Diabetes
Recent research shows that the gut microbiota is definitely linked to type 2 diabetes (T2D), and to both obesity and chronic inflammation, which are independent risk factors for T2D (this science-heavy link requires a free account), here's another good explanation of the diabetes-obesity-inflammation connections. However, the exact nature of all these relationships isn't yet clear, and other factors like diet and exercise are also clearly involved. It seems likely that there are complex feedback loops between some or all of these factors that may prevent any simple linear description of causation, but on the other hand that complexity may provide multiple avenues for therapies to interrupt the course of the disease. Given the tremendous morbidity and economic costs of diabetes (which are projected to continue to increase in the U.S. and worldwide), this is now a high priority for research. You may want to check out the Obesity Advanced Topic page also.
There are two main approaches to investigating these relationships. The first is mechanistic investigations to identify the specific microbial strains, molecules (including hormones), cell signalling pathways, and genes that are involved. This is almost always done using mice (and occasionally rats), since the microbiota of laboratory mice can be completely controlled, and many lineages of mice exist with known genetic deletion mutations that specifically eliminate individual proteins, allowing the involvement of that protein to be demonstrated conclusively. This works reasonably well even though we really want to know about humans, not mice, since metabolism is regulated similarly for most mammals. The second approach is to compare the microbiota of people with and without the disease, or of people with different degrees of disease, or perhaps best of all, of individual people over time who are treated for the disease and show improvement (or not). This has the obvious advantage of collecting data in the most relevant possible context (i.e. in humans), but some experiments are impossible or unethical with human subjects. When both approaches support the same conclusions, we can be pretty confident that we're on the right track.
Some of the ways that gut bacteria may influence T2D directly or indirectly include:
Butyrate, which is produced by some gut bacteria (e.g. Faecalibacterium prauznitzii, Roseburia sp.) that grow mostly on complex plant carbohydrates, has anti-inflammatory effects which may decrease diabetes risk. Other short chain fatty acids (all of which are by-products of bacterial fermentation in the colon) may have similar effects, but butyrate appears to be the strongest.
Another effect of short chain fatty acids is to increase hormones that regulate appetite, which may reduce overeating; these hormones also increase the metabolic rate so that additional energy is expended. Both effects will tend to reduce obesity.
A class of receptors (toll-like receptors) in intestinal cells that recognize bacterial components as part of the innate immune system are stimulated to different extents by different members of the gut microbiota. Some bacteria that usually increase in response to a high-fat, high-sugar diet, including members of the Proteobacteria phylum (and particularly the Enterobacteriaceae family), tend to generate more inflammation through these receptors, which in turn tends to perpetuate a higher abundance of the same bacteria since they tolerate inflammation better than most other gut microbes. Chronic inflammation increases the risk of diabetes.
Proteobacteria may also decrease the effectiveness of the gut barrier in keeping both bacteria and bacteria-derived compounds such as lipopolysaccharide (LPS) inside the gut, which also has the effect of producing chronic inflammation.
Unknown members of the gut microbiota may suppress the hormone Fiaf which is produced by the liver, intestinal cells, and fat cells. This in turn allows greater fat accumulation in adipocytes, which promotes obesity, which increases diabetes risk.
Unknown members of the microbiota may generate particular conjugated, long-chain fatty acids from dietary compounds, which in turn stimulates the liver to synthesize additional fatty acids from glucose, resulting in higher levels of triglycerides and more fat storage.
Unknown members of the gut microbiota may downregulate the enzyme AMPK produced by the liver, muscles and brain in response to metabolic stress. This pushes metabolism towards the storage of more fat, and away from oxidizing fat for energy.
And, in a striking study published in October 2014, the use of several common artificial sweeteners (often recommended for diabetics and those at risk for diabetes in lieu of sugar) was shown to alter the gut microbiota in ways that increase obesity and decrease glucose tolerance - precisely the opposite of the desired changes.
There are two main approaches to investigating these relationships. The first is mechanistic investigations to identify the specific microbial strains, molecules (including hormones), cell signalling pathways, and genes that are involved. This is almost always done using mice (and occasionally rats), since the microbiota of laboratory mice can be completely controlled, and many lineages of mice exist with known genetic deletion mutations that specifically eliminate individual proteins, allowing the involvement of that protein to be demonstrated conclusively. This works reasonably well even though we really want to know about humans, not mice, since metabolism is regulated similarly for most mammals. The second approach is to compare the microbiota of people with and without the disease, or of people with different degrees of disease, or perhaps best of all, of individual people over time who are treated for the disease and show improvement (or not). This has the obvious advantage of collecting data in the most relevant possible context (i.e. in humans), but some experiments are impossible or unethical with human subjects. When both approaches support the same conclusions, we can be pretty confident that we're on the right track.
Some of the ways that gut bacteria may influence T2D directly or indirectly include:
Butyrate, which is produced by some gut bacteria (e.g. Faecalibacterium prauznitzii, Roseburia sp.) that grow mostly on complex plant carbohydrates, has anti-inflammatory effects which may decrease diabetes risk. Other short chain fatty acids (all of which are by-products of bacterial fermentation in the colon) may have similar effects, but butyrate appears to be the strongest.
Another effect of short chain fatty acids is to increase hormones that regulate appetite, which may reduce overeating; these hormones also increase the metabolic rate so that additional energy is expended. Both effects will tend to reduce obesity.
A class of receptors (toll-like receptors) in intestinal cells that recognize bacterial components as part of the innate immune system are stimulated to different extents by different members of the gut microbiota. Some bacteria that usually increase in response to a high-fat, high-sugar diet, including members of the Proteobacteria phylum (and particularly the Enterobacteriaceae family), tend to generate more inflammation through these receptors, which in turn tends to perpetuate a higher abundance of the same bacteria since they tolerate inflammation better than most other gut microbes. Chronic inflammation increases the risk of diabetes.
Proteobacteria may also decrease the effectiveness of the gut barrier in keeping both bacteria and bacteria-derived compounds such as lipopolysaccharide (LPS) inside the gut, which also has the effect of producing chronic inflammation.
Unknown members of the gut microbiota may suppress the hormone Fiaf which is produced by the liver, intestinal cells, and fat cells. This in turn allows greater fat accumulation in adipocytes, which promotes obesity, which increases diabetes risk.
Unknown members of the microbiota may generate particular conjugated, long-chain fatty acids from dietary compounds, which in turn stimulates the liver to synthesize additional fatty acids from glucose, resulting in higher levels of triglycerides and more fat storage.
Unknown members of the gut microbiota may downregulate the enzyme AMPK produced by the liver, muscles and brain in response to metabolic stress. This pushes metabolism towards the storage of more fat, and away from oxidizing fat for energy.
And, in a striking study published in October 2014, the use of several common artificial sweeteners (often recommended for diabetics and those at risk for diabetes in lieu of sugar) was shown to alter the gut microbiota in ways that increase obesity and decrease glucose tolerance - precisely the opposite of the desired changes.
Selected Scientific Literature
Everard and Cani, 2013: Review on diabetes, obesity and the gut microbiota, focused mostly on the issue of gut barrier function. (full text)
Cani et al., 2008: Paper establishing the link between the gut microbiota and diabetes via the mechanism of increased gut permeability to LPS. (full text)
Parekh et al., 2014: Another brief review on diabetes/metabolic syndrome, obesity and the gut microbiota. (full text)
Larsen et al., 2010: Comparison of the gut microbiota between adults with and without T2D. (full text)
Zhang et al., 2013: Comparison of the gut microbiota between people at three levels of glucose resistance. (full text)
Qin et al., 2012: A large study in a Chinese cohort comparing the relative abundance of different gut microbial genes in people with or without T2D. A reduction of butyrate-producing bacteria is clearly evident and consistent with other studies; the finding in this paper that the bacterium Akkermansia muciniphila is associated with T2D is in conflict with other studies. (abstract only)
Karlsson et al., 2013: A large study in a European cohort also comparing the relative abundance of different gut microbial genes in people with or without T2D. The study finds that microbial marker genes for T2D in Europeans are somewhat different than in the Chinese cohort of the study above. This might result from differences in methodology, but it does not necessarily mean that one of the studies is right and the other wrong. It's quite possible that different populations have somewhat different gut communities associated with T2D. (abstract only)
Suez et al., 2014: The bombshell study linking artificial sweeteners to increased diabetes risk. (abstract only)
Cani et al., 2008: Paper establishing the link between the gut microbiota and diabetes via the mechanism of increased gut permeability to LPS. (full text)
Parekh et al., 2014: Another brief review on diabetes/metabolic syndrome, obesity and the gut microbiota. (full text)
Larsen et al., 2010: Comparison of the gut microbiota between adults with and without T2D. (full text)
Zhang et al., 2013: Comparison of the gut microbiota between people at three levels of glucose resistance. (full text)
Qin et al., 2012: A large study in a Chinese cohort comparing the relative abundance of different gut microbial genes in people with or without T2D. A reduction of butyrate-producing bacteria is clearly evident and consistent with other studies; the finding in this paper that the bacterium Akkermansia muciniphila is associated with T2D is in conflict with other studies. (abstract only)
Karlsson et al., 2013: A large study in a European cohort also comparing the relative abundance of different gut microbial genes in people with or without T2D. The study finds that microbial marker genes for T2D in Europeans are somewhat different than in the Chinese cohort of the study above. This might result from differences in methodology, but it does not necessarily mean that one of the studies is right and the other wrong. It's quite possible that different populations have somewhat different gut communities associated with T2D. (abstract only)
Suez et al., 2014: The bombshell study linking artificial sweeteners to increased diabetes risk. (abstract only)