The National Advisory Committee on Microbiological Criteria for Foods (NACMCF or Committee) was asked to report on (i) what is currently known about virulence and pathogenicity of Shiga toxin–producing Escherichia coli (STEC) and how they cause illness in humans; (ii) what methods are available to detect STEC and their specific virulence factors; and most importantly (iii) how to rapidly identify foodborne STEC that are most likely to cause serious human disease. Individual working groups were developed to address the charge questions, as well as to identify gaps and give recommendations for additional data or research needs. A complete list of Committee recommendations is in Chapter 4.
STEC infections cause illnesses that range in severity from diarrhea to diarrhea with grossly bloody stools, called hemorrhagic colitis (HC), to the life-threatening sequela of infection, the hemolytic uremic syndrome (HUS). STEC are ingested in contaminated food or water or through direct contact with infected animals or people. Of all STEC that cause disease in the United States, E. coli O157:H7 (O157) causes the most outbreaks and the largest number of cases of serious illness (as assessed by the number of patients hospitalized or with HUS). The infectious dose 50% (ID50) of O157 is low (estimated to be 10 to 100 bacteria). As determined in animal models, these bacteria bind to enterocytes in the large intestine through the intimin outer membrane protein (the gene for intimin is eae), attach and efface the mucosa, and elaborate Shiga toxin (Stx) that passes from the intestine through the bloodstream to sites in the kidney. Certain Stx subtypes are more commonly associated with severe STEC human illness, e.g., Stx2a, Stx2c, and Stx2d. The serogroups (O antigen type only) linked to most cases of illness in the United States are O157, O26, O103, O111, O121, O45, and O145 in order of decreasing incidence. STEC disease is linked most often to foods of bovine origin and fresh produce; disease burden attributed to beef and dairy products is broadly similar in numbers to that attributed to fresh produce.
Stx production, a phage-encoded trait, and intimin, but not the O antigen type, are major drivers of pathogenicity. Thus, predictions of the pathogenic potential of STEC can be made based on Stx subtype and the potential of the bacteria to attach in the intestine. The combination of virulence genes in E. coli that has led to the most severe disease is stx2a with aggR (a genetic marker for enteroaggregative E. coli [EAEC]). The second-highest risk group are those O157 STEC that have stx2a and eae, followed by that same combination in O26, O103, O111, O121, O45, or O145. The combinations of stx1a and stx2a, or stx2a and stx2c, or stx2d with eaeare also of particular concern. The lack of eae suggests a reduced potential for human disease except when aggR or stx2d is present. There have been a few exceptions to this hierarchy, such as O103 that produce only Stx1 and O113 that is eae negative.
The protocols currently used by the U.S. Food and Drug Administration (FDA), U.S. Department of Agriculture–Food Safety and Inspection Service (USDA-FSIS), clinical laboratories and public health laboratories (PHLs), and the food industry include enrichment, culture, multiplex real-time PCR (RT-PCR), toxin immunoassays, biochemical characterization, DNA-based serotyping, DNA microarray, and whole genome sequencing (WGS). The advantages and limitations of each method are summarized in this report. New and developing high-throughput methods are discussed and include metagenomics, digital PCR, biosensors, and microarray.
STEC disease prevention has been and will continue to be driven by improvement in outbreak detection, investigation, and food industry practices. Highlights of Committee recommendations include the following:
Develop a new universal enrichment culture medium that can be broadly used for all STEC in any food.
Explore high-throughput methods that can detect STEC virulence factor genes directly from enrichment medium and develop and/or improve methods that can ascertain that all critical STEC markers found in the enrichment broth are within the same cell to eliminate the need to isolate the organism.
Expand systematic sampling of food, animals, and water for STEC.
Explore ways for industry to share test data anonymously.
Fund academic research on (i) the regulation of toxin expression and the phages that encode toxin; (ii) mechanisms of attachment by eae-negative STEC; (iii) oral-infection animal models or cell culture models that are more reflective of human disease; and (iv) human host factors that influence the outcome of STEC infection.
Link standardized epidemiological, clinical, and STEC WGS data to monitor trends in recognized and emerging virulence attributes such as Stx type and phage profiles.
Further develop WGS methods to (i) predict toxin levels produced by an STEC and (ii) generate a classification scheme based on genomic clusters.
The Committee agrees that a combination of genetic characteristics (attributes) exist that signal potentially high-risk STEC and that these STEC will eventually be identifiable using high-throughput techniques that analyze gene profiles. Thus, to rapidly identify foodborne STEC that are most likely to cause serious human disease, the Committee recommends that STEC analyses move toward using virulence markers rather than serogroup or serotype to identify pathogens. The Committee concurs that as ease of use increases and costs decrease, culture-independent diagnostic tests (CIDTs) based on genomic clusters or lineages will be more broadly used to predict whether an STEC isolate is likely to cause serious human disease.
STEC are a large, diverse group of bacteria that are characterized by the production of Stx. There are two main Stx types, designated Stx1 and Stx2, and within each are many subtypes. Currently, there are three known Stx1 (Stx1a, Stx1c, and Stx1d) and seven known Stx2 (Stx2a, Stx2b, Stx2c, Stx2d, Stx2e, Stx2f, and Stx2g) subtypes, but some of these are produced mostly by environmental- or animal-associated strains. Thus far, Stx1a, Stx2a, Stx2c, and Stx2d are the subtypes most frequently implicated in human illness. There are estimated to be >400 known STEC serotypes that can produce any of the Stx types, subtypes, or combination of subtypes. However, only a subset of these STEC serotypes have been associated with human illness. Furthermore, the production of Stx alone without other virulence factors, such as intimin, has been deemed to be insufficient to cause severe human illness.