""BARDOT could differentiate micro-colonies of Salmonella, E. coli, and Listeria after 8-12 hours of growth. This reduced time to detection using the BARDOT system could improve the response time to foodborne outbreaks.""

Improved Detection Techniques for Foodborne Pathogens: Bacterial Rapid Detection using Optical Scattering Technology (BARDOT)

Investigator: Arun K. Bhunia (Department of Food Science)

Project Report 2010 - 2011

» Download Project Report 2010 - 2011

Project Rationale

The optical light scattering sensor, BARDOT, is a noninvasive label-free detection system that enables identification of bacterial colonies in real-time. In this semi-automated system, a Petri dish containing bacterial colonies is placed in the plate holder and the colony map of the plate is acquired. An incident laser beam sequentially runs through each preselected colony and generates a scatter signature that is collected and compared to the image library for identification. BARDOT has been demonstrated to detect pathogens from inoculated food samples, thereby validating the system's ability to detect pathogens from food. BARDOT has shortened detection time for Salmonella, Listeria, E. coli, and Vibrio by 2-5 days. Continuing enhancement of BARDOT's capabilities should include the generation of a scatter signature library for serovars of Salmonella and Shiga-toxin producing E. coli, and the improvement of BARDOT-based detection of Listeria and Salmonella from food samples using an immunomagnetic bead separation (IMS) method and sample preparation device, along with a pathogen enrichment detection (PED) device. Additionally, a reduction in the size of colony needed for BARDOT analysis would reduce the overall time required for pathogen detection and identification.

Project Objectives

  • Automation of the BARDOT system.
  • Capture of scattering images from different bacterial species and expansion of the database.
  • Begin studies with the immunomagnetic separation (IMS) and pathogen enrichment detection (PED) device.
  • Integration of IMS and PED to BARDOT system.
  • Capture of scattering images from E. coli, Salmonella, and L. monocytogenes grown in meat samples.
  • Improvements in rapid detection by scattering small (<1 mm) bacterial colonies and identification to the genus, species, and strain level.
  • Improve and automate the image processing and analysis software.
  • Integration of image analysis software into the packaged BARDOT system.
  • Sending BARDOT prototype for validation by other laboratories.

Project Highlights

Previously, we showed that BARDOT can detect and identify colonies of E. coli O157:H7, L. monocytogenes, Salmonella, and Vibrio. During this past year, we demonstrated that BARDOT is able to differentiate and classify 26 different serovars of Salmonella when cultured on selective xylose-lysine deoxycholate (XLD) agar plates. It was essential to build a scatter image library representing most of the common Salmonella serovars. At present, the library contains scatter signatures of 26 serovars representing 64 strains. Both Salmonella Enteritidis and S. Typhimurium were identified with 98-99% accuracy when cultured on XLD plates prior to BARDOT analysis.

Similarly, BARDOT was evaluated for its ability to distinguish seven common Shiga-toxin producing E. coli (STEC) serovars cultured on four commonly used chromogenic/selective media. BARDOT efficiently differentiated all seven serotypes when grown on SMAC (Sorbitol MacConkey) and Rainbow media, suggesting it could possibly be used for initial typing and screening of suspect Salmonella and E. coli colonies. BARDOT was successfully used with samples that were prepared by both the IMS and the PED device.

A limitation in the total time to detection from a food matrix is the length of time required for colony growth on a Petri dish. We examined the possibility of reducing the detection time of the current BARDOT system by detecting and identifying micrometer sized bacterial colonies. Pathogens representing different genera (Salmonella, E. coli, and Listeria) were grown for 7 to 8 hours (Salmonella and E. coli) or 12 hours (Listeria) with resulting colony diameters of around 100-150 µm. The existing BARDOT system was modified with a focusing lens such that, at the focal plane, the beam diameter was equivalent to the colony diameter. The three genera were differentiable via different number of rings and the circularity of the scattering patterns. BARDOT could differentiate micro-colonies (100 - 150 µm) of Salmonella, E. coli, and Listeria after 8-12 hours of growth. This reduced time to detection using the BARDOT system could improve the response time to foodborne outbreaks.