The professor of chemistry and chemical biology at Harvard University and Blavatnik National Laureate discussed the work she and colleagues have done on microbiome metabolism and what upcoming plans to pursue this further could achieve.
Emily Balskus, MPhil, PhD
In June, Emily Balskus, MPhil, PhD, was named a Laureate of the Blavatnik National Awards for Young Scientists for recent work into gut microbiota, which identified novel chemistry and unveiled more of its role in disease. The professor of chemistry and chemical biology at Harvard University received the largest unrestricted scientific prize in the US from the Biocodex Microbiota Foundation, for her work.
The work, published in Science, helped Balskus and colleagues identify a specific organism in the gut which may be responsible for peripheral levodopa metabolism in some patients with Parkinson disease. To find out more about the work, how these discoveries were made, and what Blaskus and her colleagues are doing going forward, NeurologyLive spoke with her in an interview.
Emily Balskus, MPhil, PhD: I am a chemist by training, I’m in the chemistry department at Harvard, and my training initially was in organic chemistry, so as a chemist I was trained to make molecules in the lab, and I became really curious as a grad student about how nature made molecules. So, I became interested in kind of understanding how microbes did chemistry. Initially, I studied pathways and enzymes from environmental microbes, but then, towards the end of my postdoc, I started to get really curious about human-associated microbes in the microbiome. When I started my lab, probably over 8 years ago, now, I decided to study the chemistry of human gut microbes, and we've been working on this ever since.
We've been studying many, many facets of gut bacterial chemistry. As we've gotten more and more into this field, we've become more and more aware of how much we really don't understand it. We take the opportunity to study everything from how the microbiome metabolizes drugs and dietary compounds, to how they make complex toxic molecules. There’s a whole range of different chemistry with different implications for human health and disease.
The ultimate goal of the work is to improve the efficacy of levodopa treatment for Parkinson patients by, hopefully, developing a therapeutic that prevents gut bacteria from consuming the drug.
The work that my lab and I have done, that led up to the application for this grant and receiving this really generous award—we had become curious in the context of just thinking about previously reported examples of natural drug metabolism. We'd recognize that levodopa was among the therapeutics where there was some evidence that the gut microbiome might be actually contributing to drug metabolism. But the details—things like which microbes are responsible, which genes and enzymes within the microbes were responsible for forming chemistry, the variability of those organisms that came from patient to patient—this was sort of totally unknown.
We became really excited about the idea of trying to first, just really characterize this metabolic process. I should mention that this is a project that was led by a senior graduate student in my lab, Vayu Maini Rekdal. He is the first author on a paper we recently published in Science that describes the discoveries. We were able to, first, figure out that it was, we think primarily, one organism in the gut microbiome, Enterococcus faecalis, that's responsible for metabolizing levodopa. This particular bacterium uses an enzyme to convert levodopa to dopamine in the small intestine. This is the same transformation that host enzymes perform. In the host, it's critical that dopamine production happens in the brain, but when it happens in the periphery—and a lot of the peripheral metabolism was known to happen in the gut—that's very problematic because once dopamine is generated outside of the brain, it can't cross the blood-brain barrier, where it needs to be to have a beneficial effect. Once we had found the specific bacteria and really understood the specific enzyme, we could sort of study how that worked. And being chemists, we're really excited about that.
We also then, by examining Parkinson's patient gut microbiome samples, could sort of survey different samples for their ability to metabolize levodopa, and then ask the question of whether the abundance E. faecalis or the abundance of the gene encoding this particular enzyme really correlated with metabolism. Things got interesting when we ask those questions. We found that levodopa metabolism, in a small number of Parkinson's patient gut microbiomes, as well as gut microbiomes from neurologically healthy individuals, really was variable. I believe it was about 60% of patients that had metabolism and then the rest had no metabolism.
Then, in the abundance of the specific bacteria and gene we’d identified also correlated with that activity, which is exciting to us because it suggests that organism really was the one that was responsible. With the organism identified, we then could sort of think about how could we possibly stop this metabolism? Could there be a possibility that we could do something like identify a small molecule that might prevent this gut microbiome metabolism from happening?
Before we really started to think about designing new molecules, we actually took a look at some of the existing drugs that Parkinson's patients take to prevent peripheral decarboxylation by the host enzyme. These are drugs like carbidopa, and we're intrigued to find that drugs like carbidopa actually weren't effective for the gut bacteria. They didn't stop the conversion of levodopa into dopamine, and we sort of showed through examining the purified enzyme that this is due probably, in large part, to the fact that they're not actually great inhibitors of the bacterial enzyme, and the fact they probably don't get into bacterial cells very well. What this means is that patients who are taking levodopa- and carbidopa-based combination therapies, if they have a microbiome that can consume the drugs, the therapies they're taking probably are not preventing that peripheral decarboxylation. That would be consistent with some of the things we know about levodopa therapy. We know that, that even in patients taking these combination therapies, up to 60% of levodopa can still be metabolized in the periphery. This observation, in combination with the known variability of patient response to levodopa and the variability we see in the metabolisms in the gut microbiome, it altogether supports the hypothesis that differences in patient response to levodopa-based therapies could have their roots in differences in metabolism by the microbiome.
Then we asked ourselves, all right, using existing drugs doesn't work to prevent this from happening, could we design or develop a molecule that could stop this from happening? In our preliminary work, we were able to take what we knew—about how the E. faecalis enzyme worked and the substrate that it preferred and how it did chemistry—to actually find a molecule that will work to inhibit the activity of this enzyme and prevent metabolism of levodopa by the microbiome. We showed that showed this in complex microbiome samples, we should also show that at least in a very, very, very simple animal model, that you could coadminister levodopa, carbidopa, and our compound—which is called α-fluoromethyltyrosine (AFMT)—that you could increase serum levodopa levels.
Now we plan to use this grant money to fund efforts to really optimize that lead compound into something that could potentially be a candidate therapeutic to administer alongside levodopa and carbidopa. It would be a therapeutic that would specifically prevent peripheral metabolism of levodopa by the microbiome. We also would like to learn more with this grant about how this how AFMT and any optimized molecules, how they're actually working in the microbiome, so to test them in more complex and more relevant animal models than the one we've looked at so far.
It's always really challenging to predict how rapidly drug and development efforts will happen in this case. And I'm hopeful that this could potentially be on the faster side of the drug development process, in part, because I think we have a good understanding of this type of enzyme. There have been drugs that have been developed for related enzymes, so there is a therapeutic proof of concept for engaging this type of target. There's also a lot of therapeutic precedent for drugs that inhibit peripheral metabolism by the host. I think that in terms of drug mode of action, and the therapeutic concept, it's not too much of a leap to move to thinking about targeting the microbiome.
Hopefully, we'll be able to do enough in my lab to interest either industrial partner or potentially think about doing something like building our own company around this idea in the future. I'm excited by our preliminary results, and I think I have a terrific team within my lab that's also very excited about doing the medicinal chemistry needed to optimize this compound. Hopefully, we'll be able to do the studies in the next few years that will really tell whether this really has the potential to move forward into clinical studies. That's really our goal.
Maini Rekdal V, Bess EN, Bisanz JE, et al. Discovery and inhibition of an interspecies gut bacterial pathway for Levodopa metabolism. Science. 2019;364:eaau6323. doi: 10.1126/science.aau6323