病害胁迫下桑树叶片防御酶活性变化和高光谱特征响应研究

Mulberry Bacterial Blight (Pseudomonas syringae pv. mori) is a kind of widespread bacterial disease, which caused significant reduction in mulberry leaf yield and quality. It is one of the main obstacles of the sustainable development of sericulture industry. At present, the prevention and control of mulberry blight mainly depend on chemical pesticides in addition to selective breeding of resistant varieties. However, The environmental pollution caused by chemical pesticides, and the influence of pesticide residues on sericulture have been gradually serious. For the early detection of Mulberry Bacterial Blight that can reduce the adverse effects of chemical pesticides, this research focuses on early and fast diagnosis of Mulberry Bacterial Blight based on hyperspectral imaging. Hyperspectral image information of infected mulberry leaves will be get, and early identification model of Mulberry Bacterial Blight will be established with chemometrics and digital image processing. At the same time, activities of chemical defense enzymes including Superoxide Dismutase (SOD), Catalase (CAT), Peroxidase (POD) in infected mulberry leaves will be detected, and the correlation between the activity change of chemical defense enzymes and the disease information will be illustrated, and finally the prediction model of chemical defense enzymes’ activity will be established. The results will provide information and technical support for on early prevention and control of mulberry diseases research, ecological sericulture industry promotion and comprehensive utilization of mulberry resources development.

桑疫病是一种广泛流行的细菌性疾病，严重影响桑叶的产量和质量，已经成为阻碍蚕桑产业可持续发展的主要障碍之一。目前桑疫病的防治除选育抗性品种外，主要依赖化学农药，而化学农药带来的环境污染以及农药残留对蚕业生产的负面影响已逐渐凸显。为了实现病害早期监测，进而可根据病害情况采取清除措施或变量实施农药，减少化学制剂带来的不利影响，本项目提出病害胁迫下桑树叶片防御酶活性的变化及高光谱特征响应的研究。通过获取受感染桑树叶片的高光谱图像信息，利用数字图像处理技术和化学计量学方法建立桑疫病的响应模型；同时获取桑树在桑疫病胁迫下叶片内超氧化物歧化酶、过氧化氢酶和过氧化物酶等化学防御酶活性指标，探明化学防御酶活性指标的变化与病害程度的相关性，并建立化学防御酶活性指标的响应模型，实现桑树病害的早期诊断，为研究桑树病害早期防治、促进蚕桑产业生态化、拓展桑树资源综合利用提供信息和技术支持。

本研究以高光谱成像技术为主要技术手段，以桑树叶片作为研究对象，系统地研究了桑树叶片在桑疫病病害胁迫下光谱信息、图像信息以及叶片SOD、POD和CAT等化学防御酶指标的变化。建立了基于高光谱信息的桑疫病胁迫时间鉴别模型，选取了490 nm、602 nm、678 nm、741 nm、882 nm和885 nm6个特征波长高光谱数据进行Bayes判别分析，对校正集和预测集样本的判别准确率分别为76.14%和62.33%。建立了基于高光谱图像信息的桑疫病胁迫时间鉴别模型，提取了上述6个特征波长图像的8个纹理特征数据进行Bayes判别分析，对校正集和预测集样本的判别准确率分别为60.43%和43.44%。选取了457 nm、566 nm和738 nm3个特征波长的光谱数据，建立了基于高光谱信息的桑叶SOD活性PLS检测模型，该模型的Rc为0.7971，Rp为0.5869。选取了454 nm、503 nm、520 nm、538 nm、610 nm、654 nm、886 nm和898 nm8个特征波长的光谱数据，建立了基于高光谱信息的桑叶POD活性PLS检测模型，该模型的Rc为0.6785，Rp为0.5738。选取了465 nm、482 nm、533 nm、548 nm、644 nm、734 nm和890 nm7个特征波长的光谱数据，建立了基于高光谱信息的桑叶CAT活性PLS检测模型，该模型的Rc为0.7676，Rp为0.6685。探明了桑叶SOD、POD和CAT活性与桑疫病胁迫程度存在相关性。本研究为实现桑树病害早期诊断和防治、促进蚕桑产业生态化、拓展桑树资源综合利用提供信息和技术支持。