Among bacteria, Salmonella is notorious: One strain causes typhoid and spreads readily among humans, often because of inadequate sanitation or hygiene, leading to millions of cases a year. Think Typhoid Mary, the infamous New York cook. Another strain, known as nontyphoidal Salmonella (NTS), causes millions of cases of food poisoning each year and is usually contracted by eating undercooked meat or eggs. There's a lesser-known relative as well, invasive nontyphoidal Salmonella (iNTS), which also causes serious disease in humans—and appears to have exploded in sub-Saharan Africa over the past 50 years.
Often confused with malaria because it causes high fevers, iNTS kills up to half of the adults it infects. A new study published online yesterday in Nature Genetics offers a provocative explanation for its success in sub-Saharan Africa: Strains have evolved that appear to spread from person to person, unlike other NTSs. These strains may have taken advantage of a "niche" created by the emergence of HIV, which cripples immune systems and makes people more vulnerable.
A team led by Gordon Dougan, a molecular microbiologist at the Wellcome Trust Sanger Institute in Cambridge, U.K., reports that the group fully sequenced 129 isolates of Salmonella Typhimurium from seven sub-Saharan African countries that were collected between 1988 and 2010. They then grouped the isolates into family trees and found two distinct lineages. Further analyses revealed that the emergence and spread of both lineages overlapped with that of HIV on the continent. "It's a compelling demonstration of how techniques of evolutionary analysis developed for viruses can be applied to bacterial pathogens, provided full genome data are available," says Oliver Pybus, an evolutionary biologist at the University of Oxford in the United Kingdom, who studies HIV, influenza, and hepatitis C viruses.
Dougan and his team delved into the evolution and spread of iNTS because many questions remain about its prevalence and routes of transmission. The origin of epidemics also fascinates many researchers because it enables them to use molecular and epidemiologic forensics to flesh out historical details.
It's well known that HIV and the immune damage it causes makes people more vulnerable to iNTS, as do malnutrition and malaria in children. Estimates of iNTS fatalities in children go as high as 27% and reach 47% in adults. But much of the epidemiology of iNTS is unknown because the infection is difficult to diagnose in the places it occurs. "Often, disease surveillance in these areas is not good," Dougan says. "You need to culture the organism to unequivocally identify it. But we're clearly talking about a highly significant problem in Africa, potentially tens of thousands of patients each year."
Dougan worked with an international consortium to collect samples from Malawi, Mali, Kenya, Nigeria, the Democratic Republic of the Congo, Mozambique, and Uganda. His group compared the genetic sequences of these Typhimurium serovars—short for serotype variants, a more sophisticated description of strains—to the more common ones that cause gastroenteritis and typhoid. While the common NTS serovars have wide genetic divergence, suggesting multiple introductions from various animal sources, those for iNTS in sub-Saharan Africa are closely related and were likely introduced few times and then spread through human populations. "It's remarkable that these are such tightly grouped sets of organisms, and it's quite clear that it's an epidemic," Dougan says.
Better understanding of disease transmission is key to controlling it. Many scientists long believed that Typhimurium spreads the same way other NTSs do: jumping from animals to humans and then hitting a dead end. But the new study suggests that these two Typhimurium lineages have had genetic changes that make them more typhoid-like and thus capable of spreading from person to person.
The findings support the work of Samuel Kariuki, who heads the Centre for Microbiology Research at the Kenya Medical Research Institute in Nairobi. "He was saying for many years that iNTS was human-to-human transmission and nobody believed him," Dougan says. "Western scientists couldn't accept this because it's so dogmatic that NTS is zoonotic."
Although it's well known that Typhimurium can piggyback on HIV, the case made in the Nature Genetics paper is that this Salmonella serovar got a foothold in the human population and became epidemic only because of the AIDS virus—a thesis that relies on several assumptions. "It's great to suggest the possibility that there's a causal relationship but at the end of the day this is pretty classic correlation," says Michael Worobey, an evolutionary biologist at the University of Arizona in Tucson who specializes in HIV's origins. "It's really quite possible that the dating is off by enough so that [causal argument] doesn't make sense."
The new study shows that what the researchers call lineage I first surfaced in Malawi as early as 1920, and that lineage II emerged as early as 1957 in the Democratic Republic of the Congo. The earliest documented HIV infection is 1959 in the Congo's capital, Kinshasa. "It certainly doesn't geographically leap out at me and say it's following the geographic and temporal pathway than HIV did," Worobey says.
Worobey compliments Dougan and colleagues for their "massive bacterial sequencing analysis." But he notes that there is a good deal of uncertainty in their calculations: Lineage I, for example, could have emerged anywhere between 1920 and 1960.
Dougan counters that "the temporal correlates are very strong." But he agrees with Worobey that their analyses would have benefited from more and older samples. "This is the most comprehensive data to date," Dougan says. "At the field sites we're working with, the collaborators are very high quality, but there's an awful lot more work to be done on this issue."