Prevalence of virulence genes in Escherichia coli strains isolated from piglets in the suckling and weaning period in Mexico

Transcription

1 Journal of Medical Microbiology (2012), 61, DOI /jmm Prevalence of virulence genes in Escherichia coli strains isolated from piglets in the suckling and weaning period in Mexico Andrea Toledo, 1 Daniela Gómez, 1 Celene Cruz, 2 Rosalba Carreón, 2 Jorge López, 2 Silvia Giono 3 and Ana María Castro 1 Correspondence Andrea Toledo antoro06@yahoo.com 1 Departamento de Salud Pública, Facultad de Medicina, Universidad Nacional Autónoma de México, México DF 04510, Mexico 2 Departamento de Producción Animal: cerdos, Facultad de Medicina Veterinaria y Zootecnia, Universidad Nacional Autónoma de México, México DF 04510, Mexico 3 Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, DF, Mexico Received 15 February 2011 Accepted 17 August 2011 Faecal Escherichia coli isolates from suckling (n5503) and weaning (n5450) piglets with and without diarrhoea from 10 farms in Mexico were examined for identification and prevalence of virulence genes. E. coli isolates were tested further for enterotoxin (LT, STa, STb, Stx1, Stx2 and EAST1), fimbrial (F4, F5, F6, F17, F18 and F41) and eae adhesin genes by multiplex PCR. Of the 953 isolates of E. coli examined by multiplex PCR, 650 (68.2 %) isolates were positive for at least one adhesin gene. Among the isolates from diarrhoeic piglets, F41 (72 %) was the most prevalent adhesin followed by eae adhesin (27 %), F6 (12 %), F17 (9 %), F18 (9 %), F5 (8 %) and F4 (3 %). Enterotoxin genes were detected in 424 (44.5 %) of the isolates, of which EAST1 (38 %) and STa (30 %) were the most common, followed by STb (17 %), Stx2 (6 %) and LT (5 %). Twenty-three per cent of isolates from suckling piglets and 43 % of isolates from weaned piglets carried both enterotoxin and adhesin genes, the most common virotypes being F41 : STa, F41 : EAST1, EAST1 : eae, F41 : F6, F41 : STa : STb, F41 : eae and F17 : eae. The present study examined for the first time, to our knowledge, the prevalence of 13 virulence genes among E. coli strains isolated from piglets with and without diarrhoea in Mexico. The results suggested that there are a wide variety of virulence genes associated with diarrhoea in piglets. This study provides baseline information on the significance of specific virotypes associated with suckling and weaning periods in piglets in Mexico. INTRODUCTION Pathogenic Escherichia coli strains are common agents responsible for a variety of intestinal disorders. Diarrhoea remains an important cause of morbidity and mortality in livestock and is one of the most common diseases of suckling and weaning piglets worldwide. These diseases have negative economic effects in the pork industry due to high mortality and reduced growth rate (Vu-Khac et al., 2004; Zhang et al., 2007; Madoroba et al., 2009). In neonatal and weaning piglets, E. coli induces acute and watery diarrhoea that may be followed by terminal septicaemia, an important cause of economic loss for swine producers (Lee et al., 2008). Most types of diarrhoea are caused by strains of enterotoxigenic E. coli (ETEC) and Abbreviations: EAST1, enteroaggregative E. coli heat-stable enterotoxin 1; EPEC, enteropathogenic E. coli; ETEC, enterotoxigenic E. coli; LT, heat labile; OR, odds ratio; ST, heat stable; STEC, Shiga toxin-producing E. coli. remain a major cause of both death and disease in newborn or newly weaned pigs (Holoda et al., 2005; Vu-Khac et al., 2006). The main virulence factors associated with ETEC in diarrhoea are enterotoxins and fimbrial adhesins. Fimbrial adhesins mediate attachment of bacteria to the surface of the intestinal epithelium cells, thus allowing bacterial colonization. Both heat-labile (LT) and heat-stable (STa and STb) enterotoxins cause an imbalance in intestinal homeostasis, causing hypersecretion of fluids that results in diarrhoea. The known porcine fimbriae are F4, F5, F6, F17, F18 and F41 (Vu-Khac et al., 2007; Kim et al., 2010). According to the work of Fairbrother et al. (2005) and Kim et al. (2010), different types of adhesin can be associated with ETEC diarrhoea in animals of different ages. F4 causes more severe diarrhoea in weaned piglets, causing postweaning diarrhoea, but is also associated with diarrhoea in nursing pigs. The F5, F6 and F41 genes are usually associated with ETEC from neonatal pigs (between 4 and G 2012 SGM Printed in Great Britain

2 Porcine E. coli pathotypes in Mexico 14 days of age), but may be associated less frequently with diarrhoea in older pre-weaned pigs (Kwon et al., 2002). Post-weaning diarrhoea is the most constant disease problem in large-scale farming, particularly among piglets weaned at 4 weeks of age. F4 and F18 are the fimbrial adhesins of ETEC associated with post-weaning diarrhoea (Ojeniyi et al., 1994; Imberechts et al., 1997; Bertschinger et al., 2000; Vu-Khac et al., 2007; Kim et al., 2010). F18 is the fimbrial adhesin most commonly found on Shiga toxin-producing E. coli (STEC) causing oedema disease (Vu-Khac et al., 2007). F18 adhesin occur as two antigen variants, F18ab and F18ac. F18ab is usually found on STEC, ETEC or ETEC/STEC and F18ac is usually expressed by ETEC (Fairbrother et al., 2005). This typically produces STa and STb toxins, sometimes associated with Stx2e toxin (Rippinger et al., 1995; Francis, 2002). Another virulence determinant of the ETEC virotype is the enteroaggregative E. coli heat-stable enterotoxin 1 (EAST1), which has been found to be highly prevalent in isolates from diarrhoea in humans and animals. The asta gene encoding EAST1 has also been detected in E. coli of different pathogenic strains, such as ETEC, enteropathogenic E. coli (EPEC) and STEC from humans and animals (Frydendahl, 2002; Ngeleka et al., 2003; Osek, 2003; Vu- Khac et al., 2007; Zhang et al., 2007). Through PCR tests, Choi et al. (2001) found that 31 % of E. coli strains isolated from weaning piglets with diarrhoea or oedema disease had the EAST1 gene. STEC comprises a serologically diverse group of pathogens that cause disease in weaning pigs (Barman et al., 2008). The two main groups of Shiga toxin are Shiga toxin 1 (Stx1) and Shiga toxin 2 (Stx2), and their variants (Stx1c, Stx1d, Stx2c, Stx2d, Stx2e, Stx2f, Stx2g). The Stx2e variant has been associated with oedema disease in pigs (Imberechts et al., 1992; Kim et al., 2010). Additionally to ETEC and STEC strains, some strains cause attaching and effacing lesions, very similar to those produced by EPEC in humans, and have been associated with post-weaning diarrhoea in pigs (Vu-Khac et al., 2007). These attaching and effacing E. coli strains carry the eae gene encoding the outer-membrane protein known as intimin, which is involved in the attachment of the bacteria to the surface of epithelial cells in the gastrointestinal tract. This gene is a marker for porcine EPEC; however, the pathogenic significance of porcine eae strains in weaning pigs is unclear because some eae-positive isolates may also be found in healthy pigs (Fairbrother et al., 2005; Malik et al., 2006; Cheng et al., 2006; Zhang et al., 2007). In Mexico, there are no previous reports about the prevalence of E. coli virulence genes in pigs for diagnostic purposes. It should be noted that this study is among the first in Mexico to look at the prevalence of these virulence genes. A better understanding of the repertoire of virulence genes can provide us with important information for diagnosis and prevention in order to control diarrhoeal diseases in swine. METHODS Bacterial samples and their characterization. A total of 953 E. coli isolates were obtained from faecal swabs (643) collected from suckling (0 3 weeks of age) and weaning (3 5 weeks of age) piglets with watery diarrhoea (brown or yellow) or without diarrhoea (Table 1). The sows of the piglets sampled in this study were not vaccinated against E. coli. The strains were obtained from 10 geographically separated pig farms in Mexico. The farms are located in the central region of Mexico, which produces about 50 % of the country s pork (SAGARPA, 2010) (Table 1). The samples collected were transported in Stuart s medium (DIBICO) at 4 uc, and after arriving at the laboratory, were spread on MacConkey agar and Eosin Methyl Blue Agar and incubated at 37 uc aerobically overnight. Two fermenting colonies with the Table 1. Total number of faecal samples collected from different farms in Mexico The total number of E. coli isolates from faecal swabs is shown in parentheses. Farm location Lactation type Suckling with diarrhoea Suckling without diarrhoea Weaning with diarrhoea Weaning without diarrhoea Total isolates examined per farm México City 21 (28) 10 (9) 33 (45) 9 (9) 91 Edo. de México 26 (48) 13 (26) 25 (36) 2 (4) 114 Guanajuato 23 (42) 11 (12) 30 (43) 10 (11) 108 Guerrero 30 (52) 10 (18) 30 (56) 10 (18) 144 Hidalgo 31 (32) 10 (10) 9 (16) 9 (14) 72 Jalisco 30 (36) 10 (11) 30 (42) 8 (12) 101 Michoacán 20 (29) 7 (14) 7 (13) 12 (15) 71 Morelos 30 (35) 10 (16) 30 (51) 10 (21) 123 Puebla 10 (13) 5 (8) 21 (32) 11 (12) 65 Querétaro 30 (48) 10 (16) Total 251 (363) 96 (140) 215 (334) 81 (116)

3 A. Toledo and others appearance of E. coli were randomly selected from each agar medium and were confirmed to be E. coli using conventional biochemical procedures (MacFaddin, 1980). One to four isolates of E. coli were obtained from each faecal swab; virulence genes were identified in only one or two of these E. coli isolates. Isolates were then cultured on sheep blood agar plates aerobically overnight at 37 uc to evaluate b- haemolysis. b-haemolysis was defined as a zone of complete erythrocyte lysis surrounding a bacterial colony. After all of these E. coli bacteria were isolated, they were stored on tryptic soy agar at 4 uc and in Luria Bertani broth containing 20 % glycerol at 220 uc, until further use. Reference strains. A panel of E. coli reference strains were used as positive controls for the multiplex PCR and a non-pathogenic commensal E. coli strain was used as negative control. Strains FV10186 (eae), FV10187 (STa, STb, Stx1, Stx2, F18), FV10188 (LT, STb, F4), FV10190 (F6) and FV10191 (F41, F5) were kindly provided by Dr Jorge Blanco (Laboratorio de Referencia de E. coli, Spain). O157 : H7 (Stx1, Stx2), H1O407 (LT, ST) and C600 (K12) were provided by the Instituto Nacional de Referencia Epidemiológica, Mexico. DNA extraction and PCR. Genomic DNA of E. coli isolates was extracted by the boiling method, as described previously (Al-Gallas et al., 2007). PCR (single or multiplex) to detect toxin genes (LT, STa, STb, Stx1, Stx2 and EAST1) and adhesin genes [F4, F5, F6, F17, F18 (F18ab and F18ac), F41] as well as intimin encoded by the eae gene was carried out as previously described (Table 2). Four different multiplex PCRs for toxin and adhesin genes were carried out using Apex Taq DNA polymerase (Genesee Scientific) and primers, as shown in Table 2. The 50 ml reaction mixtures contained the following: 16 PCR buffer (100 mm Tris/HCl, ph 8.5; 500 mm KCl; 15 mm MgCl 2 ; 1 % Triton X-100), 0.2 mm each dntp (Apex dntp mix, 10 mm), 0.25 mm each primer, 2.0 U Apex Taq DNA polymerase, 5 ml template DNA and 50 ml nucleasefree water. Multiplex PCR products were separated in a 2 % agarose gel (Amresco). Reference positive and negative control strains (kindly donated by Dr Jorge Blanco) and a 100 bp ladder (Amresco) were used to identify amplified products. Amplified PCR products were visualized with ethidium bromide under a UV illuminator and recorded using a Digidoc imaging system (UVP). Some samples of each identified gene were subjected to DNA sequencing methods. For this purpose, the PCR products were purified using the QIAquick PCR purification kit (Qiagen). Automated DNA sequencing. Automated DNA sequencing was carried out using capillary-based electrophoresis sequencers. The unit utilizes ABI Prism 310 (1-capillary) and ABI Prism 3100 (16- capillary) Genetic Analyzers from Applied Biosystems, with BigDye Terminator Cycle Sequencing chemistry. Samples of PCR products were sequenced with readings of length up to 700 bases of highquality sequence data per reaction, providing high-quality templates. Statistical analysis. An analysis of strength of association with odds ratio (OR) was made. x 2 was calculated to determine the statistical significance at 95 % confidence. Management of the database was done in the SPSS 10 program. Table 2. PCR primers used to amplify specific fragments from enterotoxin and fimbrial adhesin genes Target gene Amplified products (bp) Primer (sense, antisense) Annealing temp. (6C) Reference LT ATT TAC GGC GTT ACT ATC CTC TTT TGG TCT CGG TCA GAT ATG-39 EAST CCA TCA ACA CAG TAT ATC CGA GGT CGC GAG TGA CGG CTT TGT-39 STa TCTTTCCCCTCTTTTAGTCAG ACAGGCAGGATTACAACAAAG-39 Stx CGC TGA ATG TCA TTC GCT CTG C CGT GGT ATA GCT ACT GTC ACC-39 Stx CTT CGG TAT CCT ATT CCC GG CTG CTG TGA CAG TGA CAA AAC GC-39 F GGT CGC GAG TGA CGG CTT TGT CCA CTG AGT GCT GGT AGT TAC AGC C-39 F TGC GAC TAC CAA TGC TTC TG TAT CCA CCA TTA GAC GGA GC-39 F TCT GCT CTT AAA GCT ACT GG AAC TCC ACC GTT TGT ATC AG-39 F GTG AAA AGA CTA GTG TTT ATT TC CTT GTA AGT AAC CGC GTA AGC-39 F GAG GGA CTT TCA TCT TTT AG AGT CCA TTC CAT TTA TAG GC-39 STb GCC TAT GCA TCT ACA CAA TC TGA GAA ATG GAC AAT GTC CG-39 F GGG CTG ACA GAG GAG GTG GGG C CCC GGC GAC AAC TTC ATC ACC GG-39 eae GGA ACG GCA GAG GTT AAT CTG CAG GGC GCT CAT CAT AGT CTT TC Osek et al. (1999) 60 Yamamoto & Nakazawa (1997) 60.5 Blanco et al. (2004) 60.5 Blanco et al. (2004) 60.5 Blanco et al. (2004) 55.5 Vu-Khac et al. (2006) 55.5 Ojeniyi et al. (1994) 55.5 Ojeniyi et al. (1994) 55 Imberechts et al. (1997) 55 Ojeniyi et al. (1994) 55 Ojeniyi et al. (1994) 56.5 Vu-Khac et al. (2006) 56.5 Blanco et al. (2004) 150 Journal of Medical Microbiology 61

4 Porcine E. coli pathotypes in Mexico Table 3. Adhesin genes detected in E. coli strains from piglets with and without diarrhoea Lactation type (no. of isolates) Percentage (number) of positive strains F4 F5 F6 F17 F18 F41 eae Suckling with diarrhoea (363) 1 (4) 3 (12) 5 (19) 6 (21) 3* (9) 34 (124) 12 (45) Suckling without diarrhoea (140) 1 (2) 4 (5) 8 (11) 5 (7) 0 38 (53) 9 (13) Weaning with diarrhoea (334) 2 (6) 5 (18) 7 (23) 3 (10) 6* (19) 38 (126) 15 (51) Weaning without diarrhoea (116) 5 (6) 4 (5) 3 (3) 1 (1) 2 (2) 39 (45) 9 (10) *P RESULTS The biotyping of faecal samples from piglets allowed the isolation and identification of 953 isolates of E. coli: 697 (73.1 %) were obtained from piglets with diarrhoea, 363 and 334 from piglets in either the suckling or weaning period, respectively; and 256 (26.9 %) were from piglets without diarrhoea, 140 strains in the suckling period and 116 strains in the weaning period (Table 1). Adhesin genes Of the 953 isolates of E. coli evaluated by PCR (multiplex or single), 650 (68.2 %) were positive for at least one adhesin gene. Among the adhesin isolates, F41 fimbria (348, 54 %) was the most prevalent fimbria followed by eae adhesin (119, 18.3 %), F6 (56, 8.6 %), F5 (40, 6.2 %), F17 (39, 6.0 %), F18 (30, 4.6 %) and F4 (18, 2.8 %). With respect to associations of fimbrial genes with age or presence/absence of diarrhoea, we only found statistically significant differences among isolates from healthy piglets and piglets with diarrhoea in the suckling (P50.05, OR53.43) and weaning (P50.05) period for the F18 fimbrial gene (Table 3). Although adhesin genes were isolated from both healthy piglets and piglets with diarrhoea, in general we found a higher percentage of adhesin genes in weaned piglets with diarrhoea than in suckling piglets with diarrhoea (Table 3). Enterotoxin genes Enterotoxin genes were detected in 424 isolates of E. coli (44.5 %); of these, STa (159, 37.5 %) and EAST1 (153, 36.1 %) were the most common, followed by STb (65, 15.6 %), Stx2 (24, 6.0 %) and LT (20, 4.7 %). As for adhesin genes, we found a higher percentage of positive genes in isolates from animals with diarrhoea than from healthy animals (Table 4). In E. coli isolates from diarrhoeic piglets, we only found statistically significant differences for the EAST1 gene (P50.007, OR52.3) in suckling piglets and in weaned piglets (P50.001, OR53.98) with diarrhoea (Table 4). The results for identification of enterotoxins and adhesin genes were corroborated by DNA sequencing of the genes found. The sequencing of these genes produced alignments with an identity of.94 % for each of the genes identified in this study, which have been previously reported in the National Center for Biotechnology Information. Common virotypes identified in E. coli isolates from suckling and weaning piglets Isolates could only be considered pathogenic if carrying enterotoxin genes combined with different adhesin genes. Isolates revealed a high diversity with a varying frequency and distribution for each single pathogenic gene. In the E. coli isolates from suckling piglets with diarrhoea (363), we found that 234 (64.5 %) were positive for at least one adhesin gene and 114 (31.4 %) were positive for enterotoxin genes; in weaning piglets with diarrhoea (334), 253 isolates (75.7 %) were positive for adhesin genes and 218 (65.3 %) were positive for toxin genes; while 23 % (83/363) of isolates from suckling piglets and 43 % (83/363) of isolates from weaned piglets carried both enterotoxin and adhesin genes Table 4. Enterotoxin genes detected in E. coli strains from piglets with and without diarrhoea Lactation type (no. of isolates) Percentage (number) of positive strains LT STa STb Stx1 Stx2 EAST1 Suckling with diarrhoea (363) 1 (2) 8 (30) 3 (11) 0 (1) 1 (4) 18* (66) Suckling without diarrhoea (140) 0 24 (33) 1 (1) 1 (1) 1 (2) 9 (12) Weaning with diarrhoea (334) 4 (13) 22 (74) 14 (45) 0 (1) 5 (16) 20* (68) Weaning without diarrhoea (116) 4 (5) 19 (22) 7 (8) 0 2 (2) 6 (7) *P

5 A. Toledo and others (Table 5). This indicates that our isolates could be considered enteropathogens. With respect to the association of enterotoxin and adhesin genes, we found 97 different virotypes. We found more gene combinations in isolates from piglets with diarrhoea than without it, regardless of the suckling or weaning stage (Table 5). In piglets in the suckling stage with diarrhoea, we found 41 different virotypes in 86 isolates. The most prevalent combinations were F41 : EAST1 (4 %), F41 : STa, F41 : F6, F41 : eae, EAST1 : eae (2 % each), and F18 : F41, F18 : F41 and F5:F6: F41 (1 % each). In piglets in the weaning stage with diarrhoea, we found 66 different virotypes in 131 isolates of E. coli. Of these, the most prevalent were the virotypes F41:STa (6 %), F41:EAST1 (4 %), STb:EAST1 (2 %), F41:STa:STb (2 %), STa : STb (2 %), F6 : eae (2 %), and F41 : STb : EAST1, LT : F18 : STa, STa : EAST1, F17 : eae, F41 : eae, F41 : STx2, F5 : F6 : F41 and F18 : F41 (1 % each). We found other less frequent virotypes; interestingly, most of them in isolates from pigs with diarrhoea. Although in piglets in the suckling and weaning stages without diarrhoea (control samples) 13 and 18 different combinations were found, respectively, these were present in a very low number of samples (43 and 31 samples, respectively). The most prevalent virotypes in suckling pigs were F6/STa (5 %), EAST1 : eae (3 %), F41 : STa (3 %) and F5 : STa : eae (2 %). In weaned piglets, the most prevalent were F41 : STa (6 %) and STa : EAST1 (2 %). As shown in Table 5, the distribution and prevalence of different virotypes regarding the location of the farms was very heterogeneous. The most common and prevalent virotypes were located on farms in the states of Guerrero, Jalisco, Michoacán, Morelos, México City, Queretaro, Puebla and Estado de México. Haemolytic activity in strains of E. coli Of the 953 E. coli isolates tested, only 83 (8.7 %) showed b- haemolytic activity (Table 5), indicated by a zone of complete erythrocyte lysis around bacterial colonies after overnight incubation at 37 uc on 5 % sheep blood agar plates. Of these, 63 (76 %) were isolates from piglets with diarrhoea and 20 (24 %) were from healthy piglets. Of the isolates from piglets with diarrhoea, 57 % corresponded to the weaning stage and 86 % of them expressed adhesin and enterotoxin genes. b-haemolysis was observed in isolates from other different virotypes, and in isolates that did not show virulence genes. Table 5. Percentage of positive virotypes identified in E. coli isolated from suckling and weaning pigs Virotype Suckling with diarrhoea (27*) Suckling without diarrhoea (7*) Weaning with diarrhoea (36*) Weaning without diarrhoea (13*) Farm location F41 : STa Guerrero, Jalisco, Michoacán, Morelos, Puebla, Estado de México, México City F41 : EAST Guerrero, Michoacán, Morelos, México City EAST1 : eae Guerrero, Querétaro, Hidalgo, Jalisco, Morelos, Guanajuato, México City F41 : F Guerrero, Morelos F41 : STa : STb Querétaro, Hidalgo, Puebla F41 : eae Guerrero, Morelos, Estado de México STa : EAST Jalisco, Michoacán, Estado de México, México City STb : EAST Guerrero, Hidalgo, Morelos, Puebla, Guanajuato F18 : F Guerrero, Michoacán, Morelos, Estado de México F17 : eae Jalisco, Estado de México, México City STa : STb Querétaro, Estado de México F6 : eae Jalisco F41 : F Morelos F41/STx Morelos F41 : STb : EAST Guerrero, México City F41 : F5 : F Morelos LT : F18 : STa Puebla F5 : STa : eae Querétaro, México City Other virotypes All farms None All farms *Number of b-haemolytic isolates is shown in parentheses. 152 Journal of Medical Microbiology 61

6 Porcine E. coli pathotypes in Mexico DISCUSSION The present report describes what is believed to be the first study in Mexico of a large collection of pathogenic E. coli strains, as well as the design and evaluation of a multiplex PCR assay for the identification of different enterotoxin and adhesin genes that cause diarrhoea in strains isolated from piglets in the suckling and weaning stages in Mexico. Likewise, this work shows preliminary epidemiological information about the importance of E. coli as a causative agent of diarrhoeal disease in piglets in Mexico. It is commonly accepted that serotypes and virotypes of ETEC are responsible for neonatal diarrhoea and postweaning diarrhoea in piglets and that STEC is a cause of diarrhoea only if enterotoxins are expressed by STEC strains. However, the distribution and frequency of serotypes and virotypes can vary considerably from region to region and over time in a given region, while only a few of these strains have been associated with enteric infections in pigs (Blanco et al., 1997; Vu-Khac et al., 2006). This study showed that, for isolates from weaning piglets with diarrhoea, 65.3 % were positive for enterotoxin genes and 75.7 % were positive for adhesion genes. In suckling piglets, this positive percentage was lower: 31.4 % and 64.5 % were positive for enterotoxins and adhesin genes, respectively. A higher proportion of virulence genes were present in the weaning stage. This might be explained because piglets in this period lack maternal immunity, which may make them non-immune to pathogenic E. coli (Barman et al., 2008). In the same way, temperature variation of the environment and the stress associated with changes in both accommodation and diet may be important factors that trigger the proliferation of pathogenic E. coli in the intestine (Leman et al., 1999). Interestingly, genes encoding E. coli virulence factors are found in piglets with diarrhoea and in healthy piglets with a higher prevalence in weaned piglets. Suckling and weaned piglets with diarrhoea showed significant differences only in the gene encoding the toxin EAST1, P and P50.004, respectively (Table 4). In relation to genes for fimbrial adhesins, we only found statistically significant differences in the suckling (P 0.05) and weaned (P 0.05) piglets with diarrhoea for the F18 gene (Table 3). It is noteworthy that the F41 fimbrial gene was found in isolates from both healthy and diarrhoeic piglets. Also, the F41 gene is present in different virotypes which are present in a greater percentage of isolates from piglets with diarrhoea than from healthy piglets (Table 5). We think that these virotypes present in healthy piglets might be expressed if the immune response is induced by stress, temperature changes and diet, and consequently could cause disease. The role of EAST1 toxin in swine colibacillosis has not been clearly determined, but this toxin has been commonly found in E. coli strains associated with diarrhoea in suckling and weaning piglets (Vu-Khac et al., 2004). Interestingly, our results are compatible with these observations because we found the EAST1 toxin gene in 32 % of 424 isolates of E. coli from diarrhoeic piglets, and most of these isolates were associated with the genes F41 (8 %) and eae (3 %). About 8.4 % of the isolates carried the EAST1 toxin gene as the only toxin gene. These results agree with those of Choi et al. (2001), Osek (2003) and Vu- Khac et al. (2006), who reported the presence of the EAST1 gene in 31 %, 46 % and 68 % of isolates, respectively. They also found EAST1 as the only toxin gene in about 14 %, 4 % and 3 % of the strains. With respect to healthy pigs, we found that 7 % of the isolates expressed the EAST1 gene. This is consistent with the study by Ngeleka et al. (2003), who identified the presence of the EAST1 gene in isolates from healthy piglets, suggesting that when EAST1 is expressed alone, it is probably insufficient to cause diarrhoea in piglets. Whether EAST1 toxin alone or in association with other toxins is sufficient to cause diarrhoea in pigs is currently under investigation. The high frequency of the EAST1 gene along with its high OR in isolates of E. coli is an incentive for further studies to investigate the significance of this toxin in porcine diarrhoea in Mexico. STa and STb positive isolates were more prevalent than LT positive isolates. We found that 8 22 % of isolates from sick animals carried the gene for STa; most of the isolates (8 %) were associated with F41, and in 20 % it was detected as the only toxin gene. Our observation agreed with other studies in which 9 and 11 % of isolates were positive for STa toxin in suckling and weaned piglets, respectively (Lee et al., 2008). Do et al. (2006) reported the presence of STa in 92.1 % of isolates from weaning piglets with diarrhoea; Parma et al. (2000) reported the presence of this gene in 24.5 % of isolates, it being the predominant gene in isolates from piglets from 1 week to 3 months of age. Particularly striking was the low presence of the LT gene in E. coli isolates from Mexican piglets, since the gene was present in only 1 % and 4 % of the isolates from piglets with diarrhoea. Most isolates were associated with other toxins and adhesins, and all of them were b-haemolytic also. We found Shiga toxins (Stx1 and Stx2) in only a few isolates. Stx1 was identified in 0.4 %, and Stx2 in 1 % and 5 % of the isolates from the diarrhoeic piglets. These toxins are produced by isolates of human, bovine and ovine STEC associated with haemorrhagic colitis, and haemolytic uraemic syndrome in humans. Our results confirm the observation made by other authors that piglets are not a reservoir of STEC pathogenic for humans (Vu-Khac et al., 2006). The outcome of multiplex PCR for adhesin genes provides evidence for the presence of F41, F6, F5, F17, F18 and F4 fimbriae genes in piggeries in Mexico. These fimbriae are highly associated with ETEC diarrhoea in piglets in experimental animals and in the field (Vu-Khac et al., 2006). The F41 adhesin gene is not as prevalent in 153

7 A. Toledo and others countries such as Slovakia, Korea, USA and Vietnam (Vu- Khac et al., 2006; Do et al., 2006; Zhang et al., 2007; Lee et al., 2008). In our study, the F41 adhesin gene was strongly associated with the population of piglets, since on average 36 % of the isolates from piglets with diarrhoea were positive, regardless of the suckling or weaning stage. Most of the F41 positive strains were associated with EAST1 (8 %), STa (8 %) and eae (3 %) genes. We also found F41 to be the only adhesin gene in 35 % of the E. coli isolates and in 27 % of the isolates that produce b-haemolysis. While this gene was also found in healthy animals, it is interesting to note that in the vast majority of isolates from pigs with diarrhoea, F41 was associated with one or more enterotoxins (Table 5). Recently, in another study performed by Castro & Toledo (2010) on E. coli isolates from piglets with diarrhoea, suckled on breast milk and milk substitute, the most prevalent fimbrial adhesin was also F41. This fact has great importance particularly because it allows us to consider this fimbria as an important marker of pathogenicity of E. coli in Mexican isolates, and it may be useful for both diagnostic purposes and for the development of new immunogens for the prevention of disease in Mexican piglets. In the future, we intend to test F41 and eae fimbria along with EAST1 and STa toxins as immunogens for the prevention of diarrhoea caused by E. coli in piglets. F6 adhesin was present in E. coli isolated from piglets in the suckling period. These data are consistent with those previously reported by Wilson & Francis (1986) and Kwon et al. (1999). However, we also found this adhesin in weaned piglets with diarrhoea (Table 3). In Mexico, piglets are weaned between 4 and 5 weeks of age, which agrees with the results of Harel et al. (1991), who associated fimbrial adhesins F5, F6 and F41 alone or in various combinations in porcine ETEC strains in pigs aged between 7 and 27 days. Recently, Kim et al. (2010) reported that the presence of adhesin F6 could cause milder diarrhoea with later onset (between 4 and 14 days of age), but it may also appear occasionally in older pigs, taking into account that the age of weaning has been reduced substantially over the years (Fairbrother et al., 2005). We found an association of F6 with eae in 2 % of the E. coli isolates from weaned pigs with diarrhoea, which may explain the presence of diarrhoea in these pigs (Table 5). The F5 adhesin gene was present in 3 % and 5 % of the isolates from piglets with diarrhoea. Some reports have shown the association of F41 with F5, because these genes are co-expressed (Madoroba et al., 2009); nevertheless, we only found this association in two isolates from piglets with diarrhoea. The gene encoding F18 adhesin is one of the most important to have been recently associated with post-weaning diarrhoea and with oedema disease (Bertschinger et al., 2000; Frydendahl, 2002; Vu-Khac et al., 2006). Interestingly, we found F18 adhesin in 28 isolates (4.3 %) from diarrhoeic piglets, most of them in the weaning stage (OR53.43). This finding agrees with Zhang et al. (2007), who found that 15 isolates carried the F18 gene as the only adhesin gene. In our study, we found the F18 gene associated with the F41 adhesin gene in 2 % of the isolates and with LT and STa in 1 % of the isolates, which were also b-haemolytic. There are two variants of the F18 family, F18ab and F18ac, and some authors have found associations in isolates between STa and/ or STb with or without Stx2 which possess F18ac (Vu-Khac et al., 2004). Interestingly, our results are similar to these observations, because one isolate contained genes for both enterotoxins and Stx2, another had both enterotoxins Stx2 and the eae gene and another had the virotype F18 : F41 : STb : STa : Stx2. The F17 adhesin gene of E. coli has been isolated frequently from calves with or without diarrhoea. We found it in approximately 9 % of the isolates from piglets with diarrhoea, most of them from suckling piglets. We also found it associated with eae adhesin (2 %). F17 is frequently associated with cattle isolates, but recently Vu-Khac et al. (2007) and Sting & Stermann (2008) found it in E. coli isolates from post-weaning piglets with diarrhoea. The eae gene is necessary for the intimate union of the bacterium to epithelial cells in vivo. It encodes the outermembrane 94 kda protein intimin, and has been recovered from E. coli associated with porcine neonatal diarrhoea as well as post-weaning diarrhoea (Vu-Khac et al., 2006). In Canada, the eae gene was detected in 3 % and 15 % of E. coli samples taken from diarrhoeic and non-diarrhoeic piglets, respectively (Ngeleka et al., 2003). In our study, we found that 12 % and 15 % of isolates from diarrhoeic piglets carried the eae gene. In 30 % of the isolates from diarrhoeic piglets, the eae gene was present alone. This confirms the observations made by other authors suggesting that E. coli inducing attaching and effacing lesions is isolated from diarrhoeic piglets (Malik et al., 2006; Vu- Khac et al., 2006). We also found it associated with EAST1 (3 %), F41 (3 %), F17 and F6 (2 %, each one). The gene that encodes F4 adhesin is one of the most frequently found genes in E. coli isolates from suckling and weaning piglets from different countries (Frydendahl, 2002; Vu-Khac et al., 2006; Zhang et al., 2007). Contrary to these results, we found that only 1 and 2 % of isolates from piglets with diarrhoea contained this virotype. Castro & Toledo (2010) did not find F4 adhesin in the isolates in their study, indicating that F4 has a low prevalence in E. coli in Mexico. E. coli isolated from pigs in Mexico shows a different virulence gene profile from those reported in other countries. This supports the fact that the presence of virulence genes in E. coli varies with the geographical area and could be useful as a marker of pathogenicity for diagnostic purposes or to design vaccines suitable for porcine enteric infections in Mexico. This study was performed to determine the prevalence of virulence genes associated with the suckling and weaning 154 Journal of Medical Microbiology 61

8 Porcine E. coli pathotypes in Mexico periods in piglets; we found a higher prevalence of genes in the weaning period than in the suckling stage. The multiplex PCR used in this study for the detection of adhesin and toxin genes was fast and sensitive for identifying virulence markers of E. coli strains in piglets. The results show preliminary epidemiological information about the presence of known virulence factors in Mexico s piglets and constitutes an important database for the implementation of prevention, diagnosis and treatment measures. ACKNOWLEDGEMENTS This investigation was partially supported by Consejo Nacional de Ciencia y Tecnología (Grant 89578), as well as by Facultad de Medicina, Universidad Nacional Autónoma de México. The authors thank Dr José Luis Pérez García for his help in the correct usage of English and Dra. Isabel Cristina Morán for her help with statistical analysis. REFERENCES Al-Gallas, N., Abbassi, S. M., Hassan, A. B. & Aissa, R. B. (2007). Genotypic and phenotypic profiles of enterotoxigenic Escherichia coli associated with acute diarrhea in Tunis, Tunisia. Curr Microbiol 55, Barman, N. N., Deb, R., Ramamurthy, T., Sharma, R. K., Borah, P., Wani, S. A. & Kalita, D. (2008). Molecular characterization of shiga like toxin-producing Escherichia coli (STEC) isolates from pigs oedema. Indian J Med Res 127, Bertschinger, H. U., Nief, V. & Tschäpe, H. (2000). Active oral immunization of suckling piglets to prevent colonization after weaning by enterotoxigenic Escherichia coli with fimbriae F18. Vet Microbiol 71, Blanco, M., Blanco, J. E., Gonzalez, E. A., Mora, A., Jansen, W., Gomes, T. A., Zerbini, L. F., Yano, T., de Castro, A. F. & Blanco, J. (1997). Genes coding for enterotoxins and verotoxins in porcine Escherichia coli strains belonging to different O : K : H serotypes: relationship with toxic phenotypes. J Clin Microbiol 35, Blanco, M., Blanco, J. E., Mora, A., Dahbi, G., Alonso, M. P., González, E. A., Bernárdez, M. I. & Blanco, J. (2004). Serotypes, virulence genes, and intimin types of Shiga toxin (verotoxin)-producing Escherichia coli isolates from cattle in Spain and identification of a new intimin variant gene (eae-j). J Clin Microbiol 42, Castro, A. M. & Toledo, A. A. (2010). Identification of virulence genes of Escherichia coli isolated from piglets with diarrhoea. Primer Congreso de Bioquímica y Biología Molecular de Bacterias, 22 25, pp. 78. San Miguel Regla, Mexico. Cheng, D., Sun, H., Xu, J. & Gao, S. (2006). PCR detection of virulence factor genes in Escherichia coli isolates from weaned piglets with edema disease and/or diarrhea in China. Vet Microbiol 115, Choi, C., Cho, W. S., Chung, H. K., Jung, T., Kim, J. & Chae, C. (2001). Prevalence of the enteroaggregative Escherichia coli heat-stable enterotoxin 1 (EAST1) gene in isolates in weaned pigs with diarrhea and/or edema disease. Vet Microbiol 81, Do, T. N., Cu, P. H., Nguyen, H. X., Au, T. X., Vu, Q. N., Driesen, S. J., Townsend, K. M., Chin, J. J. & Trott, D. J. (2006). Pathotypes and serogroups of enterotoxigenic Escherichia coli isolated from preweaning pigs in north Vietnam. J Med Microbiol 55, Fairbrother, J. M., Nadeau, E. & Gyles, C. L. (2005). Escherichia coli in postweaning diarrhea in pigs: an update on bacterial types, pathogenesis, and prevention strategies. Anim Health Res Rev 6, Francis, D. H. (2002). Enterotoxigenic Escherichia coli infection in pigs and its diagnosis. J Swine Health Prod 10, Frydendahl, K. (2002). Prevalence of serogroups and virulence genes in Escherichia coli associated with postweaning diarrhoea and edema disease in pigs and a comparison of diagnostic approaches. Vet Microbiol 85, Harel, J., Lapointe, H., Fallara, A., Lortie, L. A., Bigras-Poulin, M., Larivière, S. & Fairbrother, J. M. (1991). Detection of genes for fimbrial antigens and enterotoxins associated with Escherichia coli serogroups isolated from pigs with diarrhea. J Clin Microbiol 29, Holoda, E., Vu-Khac, H., Andrasková, S., Chomová, Z., Wantrubová, A., Krajnák, M. & Pilipcinec, E. (2005). PCR assay for detection and differentiation of K88ab(1), K88ab(2), K88ac, and K88ad fimbrial adhesins in E. coli strains isolated from diarrheic piglets. Folia Microbiol (Praha) 50, Imberechts, H., De Greve, H. & Lintermans, P. (1992). The pathogenesis of edema disease in pigs. A review. Vet Microbiol 31, Imberechts, H., Bertschinger, H. U., Nagy, B., Deprez, P. & Pohl, P. (1997). Fimbrial colonisation factors F18ab and F18ac of Escherichia coli isolated from pigs with postweaning diarrhea and edema disease. Adv Exp Med Biol 412, Kim, Y. J., Kim, J. H., Hur, J. & Lee, J. H. (2010). Isolation of Escherichia coli from piglets in South Korea with diarrhea and characteristics of the virulence genes. Can J Vet Res 74, Kwon, D., Kim, O. & Chae, C. (1999). Prevalence of genotypes for fimbriae and enterotoxins and of O serogroups in Escherichia coli isolated from diarrheic piglets in Korea. J Vet Diagn Invest 11, Kwon, D., Choi, C., Jung, T., Chung, H. K., Kim, J. P., Bae, S. S., Cho, W. S., Kim, J. & Chae, C. (2002). Genotypic prevalence of the fimbrial adhesins (F4, F5, F6, F41 and F18) and toxins (LT, STa, STb and STx2e) in Escherichia coli isolated from postweaning pigs with diarrhoea or oedema disease in Korea. Vet Rec 150, Lee, S. I., Kang, S. G., Kang, M. L. & Yoo, H. S. (2008). Development of multiplex polymerase chain reaction assays for detecting enterotoxigenic Escherichia coli and their application to field isolates from piglets with diarrhea. J Vet Diagn Invest 20, Leman, A. D., Straw, B. E., D Allaire, S., Mengeling, W. L. & Taylor, D. J. (1999). Diseases of Swine, pp Ames: Iowa State University Press. MacFaddin, J. F. (1980). Biochemical Tests for Identification of Medical Bacteria, 2nd edn. Baltimore, London: Williams & Wilkins. Madoroba, E., Van Driessche, E., De Greve, H., Mast, J., Ncube, I., Read, J. & Beeckmans, S. (2009). Prevalence of enterotoxigenic Escherichia coli virulence genes from scouring piglets in Zimbabwe. Trop Anim Health Prod 41, Malik, A., Tóth, I., Beutin, L., Schmidt, H., Taminiau, B., Dow, M. A., Morabito, S., Oswald, E., Mainil, J. & Nagy, B. (2006). Serotypes and intimin types of intestinal and faecal strains of eae+ Escherichia coli from weaned pigs. Vet Microbiol 114, Ngeleka, M., Pritchard, J., Appleyard, G., Middleton, D. M. & Fairbrother, J. M. (2003). Isolation and association of Escherichia coli AIDA-I/STb, rather than EAST1 pathotype, with diarrhea in piglets and antibiotic sensitivity of isolates. J Vet Diagn Invest 15, Ojeniyi, B., Ahrens, P. & Meyling, A. (1994). Detection of fimbrial and toxin genes in Escherichia coli and their prevalence in piglets with 155

9 A. Toledo and others diarrhoea. The application of colony hybridization assay, polymerase chain reaction and phenotypic assays. J Vet Med 41, Osek, J. (2003). Detection of the enteroaggregative Escherichia coli heat-stable enterotoxin 1 (EAST1) gene and its relationship with fimbrial and enterotoxin markers in E. coli isolates from pigs with diarrhoea. Vet Microbiol 91, Osek, J., Gallien, P., Truszczyñski, M. & Protz, D. (1999). The use of polymerase chain reaction for determination of virulence factors of Escherichia coli strains isolated from pigs in Poland. Comp Immunol Microbiol Infect Dis 22, Parma, A. E., Sanz, M. E., Viñas, M. R., Cicuta, M. E., Blanco, J. E., Boehringer, S. I., Vena, M. M., Roibon, W. R., Benitez, M. C. & other authors (2000). Toxigenic Escherichia coli isolated from pigs in Argentina. Vet Microbiol 72, Rippinger, P., Bertschinger, H. U., Imberechts, H., Nagy, B., Sorg, I., Stamm, M., Wild, P. & Wittig, W. (1995). Designations F18ab and F18ac for the related fimbrial types F107, 2134P and 8813 of Escherichia coli isolated from porcine postweaning diarrhoea and from oedema disease. Vet Microbiol 45, SAGARPA (2010). Paginas/default.aspx. Sting, R. & Stermann, M. (2008). Duplex real-time PCR assays for rapid detection of virulence genes in E. coli isolated from post-weaning pigs and calves with diarrhoea. Dtsch Tierarztl Wochenschr 115, Vu-Khac, H., Holoda, E. & Pilipcinec, E. (2004). Distribution of virulence genes in E. coli strains isolated from diarrhoeic piglets in the Slovak Republic. J Vet Med B 51, Vu-Khac, H., Holoda, E., Pilipcinec, E., Blanco, M., Blanco, J. E., Mora, A., Dahbi, G., López, C., González, E. A. & Blanco, J. (2006). Serotypes, virulence genes, and PFGE profiles of Escherichia coli isolated from pigs with postweaning diarrhoea in Slovakia. BMC Vet Res 2, 10. Vu-Khac, H., Holoda, E., Pilipcinec, E., Blanco, M., Blanco, J. E., Dahbi,G.,Mora,A.,López, C., González, E. A. & Blanco, J. (2007). Serotypes, virulence genes, intimin types and PFGE profiles of Escherichia coli isolated from piglets with diarrhoea in Slovakia. Vet J 174, Wilson, R. A. & Francis, D. H. (1986). Fimbriae and enterotoxins associated with Escherichia coli serogroups isolated from pigs with colibacillosis. Am J Vet Res 47, Yamamoto, T. & Nakazawa, M. (1997). Detection and sequences of the enteroaggregative Escherichia coli heat-stable enterotoxin 1 gene in enterotoxigenic E. coli strains isolated from piglets and calves with diarrhea. J Clin Microbiol 35, Zhang, W., Zhao, M., Ruesch, L., Omot, A. & Francis, D. (2007). Prevalence of virulence genes in Escherichia coli strains recently isolated from young pigs with diarrhea in the US. Vet Microbiol 123, Journal of Medical Microbiology 61