Mysteries of the Microbiome

Resonance is a student-run podcast aimed at showcasing the science at Baylor through the eyes of young professionals. Each episode is written and recorded by students who have a passion for research and the medical community. Guests on the show include both clinical and basic science research faculty who are experts in their fields.

In this episode, Dr. Kjersti Aagaard discusses her journey into microbiome research and specifically into placental microbiome research.

Erik: And we're here.

Brandon: We are here.

Erik: We are the Baylor College of Medicine Resonance podcast. I am one of your hosts, and the chief sound engineer, Erik Anderson.

Brandon: And I'm another host, Brandon Garcia, as well as the admin.

Jennifer: I am head writer, and my name is Jennifer Deger.

Erik: And we are all students here at Baylor College of Medicine, and this is the inaugural episode of the new student hosted and run podcast, with our aim being to offer an additional medium to showcase the research and clinical work of the faculty here at the college.

Brandon: Yeah, and we just want to give a perspective of the students of what's going on here at Baylor. There's tons of crazy and awesome and cool things are going on that really we think people need to know about. There's a lot of other ways, so you hear somewhat about it but we wanted to give a student perspective because we want students to know what's out there.

Jennifer: Yeah, and if anyone's interested in research, it's a really good way to figure out not only the work of different faculty, but also their personalities.

Erik: Yeah, and so to give a quick background about the format of the podcast, what we're going to do here is start with a quick introduction, we call it the roundtable, where we talk about the podcasts that are the faculty member and their research, and give a little bit of background. And then we will go to about a 30-minute interview with a faculty, or really anybody here at the college, on whatever topic that we're gonna be talking about. So, for instance, today we're going to be talking with Dr. Kjersti Aagaard about the microbiome, so the three of us have worked on the episode to help write it, so we're going to give a little bit background on it.

Brandon: Yes, so the microbiome is this collection of microbial symbiotes that live within us and on us that kind of interact in our whole lives. Instead of us instead of thinking of the human or human as being a single entity that's going about and living on the earth, we're actually an ecosystem.

Erik: Definitely, as I was researching for this episode I learned that we've basically known about the microbiome or microbes on our body in some sense, since the 17th century with the great Antoni van Leeuwenhoek. Right? I think I said that right. Yeah, and so he apparently was comparing his oral and fecal even microbiota in some of his early categorizations with his with his microscope

Jennifer: Someone had to do it.

Erik: Somebody had to do it. So we've really known about microbes living on the body in some fashion for a very long time, but it wasn't really until computers and everybody hears Big Data until all of that came along that really allowed us to start to expand and detect microbes in a new fashion and a new light.

Brandon: You're talking about like whole genome sequencing like Illumina type stuff, yeah?

Erik: 16s RNA sequencing really, I mean that's I think what we're gonna be talking to Dr. Aagaard about. And so yeah, with just pipelines and mathematics as we are able to analyze larger datasets, this has really opened up our ability to study the microbiome from just from just the 19th century put a bacteria on to a plate and then see what grows.

Jennifer: Yeah but it's crazy how computers have helped with the sequencing technology so much

Erik: Yeah.

Jennifer: Just the growth in the last like 50, now I guess almost 70 years.

Brandon: It's really been amazing and it can't be understated just how beneficial this information is going to be because you can go and read papers about the gut-brain axis, and Dr. Aagaard is gonna talk a little bit about how the metabolism of these bacteria, this ecosystem within us can actually impact how we think, how we feel, and some of the things that we've got going on it's really fairly interesting. And she got her start from something called the human microbiome project, which was headed up here at Baylor along with several other institutions like the Broad Institute, Washington University, and the Craig Venter Institute. They wanted to, just like how we had the human microbiome, not the human microbiome, the Human Genome Project, the human microbiome project plays a similar role. They started in 2008, decided to do a reference library of like 3000 different samples, and Dr. Aagaard is going to talk a little bit about her involvement and the role she played here at Baylor in accomplishing that. And now we have this reference metabolomic and microbiome genome set that we can compare and start looking at how our microbiome can influence disease.

Erik: Yeah that's exactly right, and so I guess to get to the next part of the podcast we would like to introduce Dr. Kjersti Aagaard. Dr. Aagaard completed her MD from the University of Minnesota and Ph.D. from the Mayo Clinic in their joint degree program. She completed her residency training at the University of Minnesota, fellowship in maternal-fetal medicine at the University of Utah, and a master's in Clinical Investigation from the University of Utah. She is a professor of obstetrics and gynecology, and subspecializes in maternal-fetal medicine. Her lab's focus is on the interaction that the microbiome has on preterm birth and in utero environment, and epigenetics, and fetal programming and development. Her lab was the first to provide sequencing evidence for a placental microbiome, so without further ado, here is our interview with Dr. Aagaard.

Erik: Thank you, first off for doing this and being here, and so Dr. Aagaard is close to our hearts as the co-director, one of the co-directors of the MD/Ph.D. program here at Baylor. Before we get going, if we could just hear kind of how's your research developed over your career from, what maybe you were doing in your Ph.D. to your fellowship - to now.

Dr. Aagaard: Right, and thanks you, guys for doing this. This is really, this is going to be fun and going to be a nice new contribution to the Baylor community.

Brandon: We hope so.

Dr. Aagaard: Yeah, this is awesome! I love it, great innovation on your part. So I think maybe how all of the different work and research I've done will look more like a tapestry and less like a kind of scattered rug on the floor by the time I get to the end of my career, but I can definitely see how the strands are weaving together at this point. So early, very early on, my kind of first foray into biomedical research was with Dr. Dorothy Patton at the University of Washington, when I was a summer undergraduate student. She was a phenomenal investigator, she worked in non-human primates, and we studied how chlamydial infections in bulbar tissue of the eye and in fimbria from the fallopian tubes could lead to things like blindness in babies, and also pelvic inflammatory disease and infertility. And that created an initial love of both reproductive science and infections and immunity. And so I, after that summer experience, went back to my small undergraduate liberal arts college and continued my undergraduate career as a rower and as a biology major, and then the next summer went to the Mayo Clinic in Rochester Minnesota where I started looking at T cell immunity and became really interested in how our immune systems become educated and poised over time to either battle infection, or tolerate ourselves. And during that time I really came to love the immunology program at Mayo, it was an incredibly exciting time to be there and launched into eventually what became my graduate training. There was an incredible cadre of investigators at the time at Mayo, people like Jerry Gleich, who really understood how eosinophils worked, and Connie Weyand, who were doing groundbreaking work on Rheumatology, my own thesis advisor Diane Jelinek, who was a expert in B cell biology. Let's think Paul Liebman, who discovered natural killer cells and how you can silence parts of the immune system. These were just really, really exciting times and so eventually went on did my Ph.D. in immunology and was really interested in the development of your humoral immune system and B cell biology.

Brandon: Okay, so you started off in immunology for your Ph.D., I guess because of your interest in infectious disease it's kind of the direction you went. So how did you get to the Microbiome? And I have this question too of like how do, is it typical for people who are doing most of the research of the microbiome, did they start out as Immunologists and make the jump, or where does that field come from?

Dr. Aagaard: Yeah, so that's a really great question. So lots of folks who eventually get around to doing work in the microbiome either come out of microbiology, they're microbiologists that have a long-standing interest in bacteria and viruses, which has always been a clinical interest of mine. A lot of them are immunologists because they're fundamentally interested in how do you know the difference between a commensal or a healthy microbe, and a potentially pathogenic, or pathobiont, or something can cause harm, and computational biologists. So it's a really heavy computational field, and so a lot of people who do microbiome research are computational biologists. So those are kind of the three big-picture team of folks who foray into that field.

Brandon: What do you mean by computational?

Dr. Aagaard: Oh, good question. So computational biologists, some people call them data scientists or big data researchers. They're people who are used to not just doing bioinformatics or moving data from point A to point B, but also people who have some usually pretty strong biostatistical and mathematical background so that they can analyze and account for variants and things like that in large datasets, great question.

Erik: Well, so I'm kind of curious, when did people start? Because you were talking about kind of good bacteria versus bad bacteria and then that's sort of now really in the mainstream. But like when did this thinking really start to, like, when did you notice people start actually talking about like, “oh this thing is the microbiome.” I guess I'm not looking so much for the coinage, but like when it sort of came on your radar?

Dr. Aagaard: Yeah, so there's kind of two answers to that. So I think for a lot of us it was always on our radar really because we've recognized since we could first look through a microscope, right, that microbes lived in communities and that we had bajillions of inhabitants on our body that were microbial. So we've known that for a long time, and it really is a fundamental question over how is it that we don't try to get rid of some microbes, but we do try to get rid of others.

Erik: Right.

Dr. Aagaard: And that has a lot of parlance into cancer as well, right?

Erik: Yeah.

Dr. Aagaard: Why does our immune system not do anything when it's not a cancer cell, and then when it goes a little bit out of whack, now we know how to do something? So I think those are really part and parcel of each other, and so it's been on people's horizon for a long time. I would say I'd answer the question about how did I become interested in it in kind of three ways. So we talked about each of kind of the three different big-picture of folks who certainly count for a reasonable populace within the microbiome science field. So I definitely had the immunology background, I definitely have a long-standing interest in subspecialty with infectious diseases and pregnancy, and I did have an opportunity to learn something about big data science through the Utah population database. So when I was a fellow at the University of Utah, we worked with very, very, large data sets and how did we really integrate those data sets, and had a pretty strong background in the biostatistics of big data. So when we were first kind of looking at the human microbiome project here at Baylor, I got brought on board frankly because they needed somebody who could a do vaginal sampling, and B had been involved in a lot of clinical trials and big data science work, but more on the clinical trials arm than on the genomic science arm. So I kind of fit that bill and was my first couple weeks of Baylor I got pulled into the project and I never looked back, yeah it was really early on when I came here.

Brandon: You mind telling us a little bit about what the human microbiome project is?

Dr. Aagaard: Yeah so the human microbiome project was a very large, what we call, roadmap initiative with the National Institutes of Health. So other examples of that are that Human Genome Project, epigenomics project all of which have had its roots here at Baylor College of Medicine because of our genome Center and really the eminent work of Richard Gibbs in bringing in these types of science and investigations and we have so much to thank him for along those lines. So the human microbiome project, they really needed our computational and genomic science folks here at Baylor to run, but those folks aren't necessarily quite as fluent in how do you convince a woman to come in and loan us or vaginal samples every couple of months for a long period of time, and how do you actually design a study where you could create a quote/unquote healthy reference population. So the goals of the human microbiome project were to enroll 300 people across two sites, 150 at Washington University, and 150 at Baylor College of Medicine. They were to be longitudinally collected, meaning we'll see the same person multiple times and collect samples and they were ultimately to function as a reference population. So not necessarily normal right, we wanted them be referenced meaning they reflected what the population of the US looks like, and so we had kind of some healthy individuals and we used some clearly defined criteria that we set up a priori, or in advance of the study, but then we also went ahead and had some branch points off where people could do demonstration projects so they could do things like study pediatric populations. We studied some pregnant populations those types of things could go on simultaneously.

Brandon: You, you mentioned something about convincing people to come in for a vaginal swabbing every couple of months, and I have a couple of questions. First of all, can you comment a little bit about what you mean by building that trust and how were you able to convince people to participate in the study long term?

Dr. Aagaard: Great question.

Brandon: And then after that, I want to know, because we know a little bit about how you got into looking at the placental microbiome, could you didn't tell us a little bit about that?

Dr. Aagaard: Sure, of course, so with the human microbiome project per se, there was a pretty good cash incentive to participate in this study and that always helps...and because we were looking for a reference population including young, healthy people and we certainly had a reasonable number of folks who were students whether they were students here Baylor or students at the University of Texas Houston they were from a number of different situations. But they weren't all students, they really represented cross-sectional community, we screened over twice as many people as we actually enrolled because one really interesting component of this was we did a lot of oral sampling, so we had skin microbiome we had the gut or stool microbiome we had the vaginal microbiome in women and then we had the oral microbiome and those involved quite a few sites in the mouth. So we had to rely upon our dental colleagues and really great collaborations with their dental colleagues, and there were a lot of folks who ended up being excluded because of not being able to meet some of the dental health criteria. And so that was a lot of the sampling, it usually would take him a couple of hours every time they came in for sampling.

Erik: Oh wow, and then yeah we were we're kind of curious like, how did this feed in then. Did this kind of initiate your research into the placental microbiome or was it maybe, I don't know did it or were you thinking of you were gonna go that direction before you came on because it seems like at least from reading that paper that served as your reference population?

Dr. Aagaard: That's a great so, that's a great question, so yes, we did use a human microbiome project reference cohort as it was designed and ended to say this is what if you had a non-pregnant population what they look like. So I'm going to answer that question in two ways, so one is a lot of the work that we had done ultimately involved something called the developmental origins of disease, so we're very interested in how pregnancy exposures can change the future risk of health especially metabolic health in the children and we look at a variety of different situations of that. So we've done a lot on understanding the epigenomic or upon the genome changes we understand something about the genomic changes but this development origins of health and disease was a major focus of my research prior to the start of the human microbiome project. What became obvious as least kind of two parallel world worlds were chugging along the more work I was doing on the human microbiome project, the more I became curious, could our understandings of the microbiome explain a lot of this development origins of disease? In other words, could there be exposures during pregnancy that changed her microbiome and subsequently changed your metabolic health right. And so they were really parallel research interests that I started to see where they potentially connected. So we did a study that we published in the HMP collection on looking at the vaginal microbiome during pregnancy, and we compared it to non-pregnant individuals that arose from the human microbiome project. And one of the things we learned was that the vaginal microbiome during pregnancy was a little bit less rich a little bit less diverse than non-pregnant, but it's not like you suddenly had entirely different species and genre there, it was kind of some subtle differences. About that time a couple papers started emerging that were character from other investigators around the world that started characterizing the microbiome in infants in their first week of life. What we noticed was very little of those infant microbiomes were present in the vagina including if you looked in pregnancy, so we kind of had this question of, well if they're there when the baby's born A.) How long have they been there? And B.) Where they come from? Because they don't look like the vagina that's like 80% lactobacillus, and so we started to hypothesize that maybe our premise that babies are born sterile isn't actually true. Now there was quite a bit of evidence if you went into mice and into some other settings that that may be the case, but we initially ran those placental samples that we first did as controls, we weren't expecting to really find anything there.

Brandon: So quick question, when you say that like the prevailing thought was that babies were born sterile, do you mean like they were, they're born, and they had no microbiome on them at all? Or were they before the birthing process sterile while they were in Utero and then they were kind...

Dr. Aagaard: Both, together, exactly good question. So, so the intra-uterine environment, the premise was the intra-uterine environment was relatively quote-unquote sterile. And so babies are born without an actual microbiome present yet. That was kind of the prevailing thought.

Brandon: Okay, and if that was the accepted thought did you get a lot of pushback when you said hey babies aren't sterile?

Dr. Aagaard: Well, so we never said, with that paper, we never said hey babies aren't sterile. We really kept... we were pretty conservative with what we said, and what we said was we can detect a low biomass low abundance metagenomic community in the 320 placentas we looked at. So we've always been very cautious that we don't know if it's an actual viable microbiome community, meaning are there live colonizing microbes? We remain agnostic to that at this point; we don't know the answer to that. Certainly, some individuals, some other research groups, including quite a few really really talented research programs in Finland and Iceland and Norway have done some very nice work where they've been able to cultivate out microbes from the placenta. We have not been able to, but we have only tried to use clinical cultivation pipelines to do so, but to this day what we've really limited ourselves to doing is describing that metagenomic community in the placenta. We also went on and published a paper with one of our incredibly talented MSTP students Derek Chu really led this initiative, and we looked at the microbiome in neonates within an hour of birth, and then those same neonates at four to six weeks of age. And we are able to...to further our findings in the placenta and show first of all we can detect again metagenomically microbes in babies right at birth, and second of all we found that we really see quite a bit of diversification and body niche speciation, meaning they look more adult-like, so their oral microbiomes different than their skins different than their stool as early as four to six weeks of age. So we've been able to expand out these findings certainly in humans.

Brandon: So it sounds like when children are born, they have this bit of a microbiome and then over time it differentiates and becomes separate from?

Dr. Aagaard: Exactly, a little bit more mature, so they'll, you'll see these body sites speciation. And we've also been doing some parallel studies and non-human primates using a model that we developed for epigenomic research to try to understand, initially we thought we'd be understanding how maternal obesity leads to obesity in subsequent generations. What we found instead in our non-human primate work is that maternal high-fat diet or caloric dense diet is what actually lends to that, it's not the maternal obesity per se, which was kind of an a big exciting change in the literature. But we've done these parallel studies in non-human primates, and we published several papers now where we've been able to show that a maternal high-fat diet changes that offspring's microbiome detectable at birth and still prevalent with a strong footprint of that maternal diet at three years of age even when you switch that offspring on to a controlled diet at the time of weaning. So you can still see the impact of a maternal high-fat diet on and offspring's microbiome in monkeys at three years of age even if they've been fed a controlled diet for two-and-a-half years, so that also now starts being indirect evidence that what they're exposed to during gestation and lactation we can't separate the two in our primate model, has a persistent and lasting influence on the offsprings microbiome consistent with the notion that something is going on in that intrauterine environment. Whether it's truly colonizing the fetus or it's just creating a milieu that enables live microbes to see the fetus immediately at or after birth, again we don't have the data around at this moment in time.

Erik: Well, and you bring up a good point that this is based off of sequencing data and a lot of people have trouble culturing some of this stuff. And so I guess that brings into another question, it's clear that a lot of microbiome research is kind of on the shoulders of the development of sequencing technology and computational pipelines and everything like that. What do you think like right now, what do you think some of the main downsides or issues with the technology that are maybe holding further developments?

Dr. Aagaard: Great question. So I'd say I think there's really kind of three things that that I always try to keep in mind. First, I always try to remember that we developed metagenomics or sequencing technology to compensate for the fact that we can only cultivate or culture 10 to 15% of microbes at best because you have great variation between aerobic and anaerobic knowing, what their requirements are for growth, and that they exist in complex communities. So kind of the analogy I think about is we know that if we take a lion off the savanna grasslands of Africa and plunk it into our Houston Zoo, right, it's not going to reproduce as well and in fact we're going to fundamentally change that lion in the properties once we take it out of its nichek, right? Doesn't not make it a lion, but we're going to have fundamental changes, its behavior will be different. So microbes aren't that different, you take them out of their niche it's very hard to recapitulate that niche in ways that we can really study. There's a lot of work being done that's a lot of what the organoid and enteroid work is all about, but that you know as it exists. So I would say that's kind of statement number one, which brings up problem number one, how do we recapitulate those niches in ways that we can study these microbes and their function better? The second thing is when we're doing the metagenomic sequencing the resolution the degree to which we have to sequence a sample to get the strain resolution right, and strains matter, yeah so there's you know...

Erik: Genus and then...

Dr. Aagaard: Exactly, exactly, King Philip etc.

Brandon: You say strain, you mean even past the species level?

Dr. Aagaard: Right, right so one strain of streptococcus pneumoniae is different than another strain of streptococcus pneumoniae right, and we're down to strain levels when we think about pathogen and commensal at times. So those are really important differences because they make different metabolites, which we then utilize or don't utilize, and they utilize our metabolites or don't use. You have to get down to some pretty deep sequencing to do that which is then leads to the next kind of challenge we're all struggling with. You know, we will go on, and we'll sequence at 5, 10, 20 GB per sample well, before you know it you're having to deal with 30, 40, 50 TB data sets, that was just like a lot of money in space!

Brandon: Yeah, you just said TB and GB

Dr. Aagaard: Oh, sorry, sorry (laughs).

Brandon: I think it's really interesting talking to you about the microbiome and stuff like that because I get the microbiology and the immunology, but the second you start saying anything computer wise I kind of get that glassy-eyed look.

Dr. Aagaard: Yeah, so gigabytes and bases. Yeah so we measure our sequencing not necessarily in bases but in space, and then the data can get measured in either bases or space. So, so we talked about both of those interchangeably.

Erik: Yeah, I mean cause, I've only, I've done a little bit of sequencing work but not a lot, but even just in that experience it's like you have to have a server to house the information on, and so it takes a lot of resources. So I mean that's probably another limitation of this right?

Dr. Aagaard: Right, and it, so that's something Baylor has been, Baylor College of Medicine and Texas Children's Hospital have been very thoughtful around, so we have kind of two microbiome centers. Joe Petrosino runs the Center for microbiome metagenomics research here at Baylor. He's done a really nice job about making sure he's continued to develop those metagenomic pipelines and the capacity to do super high-end sequencing and try to get us to strain resolution. And then the Texas Children's microbiome center is run by Jim Versalovic, and that one has taken a little bit different approach and done a lot of work on the metabolomics, and so has been very interested in how do you do the functional readouts of these metabolites especially in pediatric populations.

Brandon: One thing I think would be interesting to talk about, just outside of scientists, I think that you've got a lot of groups that kind of look at like the microbiome and stuff like that, and kind of think of it as like being able to explain a lot of the weird things we kind of see in society. Like with reactions with things like medicines and vaccines and all this other stuff. What kind of implication do you think the microbiome has on other fields in medicine aside from just infectious disease?

Dr. Aagaard: So that's a great question. So, so I think one of the things I constantly try to remind myself is antibiotics aren't the only anti-microbials we deal with. So metformin, or Glucophage, is a great example of an antimicrobial agent, we use it for diabetes management, but it's known to have antimicrobial properties. It's not the only one, proton pump inhibitors have antimicrobial properties, so typically we think of antibiotics as being the only antimicrobial, but there's actually quite a few others we can think about. There was a big story that we....that I think we've kind of calmed down a little bit and provided a different perspective, maybe C-sections, delivering a baby by cesarean is antimicrobial. We don't find that to be true in any of our data and research and larger longitudinal studies support what we've observed as well, but those are examples of things that we have to think beyond just antibiotics as being antimicrobial. I think that's a really good example where kind of this field can be paradigm-shifting is thinking about how important good microbes are for us and understanding antibiotics aren't our only antimicrobials.

Erik: That makes a lot of sense, and so I guess again thank you for kind of coming here and talking about all this, and just to wrap it up if I don't know...just a broad question of if you have any advice for really just MDs or PhDs or PAs in terms of deciding kind of like what field you're you want to go into and kind of devote your career to? Because I think a lot of us have anxiety at this early age in our career where like there's so many options, it's almost a paradox of choice, and it's like we don't really know what to do and I don't know if you have any advice for how to navigate those waters?

Dr. Aagaard: Just do what you love to do, if you go into a field because you love it, and you can be passionate about it, you can advocate for your patients for the rest of your life, you can feel like okay maybe there's me and one other person on the planet that care about the answer to this question but at least that's two of us. so whatever you do just do it with passion do it with good intent and...and all of this rigmarole about burnout and bloody yada yada...I don't know first of all you got be on fire before you burn out and second of all it seems like those flames just keep cropping up if you're not doing something you're passionately interested in. I didn't ever go into this business thinking I'd become a high-risk obstetrician. I thought I'd become an infectious disease doctor or a hematology oncologist or something along those lines and I just love taking care of women and their babies. I mean, I have like the best job in the whole universe, and then I get to find out some secrets about how can they help their children be healthier, smarter, better than they were. You know because I don't know, sometimes I'm worried about the shape we're leaving our world in for you guys, you guys have a lot of our messes to still clean-up and a lot of questions we haven't been able to solve, but I think if we can arm you guys well you're gonna be able to come up with those solutions we failed to come up within our career.

Erik: Well, thank you again.

Brandon: Yeah, thank you so much!

Erik: All right, and that is it, for now, Brandon, and I would like to thank everybody out there who took the time to listen to this episode. Special thanks to Jennifer Deger for helping write some of the questions. Thank you to Dr. Poythress for serving as our faculty advisor and helping us put this whole thing together. Thank you to the Baylor College of Medicine communications department for help with the production of this, thank you again to Dr. Aagaard for taking the time to interview with us. So I guess I hope everybody enjoyed it and we hope you tune in again soon. Goodbye for now!