Skip to main content
Premium Trial:

Request an Annual Quote

GenomeWeb Feature: Researchers Scour the Gut Microbiome for Diabetes Associations

NEW YORK (GenomeWeb News) – The human gut is teeming with microorganisms. By some estimates, it is made up of between 10 trillion and 100 trillion microorganisms, including more than 1,000 bacterial species, the vast majority of which belong to either the Firmicutes or the Bacteroidetes phyla.

While the Human Microbiome Project surveyed microbial residents of a number of body sites, coming up with a reference set of microbial genomes, its initial focus was on the healthy microbiome.

But there is increasing evidence that the microbiome also contributes to making people sick, including influencing susceptibility to complex diseases.

"I do think that we'll find a lot of complex diseases that have a human genetic component, an environmental component, and a microbial component," said Joseph Petrosino, the director of the Alkek Center for Metagenomics and Microbiome Research at Baylor College of Medicine in Texas. "That many organisms living on us and in us having no impact whatsoever on health and disease, I think, is something that we can all put behind us. I think there is definitely a contribution."

That contribution may be particularly influential for metabolic diseases. For example, the makeup of the gut microbiome has been linked on several occasions to obesity. A study from Jeffrey Gordon's group at Washington University in Proceedings of the National Academy of Sciences in 2005 examined the microbiome of obese and lean mice, finding a difference in the ratio of Firmicutes and Bacteroidetes present, indicating a link between the microbiome and host health. In a follow-up Nature study, they found the microbiome of the obese mice was also associated with increased energy harvesting from their diet.

Another study, which appeared in PLOS One last summer from Claire Fraser's group at the University of Maryland School of Medicine linked some gut bacteria taxa to metabolic syndrome symptoms in Amish populations.

And more recent studies have tightened the connection between the microbiome and disease, finding that the microbiomes of people with type 2 diabetes have distinct features. The implications of those differences have yet to be worked out, but they do imply that there's a role for the microbiome — as well as genetics and the environment — in the clinic for diabetes care, and possibly for other disorders as well.

"[The] gut microbiome has quite a significant contribution on diabetes risk [and] development," Jun Wang, the director of BGI, said in an email to GenomeWeb Daily News, adding that "type 2 diabetes is a complex disorder influenced by both genetic and environmental components, and it has become a major public health issue throughout the world. The microbiome plays an important role [as an] environmental factor which can cause many diseases including T2D."

The type 2 diabetes microbiome

Two groups of researchers recently delved into the gut microbiome of people with type 2 diabetes to try to tease out whether there were any differences between their microbiomes and those of people without diabetes, and how such differences may have led to disease.

"In China, T2D becomes more and more [common]," Wang added. "T2D is a complex disorder influenced by both genetic and environmental components, but the genetic background of Chinese people is impossible to change that much in only dozens of years, while, in contrast, daily life has really changed much in China including food and medicine conditions. That's why we [concentrated] on the gut microbiome, as an environmental cause."

Wang and his team at BGI examined the gut microbes of 345 Chinese people with and without T2D to identify metagenomic markers linked to diabetes. Their approach, which they dubbed a metagenome-wide association study, aims to better capture the complex association between genes and disease in a metagenomic study, Wang said.

For the first step of their two-stage MGWAS, Wang and his colleagues used shotgun sequencing to characterize the gut microbiomes of 145 fecal samples from people with and without type 2 diabetes. For those samples, they examined both the types of microbes and the genes making up the gut microbiomes.

Then the researchers folded in an additional 200 cases and controls to validate the findings from the first stage of the study. In this way, they validated nearly 60,000 T2D-related gene markers, as they noted in their October Nature paper.

Instead of characterizing all the species in the two sets of microbiomes, the BGI team focused on what they called metagenomic linkage groups, or MLGs — groups of genes found in the various samples at consistent abundance levels. By this definition, they identified some 47 MLGs among the T2D-associated markers.

Interestingly, they noted that those markers were enriched in a pathway involved in the membrane transportation of sugars. People with T2D, the researchers also found, had lower numbers of butyrate-producing bacteria in their guts. Additionally, they reported that people with T2D also had higher levels of opportunistic pathogens in their guts.

This indicated to the BGI team that people with T2D have a moderate level of dysbiosis in their gut.

"In our study, functions lacked or reduced in T2D patients include butyrate biosynthesis, cell motility, and metabolism of cofactors and vitamins," Wang said. "But don't forget, [there] are also some functions increased in T2D patients such as membrane transport of sugar, BACC transport, methane metabolism, and sulfate reduction."

Similarly, a group of Swedish researchers, including Chalmers University of Technology's Jens Nielsen, examined the gut microbiomes of 70-year-old European women with normal, impaired, and diabetic glucose control, as they reported in Nature in June.

Drawing on this cohort of 145 women, the Swedish group also used shotgun genome sequencing using the Illumina HiSeq 2000 platform to examine the composition and functional changes within the gut microbiomes of these women, finding a number of differences between the three groups. For example, women with diabetic glucose control had lower levels of some Clostridium species and higher numbers of Lactobacillus species than the women with normal glucose control.

To study the gene content of these microbiome sets, Nielsen and his colleagues looked at metagenomic clusters — genes that tended to show up together in the research subjects — a concept similar to the MLGs that the BGI team used.

Twenty-six such clusters, the Swedish team found, had differential abundances between the T2D and control groups.

"We found that it's not a single or a few sets of microorganisms that [are] conferring the function of type 2 diabetes, but it really is multiple sets. It was a surprising finding for us because we needed so many bugs," Nielsen said.

The clusters enriched in women with T2D involved genes associated with starch and glucose metabolism as well as with fatty acid biosynthesis and cysteine and methionine metabolism, which the researchers said may be important for oxidative stress response.

"There is a clear correlation demonstrated in our studies, but also others … between people having type 2 diabetes and [the gut microbiome]," Nielsen added.

The two studies, the Swedish researchers pointed out, came up with similar results. The control populations of both studies had enriched microbial functions in flagellar assembly, as well as in the metabolisms of cofactors and vitamins. Both also noted similar changes in the levels of Clostridium species and lower levels of butyrate biosynthesis in people with T2D.

Cause or effect?

Whether the gut microbiome differences are the cause or the result of developing T2D isn't yet known — and may not even matter.

"We found these differences are associated with T2D, but it's still too early to make a conclusion for a cause or a result," BGI's Wang said.

He added that their finding of reduced butyrate biosynthesis in T2D patients could be related to insulin sensitivity. Through the control of butyrate biosynthesis, the microbiome could change how sensitive patients are to insulin. Wang cautioned, though, that genetic differences between people could also affect the gut environment, making it more or less suited for microbes involved in butyrate biosynthesis. "To discover the whole mechanism and to predict the causality, we still have a lot of further work to do," Wang added.

Nieslen pointed out that their results indicate that such information about the microbiome could be used to guide treatment, no matter if it is the cause or the effect of T2D, as it can identify those with or at risk of developing the disease. "You can use a healthy microbiome, use part of that as a probiotic for treatment … this may be the way to treat type 2 diabetes in the future," he said.

Indeed, for other microbiome-associated diseases, researchers are exploring probiotics and even fecal microbiome transplants as possible therapeutics, to upend the microbiome and bring it back to a healthier state. Clinicians, for example, have used fecal transplants to treat recalcitrant Clostridium difficile infections with success.

Creating classifiers

Wang and his colleagues sifted through the markers they uncovered, homing in on 50 gene markers that they determined was an optimal gene set to classify people with T2D.

As they reported in Nature, they tested their classifier on an independent group of 11 people with T2D and 12 controls. The top eight in their index, the researchers said, all had T2D, indicating that gut health could possibly be monitored for signs of changes related to diabetes.

Nielsen and his colleagues, meanwhile, developed their own classifier by training a random forest model on 50 people with type 2 diabetes and controls, and validating it on another cohort of 50.

They then tested their classifier on the 49 women in their cohort with impaired glucose function, but which had not yet progressed to the level of being diabetes. It divvied those women up, placing 10 of them in the non-diabetic subgroup and 34 of them in the T2D group.

Nielsen said that among those with impaired glucose tolerance, it can predict if they have a higher risk of developing type 2 diabetes. "We do have some preliminary results, but we do need to look further into that to see if it really is like that," he added.

Nielsen and his colleagues also tested their classifier on the Chinese cohort, and while it could, for the most part, distinguish those with T2D from those without, the investigators noted that the MGCs that differentiated the those with and without T2D differed between the Swedish and Chinese cohorts. In particular, they noted differences in the contribution of Lactobacillus and Akkermansia to the patient classification.

While there were differences in the make-up of the cohorts — the Swedish cohort consisted of 70-year-old women and the Chinese cohort of both men and women who were older — the researchers speculated that population differences could also be at play, and that diagnostics tools may have to be tailored to specific populations.

"When talking about the differences between their findings and ours, we should be conscious that a lot of things could change the structure of our gut microbiome including food, environment, lifestyle, medicine, et cetera," Wang noted. "In Europe and in China, people have different habits in daily life, so it is reasonable enough to find that their gut microbiome adapted to, for example, [having] different metabolic pathways."

The other diabetes, too

Of course, the contents of the microbiome may influence a number of diseases, including the related type 1 diabetes.

While T1D is an autoimmune disorder with a defined genetic component, Baylor's Petrosino noted that it, too, appears to have an environmental trigger that may be bacterial or viral in origin. "We've been accumulating data in the area of autoimmunity, and the microbiome specifically, and how the microbiota associates with T-cell populations that can drive inflammation," he said. "It's easy to extrapolate that into thinking that the microbiota may have a significant contribution in the type 1 diabetes disease manifestation."

The Environmental Determinants of Diabetes in the Young, or TEDDY, project, a 15-year prospective study, is searching for triggers leading to T1D development, and is about halfway done. One subset of that project is examining the gut microbiome of children as they develop T1D to see if there is a microbial influence.

So far, the project has enrolled some 130 children with T1D, and Petrosino's lab is studying stool and plasma from those kids and controls.

The hope, he said, is to develop a vaccine — if the trigger is viral — or probiotic or antibiotic treatment — if the trigger is bacterial — that prevents or delays the onset of disease. Or, he added, like the T2D researchers, they might be able to develop a classifier that predicts which children with a predisposition for T1D are more likely to develop the disorder.

The microbiome in the clinic

In the future, researchers see the microbiome as another source of information to help clinicians and patients figure out how to prevent or treat disease. "Many physicians are beginning to embrace this and really see this as something, new information for the future," Nielsen said.

Indeed, Wang said that as sequencing costs decline metagenomic sequencing may play a larger and larger role in disease diagnosis.

"Microbiome data is a new, but promising field in the clinic [and] a lot of hard work [is] waiting for us, but we really have faith in it," said Wang.

Petrosino added that a person's microbiome might become another data point in the personalized medicine portfolio.

"I think people will eventually have some of the features of their microbiome stored in their clinical data file as well because, as we know from research that's been published, that can have an impact on anything from disease onset to how certain drugs are metabolized," he said. "It may be a little bit down the road before that's realized, but I think that's going to be a part of personalized medicine."