The rapid sample processing of extracts from food matrices is an essential component for rapid detection of food pathogens and food safety. The objective of our research is to develop and integrate operational technologies that rapidly and effectively concentrate viable target cell from food matrices and to couple concentration with interrogation for the presence of pathogens. Rapid detection requires rapid sample concentration and amplification of the target population, interrogation of a concentrated sample of cells containing both non-pathogenic and pathogenic organisms, and identification of the type of pathogenic organism should a target population such as Salmonella sp., Listeria sp., or E. coli be detected. This work shows the development of an automated instrument to concentrate and recover cells from both natural flora and artificially spiked organisms from foods is possible using membrane separations. This work describes the properties of food extracts containing microbial cells with respect to fouling of membranes as well as methods that overcome the fouling issues so rapid concentration in less than 30 min may be achieved. A combination of pre- and post-filtration protocols are required so that the instrument itself can be automated, and membrane filtration devices cycled through repeated uses. The utility of this approach has been demonstrated with microorganisms recovered from food samples (specifically chicken rinse), where 500 x in less than 60 min.

The detection of low numbers of organisms in large volumes of liquids is a challenge for both the fermentation and food industries. The detection of microbial contamination or the presence of pathogens requires that the sample be processed, concentrated, and assayed to detect living cels. The rapid concentration and detection of the pathogen, Listeria monocytogenes, from liquid extract of meat is one application where sampling size to achieve adequate detection confidence levels is crucial. The prediction of the minimal sample volume required to enable detection of a specified microorganism must be carefully carried out so that the probability of detection meets pre-determined criteria. We show that detection of 10 to 50 living cells extracted from a 50 g meat sample into 250 mL of buffer can be calculated using the Poisson distribution equation. Using GFP expressing E. coli that can be individually visualized microscopically, we show that a random distribution model accurately represents the probability of detection as a function of sample volume and concentration. This work is generalized to the detection of bacteria in meat, vegetable, and fermentation broth. The significance of these results in the context of rapid detection of pathogens using microfluidic devices for purposes of bioprocess monitoring and control is discussed.

We demonstrate a simple and rapid (1 hour) technique for fabricating microfluidic flow channels using microfibers positioned on a galss or silica surface, and covered with a preformed poly(dimethylsiloxane) (PDMS) that binds to the surface to give a liquid seal. We used this technique to construct hydrohobic microchannels with a microfiber at its center. This design allows a 5 micron wide stream of liquid to be focused along the side of the microfiber. This phenomenon, utilized in combination with a conventional epi-fluorescence microscope and a photometer allows us to count fluorescently labeled bacteria. A model that quantitates both bacterial motility and convective motion due to fluid movement predicts movement of cells in this microfluidic device. Application of the model, combined with the facile assembly of microfluidic channels, enables biosensors to be designed that integrate microfluidic transport, separation, and detection of pathogenic and non-pathogenic microbes.

Rapid concentration and recovery of bacterial cells for the purpose of pathogen detection fluid, derived from meat, may be achieved in less than 30 min using polycarbonate membranes. Concentration of microbial cells accompanied by selective capture of a pathogenic microbe, such as Listeria monocytogenes, from non-pathogenic E. coli, requires a membrane that is functionalized with an antibody capable of selectively capturing the target microbe. We report immobilization techniques that enable attachment of an antibody that is specific to our target organisms, L. monocytogenes, based on Poly-L-Lysine activation of the membrane surface. Subsequent immobilization of the antibody in the presence of glutaraldehyde resulted in a functionalized membrane surface for selective capture of L. monocytogenes from a mixture containing E. coli. The selectivity of the membrane is demonstrated using both imaging and culture techniques. Retention characteristics are modeled based on equilibrium binding of microbe to membrane.

Membrane filtration has been used widely in separation processes for a long time. It has the ability to sort out different substances based on size difference and also concentrate the target into a smaller area. Our previous study showed success in using polycarbonate membrane (PC) to recover our target organism Listeria monocytogenes, it gives us ideas how we can expand this membrane from simple separation tool to detection platform where separation and pathogen detection can be done simultaneously. PC has defined pore size and pore pathway has made PC a very good candidate for screen filter and our antibody immobilization. The microorganism of interest is Listeria monocytogenes. It is detrimental for immunocompromised people, like the elderly, infants and pregnant women. The average death rate is ~28%, but can be as high as 70% for these immunocompromised persons. L. monocytogenes usually occurs in ready-to-eat (RTE) dairy food, such as hotdogs, cheese and milk. All the facts above have led the U.S. Department of Agriculture (USDA) to set up a "zero tolerance" for L. monocytogenes.

There has been a growing interest to combine microbeads-based surfaces with microfluidic devices to provide bead-based surfaces with microfluidic devices to provide bead-based separation, detection or analysis of specific biological species. This paper reports fabrication of functionalized particulate monolayer on a C18 coated SiO2 surface via bio-mediate self-assembly or adsorption. A microchip with C18 surface pre-adsorbed with biotinylated BSA enables rapid self-assembly of streptavidin coated microbeads through specific biotin-streptavidin interaction. When coated with BSA, this microchip surface immobilizes polystyrene or dimethylamino beads through possible non-specific hydrophobic or electrostatic interactions. Protein coated microbeads such as ones coated with anti-Listeria antibody cana lso be immobilized onto a bare C18 surface through hyrophobic interactions. A microbead patterned surface results where the only area capable of binding proteins or microbes is the microbead itself, since biotinylated BSA or BSA pre-adsorbed onto a C18 surface blocks non-specific adsorptions. Fluorescnece microscopy was used in this work to study the adsorption of Escherichia coli and the food pathogen Listeria monocytogenes onto various types of microbeads immobilized on microchip surfaces. Four types of microbead coated surfaces were used: (1) streptavidin microbeads surface pre-adsorbed with biotinylated anti-Listeria antibody; (2) anti-Listeria microbead surfaces pre-blocked with BSA; (3) polystyrene microbead surfaces; and (4) dimethylamino microbead surfaces. Streptavidin microbeads pre-adsorbed with biotinylated anti-Listeria antibody and anti-Listeria antibody coated microbeads showed specific capture of L. monocytogenes while polystyrene microbeads and dimethylamino microbeads captured E. coli and L. monocytogenes non-specifically. The combined use of functionalized microbeads for specific capture and biotinylated BSA or BSA for blocking non-specific adsorption enables development of fully functional microfluidic devices for separation, detection or analysis of specific biological species.

Membrane filtration has offered the advantage of concentrating substances into small volumes or areas. In this study, we utilized polycarbonate membrane filters with defined pore sizes, and paths for filtering innoculated samples and carried out membrane-based detection for Listeria monocytogenes, which is a foodborne pathogen related to several food recals and listeriosis outbreaks. Membranes immobilized with specific antibodies to L. monocytogenes were used to filter innoculates samples. The chemistry utilizes direction reaction of a spacer with the membrane surface, followed by reaction with a bifunctional cross-linker, glutaraldehyde. Polyclonal anti-Listeria antibody was reacted and covalently bound on this surface. Tests with L. monocytogenes showed capture of this bacteria, which is reduced when the blocking agent BSA is added to the mix. Mechanisms for bacterial capture during microfiltration will be discussed.

The design and fabrication of protein biochips requires characterization of blocking agents that minimize non-specific binding of proteins or organisms. Non-specific adsorption of E. coli, Listeria innocua, and Listeria monocytogenes is prevented by BSA or biotinylated BSA adsorbed on SiO2 surfaces of a biochip that had been modified with C18 coating. Biotinylated BSA forms a protein-based surface that in turn binds streptavidin. Since streptavidin has multiple binding sites for biotin, it in turn anchors other biotinylated proteins including antibodies. Henve, biotinylated BSA simultaneously serves as a blocking agent and a foundation for binding an interfcing protein, avidin or streptavidin, which in turn anchors biotinylated antibody, which in our case is antibody C11E9, that binds Listeria spp. Non-specific adsorption of another bacterium, E. coli, is minimized due to the blocking action of the BSA. Derivatization of the chip's surfaces and preparation of protein coated chips for anchoring of antibodies is discussed.

Detection of foodborne pathogens requires that food samples have to be processed in order to remove interferring factors, including food particles, proteins and lipids, and concentration microorganisms that are to be probed for the presence of pathogens. Conventional involving culture stes may require up to 7 days. In the current study, membrane filtration is able to concentrate the foodborne pathogen, Listeria monocytogenes by a factor of 95 x, with 90% recovery of microorganisms by filtering 50 mL of food samples innoculated with Listeria monocytogenes using a syringe filter. Tween 20 was required to prevent irreversible adsorption of the microorganism to the membrane, due to hydrophobic interactions. Polycarbonate, cellulose, nylon and PVDF membranes were tested for their ability to retain Listeria monocytogenes and to separate proteins from microorganisms. The polycarbonate membrane filters with straight through, mono-radial pores were proved to be the most successful one. The results show that Listeria monocytogenes concentrated in this manner gives sufficient volume of sample for processing on a protein biochip where as little as 1 uL of sample is needed.

Biochips offer the promise of quickly detecting foodborne pathogens with time to result of 3 hours or less if the time-consuming microbial enrichment of samples can be avoided. The work addresses the concentration of microbial cells using membrane filtration that can be carried out in 15 minutes. Samples collected from foods are chemically and biologically complex and contain proteins, lipids and fine particles that must be removed to avoid fouling the surface of the biochip. PVDF, nucleopore, nylon and cellulose membranes have been studied for separation of contaminating components and for concentration of cells. The pore size ranges of 0.22 um and 0.45 um were examined. The surface charge of the membrane was negative and did not absorb the cells, since the pH of the samples gave the cells a negative charge, as well. However, capture of the cells by the membrane was also observed due to the membrane structure.