For some time now, the worldwide food supply has been under attack by a bacterium we know well as a generally non-trouble-making resident of the human colon (Escherichia coli), which has lately been possessed of a terrible “new” weapon in the form of Shiga toxin. We call this beast “Shiga toxin-producing E. coli” (STEC). We wonder where it came from and why it was sent to plague us.
However, E. coli has its problems too. It serves as prey to a number of viruses (bacteriophages). Most bacteriophages use the E. coli cell to make new virus particles, killing the infected bacterium in the process. This is called a lytic cycle. There is another, much more sinister pattern of bacteriophage infection termed a “lysogenic” cycle. Once inside the host cell, lysogenic viral DNA can integrate itself within the chromosome of the host and stay there, dividing whenever the bacterial chromosome divides.
To see a video of bacteriophage attack upon a host bacterial cell, click here.
To see the lysogenic and lytic cycles compared, click here.
People may prefer to measure the prevalence of STEC rather than the toxin carried by STEC. Such a result could be arrived at using relatively simple and inexpensive tools with a theoretical analytical sensitivity in the range of parts per trillion. By comparison, a chemical or biochemical test run at an analytical sensitivity of parts per trillion, usually involves a roomful of equipment that is expensive to acquire, expensive to maintain and operated by highly-trained personnel who want not just a good salary, health care, vacations and coffee breaks, but who also want to be loved.
Several investigators engaged in the study of STEC have remarked that they did not find this organism to be particularly invasive. This finding is consistent with the notion that enterohemorraghic colitis and hemolytic uremic syndrome is the result of the toxicant and not of the “pathogen” carrying it.
The premise that STECs are not so much pathogens as carriers for a very potent toxin and that therefore they are not so much examples of infectious diseases as they are of a potent colonic toxicosis (which develops later into a toxemia) would be more convincing if evidence were in hand showing that these disease states could be attained without the detectable presence of STEC at all. Such evidence exists. For example, the first instance of an E. coli O157 outbreak in Denmark was actually attributed to milk coming from a dairy where it had been pasteurized . This report recalls a previous episode from the UK where the drinking of pasteurized milk resulted in intoxication with Shiga toxin, even though microbiological testing of the milk involved had shown that the milk had almost no live microorganisms left in it .
Both of these papers appeared to suggest that the pasteurization process at issue was somehow defective even though no bacterial contamination of the milk was found in either case. Years later, it was discovered that Shiga toxin is sufficiently heat stable to survive the usual rigors of pasteurization . If we know that elimination of the presumed infectious agent does not eliminate the disease, then why should food HACCP plans and clinical management personnel assume that an infectious disease is the issue?
A recent paper, “Mouse model of hemolytic-uremic syndrome caused by endotoxin-free Shiga toxin 2 (Stx2) and protection from lethal outcome by anti-Stx2 antibody,” by Sauter et. al  showed that hemolytic uremic syndrome (HUS) is clearly the result of an intoxication with Shiga toxin as were the previously mentioned two “experiments of nature,” published three years prior to the date of this review, which describes two critically-important phenomena: (i) hemolytic uremic syndrome can be induced by Shiga toxin alone (no STEC needed) and (ii) passive immunization with antitoxin given early enough in the cycle can cure the disease so many previous authorshave said is incurable.
The phenomena observed by Sauter et. al by no means constitute the first instance of HUS induction by Shiga toxin alone. Nevertheless, this paper and others of its kind are not merely some arcane addendum to the story of STEC “infection.” They are, in fact, the tail that wagged the dog. They tell us, with convincing clarity, that HUS may be contracted without the help of any infectious agent at all. All that is necessary is the toxin, although other bacterial components (such as the lipopolysaccharide – LPS) may either increase or suppress the activity of the toxin depending upon the details of their administration.
Sauter et al. found that a one nanogram dose, given at day zero, day three and day six was sufficient to induce HUS in mice on day eight (and without showing signs of neurological damage). Now a mouse only weighs about twenty grams. So, the level of the Stx2 intoxicant was actually quite high (50 nanograms per kilogram repeated three times). It was found that a single five nanogram dose killed the mice outright, but not from HUS. Rather, the mice died from neurological damage. Because of the way the dose was administered, these results could only be possible if Shiga toxin was able to, unaided, cross the peritoneal membrane to penetrate the blood (from where it could easily cause HUS). Such results also suggested that Shiga toxin was able to quite effectively move across the blood-brain barrier to cause its lethal effects on brain function.
The “conventional” view held by the U.S. Department of Agriculture and its food safety arm, the Food Safety and Inpection Service, as deduced from actions they have taken and the statements they have made, is that a bloody stool followed by HUS is the result of an infectious disease caused by STEC. If one accepts that view, then one is obliged to consider all the STEC variants as agents of infectious disease (adulterants) and seek to identify and control their living presence in meat and other foodstuffs. Holding this viewpoint is inconsistent with reports that HUS may be caused without the benefit of STEC at all. Those who hold the infectious disease viewpoint must pay attention to and explain how these observations square with the concept of STEC as an infectious agent. These reports, published in peer-reviewed literature, should be neither ignored nor dismissed.
The other viewpoint is that STEC is a relatively noninvasive enteric bacterium, which has lately been hijacked into carrying a very dangerous toxin. This toxin is able to cause the observed sequelae (including HUS) all by itself, which must mean that STEC, while certainly a carrier of this toxin, is not a disease-causing agent as such. STEC merely carries its stx1 and/or stx2 containing prophage payload into the colon, where it may even reproduce somewhat.
However, any STEC’s most salient role is as the victim of the deadly prophage it carries. When that prophage goes lytic, i.e. reproduces as a separate entity within the cell it not only destroys most of the population of its carriers but, in the process, generates the release of sufficient quantities of a deadly poison to the environment where the carriers once resided. The sequelae of that release are bloody stool, the result of a toxicosis of the colon, which then may (or may not) lead to a general toxemia (of the blood).
Once that toxin is in the blood, it attacks the kidney merely beca use the cells there happen to be much more sensitive to this toxin than other organ tissues in the human body. We recognize this result as HUS. I am unaware of even one report making the claim that any STEC ever caused a septicemia (bacteria in the blood) in those it made ill. This point would seem to confirm the observation that STEC is not very invasive.
Whichever viewpoint one subscribes to, it seems clear that E. coli (STEC or non-STEC) should not be tolerated in meat. Of course, good, sanitary manufacturing processes should be followed in the slaughterhouse. Regardless, meat must be pasteurized. Radiation is not the answer. Not only would it trigger customer discontent but within current dose limitations it may only be expected to kill about 99.9% of the population present. This level of effectiveness is not adequate.
High pressure pasteurization (HPP) can deliver the effectiveness needed (99.999% kill or greater). Some meat producers (e.g., Hormel) are already using this method, as mentioned in their website. This will certainly eliminate any O157 or non-O157 STEC “adulterants” from meat. However, the possibility left open by treating STEC as an infectious agent and “adulterant” is the possibility that eliminating this so-called “pathogen” through pasteurization without testing for contamination by Shiga toxin may still leave meat unsafe to eat and consumers at risk for HUS.
The Centers for Disease Control and Prevention (CDC) has already provided a method with which to measure ricin, a toxin with the same mode of action as Shiga toxin, by use of a clever synthetic substrate for N-glycosidase activity .
If we find Shiga toxin in meat, we must have a way to deal with it. Reliance upon cooking by the consumer does not seem like a good idea. Shiga toxin is a protein and enzyme we already know will survive conditions that pasteurize (HTST) milk. We already have experience trying to cook other protein poisons, such as prions. We know the process often ends in failure. Part of the problem is that while the polypeptide chains of an enzyme may be denatured by heat, we know that often, such denatured polypeptides may autonomously re-fold themselves so as to recreate the very enzymatic activity we were trying to destroy.
Recently, it has been shown that Shiga toxin triggers inflammation through interaction with the complement cascade. Block the complement cascade and one ameliorates HUS. It so happens that there is already a drug that does exactly this (Eculizumab – trade name Soliris) . Designed by gene jockeys and cloned for relative ease of manufacture, this drug is a synthetic antibody fragment. It is currently undergoing clinical trials as a treatment for HUS.
Food manufacturers, however, may want to hold off on their collective sigh of relief that a “cure” for HUS has been found and that consequently HUS is not as dangerous as it is now when there is nothing even resembling a cure. Forbes magazine has calculated that the average yearly course of this drug is likely to cost about $409,500 USD, making it the current title-holder for most expensive drug sold in the U.S.
On the other extreme (as far as cost is concerned) is manganese . Nodules comprising up to 30% by weight of manganese are said to litter the ocean floor from one end of this earth to the other, so the raw material is there for the taking. This substance has been shown to protect against Shiga toxicosis. The reader will remember that HUS is caused by Shiga toxin after it reaches the blood (toxemia). Nevertheless, if toxicosis of the colon may be controlled by manganese, then perhaps the toxin may be kept from ever getting into the bloodstream to cause a toxemia (and HUS).
What is noteworthy about both the Eculizumab and manganese reports is that these presumed curative schema work against the toxin itself. This is an indication (but not proof) that the problem at hand has everything to do with the Shiga toxin and little, if anything, to do with the presumed infectious agent, STEC, regardless of its serotype.
The regulatory environment has us currently set to perform handsprings to identify STEC in meat but almost nothing to address the most proximate problem – Shiga toxin. I think we would be better advised to identify (using a variant of CDC’s analytic method for ricin) and then solve the actual problem: Colonic toxicosis followed by toxemia with Stx1 and/or Stx2.
 Jensen, C., S. Ethelberg, A Gervelmeyer, E. M. Mielsen, K. E. Olsen and K. Molbak (2006) “First general outbreak of Verocytotoxin-producing Escherichia coli O157 in Denmark” Euro. Surveill. 11:55-58.
 Goh, S., C. Newman, M. Knowles, F. J. Bolton, V. Hollyoak, S. Richards, P. Daley, D. Counter, H. R. Smith and N. Keppie (2002) “E. coli O157 phage type 21/28 outbreak in North Cumbria associated with pasteurized milk,” Epidemiol Infect. 209:451-457.
 Rasooly, R. and P. M. Do (2010) “Shiga toxin Stx2 is heat-stable and not inactivated by pasteurization” International Journal of Food Microbiology 136:290-294.
 Sauter, K. A. D., A. R. Melton-Celsa, Kay Larkin, M. L. Troxell, A. D. O’Brien and B. E. Magun (2008) “Mouse model of hemolytic-uremic syndrome caused by endotoxin-free Shiga toxin 2 (Stx2) and protection from lethal outcome by anti-Stx2 antibody,” Infection and Immunity 76(10):4469-4478.
 Kalb, S. R. and J. R. Kalb (2009) “Mass spectrometric detection of ricin and its activity in food and clinical samples” Analytical Chemistry 81:2037-2042.
 Lapeyraque, A. L., M. Malina, V. Fremeaux-Bacchi T. Boppel, M. Kirschfink, M. Oualhua, F. Proulx, P. Niaudet and F. Schaefer (2011) “Eculizumab in severe shiga- toxin-associated HUS,” New England Journal of Medicine 364:2561-2563.
 Mukhopadhyay, S. and A. D. Linstedt (2012) “Manganese blocks intracellular trafficking of Shiga toxin and protects against Shiga toxicosis,” Science, 20 January, pp. 332-335, DOI: 10.1126/science 1215930.© Food Safety News