Description: Peter J. Turnbaugh, PhD (Professor of Microbiology & Immunology, University of California, San Francisco) provides his insights to creating better figures, graphical abstracts, and cover art to visually communicate the work from his interdisciplinary group of microbiome researchers.
Nutrition, pharmacology, human microbiome, metagenomics
This webinar was recorded at VISUALIZE 2021, a virtual BioRender event dedicated to advancing communication in science.
Dr. Peter J. Turnbaugh from the University of California San Francisco and UCSF is near and dear to my heart. I have spent the majority of my career working for Bay Area technology companies and have actually done quite a bit of philanthropic work volunteering for the UCSF Children's Hospital.
Dr. Turnbaugh is our first principal investigator speaking today. He is known for his work on the metabolic activities performed by the trillions of microbes that colonize human adult bodies. As a publishing powerhouse with over 150 publications and approximately 100,000 citations, his expertise is sure to help level up how you visually communicate science. I believe he is internally ranked one of the top five researchers at his university, which is an amazing feat considering all the amazing researchers at UCSF. I look forward to his insights on creating better figures, graphical abstracts, and cover art in this next session. So with that, I'll pass it over to you, Dr. Turnbaugh.
Hey, thanks so much, and definitely have a lot to live up to after that introduction. So I'm happy to share with you the work that we've been doing related to the complex microbial world that lives in and on the human body. I'll try to highlight some of our efforts to use tools like BioRender and others to illustrate. It's a little bit intimidating giving and talking to a conference like this where there are also so many great artists, so I hope you won't judge me so much for the examples I'm showing. For example, the cover art here on the slide. If there was an outline, as the previous speakers have done, I'm going to share a little bit about how I learned to love the microbiome and how I got into this area of research. I'll give you some highlights from our recent research in our lab and then also talk a little bit about the challenges and examples of communicating microbiology research.
Unlike the previous speaker, I really did not have a seminal moment as a kid that told me I needed to be a microbiologist, and I sort of didn't know what I wanted to do. I also went to a liberal arts college in the northwest and I really loved the ability to sort of do all sorts of different things, including science and art. We were treated to all sorts of weird and inspiring statues on campus, such as an example of this fish sucking a lemon and some sort of weird bony-looking horse, and then we also had access to all the natural beauty like the Blue Mountains nearby. So that really gave me a strong foundation for the writing and illustration aspects of science that I do today in my career.
After I graduated from Whitman, I grew up in Oregon and so sitting at the beach thinking about what I should do with my life, and at the time, I was reading a lot about this fellow called Craig Venter that was really famous for popularizing this idea of metagenomics where he was sailing all over the world collecting seawater and sequencing the microbial communities that live on the surface of the ocean. This sort of global ocean survey got a lot of press attention, and it really sort of inspired me to think about microbial communities. Unfortunately, when I applied to grad school, I didn't get into San Francisco or Los Angeles. Instead, I got into St. Louis, Missouri, at a really great institution, Washington University. This sort of forced me to redirect my attention towards medicine, and I was really lucky to meet Jeffrey Gordon. He's a trained gastroenterologist and a medical doctor and is often referred to as the father of the microbiome field. He's really helped sort of push this area of science forward in the last couple of decades.
I was really lucky in that right around the time when I joined the Gordon lab, certainly before that, Ruth Ley had joined. She was a postdoctoral fellow with a really deep training in microbial ecology, and so I sort of think of myself as the offspring of Jeff and Ruth. So, you know, I think that sort of gave a really rich training both in the medical aspects of what we were working on in the gut and the more ecological theory.
As many of you probably appreciate, this area of microbiome research has really exploded over the last 15 to 20 years. What really was the first sort of study that used metagenomic sequencing, where you sort of take all the DNA in the sample and sequence the community, was published back in 2006. You can see a similar trend if you look at publications, but this is just a Google search term for microbiome, and so you can see back in 2006, really nobody was talking about the microbiome, and that interest has sort of continued to increase every time.
So one of the areas that we sort of immediately got interested in is what the relationship is between the food that we consume and the complex microbial communities that live within the gut. The first project that I worked on was led by Ruth, and she was interested in looking across the mammalian evolution at carnivores, herbivores, and omnivores and comparing their gut microbial communities. What we found is that you could sort of predict the types of microbes that are in the gut based on the diet of the mammal and, sort of, spanning all different types of animals and humans, look just like other omnivores. We went on to show that if you change the diet of an animal, you can rapidly change the gut microbial community. This is true in lab mice within a single day. If you feed a lab mouse a different diet, you can see a really rapid and dramatic change in the microbial community and later on we said that that's also true for wild mice. We found that if you sort of tracked the gut microbiome over time, it shows seasonal patterns that match the food that the mice are eating. Of course, humans, as I mentioned, are mammals. So the same thing is true in humans as we change the food that we eat, and there's a corresponding change in the microbes within the gap. So the question now is really, what is the impact of these diet-induced changes in the microbiome? One area that I've been personally really fascinated with is nutrition, sort of thinking about the components of these nutritional labels that we all eat or we all see on a given day when we choose what to eat. So you can see in this example the 230 number for calories. A lot of us appreciate the calories are important, but we maybe don't appreciate where these numbers come from. Calorie is defined by the energy needed to raise the temperature of a kilogram of water by one degree Celsius. This is a chemical concept that comes from measurements done in a machine called the calorimeter, which is basically just a sealed device where you put the food and fill it with oxygen and then ignite it. Then you measure the water in the bucket around the device and changes in temperature. That is a really useful sort of abstract way to think about diet, but it obviously ignores all the physiology that happens in the body and, in particular, how the food interacts with the microbiome. We can hypothesize the differences in the microbes in our gut may impact the energy we gain from our diet. So for some people, those 230 calorie I'm serving that may lead to weight gain, whereas for other people, that may lead to weight loss.
One of the ways that we can test whether or not differences in the microbiome translate to differences in a mammalian host is through the use of what's called motivic mice. In a microbiome transplantation experiment, we started with mice that were obese and due to either a genetic mutation in the leptin gene or the fact that we had fed them the high-fat, high-sugar diet. We then took the microbiomes out of the gut of these different animals and transplanted them into a new set of germ-free or sterile mice. What was really exciting is that the mice that got the microbiome from the obese owners ended up gaining twice as much body fat as the mice that had gotten the control microbial communities. That shows that even those recipient mice were eating exactly the same diet and actually did not change the amount of food they ate. They had a really different phenotype or physiology. Exciting publications, but I think the most exciting thing that came out of this research was that we got a Kathy cartoon. This is just the first line and sort of a larger cartoon. I'll walk you through it. Kathy's friend who I still didn't know her name says the fat Gene makes me overweight. Kathy says no, tried that one, only worked for a while. Her friend says the stress hormone makes me overindulge. Kathy says oh, haven't been there, believe that. Then her friend says the overly efficient intestinal microbes in my system make me gain while others lose goodbye miracle diets, hello miracle excuses. We found one that we can really stick with this time. That was really exciting that we got the validation of Kathy, but obviously, the debate within the scientific field continued. One of the things that we really lacked, although research advice has progressed a lot since then, we haven't had a lot of data in actual humans that this mechanism matters. One of the exciting papers that we published last year, together with the team of clinical collaborators, looked at the effects of underfeeding and oral vancomycin on the gut microbiome and nutrient absorption in actual human subjects. What was exciting about this is that we found that if you perturb the microbiome with vancomycin, that leads to measurable differences in the amount of calories taken out of the diet. That really fits with this idea that arose from these mice experiments. Hopefully, for where I came from and why we're excited about the microbiome, I want to give you a few highlights from the more recent research that we've been doing in ECSF. After starting my own lab, I decided that I wanted to shift our attention to what happens in the treatment of disease. So, I think one way to think about the most of the work in the microbiome field is very similar to classical studies of bacterial or viral pathogenesis. People are looking for individual microbes or sets of microbes that cause disease, but we think it's really important to think about what happens once somebody's sick and how differences in the gut microbiome might alter their response to medication or to dietary or surgical interventions. We've looked at this now across many of the common interventions that are used to treat obesity and other metabolic diseases. We published a paper last year looking at ketogenic diets, which are incredibly low-carbohydrate, high-fat diets that have been used in many different disease areas. We've looked at differences in the microbiome and responses to raw and cooked foods in response to gastric bypass surgery and also in response to a very low-calorie liquid diet sitter and referred to as a medically supervised fast. In each of these interventions, we've described differences in the gut microbial community that then go on to have important effects on the host. We are really excited about thinking beyond diet to the other more exotic compounds that we're all consuming and many of us on a daily basis. Work in our lab, as well as many others, has found that hundreds of different drugs that are used to treat disease can be modified through metabolism that get bacteria. One of the things that we're really excited about is that this spans many different disease areas. So, we're not just talking about diseases within.
The gut, like inflammatory bowel disease, but also diseases that occur in other organs like cancer, heart disease, or the brain for Parkinson's, is also sort of interesting just from a basic science perspective. These are really different chemicals and so presumably are acted on very, very different enzymes and perhaps many different microbes. To give you an example of one of the drugs that we've studied in detail, we collaborate a lot with Emily Belscus's lab in the Harvard Department of Chemistry. She’s an enzymologist really interested in the detailed mechanism through which bacteria metabolize drugs. One of the students in Emily's lab got really interested in levodopa, which is a really important drug used to treat Parkinson's that's been known for decades that it can be metabolized by bacteria in the gut prior to reaching the brain, its intended target. We really didn't understand how exactly this occurred. So what we were able to describe is that this Levodopa gets converted to dopamine by a bacterium called Enterococcus Vitalis and identifies a gene called or enzyme called tyrosine decarboxylase that's responsible. In addition to that, we found that dopamine then goes on to get converted to a compound called m-tyramine and that reaction is actually performed by a second bacteria named Eggerthella lenta, which is our sort of pet species in the lab and one of our favorite members of the gut microbiome. Interestingly, we found an enzyme called which we named ATH that's found in all strains of E. lenta, but in some strains, there's a point mutation that makes it inactive, and so it's only certain strains that carry the functional form of the enzyme that are capable of metabolizing the opening. This peripheral metabolism of the drug is really an issue clinically, and so there are other drugs that have been developed to block this reaction, and an example of that is Carbidopa. It's an FDA-approved medication that's often given to patients along with L-dopa to prevent this activity, and it works really well against mammalian enzymes, but it does not inhibit the bacterial enzyme so if you add lots of carbidopa, it will still metabolize. In contrast, Emily was able to identify a compound called AFMT, which is related to carbidopa and efficiently blocks the bacterial enzyme. Hopefully, this example sort of gives you a sense of how complex this system is but also how we now have the tools to turn on and off both the host and microbial metabolism of a given drug, and that could open up a lot of interesting new experiments in the lab. As I mentioned, we think that these interactions with drugs are really broadly relevant. A drug that we've worked on a lot is a cardiac drug called digoxin and where we showed the dietary supplementation the third genome can prevent the bacterial inactivation of these drugs, so there's sort of a diet microbiome drug interaction. We've also recently published Methotrexate, a really important drug used to treat rheumatoid arthritis and have found the differences in the microphone can help predict which patients will respond to Methotrexate. In addition to that, the changes in the microbiome in response to treatment may contribute to some of the beneficial effects of the drug, and we're also really interested in the role of the microbiome in chemotherapy or the treatment of drugs for cancer. We are studying how the microbiome may alter both the side effects of treatment and also interact with cancer cells to make them more sensitive to drugs. Finally, I want to share a few reasons why we think microbiome research is particularly challenging to communicate in and give you some examples of what we've done in the lab. I think, similar to the first couple of talks, the microbiome field is really complicated. So, for example, a diagram from a review article is shown, and what they're trying to do, I think, is to conceptualize all the different things that have been studied related to the microbiome, so thinking about age, thinking about different diseases, and all sorts of different bacteria involved. Another reason why I think the microbiome field is really complicated is that microbes interact with all the different organ systems in the body, and this is a pretty summary of that made by June Round at the University of Utah. You can see that the nervous system, skin, cardiovascular, all these different organ systems have been studied in their relationship to the microbiome. So, even though a microbe might be in the gut, it can have all these different effects throughout the body, and they can be really hard to think about, study in mechanistic detail, and also to visualize.
and then you also know a lot of the standard approaches we use to describe the complex metagenomic data are very obtuse. This is an example from a really highly cited paper describing sort of clusters of different microbial community structures. Some of these panels are sort of obvious to me as somebody who's deep into this area, but hopefully many of you would agree that it's sort of hard to know exactly what we're looking at. Even I'm not sure I fully appreciate without the figure legends what the differences are between channels a, b, and c. Panel e is something that is in a lot of microbiome papers, but even having worked in this area for so long, I have a hard time thinking through what exactly we're looking at. So the question is, can we use art to better communicate these complex findings and not just in terms of figures, but also how we talk about microbiome research? One of the really fun things that I got to do early on when we started my lab was to the company Millipore and they actually hired a cartoonist to try to illustrate the concepts that we're studying. This is sort of my favorite cartoon that they drew. Hopefully, it makes sense now after the introduction that we think of the bacteria as really being the garbage disposal in the gut, so they can eat anything that we feed them. That includes components of our diet that are left over and undigested from our own enzymes, but also many of the drugs that we take orally. They went on to draw all sorts of cool diagrams. For those of you that have thought about the microbiome, hopefully, these all make sense. There are trillions of microbes in the body, so every time we have a meal, we're eating together with trillions of other organisms. The 10 times number is sort of debated. There's a question of what the ratio is between microbial and human cells. These microbes are actually really heavy, so just the weight of the microbes is a lot in their body. Questions and sort of how we define humanity. One of my favorites is realizing that there's a lot of money to be made in the microbiome. A lot of people think stool is sort of a waste product, but there's a lot of exciting science out there. What was really fun is that many years later, Kristie Yu was a summer intern working in the lab and she also really had a gift for art. She was inspired to draw little cartoons of each of the people working in the lab at the time. I think these really nicely summarize what each person is so uniquely good at as well as what they were working on at the time. We've also tried to incorporate better art into our research reports. This is an example of a diagram that was in the supplement of a recently published paper by Elizabeth and the chemist. I think this is not made in BioRender, it's an Illustrator one, but hopefully, you can appreciate what she was studying.
Elizabeth was studying a sort of metabolic pathway taking pinoresinol, which is a compound found at high levels in flax and sesame seeds and converting it to intradial and interlectin, which are two compounds that are called phytoestrogens and have been studied a lot in the context of breast cancer and other diseases. What's really interesting is that multiple bacteria, and so here are four different species, have to come together to complete this pathway. Then to sort of illustrate each of the enzymes and transporters that we've identified and where we think they might be in the cell. But obviously, that may be too much detail for people, so we tried to be more abstract when we were starting to make cover art for this study. Working with Ella Imru Studio lead, you know, thought that this is really just a conveyor belt for metabolism and so thinking about how each component in the pathway is sort of handed off to these different bacteria. Another example comes from our work on these ketogenic diets, and this is actually the graphical abstract from the paper that shows that on a control diet, we found that there's a particular type of bacteria called bifidobacteria at higher levels induce a type of T Cell called T helper 17 cells. On the other hand, a ketogenic diet production of these specific chemicals called ketobodies or beta-hydroxybutyrate specifically inhibit the growth of these bacteria and lift that induction of T cells.
I think maybe because it was the start of a pandemic and we were a little stir crazy we decided to make some art based on those findings. This is sort of, you know, meant to show you that the Ketone bodies are sort of raining down from the sky, staring away the bifidobacteria and causing the T cells to sort of stay at home.
But I think for those of you who are academic scientists in the audience, you'll appreciate that these tools are really best for getting grants. This is something my daughter drew when she was seven and she says, "When my dad doesn't work, he has grants to write, sometimes he does not write grants." We really found that these better graphics can help us get grants funded.
This is an example of a diagram that Cecilia Noeker, a postdoc in our lab, submitted to the NIH. I think it nicely summarizes her overall goal as a postdoctoral fellow. In her first name, she was focused on a deep dive into the metabolism of one particular species, E. lenta. In Aim 2, she plans to expand out and think about how that fits within the broader community. In Aim 3, she is looking at that across human populations. Reviewers also appreciated this art and she got a score of 15, which is if you don't know the NIH, the best score is 10 and the worst score is 90.
And Than, an MD PhD student in the lab, also used BioRender to create some art for his grant. This is a little bit more complicated, but he found that gut bacteria make inhibitors of an efflux transporter called P-glycoprotein. That's really important for drug resistance and cancer as well as just the absorption of compounds in the gut. He's really interested in understanding the mechanisms responsible, and so each of these arrows indicates one potential hypothesis. It could be blocking transcription, trafficking, the direct activity of the transporter, or maybe permitting its degradation. The reviewers were so excited that they gave him a perfect score of 10.
To recap, hopefully, I've convinced you that it's a great time to be a microbiome scientist. Microbiome science is changing how we view nutrition and pharmacology, as well as many different areas of study. Our success really requires a commitment to communication, and I think we'll be aided by the really cool tools we saw earlier today in BioRender. So with that, thanks so much for inviting me to this really exciting meeting!