基于SERS技术的基底增强效应对比及食源性芽孢快速检测

为了实现食源性芽孢的快速识别，该研究利用表面增强拉曼光谱技术，以产气荚膜梭菌芽孢（C.perfringens spores）、艰难梭菌芽孢（C.difficile spores）和生孢梭菌芽孢（C. sporogenes spores）为研究对象，将3种不同纳米溶胶材料AuNPs、AgNPs和Au@AgNPs作为表面增强拉曼散射（Surface-enhanced Raman scattering，SERS）基底对食源性芽孢的增强效应进行对比，确定较佳增强效果的纳米材料，进一步开展基于较优纳米溶胶材料为基底的SERS技术对不同种属食源性芽孢的光谱解析及快速识别研究。结果表明，Au@AgNPs对食源性芽孢的SERS增强效果最较好，其增强因子达3.49×104，且光谱特异性和重现性良好。经拉曼光谱解析，Ca2+-DPA的拉曼峰是三种食源性芽孢的共有标志特征峰，拉曼振动峰在1 017、1 440和1 570 cm-1波段显示且峰强度不同。C. perfringens芽孢在740、787、821、1 203、1 308和1 627 cm-1有独特的拉曼峰，C. sporogenes芽孢在852 cm-1有独特的拉曼峰，在1 017 cm-1（Ca2+-DPA）、1 081 cm-1（DNA、O-P-O）处峰位强度C. perfringens芽孢>C. sporogenes芽孢>C. difficile芽孢，在1 332 cm-1（核酸中腺嘌呤核酸）、1 440 cm-1（Ca2+-DPA）处峰位强度C. perfringens芽孢>C. difficile芽孢>C. sporogenes 芽孢，3种食源性芽孢的SERS图谱出峰位置和峰强度差异明显。结合主成分分析（Principal Component Analysis，PCA）和线性判别分析（Linear Discriminant Analysis，LDA）构建定性识别模型，经 PCA分析，其前三个主成分的累计贡献率达92.90%，然后利用LDA进行分析，其识别率可达到100%。结果表明，以Au@AgNPs为基底的SERS技术检测方法可以实现对食源性芽孢的快速识别，为食源性芽孢检测和食品安全检测提供了有效手段。

Enhanced performance comparison of substrate materials and rapid detection of foodborne bacterial spores based on SERS

Abstract: The spores of foodborne pathogens are hardly eradicated in the food industry, due to the extreme resistance. The potential risks to food safety have been from the serious foodborne diseases upon germinating and infecting. Taking three common pathogenic C.perfringens, C.difficile, and C. sporogenes spores as the research objects, rapid identification was proposed using Surface-enhanced Raman spectroscopy (SERS). A comparison was made on the enhancing ability of three noble-metal colloids substrate of AuNPs, AgNPs, and Au@AgNPs. An optimal SERS substrate material was then achieved to rapidly identify the fingerprints of foodborne spores. Three noble metal nanoparticle colloids SERS substrates were fabricated via hydrothermal reduction. The irregular spherical structure was obtained with an average size of 82.26 nm AuNPs, 49.83 nm AgNPs, and 57.85 nm Au@AgNPs. Notably, the structure of Au core and Ag shell indicated the successful preparation of Au@AgNPs. The uniform AgNPs and Au@AgNPs with the nano-scale distribution were superior to the AuNPs. Particularly, the homogeneous nanostructure provided a solid foundation for excellent SERS performance and signal reproducibility. The nanoparticle morphologies were characterized by a field-emission scanning electron microscope. The prominent characteristic peaks were selected in the spectra of spores for further analysis of enhancement performance. Among them, the Ca2+-DPA was known as the biomarker for the inner core of spores. The results showed that the optimal SERS performance of Au@AgNPs was achieved to detect the foodborne pathogenic spores with an analytical enhancement factor (AEF) of 3.49×104, indicating excellent spectroscopic specificity and signal reproducibility. According to the Raman shift and characteristic peaks, the characteristic peaks of Ca2+-DPA around 1 017, 1 440, and 1 570 cm-1 were the common specific peaks but with different intensities. Specifically, the C. perfringens spores presented the specific characteristic peaks at around 740, 787, 821, 1 203, 1 308, and 1 627 cm-1, whereas, the specific characteristic peaks at 852 cm-1 were found in the C. sporogenes spores. The intensities of characteristic peaks at 1 017 cm-1(Ca2+-DPA), and 1 081 cm-1 (DNA, O-P-O) were ranked in the order of the C. perfringens>C. sporogenes>C. difficile spores. However, the intensity at around 1332 cm-1 (adenine nucleic acid in nucleic acid) and 1 440 cm-1 (Ca2+-DPA) of C. difficile spores was stronger than that of C. sporogenes spores, but lower than that of C. perfringens spores, indicating the remarkable differences of Raman shifts and intensity in SERS spectra of three spores. A qualitative identification and discrimination model was established to combine the SERS spectra of three spores with the multivariate statistical analysis (PCA and LDA). The PCA demonstrated that the accumulated variance contribution rate for the top three principal components was up to 92.90%, and the recognition rate of LDA analysis reached 100%, fully meeting the specificity and sensitivity. The SERS performances of three noble-metal nanoparticles substrates were utilized to analyze the fingerprints spectra of three spores using SERS and multivariate statistical analysis. A rapid and accurate performance was achieved in the classification and identification of the different strains, indicating the favorable purpose of detection. An effective and reliable risk detection can be expected to serve as the food safety control and rapid identification of microorganisms as a bio-sensor platform. The finding can provide a strong reference for safety control in the subsequent food processing.