Purchase this article with an account.
Ching-Ju Hsieh, Fung-Rong Hu, Da-Wei Wang, Juen-Kai Wang, Yuh-Lin Wang, Chi-Hung Lin; A High Speed Detection Platform Based on Surface-Enhanced Raman Scattering for Rapid Diagnosis of Bacterial Endophthalmitis. Invest. Ophthalmol. Vis. Sci. 2013;54(15):2879.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Bacterial endophthalmits (BE) is a vision-threatening disease. Early diagnosis of causative pathogens can be crucial to optimize final visual prognosis. Conventional culture-based method for diagnosing BE is considered as the gold standard but inevitably takes time ranging from days to weeks or even months. Using the surface-enhanced Raman scattering (SERS) platform developed by our group, the pathogens can be differentiated on the basis of their SERS spectra which are believed related to their surface chemical components. The aim of this study was to develop SERS as a rapid whole-organism fingerprinting method for the characterization of bacteria associated with BE.
We collected the SERS spectra of Gram-positive bacteria (GPB), and Gram-negative bacteria (GNB), including coagulase-negative Staphylococci, Staphylococcus aureus, Streptococcus viridan, Enterococcus faecalis, Pseudomonas aergonisa, Klebsiella pneumoniae, and Proteus mirabilis. These samples were obtained from patients at hospital in Taiwan and were believed to represent the real diversity of clinical pathogens. The Raman signals of bacteria were enhanced by silver/aluminum anodic oxide (Ag/AAO) substrate and collected between 400 and 1600 cm-1 by Raman microscope. The multivariate statistical techniques of linear discriminant analysis (LDA), hierarchical cluster analysis (HCA) and support vector machine (SVM) were applied in order to group these organisms based on their spectral fingerprints.
Each of the individual species had its specific SERS spectrum, and the spectra of GPB also differed from those of GNB. By analyzed the specific SERS spectrum of each individual species using HCA and LDA methods, we found that the individual GPB strains have highly similar spectra with each other and the spectra of P. aergonisa showed most between-strain differences. Using SVM method, the classification accuracy of identifying different bacterial strains can achieve 89% and the mean accuracy of differentiating GPB from GNB was 90%.
The SERS profiles of BE recorded by such a platform are sensitive and stable that could readily reflect different bacterial cell walls found in GPB, GNB, or individual species. We believe this would be the first report showing bacterial discrimination of BE using SERS. This high-speed SERS detection could develop to a novel approach for microbial diagnostics.
This PDF is available to Subscribers Only