A physicist at the University of Guelph is discovering antimicrobial peptides that may be effective at killing bacteria significant to the swine industry
By Jackie Clark
Staff Writer
Farms.com
Researchers at the University of Guelph are using machine-learning to build novel molecules that could replace the use of traditional antibiotics.
“There’s a tremendous need worldwide for alternatives to antibiotics,” Dr. Chris Gray, professor emeritus in the department of physics, told Farms.com.
“Antibiotics are losing their effectiveness in both humans and animals, because there’s mutation among these bacteria; they adapt and they change. We’ve overused (antibiotics) so (the bacteria are) adapting faster than we can produce new (antibiotics),” he explained. The result is antibiotic-resistant bacteria.
The researchers in Gray’s lab are conducting studies with funding from Ontario Pork to investigate anti-microbial peptides (essentially a small protein) as an alternative to antibiotics.
“Peptides are small protein-like molecules made up of amino acids. They tend to be anywhere from 10 amino acids to 50, whereas proteins can be thousands of amino acids,” Gray said.
“The beauty of these molecules is that some of them, the trick is to find which ones, but some of them will attack bacteria … but not kill the host,” he explained.
He originally started this research to understand the biophysical reasons why effective peptides kill harmful bacteria, but don’t damage the cells of the human or animal they’re working inside. Gray and other scientists found that the effective peptides are positively charged and strongly attracted to bacterial membranes which are negatively charged, but only weakly attracted to uncharged animal cell membranes.
“That’s how (peptides) discriminate the bacteria from other things,” Gray added.
It occurred to him that this knowledge may be useful for animal health.
“Certain bacteria are important in the swine industry,” Gray said. So his team put together a research proposal for Ontario Pork, suggesting that they work to identify peptides that are effective against industry-relevant bacteria.
That work involves building up a database of effective, naturally-occurring peptides, and then eventually synthesizing new ones.
“Peptides occur naturally,” Gray explained. “Every living being has (peptides) as part of its own immune system.”
The scientists can “figure out what the effective part of (peptides) is, and try to design new ones … in other words you can find the potent part of these molecules, isolate it, and then design a new molecule that was even smaller and more powerful than the natural ones,” he said.
To do so, Gray’s team is collaborating with “artificial intelligence or machine-learning specialists,” he said.
The researchers “have to train a program to recognize patterns by giving it lots of examples” of effective peptides, and then the program can scan other possible amino acid combinations for ones that have the potential to be useful antimicrobials, Gray explained.
“It does the work of finding a needle in a haystack,” he said.
The research is still in this candidate discovery step. The next stages of the research will be trials in the lab to evaluate effectiveness and safety, and then field trials with pigs, Gray said.
Field tests will “look at how well these animals are cleared of their bacterial infection and how non-toxic the (peptide) is in practice,” he explained.
The process is a long one, with a year or two of lab testing after candidates are identified before field trials can begin. The eventual goal would be a commercially available antimicrobial peptide for farmers to give their pigs.
“It’ll probably take the form of a capsule eventually,” Gray added.
dusanpetkovic\iStock\Getty Images Plus photo