a16z Podcast - Journal Club: A New Path to Antibiotic Resistance

Episode Date: July 19, 2020

Ever since the discovery of antibiotics, microbiologists have worried about and studied how bacteria acquire resistance to these drugs. Adding to the complexity of this problem is the fact that it is not always clear whether the conditions that drive the evolution of resistance in the lab occur in patients suffering from bacterial infections.This is where the work of Nathalie Balaban -- Professor at the Hebrew University, and our guest on this episode -- comes in. The article we discuss is based on a foundation of research done in her laboratory, but this study makes the important step into the clinic by using samples from a patient with a life-threatening bacterial (MRSA) infection. By analyzing these patient samples, Dr. Balaban and her team were able to understand the conditions that lead to multi-drug resistance in a hospital setting. The work reveals how the ability of bacteria to enter a state of dormancy, also known as tolerance, can act as a stepping stone to resistance and can interfere with the efficacy of drug combinations. Our conversation covers what tolerance is, the conditions that promote tolerance, how it can lead to resistance and impact drug combination therapies, and lastly, integrating this new understanding into clinical microbiology protocols."Effect of tolerance on the evolution of antibiotic resistance under drug combinations" in Science (January 2020) by Jiafeng Liu, Orit Gefen, Irine Ronin, Maskit Bar-Meir, Nathalie Q. Balaban.a16z bio Journal Club (part of the a16z Podcast), curates and covers recent advances from the scientific literature -- what papers we’re reading, and why they matter from our perspective at the intersection of biology & technology (for bio journal club). You can find all these episodes at a16z.com/journalclub.

Hello and welcome to the A16Z Journal Club. I'm Lauren Richardson. This is our podcast where we cover recent scientific advances, why they matter, and how to take them from proof of principle to practice. Antibiotic resistance is an urgent problem, but many aspects of how bacteria acquire this ability to evade the drugs designed to target them are still mysterious, and it is not always clear whether the conditions that drive the evolution of resistance in the lab occur in patients suffering from bacterial infections. This is where the work of Natalie Balaban, a professor at the Hebrew University and my guest today, comes in.

She joins me to discuss her article published in science titled, Effective Tolerance on the Evolution of Antibiotic Resistance under Drug Combinations. The work is based off of previous research done in her laboratory, and importantly, uses samples from a patient with a life-threatening bacterial infection to understand the conditions that lead to multi-drug resistance in a hospital setting. The work reveals how the ability of bacteria to develop tolerance, which we also refer to as dormancy or slow growth, can act as a stepping stone to resistance

and can interfere with the efficacy of drug combinations. Our conversation covers what tolerance is, the conditions that promote tolerance, how it leads to resistance and impacts drug combination therapies, and lastly, integrating this new understanding into clinical microbiology protocols. Here's our conversation. How would you describe the big question that this work sets out to address? What we were asking, are there other mutations that are occurring before when bacteria are exposed to antibiotics and actually promoting the fixation of

resistance mutation? And this would typically go completely undetected because they don't really confer resistance, but they are making the resistant mutation fix and develop much easier. So there are mutations that happen before resistance is developed and that promote resistance. Exactly. And then we found that it was more general that. It was not only mutation. It was a behavior of bacteria that can be conferred by a mutation, but can be also conferred by external condition and this behavior of bacteria is the ability to become dormant or to grow slower. And once they enter this state of dormancy or slow growth, it turns out that they have the ability to become resistance at a much higher rate than if they don't have disability.

Okay, the bacteria, they become dormant either through a change to their genotype or just a change to their phenotype. So how does becoming dormant allow them to become resistant? So at the beginning, it's counterintuitive, right? Because if you are dormant, then you replicate less. And if you replicate less, you require less mutations, right? But because our bags actually became dormant, the antibiotics is not acting upon them as it should act. And therefore, these bags just survive antibiotics. They're not able to grow in the presence of antibiotics. They survive an antibiotic treatment after a few hours after an antibiotic treatment concentration goes down very much. And then these dormant bacteria, they have the ability to exit these dormancy, start growing again. Now they can acquire

resistant mutations that are typically very rare to acquire, and now the bacteria become completely resistant to the treatment. Let me see if I've got this. So antibiotics usually target bacteria that are actively dividing. And when they're in the storm at state, all of a sudden now they're not affected by the antibiotics because they're not growing. And then once the concentration of the antibiotic in the environment is down, now they have this ability to grow. And since they were exposed to the antibiotic, they will acquire resistance to it. And now when you put the antibiotic back, maybe the next day when you have your second dose, it's too late. The bugs already have the resistance mutations and now they don't have to be dormant. Now they can actively divide and be

resistant. Exactly. What we see is that these tolerance promotes evolution of resistance in places where we would not expect resistance to evolve. I see. So the treatment regimes that we've developed to prevent resistance can actually lead to the development of tolerance, which can then lead to the development of resistance. Right, exactly. Yes, you can put it that way. Is there a set of conditions that lead to bacteria becoming tolerant instead of resistant? Our main guess is that the important factor here is immunosuppression of the patient.

So if you have a very good immune system, and even if the antibiotics are not killing all the bacteria and some of the dormant bacteria are still lingering on there or slowly growing, the immune system will take care of them, aiding the antibiotics, and infection should not be so long to allow this kind of evolution to occur. But in patients that are immunocompromised, the conditions are actually promoting this evolution of tolerance and then of resistance and allowing all these mutations to accumulate in the bacteria within the host. Antibiotics can only target growing cells. The immune systems can target non-growing cells.

So if you have that your immune system is working well, it doesn't matter if the bacteria are growing or not will still be able to recognize and destroy them. It's in cases where people don't have a really active immune system that the bacteria are able to evade and develop this dormancy phase that's not being hit by the antibiotics and can get this stepping stone to resistance. Yes, exactly. And so, you know, when we talk about antibiotics, we call them the magic bullets.

But we forget that a lot of the magic is actually our immune system, right? And if you are trying to use antibiotics to eradicate infections without an immune system, you're going to look. That's a really good point. And so this understanding came from experiment. we did in Bikers in the lab, this dormancy mutation, which we call tolerant mutation, is actually acting as a stepping stone for the acquisition of resistant mutations. And if you take away this ability to become dormant, they would never evolve resistance. So after this first series of work in the lab and in an environment that we control completely,

we wanted to know whether similar effects were going on in patients. And this is where we acquired from patients that were hospitalized under life-threatening blood infection. So we should say that the infection was a MRSA infection. So that's a methicillin-resistant dastaloccus orria. So what we commonly known as drug-resistant staff. So they knew from the jump that this bacteria was going to be hard to treat. It already had drug resistance to one drug, and the doctor started taking samples before any

treatment actually happened, and then every day following. And then you were able to recapitulate that in your lab setting. I want to get a little bit deeper into what you saw in the patient samples in terms of the treatment regime, and then what you learned from recapitulating that treatment regime in the lab. We worked in close collaboration with Dr. Mosquit Barmei, or is an individual. infectious disease doctor, and we reproduced exactly the treatment that was given to the same patient in the lab, and we could reproduce, actually, the evolutionary pass to tolerant mutation and then to a resistant mutation when the same drug regime was administered.

But actually, the story was more complicated. Now, what happened is that the initial drug regime was actually promoting this evolution of tolerance, Right. And then afterwards, what happens is that as they saw that the bacteria were not dying well enough, they changed a drug regime to this combination. Two drugs were who administered in the patients. But it was too late because once tolerance has evolved under the first treatment, having this drug combination was not helpful anymore because tolerance was messing around this drug combination. At the end of the day, the bacteria found yet another way to surprise us and to evolve resistant, not to have single drug, but even when two drugs were who administered. The drug combination work is really interesting here, and just another wrinkle to the study. So you knew the treatment regime that the person received wasn't successful, and that led to this kind of persistent infection. And now that you've studied the effect of those drug combinations in vitro,

what did you learn about how the drug combinations work? So when we went back with this understanding to the lab, then we expanded to also other bacteria and other drug combinations. And what we see is that there are specific drug interactions that are actually very good at blocking the evolution of resistance. But once you have these mutations for tolerance, it's another vulgar. right? So combination that were seemingly effective in vitro before are turning actually instead of blocking the evolution of resistance, they promote the pollution. Right. So it gives

a new understanding. It means that in order to pick up the right combination for the patient that you're treating, you need more information on the bacteria that are in the infection. If you're a doctor and you have discovered that your patient, their infection is now tolerant, so not resistant, but tolerant. What can they do? What's kind of the suite of treatments that they should pursue? As a guideline, if you have a patient that is immunosuppressed, give the combination even before tolerance have a ball, and then the combination prevents both resistance and tolerance. It's actually good for both. The other guideline is, okay, if you have a patient and you didn't catch up in time that it was immunosurprised, and now it has tolerance,

This knowledge actually can guide you now to a completely different combination that is actually able to kill this tolerant bacteria. There are certain antibiotics, not a lot of them, but certain antibiotics that are actually able to kill bacteria even if they don't grow. So these are the combination that you want to use. So how do you think about integrating this new knowledge of how tolerance of all and its impact on resistance into a clinical protocol or into clinical microbiology.

So, of course, this really very preliminary work that was done only on a very small number of patients. So clearly it needs to be expanded to a much larger number of patients and also on other infections. But our aim is actually within the toolbox that doctors already have at their disposition, They need to pick up drug combinations, and it's really not an easy task.

But within this toolbox, if you ask two different doctors, they will pick different combinations, right? So there is no clear rationale for the treatment. And what we propose is that maybe the fact that torrent strain evolve and go completely under the radar in the clinic is actually a factor to take into account that may make these choices much more. clear and unified among hospitals for the same infections, right? And therefore, one of the first thing that we'd like to do is a study linking this tolerant phenotype to clinical outcome in a supervised manner and to implement a routine detection of tolerant strains in the clinic. And for that, we develop this easy tolerance detection test. It's based on the disc diffusion an assay. So that actually should be able, within the tools that the microbiology lab has,

to tell the doctors whether the back that they are treating is actually tolerant or not. So this sounds like a similar protocol that's used to determine resistance. So it would be an easy addition to kind of what's already happening in a clinical microbiology lab. Exactly. Yeah. Now that we've talked about the background of tolerance, what happened in this patient, what you learned from the research, let's take a step back and talk about the implications of this research. This study was focused on, as we mentioned, this one patient with MRSA, but are there other pathogens or other types of bacteria that it would be important to think about drug

tolerance in the evolution of resistance? Yes. So there are pathogens that have been identified already in the evolution of tolerance, such as Pseudomonas, klepsiella, all kind of pathogen that infects cystic phoborosic patients. So it's another case where the immune system is not active enough and therefore tolerance can evolve. It's well known also that recurrent urinary track infections develop tolerance and it would be interesting to see whether it promotes evolution of resistance in these bags too. Are there other non-pathogenic-based conditions that this would be important for considering?

So cancer cells, the fact that they go into phenotypic state that may be much more tolerant to the treatment is something that has been done also by many other groups. So actually we thought about cancer very early on and it was really clear to us that anti-cancer drugs were actually like using antibiotics. bacteria without the help of the immune system, right? It's a lost battle. So clearly there the way to go was to recruit the immune system to help the drug. So to have a joint effort, as we have in typical bacterial infection between the immune system and the drug. And it's quite clear that the recalcitrance of cancer cells to treatment is really something that occurs a lot. And resistant develops almost inevitably.

Like antibiotics, anti-cancer drugs target fast-dividing cells and then target mechanisms that are involved in cell divisions. If the cells go dormant and stop dividing now, then the anti-cancer drugs aren't going to work. So there could be kind of a similar understanding where you need to flow the growth of the cells and also have the immune system involved to actually targeting and destroying the cells. which seems like it would have really important implications in things like CAR-T therapy, which are training your immune system to target your cancer. Your cancer cells, yes. So a combination therapy of the immune system targeting therapies

together with the drug that takes the load of the cancer cells down so that the immune system can do the rest of the work seems a priori as a good way to think about that. Of course, there are also anti-cancer drugs that act on non-growing cancer cells, but yeah, they are usually also much more toxic, right? If you could have our listeners take away one thing, what is the key take-home message of the work?

The key take-home message is that there is hope that by combining basic research in the lab that is really done in close proximity to what's going on in the patient, we will gain an understanding that should help direct treatment much better, and even tailor the treatment to the specific bag of the specific patients. And that's it for Journal Club this week. To recap, in the absence of a strong immune system, bacteria can develop tolerance, which allows them to survive antibiotic treatment and subsequently

evolve drug resistance. Further work is needed to confirm and extend the results described in this article, but they could be used as the basis for developing a standardized antibiotic prescription practice. Thank you to Natalie for joining me and thank you for listening.