青枯菌N–醯基高絲氨酸內酯醯化酶之特性分析及其AHLs分解活性的提升

Abstract

N–醯基高絲胺酸內酯醯化酶(AHL–acylases)可藉分解AHLs達到阻止病原菌致病的目的，因此，AHL–acylases是目前抗感染藥物開發上極具潛力的標的。本研究自植物病原菌青枯菌(Ralstonia solanacearum GMI1000)選殖出aac基因，ESI–MS (electrospray ionization mass spectrometry)分析證實aac基因轉譯的795–aa Aac (NP 520668)為AHL–acylase，而aculeacin A對Candida tropicalis的MIC試驗證實，Aac並非aculeacin A acylase (AAC)，因此更名為AlaS。alaS基因表現株Chromobacterium violaceum CV026 (pS3aac) 可阻止受C7–HSL誘導的幾丁質酶及紫色素生成，證實AlaS可做為定額感應子清除者。 純化後之AlaS經N–端定序及ESI–MS分析，證明propolypeptide AlaS 由28–aa signal peptide、191–aa α–subunit、14–aa spacer peptide及562–aa β–subunit所構成，且在auto–processing修飾後可得由α–subunit (20.4 kDa)及β–subunit (60.8 kDa)所摺疊成的mature AlaS，故確認AlaS屬於N–terminal nucleophile (Ntn) hydrolase superfamily。AlaS最適反應溫度及pH值分別為35℃及pH 8.0，不需二價陽離子輔酶，且可在4°C下，穩定保持活性超過7週以上。經500 mM DTT (dithiothreitol)處理的AlaS仍保有50%的活性。 提升AlaS對短鏈N–(hexanoyl)–homoserine lactone (C6–HSL)分解活性，可擴大其應用範圍。本研究以25.3% identity的模板glutaryl 7–aminocephalosporanic acid acylase 1OQZ及docking程式建構含N–(heptanoyl)–homoserine lactone (C7–HSL)之AlaS結構模型(AlaS–C7–HSL–modeling)，並以預測活性區作為循理性設計工具。空間填補策略成功篩選到C6–HSL分解活性高於AlaS 5.6倍的突變酵素AlaSS290I及AlaSFS290I，相較於具短鏈AHLs分解活性的指標酵素AHL–acylase HacB，AlaS突變酵素對短鏈N–(butanoyl)–homoserine lactone (C4–HSL)、N–(β– Ketocaproyl)–homoserine lactone (3OC6–HSL)及C6–HSL的比活性分別可提升至HacB的2.2、2.8及1.04倍。由篩選到的突變酵素證明I283及S290是增強短鏈AHLs分解活性的關鍵殘基。另外，經序列比對、結構疊合、pH profile、突變酵素比活性及動力學分析結果顯示，AlaS催化機制應是類似1OQZ的pseudotriads S234/H256/E704。此外，本研究開發快速簡單的紫白接合篩選法(violet–white conjugation selection; VWCS)，以因應大規模的alaS突變基因庫的篩選。模擬基因庫試驗，證實VWCS是可行的篩選法。本研究是第一篇從植物病原菌中發現AHL–acylase並完成其酵素特性及動力學分析的報告，而關鍵殘基的發現可提供其它AHL–acylases增強短鏈AHLs時的重要突變標的。AlaS結構模擬證明，雖然全序列identity低至25.3%，但局部活性區為高保守性序列時，可取其活性區模型作為循理性設計工具。

N–acylhomoserine lactones (AHLs) acylase possesses high development–values in medical and agricultural application since it can quench AHLs. In this study, an aac gene from Ralstonia solanacearum GMI1000 was cloned. ESI–MS (electrospray ionization mass spectrometry) analysis demonstrated that 795–aa Aac (NP 520668) encoded by aac gene is an AHL–acylase. The MIC test of aculeacin A for Candida tropicalis suggest that Aac isn’t an aculeacin A acylase (AAC), predicted at NCBI. Consequently, aac gene was renamed as ”alaS”. The alaS–expressing clone Chromobacterium violaceum CV026 (pS3aac) effectively inhibited violacein and chitinase activity induced by exogenous N–(heptanoyl)–homoserine lactone (C7–HSL), demonstrating that AlaS can be a quorum–quenching agent. AlaS was furthermore over–expressed and purified in E. coli Star (pET41–aac) under glucose regulation and low temperature (17.5℃). N–terminal sequencing and ESI–MS analysis demonstrated that propolypeptide AlaS, consisting of 28–aa signal peptide, 191–aa α–subunit, 14–aa spacer peptide, and 562–aa β–subunit; and undergoing autoprocessing modification, α–subunit (20.4 kDa) and β–subunits (60.8 kDa) fold into mature AlaS. Consequently, AlaS belongs to N–terminal nucleophile (Ntn) hydrolase superfamily. Temperature 35℃ and pH 8.0 are optimal for reaction, no requirement for divalent cations cofactor, and AlaS activity can be stably maintained over seven weeks at 4°C storage. 500 mM DTT (dithiothreitol)– treated AlaS still can maintain 50% of activity. Improving C6–HSL–degrading activity of AlaS is necessary to control C6–HSL–dependent pathogenicity. One ideal AlaS–modeling was firstly built using only 25.3%–identity–sharing glutaryl 7–aminocephalosporanic acid acylase 1OQZ as a template. The local high–conserved active–site modeling was utilized as a rational design tool. The space–filling rational design strategy can successfully acquire both AlaSS290I and AlaSFS290I which exhibit over 5.6–folds high activities towards to C6–HSL than AlaS. Relative to index AHL–acylase HacB with short–chain degrading activity, mutated–AlaS can exhibit 2.2–, 2.8–, and 1.04–folds activities of HacB towards to N–(butanoyl)–homoserine lactone (C4–HSL), N–(β–Ketocaproyl)–homoserine lactone (3OC6–HSL), and C6–HSL, respectively. Apparently, these screening mutated–AlaS demonstrated that residues I283 and S290 are key residues for enhancing its degrading activities towards to short–chain AHLs. Additionally, superposition, structural–based alignments, pH profile, specific activity, kinetic parameters, and site–directed mutagenesis analysis revealed that the catalytic mechanism of AlaS shall be catalytic pseudotriads S234/H256/E704, similar to 1OQZ. For screening large–scale mutation library, one high–throughput violet–white conjugation selection (VWCS) method, was established in this study. The mimicked mutation library demonstrated that VWCS is realizable. To our knowledge, this is the first report to find an AHL–acylase in a phytopathogen and to investigate its biochemical characteristics and kinetic analysis. The key residues founding can provide other AHL–acylases with important rational design residue–targets for short–chain AHLs degrading activities enhancing. AlaS–modeling also provides a successful exemplification that even though the sharing–identity of global template is in low level (25.3%), if the sharing–identity of local desired–fragment is enough high, the built modeling structure still can be utilized as a rational design tool.