Passion in Science Humanitarian Award winner, Nathan Schoepp, describes his research to develop rapid point-of-care diagnostics to determine antibiotic resistance in clinical samples.
Thanks for joining us. I'm Lydia Morrison, the host of the NEB podcast, Lessons from Lab and Life, and I hope that our podcast offers you some new perspective. Today I'm joined by my colleague Nathan Tanner. Nathan's a Senior Scientist in the NEB Nucleic Acid Replication Research Group. Thanks for joining me, Nathan.
Thanks for having me.
Nathan and I had the opportunity to interview another Nathan, Nathan Schoepp, a graduate student at Caltech, who is one of the 2019 Passion and Science Award winners.
Nathan's work is a great demonstration of how loop-mediated isothermal amplification, or LAMP, can be used for point-of-care testing. He developed an assay that takes under 30 minutes. It can accurately measure antibiotic resistance from real clinical samples, and this allows antibiotics to be prescribed more conservatively for better antibiotic stewardship, and used more effectively for better patient outcomes.
Hi, Nathan and Nathan. Thank you both so much for being here today.
H2 Header: Antibiotic Resistance
So I wanted to ask you a little bit more about antibiotic resistance, and how concerned the public should be with antibiotic resistance. Is it something we're talking about being very concerned about right now, or something we're thinking about being concerned about in 10 years?
Yeah, I think it's definitely something we need to be concerned about right now. So I think what helps to realize that is an appreciation for how important antibiotics are to modern medicine in a lot of scenarios that maybe your average healthier person doesn't think about. But in everything from neonatal care to postoperative care to any immunocompromised case, antibiotics serve a critical role. When we're running out, or when resistance rates are rising to the point that physicians can't prescribe the antibiotics they're used to using, and so they have to escalate treatment, that becomes a really big problem. It's already becoming a problem. I think we're just sort of realizing that recently and now.
So as of today, how does a physician decide what antibiotic they're going to prescribe?
It depends on the situation, obviously. In your outpatient scenario or an STI clinic, you can imagine people come in, they get diagnosed with an infection, a bacterial infection, and the physician doesn't necessarily have a relationship with them. So what that means is they have to make a prescription on the spot, and so they do empirical prescribing, which means they take the evidence at hand and then decide which antibiotic to prescribe. But obviously in those situations they're going to prioritize patient outcome over anything else. Even stewardship, which a lot of physicians care about and are very aware of, but it's just a hard balance to strike when they don't have the information they need before prescribing the antibiotic.
H2 Header: Antibiotic Stewardship
So what does stewardship mean in the realm of antibiotics?
Yeah, that's a good, good question. So stewardship just means maximal rational use of antibiotics. So preserving the antibiotics we have so that we can maximize their lifetimes and reduce the emergence and spread of antibiotic resistance.
Are there alternative methods for treatment that are available other than antibiotics?
Yeah. Again, it depends on the situation. So in some cases antibiotics are going to be the treatment option. They're sort of the only treatment option, and that's what has made them so important. But there are a lot of preemptive measures we can take. So reducing antibiotic usage in the environment is one, vaccines are another huge one. So if you can prevent the infections beforehand, obviously you don't need to use antibiotics. So those strategies are really the most effective, and most promising for reducing antibiotic usage in the longterm. But that's not going to prevent many, many cases where you still have a bacterial infection that needs to be treated with antibiotics.
So what does that mean for society as a whole? It seems like we're sort of climbing this ladder toward more and more antibiotic resistance?
So basically, it means that as resistance emerges, so you get new mechanisms or transfer of mechanisms from one bacteria to another that physicians ... So as resistance rates increase, physicians have fewer and fewer options that they can safely use. By safely, I mean that they can guarantee the antibiotic they're prescribing is going to treat their patients infection. So when you can't guarantee the outcome, what is done, again depending on the scenario, but quite often is, treatment will be escalated to a broader spectrum or a second or third line antibiotic. Then obviously we start using those more and the cycle continues, and you get new mechanisms of resistance emerging and spreading to those antibiotics as well. So it speeds up the resistance cycle, which is a natural cycle I think it's important to remember, but that's really the issue.
Then all of a sudden you end up where you only have one viable treatment option left, in terms of the recommended treatments, and so that's really the main issue.
H2 Header: Antibiotic Resistance is Natural
Are there a lot of new antibiotics in the pipeline, new treatments that come to hopefully help with some of this, or is that all a long way off?
Yeah, you would hope. There aren't a lot. That's a whole other issue, and one that absolutely needs to be addressed as well. But there's not necessarily a lot of motivation for making new antibiotics. This sort of goes back to another solution to the problem, or a piece of the puzzle sort of is, we need increased capital going towards this problem, both human and financial. So there are incentive programs that basically make it profitable immediately if a pharmaceutical company for example, is to develop a new antibiotic, they can get a reward or something like that, so that they can pay for a lot of their R&D, because what happens now if they were to do that, a lot of times that antibiotic will just sit on the shelf.
It will become the new last line antibiotic that everyone wants to preserve, and so it's not being used, they're not making money off it. So it's sort of understandable that there's not a lot of motivation there, even though it's a huge problem.
I know there's a good deal of awareness with physicians and medical treatment, but what about in say, agricultural uses and things like that? Where there's a lot of antibiotic use that perhaps isn't necessary, but it's certainly a big driver of this problem?
Right, yeah. Again I think the problem there is anytime you use antibiotics, you're giving all these bacteria the chance to develop resistance, and as soon as you develop resistance in one pathogen or a new mechanism emerges, a lot of times it can spread and then eventually find its way back to cases where bacteria are infecting people, and again, that's really the heart of the issue. So I think that's an excellent point as well, environmental stewardship as well as, I don't know what you might call, personal stewardship in treating people.
And developing resistance, it's hard to fight against. It's natural, right? The bacteria are supposed to do that.
Right, yeah. They're pretty incredible in that way. I sort of have a lot of respect you could almost say, for bacteria. Yeah. It's totally natural, and that's another thing to appreciate about antibiotics. They're one of the few drugs that the more you use them, the less efficacious they become, which is just not the case for many, many drugs. So they do have a fixed lifetime. But we do have enough that if we were to really maximize their lifetime, or if you knew every time you prescribed the full profile, sorry, the full susceptibility profile, so you knew exactly which antibiotics you could use, then physicians could make really rational choices about which to prescribe, and they can guarantee patient outcome, which is obviously the priority.
H2 Header: Monitoring Antibiotic Resistance
So how do we currently monitor antibiotic resistance?
So currently what's done, if you want to find out which antibiotics a pathogen is resistant or susceptible to, they'll take a patient sample like urine for example. They'll ship it to a central laboratory that will isolate the bacteria from the sample. They'll grow it up, and then they expose it to a bunch of different antibiotics at different concentrations. That gives you basically what's called a susceptibility profile for that infecting bacteria, and they know for a given antibiotic, what concentration will kill that bacteria. Then from that they can categorize it as either susceptible or resistant or intermediate, and then they can prescribe based on that information. The problem is obviously the speed at which that happens. A lot of times physicians and/or patients can't wait long enough, two to three days, and so that information doesn't end up getting used at all. But yeah, so that determines the bacterial phenotype, which is really what physicians care the most about.
So can you tell our listeners how your method of antibiotic resistance detection works?
Sure. So what we started thinking about in the lab, a lot of other people including myself when this started, was a way we could determine phenotype like I described. So that means the response of bacteria to antibiotics from measuring DNA concentration was the initial method. The reason that's so advantageous is you can make those measurements really quickly after a very short antibiotic exposure. So we're still looking at the phenotype, but now we can shorten the assay time really significantly to a point under half an hour that we think is applicable to what's called the point of care, which is the place antibiotics are being prescribed.
So in theory, if we were to translate this methodology to a device that we could put into clinics, doctors could take the patient sample and in half an hour run it, and now know the susceptibility profile of the bacteria that is infecting the patient that are working with, and then make a very informed prescription initially. So they don't have to go back, they don't have to wait for results, they don't have to change the prescription, they don't have to recontact that patient, and they still have all the information that they need to make the call.
So can you test multiple antibiotics in the same test, so you'll know sort of like where the threshold is, and then the possible antibiotics that would treat an infection?
Yeah. So right now we can test multiple antibiotics. It is really just a matter of sort of multiplexing the same methodologies. So the important thing is validating that you get a response, and that we can actually test for susceptibility to a lot of these different antibiotic classes, which wasn't obvious immediately. That's one of the first proofs of concept we performed, was showing that even for antibiotics that don't directly impact nucleic acid replication, you can still measure nucleic acid concentration on the backend to determine susceptibility. So that is really more of sort of piecing it all together at the end, figuring out how to put this all on a device. But it's such a simple workflow that we're pretty confident that that's possible.
H2 Header: Point of Care Antibiotic Resistance Testing
You mentioned doing things at the point of care and assays for that kind of setting. You touched on speed, they've got to be fast, they've got to be really simple, and they can't be too expensive or requires sophisticated instrumentation. So how does your method handle those hurdles?
That's a really good question, and I think also something that most people, including myself for sure did not appreciate. But if anything is going to be a method that's used actually at the point of care, it needs to be pretty much push button. You put a sample in, you get a result out. We were talking to one of our collaborators, at a big microbiology lab, and they said if it's bigger than a shoe box, they're just not going to have space for it. That was also eye opening cause that's obviously not very large. So these have to be very compact, very simple and very rapid.
We touched on the time earlier. So the way our method is addressing these issues, which you really have to care about if you're going to make a point of care diagnostic or assay, is basically thinking about them up front, and trying to stick with something that's a very simple workflow. So for example, our method only requires one splitting step with the control, and then a treated or antibiotic treated sample for each antibiotic you want to test. So that's relatively simple fluidic handling there. Then you need an extraction step to isolate the DNA, in the case of our UTI work. So again, that's relatively simple. We use a one-step extraction buffer that's compatible with our readout methodology.
So yeah, I think to answer the question, basically we've thought about these things up front. We don't require a microscope on the backend or a mass spec or things like that, which obviously give you incredible capabilities in certain scenarios. But for something, again, at the point of care, we're really trying to stick with things that we think can be incorporated into again, that sort of shoe box size device.
How does affordability factor into thinking about fitting things into that shoe box size, and making them achievable in every doctor's office?
Right. Again, that's a great question, and we think primarily we're focusing mainly on the methodology right now. But again, we don't require any fancy instrumentation to put it plainly. Then in terms of reagent cost we consider it, but it's a down the road problem that we think with economies of scale etc. wouldn't be too much of an issue. Again, there's nothing super expensive in our assays. These would be about as cheap hopefully as the current broth microdilution method. You've got pay for the antibiotics that you're going to test, and then you're obviously going to have some consumable part of your device that would probably be injection molded plastic. But beyond that you could think about having a reusable base station or section of the device that you don't throw away, that has the heating and the analysis components in it. So hopefully the test itself, and the cost of each single test would be pretty low cost.
So a big part of what you're trying to do is doing everything in 30 minutes or less, and detecting DNA in that kind of window. That's pretty fast. So how do you go about making an amplification reaction that can work in that kind of time?
Yeah, that's a really great point. So I mentioned earlier our test is phenotypic, which gives it its generality. But it means that a big chunk of our 30 minute window is already taken up just exposing the living bacteria to antibiotics, because we're measuring the actual response. What that means is we have even less time to do the DNA amplification in order to quantify it, which we have to do on the back end, but still in that 30 minute window. So what we've come up with, or what we've used, utilized I guess, are these isothermal DNA amplification methodologies.
There's quite a few these days. We optimize one called loop mediated isothermal amplification or LAMP, using NEB reagents. And what we found was that through a little bit of optimization and careful primer design, we could get really, really rapid amplification reactions such that we could amplify and count single molecules of DNA in five to eight minutes. That sort of power of speed I think is really special, and really unique to DNA. I can't think of another biomolecule where you can quantify single molecules that rapidly, and that allowed us really to get under that 30 minute window. We're really tight on time, and so every minute counts. So every time you have an incremental improvement in your nucleic acid measurement, your DNA measurement, that means you either have more time to do a longer exposure, which can be important depending on the organism you're looking at, or it's just enabling. You just have more time in that 30 minute window where you're so tight for time. So those reagents and the optimization, and the methodologies themselves have been really critical in meeting that bar.
In addition to the speed factor, I would think an isothermal method is attractive because doesn't require any sophisticated heating and cooling, you can just put it at a temperature and that's it.
Right, that's a 100% correct. Just like we'd like to not use any large microscopes or fancier equipment. It's the exact same with heat control, and if we don't have to thermal cycle, everything becomes a lot simpler. There are also methodologies is as you know, very well out there for doing a really robust but rapid and simple, colorimetric readout, which you could do by eye, or there are lateral flow assays for looking at amplification of DNA as well. We've thought about incorporating those, one issue we have is we do need to make a very sensitive measurement ... Sorry, a very specific measurement of DNA concentration. So right now to achieve that sensitivity and specificity, we have to stick with measuring basically counting single molecules doing what's called a digital assay. But we've thought about ways to incorporate even simpler readout, because again, you can't be too simple for point of care. So that's been exciting to think about.
So it sounds like a lot of the proof of point research that you've done for this has been with UTIs and E. coli, but is this the type of technology that could easily be applied to other point of care diagnostics?
Yeah, I think there's certainly a lot of ... I mean you always learn something when you build a new diagnostic, and there's certainly a ton of applications for rapid or optimized LAMP methodologies for example, for measuring DNA in terms of identification of different infections. That alone is very helpful in a lot of situations. So that's one alternative application I can imagine. Yeah, and then beyond that, another challenge I thought you might ask was different organisms, cause E. coli is pretty fast growing, rapidly growing and that is certainly a challenge. But it's one we're working on, and we've come up with some, I think of as fun methodologies to, that weren't trivial to develop.
But now we've got them working pretty well where we can look at basically secondary effects. We can look at what the antibiotic impacts, and then think of ways to link that to DNA concentration, which we can obviously measure, as I mentioned really quickly on the back end. So now we've applied that to some slower growing organisms like gonorrhea, and then a family of bacteria called Enterobacteriaceae. So those are both exciting directions that should be coming out soon.
Thanks so much for being here today, Nathan.
And honestly, thank you for being a champion of antibiotic stewardship, and for all your efforts in applying LAMP technology to point of care diagnostics, in order to preserve antibiotics for future generations.
That's what we hope.
Thanks for enjoying this episode of the NEB podcast. Be sure to tune in next time when I'll be joined by Jonathan Gootenberg and Omar Abudayyeh, the first McGovern fellows of the McGovern Institute for Brain Research at MIT. They'll share with us their thoughts on applying CRISPR gene editing technology to study the mechanisms of aging