Glen D. Armstrong, Ph.D.
In Search of E. coli’s Silver Bullet
Glen Armstrong, a professor of medical microbiology at the University of Alberta in Edmonton, Alberta, finds himself serving as biotech consultant on one of the most intriguing projects embarked upon in contemporary food safety research: the development of an effective vaccination/treatment program for victims of E. coli O157 bacteria. Armstrong has been engaged in the study of gastrointestinal pathogens for over 17 years. His E. coli infection treatment program, which employs a compound called Synsorb-PK®, is presently in Phase III clinical trials and was used in the field for the first time in the recent Walkerton, Ontario water contamination tragedy.
FoodTechSource: How far off is a viable vaccine against E. coli?
Glen D. Armstrong, Ph.D.: Right now the potential for a vaccine is somewhere down the road. There’s still a lot of work to be done before we can start bringing these products into the marketplace. But in the interim there is a need for a therapeutic treatment that could be given to patients who get sick, and that’s what we’re working on first.
FTS: What sort of treatment?
Armstrong: Treatment for patients who have been hospitalized, and who face serious kidney damage or death.
FTS: What is it? How does it work?
Armstrong: E.coli O157 produces toxins. It is these potent toxins that cause the serious complications, such as kidney failure and central nervous system complications. The therapy we are developing is an insoluble, indigestible substance that has been chemically modified so it soaks up these toxins. The idea is that when a patient becomes ill they can consume this indigestible material. It goes down into the intestines, it mops up all of the toxin being produced in the intestines and allows it to be harmlessly flushed out of the body. Those are the basics.
FTS: Is the vaccine work based on the same principal?
“The Synsorb-PK
mops up all of
the toxin being produced
in the intestines
and allows it
to be harmlessly
flushed out
of the body.”
Armstrong: Somewhat. The vaccines are based on the toxins as well. These toxins are made up of several subunits, and all of these subunits have to be present and assembled in order for the toxin to have its activity. If you break the toxin into separate subunits, or eliminate a subunit, it is no longer toxic. A graduate student in my lab is using genetic cloning technology to create an e-coli organism that only uses one of the toxin subunits. When introduced into the body, this nontoxic E. coli will stimulate the immune system to create antibodies that will neutralize the assembled toxin should the patient get infected with the organism.
FTS: Because it will go after that one subunit of the toxin?
Armstrong: That’s right. By doing that it actually destroys the toxin activity and makes the toxin nontoxic.
FTS: That is an interesting concept. How did you come up with that approach?
Armstrong: The concept is actually quite old. It was originally the concept used to develop diphtheria vaccine, which is given in childhood. We’re basically just applying the concept to a different toxin.
FTS: Is that principally how all vaccines work?
Armstrong: Not all of them, but a lot of vaccines work on that principal. The trick is in the cloning technology. It sounds simple, but when you actually try to do it there are often tricks you have to come up with in order to get the organisms to make the protein you are interested in, and to make it in a form that can be recognized by the immune system.
FTS: How long have you been working on it?
Armstrong: That portion of our work has been going on for two years.
FTS: How did you become interested in E. coli treatments and vaccines?
Armstrong: I started working seriously on E. coli O157 in about 1989. Before then I was keeping up with the scientific literature so I could bring this information to the students in my classes at the university. It was through actually teaching students about this organism that the idea came to me that if we could somehow eliminate the toxins from the body this would be beneficial to patients infected with these organisms. At the same time I was considering how this might be used to develop a vaccine. But in this case I thought the need for a therapy was more acute, so I started to focus on that aspect. Through the literature I became aware that the toxin E. coli makes binds to specific receptor molecules that are on the surface of the cells that line the blood vessels in the human body. This is how the toxin causes the damage: It gets into the circulatory system and it damages those cells lining the blood vessels. In young children, the highest concentration of receptors these toxins bind to are found in the blood vessels in the kidneys—which is why the kidney is a target organ. We knew what the biochemical structure of those receptor molecules was and it turns out they are carbohydrate molecules—sugar building blocks—. There was a company at the time here in Edmonton that had a product they were developing principally to remove blood group antibodies from blood transfusions. This product was called Synsorb-PK, and because it had the same carbohydrate receptors on it that E. coli toxins bind to, I determined that the Synsorb molecule would probably bind the toxins from E. coli. I did a little experiment in the lab that showed that the toxins from E. coli in fact do bind to Synsorb-PK. Because this was an indigestible material and binds the toxins very tightly, I suggested in a paper I published in the Journal for Infectious Diseases that if patients infected with E. coli were to ingest this material it may absorb the toxins in the intestines, which could then be flushed out. This paper was read by some clinical researchers and they approached me about running clinical trials to test the theory. We are currently in Phase III clinical trials.
“The Phase II
results indicated
that children in the treatment group
had fewer
complications than
children in the
placebo group.”
FTS: Did Synsorb-PK in fact absorb toxin sufficiently? Or did it just give it more receptacles, and thus lead to the development of more toxin?
Armstrong: The former is the theory we are working on, and what the clinical trials are designed to prove. We did some preclinical work in which we got stool samples from children in the hospital who had an infection and we mixed the Synsorb with the stool samples and showed the Synsorb did bind the toxins. So we know if there is toxin present, Synsorb-Pk is capable of absorbing that toxin. The clinical trials so far have shown, and these are double-blind placebo trials, the Phase II results indicated that children in the treatment group had fewer complications than children in the placebo group.
FTS: You received “Fast Track Designation” on Phase III from the FDA, correct?
Armstrong: The company did, yes.
FTS: How long before the trials are finished?
Armstrong: Within a year is what I’m hearing.
FTS: Will the treatment only be effective with the E. coli O157:H7 serotype?
Armstrong: No; any form of E. coli that produces the toxin. In North America the O157:H7 form is responsible for anywhere from 80% to 95% of the cases of what’s called "Hamburger disease." But we know now there are close to 100 different E. coli types that are able to cause the disease. And in some parts of the world the O157:H7 E. coli is actually a very minor problem compared to some of these other serotypes. But the one thing they all have in common is they all make toxins that are very similar. And all of them make one or two of this family of toxins we focused our vaccine on, which should give it broad spectrum protection.
FTS: Does this include drug resistant strains?
Armstrong: Yes it would. It gets around that little problem. Plus, currently antibiotics are not recommended for treating these infections because there are good data in the literature showing that many of the E. coli that cause hamburger disease, if you treat them with inhibitory concentrations of antibiotics, they actually release more of the toxin. If you start to treat a patient with antibiotics, there’s a period of time, during which the antibiotic concentration is ramping up, when it’s not killing the organisms but causing them to produce a spike in toxin levels. We are concerned that at this time the spike of toxin may actually increase the probability of complications occurring in those patients. So another theory I am working on is that perhaps if we use antibiotics in combination with the Synsorb-PK we’ll be able to deliver a double-punch therapy where you kill the organisms with the antibiotics while at the same time you have an agent that can absorb the toxin produced by the dying organisms.
FTS: Is the toxin produced during the death of the organisms?
Armstrong: We’re not really certain why the organisms make this toxin. And when you look at the organisms, most of the toxin they make does not actually get out of the bacterial cell; it is trapped inside the bacteria. Plus, only a small amount of the toxin the organisms make actually gets outside the cell where it can be absorbed into the human body. The bacteria themselves are not invasive; they don’t go into the human body. They stay on the surface of the intestinal cells. The only way the toxin can get into the body is if it is excreted from the bacterial cell and absorbed by the cells in the intestine. And when you treat these organisms with antibiotics, the antibiotics cause the bacteria, as they die, to break open, at which point they release this intracellular store of toxin they have made. There’s another factor, too. The genes in the bacteria that cause the toxin to be made are carried on a bacterial virus that has inserted itself into the chromosome of the bacteria, and many antibiotics cause this virus to replicate in the bacteria. If a virus replicates it increases the number of genes for the toxin, and this causes the bacteria to produce more toxin.
“The danger
is that
antibiotics
cause the bacteria,
as they die,
to break open
and release an
intracellular
store of
toxin.”
FTS: If Synsorb-PK is only absorbing the toxin, how is it getting rid of the organism?
Armstrong: It’s not getting rid of the organism at all, which is the reason why I’m starting to think about using a combination of therapies in order increase the effectiveness of the Synsorb-PK treatment.
FTS: But the antibiotics also kill off the good E. coli, don’t they?
Armstrong: Yes. And that is a problem, because the good E. coli are part of our gastrointestinal defense mechanisms. One of the benefits of having good E. coli in your intestines is that they can compete successfully for nutrients in the intestine, which prevents pathogens from becoming established. So we are also researching a way of eliminating the bad E. coli that would not effect the good E. coli.
FTS: And what would that be?
Armstrong: We know E. coli O157:H7 produces molecules on its surface that act like grappling hooks and enable the organisms to attach to the intestinal cells. This is one of the ways O157:H7 can establish itself in the gut environment in the presence of normal E. coli—it has to have something that gives it an advantage. Well, we believe the molecules these grappling hooks are attacking also have carbohydrate structures. So, we’re in the early stages of identifying Synsorb-like molecules that can perhaps antagonize the activity of these E. coli O157 grappling hooks. That would then cause the organisms to fall off the intestinal walls and lose their advantage in the presence of the normal E. coli, allowing the body’s normal defense mechanisms to kick in and eliminate the E. coli O157.
FTS: Did the Synsorb-PK treatment program help victims of the Walkerton tragedy? [Walkerton, Ontario, was the recent site of widespread water contamination.—ed]
Armstrong: There is really no way of knowing. In the Walkerton situation the amount of material the company could release was limited to the treatment of about 20 individuals because the drug is still in clinical trials and the company hasn’t gotten its full manufacturing facility up to speed. Plus, the number of people effected in Walkerton was in the thousands, and it was not a placebo-control trial. So it’s impossible to say whether it helped or not. However, one thing we can say for certain is, based on all clinical trials and Walkerton, this is a very safe product. The number of drug-related adverse reactions is nil.
FTS: It’s a shame your work has received such sparse coverage. Aside from the Walkerton connection there is almost no information out there.
Armstrong: It’s a bit of a niche area. It’s unfortunate when terrible tragedies like Walkerton occur. But in actual fact the number of outbreaks like Walkerton is really quite low compared to other infectious diseases. It only attracts public attention when you have outbreaks like Walkerton.
FTS: But it’s something the food industry should be all over. You have companies paying huge sums in damages to victims who’ve consumed E. coli-contaminated foods. These are damage payments that would be much less severe if the pain and suffering were less severe.
Armstrong: Companies are also paying out huge sums to recall contaminated lots of beef and other food stuffs. It’s having a major financial impact on the food processing industry.
FTS: Yes, I’m sure everyone will be much happier if it turns out you’ve actually found that magic bullet.