Submitted to: Journal of Food Measurement & Characterization
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/17/2012
Publication Date: 4/15/2013
Citation: Sundaram, J., Park, B., Hinton Jr, A., Lawrence, K.C., Kwon, Y. 2013. Detection and differentiation of salmonella serotypes using surface enhanced Raman scattering (SERS) technique. Journal of Food Measurement & Characterization. 7(1):1-12.
Interpretive Summary: Rapid pathogen detection and measurement of cell counts is still needed in food processing to avoid or minimize foodborne pathogenic illness from contaminated food. Surface Enhanced Raman Scattering (SERS) has the potential to rapidly detect pathogens. SERS uses a background substrate to enhance detection limits. In this work, a silver nano particle encapsulated biopolymer colloid was used for the SERS substrate. Bacteria samples with varying cell numbers were prepared and loaded onto the substrate and individually scanned with the laser-light scattering Raman spectroscopy system. Scattered spectral data were then collected and statistical regression analysis was carried out on the spectral data to develop a model that can predict bacterial cell counts. This information is useful for developing methods to detect pathogens without noise from complicating background microflora.
Technical Abstract: Surface Enhanced Raman Scattering (SERS) can detect pathogens rapidly and accurately. The metal surface for the SERS spectroscopy was a silver nano-particle encapsulated biopolymer polyvinyl alcohol nano-colloid deposited on a stainless steel plate. Salmonella Typhimurium and Salmonella Enteritidis were the food pathogens selected for this study and were prepared in different concentrations for testing detection limits. Then a small amount of each concentration was loaded onto the metal substrate, scanned and the spectra were recorded with the confocal Raman spectroscope. A 785-nm excitation laser diode with a 50X object was used to focus the laser light on the sample. Raman-scattered spectral data were obtained from 400 to 2000 cm-1. Partial least square regression (PLS) analysis was used to predict the concentration of Salmonella cells. The PLS calibration model had an r2 of 0.99 and 0.97 for Salmonella Typhimurium and Salmonella Enteritidis, respectively. Likewise, the root mean square error of calibration was 0.034 and 0.106, respectively. Cross validation of the model also gave an r2 of 0.99 and 0.97 for Salmonella Typhimurium and Salmonella Enteritidis, respectively. Root mean square error of calibration and validation had little difference indicating there was no over-fitting of the data. The results showed the feasibility of detecting Salmonella cells at low concentrations with a silver-encapsulated biopolymer substrate for SERS.