作　　者： ; ; ; ;

机构地区： 中国水产科学研究院南海水产研究所

出　　处： 《食品科学》 2011年第20期63-68,共6页

摘　　要： 利用具有自学习特点的人工神经网络可实现对酶解过程的模拟仿真,研究从合浦珠母贝肉中制备抗菌肽的最佳工艺条件。采用3层(5-9-3)人工神经网络法对风味蛋白酶水解合浦珠母贝肉的工艺过程进行模拟和优化,并通过管碟抑菌法对产物的抑菌性质进行分析。结果表明:pH7.0、水解温度55℃、酶添加量1.6%、水解时间4h、料液比7:5,制备得到肽A,抑制鼠伤寒沙门氏菌最强,抑菌圈直径14.20mm,平均肽链长度2.6;pH7.0、水解温度55℃、酶添加量1.7%、水解时间4h、料液比3:2,制备得到肽B,抑制痢疾志贺氏菌最强,抑菌圈直径23.42mm,平均肽链长度2.8;pH6.5、水解温度60℃、酶添加量2.5%、水解时间4h、料液比7:5,制备得到肽C,对单核细胞增生李斯特菌的抑制效果最强,抑菌圈直径16.60mm,平均肽链长度2.5。3种抗菌肽对大肠杆菌、金黄色葡萄球菌也具有较强的抑菌效果,抑菌率为74.3%～80.8%。本研究利用人工神经网络优化制备的贝肉抗菌肽克服了纯度低、提取率低等缺点,为合浦珠母贝肉抗菌肽的开发利用提供技术支撑。 In this paper,artificial neural network（ANN）,which can simulate an enzymatic hydrolysis process based on self-learning,was applied to study the optimum conditions for enzymatic preparation of antibacterial peptides from Pinctada fucata muscle with flavourzyme.The hydrolysis of Pinctada fucata muscle by flavourzyme was simulated and optimized using a three-hierarchy ANN structure.The antimicrobial activity of Pinctada fucata muscle hydrolysates was tested by cylinder-plate method.The results demonstrated that the peptide A was obtained after 4 h of hydrolysis under the conditions of pH 7.0,55 ℃,7：5 material/liquid ratio and 1.6% enzyme/substrate ratio,which showed the strongest antibacterial activity against Salmonella typhimurium with an inhibition zone diameter of 14.20 mm and an average peptide chain length of 2.6.Four-hour hydrolysis under the conditions of pH 7.0,55 ℃,3：2 material/liquid ratio and 1.7% enzyme/substrate ratio generated the peptide B,which had the strongest antibacterial activity against Shigella dysenteriae with an inhibition zone diameter of 23.42 mm and an average peptide chain length of 2.8.The peptide C was obtained under the conditions of pH 6.5,60 ℃,7：5 material/liquid ratio,2.5% enzyme/substrate ratio and 4 h hydrolysis,which displayed the strongest antibacterial activity against L.monocytogenes with an inhibition zone diameter of 16.60 mm and an average peptide chain length of 2.5.The three peptides also had strong antibacterial activity against Escherichia coli and Staphyloccocus aureus with an inhibitory rate ranged of 74.3%-80.8%.The application of ANN for optimizing antibacterial peptides preparation from Pinctada fucata muscle overcomes many shortcomings such as low purity and low extraction efficiency and can therefore provide technical supports for the development and exploitation of antibacterial peptides derived from Pinctada fucata muscle.

关 键 词： 合浦珠母贝肉 抗菌肽 制备 人工神经网络优化法

领　　域： [轻工技术与工程] [轻工技术与工程]