Antibiotics are becoming a Red Queen’s Race. So how do we keep bacteria from killing us?

Bacterial infections are growing more dangerous. 

There is an annual increase of about 70% in the death rate from bacteria that are resistant to treatment with known antibiotics. This happens because bacteria, like all other organisms, strive to survive, and pass their survival tips on to their children. 

The more a drug is used, the more the pressure is on, and the more bacteria will be resistant to it. Because Penicillin and similar old workhorses are used so often — in clinical settings, in veterinary settings, and in agricultural settings — many bacteria have now become resistant to them.

The obvious thing to do would be to create new drugs. A bacteria can’t have evolved to resist a drug it’s never seen — that would be like passing a test by guessing on every answer. 

But this is a Red Queen’s Race, and it’s one that the human race is losing. 

While it’s not entirely clear how much money it takes to bring a new drug to market, costs are increasing, and estimates range from $161 million to $4.54 billion in 2021. It’s very difficult to come up with a new, safe, effective drug. So what do we do?

We need new strategies and new policies. Strategies include new ways of finding drugs, as well as the new drugs themselves.  

Artificial intelligence has gained a great deal of traction in drug development over the last few years, but it’s not as easy as just telling a computer you want a replacement for Penicillin. AI is fast, but despite the name, it’s not smart. It’s only as good as the data and instructions it’s given by humans. 

One strategy that my lab employs combines an AI that tries to dream up new peptide sequences in much the same way LLMs dream up new sentences with an AI that recommends the most interesting sequence, similar to a Youtube recommendation algorithm. 

To really find drugs that matter, we have to understand whether the AI’s ideas will work, and how, so we use our expertise in biophysics to study those drugs with simulations. Simulations are like little video game engines that let us predict how drugs will behave near a bacterium or near a human cell. Will they interact with the bacterium? Will they kill it? 

Once we’ve determined this, we can pass our likeliest candidates on to our experimental collaborators. While we rapidly find decent candidates, this is much faster than just searching blindly through different types of sequences. 

Another related possibility is finding drugs that protect people from the negative effects of bacteria, but don’t kill the bacteria while they’re doing it. This way, the bacterial population doesn’t need to protect itself, and so resistance doesn’t develop to begin with. 

For example, some bacteria cause harm to humans by producing a poison called endotoxin, which can trick a person’s immune system into overreacting so hard it damages them. Drugs that interrupt the process by which this poison works won’t kill the bacteria, but they will stop it from killing its human host.  We can use the same strategy as before, but search for drugs that render the endotoxin harmless. 

On the policy front, Canada is working hard. For a number of years, our government has had in place an action plan on combatting antimicrobial resistance. The current approach considers multiple different angles of attack, recognizing that antimicrobial resistance is a complex web of interlinked problems. 

One angle is related to the tactic of my lab: that’s the basic science and understanding angle. The government also encourages tracking data on resistance, educating people and health providers about how to use antibiotics appropriately, and putting strategies in place to keep infections from developing to begin with. 

We want to know which antibiotics are becoming less effective, under what circumstances, and to track emerging threats. We also want everyone to do their part: you shouldn’t take someone else’s antibiotic, especially if you don’t have a viral infection.

Antimicrobial resistance is a complicated problem because it’s the old story — it’s survival of the fittest.  Bacteria will keep trying to thrive, and so will humans and their animal companions. 

We need to search for multiple ways forward that allow us to coexist in a changing world.

Rachael (Ré) Mansbach is an associate professor of physics at Concordia University and Tier II Canada Research Chair in Computational Biophysics. Their lab uses computers to understand and develop peptide-based biomaterials for various applications.

The views expressed are those of the author(s). Canada Healthwatch publishes a range of perspectives and does not necessarily endorse the opinions presented.