How can we provide a good home for our microbiomes, so they’ll keep us healthy?
Fecal transplants may be one of the more surprising health news sensations in recent years. The process of transferring small amounts of one person’s stool to another’s gastrointestinal tract to treat a dangerous bacterial disease just seems too icky to make it out of the pages of obscure medical journals.
The media’s fondness for potty humor aside, it’s the astonishingly high success rate of the procedure in curing severe Clostridium difficile infections—as high as 90 percent in some studies—that has the public, and many physicians, excited about its potential. But fecal transplants also happen to dovetail nicely with that other media darling, the microbiome.
“I have patients who come to see me all the time who don’t have C. diff.—they have fatigue, bloating—and they’re insisting their microbiome is disrupted and a fecal transplant would help,” says Colleen Kelly, MD F’06, assistant professor of medicine and one of the nation’s, if not the world’s, foremost researchers and practitioners of the procedure.
Kelly’s patients can be forgiven for believing the microbiome holds the answers to life’s most vexing medical questions. Whatever the malady, from allergies and obesity to stress and low IQ, someone has declared the microbiome plays a role, some news outlet has breathlessly reported it, and most of the public ends up hopelessly confused.
But the field of microbiome research is so new that not everyone even agrees what “microbiome” means. So anyone who states definitively what our microbial fellow travelers can and can’t do for our health is probably peddling snake oil.
“We know it’s important. We know there’s a lot of research that needs to be done,” Jason Shapiro, MD RES’08 F’11, assistant professor of pediatrics and of medicine (clinical), says. “But how it affects the day-to-day treatment of our patients? We’re not there yet.”
That’s also what makes the microbiome so exciting to study. “It’s uncharted waters,” Kelly says. “It’s kind of fun to do things that haven’t been done a zillion times before.”
Path of Resistance
“This isn’t in your textbook,” Peter A. Belenky, PhD, assistant professor of molecular microbiology and immunology, told his Introductory Microbiology class one afternoon in late March. He was about to deliver a lecture on the microbiome, but, he cautioned the hundred-plus undergraduates before him, “the science is being done now, so anything I say could change.”
Here’s what (most) scientists agree on: the human microbiome is the trillions of microbial cells—bacteria as well as fungi, viruses, and archaea—and their individual genomes that have co-evolved with us over millions of years. Most of our microbiome’s members are benign or beneficial: they help with digestion and vitamin production, prevent pathogens from establishing themselves and doing harm, and play roles in metabolism and immune function.
About 1,000 species of microbes call Homo sapiens home, and have adapted to many different communities, including the gut, mouth, skin, lungs, and virtually every other bodily surface, inside and out. The microbiome gets its start when we’re born—though whether we come into the world vaginally or by C-section changes its initial makeup—and it grows and diversifies until we’re toddlers, then stays remarkably stable for the rest of our lives. Estimates of how many microbes there are in an adult body vary wildly, from 10 times the number of human cells to a 1-to-1 ratio that tips in our favor with each bowel movement. Regardless, it’s a lot, making it all the more remarkable that the study of the microbiome is only a few decades old.
So now that they know it’s there, scientists are asking: how do all those cells function in concert with our own? What happens when the microbiome’s delicate balance is upset, a condition known as dysbiosis? How much does that affect our health, and which afflictions does it cause? How can we hone our treatments to protect ourselves, as well as our microbiomes?
Belenky is trying to understand, at a genetic level, microbial response to external stressors, like antibiotics, and the role that plays in antibiotic resistance and disease. The problem, he says, often begins when we use broad-spectrum antibiotics to target one, specific pathogen. “But with the microbiome, you actually target 1,000 organisms,” he says. “We mostly know how [an antibiotic] affects the target bacteria, but not the other 999.”
Bacteria are able to share genes by taking up DNA directly from their environment. While they most commonly swap genes with fellow microbes, any genetic material that can aid a bacterium’s survival is fair game for uptake. “They can take up DNA from a mammoth bone,” Belenky says.
Gene transfer has a community benefit much of the time—in response to environmental changes, our microbes’ genomes can adapt much more quickly than ours can. But that ability also aids antibiotic resistance: if a microbe has a gene that can protect it from a drug, other members of the microbiome will take it up, including those that cause disease. That’s why “the toxicity [of antibiotics] to nonpathogenic organisms is just as important as the toxicity to pathogenic organisms,” Belenky says.
“In the impending antibiotic crisis—or it’s already here, depending on who you talk to—our current antibiotics will no longer be functional,” he says. Nor is there much likelihood of new antibiotics being developed, given the exorbitant costs and the bleak reality that they, too, would quickly lose effectiveness. So why not work with what we have? “I want to identify ways to use our current arsenal better,” he says—perhaps by combining different drugs, playing with duration of treatment and dosage, or other variables.
At Rhode Island Hospital, Belenky is recruiting inpatients to spit into vials so he can study how narrow- and broad-spectrum antibiotics affect the oral microbiome. It’s tricky because he needs a no-treatment baseline, yet most patients receive a dose of antibiotics soon after arrival in the emergency department, he says; “so we have infectious disease docs sitting in the ER for us and following patients.” After three days they collect another saliva sample, to see how the microbial community has changed.
“We’re using the oral microbiome to look at transcriptional profiles because while most research is done on the gut, on fecal samples, transcriptional changes happen in minutes or seconds,” Belenky says. “It takes food six or seven hours at a minimum to move through the gut, so [fecal]samples are too old. Oral samples show exactly what happens at the second we collect the sample. It’s essentially a freeze frame.”
His research is possible thanks to the latest technological advances that allow him to sequence an entire genome in hours, for relatively little cost, and gather data on transcriptional, metabolic changes. “We weren’t able to do that before,” Belenky says. But scientists still can’t culture most of the microbiome’s members because they seem to depend on each other, as a community, to grow, he adds.
For the past decade the revolution in sequencing technology, coupled with the realization that our bodies housed many more microbes than we could grow in a petri dish, meant that most microbiome research was descriptive and correlational, Belenky says. Papers described what species were present, and how microbial communities differed in people with disease. But the bar for publication has been raised. “Articles now are much less descriptive,” he says. “Now you need to do real science. You need to figure out, why are changes occurring, related to health outcomes?”
Few, if any, of those “whys” have been answered definitively. Even the premise of much of Belenky’s research—that overuse of antibiotics is harming us—is a strong but as yet unproven theory. “Statements that say, we should maintain microbiome diversity, reduce antibiotic use, use probiotics, are most likely true on a total population level,” he says. “But when it comes to making this decision for a specific patient, it becomes a lot harder. We simply don’t have the studies to provide concrete guidance to physicians about risk-benefit assessment of withholding therapy.
“This puts physicians in a difficult situation,” Belenky adds. “They know that overuse of antibiotics is a problem, but they don’t have all the tools at their disposal to address it.”
Mice in a Bubble
Deep in the bowels of the BioMed Center, research assistant Irina Maglysh dons a full complement of personal protective equipment—gown, shoe and hair covers, gloves, face mask—and swipes into a large, white room. In the middle, on waist-high tables, are several large, rectangular bubbles, inflated with sterilized air, each with a half-dozen clear plastic boxes inside them.
These are the sterile living quarters of about 50 germ-free mice, so called because, since birth, not a single microorganism has inhabited their bodies. Everything in their isolated enclosures—food, water, bedding—has been autoclaved; and every two weeks, Maglysh collects and tests some of the cleanest feces on the planet to confirm their aseptic state.
This is one of only about a dozen germ-free mouse colonies in the US, a technological development that, along with affordable, rapid genome sequencing, has accelerated microbiome research in recent years. “In my lab, [these advances] let us ask how a single bacterial species interacts with a host, to really delineate the role of keystone species in the gut environment,” says Shipra Vaishnava, PhD, assistant professor of molecular microbiology and immunology. This allows her to zero in on genetic changes and larger health impacts when microbes are added or eliminated.
“Bacteria can influence many aspects of host physiology,” Vaishnava says. “But how do we go from changes in the bacteria in the gut to having [a disease]? What are the key molecular pathways?” Vaishnava has a particular interest in the cells that line our intestines, and the role they play in host-microbe interaction. She wants to figure out what bacteria live where within our guts, to understand how that might influence our health.
“Scientists haven’t really thought about bacteria in the gut as, ‘Where are they with respect to host tissue?’” she says. “Maybe some diseases [occur]because bacteria are in the wrong place. But these differences wouldn’t come out if you look at the feces for who’s there.”
To tease out this biogeography, Vaishnava tailors the microbiomes of her germ-free mice by introducing into their guts the bacteria she wants to study, a science known as gnotobiotics. Her team then dissects sections of mouse intestines and, under microscopes and with lasers, isolates cells for sequencing to determine each species’ location. Armed with that knowledge, they can then tag the bacteria with fluorescent probes and see what effect antibiotic treatment or infection have on the location and abundance of these bacteria.
The research could someday answer many questions about our microbes and our health, Vaishnava says: “how the epithelial lining is regulating our gut microbiome, how it’s negotiating these [host-microbe] interactions, what are the mechanisms that help us maintain a peaceful relationship—and if you don’t have the mechanism, what is the physiological outcome?”
Defects and other disruptions in the gut epithelial lining have been observed in many diseases, from Crohn’s to liver cirrhosis, but it’s too early to say whether the cause is genetic, environmental, or lifestyle factors, or some combination of the three. “The diseases we are studying are so complex,” Vaishnava says. “It hasn’t been figured out, but I think it’s just a matter of time.”
Every Breath You Take
Though most research and knowledge of the microbiome is related to the gut, our bodies house billions of microbes specialized to other locations, including our hair, nostrils, and urogenital tract; our skin is home to many distinct communities, from our hands to our eyelids to our navels. Dependent on pH, moisture, and other factors, our microbial populations are as different, and as specialized, as the ecosystems of a reef and a desert.
Just a few years ago, Amanda Jamieson, PhD, assistant professor of molecular microbiology and immunology, became one of the first people to focus her research on the lung microbiome. “The first Human Microbiome Project left out the lung because it was thought to be sterile,” she says, referring to a five-year NIH push to identify and map our bodies’ microbial residents and find relationships between the microbiome and disease.
Jamieson was at the University of Vienna at the time, studying bacterial pneumonia, which can arise when a small amount of pathogenic bacteria infects someone recovering from influenza. “I thought, there has to be bacteria [in the lung]because we breathe it in all the time,” she says. “So I asked if bacteria in the lung could be causing it. … I did a PubMed search and got nothing.”
It’s well known that flu suppresses the immune response, and Jamieson wants to know if our lung microbiome influences that response, how it changes with infection, and whether it can be manipulated to improve outcomes. In September she won a Defense Advanced Research Project Agency (DARPA) Young Faculty Award to further that work.
Because lung microbiome research is many years behind that of the gut microbiome, it’s still in the descriptive phase, Jamieson says—collecting samples, sequencing genomes, and identifying what’s there. She also has to tease out the permanent residents from the visitors: which bacteria are there because they were inhaled, and which are part of an established community?
She is conducting some of that descriptive work on nasopharyngeal swabs collected from flu patients, to see if there’s a correlation between illness and bacteria. But getting samples from a living human’s lung is difficult, and uncomfortable, requiring insertion of a bronchoscope and then scraping or washing cells from the airway.
Jamieson’s lab uses mouse models, though that has its own challenges. “Culturing bacteria straight out of the lung is very difficult,” she says. “The intestine has a stratified, much more organized structure than the lung, which has a lot more nooks and crannies.” So after they painstakingly identify what bacteria are present, they culture strains ordered from a scientific supplier to test immune response to influenza and whether changing the amount of bacteria makes the flu worse or better.
In vitro research is unlikely to paint a full picture of the lung microbiome, however. “In human patients, there is no evidence of bacterial pneumonia in culture—but is there something there, and we just can’t culture it?” Jamieson says. “A lot of people with symptoms don’t have diagnosable bacteria.” The lung microbiome is smaller and less diverse than that of the gut; there is evidence in mice that low levels of harmful bacteria in the lung will cause problems.
Ultimately Jamieson hopes her lab’s focus on the role of bacteria in co-infections will lead to better patient outcomes. But she’s wary of the lessons learned from broad-spectrum antibiotics; treatments must be mindful of all systems in the body, she says. “We’re trying to develop ways to affect the lung microbiome without affecting the intestinal microbiome,” she says.
Cause and Effect
C. difficile infection is one of the most clear-cut examples of the importance of our microbiome to our health. The disease nearly always arises from a course of antibiotics that wipes out much of the gut’s biodiversity, allowing C. difficile, a normally benign resident of the intestine, to flourish. It can cause severe diarrhea, colitis, dehydration, and worse; according to the CDC, of the nearly half a million people sickened by C. diff. in 2011, about 15,000 died.
In the most serious, recurrent cases that don’t respond to standard treatment, fecal microbiota transplants have been remarkably successful at restoring patients’ gut microbiomes, and their health. Kelly, a gastroenterologist at the Women’s Medicine Collaborative in Providence, has led several studies in which about 9 in 10 patients were cured, with few side effects. But she cautions that more research is needed; only a few small randomized controlled trials have been done to date, and there’s little long-term safety data.
Also, though donors are rigorously screened before their stool is accepted for transplant into a C. diff. patient, there’s still a risk of other, unforeseen complications. Kelly says the FDA has shown interest in establishing a fecal transplant registry, like that for bone marrow. It would be funded by the NIH and follow 5 ,000 patients for up to 10 years after a fecal transplant. “It would be really helpful to the field, to get that safety data,” she says.
Patients suffering from other disorders of the GI tract could be helped, too. “Our next hope is moving into these other diseases associated with alterations in gut bacteria,” such as irritable bowel disease (IBD), Kelly says. She’s cautiously optimistic about small studies done so far. But she adds: “Nothing is all good. Are there people who would become worse after a fecal transplant, or would it trigger another problem?”
A more standardized approach, in the form of a pill, could prove safer than fecal transplants in the long term. Kelly is taking part in a phase II trial of a capsule containing just a few bacteria, derived from human stool, to treat C. difficile infection. Other companies are trying to design fully synthetic formulations; if successful, Kelly says they may be able to apply that knowledge to the treatment of other diseases associated with dysbiosis. “I’m a believer this is going to happen,” she says.
So is Shapiro, a pediatric gastroenterologist at Hasbro Children’s Hospital, though he thinks such therapies are “years away.” Many of his young patients suffer from IBD, the rate of which is increasing across the US and the world, he says. “This is proof of concept of how important the study of the microbiome is,” he says. “The diversity of [gut]flora in the US is really low relative to those in developing countries. While we don’t have to deal with parasites or poor sanitation, we are seeing an overall increase in chronic diseases such as IBD. … Is it causative?”
Shapiro may have the data to figure that out. Since he was a resident in pediatrics at Hasbro, he’s been involved with the Ocean State Crohn’s & Colitis Area Registry (OSCCAR); he took over as PI last year. The group annually collects blood, urine, and stool samples from more than 400 patients in Rhode Island and is examining the longitudinal data for specific microbiome signatures and how they change over time. “Are these biomarkers of disease or potential therapeutic targets?” Shapiro says. “How does the microbiome change with treatment? Analyzing the samples from OSCCAR represents a great opportunity to complete a variety of meaningful microbiome studies.”
OSCCAR began data collection in 2008, and already has a massive amount of it. In addition to sequencing patient and bacterial genomes in blood and fecal samples (they haven’t started examining the urine samples yet), they’re looking at protein signatures, and layering that with patient metadata, such as age, sex, and type of disease. “You need a mathematician now to do this,” Shapiro says. “To integrate and analyze these huge datasets while making it clinically relevant is exceptionally challenging.”
But he believes that down the line all of that data will help them refine treatments. “Now the way we treat IBD is such a shotgun approach from an immunologic standpoint. While our current medications work, they are not without risk, including the rare chance of developing a secondary malignancy,” he says. “You’re treating a disease by wiping out an entire neighborhood, but it would be nice to find the exact house.”
The future of personalized medicine, in which therapies are tailored to individual patients, will likely depend on better understanding of the microbiome, which appears to be more individual than even our genome; humans are more than 99 percent similar to each other genetically, while our microbiomes show considerably more variability. Even identical twins can have significantly different microbiomes, due to everyday differences in diet and other environmental and lifestyle factors. But Shapiro cautions that the potential to develop individualized treatments for conditions such as IBD by manipulating the microbiome has “yet to be determined.”
“The more we understand, the more we can treat [patients]with a targeted approach,” Shapiro says. “Hopefully, 10 to 20 years from now, that’s where we’re going.”
Understanding the microbiome has big implications beyond treatment of individual patients: it is critical for the health of the population. Antibiotic resistance has brought us not just C. difficile infections but an ever-growing list of terrifying pathogens, including methicillin-resistant Staphylococcus aureus (MRSA), New Delhi metallobeta-lactamase-1 (NDM-1), and multidrug-resistant tuberculosis (MDR-TB). Many gain a foothold among already weakened patients in health care facilities, but others are infecting healthy people: MRSA, for example, is known to spread among athletes who play contact sports.
“As a society we have a responsibility to limit antibiotics in conditions where they are not absolutely indicated,” Shapiro says. Antibiotics play an important role in medicine, to be sure, and in some patients they offer the only hope for recovery. But the CDC reports that up to half of antibiotics prescribed are unnecessary or aren’t taken as directed.
Alexander Fleming, the discoverer of penicillin, saw this coming. In his acceptance speech for the 1945 Nobel Prize in Physiology or Medicine, he warned, “there is the danger that the ignorant man may easily underdose himself and by exposing his microbes to nonlethal quantities of the drug make them resistant.”
This is precisely what is occurring in US agriculture today: farmers give cattle, pigs, chickens, and other animals low doses of antibiotics preventively and to promote growth; statistics suggest that up to 80 percent of the antibiotics sold in the US is used for livestock, and despite FDA rules stipulating otherwise, investigators have found that many are sold over the counter. Pigs carry MRSA and spread it to farmworkers; it even has been found in pork for sale in supermarkets. In 2013 the CDC estimated that of the more than 2 million annual cases of antibiotic-resistant infections, one in five originated in food and animals.
Antibiotics given to the youngest patients have potential to cause long-term harm. Because the microbiome is still developing until we are between 2 and 3 years old, antibiotics may permanently alter its diversity; though, again, nothing is certain, long-term problems possibly related to a stunted microbiome range from allergies to IBD to celiac disease. This theory is of a piece with the “missing microbe hypothesis,” advanced by Martin Blaser, MD, a microbiologist at the NYU School of Medicine, who suggests that the overuse of antibiotics has ushered in modern-day Western “plagues” like obesity, asthma, and type 1 diabetes.
And then there’s the “hygiene hypothesis,” which states that a lack of exposure to infectious agents early in life—like the germs passed around in day cares, on playgrounds, and by animals—suppresses the immune system, in which the microbiome plays some as-yet unclear role. We are, essentially, too clean. “Keeping your kids dirty might be good,” Belenky says. Pulling up data plots of microbiome diversity in infants, he notes, “If you have pets, your early microbiome looks like your pet’s.”
Is that a good thing? Does it make those kids healthier adults? No one knows; at this point, it’s merely an observation, just as any implied connection between antibiotics and diabetes is correlative at best. To state otherwise is to give desperate patients false hope. But even though we don’t yet understand the mechanisms of the microbiome, we are certain it needs protecting, and many of the ways we think we can do that are harmless at worst, and who knows—they may help.
Since she began studying the microbiome, Shipra Vaishnava, whose kids are 4 and 7, says she’s made some lifestyle changes to nurture her and their inner microbes: she lets them dig in the dirt, avoids antibiotics and processed foods, schedules their vaccinations, and once in awhile she eats without washing her hands.
Eating better, playing outside, getting preventive care—sounds like a prescription for good health for all of our cells, human and microbial.