Abstract
Enterohemorrhagic Escherichia coli (EHEC) strains comprise a subgroup of Shiga-toxin (Stx)-producing E. coli (STEC) and are characterized by a few serotypes. Among these, seven priority STEC serotypes (O26:H11, O45:H2, O103:H2, O111:H8, O121:H19, O145:H28, and O157:H7) are most frequently implicated in severe clinical illness worldwide. Currently, standard methods using stx, eae, and O-serogroup-specific gene sequences for detecting the top 7 EHEC serotypes bear the disadvantage that these genes can be found in non-EHEC strains as well. Here, we explored the suitability of ureD, espV, espK, espN, Z2098, and espM1 genes and combinations thereof as candidates for a more targeted EHEC screening assay. For a very large panel of E. coli strains (n = 1,100), which comprised EHEC (n = 340), enteropathogenic E. coli (EPEC) (n = 392), STEC (n = 193), and apathogenic strains (n = 175), we showed that these genetic markers were more prevalent in EHEC (67.1% to 92.4%) than in EPEC (13.3% to 45.2%), STEC (0.5% to 3.6%), and apathogenic E. coli strains (0 to 2.9%). It is noteworthy that 38.5% of the EPEC strains that tested positive for at least one of these genetic markers belonged to the top 7 EHEC serotypes, suggesting that such isolates might be Stx-negative derivatives of EHEC. The associations of espK with either espV, ureD, or Z2098 were the best combinations for more specific and sensitive detection of the top 7 EHEC strains, allowing detection of 99.3% to 100% of these strains. In addition, detection of 93.7% of the EHEC strains belonging to other serotypes than the top 7 offers a possibility for identifying new emerging EHEC strains.
INTRODUCTION
Shiga toxin-producing E. coli (STEC) may cause food-borne infections leading to life-threatening diseases in humans. The outcomes of STEC infections may range from asymptomatic carriage to uncomplicated diarrhea to the severe symptoms of hemorrhagic colitis (HC) and hemolytic-uremic syndrome (HUS). Owing to their human pathogenicity, some STEC strains are also designated as enterohemorrhagic E. coli (EHEC) (1, 2). EHEC strains comprise a subgroup of STEC and are characterized by certain serotypes, which are frequently associated with outbreaks and severe clinical illness (3, 4). Since the early 1980s, numerous cases of HC and HUS were attributed to EHEC O157:H7 (CDC report of microbiological results of raw ground beef products analyzed for E. coli O157:H7, http://www.fsis.usda.gov/wps/portal/fsis/topics/data-collection-and-reports/microbiology/ec/summary-data/summary-data-2011/ct_index). Consequently, public health and regulatory responses have been focused largely on this serotype. In recent years, cumulative evidence from numerous countries has indicated that up to 30 to 60% of human EHEC infections are caused by non-O157 EHEC (5, 6; U.S. Department of Agriculture [USDA] risk profile for pathogenic non-O157 Shiga-toxin producing E. coli, http://www.fsis.usda.gov/wps/wcm/connect/92de038d-c30e-4037-85a6-065c3a709435/Non_O157_STEC_Risk_Profile_May2012.pdf?MOD=AJPERES). Recent studies have shown that the number of non-O157 STEC infections sometimes surpasses the number of STEC O157 infections (7, 8). Accordingly, the list of STEC associated with HUS has been extended to include serotypes O26:H11, O45:H2, O103:H2, O111:H8, O121:H19, O145:H28, and O157:H7 and their non-motile derivatives. These are the seven priority STEC serotypes most frequently implicated in outbreaks and sporadic cases of HC and HUS worldwide (9; USDA report on detection and identification of non-O157 STEC from meat products, http://www.fsis.usda.gov/wps/portal/fsis/topics/science/laboratories-and-procedures/guidebooks-and-methods/microbiology-laboratory-guidebook/microbiology-laboratory-guidebook), which are referred to as the top 7 EHEC serogroups. In the United States, regulatory testing for these STEC serogroups in meat started in June 2012. Although regulations are disparate throughout the world, many food inspection programs aim at detecting STEC strains that pose a significant threat to human health in kinds of foods that are the most likely to disseminate EHEC and to be consumed raw or undercooked. Some beef products are thus of particular interest in that aspect. The U.S. regulations consider these 7 EHEC serogroups to be adulterants in beef trim (USDA report on detection and identification of non-O157 STEC from meat products). On March 2013, a new regulation on sprouts and seeds was published in Europe that introduced for the first time in European Union legislation a microbiologic criterion for certain STEC serogroups (namely O157, O26, O103, O111, O145 and O104:H4) recognized to be those causing most cases of HC and HUS occurring in the European Union. Therefore, microbiological criteria should be considered for these six serogroups. The European Union legislation also stipulates that the possibility cannot be excluded that other STEC serogroups may be pathogenic to humans as well. In fact, such STEC may cause less severe forms of disease, such as diarrhea and/or bloody diarrhea, and may also cause HUS and therefore represent a hazard for the consumer's health.
Despite the publication of this legislation, detection of non-O157 EHEC remains particularly challenging because they have no phenotypical characteristics that distinguish them from the large number of non-STEC that share the same habitats. The basis of STEC virulence is not fully understood, but a greater risk of HC and HUS is associated with STEC serotypes carrying additional virulence factors to Stx, such as the intimin (encoded by the eae gene of the locus of enterocyte effacement [LEE]), which is directly involved in the attaching and effacing (A/E) process of the enterocyte microvilli and in bacterial colonization of the gut, thus contributing to the disease process (2). The LEE also encodes regulatory elements, a type III secretion system (TTSS), secreted effector proteins, and their cognate chaperon (10, 11). However, the LEE is not highly specific for EHEC strains and is found in Stx-negative E. coli strains as well. LEE is a hallmark of enteropathogenic E. coli (EPEC), which may cause diarrhea but not HC and HUS. Hence, current methods based on PCR detection of stx and eae genes for screening non-O157 EHEC in complex samples bear the disadvantage that a large number of non-EHEC strains are detected (12, 13, 14). Identification of genetic markers allowing a more targeted screening of EHEC strains is needed for rapidly testing food, fecal, and environmental samples.
It is likely that a variable repertoire of virulence determinants, including non-LEE-encoded effector (nle) genes that encode translocated substrates of the type III secretion system are present in highly pathogenic STEC and are carried by genetic mobile elements (15, 16, 17, 18). A molecular risk assessment approach based on the evaluation of the nle gene content has been used to predict which STEC strains pose a significant risk to human health (4, 16, 17, 18). However, the genetic targets which best support such an approach have not yet been defined. Monitoring EHEC in foods requires, in particular, selection of genetic markers that clearly discriminate EHEC from EPEC strains.
In an attempt to identify such factors, we explored the suitability of certain nle genes derived from the genomic O islands OI-43, OI-44, OI-50, OI-57, and OI-71 as candidates to distinguish STEC strains that constitute a severe risk for human health from EPEC and STEC strains that are not associated with severe and epidemic disease. We focused on ureD (urease activity) encoded by OI-43 and OI-48, espN (EspN), and espK (EspK), two genes carried by OI-50, a locus involved in the persistence of EHEC O157:H7 in the intestines of orally inoculated calves (19). Also, we focused on espM1 (EspM1), which is derived from OI-71, and on Z2098, a sequence derived from OI-57, a genomic island that may be associated with increased virulence of STEC strains in humans (14, 20). Genome sequencing of EHEC strains (EHEC O157:H7, O111, O103, and O26) has also revealed other genetic markers, such as espV, whose role in disease has not been evaluated. This gene is located on OI-44 of EHEC O157:H7 but its prevalence in other E. coli pathogroups has not been documented. In this study, we evaluated the distribution of ureD, espV, espK, espN, Z2098, and espM1 in various E. coli pathogroups to assess their association with STEC strains with high virulence for humans and to test their suitability for clearly distinguishing EHEC from other E. coli pathogroups.
MATERIALS AND METHODS
Bacterial strains.
The origin and properties of the E. coli strains (n = 1,100) used in this study were previously described (13, 14, 21). The EHEC type strains (n = 340) were defined by the presence of both stx and eae genes. STEC strains (n = 193) harbor stx only. EPEC strains (n = 392) harbor eae only. Apathogenic E. coli (n = 175) were defined as stx- and eae-negative strains. Cultivation of bacteria and preparation of DNA were performed as previously described (13, 14, 21).
High-throughput real-time PCR.
E. coli gene targets used for the real-time PCR amplification are reported in Table 1. The genes stx1, stx2, and eae were used as internal controls and for group assignment purposes. Primers and probes for the detection of stx1, stx2, eae, espK, and Z2098 were described previously (14, 22), whereas those for the detection of ureD, espV, espN, and espM1 were designed for this work. Primers and probes are listed in Table 2. A LightCycler 1536 (Roche, Meylan, France) was used to perform high-throughput real-time PCR amplifications as described previously (13, 14, 21).
Table 1.
E. coli gene targets for the real-time PCR array
| Gene (ORF name if chromosomal)a | Encoded protein or family effector | Genetic support (mobile element)a |
|---|---|---|
| ureD (Z1142) | Urease-associated protein UreD | OI-43 and OI-48 |
| espV (Z1387) | AvrA family effector | OI-44 |
| espK (Z1829) | Leucine-rich repeats | OI-50 (prophage CP-933N) |
| espN (Z1824) | Hypothetical protein | OI-50 (prophage CP-933N) |
| (Z2098) | Hypothetical protein | OI-57 |
| espM1 (Z2565) | Non-LEE-encoded type III effector | OI-71 |
Nomenclature of ORFs and mobile elements refers to the sequence of E. coli O157:H7 EDL933 (GenBank accession no. AE005174).
Table 2.
Primers and probes for the real-time PCR array
| Target gene (ORF name if chromosomal)a | Primer |
Probe sequence (5′ to 3′)b | Location within GenBank sequence AE005174 (nt range) | |
|---|---|---|---|---|
| Forward | Reverse | |||
| ureD (Z1142) | GCAATAATTGACTCTGATTGCC | 1078824–1078845 | ||
| GCTGCTGCGGTAAAATTTACT | 1078892–1078872 | |||
| [FAM]-TACGCTGATCACCATGCCTGGTGC-[BHQ1] | 1078847–1078870 | |||
| espV (Z1387) | TCAGGTTCCTCGTCTGATGCCGC | 1295446–1295424 | ||
| CTGGTTCAGGCCTGGAGCAGTCC | 1295360–1295382 | |||
| [FAM]-CTTGCAACACGTTACGCTGCCGAGTATT-[BHQ1] | 1295422–1295395 | |||
| espK (Z1829) | GCAGRCATCAAAAGCGAAATCACACC | 1673422–1673397 | ||
| TCGTTTGGTAACTGTGGCAGATACTC | 1673312–1673338 | |||
| [FAM]-ATTCAGATAGAAGAAGCGCGGGCCAG-[BHQ1] | 1673395–16673370 | |||
| espN (Z1824) | GACATATTTGTTTATGTCATCAGGAGCGG | 1666890–1666918 | ||
| CCTCAGGATATGGATGGCCTACTGGC | 1667017–1666992 | |||
| [FAM]-AATGCTCTCGGCAATCGAATCCTTGACTC-[BHQ1] | 1666991–1666963 | |||
| Z2098 | CTGAAAAGAGCCAGAACGTGC | 1888173–1888193 | ||
| TGCCTAAGATCATTACCCGGAC | 1888308–1888287 | |||
| [HEX]-TAACTGCTATACCTCCGCGCCG-[BHQ1] | 1888286–1888265 | |||
| espM1 (Z2565) | GCGCTCTATCCGCTTTAATGTTAAC | 2275378–2275354 | ||
| CCATCCATGAATATCTTTAGTACTCTGC | 2275291–2275318 | |||
| [FAM]-TGCTTACCGTCTCCAGTATACAGCCGCT-[BHQ1] | 2275320–2275347 | |||
Numbering as in E. coli O157:H7 EDL933 (GenBank accession no. AE005174).
FAM, 6-carboxyfluorescein; BHQ1, black hole quencher 1; HEX, 5′-hexachlorofluorescein.
RESULTS
Distribution of ureD, espV, espK, espN, Z2098, and espM1 and combinations thereof among E. coli pathogroups.
Distribution of the genetic markers ureD, espV, espK, espN, Z2098, and espM1 among the different E. coli pathogroups is shown in Table 3. Overall, the genetic markers investigated were mostly detected in EHEC strains with frequencies ranging from 67.1% (espM1) to 92.4% (espK). These markers were less commonly associated with EPEC strains with frequencies ranging from 13.3% (espN) to 45.2% (espV) and rarely detected in STEC (0.5 to 3.6%) and nonpathogenic E. coli (0 to 2.9%). Overall, we observed that 38.5% of the EPEC strains which tested positive for at least one of the investigated genetic markers belonged to the top 7 EHEC serotypes (Table 4). Interestingly, among the EPEC strains positive for at least one of the investigated genetic markers were EPEC cluster 1 strains (45.25%) that were previously reported to be frequently involved in severe illness and outbreaks (22). These included in particular EPEC serotypes that have already been associated with stx production, such as O100:H25, O103:H25, O111:H2, O119:H25, O126:H27, O128:H2, O55:H7, O70:H11, O76:H7, O118:H8, O119:H8, O156:H8, O2:H40, O3:H8, O80:H2, O86:H11, O66:H8, O111:H25, O114:H2, and O119:H2 (data not shown).
Table 3.
Distribution of the genetic markers ureD, espV, espK, espN, Z2098, and espM1 and combinations thereof among E. coli pathogroups
| Genetic markers and combinations thereofa | Distribution (%) within each pathogroupb |
|||
|---|---|---|---|---|
| EHEC | EPEC | STEC | EC | |
| Z2098 | 87.4 | 23.2 | 3.6 | 1.1 |
| espK | 92.4 | 28.8 | 0.5 | 1.1 |
| espN | 86.5 | 13.3 | 0.5 | 0.0 |
| espV | 84.4 | 45.2 | 1.6 | 0.6 |
| ureD | 89.4 | 18.1 | 3.1 | 2.9 |
| espM1 | 67.1 | 34.4 | 0.5 | 1.7 |
| espK/espV | 98.5 | 54.1 | 1.6 | 1.1 |
| espK/ureD | 98.8 | 33.4 | 3.6 | 3.4 |
| espK/Z2098 | 97.9 | 36.7 | 3.6 | 2.3 |
| espN/Z2098 | 96.5 | 26.5 | 3.6 | 1.1 |
| espN/espM1 | 93.8 | 38.3 | 1.0 | 1.7 |
espK/espV, strains giving a positive result for espK and/or espV; espK/ureD, strains giving a positive result for espK and/or ureD; espK/Z2098, strains giving a positive result for Z2098 and/or espK; espN/Z2098, strains giving a positive result for espN and/or Z2098; espN/espM1, strains giving a positive result for espN and/or espM1.
E. coli pathogroups are defined in Materials and Methods.
Table 4.
Distribution of the genetic markers ureD, espV, espK, espN, Z2098, and espM1 and combinations thereof among EPEC serotype strains
| Genetic markera | No. of strains detected/total no. of strains (%) for: |
|||||
|---|---|---|---|---|---|---|
| O103:H2 | O121:H19 | O145:H28 | O157:H7 | O26:H11 | Other EPECb | |
| Z2098 | 7/11 (63.6) | 0/5 (0.0) | 8/8 (100) | 9/9 (100) | 31/71 (43.7) | 36/288 (12.5) |
| espK | 9/11 (81.8) | 3/5 (60.0) | 8/8 (100) | 7/9 (77.8) | 30/71 (42.3) | 56/288 (19.4) |
| espN | 6/11 (54.5) | 0/5 (0.0) | 8/8 (100) | 6/9 (66.7) | 13/71 (18.3) | 19/288 (6.6) |
| espV | 6/11 (54.5) | 5/5 (100) | 8/8 (100) | 9/9 (100) | 31/71 (43.7) | 118/288 (41.0) |
| ureD | 1/11 (9.1) | 5/5 (100) | 5/8 (62.5) | 6/9 (66.7) | 20/71 (28.2) | 34/288 (11.8) |
| espM1 | 0/11 (0.0) | 5/5 (100) | 8/8 (100) | 9/9 (100) | 64/71 (90.1) | 49/288 (17.0) |
| espK/espV | 10/11 (90.9) | 5/5 (100) | 8/8 (100) | 9/9 (100) | 43/71 (60.6) | 137/288 (47.6) |
| espK/ureD | 9/11 (81.8) | 5/5 (100) | 8/8 (100) | 9/9 (100) | 30/71 (42.3) | 70/288 (24.3) |
| espK/Z2098 | 10/11 (90.9) | 3/5 (60) | 8/8 (100) | 9/9 (100) | 33/71 (46.5) | 81/288 (28.1) |
| espN/Z2098 | 7/11 (63.6) | 0/5 (0) | 8/8 (100) | 9/9 (100) | 31/71 (43.7) | 49/288 (17.0) |
| espN/espM1 | 6/11 (54.5) | 5/5 (100) | 8/8 (100) | 9/9 (100) | 65/71 (91.5) | 57/288 (19.8) |
espK/espV, strains giving a positive result for espK and/or espV; espK/ureD, strains giving a positive result for espK and/or ureD; espK/Z2098, strains giving a positive result for Z2098 and/or espK; espN/Z2098, strains giving a positive result for espN and/or Z2098; espN/espM1, strains giving a positive result for espN and/or espM1.
Most of the EPEC strains tested in this study were described by Bugarel et al. (25).
None of the genetic markers ureD, espV, espK, espN, Z2098, or espM1 is, by itself, suitable for reliable identification of all EHEC strains. Combinations of the genetic markers were explored to identify those which detect EHEC with best specificity. The results are presented in Table 3. The combined presence of these genetic markers was highly associated with EHEC, with frequencies ranging from 93.8% (espN/espM1) to 98.8% (espK/ureD). The same combinations detected EPEC strains with frequencies ranging from 26.5% (espN/Z2098) to 54% (espK/espV), STEC strains with frequencies of 1% to 3.6%, and nonpathogenic E. coli strains with frequencies between 1.1% and 3.4%.
Distribution of ureD, espV, espK, espN, Z2098, and espM1 and combinations thereof among EHEC serotypes.
The distribution of each genetic marker, ureD, espV, espK, espN, Z2098, or espM1, was different according to EHEC serotypes. Distribution of each genetic marker in various EHEC serotypes is reported in Table 5. Except for espV, which was not detected in any EHEC O45:H2, and espM1, which was absent from EHEC O45:H2 and from 48 out of the 49 O103:H2 strains tested, all the other genetic markers investigated were found to be highly prevalent in EHEC strains of the top 7 serotypes, with frequencies ranging from 71.4% (prevalence of ureD in O103:H2) to 100%.
Table 5.
Distribution of the genetic markers ureD, espV, espK, espN, Z2098, and espM1 among EHEC serotypes
| Genetic marker | No. of strains detected/total no. of strains (%) for: |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Top 7 serotypes | O103:H2 | O111:H8 | O121:H19 | O145:H28 | O157:H7 | O26:H11 | O45:H2 | Other EHECa (new emerging EHEC) | Total EHEC | |
| Z2098 | 250/277 (90.3) | 49/49 (100) | 47/51 (92.2) | 17/20 (85.0) | 30/30 (100) | 49/66 (74.2) | 44/44 (100) | 14/17 (82.4) | 47/63 (74.6) | 297/340 (87.4) |
| espK | 269/277 (97.1) | 48/49 (98.0) | 51/51 (100) | 19/20 (95.0) | 29/30 (96.7) | 62/66 (93.9) | 43/44 (97.7) | 17/17 (100) | 45/63 (71.4) | 314/340 (92.4) |
| espN | 261/277 (94.2) | 47/49 (95.9) | 51/51 (100) | 17/20 (85.0) | 30/30 (100) | 59/66 (89.4) | 40/44 (90.9) | 17/17 (100) | 33/63 (52.4) | 294/340 (86.5) |
| espV | 248/277 (89.5) | 48/49 (98.0) | 51/51 (100) | 20/20 (100) | 30/30 (100) | 65/66 (98.5) | 34/44 (77.3) | 0/17 (0) | 39/63 (61.9) | 287/340 (84.4) |
| ureD | 257/277 (92.8) | 35/49 (71.4) | 51/51 (100) | 16/20 (80.0) | 30/30 (100) | 64/66 (97.0) | 44/44 (100) | 17/17 (100) | 47/63 (74.6) | 304/340 (89.4) |
| espM1 | 206/277 (74.4) | 1/49 (2.0) | 51/51 (100) | 20/20 (100) | 30/30 (100) | 64/66 (97.0) | 40/44 (90.9) | 0/17 (0) | 22/63 (34.9) | 228/340 (67.1) |
O103:H25 (n = 2), O118:H16 (n = 4), O118:H2, O119:H25 (n = 5), O123:H11, O127:H8s, O145, O145:H25 (n = 5), O156:H21, O156:H25 (n = 11), O165:H25 (n = 2), O172:H25 (n = 2), O172:NM, O177 (n = 2), O177:H25, O182:H25, O3, O49:H16, O5 (n = 11), O55:H7 (n = 2), O76:H51, O84:H2, Ont:H2, Ont:H25 (n = 2), Or:H16, OX186:H2.
Detection of the top 7 EHEC serotypes based on different combinations of these genetic markers is reported in Table 6. With the combined detection of espN with espM1, all EHEC O157:H7, O26:H11, O111:H8, O121:H19, O45:H2, and O145:H28 strains gave a positive result for espN and/or espM1, and 95.9% of O103:H2 strains tested positive. Likewise, detection of espK and/or Z2098 allowed detection of most of the EHEC serotypes associated with human infections. Thus, all EHEC O111:H8, O26:H11, O45:H2, O103:H2, and O145:H28 strains, 98.5% of O157:H7 strains, and 95% of O121:H19 strains gave a positive result for espK and/or Z2098. With the combined detection of espN with Z2098 all EHEC O26:H11, O103:H2, O145:H28, O45:H2, and O111:H8 strains, 98.5% of O157:H7 strains, and 85% (17/20) of O121:H19 strains tested positive for espN and/or Z2098. The association of espK with either espV or ureD allowed detection of most of the strains of the top 7 EHEC serotypes as well. Hence, all strains of serotypes O157:H7, O145:H28, O111:H8, O103:H2, O45:H2, and O121:H19 gave positive results for espK and/or espV, and 97.7% of O26:H11 strains gave positive result for espK and/or espV. Data were very similar when we tested espK in association with ureD. In this case, all strains of the top 7 EHEC serotypes gave a positive result for espK and/or ureD (Tables 6).
Table 6.
Detection of EHEC according to different combinations of the genetic markers ureD, espV, espK, espN, Z2098, and espM1
| Gene associationa | No. of strains detected/total no. of strains (%) for: |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| EHEC of the top 7 | O103:H2 | O111:H8 | O121:H19 | O145:H28 | O157:H7 | O26:H11 | O45:H2 | Other EHECb (new emerging EHEC) | Total EHEC | |
| espK/espV | 276/277 (99.6) | 49/49 (100) | 51/51 (100) | 20/20 (100) | 30/30 (100) | 66/66 (100) | 43/44 (97.7) | 17/17 (100) | 59/63 (93.7) | 335/340 (98.5) |
| espK/ureD | 277/277 (100) | 49/49 (100) | 51/51 (100) | 20/20 (100) | 30/30 (100) | 66/66 (100) | 44/44 (100) | 17/17 (100) | 59/63 (93.7) | 336/340 (98.8) |
| espK/Z2098 | 275/277 (99.3) | 49/49 (100) | 51/51 (100) | 19/20 (95.0) | 30/30 (100) | 65/66 (98.5) | 44/44 (100) | 17/17 (100) | 59/63 (93.7) | 334/340 (98.2) |
| espN/Z2098 | 273/277 (98.6) | 49/49 (100) | 51/51 (100) | 17/20 (85.0) | 30/30 (100) | 65/66 (98.5) | 44/44 (100) | 17/17 (100) | 55/63 (87.3) | 328/340 (96.5) |
| espN/espM1 | 275/277 (99.3) | 47/49 (95.9) | 51/51 (100) | 20/20 (100) | 30/30 (100) | 66/66 (100) | 44/44 (100) | 17/17 (100) | 44/63 (69.8) | 319/340 (93.8) |
espK/espV, strains giving a positive result for espK and/or espV; espK/ureD, strains giving a positive result for espK and/or ureD; espK/Z2098, strains giving a positive result for Z2098 and/or espK; espN/Z2098, strains giving a positive result for espN and/or Z2098; espN/espM1, strains giving a positive result for espN and/or espM1.
O103:H25 (n = 2), O118:H16 (n = 4), O118:H2, O119:H25 (n = 5), O123:H11, O127:H8s, O145, O145:H25 (n = 5), O156:H21, O156:H25 (n = 11), O165:H25 (n = 2), O172:H25 (n = 2), O172:NM, O177 (n = 2), O177:H25, O182:H25, O3, O49:H16, O5 (n = 11), O55:H7 (n = 2), O76:H51, O84:H2, Ont:H2, Ont:H25 (n = 2), Or:H16, OX186:H2.
DISCUSSION
In many countries the emergence of O157 and non-O157 EHEC in severe and epidemic human disease is of great concern. Accordingly, rapid and specific detection of EHEC strains has become a priority for public health authorities (5, 6; Table 7 of the 2009 CDC Foodnet report of the food-borne diseases active surveillance network, http://www.cdc.gov/foodnet/factsandfigures/Top10SalmonellaSerotypes.pdf.). The availability of rapid and specific methods for testing EHEC strains and their virulence markers is also a prerequisite for establishing monitoring programs to follow EHEC contamination in animals and foodstuffs which are part of food inspection programs. Detection of non-O157 EHEC is particularly challenging, because unlike EHEC O157, these strains have few characteristics that distinguish them from the large number of harmless commensal E. coli strains that share the same niches.
The current approach for detecting EHEC in food and stool samples is to screen first for the presence of the stx1/stx2 genes and the eae gene. The CEN/ISO TS 13136 (9) and MLG 5B.01 (USDA report on detection and identification of non-O157 STEC from meat products) standard methods request the presence of both the stx1/stx2 and eae genes for further investigation of specific sequences derived from the O-antigen genes associated with the seven priority serogroups. This approach bears the disadvantage that a large number of non-EHEC strains can produce cross-reactivity with the targets used in these tests. This is of particular interest in complex and polymicrobial samples such as food, fecal, and environmental specimens (23, 24, 25). Accordingly, the simple detection of the stx1/stx2, eae, and O-antigen genes in a food product may often result in its rejection even though it does not contain EHEC strains of the top 7 serotypes. For food inspection purposes, tests are needed that not only identify the targeted EHEC serogroups but also target the most salient distinguishing features of the priority STEC associated with human illness.
In an attempt to identify discriminative genetic markers associated with STEC strains related to the world's most frequent clinical cases, we evaluated the distribution of the genes ureD, espV, espK, espN, Z2098, and espM1 in various E. coli pathogroups. Based on a high-throughput real-time PCR approach, a very large panel of E. coli strains (n = 1,100) that comprised EHEC (n = 340), EPEC (n = 392), STEC (n = 193), and apathogenic E. coli (n = 175) was examined for these genetic markers. Distributions of the genetic markers ureD, espV, espK, espN, Z2098, and espM1 were significantly different among the various E. coli pathogroups. These genetic markers were highly prevalent in EHEC (67.1% to 92.4%) but were individually insufficient to identify all EHEC isolates. Even though none of the genetic markers was, by itself, capable of reliably detecting all EHEC isolates, when used in association with each other they identified EHEC strains with high confidence. Hence, only a few EHEC strains did not react with the combinations of the genetic markers tested here. These might be aberrant strains, not representative for the classical EHEC types. Looking at other genes in these anecdotal strains or sequencing their genomes might reveal more differences, which may make things clearer regarding their status. We should assume, on principle, that it is not necessarily the case that all members of a particular serotype would be EHEC.
The genetic markers examined in this study were also detected in some EPEC strains (13.3% to 45.2%) and very rarely in STEC (0.5% to 3.6%) and apathogenic E. coli (0 to 2.9%) strains. It is noteworthy that 38.5% of the EPEC strains that tested positive for at least one of the genetic markers ureD, espV, espK, espN, Z2098, and espM1 belonged to the top 7 EHEC serotypes. Interestingly, other EPEC strains having a known EHEC serotype, such as O55:H7 and O103:H25, were also found positive for at least one of these genetic markers. Additionally, some of the EPEC strains that were found positive for at least one of these genetic markers were previously found to be closely related to EHEC strains according to their virulence gene content (22). These findings indicate that such isolates might be Stx–negative derivatives of EHEC that are also designated as EHEC-like strains (14). We assumed these isolates were EHEC derivatives according to their serotypes and nle gene content but they might also be EPEC strains that we are as yet unable to discriminate from EHEC derivatives. Further investigation using whole-genome sequencing may clarify the exact designation of these strains in the future.
The genetic markers ureD, espV, espK, espN, Z2098, and espM1 were detected at different frequencies among the EHEC serotypes. We explored the various associations of these genetic markers to search for the best combinations of markers giving higher specificity and sensitivity for detecting EHEC. The genetic markers espK, espV, ureD, and Z2098 were shown to be the best candidates as genetic markers for detecting EHEC. Taken individually, they were not detected in all strains of the top 7 EHEC serotypes, while in association they were detected in 98.6% to 100% of the top 7 EHEC strains. The association of espK with either espV, ureD, or Z2098 proved to be the best combinations for more specific and sensitive detection of EHEC strains. Hence, a positive result for espK and/or espV was observed in 99.6% of EHEC strains belonging to the seven major serotypes of EHEC reported worldwide in human infections (only one EHEC O26:H11 isolate tested negative). Also, 93.7% of EHEC strains with serotypes other than those of the top 7 serotypes tested positive for espK and/or espV. Only a subset (54.1%) of EPEC strains tested positive for espK and/or espV. Most STEC and avirulent E. coli strains were negative for both espK and espV. Another interesting approach was to associate espK with Z2098. This combination of genetic markers resulted in the detection of 99.3% of EHEC strains belonging to the 7 major EHEC serotypes and 93.7% of EHEC strains with serotypes other than the top 7. The presence of espK and/or Z2098 was found in only 36.7% of EPEC, 3.6% of STEC and 2.3% of apathogenic E. coli strains. The best approach for detecting EHEC with the highest specificity and sensitivity was to combine espK with ureD. This association allowed detection of 100% of EHEC strains of the top 7 serotypes and 93.7% of EHEC strains belonging to other serotypes. Detection of espK and/or ureD was also reported for only 33.4% of EPEC, 3.6% of STEC, and 3.4% of apathogenic E. coli strains.
These findings showed that combining detection of espK with either espV, ureD, or Z2098 is a highly sensitive and specific approach for identifying with ≥99% confidence EHEC serotypes related to the world's most frequent clinical cases. Detection of these genetic markers in combination with stx in complex samples (food or fecal specimens) would provide a more EHEC-targeted diagnostic approach than combining only stx and eae. Given the rapidity of these PCR assays, this approach should have a major impact on top-7 EHEC surveillance and outbreak investigations and is likely to be of benefit to public health. Moreover, detection of these sets of genetic markers in 93.7% of EHEC strains having serotypes other than the top 7 may be helpful for identifying new emerging EHEC strains. The number of strains and the diversity of serotypes and pathogroups that were investigated in this study provide a solid basis for future utilization of these tests for the development of analytical methods and risk characterization of STEC. This should be confirmed with further evaluation of these assays on spiked and naturally contaminated samples. A complete evaluation of these tests in real samples will be crucial to definitively know how they can be applied in EHEC surveillance and outbreak investigations.
ACKNOWLEDGMENTS
We are grateful to F. Scheutz (WHO Collaborating Centre for Reference and Research on Escherichia and Klebsiella, Copenhagen, Denmark), A. Gill (Health Canada, Ottawa, Canada), M. Rivas (INEI-ANLIS, Buenos Aires, Argentina), G. H. Loneragan (Texas Tech University, Lubbock, TX), C. DebRoy (Pennsylvania State University, State College, PA), Roxane M. F. Piazza (Instituto Butantan, Sao Paulo, Brazil), P. Feng (U.S. FDA, College Park, MD), and N. Strockbine (CDC, Atlanta, GA) for providing some E. coli isolates or DNA extracts from E. coli.
Footnotes
Published ahead of print 24 July 2013
REFERENCES
- 1.Levine MM. 1987. Escherichia coli that cause diarrhea: enterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. J. Infect. Dis. 155:377–389 [DOI] [PubMed] [Google Scholar]
- 2.Nataro JP, Kaper JB. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142–201 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Karmali MA, Mascarenhas M, Shen S, Ziebell K, Johnson S, Reid-Smith R, Isaac-Renton J, Clark C, Rahn K, Kaper JB. 2003. Association of genomic O island 122 of Escherichia coli EDL 933 with verocytotoxin-producing Escherichia coli seropathotypes that are linked to epidemic and/or serious disease. J. Clin. Microbiol. 41:4930–4940 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bugarel M, Beutin L, Martin A, Gill A, Fach P. 2010. Micro-array for the identification of Shiga toxin-producing Escherichia coli (STEC) seropathotypes associated with hemorrhagic colitis and hemolytic uremic syndrome in humans. Int. J. Food Microbiol. 142:318–329 [DOI] [PubMed] [Google Scholar]
- 5.European Food Safety Authority (EFSA) 2013. Scientific opinion on VTEC-seropathotype and scientific criteria regarding pathogenicity assessment. EFSA J. 11:3138 http://www.efsa.europa.eu/en/efsajournal/pub/3138.htm [Google Scholar]
- 6.European Food Safety Authority (EFSA) 2007. Scientific opinion of the panel on biological hazards on a request from EFSA on monitoring of verotoxigenic Escherichia coli (VTEC) and identification of human pathogenic types. EFSA J. 579:1–61 [Google Scholar]
- 7.Fey PD, Wickert RS, Rupp ME, Safranek TJ, Hinrichs SH. 2000. Prevalence of non-O157:H7 shiga toxin-producing Escherichia coli in diarrheal stool samples from Nebraska. Emerg. Infect. Dis. 6:530–533 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Vally H, Hall G, Dyda A, Raupach J, Knope K, Combs B, Desmarchelier P. 2012. Epidemiology of Shiga toxin producing Escherichia coli in Australia, 2000–2010. BMC Public Health 12:63. 10.1186/1471-2458-12-63 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.European Food Safety Authority (EFSA) 2009. Scientific report of EFSA: technical specifications for the monitoring and reporting of verotoxigenic Escherichia coli (VTEC) on animals and food (VTEC surveys on animals and food). EFSA J. 7:1–43 [Google Scholar]
- 10.Elliott SJ, Wainwright LA, McDaniel TK, Jarvis KG, Deng YK, Lai LC, McNamara BP, Donnenberg MS, Kaper JB. 1998. The complete sequence of the locus of enterocyte effacement (LEE) from enteropathogenic Escherichia coli E2348/69. Mol. Microbiol. 28:1–4 [DOI] [PubMed] [Google Scholar]
- 11.Perna NT, Mayhew GF, Pósfai G, Elliott S, Donnenberg MS, Kaper JB, Blattner FR. 1998. Molecular evolution of a pathogenicity island from enterohemorrhagic Escherichia coli O157:H7. Infect. Immun. 66:3810–3817 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Beutin L, Jahn S, Fach P. 2009. Evaluation of the ‘GeneDisc' real-time PCR system for detection of enterohaemorrhagic Escherichia coli (EHEC) O26, O103, O111, O145 and O157 strains according to their virulence markers and their O- and H-antigen-associated genes. J. Appl. Microbiol. 106:1122–1132 [DOI] [PubMed] [Google Scholar]
- 13.Delannoy S, Beutin L, Fach P. 2012. Use of clustered regularly interspaced short palindromic repeat sequence polymorphisms for specific detection of enterohemorrhagic Escherichia coli (EHEC) strains of serotypes O26:H11, O45:H2, O103:H2, O111:H8, O121:H19, O145:H28 and O157:H7 by real time PCR. J. Clin. Microbiol. 50:4035–4040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Delannoy S, Beutin L, Fach P. 2013. Towards a molecular definition of enterohemorrhagic Escherichia coli (EHEC): detection of genes located on O island 57 as markers to distinguish EHEC from closely related enteropathogenic E. coli strains. J. Clin. Microbiol. 51:1083–1088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wickham ME, Lupp C, Mascarenhas M, Vazquez A, Coombes BK, Brown NF, Coburn BA, Deng W, Puente JL, Karmali MA, Finlay BB. 2006. Bacterial genetic determinants of non-O157 STEC outbreaks and hemolytic-uremic syndrome after infection. J. Infect. Dis. 194:819–827 [DOI] [PubMed] [Google Scholar]
- 16.Coombes BK, Wickham ME, Mascarenhas M, Gruenheid S, Finlay BB, Karmali MA. 2008. Molecular analysis as an aid to assess the public health risk of non-O157 Shiga toxin-producing Escherichia coli strains. Appl. Environ. Microbiol. 74:2153–2160 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Konczy P, Ziebell K, Mascarenhas M, Choi A, Michaud C, Kropinski AM, Whittam TS, Wickham M, Finlay B, Karmali MA. 2008. Genomic O island 122, locus for enterocyte effacement, and the evolution of virulent verocytotoxin-producing Escherichia coli. J. Bacteriol. 190:5832–5840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bugarel M, Beutin L, Fach P. 2010. Low-density macroarray targeting non-locus of enterocyte effacement effectors (nle genes) and major virulence factors of Shiga toxin-producing Escherichia coli (STEC): a new approach for molecular risk assessment of STEC isolates. Appl. Environ. Microbiol. 76:203–211 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Vlisidou I, Marchés O, Dziva F, Mundy R, Frankel G, Stevens MP. 2006. Identification and characterization of EspK, a type III secreted effector protein of enterohaemorrhagic Escherichia coli O157:H7. FEMS Microbiol. Lett. 263:32–40 [DOI] [PubMed] [Google Scholar]
- 20.Imamovic L, Tozzoli R, Michelacci V, Minelli F, Marziano ML, Caprioli A, Morabito S. 2010. OI-57, a genomic island of Escherichia coli O157, is present in other seropathotypes of Shiga toxin-producing E. coli associated with severe human disease. Infect. Immun. 78:4697–4704 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Delannoy S, Beutin L, Ylanna Burgos Fach P. 2012. Specific detection of enteroaggregative hemorrhagic Escherichia coli O104:H4 strains using the CRISPR locus as target for a diagnostic real-time PCR. J. Clin. Microbiol. 50:3485–3492 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bugarel M, Martin A, Fach P, Beutin L. 2011. Virulence gene profiling of enterohemorrhagic (EHEC) and enteropathogenic (EPEC) Escherichia coli strains: a basis for molecular risk assessment of typical and atypical EPEC strains. BMC Microbiol. 11:142. 10.1186/1471-2180-11-142 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Perelle S, Dilasser F, Grout J, Fach P. 2007. Screening food raw materials for the presence of the world's most frequent clinical cases of Shiga toxin-encoding Escherichia coli O26, O103, O111, O145 and O157. Int. J. Food Microbiol. 113:284–288 [DOI] [PubMed] [Google Scholar]
- 24.Fratamico PM, Bagi LK, Cray WC, Naragan N, Yan X, Medina M, Liu Y. 2011. Detection by multiplex real-time polymerase chain reaction assays and isolation of Shiga toxin producing Escherichia coli serogroups O26, O45, O103, O111, O121 and O145 in ground beef. Foodborne Pathog. Dis. 8:601–607 [DOI] [PubMed] [Google Scholar]
- 25.DebRoy C, Roberts E, Valadez AM, Dudley EG, Cutter CN. 2011. Detection of Shiga toxin producing Escherichia coli O26, O45, O103, O111, O113, O121, O145, and O157 serogroups by multiplex polymerase chain reaction of the wzx gene of the O-antigen gene cluster. Foodborne Pathog. Dis. 8:651–652 [DOI] [PubMed] [Google Scholar]