try to establish a cell-based screening system to find some anti-inflammatory inhibitors and then identify their molecular targets. LPS (lipopolysaccharide), one of exogenous inflammatory agent, stimulates PBMC to synthesize or release pro-inflammatory cytokines. Therefore, our strategy was to screen test compounds using LPS stimulated-PBMC and find whether some of them can inhibit the release of TNF-α and/or IL-1β. Then we can examine this kind of potential anti-inflammatory candidates to understand their inhibitory targets. Among nearly 500 test compounds, we found that MT4 had the suppressive action on the release of TNF-α and IL-1β, with IC50 values of 22 and 44 nM, respectively. After we evaluated the anti-cytokine effect of MT4 in terms of the inhibition of p38 MAPK, JNK and ERK activity using in vitro kinase assay, MT4 inhibited the activity of p38α and p38β in a concentration-dependent manner. It also displayed moderate inhibitory activity on p38γ and δ. The IC50 values were 0.13, 0.55, 5.47 and 8.63 µM for p38α, β, γ and δ respectively. Further characterization of enzyme kinetics showed that the binding mode of MT4 was competitive with the ATP substrate-binding site of p38α MAPK. Beside the cytokine synthesis and release pathway, we also studied MT4 effect on other inflammatory enzymes and found that it could inhibit cycloxygenase-2 to reduce the prostaglandin-2 production. In summary, our studies offer simple and excellent screens to identify anti-SARS and anti-inflammatory inhibitors. These identified small molecules that can serve as chemical starting points or high quality leads for further optimization to provide a good opportunity for developing novel and potent drug candidates.