Pathogenic bacteria in humans develop resistance to antibiotics much faster than expected. Computer-aided studies at the Swedish Technical University of Chalmers now show that one reason could be a significant genetic transfer between bacteria in our ecosystems and to humans. This work has also led to new tools for resistance researchers.
According to the World Health Organization, antibiotic resistance is one of the greatest threats to global health, food safety and development. In Europe alone, over 33,000 people die every year.
Completely different types of bacteria can spread resistance genes against one another using plasmids – small DNA molecules in which bacteria store some of their genes outside the chromosome. When two bacterial cells come into contact, they can copy plasmids together. This is known as conjugation and is the main mechanism for spreading antibiotic resistance.
"In recent years we have seen resistance genes spread to human pathogens much more than expected," says Jan Zrimec, researcher in systems and synthetic biology at Chalmers University of Technology. "Many of the genes appear to come from a variety of bacterial species and environments, such as soil, aquatic, and plant bacteria.
'This was difficult to explain because, although conjugation is very common, we assumed that there was a distinct limitation on which bacterial species can transfer plasmids to one another. Plasmids belong to different mobility groups or MOB groups, so they can't between transmitted to any type of bacteria.
Zrimec has developed new methods of data analysis which show that genetic transfer can be much more limitless and more widespread than previously expected.
Among other things, he has developed an algorithm that can identify certain DNA regions that are necessary for conjugation – so-called oriT regions – in large amounts of data, which consist of genetic sequences from the DNA of thousands of plasmids. The algorithm can also sort plasmids into MOB groups based on the identified oriT regions.
He used the algorithm to examine known gene sequences from over 4,600 naturally occurring plasmids from different types of bacteria, which was previously not possible systematically. The results show, among other things, that:
- The number of oriT regions can be almost eight times higher than the standard method used today.
- The number of mobile plasmids can be twice as high as previously known.
- The number of bacterial species with mobile plasmids can be almost twice as high as previously known.
- Over half of these plasmids have oriT regions that match a conjugation enzyme from another plasmid that was previously divided into a different MOB group. This means that they could be transmitted by one of these plasmids that happens to be in the same bacterial cell.
The last part means that there may be mechanisms of transmission between large numbers of types of bacteria and environments where we previously thought there were barriers.
"These results could mean that there is a robust network for the transfer of plasmids between bacteria in humans, animals, plants, soil, water and industries, to name a few," explains Zrimec. 'Resistance genes are naturally found in many different bacteria in these ecosystems, and the hypothetical network could mean that genes from all of these environments can be transferred to bacteria that cause disease in humans.
"This could be a possible reason for the rapid development of resistance in human pathogens observed in recent years. Our extensive use of antibiotics selects resistance genes that could thus flow from a much larger naturally occurring genetic reservoir than previously estimated."
The results will have to be verified experimentally in the future, but the data analysis methods developed by Zrimec can already be used by many researchers working with antibiotic resistance in various medical and biological fields. They offer a powerful new tool for systematically mapping the potential portability of various plasmids.
"This has been a major limitation of the research area so far," says Zrimec. "I hope the methods can benefit large chunks of antibiotic resistance research, which is an extremely interdisciplinary and fragmented area. The methods can be used for studies aimed at developing more effective restrictions on the use of antibiotics, instructions for the use of antibiotics and new types of substances that can prevent the spread of resistance genes at the molecular level. "
More on the topic: Genetic transfer through conjugation
In order for conjugation to begin, an enzyme – a relaxase – is required that fits into a specific location on the plasmid. The relaxase has to recognize and bind to a region in which the DNA ring can be nicked and a strand can be transferred to the next bacterium. This region of DNA is known as the origin of transfer or oriT.
Previously it was assumed that a single plasmid must contain both the gene for the relaxase and a suitable oriT in order to be transferred to other bacteria. However, a bacterial cell can contain multiple plasmids, and in recent years various researchers have shown that a relaxase from one plasmid can fit into an oriT region on another plasmid in the same cell and activate conjugation of that plasmid.
This means that it may be enough for a plasmid to have only one oriT in order to be able to conjugate, which in turn means that many plasmids previously classified as immobile because they lack the relaxase gene can be conjugative. However, it is not yet known how often the phenomenon occurs in bacteria. This is one of the knowledge gaps that Zrimec's findings fill.
More on the topic: The new method compared to the current standard
The current standard tools for assessing the transferability of plasmids are based on the search for DNA sequences for the relaxase enzyme or for oriT regions to which the enzyme can bind. There are several major limitations to this. For one thing, some tools produce incomplete results, while others require extremely time-consuming and resource-intensive laboratory tests.
The new data analysis method from Zrimec is based exclusively on the identification of oriT regions using special physiochemical properties that are specifically found in oriT regions of the DNA. Through previous research, he has shown that these physiochemical signatures – which determine which relaxase can bind to the oriT region – are more stable and specific than the DNA sequences themselves. This allows the plasmids to be classified into the correct MOB group based on of the oriT region, independent of the relaxase, which allows researchers to map the overall transferability between different bacterial species and environments.
The method can handle large amounts of data and can be used to effectively search for oriT regions on plasmids in their entirety.