Detection of pathogenic microorganisms in food is often a tedious and time-consuming exercise. Developing rapid and cost-effective techniques for identifying pathogens to subspecies is critical for tracking causes of foodborne disease outbreaks. The objective of this study was to develop a method for rapid identification and differentiation of Salmonella serovars and strains within these serovars through isolation on hydrophobic grid membrane filters (HGMFs), examination by infrared (IR) spectroscopy and microspectroscopy, and data analysis by multivariate statistical techniques. Salmonella serovars (Anatum, Enteritidis, Heidelberg, Kentucky, Muenchen, and Typhimurium), most of which were represented by multiple strains, were grown in tryptic soy broth (24 h at 42°C), diluted to 102 to 103 CFU/ml, and filtered using HGMFs. The membranes were incubated on Miller-Mallinson agar (24 h at 42°C), and typical Salmonella colonies were sonicated in 50% acetonitrile and centrifuged. Resulting pellets were vacuum dried on a ZnSe crystal and analyzed using IR spectroscopy. Alternatively, the membranes containing Salmonella growth were removed from the agar, vacuum dried, and colonies were analyzed directly by IR microspectroscopy. Soft independent modeling of class analogy (SIMCA) models were developed from spectra. The method was validated by analyzing Salmonella-inoculated tomato juice, eggs, milk, and chicken. Salmonella serovars exhibited distinctive and reproducible spectra in the fingerprint region (1,200 to 900 cm−1) of the IR spectrum. SIMCA permitted distinguishing Salmonella strains from each other through differences in bacterial lipopolysaccharides and other membrane components. The model correctly predicted Salmonella in foods at serovar (100%) and strain (90%) levels. Isolation of Salmonella on HGMF and selective agar followed by IR spectroscopic analysis resulted in rapid and efficient isolation, identification, and differentiation of Salmonella serovars and strains.
Differentiation of Salmonella enterica Serovars and Strains in Cultures and Food Using Infrared Spectroscopic and Microspectroscopic Techniques Combined with Soft Independent Modeling of Class Analogy Pattern Recognition Analysis