171 O Ags : > 100 K Ags (groups A, L and B) : 56 H Ags formula.
Mobile genetic elements :

Virulence factors gene expresssion :

=> colibacillosis

• diarrhoeagenic Escherichia coli


• enteropathogenic Escherichia coli (EPECs) : O55, O86, O111, O114, O119, O125, O127, O128, O142. They produce a type IV fimbria known as the bundle-forming pilus (BFP), through which is secreted Tir, the receptor for the plasmid-coded, membrane-attached non fimbrial adhesin intimin / Escherichia adhesion factor (EAF), present only in type I but not in type II) => "attaching and effacing". They produce an enterotoxin similar to Stx, causing an inflammatory response. They have localized adhesion (LA) to HEp-2 cells. Serotype O86 is cross-reactive with human blood group Ag B.
• "typical" EPEC (tEPEC) have both chromosomal (the locus of enterocyte effacement (LEE) island) and plasmid-borne (residing on the EPEC adherence factor (EAF) plasmid) virulence factors, which are linked by common gene regulators^ref
• "atypical EPEC" (aEPEC) carry only the LEE island (Kaper JB. Defining EPEC. Rev Microbiol Sao Paulo. 1996;27(Suppl1):130–3). Previously, these investigators reported a high prevalence of aEPEC among pediatric diarrhea patients in Melbourne, including both infants and older children (in contrast to the strong tendency for infants to be infected with tEPEC)^ref. In contrast to patients infected with other pathogens, patients infected with aEPEC are far more likely to experience diarrhea past 14 days, the point long recognized as a clinical watershed that heralds increased risk for illness and death^ref. aEPEC's prevalence among diarrhea patients, the pathogen's strong association with diarrheal symptoms, and the infection's distinctively persistent nature argue for a high disease burden in Melbourne. Although the authors define aEPEC strictly on the basis of positivity for the LEE eae gene and failure to amplify a bfpA pilin gene (not assessing additional plasmid loci), the absence of tEPEC serotypes and the occurrence of disease in children older than infants suggest that these are indeed aEPEC
=> endemic diarrhea in babies (cholera nostras) (lethality : 16%)
• Some types of EPEC are referred to as dherent E. coli (EAEC), based on specific patterns of adherence.

=> traveler's diarrhea in Mexico and in North Africa.
Chemotherapy : doxycycline
• enterohemorrhagic Escherichia coli (EHECs) after 3-4 days incubation => abdominal pain and watery diarrhea => hemorrhagic colitis => bloody diarrhea (often the patient is afebrile) => self-limiting in 6-7 days or evolving to hemolytic uremic syndrome (HUS) / Gasser syndrome in 5-15% of patients aged < 10 years (those harbouring symbiont Escherichia coli strains susceptible to lytic infection by the bacteriophage encoding Stx2, as once infected they become toxin producing cells, too; lethality 3-5%; death or ESRD in 12%; 25% of survivors have long term kidney disease as a sequel)
• O104:H7 : outbreak in Spring 2011 in Germany (1733 cases; 520 HUS; 18 deaths,1 in Sweden) and 9 other European nations (80 people sick from the bacterium, most of whom
  had recently visited northern Germany)
  
Prevention : probiotic therapy with phage-resistant Escherichia coli strains
Therapy : hemodyalisis. Antimicrobials are uneffective.
Epidemiology :  in 1996 21 elderly people died and another 500 were made ill after eating contaminated meat supplied by a butcher's shop in Wishaw, Lanarkshire, UK.
Clinically, the O157 infections had a higher frequency of bloody stools, fecal leukocytes, abdominal pain with shorter symptom duration and far higher frequency of hemolytic-uremic syndrome (although a cluster of O121 HUS cases associated with a lake in Connecticut, USA, has been reported). Since toxin assay did not identify all O157 strains found on sMAC plates, the investigators did not advocate performing toxin assay alone.
Prevention : keeping animals off the fields for the preceding 3 weeks prior to recreational use, mow the grass, keep it short. Any visible animal dropping should also be removed at the beginning of the 3-week period.
It is considered by CDC as a category B biological weapon.
• enterotoxigenic Escherichia coli (ETECs) : O6, O16, O18, O25, O27, O78, O148, O159. The same or separate plasmid has ToxA and ToxB genes coding respectively for
• p83^{heat-labile (LT)} AB₅ toxin (~ cholera toxin) binds to GM₁ and cause ADP-ribosylation of Ga_s subunits.
• heat-stable (ST) toxins
• p8^{STI / Sta} (which can occur as either STaI or STaII) is coded by the plasmid Tn1681 : they act as agonist for the membrane-bound guanylyl cyclase C => increase in [cGMP]_i. => activation of CNG channels => influx of Ca²⁺ => inhibition of DNA synthesis and proliferation. As part of the tumor-suppressive guanylin, uroguanylin=>guanylyl cyclase C pathway is frequently lost during colorectal cancer (CRC) progression, there is an inverse correlation between CRC and ETEC infections.
• p5^{STII / STb} is coded by a R-Ent plasmid :
• colonization factors (CFA) I-V are fimbrial Ags
=> diarrhea without fever in both babies and adults (80% of "traveller's diarrhoea")
• enteroinvasive Escherichia coli (EIECs) : O28, O112, O115, O124, O136, O143, O144, O147, O152. They produce Shigella adhesin (but not toxin !).

=> dysentery-like syndrome with fever
Diagnosis : Sérény test
• ggregative Escherichia coli (EAggEC) have aggregative adhesion (AA) to HEp-2 cells and create aggregates in vitro . They also produce :
• hemolysin
• cytolethal distending toxin (CLDT)
• ggregative E. coli heat-stable toxin (EAST)
=> persistent diarrhea in babies (cholera nostras)
• diffusely adherent Escherichia coli (DAEC) have diffuse adhesion (DA) to HEp-2 cells but don't create aggregates in vitro

=> transient or persistent diarrhea
• Shiga toxin-producing Escherichia coli (STEC) / Vero-cytotoxin-producing Escherichia coli (VTEC) :
• O157:H7 (70%; don't ferment sorbitol, and lactose-fermenting colonies shown to be E. coli that are selected for being sorbitol-negative are tested for the somatic O157 antigen by latex agglutination.  Rarely, a O157:H7 strain is sorbitol-fermenting and, more commonly, other strains that produce Shiga toxin, such as E. coli that contain an alternative somatic antigen, will cause HUS. In Germany, non-O157 strains may cause more HUS than O157 strains^ref). Toxin production can be detected by immunoblot, and O157 antigen can be detected by ELISA. Not only do these isolates ferment sorbitol, but they also are tellurite-susceptible^ref and will not be isolated on cefixime-tellurite Sorbitol-MacConkey agar (CT-SMAC), a selective medium for the usual E. coli O157. Sorbitol-fermenting VTEC O157 grow as pink colonies on SMAC agar and some may grow relatively poorly on CT-SMAC, although this property appears variable. If clinical symptoms are suggestive of VTEC infection (particularly in children aged under 15 years and adults aged over 65 years), and presumptive sorbitol non-fermenting E. coli O157 colonies are not observed on SMAC or CT-MAC agar, then sorbitol-fermenting colonies should be tested for agglutination with E. coli O157 antiserum. Colonies giving agglutination should be identified as E. coli and sent for confirmation, phage typing, and detection of VT genes. Fecal samples from appropriate clinical cases from whom VTEC O157 has not been isolated may be submitted for detection of non-O157 VTEC. Lactose-fermenting Salmonella species represent a similar diagnostic problem^ref. Although first recognized in Germany in 1988 during an outbreak of HUS^ref, sorbitol-fermenting, verotoxigenic E. coli O157 was thought to be restricted to continental Europe^ref until an isolation in Australia in 2002. The first isolation in the UK was reported in 2002 (1 case in Scotland), then in 2003 (1 case in Scotland), 2004 (1 case in England)^ref, and 2006 (4 cases in England)^ref. All isolates possessed genes for verocytotoxin (VT) 2, and pulsed field gel electrophoresis of DNA fragments gave indistinguishable profiles^ref.
• >> O26:H1 (motile), O103, O104:H21 (motile), O111:H2, O111:NM (or H-) (nonmotile), O113, O121, O127, O128, O142, O145, O153, O163 (30%: don't ferment sorbitol). O145 infections make up only 224 of 9358 reports (2.4%) in the Enter-net VTEC database between the years 2000-2004. When data from the United Kingdom and Ireland (where O157 infections predominate) are excluded, O145 is the 4th most common serogroup reported in mainland Europe, making up 224 of 5123 cases (4.4%). Underascertainment of non-O157 VTECs is thought to be common, as stool samples are not routinely tested for VTEC strains other than serogroup O157. Such non-O157 isolates can be obtained from sheep and cattle and, although causing as much as 30% of outbreaks of VTEC^ref, appear to be somewhat less virulent -- or at least more variable in virulence -- in a variety of in vivo and in vitro assays^{ref1, ref2, ref3}. In analyzing the genetic and phenotypic profiles of non-O157 groups, it has been found that they belong to their own lineages and have unique profiles of virulence traits different from O157^ref. The serogroups appearing to be most prominent are O26, O111, O128, and O103^ref. If a laboratory is using sorbitol-MacConkey (sMAC) plates to identify VTEC by virtue of O157's ability to ferment sorbitol, the non-O157 strains will be missed.  In a 3-year pediatric study from the University of Washington, USA^ref, 1851 stool samples were processed for sorbitol fermentation as well as toxin production by EIA, and 28 strains of O157 were found along with O103 (4 strains), O118 (2 strains), O111 (2 strains), and 3 other strains. Clinically, the O157 infections had a higher frequency of bloody stools, fecal leukocytes, and abdominal pain with shorter symptom duration. 5 (18%) of O157 infections developed hemolytic-uremic syndrome; none of the non-O157 strains did.  Since toxin assay did not identify all O157 strains found on sMAC plates, the investigators did not advocate performing toxin assay alone. Non-O157 can produce hemolytic-uremic syndrome, as demonstrated by a cluster of O121 cases associated with a lake in Connecticut, USA^ref. Since toxin assays are not uniformly performed in many areas, and most cases do not produce HUS, it is likely that cases due to non-O157 strains are being missed.  How frequent this phenomenon will become over time is unclear
• enteroaggregative E. coli O104:H4 horizontal genetic exchange allowed for the emergence of the highly virulent Shiga-toxin–producing enteroaggregative E. coli O104:H4 strain that caused the German outbreak. More broadly, these findings highlight the way in which the plasticity of bacterial genomes facilitates the emergence of new pathogens^ref.
They arose from E.coli O55 + Shigella Stx gene => verotoxins (i.e. toxic for Vero cell line)^ref :
• verotoxin 1 (VT-1) / Shiga toxin 1 (Stx1) / Shiga-like toxin I (SLT-I), very similar to Shiga toxin (Stx) produced by Shigella dysenteriae type I, group consists of
• Stx1
• Stx1c
• Stx1d
• verotoxin 2 (VT-2) / Shiga toxin (Stx2) / Shiga-like toxin II (SLT-II), dissimilar from Shiga toxin (Stx) group consists of
• Stx2
• Stx2c
• Stx2d
• Stx2dactivatable
• Stx2e
• Stx2f
They are AB₅ toxins : B subunits bind to CD77 / Pk blood group antigen / globotriaosylceramide (Gb3), while A subunit is an N-glycosylase that cleaves an A residue in rRNA 28S. As for Stx, the operator region of Stx1 (but not Stx2) contains a consensus fur box that is responsible for iron-regulated production. Neither iron nor any other environmental factors examined affect the expression of Stx2. However, intestinal mucus enhances the activity of some Stx2 variants. The Stxs, which carry typical N-terminal leader sequences, are not actively secreted from the bacterial cell and are thought to be released into the milieu during cell lysis.
Subtilase cytotoxin, which is produced by certain Shiga toxigenic strains of Escherichia coli. The extreme cytotoxicity of this toxin for eukaryotic cells is due to a specific single-site cleavage of the essential endoplasmic reticulum chaperone BiP/GRP78. The A subunit is a subtilase-like serine protease; structural studies revealed an unusually deep active-site cleft, which accounts for its exquisite substrate specificity. A single amino-acid substitution in the BiP target site prevented cleavage, and co-expression of this resistant protein protected transfected cells against the toxin. BiP is a master regulator of endoplasmic reticulum function, and its cleavage by subtilase cytotoxin represents a previously unknown trigger for cell death^ref.
Other virulence factors are eae, intimin (responsible for attachment of the organism and mucosal effacing lesions), E-hly, espA, etp and katP^ref. There is an increasing body of evidence that the stx genotype is associated with the severity of disease and strongly determines the likelihood of progression to HUS. Patients infected with EHEC strains that harbor stx2 as the sole stx gene have been found to develop HUS significantly more frequently than patients infected with strains harboring stx1 alone or stx1 in combination with stx2. In addition, stx2c has been associated with HUS^ref. Although the presence of stx2 and stx2c is associated with an increased probability of HUS development, STEC that carry stx2d or stx2e can be considered to be LP-STEC, irrespective of the presence of other virulence factors. The former strains have been associated with non-bloody diarrhea but not with HUS^{ref1, ref2}; the latter strains mainly cause edema in pigs, and the rare human isolates originate from patients with mild diarrhea or from asymptomatic carriers^ref. In their report, Bielaszewska et al.^{ref1, ref2} focus on a very interesting Stx: namely, a type of Stx with cytotoxicity that is activated by human and mouse intestinal mucus and, therefore, is designated Stx2d-activatable. Bielaszewska et al. show that the activatable phenotype is regularly expressed by the clinical isolates that harbor the corresponding gene. Therefore, the activation of this Stx by human intestinal mucus in vivo during infection may increase the virulence of such strains. The stx2d-activatable gene was identified in 60 (6.5%) of 922 STEC strains [not all O157:H7]; in 31 of these strains, it was the sole stx gene. 30 of these 31 strains produced Stx2d-activatable. All of them lacked the intimin-encoding eae gene. Among eae-negative STEC, which typically cause mild diarrhea or asymptomatic infection, production of Stx2d-activatable was significantly associated with the ability to cause severe disease, including bloody diarrhea (p < .001), and with systemic complications, such as hemolytic uremic syndrome (p < .001). This activatable genotype of Stxd appears to be an important marker to assess risk of serious outcomes in STEC infections. Additionally, in the editorial, the authors report that the EU-ENTERNET meeting held in Prague in September 2006 felt that the following terms for STEC organisms are useful:
• HUS-inducing STEC (HI-STEC)
• diarrhea-inducing STEC (DI-STEC)
• low pathogenicity STEC (LP-STEC), i.e. that does not produce diarrhea
Epidemiology : there have been at least 26 reported outbreak of E. coli O157:H7 infection that has been traced to contaminated leafy green vegetables since 1993, and > 400 Shiga toxin–producing E. coli outbreaks have been reported since enterohemorrhagic E. coli O157:H7 was identified as a cause of foodborne disease in 1982 after fast food restaurant-related clusters in 2 states in the USA^ref.  Those in rural environments, with continuous exposure to cattle, have a lower risk as those who do not have this kind of continuous exposure (adaptive immunity ?). Human VTEC infection in dairy farm families was negatively associated with age (P < 0.05) and was not associated with clinical illness. Many dairy farm residents experience subclinical immunizing VTEC infections at a young age, which frequently involve non-O157 VTEC found in cattle^ref.
• Argentina has the highest incidence of HUS in the world, 400 cases per year, 7 times higher compared to other affected countries
• UK : the techniques of ELISA and immunoblotting were used to examine a total of 1667 serum specimens, from apparently healthy members of rural communities in England, for antibodies to the LPS of E. coli O157 and VTs. 29 specimens from 22 individuals were shown to have antibodies specific for E. coli O157 LPS. Some of these lived on livestock farms and had occupational contact with cattle, suggesting that personnel working with farm animals may produce serum antibodies to the O157 LPS antigens. 15 people had IgG class antibodies to O157 LPS, suggesting long-term exposure to E. coli O157, and 5 people had serum antibodies, on more than one occasion, showing evidence of persistent antibodies to O157 LPS. 13 specimens, from 12 of 22 individuals, also contained antibodies to VT1, VT2, or both toxins. 10 specimens contained antibodies to VT1 and VT2, 3 contained antibodies to VT2 only^ref. In rural areas, most individuals carrying VTEC isolates were asymptomatic, whereas in urban areas, a significant association was found between VTEC isolation and diarrhea (p <  0.01)^ref. There is a greater level of exposure of dairy farm residents to VT-producing E. coli (VTEC) strains. The high rate of seropositivity to VT1 in farm residents probably reflects the booster effect of repeated VTEC exposures and argues against a sustained generalized immunosuppressive effect of VT1. Seroepidemiological studies may help in assessing the level of exposure of different populations to VTEC strains^ref. Cases in Cornwall and west Devon associated with animal contact included farm visitors, holiday-makers, and members of farming families and farm workers^ref.
• 157 cases (97 microbiologically confirmed : all phage type (PT) 21/28 and produce verotoxin (VT) 2, with the exception of 1 case that is PT32 VT2), 65% in children of school age, occurred in Wales from 10 to 30 Sep 2005 in over 40 schools, with one death in a 5-year-old boy, caused by contamination of sliced cooked meat^{ref1, ref2, ref3, ref4, ref5}
• the finding of asymptomatic individuals with stool cultures positive for a VTEC organism is not uncommon either in a cluster setting or an endemic one. In one study, fecal samples from 350 farm workers on 276 dairy farms and 50 abattoir employees from 7 different operations were examined for the presence of VTEC O157 by an O157-specific enzyme-linked fluorescent assay followed by immunoconcentration. VTEC O157 was isolated from 4 (1.1%) of the farm workers. A 2nd stool sample was obtained from the positive farm workers as well as from their household contacts. VTEC O157 was isolated from the wife of one of them. The strains from the same household shared the same Verocytotoxin genes profile, phage type and pulsed-field gel electrophoresis pattern. The VTEC O157⁺ subjects had neither intestinal symptoms at [that time] of sampling nor a history of bloody diarrhoea or renal failure. Farm residents often develop immunity to VTEC O157 infection, possibly due to recurrent exposure to less virulent strains of VTEC^ref
Chronology of E. coli O157 outbreaks in Scotland :
• 1990 : an outbreak (16 cases, 4 HUS) of E. coli O157 infection was associated with a restaurant in Lothian.
• 1992 : an outbreak (6 cases, 1 HUS) of E. coli O157 infection was associated with a paddling pool.
• 1994 : an outbreak (22 cases) of E. coli O157 infection involved 6 health boards.
• 1996 : an outbreak (512 confirmed cases, 151 hospitalized; 34 HUS or TTP; 22 fatal) was traced to meat from a local butcher.  75% of the cases were reported from Lanarkshire. During this period, 87 suspected or confirmed cases were also registered in Forth Valley, Greater Glasgow and Lothian.
• 1996 (publication year) : an outbreak (711 cases) of Campylobacter and E. coli O157 infections in Fife was caused by contaminated water.
• 1997 : an outbreak (37 cases) at Falkirk and District Royal Infirmary was caused by E. coli O157 phage type 8.
• 1999 : an outbreak (8 cases) at a primary school in Aberdeenshire was ascribed to unpasteurized goat cheese.
• 1999 : an outbreak (30 cases, 1 with HUS) at a school in the Grampian region was ascribed to goat cheese.
• 1999 : an outbreak (6 cases) in Wester Ross (western coast) was traced to contaminated water.
• 2000 : an outbreak (40 clinical cases, 2 confirmed) of E. coli O157 infection was reported at a scout camp.
• 2001 : an outbreak (30 cases, 2 HUS, 0 fatal) of E. coli O157 phage type 1/28 infection in Lancashire was ascribed to cross-contamination on a butcher shop counter.
• 2002 : an outbreak (5 cases) of E. coli O157 infection was reported among members of a family in Aberdeenshire.
• 2002 : an outbreak (15 cases) of E. coli O157 infection was reported among Highland campers.
• 2005 : an outbreak (10 confirmed, 5 suspected cases) was reported among Scottish tourists returning from resorts on Turkey.
Notes: E. coli O157 infection accounts for > 90% of hemolytic-uremic syndrome (HUS) cases in Scotland. 45 cases (7 fatal) of HUS were reported during 2003 to 2004 - including 43 cases related to E. coli infection.
Phage type 21/28 is the most common phage type of VTEC O157 in the UK^ref : 2 outbreaks occurred in July 2006 in Bromley and Bexley in southeast London ^ref
• USA : causes about 73 000 illnesses and 61 deaths annually in USA (in Hawaii : 9 cases in 2005, 10 cases in 2004, 9 cases in 2003). Escherichia coli O157:H7 was isolated in 0.39% (range 0.07% to 2.13%) of diarrheal stools in a multicenter survey conducted during 1990 to 1992 - isolation rates were 0.13% in the South and 0.57% in the North. The true national incidence is estimated at 725,000 per year, with 100 to 200 fatalities.
• a major outbreak of hemorrhagic colitis involving 501 cases (151 hospitalized, 4 fatal) in Washington State during 1992 to 1993 was traced to hamburger.
• E. coli O157:H7 is a leading cause of bloody diarrhea and of childhood renal failure in this country. 204 children below age 5 died of E. coli diarrhea during 1968 to 1991.
• 10,000 to 20,000 annual infections are estimated. 16 outbreaks were reported during 1993, and 11 during January to June 1994. 1,420 cases of E. coli O157:H7 infection were reported in 1994; 2,139 in 1995; 2,705 in 1996; 615 first half of 1997.
• 21 cases of post-diarrheal HUS were reported during the first half of 1997.
• unpasteurized apple juice was implicated in 3 outbreaks (and 1 fatality) during 1996.
• alfalfa sprouts were responsible for 60 cases in Michigan (25 hospitalized; 2 HUS; none fatal) and 48 cases (11 hospitalized) in Virginia during 1997.
• an outbreak of 15 cases during 1997 resulted in a massive recall of commercially-distributed frozen beef patties
• 2006 : 192 cases, including 30 cases of HUS, 98 hospitalizations, and 1 death due to contaminated spinach from California's Salinas Valley. 72% were female and 10% were children under 5 years old. The proportion of patients who developed HUS was 29% in children (under 18 years old), 7% in those 18 to 59 years old, and 14% in those 60 years old or older. 82% of patients who provided the date when their illnesses began became ill between 19 Aug and 5 Sep 2006. The peak time when illnesses began was 30 Aug to 1 Sep 2006 ­ 32% of patients with the outbreak strain became ill on one of those 3 days^ref (including 1 case in Canada^ref). 8 samples that tested positive for E. coli O157:H7 were taken from cattle feces collected from pastures adjacent to 4 fields, located in Monterey and San Benito counties^{ref1, ref2}. State and federal investigators have narrowed their focus to the ranch where boar trampled fences that had protected a spinach field. Samples taken from a wild pig, as well as from stream water and cattle on the ranch, have tested positive for the same strain of E.coli implicated in the outbreak^ref
List of verotoxin-producing E. coli outbreaks associated with fresh lettuce and spinach since 1993 in the USA^ref :
• Date / Source / Serotype of E. coli / Cases / Setting-Type / State
• July 1993 / Salad / O157:H7 / 10 / Unknown / WA
• August 1993 / Salad / O157:H7 / 53 / Salad Bar/ WA
• September 1994 / Salad / O157:H7 / 26 / School / TX
• July 1995 / Lettuce / O153:H48 / 74 / Lettuce / MT
• September 1995 / Lettuce / O153:H47 / 30 / Scout camp / ME
• October 1995 / Lettuce. O153:H46 / 11 / Lettuce / OH
• May 1996 / Lettuce / O157:H10 / 61 / Mesclun mix / (unclear state - Mod.LL]
• June 1996 / Lettuce / O153:H49 / 7 / Mesclun mix (susp) / NY
• May 1998 / Salad / O157:H7 / 2 / iceberg lettuce / --
• February 1999 / Lettuce / O157:H9 / restaurant / NE
• June 1999 / Salad / O111:H8 / 58 / camp / TX
• September 1999 / Lettuce / O157:H11 / 6 / iceberg lettuce / WA
• October 1999 / Lettuce / O157:H7 / 40 / nursing home / PA
• October 1999 / Lettuce / O157:H7 / 47 / restaurant / OH
• October 1999 / Salad / O157:H7 / 5 / Caesar salad / OR
• October 2000 / Salad / O157:H7 / 6 / deli / IN
• November 2001 / Lettuce / O157:H7 / 20 / salad / TX
• July-02 / Lettuce / O157:H8 / 55 / bagged romaine / WA
• November 2002 / Lettuce / O157:H7 / 13 / restaurant / IL
• December 2002 / Lettuce / O157:H7 / 3 / restaurant / MN
• September 2003 / Lettuce / O157:H7 / 51 / bagged salad / CA
• November 2003 / Spinach / O157:H7 / 16 / nursing home / CA
• November 2004 / Lettuce / O157:H7 / 6 restaurant / NJ
• September 2005 / Lettuce / O157:H7 / 11 / Dole bagged salad / not stated
Outbreaks of E. coli non-O157 infection :
• 1975 : an outbreak (2200 or more cases) of E. coli O6:K15:H16 infection at Crater Lake National Park was attributed to contaminated drinking water.
• 1983 : outbreaks of gastrointestinal illness associated with eating imported French Brie cheese occurred in the District of Columbia, with subsequent cases in Colorado, Georgia, Illinois, and Wisconsin. The responsible organism in this case was a strain of E. coli serotype O27:H20 which produced a heat-stable toxin.
• 1998 : an outbreak (3300 estimated cases) of E. coli O6:H16 infection was caused by delicatessen foods supplied by a caterer.
• 1999 : an outbreak (11 cases, 3 HUS) of E. coli O121 infection was associated with a lake in Connecticut.
• 1999 : an outbreak (58 cases; 2 HUS) of E. coli O111:H8 infection at a cheerleading camp in Texas was ascribed to contaminated ice and salads.  This is the first reported community outbreak attributable to this strain for the USA.
• 2004 : an outbreak (111 cases) of ETEC infection caused by 3 serotypes of E. coli was reported among employees of a company in Illinois.
• 2006 (publication year) : an outbreak (36 cases) of E. coli O169:H41 infection among employees of a company was ascribed to contaminated cole slaw.
• Canada : in the spring of 2000, in Walkerton, a town of 5000 in southern Ontario, an outbreak of E. coli O157:H7 infection claimed 7 lives -- 6 adults and a child -- and over 200 were seen at local area hospitals
Transmission : reservoir : Bos taurus rectum (for non-O157 isolates also Ovis aries). Cattle are the animal most frequently identified as a source of E. coli O157:H7, although other livestock may also carry the bacterium. In a recent report, E. coli O157:H7 was isolated from livestock at 31 of 32 fairs, including 11.4% of 1407 cattle, 1.2% of 1102 swine, 3.6% of 364 sheep and goats, and 5.2% of 154 fly pools. Cattle, swine, and flies at some fairs shared indistinguishable O157:H7 isolate subtypes^ref => orofecal route
• animal or human feces : secondary cases of E.coli O157 in a household may still occur, since excretion of the bacillus in the stool of infected children can be seen for as long as 3 weeks. Day care centers have long been recognized as sites of E. coli O157 outbreaks. As with many other enteropathogens found in day care center outbreaks, spread is frequently hand-to-mouth or fomite to mouth, without any food or water source being recognized. Toddlers are marvelous sharers of toys as well as fecally contaminated hands. In one study^ref, all companion animal (pigs, dogs, parrot and iguanas (n = 48)) isolates of E. coli O157 strains were Shiga-toxin negative. In another^ref, the toxin producing strains from healthy companion animals constituted a very heterogeneous group of E. coli, and many appeared to be specific for their hosts. The absence of eae (attaching and effacing) sequences in most animal STEC strains was felt to indicate that these strains are less virulent for humans than the classical eae-positive enterohemorrhagic types. The Shiga- like toxin II genes were frequent in these strains, were, however, found in strains from healthy dogs, but were rarely found in isolates from other animals. Disease can occur in companion animals, however, as experimental inoculation of dogs was found to cause an illness similar to HUS in a non-O157 STEC strain^ref. To this point, however, Shiga toxin producing E. coli disease in dogs, cats or birds has not at all been a risk factor in human cases. The closeness of dogs and cats to young children in a household would likely have produced such an association if such transmission occurred commonly.
• contaminated beef in undercooked fast-food hamburgers (people need to cook hamburger thoroughly, to 155-160°F = 68-71°C to kill the bacteria). The bacteria associated with the steak are usually external, meaning that the pathogen would be killed by heat, which would not inactivate internal bacteria. But since a lot of meat is injected with tenderizers or flavor-enhancing solutions, the bacteria could be inoculated internally. The donair is a Mediterranean-style pita-bread sandwich of uncertain provenance. It is similar to a gyro, except that the meat is beef, not lamb, and the sauce more resembles honey mustard salad dressing than the cucumber dill of tzatziki. While it most likely came from Greece, Turkey, or Lebanon, the donair has achieved ubiquity in Halifax, Nova Scotia, and other parts of Canada starting in the 1970s^ref
• petting zoos
• other material can be contaminated, including
• salad ingredients (leaf lettuce, iceberg lettuce, mesclun lettuce, romaine lettuce, alfalfa and clover sprouts, restaurant coleslaw). E. coli O157:H7 in spinach can be killed by cooking at 160°F for 15 seconds.
• unpasteurized apple juice
• rarely spoled pasteurized milk^ref
• landed directly in their mouth when air is dusty from the rafters, the walls and the sawdust in the building. A case-control study of 23 patients

(median age, 15 years) and 53 age-matched controls who had attended the Lorain County, Ohio, fair between 20 and 26 Aug 2001. Case-patients had laboratory-confirmed E. coli O157 infection, HUS, or bloody diarrhea within 7 days of attending the fair; controls attended the fair and did not have diarrhea. 6 (26%) case-patients were hospitalized and 2 (9%) developed hemolytic uremic syndrome. Case-patients were more likely than controls to have visited building A (a multipurpose community facility on the fairgrounds; MOR = 21.4 [95% confidence interval {CI}, 2.7-170.7]). Among visitors to building A, illness was independently associated with attending a dance in the building (MOR, 7.5; 95% CI, 1.4-41.2), handling sawdust from the floor (MOR, 4.6; 95% CI, 1.1-20.0), or eating and/or drinking in the building (MOR, 4.5; 95% CI, 1.2-16.6). 24 (44%) of 54 specimens collected from building A 6 weeks after the fair grew Shiga toxin-producing E. coli O157. Isolates from sawdust, the rafters, and other surfaces were identical by molecular fingerprinting to patient isolates. Sawdust specimens collected 42 weeks after the fair also grew the same E. coliO157 strain. Absence of evidence implicating specific food or beverage sources and the recovery of E. coli O157 from the rafters suggest that airborne dispersion of bacteria contributed to the contamination
• when people drink, or swim in^{ref1, ref2, ref3}, tainted water : in the spring of 2000, in Walkerton, a town of 5000 in southern Ontario, an outbreak of E. coli O157:H7 infection claimed 7 lives -- 6 adults and a child -- and over 200 were seen at local area hospitals
Since grain-feeding to enhance meat production — as contrasted with traditional pasturing — promotes enteric colonization by acid-resistant Shiga toxin–producing E. coli^ref, up to 2 to 3% of domestic cattle carry E. coli O157 at the time of slaughter, which is nearly universally associated with surface contamination of the carcass. The use of ground beef produced from hundreds or even several thousand animals greatly increases the risk of contamination of the pooled meat product with Shiga toxin–producing E. coli. Moreover, raising thousands of animals with the use of industrial farming techniques generates staggering quantities of manure potentially contaminated with Shiga toxin–producing E. coli, far more than any farm can use as fertilizer. Huge lagoons of stored liquid manure are the consequence — as are periodic spills of raw manure into nearby streams. During heavy rains, runoff contamination of fields of commercially raised vegetables and orchards, as well as of rivers, lakes, and wells, results in produce-associated or waterborne outbreaks of E. coli infection. Although most reported infections with Shiga toxin–producing E. coli are linked to undercooked ground beef, nearly 25% of outbreaks stem from contamination of commercial produce that is eaten uncooked — lettuce, spinach, cabbage, sprouts, or tomatoes^ref. Outbreaks have been traced to orchards that market unpasteurized apple cider, made from apples that have dropped from the trees and have become contaminated by E. coli O157 from manure used to fertilize the soil. Enteric colonization (and surface contamination) of domestic cattle has resulted in human disease from contaminated milk products and in outbreaks among children visiting petting zoos. Outbreaks at county fairs appear to have been caused by aerosolization of E. coli in the animal barns. Finally, since the infective dose of acid-resistant E. coli O157 (< 100 organisms) is much lower than that of most other bacterial enteropathogens, secondary spread through fecal–oral contact further expands the number of Shiga toxin–producing E. coli cases in most outbreaks^ref. Post-illness shedding of O157:H7 can be prolonged, contributing to secondary cases, but it is probable that transmissibility is greatest during the initial diarrheal phase^ref. Because E. coli O157 can survive in the environment for > 10 months, humans may be at risk of infection long after an environment is initially contaminated^ref

=> aspecific diarrhea => hemolytic uremic syndrome (HUS) / Gasser syndrome^ref. The risk of HUS developing after E. coli O157:H7 is < 15% (in children < 10 years old, the highest risk group). Typically HUS develops after 5 to 13 days of illness with the onset of diarrhea being day 1. Although sometimes prolonged, the usual incubation period of E. coli O157:H7 diarrhea after exposure is 3-4 days^ref.  The gastrointestinal illness is usually quite prominent, but HUS has been reported in patients with asymptomatic infection with the organism^{ref1, ref2}
=> asymptomatic infection with E. coli O157:H7 does occur but is thought to be uncommon.
The severity of the clinical condition was directly related to the number of STEC 'O' antibodies produced by the patients, suggesting that the outcome was related to the number of infecting STEC serotypes^ref. The fecal E. coli flora of cattle is very diverse and can contain a variety of STEC serotypes^ref. Finally, food like spinach, that is somehow contaminated by STEC, may also carry other enteric pathogens such as viruses and combined STEC and Norovirus outbreaks have been reported^ref.
Therapy : antimicrobial agents, antimotility agents, narcotics (also antimotility), and non-steroidal anti-inflammatory agents (can diminish renal blood flow) should be avoided^ref.
• antimicrobial therapy : although a number of studies^ref suggest that it increases the risk of HUS, a more recent meta-analysis^ref revealed a pooled odds ratio of only 1.15 with CI₉₅ of 0.79-1.68, suggesting no higher risk. The bottom line in this issue is not yet known and may relate to the particular strain of O157, the antimicrobial used, and the timing of antimicrobial treatment^ref
• antimotility agents : a retrospective analysis of 278 children who developed O157 enteritis during a 1993 outbreak in Washington State USA had an odds ratio of 2.9 with CI 1.2-7.5^ref. In those who did not develop HUS, there was no difference in the median duration of diarrhea among children who received antimotility agents, but the median duration of bloody diarrhea was longer (4 vs. 3 days, p < 0.05). In another study of 118 E. coli O157 enteritis cases^ref, a relative risk of 44.11 was found for those receiving prolonged antimotility agents with CI 8.48-229.4.
• avoiding unnecessary surgery if the patient presents with significant abdominal pain and bloody diarrhea or just gross blood, which might suggest an ischemic colitis in an adult or an intussusception in a child
• intravenous plasma expansion with fluids could lower the risk of developing HUS^ref
Prevention : cook very well ground beef, avoid cross-contamination, practice correct personal hygiene. Since 2004, there has been a 43% decline in the E. coli contamination of ground-beef samples tested by the USDA — a decrease that follows intensified federal regulatory efforts to enhance food-safety systems and microbiologic testing by commercial meat producers^ref. Much less progress has been made in enhancing the safety of commercially produced vegetables. It is hoped that the Lettuce Safety Initiative, which has been expanded to include spinach, will reduce the occurrence of infection related to consumption of lettuce and spinach. Although the producers and commercial processors of high-risk produce are required to wash products before packaging, washing clearly does not reliably eliminate the risk of contamination. Undercooking is probably common: the undercooking of foods such as poultry or eggs still causes millions of cases of salmonella and campylobacter infection each year^ref. > 4 million tons of lettuce, spinach, and sprouts are consumed in North America every year, and it is unclear how much the risk is reduced by rewashing the produce, even if the consumer bothers to do it. Irradiation of high-risk foods after processing could greatly reduce the incidence of all bacterial foodborne disease and save hundreds of lives each year^ref. The efficacy and safety of food irradiation have been established through extensive research, which has demonstrated that irradiation kills or markedly reduces counts of food pathogens without impairing the nutritional value of the food or making it toxic, carcinogenic, or radioactive. Irradiation of food is already approved in the United States for most perishable foods and has been endorsed by the World Health Organization, CDC, FDA, USDA, American Medical Association, and European Commission Scientific Committee on Food. Unfortunately, because of a widespread lack of understanding of the risks and sequelae of foodborne disease and of the effectiveness and safety of irradiation — and because of intense opposition from antinuclear activists and other interest groups — irradiation of food as a public health measure has not yet achieved widespread acceptance. A number of food products are already commonly irradiated, with no evidence of harmful effects, and for decades, we have sterilized hundreds of millions of implanted medical devices through irradiation each year. The CDC has estimated that irradiation of high-risk foods could prevent up to a million cases of bacterial foodborne disease that result in the hospitalization of > 50,000 persons and kill many hundreds each year in North America. It is time to overcome our irrational fears and act to ensure the safety of our food^ref.
Web resources :
• International VTEC/STEC Club (IVC)
• 4000 reported non-O157 VTEC from around the world including 10 of O121:H19, which were reported from USA, Canada, Hong Kong, Denmark, Germany and Japan, by Dr. Karl A. Bettelheim
• uropathogenic Escherichia coli (UPEC) (O1, O2, O4, O6, O7, O75), which also produce :
• cytotoxic necrotizing factors (CNF) p115^CNF1 and 2, which catalyze deamidation of the Gln residue at position 63 of Rho to Glu. This amino acid change produces a dominant active Rho protein unable to hydrolyze bound GTP.
• peculiar K Ags
• type 1 fimbriae
• adhesive type 1 pili have a crucial role during infection by mediating the attachment to and potentially the invasion of host tissue. These filamentous, highly oligomeric protein complexes are assembled by the 'chaperone–usher' pathway, in which the individual pilus subunits fold in the bacterial periplasm and form stoichiometric complexes with a periplasmic chaperone molecule that is essential for pilus assembly. The chaperone subsequently delivers the subunits to an assembly platform (usher) in the outer membrane, which mediates subunit assembly and translocation to the cell surface. The periplasmic type 1 pilus chaperone FimC binds non-native pilus subunits and accelerates folding of the subunit FimG by 100-fold. Moreover, the FimC–FimG complex is formed quantitatively and very rapidly when folding of FimG is initiated in the presence of both FimC and the assembly-competent subunit FimF, even though the FimC–FimG complex is thermodynamically less stable than the FimF–FimG complex. FimC thus represents a previously unknown type of protein-folding catalyst, and simultaneously acts as a kinetic trap preventing spontaneous subunit assembly in the periplasm^ref.
• P fimbriae / pyelonephritis-associated pili (PAP) binds specifically to D-Gal-D-Gal disaccharides contained in the P blood group Ag on RBCs and on surfaces of uroepithelial cells in approximately 99% of the population (remaining 1% never experience E. coli UTI)
• a-hemolysins, pore forming on all kinds of blood cells
• b-hemolysins, chemotaxis inhibitor

UPECs are embedded within pod-like bulges with cytoplasm of urothelial cells whose shells contain uroplakin and a fibrous polysaccharide matrix similar to biofilms found in ECM. Based on the degree of protrusion into the bladder lumen, the biofilms approximately double the volume of each infected cell
=> urinary tract infections (UTI) and colibacilluria
• K1-containing Escherichia coli

K1 capsular polysaccharide is a Sia homopolymer.
=> neonatal meningitis (40%)
=> septicemia (colibacillemia) (40%), respiratory and abdominal infections, osteomyelitis, prostatitis.
• curlin-producing Escherichia coli

=> sudden infant death syndrome (SIDS) (?)

• EcoCyc
• EcoGene
• The E. coli index

Dr.Daniele Focosi, MD. All rights reserved  |  Terms of use  | Legal notices
About this site  |  Site map  |  Acknowledgements | Open forum  |    |  Current link partners
 Abbreviations and acronyms  |  Medical terminology  |  Add a link  |  Translate this site into your favourite language  |  Softwares

Rate Us: 10 - The Best 9 8 7 6 5 4 3 2 1 - The Poorest

---