Search

Microarrays have been used for gene expression and protein interaction studies, but recently, multianalyte diagnostic assays have employed the microarray platform. We developed a microarray immunoassay for bacteria, with biotinylated capture antibodies on streptavidin slides. To complete the fluorescent sandwich immunoassay following capture of bacteria (Escherichia coli O157:H7), a fluorescein-labeled antibody was used to label antibody-bound bacteria. The assay time was less than 4 h. As this method was developed, it became apparent that several methodological factors markedly affected the results. Therefore, a series of experiments was conducted to investigate methodological factors that affected the assay, including analysis of variation in normal printing, use of a coverslip to contain sample exposure to the microarrayed antibodies, surface chemistries of capture antibody immobilization/attachment and assay reaction time. The use of a coverslip during immunological reactions reduced the fluorescent signal by approximately 50% compared with the use of an uncovered, hydrophobic barrier that was also used to contain sample solution during exposure to the microarray. Also, when protein G was used for capture antibody attachment, a lower signal was generated than when biotinylated capture antibody was used. At high bacterial concentrations (108 or 109 cells/mL), the assay could be shortened to less than 20 min. Antibody microarrays were effectively used to detect bacteria, but assay parameters markedly affected the results and required careful design.

A sandwich method was developed for the detection of Salmonella in shell egg contents. Different outbreak strains of Salmonella were used to inoculate shell egg contents at levels between 1 and 25 cfu/egg. Spiked eggs were then mixed with proper enrichment media before incubation at 37C for 4–20 h. After enrichment, the bacteria were first captured by the use of immunomagnetic beads (IMBs) coated with anti-Salmonella antibodies. The captured bacteria were further labeled with samarium (Sm)-conjugated anti-Salmonella antibodies. Sandwiched Salmonella were then treated with fluorescence enhancement solution that contained strong Sm chelators. The processes ranging from IMB capture to Sm chelation were performed using an automated KingFisher apparatus (Thermo Fisher Scientific, Inc., Waltham, MA). The fluorescence intensity of chelated Sm was then measured by time-resolved fluorescence (TRF). Using this approach, the presence of ∼1 cfu of outbreak strains of Salmonella Enteritidis per egg (∼50 g of egg content) could be detected after enrichment for 20 h at 37C. For higher levels of S. Enteritidis contamination, e.g., 10 cfu/50 g of egg content, the enrichment time could be reduced to 5 h at 37C. Thus, depending on desired sensitivity, the whole detection process could be completed within either an 8-h shift or 24 h. The results demonstrated that a combination of IMB capture and TRF measurement could be a rapid and sensitive method for S. Enteritidis detection in eggs.

The pathogenic bacteria Escherichia coli O157:H7 has been associated with numerous cases of foodborne illnesses. Many rapid methods have been developed for the screening of this pathogen in foods in order to circumvent timely plate culture techniques. However, many of these methods are presumptive and do not claim to confirm the presence of E. coli O157:H7. We developed a rapid method (termed enzyme-linked immunomagnetic chemiluminescence [ELIMCL]) that employed a relatively affordable and portable luminometer with a superparamagnetic microparticle-based enzyme-linked sandwich immunoassay. A key feature of this assay is that it combines the highly selective synergism of both anti-O157 and anti-H7 antibodies in the sandwich immunoassay format, thus allowing for potential confirmation of the pathogen. This work presented the development of ELIMCL that incorporated a one-step sandwich immunoassay that was demonstrated to have a limit of detection of approximately 1 e5–1 e6 live cells/mL of E. coli O157:H7 in pristine-buffered saline in approximately 1 h or at least 50 cells/mL of E. coli O157:H7 inoculated into ground beef and culture enriched for 5.5 h.

A new fluorescent sandwich method for the detection of Escherichia coli O157:H7 in ground beef was developed. Immunomagnetic beads (IMBs) precoated with anti-E. coli O157:H7 antibody were used to capture and concentrate E. coli O157:H7 present in ground beef. Streptavidin-coated fluorescent beads with anti-E. coli O157:H7 to form immunofluorescent beads (IFBs) were used to detect the E. coli O157:H7 captured by the IMBs by forming IMBM–E. coli O157:H7N–IFBO sandwich complexes, where the subscripts M, N and O were variable but integral numbers. This sandwich technique was able to detect 500 cfu of the bacteria in a 0.5-mL sample taken from a quantified and diluted broth culture. The method was then applied to detect E. coli O157:H7 in artificially contaminated ground beef. Known quantities of freshly cultured E. coli O157:H7 cells were added to 25-g ground beef patties held in sterile bags with filters. The inoculated patties were stored at 7C overnight; 75 mL of E. coli broth was added to each sample, and the bags were incubated at 40C with shaking at 160 rpm for 4.5 h. After the enrichment, E. coli O157:H7 in the filtrate was captured and concentrated with IMBs and further conjugated with IFBs for detection. The results demonstrate that, after a 4.5-h enrichment period, the developed procedure can detect the presence of E. coli O157:H7 in ground beef previously spiked with 1 cfu/g.

Escherichia coli O157:H7 was inoculated into ground beef. Samples were incubated in E. coli broth plus novobiocin growth media at 37C for 5 h. Bacteria were captured by magnetic beads coated with anti-E. coli O157:H7 antibodies and were treated with menadione, a membrane-permeable electron-transfer shuttering reagent, to oxidize the cellular nicotinamide adenine dinucleotide phosphate (NAD[P]H). Alternatively, bead-captured bacteria were sandwiched by second anti-E. coli O157:H7 antibodies labeled with peroxidase. The level of NAD(P)H and the enzyme activity of antibody-bound peroxidase were then measured by luminol-linked luminescence. The treatment of the E. coli O157:H7 with a bacterial protein extraction reagent diminished the luminescence associated with cellular NAD(P)H but not the luminescence associated with the sandwiched peroxidase label. Thus, the NAD(P)H-linked luminescence, not the luminescence that was associated with the sandwiched complexes, was related to the viability of the E. coli O157:H7.

The cellular ATP content of fourteen freshly harvested bacteria including Bacillus, Campylobacter, Citrobacter, Escherichia, Lactobacillus, Listeria, Pediococcus, Pseudomonas, Salmonella, Streptococcus and Yersinia, was determined using a luciferin-Iuciferase bioluminescence approach. Incubation of bacteria with carbonyl cyanide meta-chlorophenyl hydrazone (CCCP), a membrane protonophore, prior to cell breakage substantially lowered the bioluminescence signals indicating a decrease of cellular ATP content. The addition of CCCP afrer cell breakage had no detectable effect on the ATP levels. This differential effect of CCCP was not observed using heat-killed bacteria, i.e., the ATP content was not affected by CCCP incubation. The CCCP egects on cellular ATP level were detectable in bacterial suspensions with concentrations ranging from 106 to Id CFU/mL. Upon cold storage, the ATP content, but not the population of viable bacteria, decreased. The ATP content could be paaially restored by the addition of glucose. The ATP content restored by the addition of glucose was also sensitive to CCCP treatment. These results demonstrated that viable bacterial cells can be differentiated from dead cells by their responses to membrane protonophores.

In this manuscript we present a modified kinetic model for pathogen capture efficiency (E) variation with mixing time (Tmix) which is applicable for most combinations of immuno-magnetic bead (IMB)-target organism concentrations (e.g., [IMB] ˜ 8 × 106 IMB mL-1; Salmonella enterica serovar Enteritidis, [SE] ˜ 102-108 CFU mL-1). We found that as the ratio of [IMB] to [SE] was decreased from more than 105 to ca. 10-1, the average number of cells captured per IMB SE complex increased from ca. 1 to 4 ± 1. Concurrently, using a modified kinetic model for fitting E variation with Tmix, the maximum possible level of capture (§) was observed to decline from ˜ 100% (§˜ 1; [IMB] > > [SE]) to 23% (§˜ 0.23; [IMB]:[SE] ˜ 0.1). Using a purely kinetic approach we also found that an empirical mass transport term (yobs= (3.1 ± 1.2) × 10-9 mL IMB-1 min 1; averaged across all [SE] tested: [IMB]:[SE] ˜ 200000, 200, 3, and 0.1) did not differ much from yobs reported in earlier work (3.35 ± 0.2 × 10-9 mL min-1 IMB-1) at extremely low [SE] (e.g., ± 100 CFU mL-1). Upon inserting the classically-derived equation for γ into the kinetic model and solving for the effective IMB radius (rIMB), we found that the rIMB was 2.3 ± 0.16 μ (§ S; averaged across all [IMB]:[SE]) which was only 0.5–0.9 μ greater than the rIMB reported by the manufacturer (1.4 ± 0.2 μ). These results support previous work which showed a similar small rIMB deviation possibly due to the hydro-dynamic behavior of the IMBs in relatively dilute buffer suspensions ([IMB ˜ 8 × 106 IMB mL-1) and argues that IMB-based cell capture is almost exclusively a function of mass transport.

Live cells of E. coli O157:H7 were treated with 4′,6-diamidino-2-phenylindole (DAPI), a nucleic acid stain, for observation by epifluorescent microscopy. The treated bacteria, which exhibited minimal growth activity and partially retained respiratory function, were captured by goat anti-E. coli O157 serum coated on the surface of polystyrene based immunomagnetic beads (IMB). The beads with captured bacteria were then concentrated by magnetic separators. The efficiency of this magnetic concentration step was less than that of using high speed centrifugation. The antibody-captured and IMB-immobilized bacteria were then applied on HF-treated, bovine serum albumin (BSA)-coated microscope slides mounted on an automated stage, and magnetically aligned before fluorescence distribution was measured by a cooled CCD attached to an inverted microscope. Since the beads were concentrated and linearly aligned along the edge of the magnetic field, image capture along the edge for a few field widths was sufficient to account for most of captured bacteria. This approach could reduce the enumeration time for and increase the efficiency of bacteria counting by manual efforts. The developed procedure of capturing, concentrating and enumerating E. coli O157:H7 exhibited a similar applicability in buffer, pork carcass wash solution and apple juice as described in this study. The presence of about 100 cells/mL of the bacteria could be detected in 30 min with developed procedure.