191 Parole chiave Colibacillosi, Coniglio, DNA Microarray, EPEC atipico, Escherichia coli, REPEC. Riassunto Nel presente lavoro è stato determinato il profilo dei geni di virulenza di 26 ceppi enteropatogeni di Escherichia coli con la tecnica del DNA microarray; i campioni sono stati prelevati da 17 focolai di colibacillosi rilevati in conigli provenienti da due regioni del Nord Italia. I ceppi sono stati classificati secondo il loro biotipo, il sierogruppo e il gruppo filogenetico. È stata anche precisata la distribuzione dei geni di virulenza che codificano il Locus di cancellazione degli enterociti (LEE), il quello del sistema di secrezione tipo 3 (T3SS), le proteine non-LEE traslocate tramite il T3SS e i fattori di adesione. I ceppi testati, ad eccezione di 1, appartengono ai gruppi filogenetici A e B1. È stata osservata, inoltre, un’associazione predominante tra il sierogruppo O103 e il fenotipo ramnosio-negativo (biotipi 12 o 14). Il profilo del LEE maggiormente riscontrato è stato ler/cesT/espA-1/espB-3/ tir-1/eae(beta)/espD-2/escN/eprJ. Tutti i ceppi possedevano il fattore di adesione rabbit-2 (afr/2) o il ral (Rabbit Adherence Locus) e 24 di essi un ulteriore set costituito da uno o più tra i fattori di colonizzazione efa1/lifA, lpfA e paa. Infine, la presenza singola o combinata di proteine effettrici LEE e/o non-LEE, che codificano i geni espG, cif, map e nle, ha attestato la potenzialità genetica che hanno i ceppi studiati ad indurre lesioni patologiche all'ospite. Le tecnologie basate sul microarray, applicate alla valutazione del profilo genetico dell’Escherichia coli del coniglio, si sono dimostrate convenienti e affidabili se finalizzate ad indagini a larga scala all’interno di programmi di sorveglianza per monitorare la circolazione Profilo dei geni di virulenza di ceppi enteropatogeni di Escherichia coli isolati in conigli del Nord Italia Keywords Atypical EPEC, Colibacillosis, DNA Microarray, Escherichia coli, Rabbit, REPEC. Summary The virulence gene profile of 26 rabbit enteropathogenic Escherichia coli strains, isolated from 17 colibacillosis outbreaks located in two regions of Northern Italy, was determined using an Echerichia coli virulence DNA microarray. All strains were classified according to their determined biotype, sero- and phylo-group. The distribution of virulence genes encoding for the Locus of enterocyte effacement (LEE), LEE type III secretion system (T3SS), non-LEE T3SS translocated proteins and adherence factors was also determined. All strains but one belonged to phylogroups A and B1. A prevalent association between the O103 serogroup with the rhamnose-negative phenotype (biotype 12 or 14) was found. The most prevalent LEE profile found in tested strains was ler/cesT/espA-1/espB-3/tir-1/eae(beta)/espD- 2/escN/eprJ. All strains possessed either the adhesive factor rabbit-2 (afr/2) or the plasmid Rabbit adherence locus (ral) gene and 24 of them an additional individual or combined set of colonization factors efa1/lifA, lpfA and paa genes. Finally, the combined or single presence of a set of LEE and/or non-LEE effector proteins encoding genes, namely espG, cif, map and nle family genes, attested to the genetic potential of investigated strains to induce pathologic lesions to the host. The application of microarray-based technologies in assessing the genetic profile of rabbit E. coli is a reliable, cost-effective candidate for large scale investigations in monitoring programs aimed to survey the circulation of pathogenic strains within rabbit production units, their zoonotic genetic potential and to select E. coli strains eligible for vaccinal prophylaxis in fattening rabbit production. Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.859.4260.2 Accepted: 23.04.2016 | Available on line: 30.09.2018 1 Istituto Zooprofilattico Sperimentale dell'Abruzzo e del Molise ‘G. Caporale’, Campo Boario, 64100 Teramo, Italy. 2 Istituto Zooprofilattico Sperimentale delle Venezie, Vicolo Mazzini 4 int 5/6, 31020 Fontane di Villorba, Treviso, Italy. 3 National Research Council of Canada, 6100 Royalmount, Montreal, QC, Canada H4P 2R2. * Corresponding author at: Istituto Zooprofilattico Sperimentale dell'Abruzzo e del Molise ‘G. Caporale’, Campo Boario, 64100 Teramo, Italy. Tel.: +39 0861 332407, Fax: +39 0861 332251, e-mail: p.badagliacca@izs.it. Pietro Badagliacca1*, Fabrizio Agnoletti2, Ilenia Drigo2, Valentina Merildi1, Federica Lopes1, Cinzia Pompilii1, Iolanda Mangone1, Massimo Scacchia1, Alfreda Tonelli1 and Luke Masson3 Virulence gene profiles of rabbit enteropathogenic Escherichia coli strains isolated in Northern Italy 192 Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx (adhesive factor rabbit-2), a chromosomal encoded fimbrial adhesin inducing the phenotype of ‘diffuse adhesion’ (DA) on surface of HeLa cells, is associated with virulent rabbit strains (Fiederling et  al. 1997). In addition, a putative chromosomal pathogenicity islet gene Porcine A/E-associated (Paa), which strongly correlates with the A/E phenotype, a chromosomal Long polar fimbriae gene (Lpf ), thought to contribute to epithelial cell colonization, and a plasmid-encoded fimbriae Repec adherence locus gene (Ral), involved in the early steps of adhesion, were also found in diarrheal rabbit E.  coli isolates (Batisson et  al. 2003, Badea et  al. 2003, Newton et al. 2004, Krejany et al. 2000). The eae and afr/2 genes are the main genetic markers used to define REPEC strains (Milon et  al. 1999). Another way to determine E.  coli enteropathogenicity is to assess both the serogroup and sugar fermentation character (biotype), targeted to establishing a link between biotype/serotype and high mortalities (Camguilhem and Milon 1989). An oligonucleotide virulence microarray, initially developed and validated as previously described (Bekal et  al. 2003, Bruant et  al. 2006), allowed the detection of an exhaustive list of E.  coli virulence genes as well as antimicrobial resistance genes in E.  coli strains isolated from coastal water and wastewater (Hamelin et al. 2006, Frigon et al. 2013), camel (Salehi et al. 2012), poultry (Bonnet et al. 2009), and cattle (Staji et al. 2017). In a previous study this microarray was successful in clustering a group of aEPEC strains from rabbit experiencing diarrhea from a second cluster that grouped non-pathogenic E. coli strains from healthy rabbit, both coming from the same rabbitry (Tonelli et al. 2008). The aim of the present work was to develop a methodology for defining virulent genotypes of REPEC strains by genotyping a collection of rabbit enteropathogenic E.  coli strains isolated from rabbitries affected by colibacillosis in Northern Italy. Materials and methods Strain collection, microbiology Twenty six E.  coli extracted DNA, tested for eae polymerase chain reaction (PCR) signal carried out as previously described (Agnoletti et  al. 2004), Introduction Based on their virulence gene content and overt clinical symptoms, Escherichia coli strains are classified into different pathotypes responsible for intestinal or extra-intestinal syndromes. In particular, the attachement and effacing E.  coli (AEEC) pathotype, which includes both enteropathogenic (EPEC) and enterohemorrhagic (EHEC) strains, are characterized by the presence of a chromosomal pathogenicity island called the Locus of enterocyte effacement (LEE), encoding for proteins responsible for the attaching and effacing (A/E) phenotype. AEEC strains that are classified as EHEC also possess the Shiga toxin encoding-genes stx1 and/or stx2. LEE genes are organized in five operons, named LEE1 to 5. LEE1, LEE2 and LEE3 transcribe the Type III Secretion System (T3SS) E. coli secretion (Esc) proteins and LEE4 transcribes the E.  coli secreted proteins (Esp). LEE5 operon contains eae, the gene encoding the bacterial adhesin Intimin which is involved in the intimate attachment to host epithelial cells, its receptor tir (translocated intimin receptor) and the chaperone cesT, translocated from the bacteria to the host epithelial cells through the T3SS structure. The T3SS is also used to inject LEE and non-LEE encoded effector proteins (EspCGP, Map; Cif, Nle) into the host cell, which are responsible for the effacement of host epithelial cells (Zhu et  al. 2001, Marchés et al. 2003, Schmidt and Hensel 2004, Bertin et al. 2004, Thomas et al. 2005, Garrido et al. 2006, Luo and Donnenberg 2011). EPEC strains also possess the E.  coli adherence factor (EAF) virulence plasmid, that encodes the adhesin Bfp (bundle forming pili), which induces localized adhesion (LA) of bacterial microcolonies on epithelial cells. The LA phenotype is associated with serotypes/serogroups from human diarrhea outbreaks (Giron et al. 1991). EPEC strains, which lack the EAF plasmid, are called atypical EPEC (aEPEC) (Nuyen et al. 2006). After the initial study by Cantey and Blake (Cantey and Blake 1977) on the rabbit diarrheal E. coli-1 strain (RDEC-1), strains involved in rabbit colibacillosis outbreaks in several European countries were confirmed as non-toxin producing aEPECs inducing A/E lesions (Milon et  al. 1999). The pathogenic specificity of rabbit EPEC (REPEC) strains lies primarily with the adherence mechanism that complements attachment to host epithelial cells, mediated by Intimin. Afr-2 Virulence of rabbit enteropathogenic E. coli Badagliacca et al. di ceppi patogeni nella filiera di produzione del coniglio, valutare il loro potenziale genetico zoonotico, e selezionare ceppi patogeni di E. coli per creare una profilassi vaccinale mirata nelle unità da ingrasso degli allevamenti cunicoli. Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx 193 E.  coli isolates were clustered based on their biotype/serogroup/phylogroup profile and analysed according to the virulence gene or marker patterns related to LEE structural genes (ler, eae, tir1-2-3, cesT, espABD, escJN, eprJ), LEE and non-LEE effector (espCGP, map/orf19; nleABCDEFGH, cif) and adherence/colonization factors (afr/1-2, ralG, bfpAB, paa, efa1/lifA, lpfA). Finally, LEE gene variant combinations were classified according to Afset et  al. (2008). The chi-square test was used to assess the associations between serogroup, biotype, and phylogroup. Statistical significance was expressed with p-value (α <0.05). E. coli DNA labeling and hybridization Approximately 300 ng of purified genomic DNA was labelled with fluorescent Cyanine (Cy3) dye using the Bioprime DNA labelling system (Invitrogen Life Technologies, Burlington, ON, Canada) as described previously (Bruant et  al., 2006). Labelling efficiency and the percentage of dye incorporation were then determined by scanning the DNA sample in a NanoDrop spectrophotometer from 200 to 700  nm. Cy3 dye incorporation was calculated using a Web-based percent incorporation calculator (http:// www.pangloss.com/seidel/Protocols/percent_inc. html). Pre-hybridizations and hybridizations were performed following a protocol derived from Hamelin and colleagues (Hamelin et  al. 2006). For the hybridizations, 500 ng of labelled DNA was dried under vacuum in a rotary desiccator without heating (Savant SpeedVac®, ArrayIt, USA). Dried labelled DNA was resuspended in hybridization buffer (DIG Easy Hyb Buffer, Roche Diagnostics, Laval, Quebec, Canada). Microarrays were pre-hybridized for one hour at 50  °C with a pre-heated pre-hybridization buffer containing 5X SSC, 0.1% SDS and 1.0% BSA. After pre-hybridization, the microarrays were hybridized with a solution that consisted of 25  μl of hybridization buffer, 20  μl of Bakers Yeast tRNA (10  mg/ml) (Sigma Aldrich, St. Louis, MO, USA) and 20  μl of Sonicated Salmon Sperm DNA (10  mg/ml) (Sigma Aldrich, St. Louis, MO, USA), mixed together with the labelled DNA which had previously been denatured. Microarrays were hybridized overnight at 50  °C in a SlideBooster (model SB800; Advalytix, Germany). After hybridization, stringency washes were performed with Advawash (Advalytix, Germany) using 1X SSC, 0.02% SDS preheated to 50 °C. Microarray data acquisition Microarray slides were scanned with a ScanArray Lite fluorescent microarray analysis system (Perkin-Elmer, Missasauga, Ontario, Canada) using with ScanArray Gx software (Perkin-Elmer, Foster City, CA). Fluorescent spot intensities were quantified were selected from a more extensive collection coming from strains isolated from rabbits affected by diarrheal syndrome. Rabbits were 7- to 87-day old coming from 17 rabbitries located in the Veneto and Piemonte regions (Northern Italy) with variable pathological degree of lesion ranging from liquid content in the caecum to typhlitis, enterotyphlitis and severe enterotyphlitis. An inoculum from the caecal contents of diseased animals was directly plated onto Eosin methylene blue agar (EMB, Oxoid Ltd, England) and E. coli isolates identified using the API 20E® system (BioMérieux, France) and biotyped according to the method described by Camguilhem and Milon (Camguilhem and Milon 1989). One colony was re-isolated on tryptic soy agar (Biolife Italiana, Milan, Italy), suspended in phosphate buffer saline and its DNA extracted with a GenElute Bacterial Genomic DNA kit® (Sigma-Aldrich, USA) and stored at -20 °C until use. Microarray design and E. coli virulence genes or markers The microarray version used in the present study was composed of 70-mer oligonucleotide probes printed in duplicate on Corning Ultra GAPS slides (Corning Canada, Whitby, Ontario), targeting for 348 E. coli virulence factors, covering all known E. coli pathotypes, and 95 antimicrobial resistance genes. The tryptophanase (tnaA), beta-glucuronidase (uidA), lactose permease (lacY) and beta-galactosidase (lacZ) genes were included as positive controls whilst negative controls included empty buffer spots as well as genes for the green fluorescent protein of Aequoria victoria (gfp) and the chlorophyll synthase gene of Arabidopsis thaliana (At3g). Only data related to virulence or virulence-related gene hybridizations were considered in this study, since the antimicrobial resistance profiles of the same strains were analyzed separately (Badagliacca et  al. 2014). Confirmatory tests independent of the microarray, were performed by specific molecular probes or gene sequencing, as described below. Escherichia coli isolates were assigned to a serogroup based on their wzy gene for O7, O15, O22, O24, O26, O28, O45, O53, O55, O56, O59, O66, O86, O91, O98, O103, O104, O113, O114, O117, O121, O123, O126, O127, O128, O138, O139, O141, O145, O146, O147, O148, O149, O155, O157, O172, O174, and O177. Assignment to other serogroups was based on rfc gene for O4, wzx for O6, wb for O8, rfb for O9 and O101 and wbdl for O111. E.  coli phylogenetic groups were identified on the microarray based on the combined presence/ absence of the chuA gene, the yjaA gene and the DNA fragment TSPE4.C2, according to Clermont and colleagues (Clermont et al. 2000). Badagliacca et al. Virulence of rabbit enteropathogenic E. coli Nick title First author et al. 194 Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx (Eurofins MWG Operon, Ebersberg, Germany). Confirmatory test of tir gene as well additional PCRs for variants 1, 2 and 3 of espD, not hybridized by microarray, were performed as described by Garrido and colleagues (Garrido et al. 2006). Additional PCR for the cesT gene was performed using primers from Bertin and colleagues (Bertin et  al. 2004). All PCRs were carried out using universal master mix (PCR Master Mix 2xTM, Promega Italia srl, Milan, Italy), according the provider’s instructions. Strains showing an ambiguous signal for REPEC fimbrial factors Afr or Ral were submitted to genomic sequencing. DNA samples were quantified by using the Qubit® DNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA). One ng of DNA was used for library preparation by using the Nextera XT Library Prep kit (Illumina Inc., San Diego, CA) according to the manufacturer’s protocol. Deep Sequencing was performed on the NextSeq 500 (Illumina Inc., San Diego, CA) using the NextSeq 500/550 Mid Output Reagent Cartridge v2, 300  cycles and standard 151  bp pairs-end reads. with QuantArray Version 3.0 (Packard Bioscience, Informer Technologies Inc., USA). The mean value for each set of duplicate spotted oligonucleotides was divided by the correction factor taken from the negative control spots. This value was then divided by the average of the empty spots to create a signal-to-noise ratio. Oligonucleotide spots with a signal-to-noise fluorescence ratio greater than the established threshold (2 in this study) were considered positive. These ratios were then converted into binary data where a value of 0 indicates a negative probe and a value of 1 indicates a positive probe. Confirmatory and additional tests Strains showing LEE structural genes demonstrating a signal lower than the threshold were confirmed by PCR. A confirmatory PCR test for espB was designed with Primer-BLAST™, after a gene sequence alignment of 2,267 blast hits present in GenBank. Two primers, forward 5’-GCCGTTTTTGAGAGCCA-3’ and reverse 5-TTACCCAGCTAAGCGA-3’, were synthesized Table I. Strain identification, pathological/epidemiological data, biotype and genetic grouping of enteropathogenic Escherichia coli strains collected from rabbitries of Northern Italy. Strain* Age (day) Unit Pathological lesion Biotype O-type# Phylogroup Profile° VR/A-1 65 Fattening Enterotyphlitis B30 NT B1 II VR/A-2 65 Fattening Enterotyphlitis B14 O145 A III TV/B-1 70 Fattening Typhlitis B30 O103 A I TV/B-2 70 Fattening Typhlitis B14 NT A I TO/C-1 56 Fattening Enterotyphlitis B12 O103 B1 I TO/C-2 42 Weaning Liquid content in caecum B14 O103 B1 I TO/C-3 43 Weaning Liquid content in caecum B12 O103 B1 I TV/D-1 60 Fattening Enterotyphlitis B12 O148 A I TV/E-1 68 Fattening Enterotyphlitis B12 O103 B1 I PD/F-1 50 Weaning Enterotyphlitis B12 O157 B1 I VE/G-1 50 Weaning Severe enterotyphlitis B30 O145 B1 II TV/H-1 25 Nest Severe enterotyphlitis B30 O103 B1 I TV/I-1 21 Nest Enterotyphlitis B12 O103 B1 I TV/I-2 42 Weaning Enterotyphlitis B12 O103 D I TV/L-1 55 Fattening Enterotyphlitis B14 NT B1 I PD/M-1 7 Nest Enterotyphlitis B28 O145 B1 IV PD/M-2 48 Weaning Enterotyphlitis B30 O15 B1 II TV/N-1 48 Weaning Enterotyphlitis B12 O103 B1 I TV/O-1 75 Fattening Enterotyphlitis B12 O126 B1 I TV/L-2 70 Females Enterotyphlitis B14 O103 B1 I TV/L-3 42 Weaning Severe enterotyphlitis B12 NT B1 I TV/L-4 43 Weaning Severe enterotyphlitis B14 O59 B1 I TV/P-1 87 Fattening Enterotyphlitis B12 O157 B1 I TV/P-2 87 Fattening Severe enterotyphlitis B14 O103 B1 I TV/Q-1 50 Weaning Severe enterotyphlitis B14 O103 B1 I TV/R-1 77 Fattening Liquid content in caecum B28 O177 A I * Strain identification was defined by Province code/alphabetical code of farm-number of strain; # NT, notypeable; ° See Table II. First author et al. Nick title Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx 195 strains possessing the ler/cesT/espA-1/espB-3/tir-1/ eae(beta)/espD-2/eprJ/escN LEE gene variants were classified as the Afset group D and clustered into two profile differing for afr/2G or ralG fimbrial encoding gene, respectively. Two ralG-positive strains differ from this latter group just for the variant tir-3 or espB-2. The combined presence of adherence factors, porcine A/E-associated islet, lymphostatin and long polar fimbriae encoding genes (paa, efa1/ lifA, lpfA) was found in 13 strains with an additional 11 strains showing a positive hybridization for at least one of the three genes. Table III lists the percentage of isolates possessing virulence genes related to LEE T3SS-translocated or non-LEE T3SS-secreted proteins. The EspG protein encoding gene was diffusely detected in the strain collection as well as effector genes map and cif. All strains possessed the non-LEE effector C, (nleC), together with a variable presence of nleA, nleB, nleD, nleE, nleF, nleG and nleH. The reads obtained were trimmed using a in-house script that trims the first 15 nucleotides and uses as cutoff a mean quality of 30 and the minimum length of 50 bases. Trimmed reads were de novo assembled using SPADES version 3.0.0 (Bankevich et  al. 2012). Ambiguous nucleotides were solved by local reads remapping by Bowtie  2 (Langmead and Salzberg 2012), followed by alignment visual inspection. Results Epidemiological data and strain classification Table I summarizes the epidemiological data of colibacillosis outbreaks along with the related biochemical or genetic markers used to group the E. coli isolates. No apparent associations were found between epidemiological data (age, production unit, lesion degree) and either the biotype or phylogroup. All but four nontypable strains possessed the O-serogroup gene wzy. All strains but one belonged to phylogroups A and B1 with one strain belonging to phylogroup D. B1 was the prevalent phylogroup (76.9%) and was found mostly associated with the O103, O157, O126 wzy gene as well as notypable ones. An association between the O103 wzy gene and the TSPE4.C2 DNA fragment (phylogroup B1) with the rhamnose-negative fermentation character (biotype 12 or 14) was found but wasn’t statistically significant (α >0.05). Pathotyping and virulence gene profiles Table II shows the virulence-related gene profiles of E.  coli strains and their clustering into four groups based on the presence of the rabbit-specific adherence factor encoding genes (afr/2G or ralG) and variants of the LEE structural genes. Twenty-four Table II. Profiles of Locus of Enterocyte and Effacement (LEE) and adherence factor encoding genes in rabbit enteropathogenic Escherichia coli. Profile N° of involved strains LEE structural or secreted proteins* Fimbrial adhesin# Adherence factors (n) Phylogenetic group (n) Afset group I 21 ler, cesT, espA-1, espB-3, tir-1, eae(beta), espD-2; eprJ, escN afr/2G efa1/lifA(12), paa(17), lpfA(19) A (4); B1 (16); D (1) D II 3 ler, cesT, espA-1, espB-3, tir-1, eae(beta), espD-2; eprJ, escN, ralG efa1/lifA(3), paa(1), lpfA(2) B1 D III 1 ler, cesT, espA-1, espB-3, tir-3, eae(beta), espD-2; eprJ, escN ralG - A This study IV 1 ler, cesT, espA-1, espB-2, tir-1, eae(beta), espD-2; eprJ, escN ralG efa1/lifA, paa, lpfA B1 This study * Strains TV/H1, TV/I2 and TV/N1 possess additional escJ gene; espD and cesT genes were detected by specific PCRs; signal for espB-3 variant was below the threshold signal in strains TV/B2, TO/C3, TO/D1, TV/L4, TV/P1, TV/R1, VR/A2, VE/G1 and PD/M2, confirmed by specific espB PCR; signal for the tir-1 variant was below the threshold signal in strains TV/B2 and VR/A2, confirmed by specific tir PCR; # Strains TV/L-1 and TV/O-1 showing ambiguous signal for REPEC adhesins were analysed by genomic sequencing. Both strains showed 100% similarity with the described sequence of afr/2G (accession number U77302), 72% similarity with the sequence of ralG (accession number U84144). No sequence similar to bfpA (accession number U27184) was found. Table III. Frequence of effector enteropathogenic virulence factors in rabbit Escherichia coli. Gene N° of isolates tested positive LEE T3SS-effector espG 24 map/orf19 24 Non-LEE T3SS-secreted or structural protein cif 24 nleA 20 nleB 13 nleC 26 nleD 7 nleE 22 nleF 3 nleG 21 nleH 21 Nick title First author et al. 196 Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx 19 strains (73.1%), respectively. These gene products increase the colonization capability and severity of epithelial damage (Newton et al. 2004, Batisson et al. 2003). All strains possessed the eprJ and escN genes, responsible for secretion of T3SS basal body rod component and general T3SS translocation activity, respectively, both of which are involved in the activation of the inflammation mechanism (Miao et  al. 2010). Moreover, the confirmatory PCR test for espB gene and additional test for espD gene, together with general positive hybridizations for espA, attests to the ability of examined strains to carry out pore formation allowing the translocation of effector proteins. A LEE pattern, namely espG, map, as well as non-LEE (cif, nle family gene) effector genes were individually or in combination detected in all strains, confirming the genetic basis underlying the examined strain’s ability to induce epithelial damage and an irreversible cytopathic effect. Although no strain in our study possessed a member of the Shiga toxin gene family, finding chuA gene (a marker for heme transport in EHEC O157:H7) in TV/I-2 O103 strain as well as five other strains being characterized as O157 or O145 serogroup, is a reminder of the known plasticity of E. coli. Even in light of the recent haemolytic uraemic syndrome epidemics in Germany and France caused by bacteriophage-mediated acquisition of the stx2a gene by the enteroaggregative E.  coli O104:H4 strain (Scheutz et  al. 2011), the use of the DNA microarray technique in screening studies could be a support decision tool for more specific genetic investigations about the recombination capacity of E. coli pathotypes. Conclusion Despite the low overall number of strains tested, this study confirms that in Northern Italy rabbit colibacillosis was caused by agents having a defined genotype that was consistently found within a larger and variable virulence gene background. The rabbit species-specific adherence virulence factors, afr/2 or ral, characterize the REPEC strains. Overall, a prevalent attaching and effacing gene profile consisting of espA-1/espB-3/tir-1/eae(beta) places the examined REPEC strains into the Afset D group of atypical EPEC. The diffuse detection of additional adherence factors as efa1/lifA support this classification. The link between O145 serogroup and the Ral adherence factor suggests that routine REPEC diagnosis should incorporate ral, together with eae and afr/2, as PCR amplification targets. Finally, the absence of Shiga toxin encoding genes limits the zoonotic potential of tested strains. The application of microarray-based technologies Discussion The E.  coli virulence DNA microarray technology has proven to be a powerful tool for large scale genetic characterization of rabbit enteropathogenic E. coli and has allowed us to obtain genetic profiles consistent with its pathogenesis. In particular, it has produced a database for more specific analysis of gene products. The strains examined in this study were classified as aEPEC, thus confirming that the rabbit colibacillosis agents belong to this pathotype (Milon et  al. 1999, Licois and Marlier 2008). These strains possessed the rabbit specific adherence factors afr/2G or ralG, confirming the species-specificity of rabbit colibacillosis agents (REPEC). Interestingly, the afr/2G and ralG genes weren’t detected together in our strains, suggesting that an alternative genotype is expressed by operons of REPEC adherence locus. Although ral and afr/2 genes were not two-way associated with serogroup or biotype, the O145 (3 strains) and O103 (12 strains) serogroups were ral and afr/2 positive, respectively. The classification of strains according to biotype, phylogroup and O-antigen encoding gene, agreed with previous clinical and epidemiological studies on rabbit enteropathies. Indeed, all strains, except strain TV/I-2, belonged to phylogroups B1 and A, which generally encompasses E.  coli strains derived from animal origin (Clermont et  al. 2000). The most prevalent genetic pattern related to phylogroup/O-antigen/biotype, found in nine strains (about 35% of examined strains), was B1/ O103/rhamnose negative. This is consistent with previous investigations on rabbit colibacillosis in Italy (Badagliacca et al. 2010, Agnoletti et al. 2004). A LEE-encoding set of genes covering the cesT chaperon and the five operons related to the LEE regulator (ler) gene, the eae/tir domain (eae, tir) and the LEE type III secretion system (esp, esc and epr), was detected in all strains. In particular, the most prevalent combination of LEE gene variants espA-1/espB-3/tir-1/eae(beta) corresponds to the Afset D group. Afset and colleagues (Afset et  al. 2008) classified atypical EPEC associated with diarrhea in children into 11 group (A to K group) on the basis of variant distribution of these four LEE genes, compared to the phylogenetic group and the presence of the adherence factor efa1/lifA. The Afset D group is associated with prevalent A and B1 phylogroup and efa1/lifA gene. Therefore, the presence of the efa1/lifA gene in 16 strains (61.5%) and the above referred phylogrouping results strongly correlated with this classification. Conversely, the profiles espA-1/espB-3/tir-3/ eae(beta) and espA-1/espB-2/tir-1/eae(beta), found in the ralG-harbouring VR/A2 and PD/M1 strains were not included in the Afset classification. Finally, the lpfA and paa genes were detected in 22 (84.6%) and First author et al. Nick title Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx 197 investigations in monitoring programs aimed to survey the circulation of pathogenic strains within rabbit production units, their zoonotic genetic potential and to select E.  coli strains eligible for vaccinal prophylaxis in fattening rabbit production. in the diagnosis and genotyping of pathogenic rabbit E.  coli is a powerful tool to characterize both virulence potential and genomic lineages as well as to understand bacterial pathogenesis. It is a reliable, cost-effective candidate for more extensive Afset J.E., Anderssen E., Bruant G., Harel J., Wieler L. & Bergh K. 2008. Phylogenetic backgrounds and virulence profiles of atypical enteropathogenic Escherichia coli strains from a case-control study using multilocus sequence typing and DNA microarray analysis. J Clin Microbiol, 46, 2280-2290. Agnoletti F., Favretti M., Deotto S., Passera A., Tisato E., Bano L. & Mazzolini E. 2004. Report on enteropathogenic Escherichia coli (EPEC) isolated from enteric outbreaks in italian intensive rabbit herds. In Proc.: 8th World Rabbit Congress, Puebla, Mexico, 7-10 September 2004. World Rabbit Science Association, 416-421. http://world-rabbit-science.com/WRSA-Proceedings/ Congress-2004-Puebla/Puebla-2004-a.htm#pathology. Badagliacca P., Letizia A., Candeloro L., Di Provvido A., Di Gennaro A., Scattolini S., Pompei G., Pedicone R. & Scacchia M. 2010. Clinical, pathological and microbiological profiles of spontaneous enteropathies in growing rabbits. World Rabbit Sci, 18, 187-198. Badagliacca P., Masson L., Tonelli A., Migliorati G. & Harel J. 2014. Using DNA microarray technology to monitor distribution of antibiotic resistance genes in field strains of Escherichia coli isolated from animal outbreaks of colibacillosis. In Microarray: principles, applications and technologies (J. Rogers, ed.). Nova Science Publishers Inc., New York, USA, 187-201. Badea L., Doughty S., Nicholls L., Sloan J., Robbins-Browne R.M. & Hartland E.L. 2003. Contribution of efa/lifA to the adherence of enteropathogenic Escherichia coli to epithelial cells. Microbial Pathogens, 34, 205-215. Bankevich A., Nurk S., Antipov D., Gurevich A.A., Dvorkin M., Kulikov A.S., Lesin V.M., Nikolenko S.I., Pham S., Prjibelski A.D., Pyshkin A.V., Sirotkin A.V., Vyahhi N., Tesler G., Alekseyev M.A. & Pevzner P.A. 2012. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol, 19, 455-477. Batisson I., Guimond M-P., Girard F., An H., Zhu C., Oswald E., Fairbrother J.M., Jacques M. & Harel J. 2003. Characterization of the novel factor paa involved in the early steps of the adhesion mechanism of attaching and effacing Escherichia coli. Infect Immun, 71, 4516-4525. Bekal S., Brousseau R., Masson L., Prefontaine G., Fairbrother J. & Harel J. 2003. Rapid identification of Escherichia coli pathotypes by virulence gene detection with DNA microarrays. J Clin Microbiol, 41, 2113-2125. Bertin Y., Boukhors K., Livrelli V. & Martin C. 2004. Localization of the insertion site and pathotype determination of the locus of enterocyte effacement References of Shiga toxin-producing Escherichia coli strains. Appl Environ Microbiol, 70, 61-68. Bonnet C., Diarrassouba F., Brousseau R., Masson L. & Diarra M.S. 2009. Pathotype and antibiotic resistance gene distribution of Escherichia coli isolates from broiler chickens raised on antimicrobial supplemented diets. Appl Environ Microbiol, 75, 6955-6962. Bruant G., Maynard C., Bekal S., Gaucher I., Masson L., Brousseau R. & Harel J. 2006. Development and validation of an oligonucleotide microarray for detection of multiple virulence and antimicrobial resistance genes in Escherichia coli. Appl Environ Microbiol, 72, 3780-3784. Camguilhem R. & Milon A. 1989. Biotypes and O serogroups of Escherichia coli involved in intestinal infections of weaned rabbits: clues to diagnosis of pathogenic strains. J Clin Microbiol, 27, 743-747. Cantey J.R. & Blake R.K. 1977. Diarrhea due to Escherichia coli in the rabbit: a novel mechanism. J Infect Dis, 135, 454-462. Clermont O., Bonacorsi S. & Bingen E. 2000. Rapid and simple determination of the Escherichia coli phylogenetic group. Appl Environ Microbiol, 66, 4555-4558. Fiederling F., Boury M., Petit C. & Milon A. 1997. Adhesive Factor/Rabbit 2, a new fimbrial adhesin and a virulence factor from Escherichia coli O103, enteropathogenic for rabbit. Infect Immun, 65, 847-851. Frigon D., Biswal B., Mazza A., Masson L. & Gehr R. 2013. Biological and physicochemical wastewater treatment processes reduce the virulence risk of Escherichia coli. Appl Environ Microbiol, 79, 835-844. Garrido P., Blanco M., Moreno-Paz M., Briones C., Dahbi G., Blanco J. & Parro V. 2006. STEC-EPEC oligonucleotide microarray: a new tool for typing genetic variants of the LEE pathogenicity island of human and animal Shiga toxin-producing Escherichia coli (STEC) and enteropathogenic E. coli (EPC) strains. Clinical Chemistry, 52, 192-201. Giron J.A., Ho A.S. & Schoolnik G.K. 1991. An inducible bundle-forming pilus of enteropathogenic Escherichia coli. Science, 254, 710-713. Hamelin K., Bruant G., El-Shaarawi A., Hill S., Edge T.A., Bekal S., Fairbrother J.M., Harel J., Maynard C., Masson L. & Brousseau R. 2006. A virulence and antimicrobial resistance DNA microarray detects a high frequency of virulence genes in Escherichia coli isolates from Great Nick title First author et al. 198 Veterinaria Italiana 2018, 54 (3), xxx-xxx. doi: 10.12834/VetIt.xxxxx Salehi T.Z., Tonelli A., Mazza A., Staji H., Badagliacca P., Tamai I.A., Jamshidi R., Harel J., Lelli R. & Masson L. 2012. Genetic characterization of Escherichia coli O157:H7 strains isolated from the one-humped Camel (Camelus dromedarius) by using microarray DNA technology. Mol Biotechnol, 51, 283-288. Scheutz F., Møller Nielsen E., Frimodt-Møller J., Boisen N., Morabito S., Tozzoli R., Nataro J.P. & Caprioli A. 2011. Characteristics of the enteroaggregative Shiga toxin/ verotoxin-producing Escherichia coli O104:H4 strain causing the outbreak of haemolytic uraemic syndrome in Germany, May to June 2011. Euro Surveill, 16  (24), pii=19889. Schmidt H. & Hensel M. 2004. Pathogenicity islands in bacterial pathogenesis. Clin Microbiol Rev, 17, 14-56. Thomas N.A., Deng W., Puente J.L., Frey E.A., Yip C.K., Strynadka N.C.J. & Finlay B.B. 2005. CesT is a multi-effector chaperone and recruitment factor required for the efficient type III secretion of blth LEE- and non-LEE-encoded effectors of enteropathogenic Escherichia coli. Mol Microbiol, 57, 1762-1779. Tonelli A., Badagliacca P., Bruant G., Letizia A., Di Provvido A., Harel J. & Scacchia M. 2008. Genetic characterization of rabbit Escherichia coli strains with the use of microarray technology. In Proc. 9th World Rabbit Congress, Verona, Italy, 10-13 September 2008. World Rabbit Science Association, 1097-1101. http://world-rabbit-science. com/WRSA-Proceedings/Congress-2008-Verona/ Verona-2008-a.htm#pathology. Staji H., Badagliacca P., Salehi T.Z., Tamai IA, Lopes F., Iorio M., Mazza A, Tonelli A. & Masson L. 2017. Pathotyping of diarrheagenic cattle Escherichia coli strains isolated in the Province of Tehran. Vet Ital, 53 (4), 345-356. doi: 10.12834/VetIt.728.3521.1. Zhu C., Agin T.S., Elliot S.J., Johnson L.A. Thate T.E., Kaper J.B. & Boedeker E.C. 2001. Complete nucleotide sequence and analysis of the Locus of Enterocyte Effacement from rabbit diarrheagenic Escherichia coli RDEC-1. Infect Immun, 69, 2107-2115. Lakes recreational waters. Appl Environ Microbiol, 72, 4200-4206. Krejany E.O., Grant T.H., Bennet-Wood V., Adams L.M. & Robins-Browne R.M. 2000. Contribution of plasmid-encoded fimbriae and intimin to capacity of rabbit-specific enteropathogenic Escherichia coli to attach and colonize rabbit intestine. Infect Immun, 68, 6472-6477. Langmead B. & Salzberg S. 2012. Fast gapped-read alignment with Bowtie 2. Nature Methods, 9, 357-359. Licois D. & Marlier D. 2008. Pathologies infectieuses du lapin en élevage rationnel. INRA Prod Anim, 21, 257-268. Luo W. & Donnenberg M.S. 2011. Interactions and predicted host membrane topology of the enteropathogenic Escherichia coli translocator protein EspB. J Bacteriol, 193, 2972-2980. Marchés O., Ledger T.N., Boury M., Ohara M., Tu X., Goffeaux F., Mainil J., Rosenshine I., Sugai M., De Rycke J. & Oswald E. 2003. Enteropathogenic and enterohemorrhagic Escherichia coli deliver a novel effector called Cif, which bocks cell cycle G2/M transition. Mol Microbiol, 50, 1553-1567. Miao E.A., Mao D.P., Yudkovsky N., Bonneau R., Lorang C.G., Warren S.E., Leaf I.A. & Aderem A. 2010. Innate immune detection of the type III secretion apparatus through the NLRC4 inflammasome. Proc Natl Acad Sci USA, 107, 3076-3080. Milon A., Oswald E. & De Rycke J. 1999. Rabbit EPEC: a model for the study of enteropathogenic Escherichia coli. Vet Res, 30, 203-219. Newton H.J., Sloan J., Bennet-Wood V., Adams L.M., Robins-Browne R.M. & Hartland E.L. 2004. Contribution of long polar fimbriae to the virulence of rabbit-specific enteropathogenic Escherichia coli. Infect Immun, 72, 1230-1239. Nguyen R.N., Taylor L.S., Tauschek M. & Robins-Browne R.M. 2006. Atypical enteropathogenic Escherichia coli infection and prolonged diarrhea in children. Emerg Infect Dis, 12, 597-603.