Unusual modifications of protein biomarkers expressed by plasmid, prophage, and bacterial host of pathogenic Escherichia coli identified using top-down proteomic analysis.

RATIONALE: Pathogenic bacteria often carry prophage (bacterial viruses) and plasmids (small circular pieces of DNA) that may harbor toxin, antibacterial, and antibiotic resistance genes. Proteomic characterization of pathogenic bacteria should include the identification of host proteins and proteins produced by prophage and plasmid genomes. METHODS: Protein biomarkers of two strains of Shiga toxin-producing Escherichia coli (STEC) were identified using antibiotic induction, matrix-assisted laser desorption/ionization tandem time-of-flight (MALDI-TOF-TOF) tandem mass spectrometry (MS/MS) with post-source decay (PSD), top-down proteomic (TDP) analysis, and plasmid sequencing. Alphafold2 was also used to compare predicted in silico structures of the identified proteins to prominent fragment ions generated using MS/MS-PSD. Strain samples were also analyzed with and without chemical reduction treatment to detect the attachment of pendant groups bound by thioester or disulfide bonds. RESULTS: Shiga toxin was detected and/or identified in both STEC strains. For the first time, we also identified the osmotically inducible protein (OsmY) whose sequence unexpectedly had two forms: a full and a truncated sequence. The truncated OsmY terminates in the middle of an α-helix as determined by Alphafold2. A plasmid-encoded colicin immunity protein was also identified with and without attachment of an unidentified cysteine-bound pendant group (~307 Da). Plasmid sequencing confirmed top-down analysis and the identification of a promoter upstream of the immunity gene that is activated by antibiotic induction, that is, SOS box. CONCLUSIONS: TDP analysis, coupled with other techniques (e.g., antibiotic induction, chemical reduction, plasmid sequencing, and in silico protein modeling), is a powerful tool to identify proteins (and their modifications), including prophage- and plasmid-encoded proteins, produced by pathogenic microorganisms.

Effect of drying on oxidation of membrane lipids and expression of genes encoded by the Shiga toxin prophage in Escherichia coli.

Drying processes do not eliminate pathogenic Escherichia coli in foods but induce sublethal injury, which may also induce the Shiga toxin (Stx) prophage. This study investigated the effect of drying on membrane lipid oxidation and stx expression in E. coli. Lipid peroxidation was probed with C11-BODIPY581/591; and stx expression was assayed by quantification of GFP in E. coli O104:H4 Δstx2a:gfp:ampr. Treatment of E. coli with H2O2 oxidized the probe; probe oxidation was also observed after drying and rehydration. Lipid oxidation and the lethality of drying were reduced when cells were dried with trehalose under anaerobic condition; in addition, viability and probe oxidation differed between E. coli AW1.7 and E. coli AW1.7Δcfa. Desiccation tolerance thus relates to membrane lipid oxidation. Drying also resulted in expression of GFP in 5% of the population. Overexpression of gfp and recA after drying and rehydration suggested that the expression of Stx prophage was regulated by the SOS response. Overall, C11-BODIPY581/591 allowed investigation of lipid peroxidation in bacteria. Drying causes lipid oxidation, DNA damage and induction of genes encoded by the Stx prophage in E. coli.

Combining Mass Spectrometry, Surface Acoustic Wave Interaction Analysis, and Cell Viability Assays for Characterization of Shiga Toxin Subtypes of Pathogenic Escherichia coli Bacteria.

Shiga toxin (Stx)-producing Escherichia coli (STEC) and enterohemorrhagic E. coli (EHEC) as a human pathogenic subgroup of STEC are characterized by releasing Stx AB5-toxin as the major virulence factor. Worldwide disseminated EHEC strains cause sporadic infections and outbreaks in the human population and swine pathogenic STEC strains represent greatly feared pathogens in pig breeding and fattening plants. Among the various Stx subtypes, Stx1a and Stx2a are of eminent clinical importance in human infections being associated with life-threatening hemorrhagic colitis and hemolytic uremic syndrome, whereas Stx2e subtype is associated with porcine edema disease with a generalized fatal outcome for the animals. Binding toward the glycosphingolipid globotriaosylceramide (Gb3Cer) is a common feature of all Stx subtypes analyzed so far. Here, we report on the development of a matched strategy combining (i) miniaturized one-step affinity purification of native Stx subtypes from culture supernatant of bacterial wild-type strains using Gb3-functionalized magnetic beads, (ii) structural analysis and identification of Stx holotoxins by electrospray ionization ion mobility mass spectrometry (ESI MS), (iii) functional Stx-receptor real-time interaction analysis employing the surface acoustic wave (SAW) technology, and (iv) Vero cell culture assays for determining Stx-caused cytotoxic effects. Structural investigations revealed diagnostic tryptic peptide ions for purified Stx1a, Stx2a, and Stx2e, respectively, and functional analysis resulted in characteristic binding kinetics of each Stx subtype. Cytotoxicity studies revealed differing toxin-mediated cell damage ranked with Stx1a > Stx2a > Stx2e. Collectively, this matched procedure represents a promising clinical application for the characterization of life-endangering Stx subtypes at the protein level.

Bacteriophage cell lysis of Shiga toxin-producing Escherichia coli for top-down proteomic identification of Shiga toxins 1 & 2 using matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometry.

RATIONAL: Analysis of bacteria by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) often relies upon sample preparation methods that result in cell lysis, e.g. bead-beating. However, Shiga toxin-producing Escherichia coli (STEC) can undergo bacteriophage-induced cell lysis triggered by antibiotic exposure that may allow greater selectivity of the proteins extracted. METHODS: We have developed a sample preparation method for selective extraction of bacteriophage-encoded proteins and specifically Shiga toxins 1 and 2 (Stx1 & 2) expressed from STEC strains induced by DNA-damaging antibiotics. STEC strains were cultured overnight on agar supplemented with ciprofloxacin, mitomycin-C or an iron chelator to induce the bacteriophage lytic cycle with concomitant expression and release of Stx1 and/or Stx2. Sample preparation relied exclusively on bacteriophage lysis for release Stx into the extraction solution. RESULTS: Three clinical STEC strains were analyzed by matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometry (MALDI-TOF-TOF-MS/MS) and top-down proteomics analysis: E. coli O157:H7 strain EDL933, E. coli O91:H21 strain B2F1 and E. coli O26:H11 strain ECRC #05.2217. The B-subunit of Stx1a of EDL933 was detected and identified even though it was ~100-fold less abundant than the B-subunit of Stx2a that had been identified previously for this strain. Two bacteriophage-encoded proteins were also identified: L0117 and L0136. The B-subunits of Stx2d of strain B2F1 and Stx1a of strain ECRC #05.2217 were also detected and identified. CONCLUSIONS: Bacteriophage lysis appeared to enhance the detection sensitivity of Stx for these STEC strains compared to previous work using mechanical lysis. Detection/identification of other bacteriophage-encoded proteins (beyond Stx) tends to support the hypothesis of Stx release by bacteriophage cell lysis.

Effect of the food matrix on pressure resistance of Shiga-toxin producing Escherichia coli.

The pressure resistance of Shiga-toxin producing Escherichia coli (STEC) depends on food matrix. This study compared the resistance of two five-strain E. coli cocktails, as well as the pressure resistant strain E. coli AW1.7, to hydrostatic pressure application in bruschetta, tzatziki, yoghurt and ground beef at 600 MPa, 20 °C for 3 min and during post-pressure survival at 4 °C. Pressure reduced STEC in plant and dairy products by more than 5 logs (cfu/ml) but not in ground beef. The pH affected the resistance of STEC to pressure as well as the post-pressure survival. E. coli with food constituents including calcium, magnesium, glutamate, caffeic acid and acetic acid were treated at 600 MPa, 20 °C. All compounds exhibited a protective effect on E. coli. The antimicrobial compounds ethanol and phenylethanol enhanced the inactivation by pressure. Calcium and magnesium also performed protective effects on E. coli during storage. Glutamate, glutamine or glutathione did not significantly influence the post-pressure survival over 12 days. Preliminary investigation on cell membrane was further performed through the use of fluorescence probe 1-N-phenylnaphthylamine. Pressure effectively permeabilised cell membrane, whereas calcium showed no effects on membrane permeabilisation.

Comparative study on the high pressure inactivation behavior of the Shiga toxin-producing Escherichia coli O104:H4 and O157:H7 outbreak strains and a non-pathogenic surrogate.

Enterohemorrhagic Escherichia coli strains cause each year thousands of illnesses, which are sometimes accompanied by the hemolytic uremic syndrome, like in the 2011 outbreak in Germany. For preservation thermal pasteurization is commonly used, which can cause unwanted quality changes. To prevent this high pressure treatment is a potential alternative. Within this study, the 2011 outbreak strain O104:H4, an O157:H7 and a non-pathogenic strain (DSM1116) were tested. The cells were treated in buffer (pH 7 and pH 5) and carrot juice (pH 5.1) in a pressure temperature range of 0.1-500 MPa and 20-70 °C. Flow cytometry was used to investigate the pressure impact on cell structures of the strain DSM1116. Both pathogenic strains had a much higher resistance in buffer and carrot juice than the non-pathogenic surrogate. Further, strains cultivated and treated at a lower pH-value showed higher pressure stability, presumably due to variations in the membrane composition. This was confirmed for the strain DSM1116 by flow cytometry. Cells cultivated and treated at pH 5 had a stronger ability to retain their membrane potential but showed higher rates of membrane permeabilization at pressures <200 MPa compared to cells cultivated and treated at pH 7. These cells had the lowest membrane permeabilization rate at around 125 MPa, possibly denoting that variations in the fatty acid composition and membrane fluidity contribute to this stabilization phenomenon.

Shiga-like toxin binds with high avidity to multivalent O-linked blood group P1 determinants on mucin-type fusion proteins.

The binding of Shiga-like toxin 1 (Stx1) and Shiga-like toxin 2 (Stx2) to a mucin-like fusion protein, P-selectin glycoprotein ligand-1/mouse IgG2b (PSGL-1/mIgG2b), carrying multiple copies of the blood group P1 determinant on O-glycans was investigated with western blot and the biosensor Biacore. Chinese hamster ovary K-1 (CHO-K1) cells were stably transfected with linearized plasmids encoding the PSGL-1/mIgG2b fusion protein, the pigeon α1,4-galactosyltransferase (α4Gal-T) and the core 2 ß1,6-N-acetylglucosaminyltransferase (C2GnT-I). Western blot analyses of purified PSGL-1/mIgG2b and liquid chromatography-mass spectrometry (LC-MS) of released O-glycans confirmed the presence of the P1 determinant. Western blot analysis indicated strong binding of Stx1, but not Stx2, to PSGL-1/mIgG2b. In a Biacore assay, Stx1 and Stx2 were immobilized on a dextran chip and the binding of purified PSGL-1/mIgG2b and a P(k)-albumin neoglycoprotein was analyzed. Stx1 and Stx2 bound with high avidity to both PSGL-1/mIgG2b and P(k)-albumin, while the Stx1 binding was the strongest. In summary, we have shown that the pigeon α4Gal-T can be aberrantly expressed in CHO cells together with the core 2 enzyme to generate multiple, O-linked P1 determinants on a simultaneously expressed mucin-type fusion protein. P1-decorated PSGL-1/mIgG2b bound with high avidity to both Stx1 and Stx2, and as such constitutes a potential therapeutic inhibitor of these toxins.

Shiga-like toxin B subunit of Escherichia coli as scaffold for high-avidity display of anti-immunocomplex peptides.

Small compounds cannot bind simultaneously to two antibodies, and thus, their immunodetection is limited to competitive formats in which the analyte is indirectly quantitated by measuring the unoccupied antibody binding sites using a competing reporter. This limitation can be circumvented by using phage-borne peptides selected for their ability to specifically react with the analyte-antibody immunocomplex, which allows the detection of these small molecules in a noncompetitive format (PHAIA) with increased sensitivity and a positive readout. In an effort to find substitutes for the phage particles in PHAIA, we explore the use of the B subunit of the Shiga-like toxin of Escherichia coli, also known as verotoxin (VTX), as a scaffold for multivalent display of anti-immunocomplex peptides. Using the herbicides molinate and clomazone as model compounds, we built peptide-VTX recombinant chimeras that were produced in the periplasmic space of E. coli as soluble pentamers, as confirmed by multiangle light scattering analysis. These multivalent constructs, which we termed nanopeptamers, were conjugated to a tracer enzyme and used to detect the herbicide-antibody complex in an ELISA format. The VTX-nanopeptamer assays performed with over a 10-fold increased sensitivity and excellent recovery from spiked surface and mineral water samples. The carbon black-labeled peptide-VTX nanopeptamers showed great potential for the development of a lateral-flow test for small molecules with a visual positive readout that allowed the detection of up to 2.5 ng/mL of clomazone.

Top-down proteomic identification of Shiga toxin 2 subtypes from Shiga toxin-producing Escherichia coli by matrix-assisted laser desorption ionization-tandem time of flight mass spectrometry.

We have analyzed 26 Shiga toxin-producing Escherichia coli (STEC) strains for Shiga toxin 2 (Stx2) production using matrix-assisted laser desorption ionization (MALDI)-tandem time of flight (TOF-TOF) tandem mass spectrometry (MS/MS) and top-down proteomic analysis. STEC strains were induced to overexpress Stx2 by overnight culturing on solid agar supplemented with either ciprofloxacin or mitomycin C. Harvested cells were lysed by bead beating, and unfractionated bacterial cell lysates were ionized by MALDI. The A2 fragment of the A subunit and the mature B subunit of Stx2 were analyzed by MS/MS. Sequence-specific fragment ions were used to identify amino acid subtypes of Stx2 using top-down proteomic analysis using software developed in-house at the U.S. Department of Agriculture (USDA). Stx2 subtypes (a, c, d, f, and g) were identified on the basis of the mass of the A2 fragment and the B subunit as well as from their sequence-specific fragment ions by MS/MS (postsource decay). Top-down proteomic identification was in agreement with DNA sequencing of the full Stx2 operon (stx2) for all strains. Top-down results were also compared to a bioassay using a Vero-d2EGFP cell line. Our results suggest that top-down proteomic identification is a rapid, highly specific technique for distinguishing Stx2 subtypes.

Incorporation of a non-human glycan mediates human susceptibility to a bacterial toxin.

AB(5) toxins comprise an A subunit that corrupts essential eukaryotic cell functions, and pentameric B subunits that direct target-cell uptake after binding surface glycans. Subtilase cytotoxin (SubAB) is an AB(5) toxin secreted by Shiga toxigenic Escherichia coli (STEC), which causes serious gastrointestinal disease in humans. SubAB causes haemolytic uraemic syndrome-like pathology in mice through SubA-mediated cleavage of BiP/GRP78, an essential endoplasmic reticulum chaperone. Here we show that SubB has a strong preference for glycans terminating in the sialic acid N-glycolylneuraminic acid (Neu5Gc), a monosaccharide not synthesized in humans. Structures of SubB-Neu5Gc complexes revealed the basis for this specificity, and mutagenesis of key SubB residues abrogated in vitro glycan recognition, cell binding and cytotoxicity. SubAB specificity for Neu5Gc was confirmed using mouse tissues with a human-like deficiency of Neu5Gc and human cell lines fed with Neu5Gc. Despite lack of Neu5Gc biosynthesis in humans, assimilation of dietary Neu5Gc creates high-affinity receptors on human gut epithelia and kidney vasculature. This, and the lack of Neu5Gc-containing body fluid competitors in humans, confers susceptibility to the gastrointestinal and systemic toxicities of SubAB. Ironically, foods rich in Neu5Gc are the most common source of STEC contamination. Thus a bacterial toxin's receptor is generated by metabolic incorporation of an exogenous factor derived from food.

Crystallization and sulfur SAD phasing of AggA, the major subunit of aggregative adherence fimbriae type I from the Escherichia coli strain that caused an outbreak of haemolytic-uraemic syndrome in Germany.

The outbreak of Shiga toxin-producing Escherichia coli O104:H4 infection in Germany in 2011 was associated with significant mortality and morbidity owing to the progressive development of haemolytic-uraemic syndrome. The outbreak strain emerged recently as a result of horizontal transfer events leading to the acquisition of a number of virulence factors. Among them, aggregative adherence fimbriae type I (AAF/I) are considered to be particularly important since they are involved in the initial attachment of bacteria to the intestinal mucosa. Here, the crystallization and preliminary X-ray diffraction analysis of the major subunit of AAF/I, AggA, are reported. Crystallization of recombinant donor-strand complemented AggA was performed by the vapour-diffusion method. The crystals diffracted to 1.55 Å resolution and belonged to the orthorhombic space group C222(1), with unit-cell parameters a = 77.83, b = 80.17, c = 91.42 Å. Despite a low sulfur content of the protein [0.57%(w/w)], sufficiently accurate initial phases were derived from a sulfur SAD experiment.

Shiga toxin glycosphingolipid receptor expression and toxin susceptibility of human pancreatic ductal adenocarcinomas of differing origin and differentiation.

Shiga toxins (Stxs) are composed of an enzymatically active A subunit (StxA) and a pentameric B subunit (StxB) that preferentially binds to the glycosphingolipid (GSL) globo\xadtriaosylceramide (Gb3Cer/CD77) and to a reduced extent to globotetraosylceramide (Gb4Cer). The identification of Gb3Cer as a tumor-associated GSL in human pancreatic cancer prompted us to investigate the expression of Gb3Cer and Gb4Cer in 15 human pancreatic ductal adenocarcinoma cell lines derived from primary tumors and liver, ascites, and lymph node metastases. Thin-layer chromatography overlay assays revealed the occurrence of Gb3Cer in all and of Gb4Cer in the majority of cell lines, which largely correlated with transcriptional expression analysis of Gb3Cer and Gb4Cer synthases. Prominent Gb3Cer and Gb4Cer lipoform heterogeneity was based on ceramides carrying predominantly C16:0 and C24:0/C24:1 fatty acids. Stx2-mediated cell injury ranged from extremely high sensitivity (CD(50) of 0.94 pg/ml) to high refractiveness (CD(50) of 5.8 µg/ml) and to virtual resistance portrayed by non-determinable CD(50) values even at the highest Stx2 concentration (10 µg/ml) applied. Importantly, Stx2-mediated cytotoxicity did not correlate with Gb3Cer expression (the preferential Stx receptor), suggesting that the GSL receptor content does not primarily determine cell sensitivity and that other, yet to be delineated, cellular factors might influence the responsiveness of cancer cells.

Determination of serotypes of Shiga toxin-producing Escherichia coli isolates by intact cell matrix-assisted laser desorption ionization-time of flight mass spectrometry.

Shiga toxin-producing Escherichia coli (STEC) isolates representing the serotypes O165:H25, O26:H11/H32, and O156:H25 were analyzed by matrix-assisted laser desorption/ionization (MALDI) mass spectra of whole cells, a procedure also known as intact cell mass spectrometry (ICMS or IC-MALDI MS) or MALDI-typing. We demonstrate that within the given species the three serotypes can be well discriminated by ICMS. Conditions for the construction of serotype-specific prototypic mass spectra were systematically optimized by filtering out masses that do not contribute to the discrimination of the serotypes. Binary distances between prototypic spectra and sample spectra were used to determine serotypes of unknown samples. With parameters optimized, only 0.7% of the assignments were incorrect compared to 31% when distances were calculated from alignments of unfiltered mass spectra. Within the different serotypes, clusters of genetically related E. coli most probably originating from single clones could be distinguished by restriction fragment length polymorphism analysis. Since ICMS did not reproduce these clusters, we conclude that the power of ICMS is just sufficient to discriminate E. coli serotypes under certain conditions but fails for the differentiation of E. coli below this level.

Convergent synthesis of the tetrasaccharide repeating unit corresponding to the O-antigen of the verotoxin-producing Escherichia coli O176.

A convergent synthesis of the tetrasaccharide repeating unit of the O-antigen of the verotoxin producing E. coli O176 has been achieved in excellent yield adopting a [2+2] block glycosylation strategy. The ß-D-mannosidic moiety of the tetrasaccharide was prepared from ß-D-glucoside and α-D-galactosamine moiety was derived from D-galactal. The tetrasaccharide was synthesized as its 2-trimethylsilylethyl glycoside in excellent yield. All intermediate steps are high yielding.

Thin-Layer Chromatography in Structure and Recognition Studies of Shiga Toxin Glycosphingolipid Receptors.

Glycosphingolipids (GSLs) consist of a ceramide (Cer) lipid anchor, which is typically composed of the long-chain aminoalcohol sphingosine (d18:1) and a fatty acid (mostly C16-C24) and a sugar moiety harboring to a great extent one to five monosaccharides. GSLs of the globo-series are well-recognized receptors of Shiga toxins (Stxs) released by Stx-producing Escherichia coli (STEC). Receptors for the Stx subtypes Stx1a and Stx2a are globotriaosylceramide (Gb3Cer) and globotetraosylceramide (Gb4Cer), whereby Gb3Cer represents their high-affinity and Gb4Cer their low-affinity receptor. In addition to Gb3Cer and Gb4Cer, Gb5Cer and Forssman GSL are further receptors of the Stx2e subtype rendering Stx2e unique among the various Stx subtypes. Thin-layer chromatography (TLC) is a convenient and ubiquitously employed method for analyzing GSL mixtures of unknown composition. In particular, TLC immunochemical overlay detection allows for sensitive identification of Stx-binding GSLs in complex mixtures directly on the TLC plate. For this purpose, specific anti-GSL antibodies or Stxs themselves in conjunction with anti-Stx antibodies can be used. The described protocols of antibody-mediated detection of TLC-separated globo-series GSLs and corresponding identification of Stx-binding globo-series GSLs will provide detailed advice for successful GSL analysis and particularly highlight the power of the TLC overlay technique.

Isolation and Characterization of Shiga Toxin-Associated Microvesicles.

Microvesicles are shed from cell surfaces during infectious or inflammatory conditions and may contribute to the pathogenesis of disease. During Shiga toxin-producing Escherichia coli (STEC) infection, microvesicles are released from blood cells. These microvesicles play a part in inflammation, thrombosis, hemolysis, and the transfer of the main virulence factor of STEC strains, Shiga toxin, to target organ cells. This chapter describes how to isolate blood cell- and cell culture-derived microvesicles from plasma or cell culture medium, respectively, and how to characterize these microvesicles by various methods, with special focus on Shiga toxin-associated microvesicles.

Concise synthesis of the pentasaccharide O-antigen corresponding to the Shiga toxin producing Escherichia coli O171.

An expedient total synthesis of a pentasaccharide as its 4-methoxyphenyl glycoside corresponding to the Shiga toxin producing Escherichia coli O171 has been achieved for the first time in excellent yield. Most of the glycosylation steps are highly stereoselective. Stereoselective glycosylation of sialic acid derivative was obtained exploiting the nitrile effect of the solvent used.

Zebrafish embryo sensitivity test as in vivo platform to anti-Shiga toxin compound screening.

Shiga toxin-producing Escherichia coli (STEC) pathotype secretes two types of AB5 cytotoxins (Stx1 and Stx2), responsible for complications such as hemorrhagic colitis (HC) and hemolytic uremic syndrome (HUS) in infected patients, which could lead to sequels and death. Currently, there is no effective treatment against the cytotoxic effect of these toxins. However, in order to approve any therapy molecule, an animal experiment is required in order to evaluate the efficacy and safety of therapeutic approaches. The use of alternative small host models is growing among human infectious disease studies, particularly the vertebrate zebrafish model, since relevant results have been described for pathogen-host interaction. In this sense, the present work aimed to analyze the toxic effect of Shiga toxins in zebrafish embryo model in order to standardize this method in the future to be used as a fast, simple, and efficient methodology for the screening of therapeutic molecules. Herein, we demonstrated that the embryos were sensitive in a dose-dependent manner to both Stx toxins, with LD50 of 22 µg/mL for Stx1 and 33 µg/mL for Stx2, and the use of anti-Stx polyclonal antibody abolished the toxic effect. Therefore, this methodology can be a rapid alternative method for selecting promising compounds against Stx toxins, such as recombinant antibodies.

Identifying non-O157 Shiga toxin-producing Escherichia coli (STEC) using deep learning methods with hyperspectral microscope images.

Non-O157 Shiga toxin-producing Escherichia coli (STEC) serogroups such as O26, O45, O103, O111, O121 and O145 often cause illness to people in the United States and the conventional identification of these "Big-Six" are complex. The label-free hyperspectral microscope imaging (HMI) method, which provides spectral "fingerprints" information of bacterial cells, was employed to classify serogroups at the cellular level. In spectral analysis, principal component analysis (PCA) method and stacked auto-encoder (SAE) method were conducted to extract principal spectral features for classification task. Based on these features, multiple classifiers including linear discriminant analysis (LDA), support vector machine (SVM) and soft-max regression (SR) methods were evaluated. Different sizes of datasets were also tested in search for the suitable classification models. Among the results, SAE-based classification models performed better than PCA-based models, achieving classification accuracy of SAE-LDA (93.5%), SAE-SVM (94.9%) and SAE-SR (94.6%), respectively. In contrast, classification results of PCA-based methods such as PCA-LDA, PCA-SVM and PCA-SR were only 75.5%, 85.7% and 77.1%, respectively. The results also suggested the increasing number of training samples have positive effects on classification models. Taking advantage of increasing dataset, the SAE-SR classification model finally performed better than others with average accuracy of 94.9% in classifying STEC serogroups. Specifically, O103 serogroup was classified with the highest accuracy of 97.4%, followed by O111 (96.5%), O26 (95.3%), O121 (95%), O145 (92.9%) and O45 (92.4%), respectively. Thus, the HMI technology coupled with SAE-SR classification model has the potential for "Big-Six" identification.

Using Nanospray Liquid Chromatography and Mass Spectrometry to Quantitate Shiga Toxin Production in Environmental Escherichia coli Recovered from a Major Produce Production Region in California.

A set of 45 environmental strains of Shiga toxin producing Escherichia coli (STEC) from three California counties were analyzed for Shiga toxin production by nanospray liquid chromatography-mass spectrometry and Vero cell bioassay. The STEC in this set comprised six serotypes ((O113:H21, O121:H19, O157:H7, O6:H34, O177:H25, and O185:H7) each containing either the stx2a or stx2c operon. Six of the seven O113:H21 were found to contain two distinct stx2a operons. Eight strains of O157:H7 possessed a stx2c operon whose A subunit gene was interrupted by an insertion sequence (IS1203v). Shiga toxin production was induced by nutrient depletion and quantitated by mass spectrometry. The 37 strains produced Shiga toxins in a near 50-fold range (1.4-49 ng/mL). The IS-interrupted strains expressed low but measurable amounts of the B subunits (0.5-1.9 ng/mL). Another strain possessed an identical stx operon without an IS interruption and produced intact Stx2c (5.7 ng/mL).