NHGRI logo

Microbes in us and their role in human health and disease

News May 29, 2019

Final outcomes from the most comprehensive analysis to-date of humans and their microbiomes definitively link microbes and microbial activities with health problems.

There is a mighty but invisible kingdom of microbes present within your body. Small yet incredibly powerful, these thousands of species and trillions of inhabitants live in all parts of your body and make up the diverse human microbiome. These microbiomes support and maintain your health but also, when the microbiome is disturbed in some fashion, have been linked to hundreds of ailments such as cancers, and autoimmune and cardiovascular diseases.

Hence it is not surprising that the human microbiome is an important avenue of health research. It is one with crucial implications for our health, as a slew of recently published papers in Nature from phase two of the Human Microbiome Project suggests.

Who’s there? Surveying the microbiome in humans

The National Institutes of Health’s (NIH) Common Fund established the decade-long Human Microbiome Project (HMP) in 2007. The HMP project was designed to be a community resource to galvanize research, in what was then a young field. The main aim was to create a toolbox with datasets and analytical and clinical protocols for the larger research community to be able to study the microbiome in specific diseases and populations.

The first phase of the project catalogued a reference dataset of the types of microbes (including bacteria, fungi and viruses) that are associated with five specific body regions: skin, oral cavity, airways, gastrointestinal and urogenital tract, using samples collected from 300 healthy adult men and women.

They used DNA sequencing technologies adapted from those first developed during the Human Genome Project to analyze these microbial communities. Microbiome-associated diseases are usually not infectious, but have a proportion that can be explained by the human microbiome.

Researchers used the reference dataset as a roadmap to identify the kinds of microbial community differences that might be associated with such non-infectious diseases or conditions.

There is an important caveat here.

Researchers don’t yet know if a change in a microbial community leads to a disease or if a microbial community changes in response to the development of a disease.

Researchers don’t yet know if a change in a microbial community leads to a disease or if a microbial community changes in response to the development of a disease.

Although a snapshot of the microbiome measurements were taken in these earlier HMP studies, an understanding of how these communities change over time was needed if these kinds of studies could begin to determine the cause-and-effect relationships of the microbiome with disease. This is especially important because microbial populations in the human body change over one’s lifetime and are affected by diet, stress and other environmental factors.

The researchers would need to, in effect, make a video of the microbiome and the human body.

What are they doing? How humans and microbes interact

The NIH Common Fund initiated a second phase, the Integrative Human Microbiome Project (iHMP) in 2014. This phase was designed to develop the methods for measuring and analyzing biomolecules such as RNA, proteins and metabolism-linked substances called metabolites from the microbiome and the human host.

The researchers applied these methods to study three model microbiome-associated conditions: preterm birth, inflammatory bowel disease (IBD) and prediabetes. The goal was to evaluate which biological properties or suite of properties were proven to be the most useful for gaining insight into these conditions.

The researchers studied the microbiome of healthy and affected individuals for each of the conditions, over a prolonged period of time.

“We had completed an inventory of the microbial species in the human body. But microbes interact with each other and with the host. Each one can influence the other,” said Lita Proctor, program director of the Human Microbiome Project. “So, it was important that we captured these time-sensitive interactions between the microbiome and the host.”

Preterm birth and missing bacteria in the vaginal microbiome

Preterm birth, a condition where women give birth before 37 weeks of pregnancy, is the second most common cause of neonatal death worldwide. Preterm birth instances also vary significantly by population. According to the Centers for Disease Control and Prevention, in 2016, the rate of premature birth among African-American women was 14% compared to white women at 9% -- a nearly 50% difference.

Because a pregnant woman’s healthy vaginal microbiome is closely associated with the healthy birth of an infant, the iHMP’s first model system, the Vaginal Microbiome Consortium Multi-Omic Microbiome Study: Pregnancy Initiative (MOMS-PI), sought to characterize the vaginal microbiomes of pregnant women to gauge their risk for preterm birth, with a particular focus on African-American women.

The project enrolled and followed a group of pregnant women that included a total of 45 preterm births and 90 full-term births. By looking at microbiome data, as well as other features, including gene expression, protein and metabolite levels from both the microbiome and the subjects, the researchers came upon something surprising.

In one study result, researchers compared the normal progression of the vaginal microbiomes in uncomplicated pregnancies to those in women giving premature births. They found that during the early stages of pregnancy, women (mostly African-American) who experienced preterm birth had significantly lower levels of an naturally occurring bacterium, Lactobacillus crispatus. It is a key member of the microbiome known to be important for maintaining a healthy vaginal environment.

However, by the end of the first trimester, the vaginal microbiomes of these women tended to return to a normal, Lactobacillus-dominated community. The researchers now want to know why this microbe effectively disappears from the mother’s vagina during the earliest days of her pregnancy.

This result also confirmed to the researchers that studying the microbial community composition of a woman early in pregnancy, and not later, may be most useful in predicting risk for preterm birth.

Studying the microbial community composition of a woman early in pregnancy, and not later, may be most useful in predicting risk for preterm birth.

“These results were particularly gratifying and initially unexpected, because they suggest to us that biomarkers for preterm birth may be more readily detected early in pregnancy,” said Gregory Buck, Ph.D., senior investigator of the study and professor at Virginia Commonwealth University. “Our predictive model has sensitivity and specificity already slightly better than currently used clinical parameters.”

If these results are replicated in other clinical trials with African-American women, they may provide doctors with microbiome-based biological markers that could predict preterm births early in a woman’s pregnancy, and so provide more time and opportunity for interventions to prevent premature birth.

Inflammatory bowel disease and the loss of specific microbial products

The so-called “Western diet,” usually characterized as a high fat, low fiber diet, and frequent use of antibiotics has been associated with shifts in the composition of the gut microbiome from a healthy one. These shifts are in turn closely associated with increased incidence of inflammatory bowel diseases (IBD) and the IBD subtypes Crohn’s disease and ulcerative colitis. Both are painful conditions, marked by fever, loss of appetite and fatigue, and affect over 3 million people in the US.

There has been little knowledge about the role of the microbiome in IBD and how the microbiome affects the progression of the disease. Until now.

Curtis Huttenhower, Ph.D., at the Harvard School of Public Health, Ramnik Xavier, M.D., at Massachusetts General Hospital and other members of the iHMP’s second model system, the Inflammatory Bowel Disease Multi-omics (IBDMBD) project, have provided the most comprehensive description of the microbial community in IBD.    

They recruited Crohn’s disease and ulcerative colitis patients who either showed low level inflammation or had fully developed disease. One hundred and thirty-two patients were studied for a year, and, for each, researchers took intestinal biopsies at the beginning of the study, collected stool samples every fortnight, and drew blood samples every four months from these subjects.

Similar to the preterm study, researchers analyzed these samples for microbiome composition as well as proteins, metabolites and other biological properties. With that data, researchers were able to study the interplay of the microbiome and disease state over time.

In many cases, the microbiome of IBD patients changed entirely in makeup over the course of a few weeks. This kind of dramatic change was rare in the non-IBD participants.

In many cases, the microbiome of IBD patients changed entirely in makeup over the course of a few weeks. 

These researchers found that microbes that are commonly found in high abundance in healthy gut microbiomes, such as F. prausnitzii and R. hominis, were reduced in the IBD patients. E. coli, another bacterium that is also commonly found in healthy gut microbiomes but usually present in very low levels, increased in the IBD patients.  

Bacteria in our gut are essential for breaking down our diet. During this breakdown, bacteria produce compounds, many of which serve as communication signals between the microbiome and the human host. Losing these bacteria can therefore reduce these important signaling molecules.

This study found a significant reduction in gut butyrate levels, a signaling molecule made by gut bacteria that reduces inflammation and strengthens the gut lining. The researchers hypothesize that the decrease in butyrate may be due to the notable decrease in microbes such as F. prausnitzii and R. hominis, which may be the key butyrate producers in the gut microbiome.

"Each individual's microbiome and immune system clearly responded in a distinct, coordinated way. While this makes the complex interrelationships that lead to disease more difficult to tease apart, it points us at new opportunities to potentially disrupt these unwanted feedback loops, leading to new ways to treat or manage IBD," said Huttenhower.  

Preemptive strike: Type 2 diabetes in its early stage

Type 2 diabetes (T2D) is nearing epic proportions in the world and is closely linked to the rising epidemic of obesity. Ten percent of the US adult population has T2D, and another 30% are prediabetic. Prediabetes is a health condition where a person’s blood sugar levels are elevated, but not sufficiently high enough to warrant a T2D diagnosis. This happens because both prediabetic and T2D individuals usually suffer from insulin resistance, where the body no longer responds to the hormone’s signals to move glucose out of the bloodstream and into cells. Almost 70% of the prediabetic population will become diabetic in their lifetime.

The iHMP’s third model system, the Integrated Personal -Omics Project (IPOP), led by Michael Snyder, Ph.D., Stanford University, and George Weinstock, Ph.D., Jackson Laboratory for Genomic Medicine, was formed to answer fundamental questions such as: what causes the progression of prediabetes to T2D? What stressors during the prediabetes state increase the risk of such a shift?

The IPOP study recruited 106 healthy and prediabetic individuals, who were studied over a period of four years. Their gut and nasal microbiomes, as well as biological factors of both host and microbiome were followed over those four years.

To start with, researchers found that healthy participants had different gut microbiome makeup than those with prediabetic symptoms, thereby already setting up these prediabetic patients with a possibly diminished or unhealthy microbiome.

Further, a subset of the healthy and prediabetic individuals experienced periods of respiratory viral infection, during which researchers immunized them both with the influenza vaccine.

Usually when a person has respiratory viral infections, their immune system is activated to fight the infection. Interestingly, the study found that when prediabetic participants with respiratory viral infection were immunized, their immune system was slow to respond, compared to healthy participants, but still effective. Since type 2 diabetes is linked to immune system-related stresses, the researchers suggested that it is possible that immunization could protect prediabetic patients against the onset of the full disease.

It is possible that immunization could protect prediabetic patients against the onset of the full disease.

“It’s incredible to see that people who are insulin resistant were different even at their healthy baseline with several microbial differences,” said Snyder.

There is evidence to suggest that prediabetic patients have an altered microbiome that also appears to be linked to a weakened immune system. Prediabetic patients may benefit from immunization to boost their immune system, in effect protecting them from developing T2D. Only larger studies will be able to verify this hypothesis.

Future insights

The iHMP has produced the largest collection of biological information on microbial community composition and gene expression, proteins, metabolites and immune system properties from both the human body and its microbiome followed over time through three microbiome-related conditions.

By tracking biological features within the human body and its microbiome, the iHMP projects revealed that loss or gain of certain microbes and their properties is closely associated with the state of preterm birth, IBD and prediabetes in individuals.

In some cases, these factors served as early indicators of the diseases and so provide promise for the future development of predictive biomarkers.

The suite of data collected in these three model microbiome conditions serve as a resource for the larger community to test which biological factors will be the most insightful in their particular research. Data from these studies has been archived and curated by the HMP Data Coordination Center and is available for unrestricted use for further research by the broader community.

Last updated: May 29, 2019