Stephen Calderwood Discusses the Infectious Mechanisms of Cholera
Special Topic of Cholera Interview, July 2011
Our Special Topics analysis of cholera research over the past decade shows that the work of Dr. Stephen Calderwood ranks at #10 by total number of papers, based on 60 papers cited 888 times. Two of these papers appear on the top 20 papers lists in the Topic.
In Essential Science IndicatorsSM from Thomson Reuters, Calderwood ranks in the top 1% among scientists in the field of Microbiology. His overall record in the database includes 77 papers cited a total of 2,179 times between January 1, 2001 and February 28, 2011.
Calderwood is Chief of the Division of Infectious Diseases and Vice-Chair of the Department of Medicine at Massachusetts General Hospital in Boston. He is also the Morton N. Swartz M.D. Academy Professor of Medicine at Harvard Medical School.
You focus your research on humans in large part because of the absence of a good animal model for cholera. Could you tell us why no such model exists and what the problems are with existing models?
There are animal models of cholera that are of use for some areas of research. An infant mouse, for instance, dies if it gets cholera, but it doesn't have diarrheal disease. Infant rabbits get a diarrheal disease and die, but they're not mature animals and so the immune responses you might expect in a human aren't going to be the same as you'll see in an infant animal model.
Some years ago, we tried to study immune responses in a germ-free adult mouse model. We could colonize and see immune responses in that model, but it's a fairly artificial model compared to human infection, so we weren't sure how firm the conclusions we could draw from that model would be. These models are helpful, though, to generate hypotheses that might correlate with what we'd see in humans.
Your most-cited paper is a 2002 Nature article, "Host-induced epidemic spread of the cholera bacterium," (Merrell DS, et al., 417[6889]: 642-5, 6 June 2002). What questions were you trying to establish in that research and what did you find?
One of the most exciting and fun parts of the NIH grant mechanism under which we work—the International Collaborations in Infectious Disease Research—is that it establishes a framework by which collaborative international research can be done. What I mean by that is that we have human studies approval at the NIH, at our own institution, and at the International Center for Diarrheal Disease Research (ICDDR), Bangladesh, our collaborator, to do a variety of studies.
We also have mechanisms of enabling money for supplies and for people to move back and forth, administratively. We also have, through the Centers for Disease Control, mechanisms of shipping samples from Bangladesh to the US for analysis. This grant mechanism has allowed us to facilitate work with collaborators who aren't part of that grant mechanism themselves, but have exciting and interesting ideas that they'd like to test directly in humans. And we have done a lot of collaborative work that way—with John Mekalanos here and Shah Faruque at the ICDDR, for example.
The particular collaboration in the Nature paper was with Andy Camilli and Scott Merrell at Tufts and Gary Schoolnik at Stanford. They had been modeling in animals what happens to cholera as it passes through that animal, and how those changes might effect transmission to the next host. But as I said, modeling that in an animal may not fully reproduce what happens in a human. And so they wanted to look directly in human stool at cholera that was exiting the host.
They wanted to look both at how the organism from the human environment differs in gene expression from what you might see in vitro or in an animal model. Then they wanted to see if they could relate those differences in gene expression to transmission. So we found specific differences in gene expression in the organisms isolated from stool samples. And perhaps more excitingly to me, we found differences in subsequent transmission as well. It was the first time that a phenomenon called hyperinfectivity was suggested.
So what is hyperinfectivity and how does it manifest itself?
"One of the most exciting and fun parts of the NIH grant mechanism under which we work—the International Collaborations in Infectious Disease Research—is that it establishes a framework by which collaborative international research can be done."
What hyperinfectivity implies is that for a brief period after passage through a human, the organism is more infectious to the next human host. During this time, its infective dose-50, i.e., the dose needed to infect 50% of a population, is lower, by between 100- and 1000-fold. The reason that's important in cholera transmission is because it is very easy for a fecal-orally transmitted organism to pass directly from one person to the next person through contaminated surfaces, water, etc., if they live in a densely populated area. And this remains the case for from several hours to as long as 24 hours while the organism maintains that phenotype.
In the developed world, transmission of that rapidity is very unlikely. The organism readapts to the environment and the infectious dose goes up to be higher, making it much less likely that cholera would be transmitted person-to-person. And so it's an important phenomenon in potentially explaining why cholera transmits so well in densely populated areas with poor sewage. Hyperinfectivity has since been documented by a mathematical model to contribute to the reason why cholera outbreaks, when they begin, spread very quickly to a large area in resource-limited areas.
Do you know what genetic changes confer this temporary hyperinfectivity?
We have some ideas. There are specific genes that are turned on that are not expressed as well once the organism readapts to the environment. One of those genes is the pilus that is responsible for human colonization. A pilus is an attachment structure on the surface of a bacterium that attaches the bacterium onto the gastrointestinal tract when it's ingested. Whether that's the sole or most important of the gene expression changes, we're not sure.
We have done a follow-up study in which we modeled hyperinfectivity in the infant mouse and showed that if you pass the organism through a mouse, it's also more infectious for the next mouse. But as I said, that may or may not be exactly what happens in a person. So the normal way you would analyze this problem—making mutants of various genes and redoing the experiment—you, of course, can't do in humans. You can't artificially infect people with different mutant strains of cholera, certainly not in a place like Bangladesh.