[ad_1]
Their algorithm swallowed the entire human proteome and gave a preliminary list of approximately 43,000 peptides. Torres narrowed it down to 2,603, these proteins are derived from proteins known to be secreted from cells. Some are complete small proteins and hormones. The rest are just fragments, the encrypted chain in a larger complex. None of them have been described as antibiotics before.
To check whether their artificial intelligence is on the right track, Torres synthesized 55 most promising candidates. He tested every “Who’s Who” of drug-resistant microorganisms in liquid samples: Pseudomonas aeruginosa, A notorious lung infection; Acinetobacter baumannii, Known to spread rampantly in hospitals; Staphylococcus aureus, The bacteria behind the dangerous staphylococcal infection-plus others, a total of eight. Of the 55, most were able to prevent bacterial replication.
Some peptides stand out, including SCUB1-SKE25 and SCUB3-MLP22. These peptides are found in regions called “CUB domains”, which are found in a long list of proteins related to fertilization, creation of new blood vessels, and tumor suppression. SCUB is only part of the whole. But for themselves, they seem to be very good at killing bacteria. Therefore, Torres extended these two SCUBs to mice for experiments.
Torres tested whether SCUB or a combination of the two can eliminate infections in mice infected under the skin or thigh muscles (a more systemic disease model). In all cases, the bacterial populations sampled from these tissues stopped growing. In some cases, as Torres noticed on his warm agar, the number of bacteria dropped sharply.
Torres also tested how difficult it is for bacteria to develop resistance to peptides compared to an existing antibiotic called polymyxin B. After 30 days of exposure, the bacteria can tolerate a polymyxin B dose that is 256 times higher than the original dose, but SCUB is still effective at the same dose. (Bacteria need a lot of genetic changes to adapt to membrane damage.) Of course, this does not mean that they will never adapt, especially in longer time intervals. “Nothing can resist resistance,” de la Fuente said. “Because bacteria are the greatest evolutionary we know of.”
Despite the systematic plan of the team, Torres was still a bit dumbfounded. “We think we will have a lot of success,” he said of the peptides revealed by AI. But to his surprise, the peptide comes from the whole body. They come from proteins in the eyes, nervous system, and cardiovascular system, not just the immune system. “They are actually everywhere,” Torres said.
The team believes that life has evolved in this way to pack as many shocks as possible into the genome. “A gene encodes a protein, but the protein has multiple functions,” de la Fuente said. “I think this is a very smart way of evolution that can keep genomic information to a minimum.”
This is the first time that a scientist has discovered an antibiotic peptide in a protein unrelated to the immune response. This idea is “very creative,” said Jon Stokes, a biochemist at McMaster University in Canada. He was not involved in this research, but has been preparing his laboratory to include artificial intelligence in the search for small molecule antibiotics. middle. “For me, what I took home: started looking for antibiotics in unobvious places.”
[ad_2]
Source link
