Our microbiome consists of a vast array of bacteria, viruses, fungi, and other microbes living on us and in us. Our gut microbiome alone contains an estimated 38 trillion bacterial cells belonging to 500 to 1,000 species. Some of those species perform useful functions, such as those that synthesize folate and other vitamins or that protect us against pathogens. While most gut bacteria cause no harm, disturbances in the levels of certain species contribute to a wide range of health problems—some extremely serious.
“Doctor, this disease is ruining my bank account, it’s ruining my social life, it’s ruining me. I’m coming to you as a last resort. Is there anything you could do?”
While it’s common for a seriously ill patient to plead for help, it’s rare for a physician to come up with a eureka moment on the spot. Yet that’s what happened in 1991, when a new patient with a devastating intestinal infection came to the office of Lawrence J. Brandt, M.D., professor of medicine and of surgery at Einstein and a gastroenterologist at Montefiore.
The elderly woman’s problems started with a bout of pneumonia. Antibiotics cleared her lungs but left her infected with Clostridioides difficile, a gut bug that is resistant to many antibiotics, causes severe diarrhea and inflammation of the colon, and is often lethal. C. diff, as it is commonly referred to, and the accompanying diarrhea would both return as soon as the woman’s antibiotic treatment was halted.
She called me at home that night and said, ‘I don’t know what you did, but I haven’t felt this good in months.’
— Dr. Lawrence Brandt
“I have an idea,” Dr. Brandt told his patient after she’d explained her predicament. “I think the antibiotics did something to the bacteria in your gut that were protecting you. But you’re sitting right next to a guy [her husband of 50 years] who likely has the same bacteria in his gut as you had. Maybe we could replace what you lost by taking some of his stool and putting it in your colon.”
The couple immediately agreed to the therapy, odd though it may have seemed. Days later, Dr. Brandt took a stool sample from the woman’s husband, diluted it, and used a colonoscope to inject the solution into her colon. “She called me at home that night and said, ‘I don’t know what you did, but I haven’t felt this good in months,’” remembers Dr. Brandt. “She never had a recurrence.”
Only later did Dr. Brandt learn that his “novel” therapy—now known as fecal microbiota transplantation (FMT)—actually dates to the fourth-century Chinese healer Ge Hong. He touted fecal material for food poisoning or severe diarrhea in Handy Therapy for Emergencies, China’s first handbook of emergency medicine.
The Western experience with FMT can be traced to 1958, when Ben Eiseman, M.D., a Colorado-based surgeon, used fecal enemas to cure four patients with infectious colitis. Despite these and several other treatment successes, FMT garnered little attention until decades later, when Dr. Brandt starting curing his Montefiore patients who suffered from recurrent C. diff.
The first clinical trial of FMT occurred in 2013, when researchers in Amsterdam compared FMT with antibiotic therapy in patients whose C. diff infections had relapsed after one round of antibiotics. FMT cured 15 out of 16 patients, while antibiotics cured just 7 out of 26, according to a report in the New England Journal of Medicine. Every patient in the antibiotics group who relapsed was subsequently cured with FMT.
Dr. Brandt co-led a clinical trial in 2014, also involving patients with antibiotic-resistant C. diff. A single FMT treatment produced lasting cures in 15 of 17 patients, and a 16th was cured after two treatments, for an overall cure rate of 94%, as reported two years later in the Journal of Clinical Gastroenterology.
How does FMT work? “C. diff and the antibiotics used to treat it reduce the diversity of bacteria in the gut,” says Dr. Brandt. “FMT immediately restores this diversity and prevents disease-causing bacteria from infecting and colonizing the GI tract.” But how?
“We know that FMT works, but not the mechanism responsible,” Dr. Brandt acknowledges. “We think the benefits result from the presence of metabolic products of bacteria found in healthy stool that maintain and regulate our metabolism.”
Better treatments for C. diff are certainly needed. C. diff infections are growing at an alarming rate, particularly among nursing home residents. Roughly a half million people in the United States are affected annually, with a recurrence rate of about 20%, leading to an estimated 29,000 deaths each year.
Lawrence Brandt, M.D., checks on the image clarity of an endoscope before beginning a procedure at Montefiore.
The U.S. Food and Drug Administration (FDA) still classifies FMT as an investigational procedure but allows its use for patients with C. diff who do not respond to conventional treatment. (In November 2019, the FDA hosted a meeting to obtain public input on the state of the science regarding FMT, which could lead the agency to reclassify the therapy.)
The FDA’s limitations aside, the major barrier to wider FMT use is the therapy’s “yuck” factor.
“In our society, stool is seen as something dirty,” Dr. Brandt says. “We make light of it with scatological humor. When I tell people I’m infusing stool as a medical treatment, they laugh and think I’m crazy.” He’s now encouraging the field to rechristen the treatment as IMT, for intestinal microbiota transplant.
“It is the microbiota that is transplanted … not feces!” he wrote in a recent issue of the American Journal of Gastroenterology. “The term ‘fecal transplant’ is often puzzling to patients and awkward for providers.”
In the meantime, Dr. Brandt cautions that physicians prescribing FMT should obtain microbiota from rigorously tested sources, such as OpenBiome, a nonprofit stool bank, to reduce the risk of transmitting disease-causing microbes—a rare but potentially deadly complication. Another worry: No one knows the long-term consequences of using FMT to manipulate the gut microbiome.
“It’s one thing to treat older patients who might die if they don’t get their
C. diff under control,” he says. “It’s quite another to treat younger patients with non-life-threatening conditions. We don’t know if FMT may set them up for other problems later in life.”
FMT may work for other conditions as well. Researchers are currently studying the technique in Crohn’s disease, ulcerative colitis, irritable bowel syndrome, obesity, diabetes, Parkinson’s disease, and autism.
To a great extent, obesity and type 2 diabetes result from what we put into our stomachs. Our microbiome also has a major influence—and is itself influenced by a third variable: acculturation. Where we were raised appears to influence our microbiome and hence our risk for metabolic disease.
Research at Einstein on the acculturation connection can be traced to 2016, when the National Institutes of Health awarded Einstein researchers a five-year, $3.9 million grant to investigate the role of the gut microbiome in type 2 diabetes among Hispanics and Latinos—the fastest-growing segment of the U.S. population—enrolled in the long-running Hispanic Community Health Study of Latinos (HCHS/SOL). The study has been following 16,000 participants ages 18 to 74 since 2010, and Einstein is one of four HCSH/SOL sites nationwide.
The Einstein researchers analyzed the microbiomes of more than 3,000 adult HCHS/SOL participants who either were born here or came to the United States later in life. In a study published in 2019 in Genome Biology, they reported that the composition of people’s microbiomes was related to their degree of acculturation. People who relocated to the United States at an early age had a lower diversity of intestinal microbes compared with those who relocated after age 45.
Acculturation’s influence on the microbiome may have major health implications. The study found that people with lower microbiome diversity were more likely to be obese (a major diabetes risk factor), which is consistent with results from other studies. In addition, a reduced ratio of Prevotella to Bacteroides bacteria was significantly associated with obesity among the HCHS/SOL participants—a link not seen in other populations.
“These findings from our study are important since they tie population-
specific microbiome associations to clinical outcomes,” says Robert Burk, M.D. He is the co-principal investigator of the HCHS/SOL gut microbiome study and also vice chair for translational research in the department of pediatrics, professor of medicine, of microbiology & immunology, of pediatrics, of obstetrics & gynecology and women’s health, and of epidemiology & public health at Einstein, and a pediatric geneticist at Montefiore.
The question is, how do you make favorable changes to the microbiome? With probiotics? With regular foods? With medications?
— Dr. Robert Kaplan
Robert Kaplan, Ph.D., the other co-principal investigator of the HCHS/SOL gut microbiome study, notes that its findings also suggest that manipulating the gut flora could help in treating or even preventing metabolic diseases such as type 2 diabetes—which usually results from obesity and is especially common among Hispanics, who have a 66% higher rate of diabetes than non-Hispanic U.S. whites (11.8% versus 7.1%).
“The question,” he says, “is how do you make favorable changes to the microbiome? With probiotics? With regular foods? With medications? These are all things to be learned in the years ahead.” Dr. Kaplan is also a professor of epidemiology & population health, the Dorothy Manealoff Foundation and Molly Rosen Chair in Social Medicine, and the principal investigator for the HCHS/SOL at Einstein.
Earlier this year, Carmen Isasi, M.D., Ph.D., the co-principal investigator for the Einstein HCHS/SOL site, took the lead in rapidly redesigning the entire HCHS/SOL project to encompass the COVID-19 pandemic.
“We have already begun collecting data on our study participants’ COVID-19-related diagnoses and hospitalizations, including how the pandemic has affected them personally from a health, emotional, and employment standpoint,” says Dr. Isasi, who is an associate professor of epidemiology & population health and of pediatrics at Einstein. That effort also includes analyzing stool specimens to determine whether the gut microbiome—which can profoundly affect the immune system—influences the body’s response to SARS-CoV-2, the coronavirus that causes COVID-19.
“We could be dealing with the effects of COVID-19 long after this pandemic is over,” Dr. Kaplan says. “We’re hopeful that our ongoing HCHS/SOL trial—and its recent redesign to study the intersection of the gut microbiome, COVID-19, and overall health—can shed light on risk factors that can lead to long-term complications and on what can be done to prevent them.”
In addition to his gut microbiome research involving HCHS/SOL participants, Dr. Burk is investigating the role of microbiomes in health problems including human papillomavirus (HPV) infection, osteosarcoma (a pediatric bone cancer), kidney stones, HIV infection, and cognitive impairment.
A decade ago, he says, he was skeptical about the importance of the microbiome in human health. “We’ve learned a lot since then,” he adds. “Now I think we have to take a step back and consider that there may be a role for the microbiome in almost every disease.”
Paul Frenette, M.D., is the principal investigator on a study involving the gut microbiome and sickle cell disease (SCD). Dr. Frenette is a professor of medicine and of cell biology and the chair and director of Einstein’s Ruth L. and David S. Gottesman Institute for Stem Cell Biology and Regenerative Medicine.
People with SCD have an inherited gene mutation that leads them to produce abnormal hemoglobin, the oxygen-carrying protein in red blood cells. Red blood cells with abnormal hemoglobin become sickle-shaped and less flexible, which causes them to clog small blood vessels, triggering attacks of severe pain called sickle cell crises, or vaso-occlusive episodes. These episodes cause major organ damage over time and contribute to the decreased life expectancy (now in the mid-40s) of people with SCD.
In 2002, Dr. Frenette found that SCD vessel blockages occur when sickled red cells bind to neutrophils (a type of white cell) that have adhered to the vessel walls. “This early work indicated that not all neutrophils are the same,” says Dr. Frenette. “Some appear to be inert while others appear overly active in promoting inflammation—which is useful for attacking microbes but causes neutrophils to capture sickled red cells inside vessels, leading to blockages.”
What was most surprising and exciting was that the antibiotics greatly reduced chronic tissue damage in the spleen and liver.
— Dr. Paul Frenette
The hows and whys of neutrophil activation became clearer in 2015. While studying a mouse model of SCD, Dr. Frenette’s lab found that neutrophils become more active and pro-inflammatory as they grow older in the blood, suggesting they receive signals that tell them to age—signals, it turns out, that come from the gut microbiome.
“Since the body’s microbiota seem to ‘educate’ neutrophils to age,” Dr. Frenette says, “we realized that purging those microbes with antibiotics might help against SCD.”
In fact, when the researchers used antibiotics to deplete the microbiota of SCD mice, they observed a striking reduction in neutrophils but not in other white cells. Moreover, the antibiotics appeared to prevent sickle cell crises in SCD mice—markedly suppressing interactions between neutrophils and red cells, improving local blood flow, and increasing the rodents’ survival.
“What was most surprising and exciting was that the antibiotics greatly reduced chronic tissue damage in the spleen and liver,” says Dr. Frenette, whose study appeared in Nature. “This is the first time that something was found to have an impact on the organ damage that can be so devastating in SCD.”
With help from the Sickle Cell Disease Program at Children’s Hospital at Montefiore (CHAM), Dr. Frenette and colleagues investigated whether their mouse findings might be relevant to people.
They obtained blood samples from nine healthy children and from 34 patients with SCD. Of the 34 SCD patients, 11 were taking penicillin daily to ward off infections, as is recommended for children 5 or younger with SCD; 23 had been off penicillin for at least two months. Consistent with the mouse findings, children with SCD who were not taking penicillin had many more circulating aged neutrophils compared with healthy controls.
“Daily penicillin for patients with SCD younger than 5 works really well in preventing infections,” Dr. Frenette says. “Our study suggests that older people with SCD could potentially benefit from preventive antibiotic therapy as well.”
Deepa Manwani, M.D., director of the Sickle Cell Disease Program at Children's Hospital at Montefiore, listens to the heart of a young patient.
His team is currently collaborating with Deepa Manwani, M.D., who directs CHAM’s Sickle Cell Disease Program, on a study evaluating penicillin’s effect on the gut microbiomes of young SCD patients. Children are enrolled and their microbiomes evaluated at age 4 to 5, while they are still taking daily penicillin; their microbiomes are sampled again around age 5 to 7 (when they have stopped taking penicillin for at least six months) to compare the microbiomes of the children while they were taking penicillin and after they stopped taking the antibiotic.
Dr. Frenette is also trying to understand how the microbiome combines with psychological stress to cause sickle cell crises. Using a mouse model of SCD, he has found that the path to sickle cell crises begins in the brain.
Stress triggers the secretion of hormones in the brain that travel to the gut and increase its permeability. The greater permeability allows gut microbes known as segmented filamentous bacteria to interact with immune cells, which are stimulated to produce pro-inflammatory molecules that enter the bloodstream. These molecules promote the aging and accumulation of neutrophils that drive sickle cell crises. The findings were published in July 2020 in Immunity.
“Importantly, we found we could markedly reduce sickle cell vaso-occlusive episodes in mice through several different interventions: inhibiting glucocorticoid synthesis, depleting segmented filamentous bacteria, or blocking the inflammatory molecules induced by these bacteria,” Dr. Frenette says. “Each of those actions could potentially limit the impact of psychological stress on people with SCD.”
Patients who develop colorectal cancer often are prescribed irinotecan—a chemotherapy drug found on the World Health Organization’s List of Essential Medicines. However, up to 40% of patients who receive irinotecan experience severe, potentially life-threatening diarrhea.
“As you can imagine, such patients are already quite ill, so giving them a treatment that causes intestinal problems can be dangerous,” says Libusha Kelly, Ph.D., associate professor of systems & computational biology and of microbiology & immunology at Einstein.
Microbes are nature’s chemists, capable of producing and metabolizing a diverse array of compounds.
— Dr. Libusha Kelly
Dr. Kelly wondered if the gut’s microbiome might be to blame for adverse reactions to irinotecan. “We’ve known for years that variations in genetic makeup can affect how people respond to a medication,” she explains. “It stood to reason that microbiome variations might also play a role. Microbes are nature’s chemists, capable of producing and metabolizing a diverse array of compounds. That’s beneficial in some cases—when gut bacteria make vitamin B12, for example. But the microbiome can also turn otherwise helpful drugs into toxins.”
Irinotecan is administered intravenously in an inactive form. Certain liver enzymes metabolize the drug into a compound toxic to cancer cells. Other liver enzymes later convert the drug back to its inactive form, which makes its way to the intestine for elimination.
But some people possess gut bacteria that use part of inactivated irinotecan as a food source by digesting the drug with enzymes called beta-glucuronidases. That’s good for the bacteria but terrible for the intestines. This enzymatic action reactivates irinotecan into its toxic form, which damages the intestinal lining.
To minimize irinotecan-related toxicity, doctors have tried using oral antibiotics to kill the offending bacteria but found that the antibiotics killed off protective gut bacteria as well.
In a search for an alternative treatment, Dr. Kelly and her colleagues collected fecal samples from 20 healthy individuals and treated the samples with inactive irinotecan. They then used metabolomics (the study of the unique chemical fingerprints that cellular processes leave behind) to group the fecal samples according to whether they metabolically reactivated the drug. Four of the 20 people were found to be “high metabolizers,” and the other 16 were “low metabolizers.”
The two groups of fecal samples were checked for differences in microbiome composition, with a focus on beta-glucuronidase-producing bacteria. The microbiomes of high metabolizers contained significantly higher levels of three previously unreported types of bacteria capable of producing beta-glucuronidases compared with low metabolizers.
The 2017 findings, published in npj Biofilms and Microbiomes, a Nature journal, suggest that analyzing the composition of patients’ microbiomes before giving irinotecan might predict whether intestinal side effects will occur. Thus forewarned, physicians would know when to prescribe diarrheal treatment preventively and to monitor patients more closely for irinotecan side effects.
Dr. Kelly is now testing her hypothesis in a study involving colorectal cancer patients who are receiving irinotecan, in collaboration with Montefiore oncologist Sanjay Goel, M.B.B.S., M.S., professor of medicine at Einstein.
The findings also suggest that the use of drugs that inhibit specific beta-glucuronidases might prevent adverse irinotecan reactions—and that prebiotics (nondigestible fiber that gut microbes feed on) might help, too. “Beta-glucuronidases have an appetite for the carbohydrates found in the inactive form of irinotecan,” Dr. Kelly says. “If we feed patients a competing source of carbohydrates when we administer irinotecan, perhaps we could prevent those enzymes from metabolizing the drug.”
Libusha Kelly, Ph.D., at work in her Einstein lab.
In a project with even broader clinical implications, Dr. Kelly and two members of her lab (Leah Guthrie, Ph.D., and postdoctoral fellow Sarah Wolfson, Ph.D.) developed MicrobeFDT, a tool that groups 10,000 drugs, foods, and other compounds according to their chemical structure and links them to gut microbial enzymes that interact with them. The goal: predicting adverse health effects from exposure to those compounds.
MicrobeFDT correctly predicted how gut microbes might alter the structure of altretamine, an ovarian cancer drug that can cause diarrhea and kidney damage. “When we incubated altretamine in fecal samples from healthy volunteers, the drug was altered as the tool predicted,” says Dr. Kelly, whose findings were published in 2019 in the journal eLife.
“We hope this research will ultimately enable us to improve people’s health by monitoring and potentially altering their microbiomes,” Dr. Kelly says.
In an npj Biofilms and Microbiomes paper published in October 2020, Dr. Kelly and her postdoctoral fellow William Chang, Ph.D., devised an analytical method that was able to distinguish between gut microbe dynamics found in “healthy” and “sick” people, despite the continually changing microbiome compositions in both conditions over time.
“How does one’s microbiome respond to a trip outside the country? Does the state of your microbiome today tell us anything about your health tomorrow, or next week? We have so many questions,” says Dr. Kelly. “It’s early days for this field, but I’m excited about the potential to translate this basic research into the clinic and hopefully help improve people’s lives.”
Betsy C. Herold, M.D. 
