Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2016 Nov 1.
Published in final edited form as: Diagn Microbiol Infect Dis. 2015 Jul 26;83(3):319–324. doi: 10.1016/j.diagmicrobio.2015.07.016

Enteropathogenic and enteroaggregative E. coli in stools of children with acute gastroenteritis in Davidson County, Tennessee

Monique A Foster a, Junaid Iqbal a, Chengxian Zhang a, Rendie McHenry a, Brent E Cleveland a, Yesenia Romero-Herazo a, Chris Fonnesbeck a, Daniel C Payne b, James D Chappell a, Natasha Halasa a, Oscar G Gómez-Duarte a,*
PMCID: PMC4618545  NIHMSID: NIHMS724387  PMID: 26298817

Abstract

This prospective acute gastroenteritis (AGE) surveillance was conducted in the inpatient and emergency room settings at a referral pediatric hospital to determine the prevalence of diarrheagenic Escherichia coli (DEC) in children<12 years of age with AGE in Davidson County, Tennessee. Subjects 15 days to 11 years of age, who presented with diarrhea and/or vomiting, were enrolled. Stool specimens were processed for detection of DEC using multiplex polymerase chain reaction. From December 1, 2011, to June 30, 2012, a total of 79 (38%) out of 206 stool specimens from children with AGE tested positive for E. coli. A total of 12 (5.8%) out of 206 stool specimens from children with AGE were positive for a DEC. Eight (67%) out of these 12 were positive for enteropathogenic E. coli, and the remaining 4 were positive for enteroaggregative E. coli. DEC clinical isolates clustered with known E. coli enteropathogens according to multilocus sequencing typing.

Keywords: E. coli, diarrhea, children, gastroenteritis, EPEC, EAEC, epidemiology

1. Introduction

The Centers for Disease Control and Prevention (CDC) estimates that 0.6 episodes of diarrhea per person occur every year in the United States (Jones et al., 2007). Rates are higher for children under the age of 5 years. CDC estimates that 1 in 6 Americans per year (or 48 million people) is infected by foodborne pathogens, resulting in 128,000 hospitalizations and 3000 deaths (Scallan et al., 2011a, 2011b). Most acute gastroenteritis (AGE) episodes (90%) were caused by 31 enteric pathogens including viruses 59%, bacteria (39%), and parasites (2%). The leading causes of AGE were norovirus (58%), nontyphoidal Salmonella spp. (11%), Clostridium perfringens (11%), Campylobacter spp. (9%), and Staphylococcus aureus (Scallan et al., 2011b). The leading causes of hospitalization were nontyphoidal Salmonella spp. (35%), norovirus (26%), Campylobacter spp. (15%), and Shiga toxin–producing Escherichia coli 0157, while the leading causes of death were nontyphoidal Salmonella spp. (28%), norovirus (11%), and Campylobacter spp. (6%).

Shiga toxin–producing E. coli (STEC) are an important cause of foodborne illnesses in the United States, and the severity of the infection results in hospitalization and death. STEC are leading causes of dysenteric diarrhea and mortality secondary to hemolytic uremic syndrome (HUS) (Bavaro, 2012; Johnson et al., 2006; Rangel et al., 2005). Limited information is available on other E. coli intestinal pathotypes. Based on their mechanism of pathogenicity and unique arrays of virulence factors, the different pathotypes are designated enterotoxigenic (ETEC), enteropathogenic (EPEC), enteroaggregative (EAEC), enteroinvasive (EIEC), diffusely adherent (DAEC), and Shiga toxin–producing (STEC)/enterohemorrhagic E. coli. Since diagnostic tools to differentiate these E. coli pathogens are typically not available at clinical laboratories or reference laboratory centers, little is known about their epidemiology.

We attempted to characterize and describe the epidemiology of E. coli pathogens in children who presented with AGE at a large, academic medical center in Nashville, Tennessee. We isolated E. coli strains from stools of children with diarrhea in a defined geographic area and tested them for genetic markers of known pathotypes of diarrheagenic E. coli (DEC).

2. Materials and methods

2.1. Study design and enrollment

This prospective, cross-sectional active surveillance study enrolled children who were 15 days to 11 years old and who resided in Davidson County, Tennessee, presenting to the Vanderbilt Children's Hospital Emergency Department and inpatient units with diarrhea and/or vomiting from December 1, 2011, through June 30, 2012. AGE cases were enrolled if diarrhea, defined as ≥3 loose stools in a 24-hour period, or vomiting, ≥1 episode in a 24-hour period, had occurred. Subjects were excluded for the following criteria: not residing in Davidson County, Tennessee, having symptoms at the time of presentation lasting ≥10 days in duration, immunocompromised, or an inability to understand English or Spanish. Demographic data, illness characteristics, medication, and travel history were obtained by chart review and standardized questionnaires. This study was approved by the Vanderbilt University Institutional Review Board (no. 120099).

2.2. Processing of stool specimens

Specimens consisted of at least 1 gram of whole stool in sterile containers or swabs of stool preserved in Cary–Blair transport media. Specimens plated on MacConkey agar (Becton Dickinson and Company, Sparks, MD, USA) and incubated overnight at 37 °C were examined for the presence of lactose-fermenting, nonmucoid colonies. Such isolates were subcultured onto eosin–methylene blue (EMB) agar (Remel, Lenexa, KS, USA) and incubated for 18–24 hours at 37 °C. Colonies with characteristic E. coli morphology on EMB agar (metallic green sheen) were inoculated into sulfide-indole-motility medium (Neogen Corporation, Lansing, MI, USA) for biochemical testing. Presumptive E. coli isolates that were indole positive, motile, gas-producing, and negative for hydrogen sulfide production were confirmed as E. coli and stored at −80 °C for further testing as described before (Gómez-Duarte et al., 2010).

2.3. Preparation of DNA and DNA amplification by polymerase chain reaction (PCR) assays

Genomic DNA was isolated from E. coli clinical isolates by culturing strains in 2-mL Luria broth (LB) at 37 °C with shaking at 225 rpm overnight. A 200-μL aliquot of bacterial culture was centrifuged and pellet resuspended in 500 μL of Tris-EDTA buffer, boiled for 5 minutes and centrifuged at 10,000×g for 3 minutes. The supernatant containing crude genomic DNA was used as template for PCR. Two separate multiplex PCR reactions were performed on each DNA sample as described before (Gómez-Duarte et al., 2010). Mix 1 detected EPEC, EAEC, and STEC, and mix 2 detected ETEC, EIEC, and DAEC. Both mix 1 and mix 2 reactions contained 18-μL PCR blue master mix (Invitrogen, Carlsbad, CA, USA), 1 μL of oligonucleotide mix 1 or oligonucleotide mix 2, and 1-μL DNA template. Reactions were preheated at 94 °C for 2 minutes and amplified for 40 cycles consisting of 30 seconds at 92 °C, 30 seconds at 58 °C, and 30 seconds at 72 °C. PCR products were separated by electrophoresis using a 2% agarose gel containing ethidium bromide. Gel images were captured and analyzed.

2.4. Serotyping

O– and H– typing of DEC isolates was conducted at Pennsylvania State E. coli Reference Center (Pennsylvania State University, University Park, PA, USA). O– serotyping was conducted using antisera generated against E. coli serogroups designated O1-O187 with the exceptions of O31, O47, O67, O72, O94, and O122 as these are not designated. H– typing was performed by PCR–restriction fragment length polymorphism of fliC flagellar gene responsible for H types.

2.5. Antimicrobial susceptibility testing

Antimicrobial susceptibility to 12 different antibiotics was tested using BD BBL™ Sensi-Disc™ Susceptibility Test Discs methods (Becton, Dickinson and Company Sparks, MD, USA). Strain activity was tested against cefazolin, ceftriaxone, ampicillin, amoxicillin/clavulanic acid, ceftazidime, cefuroxime, cefepime, ciprofloxacin, gentamicin, meropenem, trimethoprim/sulfamethoxazole, and piperacillin/tazobactam (Qin et al., 2008). The zones of bacterial growth inhibition were interpreted according to the 2014 guidelines of the Clinical and Laboratory Standards Institute. Controls used for testing included the E. coli ATCC 29522 as negative control (susceptible to all antibiotics) and Klebsiella pneumoniae ATCC 700603D-5 as positive control (resistant to all beta-lactam antibiotics).

2.6. Biofilm formation assay

Biofilm quantitative assay was performed as described previously (Wakimoto et al., 2004). In brief, cultures were diluted in 297-μL Dulbecco's Modified Eagle Medium at a ratio of 1:100 in a microcentrifuge tube and vortexed. One hundred microliters of the bacteria dilution was added to a 96-well plate, which was sealed with an adhesive film. The plate was then incubated at 37 °C for 24 hours. The plate was washed, and 125 μL of 0.1% crystal violet solution was added to each well and incubated at room temperature for 15 minutes. The crystal violet was discarded and plate rewashed and air dried. Two hundred microliters of 95% ethanol was added to each well and incubated at room temperature for 15 minutes. The solubilized crystal violet was transferred to a 96-well Immulon 2 HB plate, and the optical density (OD) was measured at 590 nm. A strain was considered positive if the OD ≥0.089.

2.7. Multilocus sequence typing (MLST)

E. coli clinical isolates were analyzed by MLST as described online (http://mlst.warwick.ac.uk/mlst/). Internal fragments from 7 housekeeping genes, adk (adenylate kinase), fumC (fumarate hydratase), gyrB (DNA gyrase), icd (isocitrate/isopropylmalate dehydrogenase), mdh (malate dehydrogenase), purA (adenylosuccinate dehydrogenase), and recA (ATP/GTP binding motif of recombinase A), were amplified by PCR and DNA sequenced as described before (Wirth et al., 2006). Sanger DNA sequencing was conducted by the DNA core facility at Vanderbilt University. Forward and reverse DNA sequences for each gene and strain were aligned for comparison and editing. Sequence editing was conducted with DNADynamo software (Blue Tractor Software, North Wales, UK). Sequences for the 7 genes of each strain were concatenated to produce an alignment sequence of 3423 bp. Alignment of concatamers used ClustalW software available online at http://www.phylogeny.fr/version2_cgi/simple_phylogeny.cgi (Dereeper et al., 2008, 2010). Phylogenetic trees for all E. coli isolate and E. coli control MLST concatamers were conducted by bootstrapping procedure. E. coli control MLSTs included STEC, EPEC and ETEC strains reported in the http://mlst.warwick.ac.uk/mlst/ database and previously isolated from human cases of diarrhea. Control E. coli ancestral phylogenetic groups sequences were also obtained from the same database, including the E. coli Reference (ECOR) collection (Ochman and Selander, 1984). MLST-based clonal groups were defined as 2 or more E. coli strains with identical MLST sequence.

2.8. Statistical analysis

Data were analyzed using descriptive statistics, and epidemiologic variable distributions were identified. Central tendency and dispersion values were calculated for all variables analyzed using STATA version 12.

3. Results

3.1. Demographic characteristics of children with diarrhea

During the study period, 410 subjects were enrolled with AGE, and 299 of them consented for further E. coli testing. From the 299 subjects enrolled, 206 (69%) provided stool specimens for further analysis (Table 1). Eighty percent of these subjects (165 of 206) were <5 years old, and 54% were <1 year old, and gender was similar (54% male, 46% female).

Table 1.

Demographics of children with acute gastroenteritis.

Negative for E. coli Positive for E. coli Pathogenic E. colia Total
Sex
Male 70 (63%) 41 (37%) 9 (8%) 111
Female 57 (60%) 38 (40%) 3 (3%) 95
Age
    <1 year old 46 (65%) 25 (35%) 5 (7%) 71
    1 to <2 years old 20 (53%) 18 (47%) 2 (5%) 38
    2 to <5 years old 35 (62%) 21 (38%) 3 (5%) 56
    5 to <10 years old 24 (67%) 12 (33%) 2 (6%) 36
    ≥10 years 2 (40%) 3 (60%) 0 (40%) 5
Race
    White 56 (54%) 47 (46%) 10 (8%) 103
    Black 61(68%) 29 (32%) 2 (3%) 90
    Asian 4 (100%) 0 0 4
    Other 6 (67%) 3 (33%) 0 9
Open in a new tab
a

Pathogenic E. coli are a subgroup of E. coli isolates that tested positive by PCR for either EPEC or EAEC.

3.2. Diarrheagenic E. coli identified among children with diarrhea

E. coli isolates were recovered from 38% (79 of 206) of the AGE specimens. Five individual E. coli isolates were obtained from each stool specimen up. Six percent (12 of 206) of the AGE specimens were positive for DEC. In those specimens, a minimum of 1 out the maximum of 5 E. coli isolates per stool specimen was positive. Of the 12 subjects positive for DEC, the distribution of pathotypes revealed EPEC (67%, n = 8) and EAEC (33%, n = 4). No STEC, ETEC, EIEC, or DAEC were identified. The mean age of subjects was 36 months (Table 2). The mean age of subjects positive for EPEC was 25 months (SD 38, range 2–114). Most cases of EPEC (6/8, 75%) were in children less than 5 years of age. The mean age of subjects positive for EAEC was 16.6 months (SD 16, range 5–39). All EAEC cases (4/4, 100%) were found in children less than 5 years of age (Table 2). None of the study subjects was found to have coinfections with multiple DEC. The majority of DEC were isolated from cases occurring during summer months.

Table 2.

Demographic features of AGE cases positive for diarrheagenic E. coli, December 2011–June 2012.

Sample Subject Age (months) Sex Racea Ethnicity Month of isolation E. coli pathotype
1 EN1E-0007 18 M W NH/NL December EAEC
2 EN1E-0139 22 M W NH/NL March EPEC
3 EN1E-0182 4 F W H/L April EAEC
4 EN1E-0191 39 M W H/L May EAEC
5 EN1I-0027 2 M W NH/NL May EPEC
6 EN1I-0032 27 F W H/L May EPEC
7 EN1E-0224 41 M W NH/NL June EPEC
8 EN1E-0227 10 M B NH/NL June EPEC
9 EN1E-0232 113 F W NH/NL June EPEC
10 EN1E-0241 65 M W H/L June EPEC
11 EN1E-0245 8 M B NH/NL June EAEC
12 EN1I-0034 4 M W NH/NL June EPEC
Open in a new tab
a

Race: W represents white and B represent black. Ethnicity: H/L represents Hispanic or Latino and NH/NL represents non-Hispanic or non-Latino.

3.3. Phenotypic and genotypic features of DEC strains

DEC isolates were characterized at the genetic and phenotypic levels. All EPEC strains (8 of 8) were positive for the eae gene and negative for the bfpA gene indicating that they were atypical EPEC (aEPEC). Diverse numbers of serotypes were associated with EPEC strains, which included 4 O groups and 5 H groups. The majority of EPEC strains, 62% (5 of 8), were resistant to ampicillin (Table 3). All EAEC clinical isolates were positive for aggR and for aaiC. Three different O and H groups were recognized in the 4 EAEC clinical isolates. Only 2 strains had an identical serotype. All EAEC were resistant to ampicillin.

Table 3.

Genotypic and phenotypic features of diarrheagenic E. coli isolates.

Sample Subject Pathotype Virulence genes Serotype Sequence type Antibiotic resistanceb
1 EN1E-0007 EAEC aggR, aaiC O86:H27 3570 A
2 EN1E-0139 EPEC eae O6:H1 342 None
3 EN1E-0182 EAEC aggR, aaiC O86:H27 3570 A
4 EN1E-0191 EAEC aggR, aaiC O3:H2 10 A
5 EN1E-0224 EPEC eae O–:H– 10 A
6 EN1E-0227 EPEC eae O119:H21 40 A, A/C
7 EN1E-0232 EPEC eae O108:H9 302 None
8 EN1E-0241 EPEC eae O–:H– 10 A, A/C
9 EN1E-0245 EAEC aggR, aaiC O92:H33 ST4599a A, A/C, S
10 EN1I-0027 EPEC eae O119:H30 ST4600a A
11 EN1I-0032 EPEC eae O35:H10 10 None
12 EN1I-0034 EPEC eae O–:H9 779 A, S
Open in a new tab
a

New MLST sequence type submitted to http://mlst.warwick.ac.uk/mlst/.

b

A = ampicillin; A/C = amoxicillin/clavulanic acid; S = sulfamethoxazole-trimethoprim.

3.4. Few EAEC clinical isolates form biofilm of inert surfaces

EAEC strains typically form biofilms over inert and biological surfaces. To evaluate if EAEC clinical isolates form biofilms, individual colonies from cases with AGE were tested in a biofilm assay. Only 2 EAEC isolates from a single subject (ENIE-0245) were positive for biofilm formation (Fig. 1). The remaining EAEC isolates from the same subject and from the remaining 3 subjects with AGE were negative for biofilm formation. Control strains of E. coli produced the expected patterns: K12 and EAEC O42 were negative and positive, respectively, for biofilm formation.

Fig. 1.

Fig. 1

Open in a new tab

Quantitative biofilm formation assay on EAEC isolates. Individual colonies for each case of diarrhea positive for EAEC were tested in the biofilm assay. E. coli K12 and DH5 strains are negative controls, and EAEC O42 is positive control for biofilm formation. The optical density is proportional to the amount of bacteria attached to the inert surfaces. Interrupted line is the threshold above which the strain is positive for biofilm formation. The threshold corresponds to the average of the negative controls plus 2 times the SD. In this assay, the threshold was set at an absorbance (A590 nm) of 0.43.

3.5. Phylogenetic analysis

MLST conducted on E. coli pathotypes from subjects with AGE in Davidson County, Tennessee, revealed that they are genetically related to EPEC and EAEC clinical isolates previously reported in the MLST database (http://mlst.warwick.ac.uk/mlst/). MLST sequences from EPEC and EAEC isolates from children with AGE in Davidson County, Tennessee, were compared to similar sequences from EPEC clinical isolates from subjects with AGE and available in the MLST database (http://mlst.warwick.ac.uk/mlst/). We identified 5 previously reported MLST sequence types and 2 novel sequence types, among DEC isolates (Table 3). The 2 new sequence types ST4599 and ST4600, submitted to the MLST database, corresponded to the EAEC EN1I-0245 and EPEC EN1I-0027 strains (Table 3). The EPEC EN1I-027 MLST also had a new icd sequence designated icd486 that was also submitted to the MLST database. A phylogenetic tree constructed based on cluster W alignments of MLST concatemer sequences from Davidson County was compared with MLST sequences from ancestral E. coli strains from groups A, B1, B2, and D. As shown in Fig. 2, most DEC from Davidson County clustered with E. coli ancestral group A and B1 strains. The only exceptions were the EPEC EN1I-0034 strain that clustered with ancestral group B2 and the EPEC EN1I-0027 strain that was not related to any of the E. coli ancestral group strains.

Fig. 2.

Fig. 2

Open in a new tab

MLST phylogenetic tree of E. coli intestinal pathotypes isolated from children with AGE. Phylogenetic tree was constructed after assembly and alignment of MLST DNA sequences from diarrheagenic E. coli isolates using ClustalW program. E. coli MLST sequences from ancestral groups A, B1, B2, and D available at the MLST database (http://mlst.warwick.ac.uk/mlst/) were used as controls. Arrows represent clinical isolates from Davidson County, Tennessee; thin arrows indicate EPEC study strains, and bold arrows indicate EAEC study strains.

EPEC isolates from Davidson County revealed close relationships with previously reported EPEC clinical isolates derived from AGE cases from different worldwide locations. EPEC EN1E-0139 strain with MLST sequence type 342 has a clonal relationship with EPEC Trh37, a Norway strain O177 isolated from a case of diarrhea (Fig. 3). EPEC strains EN1E-0224, EN1E-0241, and EN1I-0032 are also identical and cluster with E. coli ancestral group A and specifically with MLST sequence type 10 (ST10), associated with EPEC human pathogens. The remaining EPEC MLSTs we identified are not identical to EPEC from the MLST database, yet they are related (Fig. 3). The EPEC EN1I-0027 strain, for instance, is distantly related to EPEC 109 in the database, a Chinese O15 isolate from a case of diarrhea.

Fig. 3.

Fig. 3

Open in a new tab

MLST phylogenetic tree of EPEC isolates. Phylogenetic tree of the 8 EPEC MLST sequences were compared with MLST sequences from ancestral groups A, B1, B2, and D as well as EPEC sequences, available at the MLST database (http://mlst.warwick.ac.uk/mlst/). EPEC MLST controls included: EPEC HC68; EPEC DEC2A; EPEC 2348/69; EPEC HC10; EPEC HC15; EPEC Trh36; EPEC DEC6A; EPEC DEC12B; EPEC Trh37; EPEC 181; EPEC HC40; EPEC HC36; EPEC HC91; EPEC HC95; EPEC HC59; EPEC HC66; EPEC HC87; EPEC 109; EPEC 219. E. coli ancestral control strain sequences are derived from the ECOR collection and they include clonal group B1, Ecor26 and Ecor28; clonal group A, Ecor9 and Ecor10; clonal group B2, Ecor56 and Ecor60; and clonal group D, Ecor35 and Ecor36. Arrows represent EAEC clinical isolates from Davidson County, Tennessee.

The 4 EAEC clinical isolates clustered with EAEC clinical isolates reported in the MLST database (Fig. 4). Two EAEC study strains formed a clonal group, in addition to identical MLST, and they also share identical O86:H27 serotype. Interestingly, the EAEC EN1E-0191 strain clustered also with ancestral group A, ST10, which is associated with both EPEC and EAEC strains. The EAEC EN1E-0245 strain is closely related to ST10, although it belongs to its own new sequence type ST4599 (Table 3).

Fig. 4.

Fig. 4

Open in a new tab

MLST phylogenetic tree of EAEC isolates. Phylogenetic tree constructed after assembly and alignment of MLST DNA sequences using the ClustalW program. EAEC MLST sequences from study isolates: EN1E-0007, EN1E-0182, EN1E-0191, and EN1E-0245. E. coli MLST sequences from ancestral groups A, B1, B2, and D as well as STEC sequences, available at the MLST database, were used as controls. E. coli ancestral control strain sequences are derived from the ECOR collection. EAEC MLST sequences from human cases of diarrhea included: EAEC 4356/96; EAEC 101-1; EAEC 13-03250; EAEC 100; EAEC 236; EAEC C08; 42; EAEC C04; EAEC IE-3627; EAEC 10-06632; EAEC 1634; EAEC 1759; EAEC 1772-2; EAEC 1723; EAEC 108; EAEC 1711; and EAEC 219; 188. Arrows represent EAEC isolates from Davidson County, Tennessee.

4. Discussion

Diarrheagenic E. coli are a common cause of infant diarrhea in developing countries, but their impact in pediatric infectious diarrhea in industrialized nations remains unclear (Kotloff et al., 2013). Previous studies in the United States examining the prevalence of DEC have reported several E. coli pathotypes including STEC; EPEC; EAEC; and, less frequently, ETEC (Devasia et al., 2006; Denno et al., 2012; Klein et al., 2006; Nataro et al., 2006). Our study identified atypical EPEC and EAEC as the 2 DEC pathotypes represented among patients with AGE in this geographic area. In our cohort, other E. coli pathotypes, including STEC strains which can cause HUS complications and mortality (Tarr et al., 2005), were not found.

EPEC has been considered a cause of human disease since the 1940s (Walker-Smith, 1996). The identification of certain genetic determinants,eae (encodes intimin, which mediates aggregation) and bfpA (encodes the bundle-forming pilus), has aided in understanding EPEC pathogenicity (Jerse et al., 1990). aEPEC is defined as the presence of eae without bfpA (Trabulsi et al., 2002), and this represented the majority of our samples. While the pathogenic potential of aEPEC strains has been speculative in the past, a recent publication by the Global Enteric Multicenter Study (GEMS) showed aEPEC as the 5th most frequently detected pathogen in patients aged 0–11 months who died of AGE (Kotloff et al., 2013). The 3.8% EPEC prevalence in this study is similar to that reported in other US geographic regions (Denno et al., 2012; Nataro et al., 2006; Caeiro et al., 1999). Additionally, typical EPECs (eae+, bfpA+) were found most frequently in that age group. Our EPEC-positive specimens were too few to demonstrate any association with age, which has been shown previously (Sakkejha et al., 2013). Interestingly, studies conducted in England and Peru have shown that aEPEC is often found in children with and without diarrhea, whether in the developed or developing world (Hernandes et al., 2009; Ochoa and Contreras, 2011). Further surveillance studies in the United States that include healthy controls may provide clues on host risk factors as well as EPEC virulence factors that are associated with disease.

Of the 8 positive EPEC specimens, 2 (25%) had O119 serogroups consistent with recognized classic human EPEC serogrups (Levine and Edelman, 1984). One EPEC sample had an O serogroup bovine association (EPEC O35) (Blanco et al., 2005). Three EPEC isolates were nontypeable, and 2 belonged to O serogroups (O6, O108) not previously associated with EPEC strains. This may indicate that emergence of new O serogroups among EPEC strains in humans is associated with disease or with permanent or transient colonization of the human intestine without evidence of disease.

Studies from multiple cities in the United States, as well as cities in Brazil, Peru, and Burkina Faso, have repeatedly demonstrated that EAEC is the most frequently identified DEC found when specimens are tested for the 6 known pathotypes (Klein et al., 2006; Nataro et al., 2006; Bonkoungou et al., 2013; Cohen et al., 2005; Lozer et al., 2013). The 1.9% EAEC prevalence in this study was lower than the one reported in the United States previously (Denno et al., 2012; Nataro et al., 2006). EAEC is typically defined by a “stacked-brick” aggregation pattern on HEp-2 cells directly associated with biofilm formation (Kaper et al., 1997). Increasingly, PCR methods detecting aggR (a central regulator of virulence) are being used for identification. Reliable ways of determining whether EAEC causes disease are difficult, as it is also known to cause asymptomatic colonization (Nüesch-Inderbinen et al., 2013). EAEC-positive specimens in this study were observed in younger children, consistent with observations in the United States and developing countries (Kotloff et al., 2013; Klein et al., 2006; Nataro et al., 2006; Bonkoungou et al., 2013; Lozer et al., 2013). The 4 EAEC strains isolated in the current study belonged to 3 serogroups: 2 O86 and 1 each O3 and O92. These serogroups have previously been associated with EAEC diarrheal disease and shown to display aggregative adhesion and biofilm formation (Knutton et al., 1992). Interestingly, only 1 EAEC isolate from 1 subject was positive for biofilm formation compared with EAEC strain control. This may indicate that the remaining EAEC isolates unable to form biofilm may be nonpathogenic intestinal colonizers. Alternatively, they may form biofilm only under host intestinal conditions.

Overall, DEC isolates were susceptible to the antibiotics tested in our study, with the exception of resistance to ampicillin (5/12, 41.6%). Only 1 strain was resistant to at least 3 classes of antimicrobials. In contrast, multidrug resistance was reported among DEC isolates in Iran, Kenia, Mexico, and South Korea (Al Jarousha et al., 2011; Harel et al., 1991; Sang et al., 2012; Unno et al., 2011).

MLST typing of strains reveled associations between E. coli clinical isolates from Davidson County, Tennessee, and E. coli ancestral groups A and B1 known to be associated with intestinal disease (Gómez-Duarte et al., 2010; Guerra et al., 2014). EPEC clinical isolates demonstrated close MLST relationships with previously reported EPEC strains. Three EPEC and 1 EAEC Davidson County strains belong to MLST ST10, which includes EPEC and EAEC pathotypes and belongs to the ancestral E. coli group A. These phylogenetic findings are strong evidence that EPEC and EAEC isolates are true intestinal pathogens with genetic traits that may favor human intestinal colonization.

The primary limitation of this study is low overall recovery of E. coli from diarrheal samples. Specimens were sometimes plated up to 15 days after collection, reducing the efficiency of E. coli culture isolation. Our study lacked an analysis of healthy control specimens, so association between E. coli pathotypes and disease cannot be definitively ascertained. As new multipanel systems for DEC detection become available, more accurate epidemiological data on DEC in the United States will be reported (Buss et al., 2015).

In summary, we have reported for the first time the presence of EPEC and EAEC E. coli pathotypes among children with AGE in Davidson County, Tennessee. The EPEC strains, the most frequently identified, were atypical as they do not carry bundle-forming pilus genes. EAEC strains were positive for both aggR and aaiC genes, yet only a single EAEC-positive specimen was able to form biofilm. All pathotypes identified were genetically related to EPEC and EAEC pathotypes previously isolated from human cases of AGE. The role of these pathotypes in the epidemiology of AGE in Tennessee and in the United States warrants further systematic investigation. Information about common MLST types, serotypes, reservoirs, and antimicrobial resistance in prevalent EPEC and EAEC will aid in development of strategies to prevent spread and outbreaks.

Acknowledgements

This study was funded in part by funds from the Department of Pediatrics, Vanderbilt University School of Medicine, to O.G.G.-D., and by a T32 fellowship award to M.F.

References

  1. Al Jarousha AMK, El Jarou MA, El Qouqa IA. Bacterial enteropathogens and risk factors associated with childhood diarrhea. Indian J Pediatr. 2011;78:165–70. doi: 10.1007/s12098-010-0249-0. [DOI] [PubMed] [Google Scholar]
  2. Bavaro MF. E. coli O157:H7 and other toxigenic strains: the curse of global food distribution. Curr Gastroenterol Rep. 2012;14:317–23. doi: 10.1007/s11894-012-0264-6. [DOI] [PubMed] [Google Scholar]
  3. Blanco M, Schumacher S, Tasara T, Zweifel C, Blanco JE, Dahbi G, et al. Serotypes, intimin variants and other virulence factors of eae positive Escherichia coli strains isolated from healthy cattle in Switzerland. Identification of a new intimin variant gene (eae-eta2). BMC Microbiol. 2005;5:23. doi: 10.1186/1471-2180-5-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bonkoungou IJO, Haukka K, Österblad M, Hakanen AJ, Traoré AS, Barro N, et al. Bacterial and viral etiology of childhood diarrhea in Ouagadougou, Burkina Faso. BMC Pediatr. 2013;13:36. doi: 10.1186/1471-2431-13-36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Buss SN, Leber A, Chapin K, Fey PD, Bankowski MJ, Jones MK, et al. Multicenter evaluation of the BioFire FilmArray gastrointestinal panel for etiologic diagnosis of infectious gastroenteritis. J Clin Microbiol. 2015;53:915–25. doi: 10.1128/JCM.02674-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Caeiro JP, Mathewson JJ, Smith MA, Jiang ZD, Kaplan MA, Dupont HL. Etiology of outpatient pediatric nondysenteric diarrhea: a multicenter study in the United States. Pediatr Infect Dis J. 1999;18:94–7. doi: 10.1097/00006454-199902000-00003. [DOI] [PubMed] [Google Scholar]
  7. Cohen MB, Nataro JP, Bernstein DI, Hawkins J, Roberts N, Staat MA. Prevalence of diarrheagenic Escherichia coli in acute childhood enteritis: a prospective controlled study. J Pediatr. 2005;146:54–61. doi: 10.1016/j.jpeds.2004.08.059. [DOI] [PubMed] [Google Scholar]
  8. Denno DM, Shaikh N, Stapp JR, Qin X, Hutter CM, Hoffman V, et al. Diarrhea etiology in a pediatric emergency department: a case control study. Clin Infect Dis. 2012;55:897–904. doi: 10.1093/cid/cis553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dereeper A, Guignon V, Blanc G, Audic S, Buffet S, Chevenet F, et al. Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 2008;36:W465–9. doi: 10.1093/nar/gkn180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dereeper A, Audic S, Claverie J-M, Blanc G. BLAST-EXPLORER helps you building datasets for phylogenetic analysis. BMC Evol Biol. 2010;10:8. doi: 10.1186/1471-2148-10-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Devasia RA, Jones TF, Ward J, Stafford L, Hardin H, Bopp C, et al. Endemically acquired foodborne outbreak of enterotoxin-producing Escherichia coli serotype O169:H41. Am J Med. 2006;119:168.e7–168.e10. doi: 10.1016/j.amjmed.2005.07.063. [DOI] [PubMed] [Google Scholar]
  12. Gómez-Duarte OG, Arzuza O, Urbina D, Bai J, Guerra J, Montes O, et al. Detection of Escherichia coli enteropathogens by multiplex polymerase chain reaction from children's diarrheal stools in two Caribbean-Colombian cities. Foodborne Pathog Dis. 2010;7:199–206. doi: 10.1089/fpd.2009.0355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Guerra JA, Romero-Herazo YC, Arzuza O, Gómez-Duarte OG. Phenotypic and genotypic characterization of enterotoxigenic Escherichia coli clinical isolates from Northern Colombia, South America. BioMed Res Int. 2014;2014:236260. doi: 10.1155/2014/236260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Harel J, Lapointe H, Fallara A, Lortie LA, Bigras-Poulin M, Larivière S, et al. Detection of genes for fimbrial antigens and enterotoxins associated with Escherichia coli serogroups isolated from pigs with diarrhea. J Clin Microbiol. 1991;29:745–52. doi: 10.1128/jcm.29.4.745-752.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hernandes RT, Elias WP, Vieira MAM, Gomes TAT. An overview of atypical enteropathogenic Escherichia coli. FEMS Microbiol Lett. 2009;297:137–49. doi: 10.1111/j.1574-6968.2009.01664.x. [DOI] [PubMed] [Google Scholar]
  16. Jerse AE, Yu J, Tall BD, Kaper JB. A genetic locus of enteropathogenic Escherichia coli necessary for the production of attaching and effacing lesions on tissue culture cells. Proc Natl Acad Sci U S A. 1990;87:7839–43. doi: 10.1073/pnas.87.20.7839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Johnson KE, Thorpe CM, Sears CL. The emerging clinical importance of non-O157 Shiga toxin-producing Escherichia coli. Clin Infect Dis. 2006;43:1587–95. doi: 10.1086/509573. [DOI] [PubMed] [Google Scholar]
  18. Jones TF, McMillian MB, Scallan E, Frenzen PD, Cronquist AB, Thomas S, et al. A population-based estimate of the substantial burden of diarrhoeal disease in the United States; FoodNet, 1996–2003. Epidemiol Infect. 2007;135:293–301. doi: 10.1017/S0950268806006765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kaper JB, McDaniel TK, Jarvis KG, Gomez-Duarte O. Genetics of virulence of enteropathogenic E. coli. Adv Exp Med Biol. 1997;412:279–87. doi: 10.1007/978-1-4899-1828-4_47. [DOI] [PubMed] [Google Scholar]
  20. Klein EJ, Boster DR, Stapp JR, Wells JG, Qin X, Clausen CR, et al. Diarrhea etiology in a Children's Hospital Emergency Department: a prospective cohort study. Clin Infect Dis. 2006;43:807–13. doi: 10.1086/507335. [DOI] [PubMed] [Google Scholar]
  21. Knutton S, Shaw RK, Bhan MK, Smith HR, McConnell MM, Cheasty T, et al. Ability of enteroaggregative Escherichia coli strains to adhere in vitro to human intestinal mucosa. Infect Immun. 1992;60:2083–91. doi: 10.1128/iai.60.5.2083-2091.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kotloff KL, Nataro JP, Blackwelder WC, Nasrin D, Farag TH, Panchalingam S, et al. Burden and aetiology of diarrhoeal disease in infants and young children in developing countries (the Global Enteric Multicenter Study, GEMS): a prospective, case–control study. Lancet. 2013;382:209–22. doi: 10.1016/S0140-6736(13)60844-2. [DOI] [PubMed] [Google Scholar]
  23. Levine MM, Edelman R. Enteropathogenic Escherichia coli of classic serotypes associated with infant diarrhea: epidemiology and pathogenesis. Epidemiol Rev. 1984;6:31–51. doi: 10.1093/oxfordjournals.epirev.a036274. [DOI] [PubMed] [Google Scholar]
  24. Lozer DM, Souza TB, Monfardini MV, Vicentini F, Kitagawa SS, Scaletsky ICA, et al. Genotypic and phenotypic analysis of diarrheagenic Escherichia coli strains isolated from Brazilian children living in low socioeconomic level communities. BMC Infect Dis. 2013;13:418. doi: 10.1186/1471-2334-13-418. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Nataro JP, Mai V, Johnson J, Blackwelder WC, Heimer R, Tirrell S, et al. Diarrheagenic Escherichia coli infection in Baltimore, Maryland, and New Haven, Connecticut. Clin Infect Dis. 2006;43:402–7. doi: 10.1086/505867. [DOI] [PubMed] [Google Scholar]
  26. Nüesch-Inderbinen MT, Hofer E, Hächler H, Beutin L, Stephan R. Characteristics of enteroaggregative Escherichia coli isolated from healthy carriers and from patients with diarrhoea. J Med Microbiol. 2013;62:1828–34. doi: 10.1099/jmm.0.065177-0. [DOI] [PubMed] [Google Scholar]
  27. Ochman H, Selander RK. Standard reference strains of Escherichia coli from natural populations. J Bacteriol. 1984;157:690–3. doi: 10.1128/jb.157.2.690-693.1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ochoa TJ, Contreras CA. Enteropathogenic escherichia coli infection in children. Curr Opin Infect Dis. 2011;24:478–83. doi: 10.1097/QCO.0b013e32834a8b8b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Qin X, Zerr DM, Weissman SJ, Englund JA, Denno DM, Klein EJ, et al. Prevalence and mechanisms of broad-spectrum beta-lactam resistance in Enterobacteriaceae: a children's hospital experience. Antimicrob Agents Chemother. 2008;52:3909–14. doi: 10.1128/AAC.00622-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Rangel JM, Sparling PH, Crowe C, Griffin PM, Swerdlow DL. Epidemiology of Escherichia coli O157:H7 outbreaks, United States, 1982–2002. Emerg Infect Dis. 2005;11:603–9. doi: 10.3201/eid1104.040739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Sakkejha H, Byrne L, Lawson AJ, Jenkins C. An update on the microbiology and epidemiology of enteropathogenic Escherichia coli in England 2010–2012. J Med Microbiol. 2013;62:1531–4. doi: 10.1099/jmm.0.062380-0. [DOI] [PubMed] [Google Scholar]
  32. Sang WK, Oundo V, Schnabel D. Prevalence and antibiotic resistance of bacterial pathogens isolated from childhood diarrhoea in four provinces of Kenya. J Infect Dev Ctries. 2012;6:572–8. doi: 10.3855/jidc.2196. [DOI] [PubMed] [Google Scholar]
  33. Scallan E, Griffin PM, Angulo FJ, Tauxe RV, Hoekstra RM. Foodborne illness acquired in the United States—unspecified agents. Emerg Infect Dis. 2011a;17:16–22. doi: 10.3201/eid1701.P21101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson M-A, Roy SL, et al. Foodborne illness acquired in the United States—major pathogens. Emerg Infect Dis. 2011b;17:7–15. doi: 10.3201/eid1701.P11101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Tarr PI, Gordon CA, Chandler WL. Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet. 2005;365:1073–86. doi: 10.1016/S0140-6736(05)71144-2. [DOI] [PubMed] [Google Scholar]
  36. Trabulsi LR, Keller R, Tardelli Gomes TA. Typical and atypical enteropathogenic Escherichia coli. Emerg Infect Dis. 2002;8:508–13. doi: 10.3201/eid0805.010385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Unno T, Han D, Jang J, Widmer K, Ko G, Sadowsky MJ, et al. Genotypic and phenotypic trends in antibiotic resistant pathogenic Escherichia coli isolated from humans and farm animals in South Korea. Microbes Environ JSME. 2011;26:198–204. doi: 10.1264/jsme2.me10194. [DOI] [PubMed] [Google Scholar]
  38. Wakimoto N, Nishi J, Sheikh J, Nataro JP, Sarantuya J, Iwashita M, et al. Quantitative bio-film assay using a microtiter plate to screen for enteroaggregative Escherichia coli. Am J Trop Med Hyg. 2004;71:687–90. [PubMed] [Google Scholar]
  39. Walker-Smith JA. Evolution of concept of infective origin of infantile diarrhoea from Ballard to Bray (1887–1945). J Infect. 1996;33:213–8. doi: 10.1016/s0163-4453(96)92321-x. [DOI] [PubMed] [Google Scholar]
  40. Wirth T, Falush D, Lan R, Colles F, Mensa P, Wieler LH, et al. Sex and virulence in Escherichia coli: an evolutionary perspective. Mol Microbiol. 2006;60:1136–51. doi: 10.1111/j.1365-2958.2006.05172.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES