A new study by French scientists suggests the rise of ampicillin resistance in a strain of Salmonella may have begun before the antibiotic was ever used in humans.

Ampicillin, a widely used broad-spectrum, semi-synthetic derivative of penicillin, was first marketed in the United Kingdom in 1961 and introduced in other European countries shortly thereafter. But it wasn’t long before resistance to the drug began to emerge.

The first reported outbreak of an ampicillin-resistant strain of Salmonella enterica serotype Typhimurium—a common zoonotic bacterium that causes gastrointestinal disease in both humans and animals—occurred in the United Kingdom in 1962, and another one followed in 1964. Subsequent analysis revealed that the resistance mechanism, a beta-lactamase gene, was transmissible.

What caused ampicillin resistance to emerge so quickly? While widespread use of the drug and the ability of Salmonella bacteria to share resistance genes are the simplest explanations, the short amount of time between the introduction of ampicillin and the emergence of resistance left researchers at France’s Pasteur Institute wondering whether other factors may have been involved.

To find the answer to that question, they studied a large collection of historical isolates of SalmonellaTyphimurium, some more than a century old. Their findings appeared yesterday in The Lancet Infectious Diseases.

Stepping back in time

In the study, the researchers analyzed 288 Salmonella Typhimurium isolates obtained from humans, animals, food products, and animal in 31 countries and four continents from 1911 to 1969. Their analysis included antibiotic susceptibility testing and whole-genome sequencing, a method of DNA fingerprinting that can be used to identify mechanisms of resistance and genetic diversity among bacterial isolates.

Antibiotic susceptibility testing revealed that 11 of the 288 isolates (3.8%) were resistant to ampicillin, and whole-genome sequencing identified beta-lactamase genes on plasmids—the mobile pieces of DNA that can transfer resistance genes among and between different types of bacteria. While eight of the isolates were collected in the late 1960s—after the introduction of ampicillin—one was collected from France in 1959, and two were collected from Tunisia in 1960. Those three isolates, all of which came from people, carried the blaTEM-1 gene.

That finding indicated that something other than ampicillin use may have played a role in the emergence of ampicillin resistance. Additional experiments provided a potential answer. When the scientists exposed the three isolates to ampicillin, they all showed the ability to transfer their blaTEM-1 resistance gene to a laboratory-generated Salmonella Typhimurium strain. But the gene was also successfully transferred when the isolates were exposed to a small quantity of penicillin G, an antibiotic used widely in livestock and poultry production in Europe until 1969. France was among the countries that permitted penicillin G as a food additive for livestock.

Based on these findings, and the fact that penicillin G is poorly absorbed by the digestive system and can end up in manure and soil, lead author Francois-Xavier Weill, MD, PhD, and his colleagues theorize that the presence of penicillin G residues, even in small amounts, could have exerted enough selective pressure to encourage the spread of plasmids carrying the blaTEM-1 gene.

“Our findings suggest that antibiotic residues on farming environments such as soil, waste water, and manure may have a much greater impact on the spread of resistance than previously thought,” Weill, a clinical pathologist and research director at the Pasteur Institute, said in a news release.

Different vectors of resistance

Weill also found that the vectors of ampicillin resistance in the isolates, most of which were from France, differed from those that caused the initial outbreaks in the United Kingdom. That suggests that the emergence of ampicillin resistance in Salmonella Typhimurium was due to multiple independent acquisitions of the blaTEM-1-carrying plasmids by different bacterial populations.

Weill and his co-authors note that the results of the analysis provide no causal link between the use of penicillin G and the emergence of transmissible ampicillin resistance. But they are not the first to suggest the possibility. British microbiologist E.S. Anderson, who did pioneering research on plasmid-mediated resistance in Salmonella in the 1960s, also suspected that use of penicillin G in agriculture was linked to ampicillin resistance, and warned that widespread use of antibiotics in animal feed was contributing to multidrug-resistant bacteria.

Weill said the results highlight the need to re-evaluate the use of antibiotics in livestock and poultry and closely monitor antibiotic resistance in humans and food-producing animals.

See also:

Nov 29 Lancet Infect Dis study

Nov 29 Pasteur Institute news release