Health

Machine Learning Uncovers New Ways to Kill Bacteria With Non-Antibiotic Drugs : ScienceAlert

Human history changed forever with the discovery of antibiotics in 1928. Infectious diseases such as pneumonia, tuberculosis, and septicemia were widespread and deadly until penicillin made them treatable.

Surgical procedures that once carried a high risk of infection have become safer and more common. Antibiotics marked a triumphant moment in science that transformed medical practice and saved countless lives.

But antibiotics come with an inherent caveat: When overused, bacteria can develop resistance to these drugs. The World Health Organization estimates that these superbugs caused 1.27 million deaths worldwide in 2019 and will likely pose a growing threat to global public health in the years to come.

New discoveries are helping scientists address this challenge in innovative ways. Studies have shown that nearly a quarter of drugs not normally prescribed as antibiotics, such as drugs used to treat cancer, diabetes and depression, can kill bacteria at doses typically prescribed to humans.

Understanding the mechanisms underlying the toxicity of certain drugs to bacteria could have considerable implications for medicine. If non-antibiotic drugs target bacteria in a different way than standard antibiotics, they could provide avenues for the development of new antibiotics.

But if non-antibiotics kill bacteria in the same way as known antibiotics, their prolonged use, for example in the treatment of chronic diseases, could inadvertently promote antibiotic resistance.

In our recently published research, my colleagues and I developed a new machine learning method that not only identifies how non-antibiotics kill bacteria, but can also help find new bacterial targets for antibiotics.

Mycobacterium tuberculosis is one of several microbial species that have developed resistance to several antibiotics. (NIAID/Flickr, CC BY)

New ways to kill bacteria

Many scientists and doctors around the world are tackling the problem of drug resistance, including myself and my colleagues in the Mitchell lab at UMass Chan School of Medicine. We use the genetics of bacteria to study which mutations make bacteria more resistant or more sensitive to drugs.

When my team and I discovered the widespread antibacterial activity of non-antibiotics, we were consumed by the challenge it presented: understanding how these drugs kill bacteria.

To answer this question, I used a genetic screening technique recently developed by my colleagues to study how cancer drugs target bacteria. This method identifies specific genes and cellular processes that change when bacteria mutate. Monitoring how these changes influence bacterial survival allows researchers to infer the mechanisms these drugs use to kill bacteria.

I have collected and analyzed nearly 2 million cases of toxicity between 200 drugs and thousands of mutant bacteria. Using a machine learning algorithm I developed to infer similarities between different drugs, I grouped the drugs into a network based on how they affected the mutant bacteria.

My maps clearly showed that known antibiotics were closely grouped according to their known classes of killing mechanisms. For example, all antibiotics that target the cell wall – the thick protective layer surrounding bacterial cells – have been grouped together and kept well separated from antibiotics that interfere with the bacteria’s DNA replication.

Interestingly, when I added non-antibiotic drugs to my analysis, they formed clusters distinct from antibiotics. This indicates that non-antibiotic drugs and antibiotics have different ways of killing bacterial cells. While these groupings don’t reveal how each drug specifically kills antibiotics, they do show that those grouped together probably work the same way.

The final piece of the puzzle – whether we could find new drug targets in bacteria to kill them – came from the research of my colleague Carmen Li.

It grew hundreds of generations of bacteria that were exposed to different non-antibiotic medications normally prescribed to treat anxiety, parasitic infections and cancer.

Sequencing the genomes of bacteria that have evolved and adapted to the presence of these drugs allowed us to identify the specific bacterial protein targeted by triclabendazole – a drug used to treat parasitic infections – to kill the bacteria. It is important to note that current antibiotics generally do not target this protein.

Additionally, we found that two other non-antibiotics using a similar mechanism to triclabendazole also target the same protein. This demonstrated the power of my drug similarity maps to identify drugs with similar killing mechanisms, even when that mechanism was still unknown.

Helping to discover antibiotics

Our results open multiple opportunities for researchers to study how non-antibiotic drugs act differently from standard antibiotics. Our drug mapping and testing method also has the potential to resolve a critical bottleneck in antibiotic development.

The search for new antibiotics typically involves devoting considerable resources to screening thousands of bacteria-killing chemicals and understanding how they work. Most of these chemicals work similarly to existing antibiotics and are discarded.

Our work shows that combining genetic screening and machine learning can help uncover the chemical needle in the haystack that can kill bacteria in a way researchers have never used before.

There are different ways to kill bacteria that we haven’t yet exploited, and there are still avenues we can take to combat the threat of bacterial infections and antibiotic resistance.The conversation

Mariana Noto Guillen, Ph.D. Candidate in Systems Biology, UMass Chan Medical School

This article is republished from The Conversation under a Creative Commons license. Read the original article.

News Source : www.sciencealert.com
Gn Health

Back to top button