Share this article on:

Shigatoxigenic Escherichia coli in Australia: a review

Bettelheim, Karl A.a; Goldwater, Paul N.b,c

Reviews in Medical Microbiology: January 2013 - Volume 24 - Issue 1 - p 22–30
doi: 10.1097/MRM.0b013e328358ac88
Original Articles

Shigatoxigenic Escherichia coli (STEC) belonging to O serogroup O157 are predominantly reported in many parts of the world; this appears not to be the case in Australia, where other serogroups, especially O111, are more common. In this review, the incidence of human STEC infections in Australia from the first reported cases until today is reviewed. In this review, the extensive simultaneous studies on domestic food animals and the incidence of STEC in these animals are discussed in relation to human infections.

aRetired, Southgate, London, UK

bDepartment of Microbiology and Infectious Diseases, SA Pathology at the Women's and Children's Hospital

cSchool of Paediatrics and Reproductive Health, University of Adelaide, North Adelaide, South Australia, Australia.

Correspondence to Paul N. Goldwater, Department of Microbiology and Infectious Diseases, SA Pathology at the Women's and Children's Hospital, 72 King William Road, North Adelaide, SA 5006, Australia. Tel: +61 8 81617432; fax: +61 8 81616882; email:

Received 24 June, 2012

Accepted 30 July, 2012

Back to Top | Article Outline


Although shigatoxigenic Escherichia coli (STEC) were first described in 1977 [1], the importance of shiga toxins has been recognized for over a century [2,3]. The toxin described by these early studies was produced by strains called at the time bacterium Shiga and now known as Shigella dystenteriae type 1. For many years, this toxin was considered an orphan toxin in that it did not appear to play a role in the clinical condition of dysentery until the observations from Switzerland [4] over 50 years later linked this toxin to haemolytic uraemic syndrome (HUS). Although taxonomists realized for many years that the genera Shigella and Escherichia are closely related and some have even suggested that they should be combined into one genus, it took over 20 years to realize that E. coli can also produce this toxin [1].

It then took another 2 years until STEC were described in association with human disease first in the United Kingdom [5] and shortly afterwards in New Zealand [6]. The older literature suggested that STEC infections may well have occurred previously, given that an epidemic in the early 1950s was caused by E. coli O111 but its significance was not recognized [7], although it has played an important role in later outbreaks both in Australia and other countries.

Shortly thereafter, two outbreaks followed in the USA associated with consumption of hamburgers and the causative organisms were strains of STEC described by the authors as ‘a rare Escherichia coli serotype’, that is O157:H7 [8]. Some years later, retrospective studies showed that this serotype probably emerged in South American cattle [9]. As a result of the original observations [1], which showed that the toxin affected vero cells, it is also known as verotoxin. Thereafter, ‘verocytoxigenic’ (VTEC) and shigatoxigenic (STEC) became interchangeable terms.

One of the features of these strains of STEC O157 was that they lacked the ability to ferment sorbitol and a selective medium based on this observation was developed [10]. Further selective media for this serotype followed and are in regular use in many laboratories around the world [11–13]. This has led to the conclusion that the O157 STEC are the only STEC, which could and therefore should be sought in clinical specimens and, thus, the non-O157 STEC, which in some cases can be as virulent as the O157 strains, have been largely ignored. Accompanying the discovery that there are strains of sorbitol-fermenting O157 STEC [14], new light was shed on the importance of looking for all STEC and not relying on tests for only one E. coli serogroup, based on nonfermentation of sorbitol.

Subsequent studies had shown that there are two distinct shiga toxins and that STEC may carry one or both of these [15–17]. Despite a number of reports coming from around the world on the importance of non-O157 STEC associated with human disease including HUS [18–20], many laboratories have still not realized the importance of these pathogens and hence they remain underreported. Serological [15–17,21,22] and molecular [23–25] techniques are available to test for and identify these pathogens and, thus, it is surprising that these have not been used more frequently.

Nevertheless, non-O157 STEC, especially O26:H11 and O111:H-, have been regularly described from sporadic cases and outbreaks, initially by the few laboratories that were aware of their importance. Examples of such reports include O111:H− in Italy [26]; O111:H2 in Germany [26]; O103:H2 in France [27], the United States [28] and Germany [29]; O145:H5 in Japan [30]; and O104:H21 [31] and O111:H8 [32] in the United States. Finally to this list of examples has to be added the 2011 outbreak centred in Germany but which spread globally involving STEC O104:H4, which also had the characteristics of enteroaggregative E. coli [33].

Back to Top | Article Outline

The situation in Australia

The first reports of STEC infections in Australia include a single case and a small family outbreak with STEC O111 strains [34,35]. Realization that STEC serotypes other than O157:H7 should also be sought soon led to them being found. A brief study in central Australia yielded a non-O typeable strain of STEC [36]. These findings were summarized in brief reviews [37,38]. Purely coincidentally, one of the reviews [38] appeared at the same time as a major STEC outbreak. In this review were noted a number of cases of STEC O157 infections associated with both bloody diarrhoea and HUS; a significant number of non-O157 STEC infections were also reported, including (and especially) strains belonging to serogroup O111. The importance of testing for these and identifying them is stressed. Therefore, when the outbreak occurred in Adelaide, South Australia, the local laboratory was ready and able to identify the main STEC involved very rapidly, namely STEC O111:H- [39]. However, the outbreak which was both epidemiologically and microbiologically linked to a fermented salami type of sausage yielded isolates both from the patients and the salami and included a number of STEC other than the main STEC O111:H-. These included strains belonging to serogroups O23, O26 and O91 and in a few cases O157:H-. This led to the speculation that if only a selective medium for the isolation of STEC O157 had been used, this outbreak might well have been labelled as an O157 outbreak [40], and on the same basis, other internationally reported outbreaks might well have been similarly misidentified. A full analysis of this outbreak [41,42] showed that all these diverse serotypes played a role in the outbreak. Both the number of complications suffered by the patients and the severity of the symptoms were found to be proportionally related to the number of STEC ‘O’ serogroups to which they had been exposed.

The awareness in Australia of other non-O157 STEC as causative agents of human morbidity and mortality was significantly increased and a number of cases were being reported, including a case of HUS due to STEC O48:H21 [43]. This STEC serotype has only been reported once, from a case of bloody diarrhoea in California, USA [44]. Similarly in 2004, the STEC serotype O86:H27 was isolated from a small outbreak in Queensland (reported in a review in 2012) [45]. This is the only report of this STEC serotype and demonstrates the importance of fully identifying STEC serotypes. A start was also made to determine the possible sources of STEC with a brief study from Tasmania [46] revealing the presence of non-O157 STEC in raw meat and water samples; and a contemporaneous study from Queensland [47] of STEC isolated from sheep carcases revealed the presence of STEC including serotypes O5:H-, O91:H- and O163:H19, which had been implicated in cases of HUS around the world, and O113:H21, which had been implicated in human cases of diarrhoea. These results demonstrated the widespread nature of STEC in food animals and warranted further investigation. This was enhanced by the report of a case of HUS in a 6-week-old child in Melbourne [48], who did not have diarrhoea but presumably developed the HUS following a urinary tract infection, which yielded the STEC serotype O5:H-. Further comparative analysis showed a similarity between this strain and O5:H- ovine isolates from Queensland [48]. A brief review [49] summarizes the situation at that time in Australia by noting that ‘While not diminishing the role of the 0157:H7/H- clone, this review indicates that other serotypes can be responsible for outbreaks as well as cases of sporadic human disease. The current focus on O157:H7 has major implications in terms of diagnosis, the food industry and human health.’ The reports of the isolation from clinical specimens of non-O157 STEC continued with the report [50] of an unusual case of microangiopathic haemolytic anaemia associated with STEC O113:H21. Again it should be noted that this same serotype had already been isolated from sheep carcasses in Queensland [47]. Two reviews [51,52] stressed the importance of non-O157 STEC in Australia and globally. The turn of the century saw the first [53] of extensive studies undertaken to examine the animal sources of STEC, especially in Australia. This showed that bovine and ovine sources were the main reservoir of STEC, whereas none were isolated from porcine samples.

The dawning of the new century coincided with a thought-provoking review [54] which analysed the serotypes of all 90 strains isolated from human sources between 1995 and 1999, leading to the conclusion (within the Australian context) with regard to both severity of the clinical course of the infection and the range of infections caused by the various serotypes isolated that STEC O157:H7 may well be considered ‘a red herring’. With these conclusions in mind, a cheap rapid method for the detection of STEC was developed [22]. This enabled many more STEC to be isolated and characterized. It had also been demonstrated [55] in an analysis of strains of STEC O111 from different European countries and in Australia that apart from strains isolated in France, all other strains ‘isolated in different countries and over an extensive period of time from both man and cattle belong to a unique clone’. This relatively surprising conclusion suggested that the STEC O111:H- clone displayed as universal a distribution as the STEC O157:H7 clone.

As techniques for the isolation of STEC improved, it was revealed that a thorough multidisciplinary investigation of a gastroenteritis outbreak [56] demonstrated the presence of norovirus and STEC in one individual, and norovirus and shigatoxin-producing Aeromonas sobria in another individual, both part of a large outbreak of gastroenteritis. The STEC strain could be isolated and characterized and shown to be O128:H2. This demonstrated that the causes of gastroenteritis outbreaks may well be more complex than had previously been thought. Such studies of gastroenteritis outbreaks further demonstrated the problems [57]: for example, in one outbreak, the predominant pathogen demonstrated was norovirus. However, in a group of 22 faecal specimens tested for pathogenic E. coli, non-shigatoxigenic strains belonging to serotype O157:H39 and O7:H15 were found either singly or both in faecal specimens. Only one faecal specimen was positive for shiga toxin on cell culture, which could not be confirmed by PCR. These results showed that if only a superficial investigation had been carried out, it is possible that this outbreak could have been erroneously ascribed to STEC O157. Three further gastroenteritis outbreaks [58] in all of which STEC O128:H2 was involved also yielded astrovirus in one and norovirus in the other two.

A brief review of the STEC situation in Australia in 2005 [59] showed that on average there were 0.32 cases of STEC infection reported in Australia per 100 000 population; however, in South Australia, it was as high as 2.58 cases per 100 000. The authors discussed these discrepancies, which probably reflect increased awareness in some jurisdictions than in others. Until a standardized method of testing for STEC is introduced across the country, such figures should be viewed cautiously. A contemporaneous study from Canada [60] noted a similar divergence in the rate of isolations of STEC. In a 1997 review [61], it was noted that the pattern of STEC serotypes in the Australian human and animal populations may differ from that in countries in the northern hemisphere. Especially noteworthy was that serotype O157:H7 does not appear to be the predominant serotype isolated. Nevertheless, a PCR was developed for the detection of E. coli O157 based on the rfbE O-antigen synthesis genes [62].

Back to Top | Article Outline

Shigatoxigenic Escherichia coli in animals

The study of STEC isolated from lamb carcases in Queensland [47] was extended to STEC from ovine faeces [63]. By means of PCR, STEC were able to be isolated from the faeces of 45 out of 101 healthy sheep. Serotyping of these STEC revealed the presence of O5:H-, O91:H- and O163:H19, which had previously been reported as being associated with human disease including HUS. Other serotypes found included O75:H- and O75:H40 as well as O123:H- and O153:H-. At the same time [64], other studies on sheep and preslaughter lambs demonstrated that STEC were widely distributed among these animals, with 45% of sheep faeces yielding STEC and 36% of lamb faeces doing so. Strains of O157:H- were also detected. Unfortunately, further serotyping was not performed.

Importantly and problematically, STEC O111:H- had not been isolated from any animal faeces in Australia despite the likelihood that this was the source of human cases. Following an intensive investigation [65] using vancomycin-cefixime-cefsulodin blood agar (BVCCA) to test faecal samples from a 6-year-old dairy cow and four calves with watery diarrhoea from a herd with a history of ill-thrift, showed the presence of only four colonies of STEC among a profusion of other colonies. One colony from the cow and another from a calf yielded STEC O111:H-. Another calf yielded STEC O84:H2 and the fourth strain was subsequently shown to be nontoxigenic and not studied further. Further studies on the use of BVCCA to test faecal specimens from calves with bloody scours yielded strains of STEC O103:H2 [66]. The serotype is a well established pathogenic serotype, although it has not been reported from human cases in Australia.

A study looking at the dynamics of STEC in cattle noted a high diversity in STEC and it appeared that changing conditions such as type of feed or time of feed did not have any effect on the faecal carriage of STEC [67]. Similarly, an extensive study over a 12-month period examined 588 Australian dairy cattle samples as well as 147 environmental samples in southeastern Queensland [68]. In these studies, 16.7% of cattle samples and 4.1% of environmental samples yielded STEC. It was also noteworthy that 10.2% of the cattle isolates were serotype O26:H11 and 11.2% were O157:H7 indicating estimates of the general prevalence of these two STEC serotypes of 1–2%. The authors of this study noted particularly that 1–14-week-old weanling calves may be the primary reservoir for STEC in dairy cattle. Examination of cattle specimens for the presence of STEC [69] in preslaughter animals revealed STEC and especially serogroup O157 were endemic in cattle; intermittent peaks in shedding were noted. It is suggested that these peaks should be seen as the main target for intervention to reduce incidence of the spread of these pathogens. Further studies on 136 of the STEC isolated earlier [68] were undertaken. Their serotypes as well as virulence factors and genomic relationships were determined [70]. Particular note was taken on the potential of these strains to cause human morbidity or mortality. With human pathogenic serotypes O157:H7 and O26:H11 being present as 9.4 and 7.8%, respectively as well as other serotypes such as O12:H11, O35:H-, O98:H-, O165:H- and O165:H25 being identified, the potential for the presence of human pathogens in such specimens is high. Of these strains, 19.9% produced shiga toxin 1 (stx1), 36.8% produced stx2 and 16.9% produced both toxins. Thus, of these strains, 73.6% were shigatoxigenic. Using pulsed-field gel electrophoresis (PFGE), isolates with common genomic profiles were identified. This study stresses the importance of examining cattle for the presence of these pathogens. Using calves experimentally colonized with a marker test STEC strain and each calf then retested under different rearing conditions showed a significant effect on carriage and shedding of the test STEC strain as well as other acquired strains [71].

A series of systematic investigations into the carriage of STEC and related organisms by food animals, especially cattle and sheep was undertaken. A study of ovine-derived STEC serotypes [72] demonstrated a number of genetic variants of the stx 2 gene to determine whether predominance occurred. It had been noted earlier that ovine-derived STEC generally belonged to a different group of serotypes than bovine-derived serotypes. The study [72] showed that the stx 2 gene subtypes were also different from cattle strains. The finding that ovine-derived STEC carry a different subtype of the stx 2 gene led to studies on the stx 1 gene [73] and characterized the gene as subtype stx 1c which was rarely found in cattle strains. The subtype stx 1c was recovered from human isolates but predominantly from those belonging to the typical ovine serotypes such as O128:H2 and O5:H-. It was not found among the typical bovine serotypes isolated from humans (such as O26:H11; O111:H-; O113:H21 and O157:H7). The bovine-derived non-O157 STEC were found to carry the stx 2 gene variants stx 2-EDL933 and/or stx 2vhb [74]. Thus, a double host specificity was suggested, in which ovine-derived STEC serotypes differ from bovine-derived ones, but both stx 1 and stx 2 genes, when carried by ovine-derived STEC serotypes, differ from those of bovine-derived ones. The importance of these findings in understanding the role of animal-derived STEC in human disease, therefore, justifies emphasis.

Further studies on ovine specimens collected from 65 distinct geographical locations [75] further confirmed the distinct ovine pattern of STEC serotypes which included types O91:H-; O5:H-; O128:H2; O123:H- and O85:H49. Of the 90 STEC isolates, serotype O157:H- was isolated twice. This indicates that although this serotype is present, it is comparatively rare, and although an infection due to an ovine source might yield this serotype, a number of others might very well predominate. It is noteworthy that these typical ovine serotypes have been isolated from human infections, although rarely. A general survey of STEC isolates of pasture-fed and lot-fed sheep [76] revealed that 86.6% of these ovine isolates were different from those isolated from cattle [67]. It was noted that STEC serotypes including O5:H-; O75:H8; O91:H-; O123:H- and O128:H2 appeared to be especially well adapted to colonize the ovine gastrointestinal tract, as they were the most common serotypes associated with these sheep.

Having isolated potentially pathogenic STEC serotypes from sick cattle [65,66], a more detailed analysis for STEC of bovine specimens submitted to the Regional Veterinary Laboratory in New South Wales was examined [77]. Overall, 18.7% of the faecal samples yielded STEC, and included serotypes O5:H-; O26:H-; O26:H11; O91:H21; O104:H11; O111:H-; O111:H8; O113:H21 and notably O157:H8. This study confirmed that when meticulous microbiological techniques are employed, then STEC will be isolated from such specimens and that these STEC comprise a wide range of serotypes, many of which have also been associated with human disease. The finding of O157 STEC with an unusual H antigen (H8) is noted. These isolates were biochemically quite distinct from the typical O157:H7/H- clones, which have been isolated from human infections around the world. During that period, specimens from a patient with watery diarrhoea and HUS yielded sorbitol-fermenting strains of STEC O157:H- [78]; this report was the first outside Europe [14]. At the same time, detailed studies were undertaken to determine whether there were differences between human and animal isolates of STEC O157 from Australia [79]. A group of Australian isolates of STEC belonging to serotypes O157:H7 and O157:H- isolated from both human and animal sources were examined for the presence of virulence factors and compared by XbaI DNA macrorestriction analysis using PFGE. All the strains were eae-positive and only one of the 102 strains lacked the ability to produce enterohaemolysin. The results showed that the most common virulence gene carried by these isolates was stx 2c, which was present alone in 16% or in combination with stx 1 in 74% or stx 2 in 3%. As both the human and animal strains of Australian STEC O157 isolates appear to carry the same virulence factors, the authors speculate that ‘other factors must be responsible for the low rates of human infection in Australia’. However, another explanation may be that in Australia, the awareness of STEC serotypes other than O157 is very high and, therefore, these are isolated more commonly from human disease than in other countries and, therefore, the ratio of STEC O157 to other STEC isolated from humans is much lower than elsewhere.

It had been demonstrated that the shiga toxins of STEC in some conditions require to be activated by the intestinal mucus and this was demonstrated in mice and humans [80]. In a study in Queensland, the presence of this ‘activatable’ shiga toxin genotype (stx 2d) [81] was detected in STEC isolated from beef cattle faeces, beef, lamb meat as well as dairy farm environmental sources. The serotypes included O1:H20; O2:H29; O8:H19; O174:H8 and O174:H21 as well as some belonging to untypeable O serogroups. Although these STEC serotypes have been reported from human infections including HUS outside Australia, though rarely, they have not been described to date from Australia.

Shedding of STEC by healthy cattle has been studied [82] and a range of 70 serotypes were isolated, including O3:H7; O26:H11; O104:H7; O113:H21; O116:H21 as well as a number of strains whose O antigen was untypeable with the range of O antisera available. In order to assess whether and to what extent antibiotic resistance played a role in the distribution of STEC in cattle and sheep, isolates from these animals as well as recent human isolates were tested [83]; non-STEC were also included in these studies. It was noted that resistant STEC were predominantly isolated from specimens associated with sick animals. A far higher level of resistance was found among the non-STEC than among the STEC. Of 87 strains of non-STEC isolated from healthy infants, who had neither contact with antibiotics nor had gastrointestinal symptoms, 24 (28%) were resistant to one or more antibiotics. In another study on the colonization of the bovine colonic mucosa by STEC [84], in-vitro techniques were applied to determine the conditions under which the colon of cattle is colonized by STEC. Fasted, poorly or only intermittently fed cattle and preruminant calves are more likely to excrete STEC than well-fed ruminant animals. An attempt to investigate the reasons for these differences noted no differences in colonization susceptibilities between tissues derived from weanling calves and adult cattle [85]. The studies also showed that under conditions mimicking the well-fed ruminant diet, significantly less STEC colonized the mucosal surface of colonic biopsies than under conditions of poor feeding. These studies were extended in an attempt to enumerate E. coli O157 in cattle faeces [86]. By applying the most probable number (MPN) technique, it was concluded that the E. coli O157 only made up a small proportion of the total E. coli population, being between undetectable to 2.4 × 104 MPN g−1. An extension of these studies examined the prevalence and concentration of E. coli O157 in the faeces of cattle at slaughter, and subjected to a different production system [86]. Similar methods to those in the earlier study were applied [86] and a similar prevalence was noted. In addition, there was no demonstrable effect on the prevalence of E. coli O157 strains of the different production systems, including grass (pasture) fed as well as lot-fed (feedlot) cattle. As long ago as 1997, it was speculated that the reason for these disparities in STEC colonization may lie with the ecological role played by STEC in the digestive system of ruminants [87]. In the well-fed animal, rumen protozoa maintain the levels of the bacteria involved in the digestive process at their most active logarithmic phase and the STEC similarly maintain the protozoa. However, in low feeding, the digestive bacteria are reduced, but STEC, which can kill the protozoa, have, thus, gained a selective advantage and multiply. When an animal carrying STEC enters the slaughtering process, then contamination of other carcasses is highly likely. However, abattoirs maintaining good practices limit cross-contamination to a minimum [88]. Despite stringent meat production rules being applied in Australia, STEC do enter the food chain as shown by a 1-year study in which ground beef and lamb cuts were regularly monitored [89]. Although a great variety of serotypes were represented among the isolates, including ones associated with human disease in Australia and/or globally, the most prominent serogroups O157, O111 and O26 were not found.

All these studies on cattle and sheep used techniques to select specifically for STEC; however, it was considered appropriate as part of a study on the effects of different feeding regimens to select E. coli-like colonies using nonselective media and to, thus, determine the diversity of the E. coli serotypes present in cattle faeces [90]. Diverse serotypes were found and a number of strains were shown to be STEC. One faecal specimen, which yielded strains of O2:H29, included STEC and non-STEC of this same serotype.’ This phenomenon suggested that the potential STEC may be present, which under certain circumstances can acquire Stx genes to become STEC. Such observations indicated that a more detailed analysis should be undertaken to look for E. coli-like pathogens among bovine faecal specimens taken from animals with gastrointestinal infections [91]. A number of typical bovine STEC serotypes were isolated and of the non-STEC intimin-producing E. coli, the two most common serotypes were O111:H- and O177:H11. In addition to intimin, these strains also produced enterohaemolysin, typical of many STEC. It was noted earlier that although STEC O111:H- strains were the most common cause of STEC infections in Australia, they are very rarely isolated from cattle [65], which led to the speculation that the STEC serotype O111:H- may be present in the animal population predominantly in the nontoxigenic form and only acquired toxigenicity following human infection or in poorly maintained cattle from whom the first bovine isolation of STEC O111:H- in Australia was made [65]. At that time also two clusters of STEC infections in South Australia were reported [92]; of these clusters, one yielded O157 strains and the other O111 strains.

A further analysis was performed on a selection of STEC strains derived from human infections and domestic food animals [93], in which the varieties of intimin subtypes were examined. These studies showed that intimin subtype β was the one most commonly found in STEC isolated from human infections as well as cattle and sheep. Subtype ζ, although relatively common in bovine and ovine isolates, was rarely associated with human infections, whereas subtype ε (relatively common in sheep) was rare in cattle and humans, and subtype θ was common in bovine isolates but not the others. These studies further confirm the diversity of the STEC population.

These detailed studies, although interesting and important in gaining an understanding of the ecology and dynamics of E. coli in animals and human disease, could be construed to be too academic to be of much use in the field or clinic; however, a 2005 case highlighted the importance of understanding this background. A strain of E. coli O5:H-, which possessed the genes for stx 1, intimin-β and enterohaemolysin was isolated from an adult with bloody diarrhoea [94]. The strain had the additional unusual feature of being able to split urea. When it was compared to strains of the same serotype isolated from animals as well as humans, especially the case of an infant with HUS [48], it was noted that this serotype comprises two distinct clones. One clone (the ovine clone) produces toxins stx 1c /stx 1 and stx 2c, and enterohaemolysin, and is commonly found in healthy sheep and causes diarrhoea and HUS in humans, whereas the other (bovine) clone only produces stx 1, enterohaemolysin and intimin-β, splits urea, and causes diarrhoea in calves and diarrhoea in humans. Whether the bovine clone can cause human HUS is at present uncertain.

This review clearly demonstrated that detailed analysis is an essential prerequisite for the understanding of the ecology and pathogenicity of a group of microorganisms that is as diverse as the E. coli group. A recent report of a family outbreak [95] highlighted this situation by showing that a diverse collection of STEC was isolated. The father, was the only member of the family with symptoms. His faeces yielded a rough STEC. Other family members also yielded STEC of serogroup O128 and a strain of intimin producing non-STEC belonging to serogroup O106. The role of these other potential pathogens could not be elucidated.

In accord with many infectious agents, changes in epidemiology occur over time and it is, therefore, expected that the predominant serotypes observed in the future will differ from those observed today. Constant monitoring is a prerequisite for protecting our food sources and human health [96] and, without question, STEC are major pathogens causing severe disease, and as such they have an economic impact. It was recently estimated [97] that the mean cost for Australia is AUD$3132 (US$3119) per case and that annually it is AUD$2 633 181 (US$2 622 440). This may well be an underestimate, as many minor infections go unnoticed, unreported and undiagnosed.

Back to Top | Article Outline


The authors of this review have often been asked, whether Australia is unique in having found and reported so many different serotypes of STEC and relegated the universally accepted serotype O157:H7/- into the background. Although no definitive answer to this question can be given, a review [59], not by the authors, highlighted the situation within Australia by showing that in 2002, 1665 specimens were tested in South Australia, whereas far fewer specimens were tested in the other states. It was, therefore, not surprising that there were 2.58 notifications of STEC per 100 000 population compared to 0.1–0.16/100 000 in the other states. The laboratory in South Australia was more aware of the importance of non-O157 STEC, especially STEC belonging to serogroup O111 [37–39] and again it is, therefore, not surprising that it is this state that reports about 20 times more STEC than the other states. If these figures can be expanded to the global picture, then again non-O157 STEC are found where they are sought. E. coli are organisms that continue to surprise us with their potential as was shown by the recent outbreak of strains belonging to the ‘rare serotype’ O104:H4 in Europe [33], and we should not forget that 30 years ago the STEC serotype O157:H7 was also called a ‘rare serotype’ [8] when it was first reported. It is only by awareness and not concentrating on only one or a few STEC serotypes that the infection rate due to these organisms can be effectively addressed and reduced.

Back to Top | Article Outline


Conflicts of interest

There are no conflicts of interest.

Funding: none declared.

Back to Top | Article Outline


1. Konowalchuk J, Speirs JL, Stavric S. Vero response to a cytotoxin of Escherichia coli. Infect Immun 1977; 18:775–779.
2. Conradi H. Concerning soluble toxins obtained through aseptic techniques from dysentery and typhoid bacilli. Dtsche Med Wchschr 1903; 29:26–28.
3. Neisser M, Shiga K. Concerning free receptors of typhoid and dysentery bacilli and about the dysentery toxin. Dtsche Med Wchschr 1903; 29:61–62.
4. von Gasser C, Gautier E, Steck A, Siebenmann RE, Oechslin R. Haemolytic uraemic syndrome: bilateral kidney cortex necrosis through acutely acquired haemolytic anaemia. Schweiz Med Wchschr 1955; 85:905–909.
5. Wade WG, Thom BT, Evans BT. Cytotoxic enteropathogenic Escherichia coli. Lancet 1979; ii:1235–1236.
6. Wilson MW, Bettelheim KA. Cytotoxic Escherichia coli serotypes. Lancet 1980; i:201.
7. Belnap WD, O’Donnell JJ. Epidemic gastroenteritis due to Escherichia coli O-111. J Pediatr 1955; 47:178–193.
8. Riley LW, Remis RS, Helgerson SD, McGee HB, Wells GJ, Davis BR, et al. Hemorrhagic colitis associated with a rare Escherichia coli serotype. N Engl J Med 1983; 308:681–685.
9. Ørskov F, Ørskov I, Villar JA. Cattle as a reservoir of verotoxin producing Escherichia coli O157:H7. Lancet 1987; ii:276.
10. March SB, Ratnam S. Sorbitol-MacConkey medium for detection of Escherichia coli O157:H7 associated with haemorrhagic colitis. J Clin Microbiol 1986; 23:869–872.
11. Bettelheim KA. Studies of Escherichia coli cultured on Rainbow agar O157 with particular reference to enterohaemorrhagic Escherichia coli (EHEC). Microbiol Immunol 1998; 42:265–269.
12. Bettelheim KA. Reliability CHROMagar for the detection of enterohaemorrhagic Escherichia coli (EHEC) O157 but not EHEC belonging to other serogroups. J Appl Microbiol 1998; 85:425–428.
13. Bettelheim KA. Reliability of O157:H7 ID agar (O157 H7 ID-F) for the detection and isolation of verocytotoxigenic strains of Escherichia coli belonging to serogroup O157. J Appl Microbiol 2005; 99:408–410.
14. Karch H, Bielaszewska M. Sorbitol-fermenting shiga toxin producing Escherichia coli O157:H_ strains: epidemiology, phenotypic and molecular characteristics, and microbiological diagnosis. J Clin Microbiol 2001; 39:2043–2049.
15. O’Brien AD, LaVeck GD, Thompson MR, Formal SB. Production of Shigella dysenteriae type 1-like cytotoxin by Escherichia coli. J Infect Dis 1982; 146:763–769.
16. Strockbine NA, Marques LRM, Newland JW, Smith HW, Holmes RK, O’Brien AD. Two toxin-converting phages from Escherichia coli O157:H7 strain 933 encode antigenically distinct toxins with similar biological activities. Infect Immun 1986; 53:135–140.
17. O’Brien AD, Holmes RK. Shiga and shiga-like toxins. Microbiol Rev 1987; 51:206–220.
18. Scotland SM, Rowe B, Smith HR, Willshaw GA, Gross RJ. Vero cytotoxin-producing strains of Escherichia coli from children with haemolytic uraemic syndrome and their detection by specific DNA probes. J Med Microbiol 1988; 25:237–243.
19. Dorn CR, Scotland SM, Smith HR, Willshaw GA, Rowe B. Properties of verocytotoxin-producing Escherichia coli of human and animal origin belonging to serogroups other than O157:H7. Epidemiol Infect 1989; 103:83–95.
20. Caprioli A, Luzzi I, Rosmini F, Resti C, Edefonti A, Perfumo F, et al. Community-wide outbreak of hemolytic-uremic syndrome associated with non-O157 verocytotoxin-producing Escherichia coli. J Infect Dis 1994; 169:208–211.
21. Acheson DWK, Keusch GT, Lightowlers M, Donohue-Rolfe A. Enzyme-linked immunosorbent assay for shiga toxin and shiga-like toxin II using P1 glycoprotein from hydatid cysts. J Infect Dis 1990; 161:134–137.
22. Bettelheim KA. Development of a rapid method for the detection of verocytotoxin-producing Escherichia coli (VTEC). Lett Appl Microbiol 2001; 33:31–35.
23. Bastian SN, Carle I, Grimont F. Comparison of 14 PCR systems for the detection and subtyping of stx genes in shiga-toxin-producing Escherichia coli. Res Microbiol 1998; 149:457–472.
24. Paton AW, Paton JC, Goldwater PN, Manning PA. Direct detection of Escherichia coli shiga-like toxin genes in primary fecal cultures by polymerase chain reaction. J Clin Microbiol 1993; 31:3063–3067.
25. Fagan PK, Hornitzky MA, Bettelheim KA, Djordjevic SP. Detection of shiga-like toxin (stx1 and stx2), intimin (eaeA), and enterohemorhagic Escherichia coli (EHEC) hemolysin EHEC hlyA genes in animal feces by multiplex PCR. Appl Environ Microbiol 1999; 65:868–872.
26. Morabito S, Karch H, Mariani-Kurkdjian P, Schmidt H, Minelli F, Bingen E, Caprioli A. Enteroaggregative, shiga toxin-producing Escherichia coli O111:H2 associated with an outbreak of hemolyticuremic syndrome. J Clin Microbiol 1998; 36:840–842.
27. Mariani-Kurkdjian P, Denamur E, Milon A, Picard B, Cave H, Lambert-Zechovsky N, et al. Identification of a clone of Escherichia coli O103:H2 as a potential agent of hemolytic uremic syndrome in France. J Clin Microbiol 1993; 31:296–301.
28. Tarr PI, Fouser LS, Stapleton AE, Wilson RA, Kim HH, Vary JC, Clausen CR. Hemolytic-uremic syndrome in a six-year-old girl after a urinary tract infection with shiga-toxin-producing Escherichia coli O103:H2. N Engl J Med 1996; 335:635–638.
29. Karch H, Geitz C, Schmidt H. Increased incidence of infections with EHEC O103:H2. Notiziaro dell’Istituto Superiore di Sanita 1997; 10:2.
30. Kudoh Y, Kai A, Obata H, Kusunoki J, Monma C, Shingaki M, et al. Epidemiological surveys on verocytotoxin-producing Escherichia coli infections in Japan. In: Karmali MA, Goglio AG, editors. Recent advances in verocytotoxin-producing Escherichia coli infections (Excerpta Medica International Congress Series 1072). Amsterdam, The Netherlands: Elsevier Science; 1994. pp. 53–56.
31. Centers for Disease Control and Prevention. Outbreak of acute gastroenteritis attributable to Escherichia coli serotype O104:H21: Helena, Montana, 1994. Morbid Mortal Wkly Rep 1995; 44:501–503.
32. Bergmire-Sweat D, Marengo L, Pendergrass P, Hendricks K, Garcia M, Drumgoole R, et al. Escherichia coli O111:H8 outbreak among teenage campers: Texas, 1999. J Am Med Assoc 2000; 283:2517.
33. Beutin L, Martin A. Outbreak of shiga toxin-producing Escherichia coli (STEC) O104:H4 infection in Germany causes a paradigm shift with regard to human pathogenicity of STEC strains. J Food Protect 2012; 75:408–418.
34. Gunzburg S, Gracey M, Forbes D, Hewitt I, Bettelheim KA. Haemolytic-uraemic syndrome and verotoxigenic Esch. coli. Med J Australia 1988; 149:54–55.
35. Pryor W, Hodson H, McIntyre P, Bettelheim KA. Toxigenic Escherichia coli in uraemic haemolytic syndrome. Med J Australia 1990; 152:221–222.
36. Albert MJ, Bettelheim KA. Enterohaemorrhagic E. coli O157:H7 in central Australia. J Diarrhoeal Dis Res 1989; 7:96–97.
37. Goldwater PN, Bettelheim KA. The role of enterohaemorrhagic E. coli serotypes other than O157:H7 as causes of disease. In: Karmali MA, Goglio AG, editors. Recent advances in verocytotoxin-producing Escherichia coli infections (Excerpta Medica International Congress Series 1072). Amsterdam, The Netherlands: Elsevier Science; 1964. pp. 57–60.
38. Goldwater PN, Bettelheim KA. The role of enterohaemorrhagic Escherichia coli serotypes other than O157:H7 as causes of disease in Australia. Comm Dis Intell 1995; 19:2–4.
39. Paton AW, Paton JC. Direct detection of shiga toxigenic Escherichia coli strains belonging to serogroups O111, O157, and O113 by multiplex PCR. J Clin Microbiol 1999; 37:3362–3365.
40. Goldwater PN, Bettelheim KA. An outbreak of hemolytic uremic syndrome due to Escherichia coli O157:H7: or was it? Emerg Infect Dis 1996; 2:153–154.
41. Goldwater PN, Bettelheim KA. Escherichia coli O group serology of a haemolytic uraemic syndrome (HUS) epidemic. Scand J Infect Dis 2000; 32:385–394.
42. Kulkarni H, Goldwater PN, Martin A, Bettelheim KA. Escherichia coli ‘O’ group serological responses and clinical correlations in epidemic HUS patients. Comp Immunol Microbiol Infect Dis 2002; 25:249–268.
43. Goldwater PN, Bettelheim KA. Hemolytic uremic syndrome due to shiga-like toxin producing Escherichia coli O48:H21 in South Australia. Emerg Infect Dis 1995; 1:132–133.
44. Strockbine N, Sowers E, Greene K, Hayes P, Griffin P, Wells J. Characterization of shiga toxin-producing non-O157 Escherichia coli from the United States 1983–1997. VTEC ’97, 3rd International Symposium and Workshop on Shiga Toxin (Verocytotoxin)-producing Escherichia coli infections [abstract V235/III]; 1997. p.67.
45. Vally H, Hall G, Dyda A, Raupach J, Knope K, Combs B, Desmarchelier P. Epidemiology of shiga toxin producing Escherichia coli in Australia, 2000–2010. BMC Pub Hlth 2012; 12:63.
46. Manandhar R, Bettiol SS, Bettelheim KA, Goldsmid JM. Isolation of verocytotoxigenic Escherichia coli from the Tasmanian environment. Comp Immunol Microbiol Infect Dis 1997; 20:271–279.
47. Sidjabat-Tambunan H, Bensink JC, Bettelheim KA. Isolation of verocytotoxin-producing Escherichia coli from mutton carcases. Austral Vet J 1998; 76:364–365.
48. Starr M, Bennettwood V, Bigham AK, Dekoningward TP, Bordun AM, Lightfoot D, et al. Hemolytic-uremic syndrome following urinary tract infection with enterohemorrhagic Escherichia coli: case report and review. Clin Infect Dis 1998; 27:310–315.
49. Goldwater PN, Bettelheim KA. New perspectives on the role of Escherichia coli and other enterohaemorrhagic E. coli serotypes in human disease. J Med Microbiol 1998; 47:1039–1045.
50. Goldwater PN, Giles N, Bettelheim KA. An unusual case of microangiopathic haemolytic anaemia associated with enterohaemorrhagic Escherichia coli O113: H21 infection, a verocytotoxin-2/shiga toxin-2 producing serotype. J Infect 1998; 37:302–304.
51. Bettelheim KA. A spectrum of verocytotoxigenic Escherichia coli. J Med Microbiol 1998; 47:1037–1038.
52. Bettelheim KA. Role of non-O157 VTEC. J Appl Microbiol 2000; 88 (Symposium Supplement):38S–50S.
53. Fagan PK, Hornitzky MA, Bettelheim KA, Djordjevic SP. Detection of shiga-like toxin (stx1 and stx2), intimin (eaeA), and enterohemorhagic Escherichia coli (EHEC) hemolysin (EHEC hlyA) genes in animal feces by multiplex PCR. Appl Environm Microbiol 1999; 65:868–872.
54. Bettelheim KA. Enterohaemorrhagic Escherichia coli O157:H7: a red herring? J Med Microbiol 2001; 50:201–202.
55. Morabito S, Karch H, Schmidt H, Minelli F, Mariani-Kurkdjian P, Allerberger F, et al. Molecular characterisation of verocytotoxin-producing Escherichia coli of serogroup O111 from different countries. J Med Microbiol 1999; 48:891–896.
56. Bettelheim KA, Bowden DS, Doultree J, Catton MG, Chibo D, Ryan NJP, et al. Combined infection of Norwalk-like virus and verotoxin-producing bacteria associated with a gastroenteritis outbreak. J Diarrh Dis Res 1999; 17:34–36.
57. Marshall JA, Yuen LKW, Catton MG, Gunesekere IC, Wright PJ, Bettelheim KA, et al. Multiple outbreaks of Norwalk-like virus gastroenteritis associated with a Mediterranean-style restaurant. J Med Microbiol 2001; 50:143–151.
58. Bettelheim KA, Bennett-Wood V, Lightfoot D, Wright PJ, Marshall JA. Simultaneous isolation of verotoxin-producing strains of Escherichia coli O128:H2 and viruses in gastroenteritis outbreaks. Comp Immunol Microbiol Infect Dis 2001; 24:135–142.
59. Combs BG, Raupach JC, Kirk MD. Surveillance of shiga toxigenic Escherichia coli in Australia. Commun Dis Intell 2005; 29:366–369.
60. Thomas MK, Majowicz SE, Sockett PN, Fazil A, Pollari F, Dore K, et al. Estimated numbers of community cases of illness due to salmonella, campylobacter and verotoxigenic Escherichia coli: pathogen-specific community rates. Can J Infect Dis Med Microbiol 2006; 17:229–234.
61. Desmarchelier PM. Enterohemorrhagic Escherichia coli: the Australian perspective. J Food Protect 1997; 60:1447–1450.
62. Desmarchelier PM, Bilge SS, Fegan N, Mills L, Vary JC Jr, Tarr P. A PCR specific for Escherichia coli O157 based on the rfb locus encoding O157 polysaccharide. J Clin Microbiol 1998; 36:1801–1804.
63. Bettelheim KA, Bensink JC, Sidjabat-Tambunan H. Serotypes of verocytotoxin-producing (Shiga-toxin-producing) Escherichia coli isolated from healthy sheep. Comp Immunol Microbiol Infect Dis 2000; 23:1–7.
64. Fegan N, Desmarchelier P. Shiga toxin-producing Escherichia coli in sheep and preslaughter lambs in eastern Australia. Lett Appl Microbiol 1999; 28:335–339.
65. Hornitzky MA, Bettelheim KA, Djordjevic SP. The isolation of Enterohaemorrhagic Escherichia coli O111:H- from Australian cattle. Austral Vet J 2000; 78:636–637.
66. Bettelheim KA, Hornitzky MA, Djordjevic SP. First bovine and ovine isolations of shiga toxin-producing Escherichia coli O103:H2 in Australia. Austr Vet J 2001; 79:289–290.
67. Midgley J, Fegan N, Desmarchelier P. Dynamics of shiga toxin-producing Escherichia coli (STEC) in feedlot cattle. Lett Appl Microbiol 1999; 29:85–89.
68. Cobbold R, Desmarchelier P. A longitudinal study of shiga-toxigenic Escherichia coli (STEC) prevalence in three Australian dairy herds. Vet Microbiol 2000; 71:125–137.
69. Midgley J, Desmarchelier P. Preslaughter handling of cattle and shiga toxin-producing Escherichia coli (STEC). Lett Appl Microbiol 2001; 32:307–311.
70. Cobbold R, Desmarchelier P. Characterisation and clonal relationships of shiga-toxigenic Escherichia coli (STEC) isolated from Australian dairy cattle. Veter Microbiol 2001; 79:323–335.
71. Cobbold R, Desmarchelier P. Horizontal transmission of shiga toxin-producing Escherichia coli within groups of dairy calves. Appl Environ Microbiol 2002; 68:4148–4152.
72. Ramachandran V, Hornitzky MA, Bettelheim KA, Walker MJ, Djordjevic SP. The common ovine shiga toxin 2-containing Escherichia coli serotypes and human isolates of the same serotypes possess a Stx2d toxin type. J Clin Microbiol 2001; 39:1932–1937.
73. Brett KN, Ramachandran V, Hornitzky MA, Bettelheim KA, Walker K, Djordjevic SP. stx 1c is the most common shiga toxin 1 subtype among shiga toxin-producing Escherichia coli isolates from sheep but not among isolates from cattle. J Clin Microbiol 2003; 41:926–936.
74. Brett KN, Hornitzky MA, Bettelheim KA, Walker K, Djordjevic SP. Bovine non-O157 shiga toxin 2-containing Escherichia coli isolates commonly possess stx 2-EDL933 and/or stx 2vhb subtypes. J Clin Microbiol 2003; 41:2716–2722.
75. Djordjevic SP, Hornitzky MA, Bailey PG, Vanselow B, Walker K, Bettelheim KA. Virulence properties and serotypes of shiga toxin-producing Escherichia coli from healthy Australian slaughter-age sheep. J Clin Microbiol 2001; 39:2017–2021.
76. Djordjevic SP, Ramachandran V, Bettelheim KA, Vanselow BA, Holst P, Bailey G, Hornitzky MA. Serotypes and virulence gene profiles of shiga toxin-producing Escherichia coli strains isolated from feces of pasture-fed and lot-fed sheep. Appl Environ Microbiol 2004; 70:3910–3917.
77. Hornitzky MA, Bettelheim KA, Djordjevic SP. The detection of shiga toxin-producing Escherichia coli in diagnostic bovine faecal samples using vancomycin-cefixime-cefsulodin blood agar and PCR. FEMS Microbiol Lett 2001; 198:17–22.
78. Bettelheim KA, Whipp M, Djordjevic SP, Ramachandran V. First isolation outside Europe of sorbitol-fermenting Escherichia coli (VTEC) belonging to O group O157. J Med Microbiol 2002; 51:713–714.
79. Fegan N, Desmarchelier P. Comparison between human and animal isolates of shiga toxin-producing Escherichia coli O157 from Australia. Epidemiol Infect 2002; 128:357–362.
80. Melton-Celsa AR, Darnell SC, O’Brien AD. Activation of shiga-like toxins by mouse and human intestinal mucus correlates with virulence of enterohemorrhagic Escherichia coli O91:H21 isolates in orally infected, streptomycin-treated mice. Infect Immun 1996; 64:1569–1576.
81. Gobius KS, Higgs GM, Desmarchelier PM. Presence of activatable shiga toxin genotype (stx 2d) in shiga toxigenic Escherichia coli from livestock sources. J Clin Microbiol 2003; 41:3777–3783.
82. Hornitzky MA, Vanselow BA, Walker K, Bettelheim KA, Corney B, Gill P, et al. Virulence properties and serotypes of shiga toxin-producing Escherichia coli from Australian cattle. Appl Environ Microbiol 2002; 68:6439–6445.
83. Bettelheim KA, Hornitzky MA, Djordjevic SP, Kuzevski A. Antibiotic resistance among verocytotoxigenic Escherichia coli (VTEC) and non-VTEC isolated from domestic animals and humans. J Med Microbiol 2003; 52:155–162.
84. Cobbold RN, Desmarchelier PM. In vitro studies on the colonization of bovine colonic mucosa by shiga-toxigenic Escherichia coli (STEC). Epidemiol Infect 2004; 132:87–94.
85. Fegan N, Higgs G, Vanderlinde P, Desmarchelier P. Enumeration of Escherichia coli O157 in cattle faeces using most probable number technique and automated immunomagnetic separation. Lett Appl Microbiol 2004; 38:56–59.
86. Fegan N, Vanderlinde P, Higgs G, Desmarchelier P. The prevalence and concentration of Escherichia coli O157 in faeces of cattle from different production systems at slaughter. J Appl Microbiol 2004; 97:362–370.
87. Bettelheim KA. Ecology of verotoxigenic Escherichia coli. Austral Vet J 1997; 75:912.
88. Fegan N, Higgs G, Vanderlinde P, Desmarchelier P. An investigation of Escherichia coli O157 contamination of cattle during slaughter at an abattoir. J Food Protect 2005; 68:451–457.
89. Barlow R, Gobius KS, Desmarchelier PM. Shiga toxin-producing Escherichia coli in ground beef and lamb cuts: results of a one-year study. Int J Food Microbiol 2006; 111:1–5.
90. Bettelheim KA, Kuzevski A, Gilbert RA, Krause D, McSweeney CS. The diversity of Escherichia coli serotypes and biotypes in cattle faeces. J App Microbiol 2005; 98:699–709.
91. Hornitzky MA, Mercieca K, Bettelheim KA, Djordjevic SP. Bovine feces from animals with gastrointestinal infections are a source of serologically diverse atypical enteropathogenic Escherichia coli and shiga toxin-producing E. coli strains that commonly possess intimin. Appl Environ Microbiol 2005; 71:3405–3412.
92. Combs BG, Wise RP, Tribe IG, Mwanri L, Raupach JA. Investigation of two clusters of shiga toxin-producing Escherichia coli cases in South Australia. Commun Dis Intell 2003; 27:517–519.
93. Ramachandran V, Brett KN, Hornitzky MA, Dowton M, Bettelheim KA, Walker K, Djordjevic SP. Distribution of intimin subtypes among Escherichia coli isolates from ruminant and human sources. J Clin Microbiol 2003; 41:5022–5032.
94. McLean C, Bettelheim KA, Kuzevski A, Falconer L, Djordjevic SP. Isolation of Escherichia coli O5:H-, possessing genes for shiga toxin 1, intimin-β and enterohaemolysin, from an intestinal biopsy from an adult case of bloody diarrhea: evidence for two distinct O5:H- pathotypes. J Med Microbiol 2005; 54:605–607.
95. Staples M, Graham RMA, Doyle CJ, Smith HV, Jennison AV. Prolonged and mixed non-O157 Escherichia coli infection in an Australian household. Clin Microbiol Infect 2012; 18:E140–E143.
96. Beutin L. Emerging enterohaemorrhagic Escherichia coli, causes and effects of the rise of a human pathogen. J Vet Med B Infect Dis Vet Public Health 2006; 53:299–305.
97. McPherson M, Kirk MD, Raupach J, Combs B, Butler JRG. Economic costs of shiga toxin-producing Escherichia coli infection in Australia. Foodborne Path Dis 2011; 8:55–62.

Australia; Escherichia coli; human infections; nosocomial; outbreaks; serotyping; shiga toxins; verotoxins

© 2013 Wolters Kluwer Health | Lippincott Williams & Wilkins