当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 感染与免疫杂志 > 2005年 > 第5期 > 正文
编号:11260290
Use of In Vivo-Induced Antigen Technology for Identification of Escherichia coli O157:H7 Proteins Expressed during Human Infection
     Division of Infectious Diseases, Massachusetts General Hospital, Boston, Massachusetts 02114

    Department of Medicine, Harvard Medical School, Boston, Massachusetts 02115

    Harvard Partners Center For Genetics and Genomics, 65 Landsdowne Street, Cambridge, Massachusetts 02139

    College of Dentistry, Department of Oral Biology, University of Florida, Gainesville, Florida 32610

    Departments of Pediatrics and Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri 63110

    Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, Massachusetts 02115

    ABSTRACT

    Using in vivo-induced antigen technology (IVIAT), a modified immunoscreening technique that circumvents the need for animal models, we directly identified immunogenic Escherichia coli O157:H7 (O157) proteins expressed either specifically during human infection but not during growth under standard laboratory conditions or at significantly higher levels in vivo than in vitro. IVIAT identified 223 O157 proteins expressed during human infection, several of which were unique to this study. These in vivo-induced (ivi) proteins, encoded by ivi genes, mapped to the backbone, O islands (OIs), and pO157. Lack of in vitro expression of O157-specific ivi proteins was confirmed by proteomic analysis of a mid-exponential-phase culture of E. coli O157 grown in LB broth. Because ivi proteins are expressed in response to specific cues during infection and might help pathogens adapt to and counter hostile in vivo environments, those identified in this study are potential targets for drug and vaccine development. Also, such proteins may be exploited as markers of O157 infection in stool specimens.

    INTRODUCTION

    Enterohemorrhagic Escherichia coli (EHEC) O157:H7 (O157) is a uniquely human pathogen that causes disease ranging from acute, self-resolving watery diarrhea to hemorrhagic colitis and the potentially fatal hemolytic-uremic syndrome (HUS). Currently, no therapies are available to lessen the potential morbidity and mortality of this infection.

    E. coli O157 is thought to have evolved from a strain of enteropathogenic E. coli (EPEC) O55:H7 bearing the pathogenicity island termed the locus for enterocyte effacement (LEE), through the acquisition of bacteriophages encoding Shiga toxins type 1 (stx1) and/or 2 (stx2), acquisition of a virulence plasmid (pO157), transition of somatic antigen O55 to O157, and loss of sorbitol fermentation and -glucuronidase activity (21). HUS as a complication of E. coli O157 infection has been associated with the presence of the stx2 gene or its variant stx2c in the infecting E. coli O157 strain (21). In addition, the characteristic attaching and effacing (A/E) lesions produced by this organism on the human colonic epithelium are a result of proteins encoded on the LEE, including the adhesion molecule intimin- (Eae), its receptor (Tir), the type III protein secretion system, which secretes a variety of LEE-encoded translocator proteins (EspA, EspB, and EspD) that translocate effectors into host cells, and effector proteins (Tir, EspG, EspF, Map, and EspH) that modulate the host cell cytoskeleton (21). The type III secretion system translocates Tir into the host cell, with subsequent trafficking to the host cell membrane. Intimin binding of Tir leads to host cell actin rearrangement and formation of A/E lesions. Other putative virulence factors are encoded on pO157 and include an enterohemolysin (Ehx), an immunomodulator (Lif), and a serine protease (EspP) (21). Hence several factors may be involved in E. coli O157 pathogenesis, and research is ongoing to understand the complexity of this infection.

    The sequenced E. coli O157 EDL933 genome shows that although this organism shares 4.1 Mb of DNA (termed backbone) with E. coli K-12, it has 1.34 Mb of DNA distributed among 177 DNA segments termed O islands (OIs) that is absent in K-12 (44). Of the genes found in these OIs, only 40% have been assigned a function and several remain to be characterized (44). Collective evidence indicates that intimin- and the Shiga toxins act in concert with other, unidentified virulence factors, encoded by both the OI and backbone sequences, to cause the spectrum of E. coli O157 disease (21, 54).

    To date, the main impediment to identifying a broader complement of virulence factors in this pathogen has been the lack of an animal model that mimics the spectrum of human disease. Also, the potentially fatal sequelae that can follow E. coli O157 infection preclude human volunteer studies. We circumvented these limitations and exploited the human immune response following E. coli O157 infection to identify a panel of microbial factors that might contribute to the pathogenicity of this organism. In particular, we used a modified immunoscreening technique called in vivo-induced antigen technology (IVIAT) (11), which enables identification of antigens expressed specifically during infection but not during growth in standard laboratory media. The rationale was that such immunogenic O157 antigens, expressed in response to unique signals encountered within the gastrointestinal tract, might contribute to pathogen adaptation and survival within the gut and hence might play important roles in the virulence of this organism. Here we report the identification of O157 proteins that are expressed during human infection. We expect the proteins identified to be potential targets for development of diagnostics, drugs, and vaccines.

    MATERIALS AND METHODS

    Recombinant DNA methods. Isolation of plasmid DNA, restriction digestions, and agarose gel electrophoresis were performed using standard procedures (48). All enzymes for restriction digestions, DNA modifications, and ligations were from New England Biolabs, Beverly, MA. DNA sequencing was performed at the DNA Sequencing Core Facility, Department of Molecular Biology, Massachusetts General Hospital, using ABI Prism DiTerminator cycle sequencing with AmpliTaq DNA polymerase FS and an ABI 377 DNA sequencer (Perkin-Elmer Applied Biosystems Division, Foster City, CA). Oligonucleotides for PCR and sequencing were obtained from the DNA Synthesis Core Facility, Department of Molecular Biology, Massachusetts General Hospital. Plasmids were electroporated into E. coli DH5 or BL21(DE3) using a Gene Pulser (Bio-Rad Laboratories, Richmond, CA) as instructed by the manufacturer. Electroporation conditions were 2,500 V at 25-mF capacitance, producing time constants of 4.8 to 4.9 ms.

    Bacterial strains, plasmids, and growth conditions. An isolate of E. coli O157, from a patient who recovered from recent, clinically diagnosed HUS and contributed a serum sample to the pool of convalescent-phase sera for probing the expression library, was used to construct the DNA expression library (see below). E. coli X21-Blue(pEB313) expressed an intracellular derivative of intimin-, His6-intimin- (35), from which the putative signal sequence of 34 amino acids had been removed (a gift of Alison O'Brien, Uniformed Services University of the Health Sciences, Bethesda, MD.). E. coli DH5(pCVD468, pREP4) expressing a genetically engineered version of EspA, His6-EspA (22), was a gift from James B. Kaper, University of Maryland School of Medicine, Baltimore, MD. Bacterial strains were grown in vitro in Luria-Bertani (LB) medium and maintained at –70°C in LB broth containing 15% glycerol. Kanamycin (Kan) and ampicillin (Amp) were used at concentrations of 50 μg/ml and 100 μg/ml, respectively.

    Patient and control sera. Convalescent-phase sera (approximately 500 μl/patient) were obtained from four patients who had recovered from HUS following E. coli O157 infection. The ages of the patients ranged from 2 to 10 years, and sera were collected on days 13 to 96 postillness. A serum sample from a healthy pediatric patient was used as the control. All of the above serum samples were collected at the Children's Hospital and Regional Medical Center, Seattle, WA, for routine laboratory investigations, and only excess sera were used for this study. The Institutional Review Board of the Children's Hospital and Regional Medical Center approved the use of these sera, which were stored at –70°C until use.

    Assessment of reactivities of pooled, unadsorbed HUS convalescent-phase and healthy control sera with immunogenic O157 proteins. Sera were assessed by examining their reactivities via colony immunoblotting (described below) against E. coli XL1-Blue(pEB313), expressing His6-intimin-, plated onto LB-Amp plates (35) and against E. coli DH5(pCVD468, pREP4), expressing His6-EspA, plated onto LB-Amp-Kan plates (22). Both EspA and intimin- are expressed during human infection and targeted by the immune response (22, 35).

    Adsorption of HUS convalescent-phase and control sera. To compensate for variations in immune responses of individual patients and identify the widest array of O157 antigens, equal volumes of HUS convalescent-phase serum samples from four patients were pooled and sequentially adsorbed against the E. coli O157 isolate recovered from one of the four patients (the same E. coli O157 isolate used to generate the expression library).

    The adsorption protocol has been described previously (12). Briefly, a protease inhibitor cocktail formulated for bacterial cells and containing 4-(2-aminoethyl)benzenesulfonyl fluoride (AEBSF; 23 mM), EDTA (100 mM), bestatin (2 mM), pepstatin A (0.3 mM), and E-64 (0.3 mM) was prepared per the manufacturer's (Sigma, St. Louis, MO) instructions and then added to intact cells and cell lysates at a dilution of 1:10. Pooled HUS convalescent-phase sera were sequentially adsorbed against in vitro-grown (LB broth, 37°C) E. coli O157 whole cells, cell lysates (prepared by three cycles of freezing and thawing, followed by sonication), and heat-denatured cell lysates (12). Adsorbed sera were stored at –70°C until further use.

    Individual HUS convalescent-phase serum samples from each of the four patients and the control serum sample were adsorbed against whole cells, cell lysates, and heat-denatured cell lysates of the in vitro-grown (LB-Kan broth, 37°C) expression host, E. coli BL21(DE3) containing the native pET-30abc expression plasmids, in a similar manner.

    The efficiency of adsorption of pooled HUS convalescent-phase sera was assessed using an enzyme immunoassay described previously (12) and detailed below. Adsorption efficiency was further evaluated by reacting sera with recombinant clones expressing His6-intimin- and His6-EspA via colony immunoblotting as described below.

    Efficiency of adsorption of pooled HUS convalescent-phase sera. Microtiter wells were coated with 100 μl of a 1:2 dilution of a lysate of in vitro-grown E. coli O157 (the same isolate used to make the DNA expression library) in 50 mM carbonate buffer (pH 9.6), prepared by three cycles of freezing and thawing followed by sonication. Following overnight incubation at room temperature, wells were washed with phosphate buffered saline (PBS) containing 0.05% of Tween 20 (PBS-T) and blocked with a 1% solution of bovine serum albumin. After a 1-h incubation at 37°C, wells were emptied, and 100 μl dilutions (1:200 to 1:25,600) of sera, removed from the pool after each adsorption step, were added to wells. The wells were incubated for 1 h at 37°C and washed, after which 100 μl of a 1:1,000 dilution of horseradish peroxidase-conjugated goat anti-human affinity-purified immunoglobulin G (IgG), reactive against all classes of human immunoglobulins (ICN, Cappel, Aurora, OH), was added to the wells. Wells were incubated for 1 h at 37°C and washed with PBS-T. Reactions were developed with a 1-mg/ml solution of 2,2'-azinobis(ethylbenzthiazolinesulfonic acid) (ABTS; Sigma, St. Louis, MO) with 0.1% H2O2 (Sigma). The optical density at 405 nm (OD405) was determined kinetically with a Vmax microplate reader (Molecular Devices Corporation, Sunnyvale, Calif.). Plates were read for 5 min at 19-s intervals, and the maximum slope for an OD change of 0.2 U was determined as milli-OD units/min (20).

    Construction of an inducible E. coli O157:H7 genomic DNA expression library. Polymorphic amplified typing sequences, a powerful and user-friendly typing methodology for bacterial pathogens that compares well with pulsed-field gel electrophoresis (24, 25), profiled the cognate E. coli O157 isolates from the four HUS patients as heterogeneous. We therefore selected one isolate at random, purified genomic DNA, and generated the DNA expression library.

    To generate the expression library, vector DNA was prepared by digesting with the restriction enzyme BamHI (New England Biolabs, Beverly, MA). The vectors used were the pET-30abc series of expression vectors (Novagen, Madison, WI), which permit the cloning of inserts in each of the three reading frames under the transcriptional control of the T7 phage promoter. The restriction enzyme-digested plasmid DNA was gel purified using the QIAEX II gel extraction kit (QIAGEN, Valencia, CA) and then treated with shrimp alkaline phosphatase. Genomic DNA of an E. coli O157 isolate from one of the four HUS patients was partially digested with the restriction enzyme Sau3AI. Following fractionation on a 1% agarose gel, DNA fragments ranging from ca. 0.5 to 1.5 kbp (insert DNA) were excised and purified using the QIAEX gel extraction kit (QIAGEN). Various ratios of insert and vector DNA were ligated and used to transform competent E. coli DH5 via electroporation according to standard protocols (48). Transformants were plated onto LB plates supplemented with 50 μg/ml of Kan (LB-Kan). After an overnight incubation at 37°C, growth was scraped off the plates and plasmid DNA was isolated using standard procedures (48) and used to transform electrocompetent E. coli BL21(DE3) (Novagen), a general-purpose expression host. To determine the percentage of transformants containing inserts, the library was plated onto LB-Kan plates, and 100 colonies were randomly picked and analyzed by colony PCR using vector-specific primers. More than 90% of transformants contained inserts ranging from 0.2 kbp to 1.8 kbp.

    Screening of the expression library and identification of clones expressing immunogenic O157 proteins. The expression library was first screened with pooled unadsorbed HUS convalescent-phase sera as follows. An optimal dilution of the library in E. coli BL21(DE3) was plated onto LB-Kan plates to yield ca. 300 to 350 colonies per plate. After 5 h of incubation at 37°C, colonies were lifted using a nitrocellulose filter and placed, colony side up, on fresh LB-Kan plates containing 1 mM isopropyl--D-thiogalactoside. Plates were incubated overnight at 30°C to induce expression of genes contained within cloned inserts. Colonies on plates were partially lysed by exposing them to chloroform vapors for 15 min in a candle jar. The filters were removed from the plates, air dried, and blocked using 5% nonfat milk in PBS (pH 7.4) for 1 h at room temperature. After a rinse with PBS-T, the filters were probed with a 1:500 dilution of pooled unadsorbed HUS convalescent-phase sera for 1 h at room temperature on a rocking platform. Filters were washed three times with PBS-T and incubated with a 1:5,000 dilution of peroxidase-labeled goat IgG directed against the human gamma globulin fraction (ICN/Cappel). Following development with an ECL chemiluminescence kit (Amersham Pharmacia Biotech), reactive clones were identified by their positions on the reference plate (the original plate from which the colonies were lifted, which was also incubated overnight at 30°C). Each reactive clone identified in the primary screen was purified further, picked with a toothpick in a grid pattern such that each test clone alternated with E. coli BL21(DE3)(pET-30a), the negative control, and processed as described above.

    Identification of O157 proteins expressed during human infection by using IVIAT. To identify proteins expressed by E. coli O157 during human infection, the clones identified above were subjected to IVIAT using pooled adsorbed HUS convalescent-phase sera. Clones were picked with a toothpick onto LB-Kan plates in a grid pattern and incubated for 5 h at 37°C. Processing of filters and screening were identical to those described above, except that a 1:100 dilution of pooled adsorbed HUS convalescent-phase sera was used as a probe. Following confirmation by four additional rounds of screening, plasmid inserts were sequenced and encoded proteins were identified via BLAST against the genomic sequences of E. coli O157:H7 strains EDL933 and Sakai. Proteins expressed from inserts within positive clones were called in vivo-induced proteins (ivi proteins), and the genes encoding them were referred to as ivi genes. Positive clones were further probed with individual adsorbed HUS convalescent-phase serum samples from each of the four patients and the control serum sample, as described above.

    Cellular localization of ivi proteins was predicted using the PSORT/PSORT-B program (http://psort.nibb.ac.jp/). Hypothetical ivi proteins were assigned putative functions using the Clusters of Orthologous Groups (COGs) database, available at http://www.ncbi.nlm.nih.gov/COG/. The online browser tool Proteome Navigator was used to compare proteins not assigned specific functions by the COGs database against Prolinks, a database of protein functional linkages derived from coevolution (2), available at http://dip.doe-mbi.ucla.edu/pronav.

    Proteomic analysis of E. coli O157 grown in LB broth using ESIμLC-MS/MS. To determine whether ivi proteins were expressed when E. coli O157 was cultured under standard laboratory conditions, E. coli O157 grown in LB broth was subjected to microcapillary high-performance liquid chromatography combined with electrospray ionization tandem mass spectrometry (ESIμLC-MS/MS) at the Harvard Partners Center for Genetics and Genomics, Cambridge, MA. The E. coli O157 whole cells for one-dimensional μLC-MS/MS analysis were prepared as follows. Mid-log-phase E. coli O157 (OD600, 0.7), cultured at 37°C in LB broth, was pelleted via centrifugation at 4°C and washed twice in deionized water. Cells were aliquoted into 1.5-ml tarred centrifuge tubes, frozen at –80°C, and lyophilized to dryness under a high vacuum. The tubes were then weighed again and the total dried cell pellet weight determined. Dried E. coli O157 pellets (2 mg) were dissolved in 200 μl of 6 M urea, 1% sodium dodecyl sulfate, 100 mM ammonium bicarbonate, and 10 mM dithiothreitol (pH 8.5). Samples were vortexed, and following incubation at 37°C for 1 h, 12 μl of 500 mM iodoacetamide, 100 mM ammonium bicarbonate (pH 8.5) was added to each 200-μl sample. The reaction was allowed to proceed at room temperature for 1 h in the dark. Alkylation was quenched by the addition of 2 μl of 2 M dithiothreitol in 100 mM ammonium bicarbonate, pH 8.5. Samples were then diluted eightfold with 5 mM CaCl2, mixed with 20 μg of Promega sequencing-grade trypsin, and incubated at 37°C for 16 h. Following quenching with 2 μl of formic acid, samples were diluted with 2 ml of 0.1% formic acid and cleaned up using a Waters Oasis MCX cartridge. Peptides were eluted with 6% ammonium hydroxide in 50% acetonitrile, frozen, and lyophilized. Samples were redissolved in 5% acetonitrile-0.1% formic acid/water and loaded onto a 96-well plate for mass spectrometry (MS) analysis.

    For MS, samples were run on an LCQ DECA XP plus Proteome X workstation from Thermo Finnigan. For each run, 85 μl of reconstituted sample was injected with a Surveyor Autosampler, while the separation was done on a 250-μm (inner diameter) by 30-cm column packed with C18 medium running at a 2-μl-per-minute flow rate provided from a Surveyor MS pump with a flow splitter, with a gradient of 5 to 72% water, 0.1% formic acid, and 5% acetonitrile over the course of 240 min (4 h). Two such runs were performed. Between each set of samples, two standards of a 5 Angio mix of peptides (Michrom BioResources) were run to ascertain column performance and observe any potential carryover. The LCQ was run in a top five configuration, with one MS scan and five MS/MS scans. Dynamic exclusion was set to 1, with a limit of 30 seconds. Peptide identifications were made using SEQUEST (Thermo Finnigan) through the Bioworks Browser 3.1. Sequential database searches were made using the E. coli O157:H7 strain EDL933 FASTA database from the European Bioinformatics Institute (http://www.ebi.ac.uk/newt/display) of the European Molecular Biology Laboratory using differential carbamidomethyl-modified cysteines and oxidized methionines. A yeast protein database was spiked in to provide noise and determine the validity of the peptide hits. In this fashion, known and theoretical protein hits can be found without compromising the statistical relevance of all the data (43). Peptide score cutoff values were chosen at cross-correlation values (Xcorr) of 1.8 for singly charged ions, 2.5 for doubly charged ions, and 3.0 for triply charged ions, along with deltaCN values of 0.1 and cross-correlation normalized values of 1. The Xcorr chosen for each peptide ensured a high confidence match for the different charge states, while the delta rank scoring preliminary cutoff ensured the uniqueness of the peptide hit; the RSP value of 1 ensured that the peptide matched the top hit in the preliminary scoring.

    RESULTS AND DISCUSSION

    Pooled unadsorbed HUS convalescent-phase sera reacted specifically with previously identified immunogenic O157 proteins. Pooled unadsorbed HUS convalescent-phase sera reacted strongly and specifically with E. coli XL1-Blue(pEB313) (35) and E. coli DH5(pCVD468, pREP4) (22), expressing His6-intimin- and His6-EspA, respectively, in contrast to E. coli BL21(DE3) expressing recombinant Vibrio cholerae PilA, an irrelevant control protein (Fig. 1A). This suggested that the pool of sera possessed sufficient reactivity for probing the expression library. On the other hand, the unabsorbed healthy-control serum did not react with either of these proteins (data not shown).

    Adsorption of pooled HUS convalescent-phase sera as per the IVIAT protocol resulted in selective depletion of antibodies against O157 antigens expressed in vitro. Adsorption efficiency was determined by examining the reactivities of serum aliquots from pooled HUS convalescent-phase sera after each adsorption step with lysates of in vitro-grown E. coli O157. There was a sharp decline in the reactivities of sera against the lysates of in vitro-grown E. coli O157 following the first adsorption step compared to those of the unadsorbed sera, indicating efficient depletion of antibodies against in vitro-expressed O157 proteins (Fig. 2).

    Although the adsorbed sera continued to react with the clone expressing His6-intimin-, it did not react with the clone expressing His6-EspA (Fig. 1B). Since both are reportedly expressed during human infection (22, 29), as well as during in vitro growth (36), we anticipated reactivity with both might be eliminated by adsorption of the sera. We attribute the residual serum reactivity with intimin- to relatively weak in vitro expression of this protein, which may be insufficient to adsorb away all of the anti-intimin- antibodies generated in response to significantly higher expression of this adhesin within the gastrointestinal tract.

    Screening of an E. coli O157 genomic expression library. Primary screening of ca. 50,000 clones of an E. coli O157 genomic expression library, using pooled unadsorbed HUS convalescent-phase sera, yielded 918 reactive clones. IVIAT of these clones using pooled adsorbed HUS convalescent-phase sera identified 223 persistently reactive clones containing unique inserts as determined from nonredundant databases.

    ivi proteins included previously identified E. coli O157 virulence-related proteins. IVIAT identified four proteins previously reported to have a putative role in E. coli O157 virulence (1) (Table 1). (i) Intimin- is a LEE-encoded outer membrane adhesin that acts in concert with other LEE-encoded proteins to generate the A/E lesion (21) and to effect binding to host nucleolin (50) to tether the bacterium to the enterocyte. (ii) QseA, a backbone-encoded LysR-type quorum-sensing E. coli transcriptional regulator, is part of the regulatory cascade that controls expression of O157 virulence factors via quorum sensing (52). QseA is also present in other gastrointestinal pathogens such as EPEC (QseA) and V. cholerae (AphB). Following activation by the furanone AI-2, QseA activates transcription of Ler, the positive activator of the LEE operon, and thereby influences expression of putative virulence factors from the LEE. A qseA mutant is impaired in the secretion of LEE-encoded proteins via the type III secretion system, also encoded on the LEE (53). (iii) TagA, a pO157-encoded inner membrane lipoprotein (41), is a protein of unknown function, but a putative role in E. coli O157 virulence has been suggested because of its presence in a diverse collection of E. coli O157 strains (41) and the fact that a homolog in V. cholerae is regulated by ToxR, a transcriptional regulator that governs expression of several V. cholerae virulence factors (13). (iv) MsbB2, a pO157-encoded inner membrane acyltransferase, facilitates the synthesis of hexaacyl lipid A, the form with maximal biological activity (23). MsbB2 reportedly functions to suppress minor modifications of lipid A. Acting in conjunction with MsbB1, another homologous acyltransferase expressed from the chromosome, MsbB2 facilitates the synthesis of lipid A of maximal biological activity, which interacts optimally with the host immune system to evoke an immune response to LPS (23). Strongly supporting a role for MsbB2 in E. coli O157 virulence is the fact that LPS reportedly acts synergistically with the Shiga toxins, especially Stx2, in the pathogenesis of HUS (19). Further support for a likely role in E. coli O157 virulence is suggested by observations that MsbB2 contributes to the virulence of related pathogens such as Shigella flexneri and septicemic E. coli O18:K1:H7 strain H16 and that it influences the expression of virulence-related surface structures in diverse pathogens (23).

    ivi proteins expressed from the E. coli O157 backbone. A total of 181 ivi proteins of diverse functional classes were expressed from the backbone (Table 2). Those involved in biosynthesis and metabolism (51.38%) may have functions essential for bacterial growth in vivo, a feature imperative for bacterial pathogenicity. Also, consistent with the anaerobic gut environment, IVIAT identified glycolytic enzymes, hydrogenases involved in fermentation of carbon compounds, an alcohol dehydrogenase, and reductases (including two cryptic nitrate reductases) involved in energy generation from carbohydrates via anaerobic respiration and fermentation (42). These results were expected and consistent with those of other studies of diverse organisms using techniques such as transcriptional profiling (59), in vivo expression technology (15, 33), recombinase in vivo expression technology (4), signature-tagged mutagenesis (14, 37, 38), selective capture of transcribed sequences (SCOTS) (7), and IVIAT of other organisms (5).

    Transport ivi proteins (18.78%) included diverse ABC-type transporters and phosphotransferase systems (PTS), permeases, transport proteins involved in the transport of diverse molecules, iron uptake proteins such as FhuA (a transporter of ferrichrome [Fe3+] and a receptor for phages and colicins), CirA, an iron-regulated receptor for uptake and transport of colicin I, uncharacterized transport proteins, and, in particular, an anaerobically induced permease, TdcC, expressed from the tdc operon, which encodes the transport and degradation of L-threonine and L-serine. Inactivation of the activator, TdcA, results in hyperadherence of E. coli O157 to cultured epithelial cells due to derepression of OmpA expression (54).

    Regulatory proteins (13.81%) comprised, among others, QseA, a LysR-type transcriptional activator (described above) (52), and several two-component regulatory systems that govern the virulence of diverse pathogens (17). Some of these are functionally interlinked, as evidenced by earlier reports (8, 10, 16, 58) and the Prolinks database, suggesting that IVIAT identified proteins that sense and integrate diverse environmental signals (such as anaerobiosis, cation limitation, acid, and excess, toxic levels of extracytoplasmic Fe3+) and help E. coli O157 mount a coordinated cellular adaptive response to counter the hostile host environment. The two-component regulatory systems IVIAT identified were (i) the sensor molecule, NarX, of the NarL-NarX system, which in the absence of oxygen responds to nitrate or nitrite and acts via NarL, the response regulator that activates expression of enzymes involved in nitrate respiration and represses enzymes involved in respiration of other electron acceptors (27, 46); (ii) the sensor kinase component, PhoQ, of the PhoP-PhoQ system, which responds to extracytoplasmic levels of Mg2+ and Ca2+ (involved in the adaptation to Mg2+ limitation) (8) and to Zn2+ excess (10); (iii) the sensor component, BasS, of the BasR-BasS system, which governs the response to excess extracytoplasmic Fe3+ (58) and mild acid pH (51); (iv) the sensor protein, GlnL, of the GlnG-GlnL system, which responds to low ammonia concentrations and stimulates ammonia assimilation (40); and (v) HydH, the sensor for HydG, which primarily responds to high periplasmic Zn2+ and Pb2+ concentrations and nonspecifically activates the expression of hydrogenase 3, an enzyme involved in hydrogen production during fermentation (28). IVIAT also identified an outer membrane lipoprotein, RcsF, that transduces a signal in response to glucose and zinc to the RcsC/YojN/RcsB/RcsA phosphorelay system, which in turn controls the rcs regulon (target genes), encoding enzymes for the colanic acid exopolysaccharide capsule (10), an acid-adaptive response that protects E. coli O157 from environmental stresses such as acid and heat (34). Other ivi proteins that were part of this functional group and likely to impact in vivo survival of E. coli O157 were (i) PrpR, a regulator of the prp operon involved in the catabolism of propionate, a short-chain fatty acid (SCFA) that can be detrimental to E. coli O157 at high concentrations (exposure to SCFA is a stress condition, and catabolism may serve to decrease the concentration of this SCFA [26, 45]); (ii) FucR, a positive regulator of the fuc operon encoding enzymes for metabolism of L-fucose, a component of both mucus and glycans on enterocytes (42) (following experimental inoculation of mice, E. coli O157 reportedly is found attached to both mucus and enterocytes [in contrast to nonpathogenic E. coli, which is found in mucus alone] [39] and may utilize L-fucose as one nutrient source to multiply and outcompete other flora to establish infection); (iii) FliS and FliT, which along with FliD negatively regulate the export of the anti-sigma factor, FlgM, to prevent expression of the flagellar regulon (which may promote in vivo survival, since overproduction of flagella is deleterious to bacterial growth) (61); and (iv) paraquat-inducible protein A (PqiA), an inner membrane protein of uncharacterized function.

    ivi proteins functioning in environmental adaptation (8.84%) included methyl-accepting chemotaxis proteins (MCPs), a protein that was part of the adaptive response to hyperosmolarity, a colicin expressed in response to iron-limiting conditions, a modulator of drug activity, and two proteins of the PhoB regulon expressed as part of the adaptive response to phosphate limitation. Specifically, IVIAT identified two MCPs, namely, Trg, a receptor for the periplasmic ribose and galactose binding proteins, and Tsr, the serine sensor receptor, both of which are regulators of chemotaxis and motility (40). Interestingly, MCPs were also highly expressed during human infection with V. cholerae as identified by IVIAT (12). IVIAT also identified OsmY, a periplasmic protein of unknown function that is induced in response to hyperosmolarity (60). This protein is expressed as part of the Rcs regulon, which includes genes encoding the synthesis of the exopolysaccharide colanic acid capsule (see above) and possibly functions in the transport of an alternative osmolyte (10). One of the ivi proteins in this subgroup was CirA, an iron-regulated receptor for colicin homologous to a siderophore iron uptake system, which is also expressed under iron-limiting conditions by other pathogens such as Salmonella enterica serovar Typhimurium (16). Interestingly, as in previous studies (16, 58), IVIAT identified proteins expressed in response to iron limitation (FhuA and CirA), as well as those expressed in response to extracytoplasmic iron excess (BasS of the BasR-BasS two-component system; see above). Other ivi proteins included MdaA, a modulator of drug activity, and two proteins that are part of the Pho regulon and are expressed as part of the adaptive response to limiting phosphate in the environment, namely, PhoE, an outer membrane porin functioning in the transport of various anions, and a periplasmic phosphate ester hydrolase, PhoA, involved in the degradation of nontransportable organophosphates (40). The PhoB regulon is required for colonization of the rabbit small intestine by V. cholerae (55) and regulates hilA and invasion genes in Salmonella serovar Typhimurium (31). A possible role in E. coli O157 virulence is also suggested by the fact that in vivo expression of PhoB in an avian-pathogenic E. coli (APEC) strain during experimental infection of chickens was identified by SCOTS (7).

    Phage-related proteins (1.11%) included an inner membrane protein of unknown function and a nonspecific protease that degrades the lambda repressor cII. Proteins of unknown function (6.08%) rounded off ivi proteins expressed from the backbone. Collectively, these results suggested that defined backbone ivi proteins not only support pathogenicity by facilitating in vivo survival but also regulate and indirectly complement pathogen-specific virulence factors.

    ivi proteins expressed from OIs. The 37 ivi proteins expressed from OIs included 13 phage-related proteins (Table 3). Because phage proteins include both Stx1 and Stx2 in EHEC, and because they also include proteins that influence every stage of infection of mammalian hosts by diverse pathogens (56), they are potential virulence factors and warrant further evaluation. Although IVIAT did not identify either Stx1 or Stx2 (both are also produced during in vitro growth in LB broth) (47), it identified two homologous ivi proteins of unknown function, one expressed from each of the phages that encode Stx1 and Stx2. These two ivi proteins are homologous to ivi proteins expressed from several other cryptic prophages (Table 3).

    Particularly interesting was the fact that certain OIs (OIs 36 and 71) containing cryptic prophages expressing ivi proteins also expressed nonphage, non-LEE (Nle) effectors (proteins encoded outside the LEE but secreted via the type III secretion apparatus encoded on the LEE) (6). The presence of Nle homologs, such as NleA and NleF (OI 71) and NleB, NleC, and NleD (OI 36), in related pathogens and the requirement of NleA for full virulence of Citrobacter rodentium, a pathogen of mice, suggest a probable role for phage proteins expressed from these OIs in E. coli O157 virulence (9).

    IVIAT identified 24 clones whose inserts encoded proteins expressed from OI sequences that are not part of phage genomes (Table 3). These included intimin-, expressed from the LEE (OI 148); WbdP, a cytoplasmic glycosyltransferase (OI 84) involved in the synthesis of the O-polysaccharide antigen (57); WaaD, a putative periplasmic glycosyltransferase (OI 145) involved in biosynthesis of the oligosaccharide core of LPS; and numerous ivi proteins with putative or unknown functions (Table 3). The identification of enzymes involved in both O-antigen and LPS core biosynthesis by IVIAT was expected, because LPS is a broadly recognized virulence determinant of pathogenic gram-negative bacteria, and transcripts of genes encoding such enzymes were identified during infection of chickens with APEC by using SCOTS (7). Other nonphage ivi proteins that could impact E. coli O157 virulence included a putative arylsulfatase (OI 40) involved in the scavenging of sulfate and implicated in the ability of E. coli K1 to invade brain microvascular endothelial cells (18) and a putative recombinant hot spot A (RhsA) protein (OI 30) that reportedly contributes to genomic plasticity (30). Transcripts of a gene encoding RhsH, a similar protein, were also detected during APEC infection of chickens by SCOTS (7).

    Many nonphage ivi proteins were expressed from seven of the nine large OIs (>15 kb) that reportedly encode putative virulence factors (Table 4) (44). Besides intimin- from OI 148, IVIAT also identified a putative acyl coenzyme A (CoA) synthetase (fatty acid:CoA ligase), expressed from OI 138, which catalyzes the formation of fatty acyl-CoA, a substrate for phospholipid biosynthesis and enzymes of -oxidation, and is involved in diverse functions such as protein transport, protein acylation, enzyme activation, cell signaling, and control of transcription (40); a putative inner membrane ABC-type transport permease expressed from OI 47 and functioning in cell wall biogenesis; a putative inner membrane ABC-type bacteriocin/lantibiotic exporter expressed from OI 28 that functions in the export of large molecules such as proteins and peptides and is homologous to ATP-binding proteins of ABC transporters and toxin secretion systems of several pathogens, including Pseudomonas putida, Salmonella enterica serovar Typhi, and V. cholerae; a cytoplasmic esterase of the - hydrolase superfamily expressed from OI 43; a conserved cytoplasmic protein of unknown function expressed from OI 7; and an inner membrane protein of unknown function expressed from OI 48. Interestingly, IVIAT did not identify clones expressing putative virulence factors encoded on OI 115 and OI 122 (Table 4). Perhaps these proteins are expressed equally in vitro and in vivo, resulting in the removal of corresponding reactive antibodies during adsorption, or these proteins are not immunogenic, or antibodies to these proteins are short-lived.

    ivi proteins expressed from pO157. ivi proteins expressed from pO157 (Table 3) included the previously discussed TagA and MsbB2; SopA, an ATPase that accurately partitions low-copy-number F plasmids into daughter cells (also identified during APEC infection of chickens) (7); a putative nickase associated with plasmid maintenance; and a putative hemolysin expression-modulating protein, a homolog (90% amino acid identity) of the E. coli regulator Hha (3), which complexes with the nucleoid-associated universal regulator protein H-NS and governs expression of the hly operon in response to changes in temperature and osmolarity (32). In addition, Hha has also been shown to repress the LEE-encoded regulator (Ler) in E. coli O157, thereby causing reduced expression of the esp operon encoding the LEE translocator proteins EspA, EspB, and EspD (49).

    A majority of clones expressing ivi proteins reacted specifically and broadly with HUS convalescent-phase serum samples from individual patients. ivi proteins for practical applications, such as the development of diagnostic markers, vaccines, and drugs, should ideally be expressed strongly during infection and evoke robust immune responses broadly in patients with E. coli O157 disease. The majority of the 223 positive clones identified earlier using pooled adsorbed HUS convalescent-phase sera reacted with each of the four individual serum samples that made up the pool but not with a control serum sample taken from a healthy pediatric patient (Tables 1, 2, and 3). However, 15 ivi proteins expressed from the backbone and four ivi proteins expressed from OIs reacted differentially with individual patient serum samples. We are currently investigating via PCR whether the failure of individual patients to respond to a particular ivi protein is attributable to heterogeneity of cognate isolates. Also, 22 backbone (Table 2) and 2 E. coli O157-specific (Table 3) ivi proteins reacted with the control serum. We speculate that this may be due to cross-reacting antibodies in the control sera or to the presence of preexisting antibodies to O157 proteins from prior, unrecognized infection. Studies are ongoing to compare the reactivities of sera from healthy individuals of different age groups to the ivi antigens.

    ivi proteins expressed from OIs and pO157 were not among the 300 O157 proteins most highly expressed during in vitro growth. The central premise of IVIAT is that the proteins identified are expressed specifically during infection but not during growth under standard laboratory conditions. Proteomic analysis using ESIμLC-MS/MS confirmed that none of the 37 ivi proteins expressed from OIs and none of the 5 expressed from pO157 were among the 300 O157 proteins most highly expressed during growth in LB broth (data not shown). This was not entirely expected, because some of the E. coli O157-specific proteins are reportedly expressed, at least to some degree, in vitro (36). We speculate that, owing to low-level expression of such proteins during in vitro culture, an MS run of longer duration would be required for their identification.

    In contrast, 18 of 181 backbone ivi proteins were expressed sufficiently in LB broth for detection by proteomic analysis (Table 5). Of these 18 backbone ivi proteins expressed in vitro, 12 were identified by ESIμLC-MS/MS during the course of both runs; 6 of which were weakly expressed at a percent protein abundance ranging from 0.09 to 0.04%, and the remaining 6 were expressed in only one run (Table 5). We hypothesize that these 18 proteins are expressed at higher levels during human infection than during growth in LB broth and attribute their identification by IVIAT to the fact that low-level protein expression during in vitro growth may not effectively deplete HUS convalescent-phase sera of antibodies against these ivi proteins during absorption.

    Graphical representations of the locations of ivi genes on the E. coli O157 chromosome and on the pO157 plasmid are shown in Fig. 3 and 4, respectively. ivi genes included 181 of 4,029 (4.5%) open reading frames (ORFs) in the backbone, 37 of 1,387 (2.7%) ORFs in the OIs, and 5 of 100 (5%) ORFs in pO157 sequences of the EDL933 genome (44).

    The 181 backbone-specific ivi genes were distributed uniformly on the E. coli O157 chromosome; however, several of the 37 OI-specific ivi genes appeared to localize in three discrete OI groups (a group contained four or more OIs, with each separated by five or fewer intervening OIs). Although in most cases only one ivi gene mapped to an individual OI, there were instances where multiple ivi genes (two or more) mapped within the same OI (OI 36, OI 52, OI 57, OI 89, OI 93). The five ivi genes that mapped to pO157 are also shown in Fig. 4.

    The apparent grouping of OIs expressing ivi proteins raises the possibility that ivi proteins (and other proteins) expressed from OIs within a particular group might act in concert to optimally influence a specific function. Particularly interesting was the fact that group I, which included OI 148, expressing the adhesin intimin-, also included OI 145, expressing the glycosyltransferase WaaD, one of the many enzymes involved in the biosynthesis of LPS core oligosaccharide. The facts that WaaD is expressed from the same operon as WaaI (another enzyme that functions in LPS core oligosaccharide biosynthesis) and that E. coli O157 waaI deletion mutants are hyperadherent to cultured intestinal epithelial cells (54) may suggest that during human infection, intimin- and E. coli O157 LPS (and possibly other ivi and non-ivi proteins, expressed from OIs within this group) might act in concert to modulate the adherence of E. coli O157 to human epithelial cells. It will be interesting to test this hypothesis experimentally and also to determine whether proteins in OI groups II and III might be functionally related as well.

    In conclusion, IVIAT identified 223 O157 proteins expressed in vivo during human infection, several of which were unique to this study. Judged by our results, IVIAT enables identification both of proteins expressed specifically during human infection but not during growth under standard laboratory conditions and of proteins expressed at significantly higher levels in vivo than in vitro. Although IVIAT for E. coli O157 was validated by the identification of previously identified potential E. coli O157 virulence factors, prior infection with E. coli O157 does not necessarily produce full protection from subsequent reinfection (1). This may reflect suboptimal antibody responses to protective antigens, and we hypothesize that robust expression and optimal delivery of relevant ivi proteins (and other O157 antigens) to the mucosal immune system might engender more-protective immune responses. Preliminary experiments demonstrated that all of the 223 reactive clones also reacted with pooled adsorbed sera from patients who had recovered from hemorrhagic colitis (data not shown), suggesting that similar pathogenic mechanisms may be operating in this illness and HUS.

    IVIAT provides a "snapshot" of O157 protein expression during infection and a glimpse of the possible mechanisms by which this pathogen might counter host defenses and adapt and establish itself within the human gut to cause disease. Studies directed toward the characterization of the role of ivi proteins in E. coli O157 pathogenesis are currently under way. The identification of ivi genes unique to diverse E. coli O157 isolates, as well as identification of those unique to non-O157 EHEC or to Shiga-toxigenic E. coli (STEC), which lacks LEE but is pathogenic to humans, and of ivi genes shared between EHEC and EPEC (unpublished data), augurs well for the future development of diagnostic tests for EHEC and STEC infection, as well as for the development of common drugs and vaccines against EHEC, STEC, and EPEC.

    ACKNOWLEDGMENTS

    This work was supported by grants R03 AI53700-01 and R21 AI055963 (to I.T.K.), R01 DE13523 (to M.H.), and R01 DK52081 and R01 AI47499 (to P.I.T.) from the National Institutes of Health.

    Equal contribution.

    REFERENCES

    1. Besser, R. E., P. M. Griffin, and L. Slutsker. 1999. Escherichia coli O157:H7 gastroenteritis and the hemolytic uremic syndrome: an emerging infectious disease. Annu. Rev. Med. 50:355-367.

    2. Bowers, P. M., M. Pellegrini, M. J. Thompson, J. Fierro, T. O. Yeates, and D. Eisenberg. 2004. Prolinks: a database of protein functional linkages derived from coevolution. Genome Biol. 5:R35.

    3. Burland, V., Y. Shao, N. T. Perna, G. Plunkett, H. J. Sofia, and F. R. Blattner. 1998. The complete DNA sequence and analysis of the large virulence plasmid of Escherichia coli O157:H7. Nucleic Acids Res. 26:4196-4204.

    4. Camilli, A., and J. J. Mekalanos. 1995. Use of recombinase gene fusions to identify Vibrio cholerae genes induced during infection. Mol. Microbiol. 18:671-683.

    5. Cao, S. L., A. Progulske-Fox, J. D. Hillman, and M. Handfield. 2004. In vivo induced antigenic determinants of Actinobacillus actinomycetemcomitans. FEMS Microbiol. Lett. 237:97-103.

    6. Deng, W., J. L. Puente, S. Gruenheid, Y. Li, B. A. Vallance, A. Vazquez, J. Barba, J. A. Ibarra, P. O'Donnell, P. Metalnikov, K. Ashman, S. Lee, D. Goode, T. Pawson, and B. B. Finlay. 2004. Dissecting virulence: systematic and functional analyses of a pathogenicity island. Proc. Natl. Acad. Sci. USA 101:3597-3602.

    7. Dozois, C. M., F. Daigle, and R. Curtiss III. 2003. Identification of pathogen-specific and conserved genes expressed in vivo by an avian pathogenic Escherichia coli strain. Proc. Natl. Acad. Sci. USA 100:247-252.

    8. Groisman, E. A. 2001. The pleiotropic two-component regulatory system PhoP-PhoQ. J. Bacteriol. 183:1835-1842.

    9. Gruenheid, S., I. Sekirov, N. A. Thomas, W. Deng, P. O'Donnell, D. Goode, Y. Li, E. A. Frey, N. F. Brown, P. Metalnikov, T. Pawson, K. Ashman, and B. B. Finlay. 2004. Identification and characterization of NleA, a non-LEE-encoded type III translocated virulence factor of enterohaemorrhagic Escherichia coli O157:H7. Mol. Microbiol. 51:1233-1249.

    10. Hagiwara, D., M. Sugiura, T. Oshima, H. Mori, H. Aiba, T. Yamashino, and T. Mizuno. 2003. Genome-wide analyses revealing a signaling network of the RcsC-YojN-RcsB phosphorelay system in Escherichia coli. J. Bacteriol. 185:5735-5746.

    11. Handfield, M., L. J. Brady, A. Progulske-Fox, and J. D. Hillman. 2000. IVIAT: a novel method to identify microbial genes expressed specifically during human infections. Trends Microbiol. 8:336-339.

    12. Hang, L., M. John, M. Asaduzzaman, E. A. Bridges, C. Vanderspurt, T. J. Kirn, R. K. Taylor, J. D. Hillman, A. Progulske-Fox, M. Handfield, E. T. Ryan, and S. B. Calderwood. 2003. Use of in vivo-induced antigen technology (IVIAT) to identify genes uniquely expressed during human infection with Vibrio cholerae. Proc. Natl. Acad. Sci. USA 100:8508-8513.

    13. Harkey, C. W., K. D. Everiss, and K. M. Peterson. 1995. Isolation and characterization of a Vibrio cholerae gene (tagA) that encodes a ToxR-regulated lipoprotein. Gene 153:81-84.

    14. Hava, D. L., and A. Camilli. 2002. Large-scale identification of serotype 4 Streptococcus pneumoniae virulence factors. Mol. Microbiol. 45:1389-1405.

    15. Heithoff, D. M., C. P. Conner, and M. J. Mahan. 1997. Dissecting the biology of a pathogen during infection. Trends Microbiol. 5:509-513.

    16. Heithoff, D. M., C. P. Conner, U. Hentschel, F. Govantes, P. C. Hanna, and M. J. Mahan. 1999. Coordinate intracellular expression of Salmonella genes induced during infection. J. Bacteriol. 181:799-807.

    17. Hoch, J., and T. Silhavy. 1995. Two-component signal transduction. ASM Press, Washington, D.C.

    18. Hoffman, J. A., J. L. Badger, Y. Zhang, S. H. Huang, and K. S. Kim. 2000. Escherichia coli K1 aslA contributes to invasion of brain microvascular endothelial cells in vitro and in vivo. Infect. Immun. 68:5062-5067.

    19. Ikeda, M., S. Ito, and M. Honda. 2004. Hemolytic uremic syndrome induced by lipopolysaccharide and Shiga-like toxin. Pediatr. Nephrol. 19:485-489.

    20. John, M., T. I. Crean, S. B. Calderwood, and E. T. Ryan. 2000. In vitro and in vivo analyses of constitutive and in vivo-induced promoters in attenuated vaccine and vector strains of Vibrio cholerae. Infect. Immun. 68:1171-1175.

    21. Kaper, J. B., J. P. Nataro, and H. L. Mobley. 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol. 2:123-140.

    22. Karpman, D., Z. D. Bekassy, A. C. Sjogren, M. S. Dubois, M. A. Karmali, M. Mascarenhas, K. G. Jarvis, L. J. Gansheroff, A. D. O'Brien, G. S. Arbus, and J. B. Kaper. 2002. Antibodies to intimin and Escherichia coli secreted proteins A and B in patients with enterohemorrhagic Escherichia coli infections. Pediatr. Nephrol. 17:201-211.

    23. Kim, S. H., W. Jia, R. E. Bishop, and C. Gyles. 2004. An msbB homologue carried in plasmid pO157 encodes an acyltransferase involved in lipid A biosynthesis in Escherichia coli O157:H7. Infect. Immun. 72:1174-1180.

    24. Kudva, I. T., P. S. Evans, N. T. Perna, T. J. Barrett, G. J. DeCastro, F. M. Ausubel, F. R. Blattner, and S. B. Calderwood. 2002. Polymorphic amplified typing sequences provide a novel approach to Escherichia coli O157:H7 strain typing. J. Clin. Microbiol. 40:1152-1159.

    25. Kudva, I. T., R. W. Griffin, M. Murray, M. John, N. T. Perna, T. J. Barrett, and S. B. Calderwood. 2004. Insertions, deletions, and single-nucleotide polymorphisms at rare restriction enzyme sites enhance discriminatory power of polymorphic amplified typing sequences, a novel strain typing system for Escherichia coli O157:H7. J. Clin. Microbiol. 42:2388-2397.

    26. Kwon, Y. M., and S. C. Ricke. 1998. Induction of acid resistance of Salmonella typhimurium by exposure to short-chain fatty acids. Appl. Environ. Microbiol. 64:3458-3463.

    27. Lee, A. I., A. Delgado, and R. P. Gunsalus. 1999. Signal-dependent phosphorylation of the membrane-bound NarX two-component sensor-transmitter protein of Escherichia coli: nitrate elicits a superior anion ligand response compared to nitrite. J. Bacteriol. 181:5309-5316.

    28. Leonhartsberger, S., A. Huber, F. Lottspeich, and A. Bock. 2001. The hydH/G genes from Escherichia coli code for a zinc and lead responsive two-component regulatory system. J. Mol. Biol. 307:93-105.

    29. Li, Y., E. Frey, A. M. R. Mackenzie, and B. B. Finlay. 2000. Human response to Escherichia coli O157:H7 infection: antibodies to secreted virulence factors. Infect. Immun. 68:5090-5095.

    30. Lin, R. J., M. Capage, and C. W. Hill. 1984. A repetitive DNA sequence, rhs, responsible for duplications within the Escherichia coli K-12 chromosome. J. Mol. Biol. 177:1-18.

    31. Lucas, R. L., C. P. Lostroh, C. C. DiRusso, M. P. Spector, B. L. Wanner, and C. A. Lee. 2000. Multiple factors independently regulate hilA and invasion gene expression in Salmonella enterica serovar Typhimurium. J. Bacteriol. 182:1872-1882.

    32. Madrid, C., J. M. Nieto, S. Paytubi, M. Falconi, C. O. Gualerzi, and A. Juarez. 2002. Temperature- and H-NS-dependent regulation of a plasmid-encoded virulence operon expressing Escherichia coli hemolysin. J. Bacteriol. 184:5058-5066.

    33. Mahan, M. J., D. M. Heithoff, R. L. Sinsheimer, and D. A. Low. 2000. Assessment of bacterial pathogenesis by analysis of gene expression in the host. Annu. Rev. Genet. 34:139-164.

    34. Mao, Y., M. P. Doyle, and J. Chen. 2001. Insertion mutagenesis of wca reduces acid and heat tolerance of enterohemorrhagic Escherichia coli O157:H7. J. Bacteriol. 183:3811-3815.

    35. McKee, M. L., and A. D. O'Brien. 1996. Truncated enterohemorrhagic Escherichia coli (EHEC) O157:H7 intimin (EaeA) fusion proteins promote adherence of EHEC strains to HEp-2 cells. Infect. Immun. 64:2225-2233.

    36. McNally, A., A. J. Roe, S. Simpson, F. M. Thomson-Carter, D. E. E. Hoey, C. Currie, T. Chakraborty, D. G. E. Smith, and D. L. Gally. 2001. Differences in levels of secreted locus of enterocyte effacement proteins between human disease-associated and bovine Escherichia coli O157. Infect. Immun. 69:5107-5114.

    37. Mei, J., F. Nourbakhsh, C. Ford, and D. Holden. 1997. Identification of Staphylococcus aureus virulence genes in a murine model of bacteremia using signature-tagged mutagenesis. Mol. Microbiol. 26:399-407.

    38. Merrell, D. S., D. L. Hava, and A. Camilli. 2002. Identification of novel factors involved in colonization and acid tolerance of Vibrio cholerae. Mol. Microbiol. 43:1471-1491.

    39. Miranda, R. L., T. Conway, M. P. Leatham, D. E. Chang, W. E. Norris, J. H. Allen, S. J. Stevenson, D. C. Laux, and P. S. Cohen. 2004. Glycolytic and gluconeogenic growth of Escherichia coli O157:H7 (EDL933) and E. coli K-12 (MG1655) in the mouse intestine. Infect. Immun. 72:1666-1676.

    40. Neidhardt, F. C., R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger. 1996. Escherichia coli and Salmonella: molecular and cellular biology. American Society for Microbiology, Washington, D.C.

    41. Paton, A. W., and J. C. Paton. 2002. Reactivity of convalescent-phase hemolytic-uremic syndrome patient sera with the megaplasmid-encoded TagA protein of Shiga-toxigenic Escherichia coli O157. J. Clin. Microbiol. 40:1395-1399.

    42. Peekhaus, N., and T. Conway. 1998. What's for dinner Entner-Doudoroff metabolism in Escherichia coli. J. Bacteriol. 180:3495-3502.

    43. Peng, J., J. E. Elias, C. C. Thoreen, L. J. Licklider, and S. P. Gygi. 2003. Evaluation of multidimensional chromatography coupled with tandem mass spectrometry (LC/LC-MS/MS) for large-scale protein analysis: the yeast proteome. J. Proteome Res. 2:43-50.

    44. Perna, N. T., G. Plunkett, V. Burland, B. Mau, J. D. Glasner, D. J. Rose, G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J. Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis, A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Anantharaman, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner. 2001. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409:529-533.

    45. Polen, T., D. Rittmann, V. F. Wendisch, and H. Sahm. 2003. DNA microarray analyses of the long-term adaptive response of Escherichia coli to acetate and propionate. Appl. Environ. Microbiol. 69:1759-1774.

    46. Rabin, R. S., and V. Stewart. 1992. Either of two functionally redundant sensor proteins, NarX and NarQ, is sufficient for nitrate regulation in Escherichia coli K-12. Proc. Natl. Acad. Sci. USA 89:8419-8423.

    47. Ritchie, J. M., P. L. Wagner, D. W. K. Acheson, and M. K. Waldor. 2003. Comparison of Shiga toxin production by hemolytic-uremic syndrome-associated and bovine-associated Shiga toxin-producing Escherichia coli isolates. Appl. Environ. Microbiol. 69:1059-1066.

    48. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

    49. Sharma, V. K., and R. L. Zuerner. 2004. Role of hha and ler in transcriptional regulation of the esp operon of enterohemorrhagic Escherichia coli O157:H7. J. Bacteriol. 186:7290-7301.

    50. Sinclair, J. F., and A. D. O'Brien. 2002. Cell surface-localized nucleolin is a eukaryotic receptor for the adhesin intimin- of enterohemorrhagic Escherichia coli O157:H7. J. Biol. Chem. 277:2876-2885.

    51. Soncini, F. C., and E. A. Groisman. 1996. Two-component regulatory systems can interact to process multiple environmental signals. J. Bacteriol. 178:6796-6801.

    52. Sperandio, V., A. G. Torres, and J. B. Kaper. 2002. Quorum sensing Escherichia coli regulators B and C (QseBC): a novel two-component regulatory system involved in the regulation of flagella and motility by quorum sensing in E. coli. Mol. Microbiol. 43:809-821.

    53. Sperandio, V., C. C. Li, and J. B. Kaper. 2002. Quorum-sensing Escherichia coli regulator A: a regulator of the LysR family involved in the regulation of the locus of enterocyte effacement pathogenicity island in enterohemorrhagic E. coli. Infect. Immun. 70:3085-3093.

    54. Torres, A. G., and J. B. Kaper. 2003. Multiple elements controlling adherence of enterohemorrhagic Escherichia coli O157:H7 to HeLa cells. Infect. Immun. 71:4985-4995.

    55. von Kruger, W. M. A., S. Humphreys, and J. M. Ketley. 1999. A role for the PhoBR regulatory system homologue in the Vibrio cholerae phosphate-limitation response and intestinal colonization. Microbiology 145:2463-2475.

    56. Wagner, P. L., and M. K. Waldor. 2002. Bacteriophage control of bacterial virulence. Infect. Immun. 70:3985-3993.

    57. Wang, L., and P. R. Reeves. 1998. Organization of Escherichia coli O157 O antigen gene cluster and identification of its specific genes. Infect. Immun. 66:3545-3551.

    58. Wosten, M. M., L. F. Kox, S. Chamnongpol, F. C. Soncini, and E. A. Groisman. 2000. A signal transduction system that responds to extracellular iron. Cell 103:113-125.

    59. Xu, Q., M. Dziejman, and J. J. Mekalanos. 2003. Determination of the transcriptome of Vibrio cholerae during intraintestinal growth and midexponential phase in vitro. Proc. Natl. Acad. Sci. USA 100:1286-1291.

    60. Yim, H. H., and M. Villarejo. 1992. osmY, a new hyperosmotically inducible gene, encodes a periplasmic protein in Escherichia coli. J. Bacteriol. 174:3637-3644.

    61. Yokoseki, T., T. Iino, and K. Kutsukake. 1996. Negative regulation by fliD, fliS, and fliT of the export of the flagellum-specific anti-sigma factor, FlgM, in Salmonella typhimurium. J. Bacteriol. 178:899-901.(Manohar John, Indira T. K)