Open Collections

UBC Faculty Research and Publications

Two distinct groups of porcine enteropathogenic Escherichia coli strains of serogroup O45 are revealed… Bruant, Guillaume; Zhang, Yongxiang; Garneau, Philippe; Wong, Justin; Laing, Chad; Fairbrother, John M; Gannon, Victor P; Harel, Josée Aug 26, 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12864_2009_Article_2286.pdf [ 636.92kB ]
Metadata
JSON: 52383-1.0223059.json
JSON-LD: 52383-1.0223059-ld.json
RDF/XML (Pretty): 52383-1.0223059-rdf.xml
RDF/JSON: 52383-1.0223059-rdf.json
Turtle: 52383-1.0223059-turtle.txt
N-Triples: 52383-1.0223059-rdf-ntriples.txt
Original Record: 52383-1.0223059-source.json
Full Text
52383-1.0223059-fulltext.txt
Citation
52383-1.0223059.ris

Full Text

ralssBioMed CentBMC GenomicsOpen AcceResearch articleTwo distinct groups of porcine enteropathogenic Escherichia coli strains of serogroup O45 are revealed by comparative genomic hybridization and virulence gene microarrayGuillaume Bruant†1, Yongxiang Zhang†2, Philippe Garneau1, Justin Wong2, Chad Laing2, John M Fairbrother1, Victor PJ Gannon2 and Josée Harel*1Address: 1Groupe de Recherche sur les Maladies Infectieuses du Porc, Faculté de médecine vétérinaire, Université de Montréal, 3200 rue Sicotte, Saint-Hyacinthe, Québec J2S 7C6, Canada and 2Laboratory for Foodborne Zoonoses, Public Health Agency of Canada, Lethbridge, Alberta, T1J 3Z4, CanadaEmail: Guillaume Bruant - guillaume.bruant@nrc-cnrc.gc.ca; Yongxiang Zhang - Yongxiang.Zhang@inspection.gc.ca; Philippe Garneau - philippe.garneau@umontreal.ca; Justin Wong - justin@brc.ubc.ca; Chad Laing - Chad.Laing@inspection.gc.ca; John M Fairbrother - john.morris.fairbrother@umontreal.ca; Victor PJ Gannon - Vic.Gannon@inspection.gc.ca; Josée Harel* - josee.harel@umontreal.ca* Corresponding author    †Equal contributorsAbstractBackground: Porcine enteropathogenic Escherichia coli (PEPEC) strains of serogroup O45 causepost-weaning diarrhea and produce characteristic attaching and effacing (A/E) lesions. Most O45PEPEC strains possess the locus of enterocyte effacement (LEE), encoding the virulence factorsrequired for production of A/E lesions, and often possess the paa gene, which is thought tocontribute to the early stages of PEPEC pathogenicity. In this study, nine O45 PEPEC strains and arabbit enteropathogenic (REPEC) strain, known to produce A/E lesions in vivo, were characterizedusing an E. coli O157-E. coli K12 whole genome microarray and a virulence gene-specific microarray,and by PCR experiments.Results: Based on their virulence gene profiles, the 10 strains were considered to be atypicalEPEC. The differences in their genomes pointed to the identification of two distinct evolutionarygroups of O45 PEPEC, Groups I and II, and provided evidence for a contribution of these geneticdifferences to their virulence in pigs. Group I included the REPEC strain and four O45 PEPECstrains known to induce severe A/E lesions in challenged pigs whereas Group II was composed ofthe five other O45 PEPEC strains, which induced less severe or no A/E lesions in challenged pigs.Significant differences between Groups I and II were found with respect to the presence or absenceof 50 O-Islands (OIs) or S-loops and 13 K-islands (KIs) or K-loops, including the virulence-associated islands OI#1 (S-loop#1), OI#47 (S-loop#71), OI#57 (S-loop#85), OI#71 (S-loop#108),OI#115, OI#122, and OI#154 (S-loop#253).Conclusion: We have genetically characterized a collection of O45 PEPEC strains and classifiedthem into two distinct groups. The differences in their virulence gene and genomic island contentmay influence the pathogenicity of O45 PEPEC strains, and explain why Group I O45 PEPEC strainsinduced more severe A/E lesions in explants and challenged pigs than Group II strains.Published: 26 August 2009BMC Genomics 2009, 10:402 doi:10.1186/1471-2164-10-402Received: 23 April 2009Accepted: 26 August 2009This article is available from: http://www.biomedcentral.com/1471-2164/10/402© 2009 Bruant et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 14(page number not for citation purposes)BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402BackgroundEscherichia coli of serogroup O45 may be isolated both inintestinal and extraintestinal sites, although they havebeen only sporadically described in the latter [1-3]. Onthe other hand, intestinal E. coli strains have been morefrequently identified as belonging to this serogroup. Intes-tinal O45 E. coli strains have been isolated from animalsand humans and have been classified as both enterotoxi-genic (ETEC) and attaching and effacing (AEEC) E. coli,the latter including both enterohemorrhagic (EHEC) andenteropathogenic (EPEC) E. coli [4-6]. Serogroup O45 isparticularly important among porcine EPEC (PEPEC)strains which cause post-weaning diarrhea (PWD) charac-terized by specific attaching and effacing (A/E) lesions [7-9]). Most O45 PEPEC strains possess the locus of entero-cyte effacement (LEE) pathogenicity island, which con-tains virulence genes necessary for the production of A/Elesions. They also often possess the paa gene (for porcineA/E associated gene), which encodes a virulence factorinvolved in the A/E phenotype and is thought to contrib-ute to the early stages of PEPEC pathogenicity [10]. Thesestrains also have the ability to produce A/E lesions inexperimentally inoculated newborn gnotobiotic pigletsand in a homologous in vitro model using newborn pigletileal explants, as well as to adhere to PK15 porcine kidneycells in vitro [10-14].Genomic islands (GIs) such as LEE are regions of bacterialgenomes that have been acquired by horizontal genetransfer and often contain blocks of genes that functiontogether in specific processes. When the genomes of thetwo E. coli O157:H7 strains EDL933 and Sakai were com-pared with that of E. coli K12 strain MG1655, the GIsfound to be present in strains EDL933 and Sakai butabsent in strain MG1655 were named O-islands (OIs) andSakai loops (S-loops), respectively. The GIs found to bepresent in E. coli K12 but absent from the two E. coliO157:H7 strains were named K-islands (KIs) and K-loops,respectively [15-17]. GIs related to the virulence of a path-ogen are also referred to as pathogenicity islands (PAIs)[18]. In E. coli O157:H7 strain EDL933, several large OIsencode virulence or putative virulence factors. These OIsinclude OI#45 (S-loop#69) and OI#93 (S-loop#153) forShiga toxin 2 and 1, respectively, OI#148 (S-loop#244)for LEE, and OI#57 (S-loop#85) for paa [16,17].A recent microarray-based study has catalogued genomicalterations in a collection of E. coli O157:H7 strains, par-ticularly in GIs, suggesting the existence of two dominantlineages, with characteristics that are unique to each ofthem [19]. Previous studies performed on various AEECstrains have also shown that, depending on their patho-type and host specificity, strains can show variations inpresent study was to examine the genotypic differences,particularly in LEE sequences and chromosomal insertionsites, and in the presence or absence of non LEE-encodedvirulence factors, such as Paa, among a collection of O45PEPEC strains which have been previously shown toinduce A/E lesions in pigs. In this study, we have charac-terized O45 PEPEC strains using a DNA-microarraydesigned specifically for detection of E. coli virulencegenes [22] and compared their genomes using compara-tive genomic hybridization (CGH) and PCR. We identi-fied two distinct groups of PEPEC O45 strains, betweenwhich there were significant variations in GI content.MethodsBacterial strains and preparation of genomic DNANine O45 PEPEC strains, which were isolated at the Fac-ulté de médecine vétérinaire, Saint-Hyacinthe, Québec,Canada, from pigs with PWD [13] were used for themicroarray studies (Table 1). These strains were selectedbased, i) on their ability to produce or not A/E lesions inchallenged pigs [13] and in an homologous ex vivo modelusing newborn piglet ileal explants (data not shown), ii)and on the severity of the A/E manifestation they pro-duced [13] (Table 1). Because of its genetic and pheno-typic similarities with the O45 PEPEC strains [23], theO103 rabbit EPEC (REPEC) strain E22, provided by EricOswald (INRA, Toulouse, France) [24,25], was includedin the study. Five E. coli reference strains were used as con-trols in PCR experiments: the two O157:H7 E. coli strainsEDL933 and Sakai, the K12 strain MG1655, the uropath-ogenic (UPEC) strain CFT073 and the REPEC strainRDEC-1.For DNA preparations, strains were grown overnight in 45mL of Brain-Heart-Infusion (BHI) broth at 37°C. The cul-tures were centrifuged at 8000 rpm for 10 minutes and thepellet was dissolved in 15 mL of 10 mM NaCl, 20 mMTris-HCl (pH 8.0), 1 mM EDTA, 100 μg/mL proteinase Kand 0.5% SDS. This suspension was incubated at 50°C for2 h and DNA was extracted with an equal volume of phe-nol:chloroform:isoamyl alcohol (25:24:1). Followingcentrifugation for 10 min at 8000 rpm, the upper phasewas removed and precipitated by adding 0.1 volume of 3M NaOAc (pH 5.2) and 2 volumes of 99% ethanol. TheDNA precipitate was then spooled out of the solutionusing a sterile glass rod, washed with 70% ethanol, anddissolved in 5 mL of TE (10 mM Tris-HCl, 1 mM EDTA,pH 8.0) buffer.E. coli DNA microarraysFor the whole genome microarray (named E. coliO157:H7 microarray), Corning Ultra-Gap II slides (Corn-ing, Acton, MA) were spotted with the MWG E. coliPage 2 of 14(page number not for citation purposes)their LEE sequences as well as in the site of insertion ofLEE in the chromosome [14,20,21]. The purpose of theO157:H7 array set (MWG Biotech). The MWG array setconsists of 6167 50-mer oligonucleotides covering theBMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402whole genomes of E. coli K-12 strain MG1655 [15] and E.coli O157:H7 strains Sakai (RIMD 0509952) [16] andEDL933 (ATCC700927) [17].The E. coli virulence microarray used in this study wasderived from the one previously developed by Bruant et al.and included 315 70-mer oligonucleotides specific for189 E. coli virulence or putative virulence genes or markersfound in various intestinal and extraintestinal E. colistrains of all known pathotypes [22]. Probes were specificfor genes encoding adhesins; toxins; bacteriocins; anti-aggregative factors; autotransporters; capsular, flagellar,and somatic antigens; hemolysins; invasins; iron acquisi-tion systems or transport proteins; and outer membraneproteins, as well as other genes recently shown to be asso-ciated with virulence in E. coli. This microarray alsodetected genetic variants of particular genes, such as theintimin-encoding gene eae (variants alpha, alpha2, beta,beta2, delta, epsilon, epsilon2, eta, gamma, gamma2,iota, iota2, lambda, mu, nu, pi, xi, and zeta), espA (vari-ants espA1, espA2, and espA3), espB (variants espB1, espB2,and espB3), and tir (variants tir-1, tir-2, and tir-3) from theLEE. Oligonucleotides specific for three variants of themajor fimbrial subunit of the long polar fimbria (LPF)-encoding gene lpfA were also included. These were basedon sequences from the lpfA genes of EPEC strains of sero-group O113 (lpfAO113), OI#141 from E. coli of serotypeO157:H7 (lpfA1), and REPEC strains and E. coli of sero-Microarray hybridizationsPrior to E. coli O157:H7 microarray hybridization, eacharray was pre-hybridized at 50°C in a solution of 5 × SSC,0.1% SDS and 0.1% BSA for 1 h. Following this step,arrays were washed completely in dH2O, rinsed with iso-propanol, and then centrifuged and dried. For hybridiza-tion, 5 μg of test genomic DNA were digested with EcoRVand PstI restriction enzymes, 3 μg of which were labeledwith ULYSIS Alexa Fluor 647 dye (Invitrogen, Burlington,ON). Genomic DNA from strains MG1655, Sakai andEDL933 was digested in an analogous fashion, and 1 μgof the preparation from each strain was combined andlabeled with Alexa Fluor 546 dye (Invitrogen, Burlington,ON). This labeled genomic DNA mixture was then used asa reference for all hybridizations. Unincorporated dye wasremoved using Qiaquick PCR purification kits (Qiagen,Mississauga, ON), according to the manufacturer'sinstructions, and DNA was eluted in 30 μl of 0.1 × TEbuffer. Labeled DNA was vacuum-dried and resuspendedin 20 μl of dH2O. A 70 μl hybridization solution consist-ing of 30% formamide, 5 × SSC, 0.1% SDS, 0.1 mg/mlsonicated Salmon sperm DNA, and equal amounts of testand reference labeled DNAs, each containing at least 30pmol of incorporated dye, was denatured at 95°C for 5min and briefly centrifuged to collect all the contents.DNA preparations were then hybridized overnight (16 h)at 42°C. After hybridization, arrays were washed accord-ing to the modified Corning method (Corning). ArraysTable 1: Characteristics of O45 PEPEC strains and REPEC strain E22 used in this study.Namea Serogroup Origin A/E lesions in explantsb A/E lesions in challenged pigletscExtentd Sitee InflammationfE22g O103:H2 Rabbit ND ND ND NDECL1001 (86–1390) O45:KE65 Pig + ++++ I/Ce/C-P/C-D SECL2004 (81–4420) O45:KE65 Pig + ++++ I/Ce/C-P/C-D LECL2017 (86–4220) O45:KE65 Pig + +++ I/Ce/C-P/C-D MECL2033 (91-19-172) O45:KE65 Pig + ++++ I/Ce/C-P/C-D -ECL2019 (88–1861) O45:KE65 Pig - - - -ECL2020 (88–4299) O45:KE65 Pig + ++ C-P/C-D LECL2027 (89-56-196) O45:KE65 Pig - - - -ECL2076h (87–4725) O45:KE65 Pig +ECL2078 (83–2315) O45:KE65 Pig + ++ D/I/Ce/C-P/C-D La Former names used in a previous study by Zhu et al. [13] are indicated in parentheses.b Unpublished data from studies performed in our laboratory. +, strain tested produced A/E lesions in explants; -, no lesion observed; ND, not determined.c Data from a previous study by Zhu et al. [13].d ++++, extensive bacterial colonization and severe effacement of microvilli; +++, large areas of bacterial colonization and heavy effacement; ++, focal lesions; +, small scattered focal lesions; -, no lesions observed; ND, not determined.e A/E lesions were observed in duodenum (D), ileum (I), cecum (Ce), proximal colon (C-P), and distal colon (C-D); ND, not determined.f L, light inflammatory response; M, mild inflammatory response; S, severe inflammatory response; -, no inflammatory response; ND, not determined.g REPEC reference strain [24,25].h Strain ECL2076 produced A/E lesions in explants but was not included in the study by Zhu et al. [13].Page 3 of 14(page number not for citation purposes)group O26 (lpfAR141). were then scanned with a GenePix 4000B scanner (AxonBMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402Instruments, Redwood City, CA) and processed usingGenePix Pro 5.0. Two slides were hybridized per strainwith a dye-swap repeat per slide.Hybridizations on E. coli virulence microarrays were per-formed as described previously [22]. Arrays were scannedwith a ScanArray® Lite fluorescent microarray analysis sys-tem (Canberra-Packard Canada, Montreal, Quebec) andacquisition and quantification of fluorescent spot intensi-ties were performed using the ScanArray Express® softwareversion 2.1 (Perkin-Elmer, Foster City, CA, USA).Microarray data analysisData obtained from E. coli O157:H7 microarrays werenormalized using the Ratio-based and Lowess methods inAcuity 3.1 (Axon instruments) before analysis. The nor-malized data for all strains were converted into log2 (Fluor647/Fluor 546) in Acuity 3.1 and subsequently analyzedin Microsoft Excel. Control, blank, and test spots with amean intensity below that of the mean of all negative con-trols were removed from the analysis. The arithmeticmean of the remaining spots across the four duplicateswas calculated to construct the dataset. GACK (forGenomotyping Analysis by Charles Kim) [26], was usedto generate a cut-off value determining the presence orabsence of genes, and a dendrogam using the Euclideandistance metric with average linkage was created withtMEV v4.1 [27].For the data obtained from E. coli virulence microarrays,the local background was subtracted from the recordedspot intensities. The median value of each set of triplicatespotted oligonucleotides was then compared to themedian value of the negative control spots present on thearray. Oligonucleotides with a signal-to-noise fluores-cence ratio greater than 2.0 were considered as positive.Microarray data accession numberThe microarray data have been deposited in NCBI's GeneExpression Omnibus (GEO accession number GSE17036)http://www.ncbi.nlm.nih.gov/projects/geo/.PCR experimentsPCR experiments were performed for the nine O45 PEPECstrains and the REPEC strain E22 to determine the locali-zation in their chromosome of the LEE and of the OI#122,as well as the integrity of the OI#122 and of the secondarytype III secretion system gene cluster designated ETT2 (forE. coli type III secretion 2). All PCR experiments were per-formed as described in previous studies carried out on theLEE, OI#122 and ETT2 gene clusters (Additional file 1:Table S1 [21,23,28-31]).CAATCCGAATTACCTC) – nleA-R (TCCATTGCGCG-TATATCAGC) and ECs1812F (CTGTCCAACAGGGATAC)– ECs1812R (CCGCAATCCGAATTACC) for nleA, andnleC-F (AAGTGTAATACGCGCCGTCC) – nleC-R(ATCAGGACTCGCCTCATATC) and ECs0847F (CCCATT-GCTCCTAATCG) – ECs0847R (CAGCGGAATACTCT-GTG) for nleC. The conditions for amplifications were aninitial denaturation of 95°C for 5 min, followed by 30cycles of 95°C for 30 s; 55°C for 30 s; 72°C for 80 s anda final elongation of 72°C for 10 min.ResultsCharacterization of O45 PEPEC strains using the E. coli virulence microarrayAll O45 PEPEC strains and the REPEC strain E22 werecharacterized using the E. coli virulence microarraydescribed previously, which includes probes targeting vir-ulence genes generally found in AEEC but also virulencegenes from the other E. coli virotypes [22]. All strains pos-sessed their own specific virulence gene profile but wereall classified as atypical EPEC (Additional file 2: Table S2[32]). They all possessed the LEE genes and shared thesame LEE profile: eae(β) – espA group I – espB group III –tir group I. In addition, each strain lacked the Shiga toxin2 encoding genes stx2A and stx2B, as well as the bundleforming pili (BFP) encoding gene bfpA and the E. coliadherence factor (EAF) virulence plasmid marker eaf.Remarkably, all O45 PEPEC strains and REPEC strain E22,although stx1A-negative, gave a positive hybridization forthe stx1B gene, which encodes the B subunit of EHECShiga-like toxin 1 and which is generally absent in EPECstrains. However, the presence of the stx1B gene was notconfirmed in PCR experiments (data not shown).The E. coli strains could be classified into two distinctgroups according to their virulence gene pattern (Addi-tional file 2: Table S2 [32]). Group I included the fourPEPEC strains ECL1001, ECL2004, ECL2017, andECL2033, and REPEC strain E22. Group II included thefive other PEPEC strains ECL2019, ECL2020, ECL2027,ECL2076, and ECL2078. Results obtained with the E. colivirulence microarray identified 19 virulence genes thatshowed a non-random distribution between Group I andGroup II strains (Additional file 2: Table S2 [32]). Genesb1121 (encoding a hypothetical protein YcfZ), set (encod-ing a probable enterotoxin, also named ent), tspE4.C2 (ananonymous fragment), efa1 (encoding the EHEC factorfor adherence Efa1), and paa were present in all Group Istrains, including REPEC strain E22, but absent from allGroup II strains. The temperature sensitive hemagglutininencoding gene tsh, the yersiniabactin-related genes fyuA,irp1 and irp2, as well as the PAI-associated gene malX werepresent in all O45 PEPEC strains from Group I but in nei-Page 4 of 14(page number not for citation purposes)PCR experiments were also performed for nleA and nleCgenes. The pairs of primers used were nleA-F (ACCG-ther REPEC strain E22 nor Group II strains. The heat-sta-ble enterotoxin encoding gene astA was present in all O45BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402PEPEC Group I strains and in Group II strain ECL2020 butabsent from all other Group II strains and REPEC strainE22.The genes aidaI (encoding the Adhesin Involved in E. coliDiffuse Adherence), chuA (an iron related gene), ECs1282(encoding a probable filamentous hemagglutinin-likeprotein), rtx (encoding a putative RTX family exoprotein),and yjaA (encoding a hypothetical protein) were presentin all Group II strains but absent from Group I strains,including REPEC strain E22. The gene fepC (encoding aferric enterobactin transport ATP-binding protein) waspresent in three Group II strains (ECL2019, ECL2078 andECL2027) but absent from all other strains.In addition, Group I and Group II strains possessed differ-ent variants of the fliC (encoding the flagellin major sub-unit) and lpfA genes. O45 PEPEC strains from Group I,including REPEC strain E22, possessed the fliC variantflmA54, whereas Group II strains possessed the fliC gene.Group I strains also possessed the lpfAO113 and lpfAR141variants, whereas REPEC strain E22 only possessed thelpfAR141 variant, and Group II strains possessed the lpfA1variant.Previous phylogenetic analyses have shown that most E.coli strains belonged to the four main phylogenetic groupsA, B1, B2, and D. Whereas extraintestinal E. coli strainsbelong mainly to groups B2 and D, most commensal anddiarrheogenic strains belong to group A and group B1[33]. Determination of the phylogenetic groups of theO45 PEPEC strains and REPEC strain E22 was based onthe presence or absence of the two genes chuA and yjaA,and the DNA fragment tspE4.C2, as described by Cler-mont et al. [32]. All Group I strains including REPECstrain E22 were classified in phylogenetic group B1 and allGroup II strains were classified in phylogenetic group B2.CGH-Genomotyping of PEPEC strainsThe CGH-based genomotyping analysis of the nine O45PEPEC strains and the REPEC strain E22 led to their clas-sification into two distinct groups in the same distributionas observed by the E. coli virulence microarray. A dendro-gram based on the analysis of the CGH data for O45PEPEC strains and REPEC strain E22, as well as for the twoO157:H7 strains Sakai and EDL933 is presented in Figure1. The distribution of GIs in O45 PEPEC strains andREPEC strain E22 was investigated by analysis of the CGHdata. Since the microarray used for CGH was not an EPEC-specific microarray and was composed of oligonucleotideprobes specific for genome sequences of O157:H7 EHECand K12 strains, it was not possible to investigate all theGIs in O45 PEPEC strains. The divergences in GIsobserved by CGH could thus indicate either the absenceof particular genes or the presence of different variants ofthese genes. As shown in Table 2, 63 GIs (islands or loops)were found to be significantly different between Group Iand Group II strains. Among these 63 GIs, 13 were KIs orCGH-based genomotyping of the O45 PEPEC strainsFigure 1CGH-based genomotyping of the O45 PEPEC strains. The nine O45 PEPEC strains and the REPEC strain E22 were classified in two distinct groups by CGH-based genomotyping, in the same distribution as observed by the E. coli virulence microarray (Groups I and II). The O157:H7 strains Sakai and EDL933 were used as controls. The tree was constructed with tMEV v4.1 and viewed in SplitsTree 4.1 [27] by using the Euclidean distance, average linkage algorithm and 1,000 bootstrap rep-Group IIGroup IEDL933SakaiECL2017ECL1001ECL2004ECL2033E22ECL2027ECL2019ECL2078ECL2020ECL2076 Group IIGroup IPage 5 of 14(page number not for citation purposes)licates. Bootstrap confidence values are indicated at the nodes.BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402Table 2: Genomic islands comparison between two genotype groups of E. coli O45 strains.K-Loop K-island O-island S-loop Group Ia Group IIa Notable function of this Island or LoopOI#1 S-loop#1 - + putative fimbrial chaperone and proteinOI#2 S-loop#3 - + CcdA-like protein, CcdB-like proteinOI#7 S-loop#14 +/- - VgrG proteinOI#9 S-loop#17 - + putative transcriptional regulator, transport proteinOI#11 S-loop#19 - + putative transcriptional regulatorOI#14 S-loop#23 - + putative invertase, hypothetical proteinOI#15 S-loop#24 + - Aida-I, adhesin-like proteinOI#19 S-loop#30 - + hypothetical proteinOI#20 S-loop#31 - + putative sensor histidine protein kinaseOI#24 S-loop#37 - + hypothetical proteinOI#25 S-loop#38 - + hypothetical proteinOI#26 S-loop#39 - + hypothetical proteinOI#28 S-loop#42 - + putative outer membrane transport proteinOI#29 S-loop#43 - + adhesin/invasin-like proteinOI#30 S-loop#44 +/- - putative Vgr proteinS-loop#51 - + putative fimbrial-like proteinOI#35 S-loop#52 - + putative transcription regulatorOI#37 S-loop#57 - + hypothetical proteinOI#41 S-loop#61 - + hypothetical proteinOI#47 S-loop#71 - + hemagglutinin/hemolysin-related proteinOI#50 +/- - putative antirepressor of prophage CP-933NOI#51 S-loop#78 +/- - putative single stranded DNA-binding proteinS-loop#85 + - putative regulatory proteinOI#52 S-loop#108 + - putative phage tail proteinOI#52 S-loop#153 + - putative host specificity proteinOI#57 S-loop#85 + - virulence factor PaaOI#65 S-loop#99 + - VgrE proteinOI#70 S-loop#87 - + putative transcription regulatory proteinOI#71 S-loop#108 +/- - BfpT-regulated chaperone-like proteinOI#76 S-loop#119 +/- - putative transcriptional regulatorOI#91 + - unknown functionOI#106 - + putative polyferredoxinOI#115 - + TTSS or ETT2OI#120 + - hypothetical lipoproteinOI#122 +/- - putative adherence factor, putative enterotoxinOI#123 - +/- putative ABC-type iron-siderophore transport systemOI#130 - + leader peptidase HopDOI#132 - + hypothetical proteinOI#134 S-loop#220 - + putative ATP-dependent DNA helicaseOI#136 S-loop#223 - + putative membrane proteinOI#137 S-loop#224 - + hypothetical proteinOI#140 S-loop#231 - + heme utilization/transport protein,OI#141 S-loop#232 - +/- putative fimbrial protein precursorOI#154 S-loop#253 - + putative type 1 fimbrial proteinOI#155 S-loop#256 - + hypothetical proteinOI#156 S-loop#257 + - hypothetical membrane proteinOI#158 S-loop#264 - + hypothetical proteinOI#159 S-loop#265 - + putative glycoproteinOI#161 S-loop#267 - + hypothetical membrane proteinOI#162 + - RhsC protein in rhs elementK-loop#28 KI #18 + - polysaccharide metabolism; YaiPK-loop#67 KI #40 + - hypothetical proteinK-loop#76 KI #44 + - hypothetical proteinK-loop#88 KI #52 + - homolog of virulence factor; b1121K-loop#97 KI #60 +b - orf, hypothetical protein; YchGK-loop#140 KI #84 - + orf, hypothetical protein; b2071K-loop#164 KI #93 - + acetyl-CoA: acetoacetyl-CoA transferase alpha subunitK-loop#178 KI #103 + - xanthosine permease; XapA, XapBPage 6 of 14(page number not for citation purposes)K-loop#203 KI #120 + - putative DEOR-type transcriptional regulator; YgbIBMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402K-loops and 50 were OIs or S-loops. Twenty GIs werepresent in Group I strains but absent in Group II strains,including OI#57 (S-loop#85), which contains the paagene, and the two virulence related GIs, OI#71 (S-loop#108) [34] and OI#122 [35,36]. On the other hand,33 GIs were absent in Group I strains but present in GroupII strains, including OI#1 (S-loop#1), OI#47 (S-loop#71)and OI#154 (S-loop#253), which contain fimbriaerelated genes. KI#60 (K-loop#97), which was absent in allGroup II strains, was present in the four O45 PEPECstrains from Group I but not in REPEC strain E22. Inter-estingly, for seven GIs, more than half of the ORFs in eachisland were present in all Group I strains whereas theseGIs were absent in all Group II strains. Conversely, for twoGIs, more than half of the ORFs in each island werepresent in Group II strains whereas these GIs were absentin Group I strains.In addition to the 63 GIs found to be significantly differ-ent between Group I and Group II strains, 26 other GIswere conserved in both Groups (Table 3). Eight of theseGIs were KIs or K-loops and 18 were OIs or S-loops.Finally, analysis of the E. coli O157:H7 microarray dataindicated that the Shiga toxin encoding genes stx1 and stx2could not be detected in any of the O45 PEPEC strains orin the REPEC strain E22, and that similarly, all strainswere lacking both the tccp (ECs2715/Z3072) and tccp2(ECs1126/Z1385) genes, which encode E. coli O157:H7type III effector proteins that couple the intimin receptorTir to the actin-cytoskeleton, and trigger actin polymeriza-tion [37-40].Analysis of LEE by the E. coli O157:H7 microarrayThirty of the 41 genes on LEE (OI#148/S-loop#244) werefound to be conserved among the O45 PEPEC strains,K-loop#218 KI #132 + - hypothetical proteinK-loop#220 KI #133 + - putative fimbrial-like protein; YgiLK-loop#255 KI #154 + - putative lipase; YiaLK-loop#267 KI #161 + - regulator protein for dgo operon; YidWa +/- indicates that more than 50% of the ORFs in this island or loop are present.b REPEC strain E22 is "-".Table 2: Genomic islands comparison between two genotype groups of E. coli O45 strains. (Continued)Table 3: Conserved genomic islands identified by CGH in O45 PEPEC strains.K-loop K-island O-island S-loop Number of ORFs Notable function of this Island or Loop (if known)OI#17 S-loop #27 5 ribose transport related proteinS-loop #45 1OI #32 1OI #61 S-loop #104 2 fimbrial proteinOI #69 S-loop #89 2OI #71 S-loop #109 1 chaperone proteinOI #94 1OI #103 S-loop #173 2OI #108 S-loop #56 1 chaperone proteinOI #110 S-loop #191 4 decarboxylase proteinS-loop #192 8OI#118 1OI#119 S-loop #202 5 ABC transport proteinOI#120 S-loop #203 1OI#126 S-loop #211 2 sugar PTS proteinOI#130 S-loop #215 1 bacterioferritin proteinOI#145 S-loop #239 7 LPS biosynthesis proteinOI#169 S-loop #278 2K-loop#90 KI #56 12K-loop#108 KI #66 20K-loop#115 KI #71 1K-loop#119 KI #73 1K-loop#126 KI #76 1K-loop#134 KI #80 1K-loop#232 KI #141 8Page 7 of 14(page number not for citation purposes)K-loop#271 KI#163 8BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402REPEC strain E22, and the two O157:H7 strains EDL933and Sakai (Table 4). These included the effector-encodinggenes espA, espF and espG, the regulator ler, and most ofthe genes of the type III secretion pathway such as sepL,escD, cesT, escN, escV, sepD, escC, cesD, escU, escT, escS, andescR. For the 11 remaining genes on LEE, no hybridizationwas observed in the O45 PEPEC strains and REPEC strainE22, possibly reflecting genetic divergences between thesestrains and the O157:H7 representative strains EDL933and Sakai. These genes were the effector-encoding genesespB, espD, and espH, intimin and the translocated intiminreceptor-encoding genes eae and tir, the genes of the typeIII secretion pathway sepQ, sepZ, and escJ, and the genesmap, mpc (for multiple point controller) and Z5117.Localization of LEE and OI#122The LEE of AEEC strains is often inserted in the vicinity ofthe tRNA loci selC or pheU. Since it has been previouslyreported that the site of insertion of LEE in PEPEC strainscould be either in selC or in pheU [23], the O45 PEPECstrains in our study and REPEC strain E22 were examinedby PCR using primers specific for these two genes and forLEE extremities (Additional file 1: Table S1 [21,23,28-31]). The LEE was found to be inserted into the tRNA pheUlocus in all examined strains. Remarkably, an amplicon of500 bp longer than the expected size was also obtainedwith primers specific for LEE extremities and selC forstrains ECL2033 and ECL2020 (Additional file 3: Table S3[22]).Similarly, the localization and integrity of OI#122 wasdetermined by PCR using primers described previously(Additional file 1: Table S1 [21,23,28-31]). All Group Istrains possessed this GI and were positive for the fourgenes tested; efa1, ent, nleB, and nleE, with the latter twoencoding non-LEE virulence factors. OI#122 was found toTable 4: Divergence in the LEE genes among O45 PEPEC strains and REPEC strain E22.Divergent genesa, bGroup I strainsc Group II strainsc FunctionECL1001 ECL2017 ECL2004 ECL2033 E22 ECL2019 ECL2078 ECL2027 ECL2020 ECL2076Z5105/ECs4554- + - - - - - - - - EspB proteinZ5106/ECs4555+ + + + - + - + - + EspD proteinZ5110/ECs4559- - - - - - - - - - Gamma intiminZ5112/ECs4561+ + + - - - - - - + translocated intimin receptor TirZ5113/ECs4562- + + + - - - + - + Map proteinZ5115/ECs4564- - - - - - - - - - EspH proteinZ5116/ECs4565- + + - + + - + - + type III secretion system SepQ proteinZ5117/ECs4566- + + + - + - + - + hypothetical proteinZ5121/ECs4570- - - - - - - - - - Mpc proteinZ5122/ECs4571- - - - - - - - - - type III secretion system SepZ proteinZ5124/ECs4573+ + + + - + - + - + type III secretion system EscJ proteina Z, nomenclature for genes from O157:H7 strain EDL933; ECs, nomenclature for genes from O157:H7 strain Sakai.b the following genes were present in all O45 EPEC strains and REPEC strain E22: Z5100/ECs4550 (EspF protein), Z5102/ECs4551 (hypothetical protein), Z5103/ECs4552 (EscF protein), Z5104/ECs4553 (hypothetical protein), Z5107/ECs4556 (EspA protein), Z5108/ECs4557 (type III secretion system SepL protein), Z5109/ECs4558 (type III secretion system EscD protein), Z5111/ECs4560 (CesT protein), Z5114/ECs4563 (hypothetical protein), Z5118/ECs4567 (hypothetical protein), Z5119/ECs4568 (type III secretion system protein EscN), Z5120/ECs4569 (type III secretion system EscV protein), Z5123/ECs4572 (hypothetical protein), Z5125/ECs4574 (type III secretion system protein SepD), Z5126/ECs4575 (type III secretion system EscC protein), Z5127/ECs4576 (type III secretion system CesD protein), Z5128/ECs4577 (hypothetical protein), Z5129/ECs4578 (hypothetical protein), Z5131/ECs4579 (hypothetical protein), Z5132/ECs4580 (type III secretion system EscU protein), Z5133/ECs4581 (type III secretion system EscT protein), Z5134/ECs4582 (type III secretion system EscS protein), Z5135/ECs4583 (type III secretion system EscR Page 8 of 14(page number not for citation purposes)protein), Z5136/ECs4584 (hypothetical protein), Z5137/ECs4585 (hypothetical protein), Z5138/ECs4586 (hypothetical protein), Z5139/ECs4587 (hypothetical protein), Z5140/ECs4588 (Ler protein), Z5142/ECs4590 (EspG protein), and Z5143/ECs4591 (hypothetical protein).c +, gene present; -, gene absent.BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402be inserted into the tRNA pheU locus in strain ECL2033and into the tRNA pheV locus in strains ECL1001 andECL2004. The site of insertion of this GI was not deter-mined for strain ECL2017 or for REPEC strain E22. AllGroup II strains lacked OI#122 (Additional file 3: TableS3 [22]).nle genes in O45 PEPEC strainsThe E. coli O157:H7 microarray used in our CGH studiescontains oligonucleotide probes specific for genes encod-ing non-LEE factors which have previously been associ-ated with the pathogenicity of AEEC strains [34,41]. ThenleA and nleC genes were present in all O45 PEPEC strainsand REPEC strain E22 as determined by the O157:H7microarray (Table 5). Nevertheless, PCR analysis usingtwo different primer sets for each gene (Additional file 1:Table S1 [21,23,28-31]) revealed amplicons of varioussizes in the different strains, showing that the nleA andnleC genes in Group I were different from those in GroupII (data not shown).Sixteen other nle genes showed non-random distributionsbetween Group I and Group II strains (Table 5). The fivegenes espY3, espX2, espR1, espL3' (Z5199/ECs4642), andespL3' (Z5200/ECs4643) were absent in all Group I strainsbut present in Group II strains. On the other hand, the fivegenes espX7, espK, espL2, nleB1, and nleE were present in allGroup I strains but absent in Group II strains. The fourTable 5: Distribution of nle genes among O45 PEPEC strains and REPEC strain E22.Group I strains Group II strainsGene IDa Family ECL1001 ECL2017 ECL2004 ECL2033 E22 ECL2019 ECL2078 ECL2027 ECL2020 ECL2076espY3 Z0521 (ECs0472)SopD-N; PRR- - - - - + + + + +espX2 Z1019 (ECs0876)PPR - - - - - + + + + +espR1 Z2242 (ECs2073)LRR - - - - - + + + + +espL3' Z5199 (ECs4642)AR - - - - - + + + + +espL3' Z5200 (ECs4643)AR - - - - - + + + + +espX7 Z1822 (ECs1560)PPR; LRR + + + + + - - - - -espK Z1829 (ECs1568)LRR + + + + + - - - - -espL2 Z4326 (ECs3855)AR + + + + + - - - - -nleB1 Z4328 (ECs3857)NleB + + + + + - - - - -nleE Z4329 (ECs3858)NleE + + + + + - - - - -nleB2-1 Z0985 (ECs0846)NleB - - - - + + + + + +nleG2-1' Z6025 (ECs1810)NleG + + + + - - - - - -espO1-2 ECs1821 OspE + + + + - - - - - -nleG Z6010 (ECs1824)NleG + + + + - - - - - -nleG9' Z2560 (ECs1828)NleG + + + + - - - - - -nleD Z0990 (ECs0850)NleD - - + + - - - - - -nleF Z6020 (ECs1815)NleF + + + + - + + + + +nleH Z0989 (ECs0848)NleH + + + + + + + + + +nleAb Z6024 (ECs1812)NleA + + + + + + + + + +nleCb Z0986 (ECs0847)NleC + + + + + + + + + +Page 9 of 14(page number not for citation purposes)a Z, nomenclature for O157:H7 strain EDL933; ECs, nomenclature for O157:H7 strain Sakai.b PCRs using two different primer sets for each gene revealed that both nleA and nleC in Group I were different from those in Group II.BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402genes nleG2-1', espO1-2, nleG, and nleG9' were present inall Group I strains with the exception of REPEC strain E22,and absent in Group II strains. The gene nleB2-1 waspresent in all Group II strains and also in REPEC strainE22 but absent in the other Group I strains. The gene nleDwas present in only two Group I strains, ECL2004 andECL2033, and absent in all the other strains, includingREPEC strain E22.Two additional nle genes, nleF and nleH, were present inall O45 PEPEC strains. nleH, but not nleF, was also presentin REPEC strain E22 (Table 5).Distribution of ETT2 genesOI#115, initially described in E. coli of serotype O157:H7and present in other EPEC and EHEC strains from animalsand humans, contains the secondary type III secretion sys-tem gene cluster ETT2 [31,42,43]. CGH data analysisrevealed that all Group II strains had the entire ETT2 locuscomprising 36 genes, with the exception of strainECL2019, which lacked most of this GI (Additional file 4:Table S4 [16]).In contrast, Group I strains possessed only a partiallyintact locus and the occurrence of the ETT2 genes washighly variable. Among the 36 ETT2 genes, strainECL2033 possessed only 21, strain ECL2004 possessed20, strain ECL1001 possessed 17, and strain ECL2017possessed 15. Finally, REPEC strain E22 possessed 21genes of this GI. These results were confirmed by PCR asdescribed previously [31], with primers specific for differ-ent regions of the ETT2 gene cluster (Additional file 1:Table S1 [21,23,28-31]).Genes required for intestinal colonization in the bovinePrevious studies have identified several genes required forEHEC intestinal colonization of the bovine [44,45].Microarray analysis in our study revealed that 13 genesassociated with colonization of either E. coli O157:H7 orE. coli O26:H- in the bovine were associated with eitherGroup I or Group II strains (Additional file 5: Table S5).Seven genes were present in Group I but not in Group IIstrains, with the exception of REPEC strain E22 which didnot possess the gene Z6010 (ECs1824). In contrast, sixother genes were present in Group II but not in Group Istrains, with the exception of REPEC strain E22 whichpossessed the gene Z1526 (ECs1270). These results wereconfirmed by PCR using primers designed for each gene(data not shown).DiscussionIn this study, we investigated the genetic relationshipsamong PEPEC strains of serogroup O45 and cataloguedpreviously characterized for their capacity to induce A/Elesions in both explants and challenged pigs, and weregrouped according to the severity of the A/E manifestationthey produced [13]. Based on their virulence gene contentas determined by the E. coli virulence microarray, O45PEPEC strains and REPEC strain E22 displayed significantdifferences from typical EPEC and could be regarded asatypical EPEC, that are defined as LEE-positive E. coli lack-ing stx1 and stx2 genes, as well as the EAF virulence plasmidwhich encodes the EPEC adhesin BFP [46,47]. In addi-tion, all O45 PEPEC strains and REPEC strain E22 unex-pectedly hybridized with the stxB1 probe of the E. colivirulence microarray, as was also observed for some atyp-ical EPEC strains isolated from children with diarrhea in arecent study in Norway [35,48]. Due to the absence ofhybridization with the corresponding stxA1 probe and thenegative PCR results obtained with stxB1 sequence specificprimers [35,48], we therefore concluded that the genesequences detected by the stxB1 hybridization probe didnot represent a complete stxB gene but rather a possibletruncated form of this gene.As observed for other atypical EPEC strains, O45 PEPECstrains and REPEC strain E22 also displayed a relativelyhigh heterogeneity in their virulence gene profiles [35,49].Based on their virulence gene content, they could bedivided into two distinct groups, Groups I and II. It hasbeen argued that atypical EPEC strains could have arisenfrom E. coli strains of different pathotypes which acquiredthe LEE by horizontal gene transfer or from certain typicalEPEC strains that have lost the EAF plasmid [49]. Trabulsiet al. have also observed that some atypical EPEC strainsare genetically closer to EHEC strains of serotypeO157:H7 than to typical EPEC [50]. Several virulencegenes showed a non-random distribution between GroupI and Group II strains. Group I strains thus possessed sev-eral virulence-related genes which were absent in Group IIstrains. Group I-specific genes included paa (which con-tributes to A/E lesion formation in PEPEC strains [10])and OI#122 genes efa1 (which plays an important role inintestinal colonization by EHEC strains in cattle [51]) andset (which encodes a putative enterotoxin highly similarto the enterotoxin ShET2 of Shigella flexneri). Genes asso-ciated with other pathotypes were also found. The genetsh, encoding a hemagglutinin which may be a virulencefactor of avian extraintestinal E. coli [52], the pathogenic-ity island marker malX, related to virulence in extraintesti-nal E. coli [53] and the yersiniabactin-related genes fyuA,irp1 and irp2, implicated in the ferric uptake system, werealso Group I-specific. In contrast, Group II strains pos-sessed only a few additional virulence-related genes whencompared with Group I strains. These included aidaI,which encodes a protein involved in the adherence ofPage 10 of 14(page number not for citation purposes)genomic alterations unique to these strains by using botha virulence gene-specific microarray and a whole genomemicroarray. The 045 PEPEC strains in this study have beenEPEC [54], and iron uptake-related genes chuA and fepC.Finally, Group I and Group II strains also possessed differ-ent variants of the long polar fimbriae encoding gene lpfA.BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402A recent study has shown that the lpfAO113 variant, foundin Group I strains in our study, was found significantlymore frequently in atypical EPEC strains associated withcases of diarrhea than in strains isolated from healthyindividuals [35].It is interesting to note that analysis by CGH using a wholegenome E. coli microarray, representing two lineage I,human outbreak-related E. coli O157:H7 strains and onenon-pathogenic E. coli K12 strain, resulted in the distribu-tion of the O45 PEPEC strains into the same two groups(Groups I and II), observed for the E. coli virulence micro-array. This genetic-based grouping, principally reflectingthe virulence gene content of the strains, was also compat-ible with the grouping based on their A/E activity [13].The O45 PEPEC strains of Group I all induced severe A/Elesions whereas those of Group II induced less severe orno A/E lesions in both pig ileal explants and challengedpigs. REPEC strain E22 was placed into Group I but wasgenetically distant from the four O45 PEPEC strainsbelonging to this group. These strains showed a relativelyhigh level of heterogeneity in their virulence gene profiles.In addition to the variations in particular virulence genes,significant variations in GIs were also observed betweenGroup I and II strains. We observed that several virulence-related OIs were present only in Group I strains. Theseincluded OI#57 (S-loop#85), which contains the paagene; OI#71 (S-loop#108), which contains the non-LEEencoded factor gene nleA, previously shown to be associ-ated with the pathogenicity of AEEC strains [34]; andOI#122 which contains the two non-LEE encoded factorgenes nleB and nleE and virulence genes efa1 and set. Inter-estingly, these OIs have also been shown to be more prev-alent in STEC strains associated with outbreaks and severedisease [35,36]. On the other hand, certain OIs were onlypresent in Group II strains. These include OI#1 (S-loop#1), containing genes encoding putative fimbrialchaperone proteins; OI#47 (S-loop#71), containing afimbrial operon and genes encoding several additionalputative virulence factors in E. coli of serogroup O157[55]; and OI#154 (S-loop#253), containing genes encod-ing putative type 1 fimbrial proteins. Finally, OI#115 washighly divergent between Group I and Group II strains.This OI contains the secondary type III secretion systemgene cluster ETT2, which resembles the SPI-1-encodedtype III secretion system from Salmonella enterica and hasbeen previously characterized in O157:H7 E. coli strains[16,17]. It has been recently shown that ETT2 influencesthe secretion of proteins encoded by the LEE and alsomodulates adhesion to human intestinal cells [56]. MostGroup II strains (4/5) possessed the entire ETT2. How-ever, Group II strain ECL2019 lacked most of the entireexplants or in challenged pigs. Group I strains, includingREPEC strain E22, have only a partial ETT2 gene cluster,possessing from 15 to 21 genes of the 36 in the intact clus-ter. The substantial variations observed for this cluster areconsistent with the findings of previous studies, that whilethe ETT2 gene cluster was present in most of the E. colistrains tested, it contained numerous inactivating muta-tions [31,42,43].In contrast to the heterogeneity of their virulence genecontent and GI distribution, O45 PEPEC strains andREPEC strain E22 showed a high level of homogeneity intheir LEE sequences and site of insertion. In all strains, theLEE was inserted into the tRNA pheU gene and no signifi-cant divergence between Group I and Group II strains wasobserved for the LEE genes. In addition, all O45 PEPECstrains and REPEC strain E22 shared the same profile forthe intimin encoding gene eae, its translocated receptor tirand the effector encoding genes espA and espB, as shownby the E. coli virulence microarray. All strains possessedthe beta variant of the intimin encoding gene, being themost widespread among the intestinal EPEC strains of dif-ferent animal species [57,58]. However, Group I andGroup II strains from our study all belonged to the phylo-genetic groups B1 and B2 whereas Ishii et al. have shownthat most EPEC strains possessing the intimin subtypebeta belong to phylogenetic groups A and B1 [59].Finally, we have also observed numerous disparities in thedistribution of non-LEE encoded genes. Several nle geneswere only present in Group I strains whereas others wereonly found in Group II strains. In addition, Group I andGroup II strains possessed two different variants of thegenes nleA and nleC. This variation in the distribution ofnle genes may influence the pathogenicity of the strainsand the type of A/E lesions they produce, since many stud-ies have shown the importance of non-LEE encoded fac-tors in the A/E phenotype [34,41,60,61].ConclusionWe have genetically characterized a collection of O45PEPEC strains using E. coli O157-E. coli K12 wholegenome and virulence gene-specific E. coli microarrays.We have shown that the strains, although showing someheterogeneity, could be classified into two groups, basedon their virulence gene and GI content. These differencesin their virulence gene content may influence the patho-genicity of O45 PEPEC strains, and explain why Group IO45 PEPEC strains induced more severe A/E lesions inexplants and challenged pigs than Group II strains[13,14].Authors' contributionsPage 11 of 14(page number not for citation purposes)cluster (only four genes were found to be present) and wasone of the two strains which did not induce A/E lesions inGB and YZ both contributed equally to the manuscript:they were involved in the conception and design of theBMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402study, in the analysis and interpretation of the microarraydata, and in drafting and revising the manuscript. PG, JW,and CL were involved in the acquisition of the microarraydata and in revising the manuscript. JMF was involved inrevising the manuscript critically for important intellec-tual content. VPJG and JH were involved in the concep-tion and design of the study, and revising the manuscriptcritically for important intellectual content. All authorsread and approved the final manuscript.Additional materialAcknowledgementsClarisse Desautels, from the Reference Laboratory for E. coli, Faculté de médecine vétérinaire, Université de Montréal, is greatly acknowledged for her information and her help regarding the collection of O45 PEPEC strains. This work was supported in part by the Natural Sciences and Engi-neering Research Council of Canada (NSERC) (STPGP 364950), and by the Fond Québécois de la Recherche sur la Nature et les Technologies (FQRNT) (PR-121927). G.B. was supported by a scholarship "Michel Sau-cier" from Fondation Canadienne Louis Pasteur (FCLP).References1. Ott M, Bender L, Blum G, Schmittroth M, Achtman M, Tschape H,Hacker J: Virulence patterns and long-range genetic mappingof extraintestinal Escherichia coli K1, K5, and K100 isolates:use of pulsed-field gel electrophoresis.  Infect Immun 1991,59(8):2664-2672.2. Plainvert C, Bidet P, Peigne C, Barbe V, Medigue C, Denamur E, Bin-gen E, Bonacorsi S: A new O-antigen gene cluster has a key rolein the virulence of the Escherichia coli meningitis cloneO45:K1:H7.  J Bacteriol 2007, 189(23):8528-8536.3. Wullenweber M, Beutin L, Zimmermann S, Jonas C: Influence ofsome bacterial and host factors on colonization and invasive-ness of Escherichia coli K1 in neonatal rats.  Infect Immun 1993,61(5):2138-2144.4. Blanco J, Blanco M, Garabal JI, Gonzalez EA: Enterotoxins, coloni-zation factors and serotypes of enterotoxigenic Escherichiacoli from humans and animals.  Microbiologia 1991, 7(2):57-73.5. Nagy B, Fekete PZ: Enterotoxigenic Escherichia coli (ETEC) infarm animals.  Vet Res 1999, 30(2–3):259-284.6. Toth I, Karcagi V, Nagy B, Gado I, Milch H: Examination of vero-cytotoxin producing capacity and determination of the pres-ence of Shiga-like toxin genes in human Escherichia coliisolates.  Acta Microbiol Immunol Hung 1994, 41(3):259-264.7. Frydendahl K: Prevalence of serogroups and virulence genes inEscherichia coli associated with postweaning diarrhoea andedema disease in pigs and a comparison of diagnosticapproaches.  Vet Microbiol 2002, 85(2):169-182.8. Harel J, Lapointe H, Fallara A, Lortie LA, Bigras-Poulin M, Lariviere S,Fairbrother JM: Detection of genes for fimbrial antigens andenterotoxins associated with Escherichia coli serogroups iso-lated from pigs with diarrhea.  J Clin Microbiol 1991,29(4):745-752.9. Helie P, Morin M, Jacques M, Fairbrother JM: Experimental infec-tion of newborn pigs with an attaching and effacingEscherichia coli O45:K"E65" strain.  Infect Immun 1991,59(3):814-821.10. Batisson I, Guimond MP, Girard F, An H, Zhu C, Oswald E, Fair-brother JM, Jacques M, Harel J: Characterization of the novel fac-tor paa involved in the early steps of the adhesionmechanism of attaching and effacing Escherichia coli.  InfectImmun 2003, 71(8):4516-4525.11. Girard F, Batisson I, Frankel GM, Harel J, Fairbrother JM: Interactionof enteropathogenic and Shiga toxin-producing Escherichiacoli and porcine intestinal mucosa: role of intimin and Tir inadherence.  Infect Immun 2005, 73(9):6005-6016.12. Girard F, Oswald IP, Taranu I, Helie P, Appleyard GD, Harel J, Fair-brother JM: Host immune status influences the developmentof attaching and effacing lesions in weaned pigs.  Infect Immun2005, 73(9):5514-5523.13. Zhu C, Harel J, Jacques M, Desautels C, Donnenberg MS, Beaudry M,Fairbrother JM: Virulence properties and attaching-effacingactivity of Escherichia coli O45 from swine postweaningdiarrhea.  Infect Immun 1994, 62(10):4153-4159.14. Zhu C, Harel J, Jacques M, Fairbrother JM: Interaction with pigileal explants of Escherichia coli O45 isolates from swine withpostweaning diarrhea.  Can J Vet Res 1995, 59(2):118-123.15. Blattner FR, Plunkett G 3rd, Bloch CA, Perna NT, Burland V, Riley M,Collado-Vides J, Glasner JD, Rode CK, Mayhew GF, Gregor J, DavisNW, Kirkpatrick HA, Goeden MA, Rose DJ, Mau B, Shao Y: Thecomplete genome sequence of Escherichia coli K-12.  Science1997, 277(5331):1453-1474.16. Hayashi T, Makino K, Ohnishi M, Kurokawa K, Ishii K, Yokoyama K,Han CG, Ohtsubo E, Nakayama K, Murata T, Tanaka M, Tobe T, IidaT, Takami H, Honda T, Sasakawa C, Ogasawara N, Yasunaga T,Kuhara S, Shiba T, Hattori M, Shinagawa H: Complete genomesequence of enterohemorrhagic Escherichia coli O157:H7and genomic comparison with a laboratory strain K-12.  DNARes 2001, 8(1):11-22.17. Perna NT, Plunkett G 3rd, Burland V, Mau B, Glasner JD, Rose DJ,Mayhew GF, Evans PS, Gregor J, Kirkpatrick HA, Posfai G, Hackett J,Klink S, Boutin A, Shao Y, Miller L, Grotbeck EJ, Davis NW, Lim A,Dimalanta ET, Potamousis KD, Apodaca J, Anantharaman TS, Lin J,Yen G, Schwartz DC, Welch RA, Blattner FR: Genome sequenceAdditional file 1Table S1. Primers and E. coli control strains used for PCR experiments.Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2164-10-402-S1.pdf]Additional file 2Table S2. Presence of virulence genes in O45 PEPEC strains and REPEC strain E22 as determined by E. coli virulence microarray.Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2164-10-402-S2.pdf]Additional file 3Table S3. Localization of the LEE and OI#122 in O45 PEPEC strains and REPEC strain E22.Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2164-10-402-S3.pdf]Additional file 4Table S4. Distribution of ETT2 genes in O45 PEPEC strains and REPEC strain E22.Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2164-10-402-S4.pdf]Additional file 5Table S5. Divergence of genes related to intestinal colonization in bovine and calves as determined by CGH.Click here for file[http://www.biomedcentral.com/content/supplementary/1471-2164-10-402-S5.pdf]Page 12 of 14(page number not for citation purposes)BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/402of enterohaemorrhagic Escherichia coli O157:H7.  Nature 2001,409(6819):529-533.18. Kaper JB, Mellies JL, Nataro JP: Pathogenicity islands and othermobile genetic elements of diarrheagenic E. coli.  WashingtonD. C.: ASM Press; 1999. 19. Zhang Y, Laing C, Steele M, Ziebell K, Johnson R, Benson AK,Taboada E, Gannon VP: Genome evolution in major Escherichiacoli O157:H7 lineages.  BMC Genomics 2007, 8:121.20. Elliott SJ, Wainwright LA, McDaniel TK, Jarvis KG, Deng YK, Lai LC,McNamara BP, Donnenberg MS, Kaper JB: The completesequence of the locus of enterocyte effacement (LEE) fromenteropathogenic Escherichia coli E2348/69.  Mol Microbiol 1998,28(1):1-4.21. Sperandio V, Kaper JB, Bortolini MR, Neves BC, Keller R, Trabulsi LR:Characterization of the locus of enterocyte effacement(LEE) in different enteropathogenic Escherichia coli (EPEC)and Shiga-toxin producing Escherichia coli (STEC) serotypes.FEMS Microbiol Lett 1998, 164(1):133-139.22. Bruant G, Maynard C, Bekal S, Gaucher I, Masson L, Brousseau R,Harel J: Development and validation of an oligonucleotidemicroarray for detection of multiple virulence and antimi-crobial resistance genes in Escherichia coli.  Appl Environ Micro-biol 2006, 72(5):3780-3784.23. An H, Fairbrother JM, Desautels C, Mabrouk T, Dugourd D, DezfulianH, Harel J: Presence of the LEE (locus of enterocyte efface-ment) in pig attaching and effacing Escherichia coli and char-acterization of eae, espA, espB and espD genes of PEPEC (pigEPEC) strain 1390.  Microb Pathog 2000, 28(5):291-300.24. Beutin L, Kaulfuss S, Herold S, Oswald E, Schmidt H: Genetic anal-ysis of enteropathogenic and enterohemorrhagic Escherichiacoli serogroup O103 strains by molecular typing of virulenceand housekeeping genes and pulsed-field gel electrophoresis.J Clin Microbiol 2005, 43(4):1552-1563.25. Camguilhem R, Milon A: Biotypes and O serogroups ofEscherichia coli involved in intestinal infections of weanedrabbits: clues to diagnosis of pathogenic strains.  J Clin Microbiol1989, 27(4):743-747.26. Kim CC, Joyce EA, Chan K, Falkow S: Improved analytical meth-ods for microarray-based genome-composition analysis.Genome Biol 2002, 3(11):RESEARCH0065.27. Saeed AI, Sharov V, White J, Li J, Liang W, Bhagabati N, Braisted J,Klapa M, Currier T, Thiagarajan M, Sturn A, Snuffin M, Rezantsev A,Popov D, Ryltsov A, Kostukovich E, Borisovsky I, Liu Z, Vinsavich A,Trush V, Quackenbush J: TM4: a free, open-source system formicroarray data management and analysis.  Biotechniques 2003,34(2):374-378.28. McDaniel TK, Jarvis KG, Donnenberg MS, Kaper JB: A genetic locusof enterocyte effacement conserved among diverse entero-bacterial pathogens.  Proc Natl Acad Sci USA 1995,92(5):1664-1668.29. Bielaszewska M, Sonntag AK, Schmidt MA, Karch H: Presence ofvirulence and fitness gene modules of enterohemorrhagicEscherichia coli in atypical enteropathogenic Escherichia coliO26.  Microbes Infect 2007, 9(7):891-897.30. Bertin Y, Boukhors K, Livrelli V, Martin C: Localization of theinsertion site and pathotype determination of the locus ofenterocyte effacement of shiga toxin-producing Escherichiacoli strains.  Appl Environ Microbiol 2004, 70(1):61-68.31. Hartleib S, Prager R, Hedenstrom I, Lofdahl S, Tschape H: Preva-lence of the new, SPI1-like, pathogenicity island ETT2among Escherichia coli.  Int J Med Microbiol 2003, 292(7–8):487-493.32. Clermont O, Bonacorsi S, Bingen E: Rapid and simple determina-tion of the Escherichia coli phylogenetic group.  Appl EnvironMicrobiol 2000, 66(10):4555-4558.33. Girardeau JP, Dalmasso A, Bertin Y, Ducrot C, Bord S, Livrelli V, Ver-nozy-Rozand C, Martin C: Association of virulence genotypewith phylogenetic background in comparison to differentseropathotypes of Shiga toxin-producing Escherichia coli iso-lates.  J Clin Microbiol 2005, 43(12):6098-6107.34. Gruenheid S, Sekirov I, Thomas NA, Deng W, O'Donnell P, GoodeD, Li Y, Frey EA, Brown NF, Metalnikov P, Pawson T, Ashman K, Fin-lay BB: Identification and characterization of NleA, a non-LEE-encoded type III translocated virulence factor of enter-35. Afset JE, Bruant G, Brousseau R, Harel J, Anderssen E, Bevanger L,Bergh K: Identification of virulence genes linked with diarrheadue to atypical enteropathogenic Escherichia coli by DNAmicroarray analysis and PCR.  J Clin Microbiol 2006,44(10):3703-3711.36. Karmali MA, Mascarenhas M, Shen S, Ziebell K, Johnson S, Reid-SmithR, Isaac-Renton J, Clark C, Rahn K, Kaper JB: Association ofgenomic O island 122 of Escherichia coli EDL 933 with vero-cytotoxin-producing Escherichia coli seropathotypes that arelinked to epidemic and/or serious disease.  J Clin Microbiol 2003,41(11):4930-4940.37. Frankel G, Phillips AD: Attaching effacing Escherichia coli andparadigms of Tir-triggered actin polymerization: getting offthe pedestal.  Cell Microbiol 2008, 10(3):549-556.38. Ogura Y, Ooka T, Whale A, Garmendia J, Beutin L, Tennant S, KrauseG, Morabito S, Chinen I, Tobe T, Abe H, Tozzoli R, Caprioli A, RivasM, Robins-Browne R, Hayashi T, Frankel G: TccP2 of O157:H7 andnon-O157 enterohemorrhagic Escherichia coli (EHEC): chal-lenging the dogma of EHEC-induced actin polymerization.Infect Immun 2007, 75(2):604-612.39. Whale AD, Garmendia J, Gomes TA, Frankel G: A novel categoryof enteropathogenic Escherichia coli simultaneously utilizesthe Nck and TccP pathways to induce actin remodelling.  CellMicrobiol 2006, 8(6):999-1008.40. Whale AD, Hernandes RT, Ooka T, Beutin L, Schuller S, GarmendiaJ, Crowther L, Vieira MA, Ogura Y, Krause G, Phillips AD, Gomes TA,Hayashi T, Frankel G: TccP2-mediated subversion of actindynamics by EPEC 2 – a distinct evolutionary lineage ofenteropathogenic Escherichia coli.  Microbiology 2007, 153(Pt6):1743-1755.41. Deng W, Puente JL, Gruenheid S, Li Y, Vallance BA, Vazquez A, BarbaJ, Ibarra JA, O'Donnell P, Metalnikov P, Ashman K, Lee S, Goode D,Pawson T, Finlay BB: Dissecting virulence: systematic and func-tional analyses of a pathogenicity island.  Proc Natl Acad Sci USA2004, 101(10):3597-3602.42. Makino S, Tobe T, Asakura H, Watarai M, Ikeda T, Takeshi K, Sasa-kawa C: Distribution of the secondary type III secretion sys-tem locus found in enterohemorrhagic Escherichia coliO157:H7 isolates among Shiga toxin-producing E. colistrains.  J Clin Microbiol 2003, 41(6):2341-2347.43. Ren CP, Chaudhuri RR, Fivian A, Bailey CM, Antonio M, Barnes WM,Pallen MJ: The ETT2 gene cluster, encoding a second type IIIsecretion system from Escherichia coli, is present in themajority of strains but has undergone widespread muta-tional attrition.  J Bacteriol 2004, 186(11):3547-3560.44. Dziva F, van Diemen PM, Stevens MP, Smith AJ, Wallis TS: Identifi-cation of Escherichia coli O157:H7 genes influencing coloniza-tion of the bovine gastrointestinal tract using signature-tagged mutagenesis.  Microbiology 2004, 150(Pt 11):3631-3645.45. van Diemen PM, Dziva F, Stevens MP, Wallis TS: Identification ofenterohemorrhagic Escherichia coli O26:H- genes requiredfor intestinal colonization in calves.  Infect Immun 2005,73(3):1735-1743.46. Kaper JB: Defining EPEC.  Rev Microbiol Sao Paulo 1996, 27:130-133.47. Malik A, Toth I, Beutin L, Schmidt H, Taminiau B, Dow MA, MorabitoS, Oswald E, Mainil J, Nagy B: Serotypes and intimin types ofintestinal and faecal strains of eae + Escherichia coli fromweaned pigs.  Vet Microbiol 2006, 114(1–2):82-93.48. Afset JE, Anderssen E, Bruant G, Harel J, Wieler L, Bergh K: Phylo-genetic backgrounds and virulence profiles of atypical enter-opathogenic Escherichia coli strains from a case-control studyusing multilocus sequence typing and DNA microarray anal-ysis.  J Clin Microbiol 2008, 46(7):2280-2290.49. Vieira MA, Andrade JR, Trabulsi LR, Rosa AC, Dias AM, Ramos SR,Frankel G, Gomes TA: Phenotypic and genotypic characteris-tics of Escherichia coli strains of non-enteropathogenic E. coli(EPEC) serogroups that carry EAE and lack the EPEC adher-ence factor and Shiga toxin DNA probe sequences.  J Infect Dis2001, 183(5):762-772.50. Trabulsi LR, Keller R, Tardelli Gomes TA: Typical and atypicalenteropathogenic Escherichia coli.  Emerg Infect Dis 2002,8(5):508-513.51. Nicholls L, Grant TH, Robins-Browne RM: Identification of a novelgenetic locus that is required for in vitro adhesion of a clini-Page 13 of 14(page number not for citation purposes)ohaemorrhagic Escherichia coli O157:H7.  Mol Microbiol 2004,51(5):1233-1249.cal isolate of enterohaemorrhagic Escherichia coli to epithe-lial cells.  Mol Microbiol 2000, 35(2):275-288.Publish with BioMed Central   and  every scientist can read your work free of charge"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central BMC Genomics 2009, 10:402 http://www.biomedcentral.com/1471-2164/10/40252. Dozois CM, Dho-Moulin M, Bree A, Fairbrother JM, Desautels C,Curtiss R 3rd: Relationship between the Tsh autotransporterand pathogenicity of avian Escherichia coli and localizationand analysis of the Tsh genetic region.  Infect Immun 2000,68(7):4145-4154.53. Johnson JR, Stell AL: Extended virulence genotypes ofEscherichia coli strains from patients with urosepsis in rela-tion to phylogeny and host compromise.  J Infect Dis 2000,181(1):261-272.54. Benz I, Schmidt MA: Cloning and expression of an adhesin(AIDA-I) involved in diffuse adherence of enteropathogenicEscherichia coli.  Infect Immun 1989, 57(5):1506-1511.55. Shen S, Mascarenhas M, Morgan R, Rahn K, Karmali MA: Identifica-tion of four fimbria-encoding genomic islands that are highlyspecific for verocytotoxin-producing Escherichia coli sero-type O157 strains.  J Clin Microbiol 2005, 43(8):3840-3850.56. Zhang L, Chaudhuri RR, Constantinidou C, Hobman JL, Patel MD,Jones AC, Sarti D, Roe AJ, Vlisidou I, Shaw RK, Falciani F, Stevens MP,Gally DL, Knutton S, Frankel G, Penn CW, Pallen MJ: Regulatorsencoded in the Escherichia coli type III secretion system 2gene cluster influence expression of genes within the locusfor enterocyte effacement in enterohemorrhagic E. coliO157:H7.  Infect Immun 2004, 72(12):7282-7293.57. Oswald E, Schmidt H, Morabito S, Karch H, Marches O, Caprioli A:Typing of intimin genes in human and animal enterohemor-rhagic and enteropathogenic Escherichia coli: characteriza-tion of a new intimin variant.  Infect Immun 2000, 68(1):64-71.58. Krause G, Zimmermann S, Beutin L: Investigation of domesticanimals and pets as a reservoir for intimin- (eae) gene posi-tive Escherichia coli types.  Vet Microbiol 2005, 106(1–2):87-95.59. Ishii S, Meyer KP, Sadowsky MJ: Relationship between phyloge-netic groups, genotypic clusters, and virulence gene profilesof Escherichia coli strains from diverse human and animalsources.  Appl Environ Microbiol 2007, 73(18):5703-5710.60. Kelly M, Hart E, Mundy R, Marches O, Wiles S, Badea L, Luck S,Tauschek M, Frankel G, Robins-Browne RM, Hartland EL: Essentialrole of the type III secretion system effector NleB in coloni-zation of mice by Citrobacter rodentium.  Infect Immun 2006,74(4):2328-2337.61. Roe AJ, Tysall L, Dransfield T, Wang D, Fraser-Pitt D, Mahajan A,Constandinou C, Inglis N, Downing A, Talbot R, Smith DG, Gally DL:Analysis of the expression, regulation and export of NleA-Ein Escherichia coli O157:H7.  Microbiology 2007, 153(Pt5):1350-1360.yours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 14 of 14(page number not for citation purposes)

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0223059/manifest

Comment

Related Items