A scoopful of soil, teeming with microscopic life, contains a rich library of genes that help bacteria thrive in the wild. Some of those genes, new research has found, are identical to those that allow disease-causing bacteria in humans to survive antibiotic treatment. The finding suggests that innocuous soil bacteria could be the original source of some antibiotic-resistant genes seen in hospitals.
"Soil ecologists have been predicting for quite a while that the soil acts as a reservoir for resistance," says molecular biologist Jo Handelsman of Yale University, who was not involved in the new study. "But until now there’s been very little evidence."
Soil-dwelling bacteria have been exposed to natural antibiotics—compounds frequently produced by competing microbes—for millions of years, often developing resistance mechanisms as they coevolved. Many of these natural killer compounds have served as the basis for commercially sold antibiotics. Because bacteria are known to swap genes when they come in contact, researchers have speculated that some resistance genes found in the soil may find their way into microbes that cause diseases in humans and animals, such as Escherichia coli or Staphylococcus.
But scanning a soil sample for specific types of genes is tricky because there are so many genes in total, and most bacterial resistance genes found in soil in the past have differed from those seen in human pathogens.
Now, combining next-generation, high-throughput sequencing with classic bacterial culturing methods, biologist Gautam Dantas of Washington University School of Medicine in St. Louis studied 11 soil samples from around the United States. The team first isolated genes from the soil that originated from Proteobacteria, a group of bacteria that’s usually disease-causing, hypothesizing that these organisms might be most likely to share genes with pathogens. Next, they inserted the genes into cells and tested which grew in the presence of any of 12 antibiotics. They then pinpointed 252 genes that survived these antibiotics and further analyzed their gene sequences, comparing them with genes from known human pathogens. One-hundred-ten of the genes had clear similarities in sequence to known antibiotic-resistance genes, the team discovered, and 18 of those were 100% identical to genes found in human pathogens.
The matching sequences likely mean that the genes were transferred at some point between the soil bacteria and the pathogens , the team reports online today in Science. But although it is more likely that they were transferred from the soil to human pathogens, the team can't rule out that it was the other way around. "It's hard to tell is who gave what to whom," Dantas says. "Our work does not speak to the direction of the transfer." Future studies may show how the genes are transferred exactly, he says—and find ways to stop that process.
Further analysis reveals that the antibiotic resistance genes are often lumped together into “resistance islands” within microbes’ DNA, with up to five genes within one island. These clumps of genes are flanked on each end by DNA sequences known as mobility units, which make them prone to moving between genomes as a whole. “In one gene transfer event, you could take a susceptible pathogen and … make it resistant to four or five antibiotics at the same time,” Dantas says.
"I think it's a really sound study with interesting results," Handelsman says. "The high degree of identity they reported is remarkable." One fascinating aspect is that so many resistance-conferring gene sequences didn't show similarity to known resistance genes, she says. Uncovering their mechanisms of resistance could reveal new molecular pathways to target with antibiotics. "Part of the work that’s going to be really interesting in the future is uncovering the functions of some of those genes," she says.