Enrichment, Amplification, and Sequence-Based Typing of Salmonella enterica and Other Foodborne Pathogens.

Detection of Salmonella enterica in foods typically involves microbiological enrichment, molecular-based assay, and subsequent isolation and identification of a pure culture. This is ideally followed by strain typing, which provides information critical to the investigation of outbreaks and the attribution of their sources. Pulsed-field gel electrophoresis is the "gold standard" for S. enterica strain typing, but its limitations have encouraged the search for alternative methods, including whole genome sequencing. Both methods typically require a pure culture, which adds to the cost and turnaround time. A more rapid and cost-effective method with sufficient discriminatory power would benefit food industries, regulatory agencies, and public health laboratories. To address this need, a novel enrichment, amplification, and sequence-based typing (EAST) approach was developed involving (i) overnight enrichment and total DNA preparation, (ii) amplification of polymorphic tandem repeat-containing loci with electrophoretic detection, and (iii) DNA sequencing and bioinformatic analysis to identify related strains. EAST requires 3 days or less and provides a strain resolution that exceeds serotyping and is comparable to pulsed-field gel electrophoresis. Evaluation with spiked ground turkey demonstrated its sensitivity (with a starting inoculum of ≤1 CFU/g) and specificity (with unique or nearly unique alleles relative to databases of >1,000 strains). In tests with unspiked retail chicken parts, 3 of 11 samples yielded S. enterica -specific PCR products. Sequence analysis of three distinct typing targets (SeMT1, SeCRISPR1, and SeCRISPR2) revealed consistent similarities to specific serotype Schwarzengrund, Montevideo, and Typhimurium strains. EAST provides a time-saving and cost-effective approach for detecting and typing foodborne S. enterica , and postenrichment steps can be commercially outsourced to facilitate its implementation. Initial studies with Listeria monocytogenes and Shiga toxigenic Escherichia coli suggest that EAST can be extended to these foodborne pathogens.

Short communication: Improved method for centrifugal recovery of bacteria from raw milk applied to sensitive real-time quantitative PCR detection of Salmonella spp.

Centrifugation is widely used to isolate and concentrate bacteria from dairy products before assay. We found that more than 98% of common pathogenic bacteria added to pasteurized, homogenized, or pasteurized homogenized milk were recovered in the pellet after centrifugation, whereas less than 7% were recovered from raw milk. The remaining bacteria partitioned into the cream layer of raw milk within 5 min, and half-saturation of the cream layer required a bacterial load of approximately 5×10(8) cfu/mL. Known treatments (e.g., heat, enzymes or solvents) can disrupt cream layer binding and improve recovery from raw milk, but can also damage bacteria and compromise detection. We developed a simple, rapid agitation treatment that disrupted bacteria binding to the cream layer and provided more than 95% recovery without affecting bacteria viability. Combining this simple agitation treatment with a previously developed real-time quantitative PCR assay allowed the detection of Salmonella spp. in raw milk at 4 cfu/mL within 3 h. To our knowledge, this is the first report of an effective method for achieving high centrifugal recovery of bacteria from raw milk without impairing bacterial viability.

Antibody Microarray for E. coli O157:H7 and Shiga Toxin in Microtiter Plates.

Antibody microarray is a powerful analytical technique because of its inherent ability to simultaneously discriminate and measure numerous analytes, therefore making the technique conducive to both the multiplexed detection and identification of bacterial analytes (i.e., whole cells, as well as associated metabolites and/or toxins). We developed a sandwich fluorescent immunoassay combined with a high-throughput, multiwell plate microarray detection format. Inexpensive polystyrene plates were employed containing passively adsorbed, array-printed capture antibodies. During sample reaction, centrifugation was the only strategy found to significantly improve capture, and hence detection, of bacteria (pathogenic Escherichia coli O157:H7) to planar capture surfaces containing printed antibodies. Whereas several other sample incubation techniques (e.g., static vs. agitation) had minimal effect. Immobilized bacteria were labeled with a red-orange-fluorescent dye (Alexa Fluor 555) conjugated antibody to allow for quantitative detection of the captured bacteria with a laser scanner. Shiga toxin 1 (Stx1) could be simultaneously detected along with the cells, but none of the agitation techniques employed during incubation improved detection of the relatively small biomolecule. Under optimal conditions, the assay had demonstrated limits of detection of ~5.8 × 105 cells/mL and 110 ng/mL for E. coli O157:H7 and Stx1, respectively, in a ~75 min total assay time.

Two methods for increased specificity and sensitivity in loop-mediated isothermal amplification.

The technique of loop-mediated isothermal amplification (LAMP) utilizes four (or six) primers targeting six (or eight) regions within a fairly small segment of a genome for amplification, with concentration higher than that used in traditional PCR methods. The high concentrations of primers used leads to an increased likelihood of non-specific amplification induced by primer dimers. In this study, a set of LAMP primers were designed targeting the prfA gene sequence of Listeria monocytogenes, and dimethyl sulfoxide (DMSO) as well as Touchdown LAMP were employed to increase the sensitivity and specificity of the LAMP reactions. The results indicate that the detection limit of this novel LAMP assay with the newly designed primers and additives was 10 fg per reaction, which is ten-fold more sensitive than a commercial Isothermal Amplification Kit and hundred-fold more sensitive than previously reported LAMP assays. This highly sensitive LAMP assay has been shown to detect 11 strains of Listeria monocytogenes, and does not detect other Listeria species (including Listeria innocua and Listeria invanovii), providing some advantages in specificity over commercial Isothermal Amplification Kits and previously reported LAMP assay.

A high-throughput, precipitating colorimetric sandwich ELISA microarray for Shiga toxins.

Shiga toxins 1 and 2 (Stx1 and Stx2) from Shiga toxin-producing E. coli (STEC) bacteria were simultaneously detected with a newly developed, high-throughput antibody microarray platform. The proteinaceous toxins were immobilized and sandwiched between biorecognition elements (monoclonal antibodies) and pooled horseradish peroxidase (HRP)-conjugated monoclonal antibodies. Following the reaction of HRP with the precipitating chromogenic substrate (metal enhanced 3,3-diaminobenzidine tetrahydrochloride or DAB), the formation of a colored product was quantitatively measured with an inexpensive flatbed page scanner. The colorimetric ELISA microarray was demonstrated to detect Stx1 and Stx2 at levels as low as ~4.5 ng/mL within ~2 h of total assay time with a narrow linear dynamic range of ~1-2 orders of magnitude and saturation levels well above background. Stx1 and/or Stx2 produced by various strains of STEC were also detected following the treatment of cultured cells with mitomycin C (a toxin-inducing antibiotic) and/or B-PER (a cell-disrupting, protein extraction reagent). Semi-quantitative detection of Shiga toxins was demonstrated to be sporadic among various STEC strains following incubation with mitomycin C; however, further reaction with B-PER generally resulted in the detection of or increased detection of Stx1, relative to Stx2, produced by STECs inoculated into either axenic broth culture or culture broth containing ground beef.

DNA extraction protocol for rapid PCR detection of pathogenic bacteria.

Twelve reagents were evaluated to develop a direct DNA extraction method suitable for PCR detection of foodborne bacterial pathogens. Many reagents exhibited strong PCR inhibition, requiring significant dilution of the extract with a corresponding reduction in sensitivity. Most reagents also exhibited much lower recovery of DNA from the gram-positive test organism (Listeria monocytogenes) than from the gram-negative organism (Escherichia coli O157:H7), preventing unbiased detection and quantitation of both organisms. The 5× HotSHOT+Tween reagent exhibited minimal inhibition and high extraction efficiency for both test organisms, providing a 15-min single-tube DNA-extraction protocol suitable for highly sensitive quantitative PCR assays.

Non-stochastic sampling error in quantal analyses for Campylobacter species on poultry products.

Using primers and fluorescent probes specific for the most common food-borne Campylobacter species (Campylobacter jejuni and Campylobacter coli), we developed a multiplex, most probable number (MPN) assay using quantitative PCR (qPCR) as the determinant for binomial detection: i.e., number of p positive pathogen growth responses out of n = 6 observations each of 4 mL (V) per dilution. Working with media washes of thrice frozen-thawed chicken pieces which had been spiked with known levels of C. jejuni and C. coli, we found that about 20% of the experiments had a significant amount of error in the form of either greater than 25% MPN calculation error (Δε) and/or a low apparent recovery rate (R less than 1 = MPN observed ÷ CFU spiked). Assuming such errors were exacerbated by an excessively small n, we examined computer-generated MPN enumeration data from the standpoint of stochastic sampling error (Δ) and found that such binomial-based assays behaved identically to Poisson-based methods (e.g., counting data) except that fewer technical replicates (n) appeared to be required for the same number of cells per test volume (µ). This result implies that the qPCR detection-based MPN protocol discussed herein should accurately enumerate a test population with a µ ≥ 1 using n = 6 observations per dilution. For our protocol, this equates to ≥ 8 cells per 400-500 g of sampled product. Based on this analysis, the error rate we saw in spiked experiments (where µ >> 1) implied a non-stochastic source. In other experiments we present evidence that this source was, at least in part, related to the cell concentration step (i.e., centrifugation). We also demonstrate that the error rate lessened (from ~38% to ~13%) at lower Campylobacter levels (µ ≤ 40) as would most likely exist in nature. Using this protocol, we were able to quantify 14 to 1,226 MPN per 450 g of naturally contaminated chicken for skinless pieces and 11 to 244 MPN per 450 g for wings, breasts, legs, and thighs (skin on) whereupon about 50% of the 29 samples tested negative for both species. Four of these chicken wash samples did have substantially lower Campylobacter levels (1 to 6 MPN per 450 g) which might be better enumerated using a larger n. However, we established that the limit of quantification of this protocol diminishes for n > 6 because one is ever more diluting the sample, or lessening V, to achieve the requisite n.

An antibody microarray, in multiwell plate format, for multiplex screening of foodborne pathogenic bacteria and biomolecules.

Intoxication and infection caused by foodborne pathogens are important problems worldwide, and screening tests for multiple pathogens are needed because foods may be contaminated with multiple pathogens and/or toxic metabolites. We developed a 96-well microplate, multiplex antibody microarray method to simultaneously capture and detect Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium (S. typhimurium), as well as a biomolecule (chicken immunoglobulin G or IgG employed as a proteinaceous toxin analog) in a single sample. Microarrayed spots of capture antibodies against the targeted analytes were printed within individual wells of streptavidin-coated polystyrene 96-multiwell microtiter plates and a sandwich assay with fluorescein- or Cy3-labeled reporter antibodies was used for detection. (Printing was achieved with a conventional microarray printing robot that was operated with custom-developed microplate arraying software.) Detection of the IgG was realized from ca. 5 to 25 ng/mL, and detection of E. coli O157:H7 and S. typhimurium was realized from ca. 10(6) to 10(9) and ca. 10(7) to 10(9) cells/mL, respectively. Multiplex detection of the two bacteria and the IgG in buffer and in culture-enriched ground beef filtrate was established with a total assay (including detection) time of ca. 2.5 h. Detection of S. typhimurium was largely unaffected by high concentrations of the other bacteria and IgG as well as the ground beef filtrate, whereas a small decrease in response was observed for E. coli O157:H7. The multiwell plate, multiplex antibody microarray platform developed here demonstrates a powerful approach for high-throughput screening of large numbers of food samples for multiple pathogens and toxins.

SPR biosensor for the detection of L. monocytogenes using phage-displayed antibody.

Whole cells of Listeria monocytogenes were detected with a compact, surface plasmon resonance (SPR) sensor using a phage-displayed scFv antibody to the virulence factor actin polymerization protein (ActA) for biorecognition. Phage Lm P4:A8, expressing the scFv antibody fused to the pIII surface protein was immobilized to the sensor surface through physical adsorption. A locally constructed fluidics system was used to deliver solutions to the compact, two-channel SPREETA sensor. Specificity of the sensor was tested using common food-borne bacteria and a control phage, M13K07 lacking the scFv fusion on its coat protein. The detection limit for L. monocytogenes whole cells was estimated to be 2 x 10(6)cfu/ml. The sensor was also used to determine the dissociation constant (Kd) for the interaction of phage-displayed scFv and soluble ActA in solution as 4.5 nM.

Development of Listeria monocytogenes-specific immunomagnetic beads using a single-chain antibody fragment.

A method for coupling single-chain antibody fragments (scFvs) to immunomagnetic beads (IMBs) was developed and evaluated using scFvs specific for Listeria monocytogenes. A plasmid vector, pBAD380, was constructed that allowed the expression of histidine-tagged biotinylated scFvs in Escherichia coli. The gene encoding a scFv specific for L. monocytogenes was cloned into pBAD380 and the 6-histidine-tagged biotinylated anti-L. monocytogenes scFvs were coupled to streptavidin-coated IMBs. The ability of the anti-L. monocytogenes scFv-IMBs to capture L. monocytogenes and other Listeria species was evaluated in comparison to commercially available anti-Listeria IMBs. The anti-L. monocytogenes scFv-IMBs displayed higher efficiencies of capture (1.38-19.04%) for most strains of L. monocytogenes than were observed for the anti-Listeria IMBs (0.05-3.35%); also, the anti-L. monocytogenes scFv-IMBs exhibited improved specificity for L. monocytogenes as determined by cell capture efficiency in pure and mixed cultures.

Lattice-Boltzmann simulations of three-dimensional fluid flow on a desktop computer.

The lattice-Boltzmann (LB) method is a cellular automaton approach to simulating fluid flow with many advantages over conventional methods based on the Navier-Stokes equations. It is conceptually simple, amenable to a wide array of boundary conditions, and can be adapted to handle thermal, density, miscibility, and other effects. The LB approach has been used to model a number of fluid systems of interest to analytical chemists, including chromatography columns, micromixers, and electroosmotic pumps. However, widespread use of this tool has been limited, in part because virtually all large-scale 3D simulations in the literature have been executed on supercomputers. This work demonstrates that such simulations can be executed in reasonable periods of time (hours) on a desktop computer using a cross-platform software package that is easy to learn and use. This package incorporates several improvements that enhance the utility of the LB approach, including an algorithm for speeding common calculations by 2 orders of magnitude and a scheme for handling convection-diffusion equations of interest in electrochemical and surface reaction studies.

Single-chain Fv antibody with specificity for Listeria monocytogenes.

Single chain antibodies (scFv) exhibiting specific binding to Listeria monocytogenes strains were isolated from a pool of random scFvs expressed on the surface of filamentous bacteriophages. Positive selection (panning) using L. monocytogenes was used to enrich for phage clones with the desired binding affinity, and negative selection using L. innocua and L. ivanovii was used to remove phages expressing cross-reactive antibody fragments. A single phage clone, P4:A8, was selected using two independent panning schemes. A rapid assay was devised to determine phage antibody binding specificity and was used to develop a selectivity profile for individual phage clones. The P4:A8 clone was screened against a panel of bacteria consisting of eight strains of L. monocytogenes, one each of the other six species of Listeria and nine other relevant bacterial species. A collection of individual clones from the penultimate panning was also screened against a subset of the panel of bacteria. The selectivity profiles indicate that multiple clones, including P4:A8, exhibit binding to one or more strains of L. monocytogenes without cross-reactivity toward any other species in the panel. This is the first report of a species-specific antibody for viable cells of L. monocytogenes (i.e., the ability to bind to L. monocytogenes without cross-reactivity toward any other species of Listeria).

Isolation and concentration of Salmonellae with an immunoaffinity column.

A method for rapidly and selectively isolating Salmonellae from buffer solutions and concentrating the bacteria by a factor of approximately 500 was developed. Anti-Salmonellae antibody was covalently linked to 40 microm polyacrylamide beads to prepare a solid phase with affinity for the bacteria. The beads were packed into 1-mm diameter glass tubes to form a column 20 microl in volume. Buffer containing Salmonellae at concentrations ranging from 10(2) to 10(6)/ml was pumped through the column to trap and concentrate the bacteria. At a flow rate of 50 microl/min, more than 95% of the bacteria introduced to the column were captured, while at 800 microl/min capture dropped to 32%. Specificity was high, with no detectable capture of Escherichia coli at a concentration of 10(5)/ml. Capture of more than 90% of Salmonellae in a 5-ml sample was achieved in 40 min by re-circulating the sample through the column at a flow rate of 500 microl/.

A simple micro-growth assay for enumerating bacteria.

A simple method for nonspecific determination of bacteria concentrations in a variety of liquid samples was developed. The assay was based on the time required for a sample grown in liquid media to reach a threshold turbidity. Samples were combined with media in a covered 96-well microwell plate and the turbidity was monitored in real time as the bacteria grew in a temperature-controlled plate reader. A significant problem with growth in microwells was condensation on the cover, which prevented accurate turbidity measurement. This problem was overcome by coating the cover with a small amount of surface-active agent. Salmonella and E. coli concentrations could be determined with a relative error of approximately 20% at levels from 10 to 10(6) cells/ml (eight replicates). An assay of 10 samples with standards required 10 min to set up and 20 min for data processing using a computer spreadsheet program. Growth time at 37 degrees C ranged from 4 h for samples at 10(7) cells/ml to 16 h for samples at 10 cells/ml.