I've been hacked! My microbes made me eat that jelly doughnut!

overweight and diet 1318798 639x715

A hot topic these days is how the millions of bacteria in the digestive system, collectively known as gut microbiota, might be influencing many aspects of health, including mental health, autoimmune diseases, metabolic diseases, and some cancers. (A very similar term, microbiome, refers to the collective bacterial genome.) The journal Nature in conjunction with Scientific American published a supplement in February 2015 titled "Innovations in the Biome", and Scientific American has published several interesting articles on connections with obesity and inflammatory bowel disease.

The microbiome, described by Joshua Lederberg as "the ecological community of commensal, symbiotic, and pathogenic microorganisms that literally share our body space", comprises approximately 10,000 microbial species (bacteria, eukaryotes, and viruses). Almost all of the gut bacteria fit into 30 or 40 different species. We each host approximately 100 trillion microbial cells in the gut that account for several pounds of our total body mass. These bacterial cells outnumber our own cells by 10 to 1, and bacterial genes outnumber our genes by 100 to 1. Some of the well-characterized useful functions of gut bacteria include fermenting unused energy stores, training the immune system, preventing growth of pathogenic bacteria, regulating gut development, and producing vitamins and hormones. But they also seem to be influential in many organs and diseases outside the digestive system. Every time I turn on the TV or radio, or look at a newspaper, a magazine, or a research journal, there's something about the microbiome, antibiotics, probiotics, prebiotics, the enteric nervous system, the second brain, fiber, fermented foods, bacterial diversity. Just like the bacteria themselves, it's everywhere.

The National Institutes of Health (NIH) created the Human Microbiome Project in 2008 to facilitate work in the area of the human microbiome. Their initial projects focused on developing metagenomics databases and computational tools for characterizing the microbiome in healthy adults and in cohorts of patients with specific microbiome-associated diseases.

In 2012, researchers at the University of Colorado Boulder launched the American Gut Project, a crowd-funded, open-assess project where anyone can pay for sequencing of their gut bacteria. Participants provide samples along with personal health information. In turn, this provides the researchers with data they can use for basic research. Michael Pollan, a bestselling author and journalist who writes about food and agriculture, is one of the thousands of people who have participated.

In the area of metabolic syndrome and obesity, I found two research papers especially interesting. In the first, researchers were able to bring on metabolic syndrome in germ-free wild-type (WT) mice by transplanting gut microbiome extracts from mice with metabolic syndrome symptoms and then reverse these symptoms with antibiotics. In the second study, scientists were able to induce obesity in germ-free mice by transplanting fecal microbiome samples from an obese human subject. These studies are described in more detail below.

Metabolic syndrome study

Metabolic syndrome is a collection of five medical conditions: abdominal obesity, elevated blood pressure, elevated fasting plasma glucose, elevated serum triglycerides, and low high-density lipoprotein (HDL) levels. Metabolic syndrome is associated with a risk of developing cardiovascular disease and diabetes. Approximately 34% of the adult population in the US have metabolic syndrome.

In their 2010 study, Vijay-Kumar and colleagues compared mice that were deficient in Toll-like receptor 5 (TLRT) (called T5 knock-out, or T5KO mice) versus WT mice. The "guardian of the gut", TLRT is a protein that plays a fundamental role in governing the interface between host and microbiome; it recognizes pathogens and activates innate immunity for immediate defense against infection. Vijay-Kumar's findings confirm that the T5KO mice have mild loss of glycemic control, a condition typically seen in humans with metabolic syndrome: At 20 weeks of age, the T5KO mice had body masses that were 20% greater than those of WT mice. MRI showed increased fat mass, particularly visceral fat, in the T5KO mice. Fat-pad mass was significantly increased in the T5KO mice, both males and females, compared with their WT counterparts. Serum cholesterol and triglycerides and systolic and diastolic blood pressures were also significantly higher in the T5KO mice. Higher production of the proinflammatory cytokines interferon-γ and interleukin-1β was noted in adipose tissue of T5KO mice in ex vivo analysis. Fasting blood glucose levels were somewhat higher in the T5KO mice, and following glucose administration, T5KO mice showed impaired ability to restore blood glucose to baseline levels. Baseline insulin levels were substantially elevated in T5KO mice, as was the amount of insulin produced in response to glucose challenge. T5KO mice showed elevated serum adipokine lipocalin-2, which promotes insulin resistance, and showed an increase in number and size of pancreatic islets that immunostained positive for insulin.

When a high-fat diet was administered for 8 weeks, both T5KO and WT mice showed significant increases in body mass and fat-pad mass, with increased serum levels of triglycerides, cholesterol, leptin, and insulin. However, only the T5KO mice became diabetic (fasting glucose > 120 mg/dL). They also exhibited inflammatory infiltrates in the pancreatic islets and displayed hepatic steatosis (fatty liver). Thus, a high-fat diet exacerbated the metabolic syndrome symptoms in the T5KO mice.

When food intake was evaluated, the researchers found that T5KO mice consumed about 10% more food than WT littermates and had greater stool output. No between-group differences in dietary energy harvest were seen. In another experiment of 4-week-old T5KO mice, food intake was restricted for 8 weeks to the amount consumed by a control group of WT mice; most metabolic abnormalities including increased body mass, fat-pad mass, and increased glucose, lipids, and insulin were prevented. However, these mice still exhibited a decreased response to exogenous insulin.

Pyrosequencing of the 16S ribosomal RNA (rRNA) genes in the ceca showed an abundance of bacterial phyla in all samples. UniFrac analysis showed a significant difference in bacterial species composition in the 2 groups. Despite the marked interindividual differences (typical both in mice and humans), they identified 116 bacterial phylotypes that were consistently either enriched or reduced in T5KO mice relative to WT mice.

Interestingly, the researchers then evaluated a possible role for the gut microbiome in the development of metabolic disease in T5KO mice. They treated the T5KO mice with a broad-spectrum antibiotic, thus lowering the gut bacterial load by 90%. This corrected T5KO metabolic syndrome symptoms (food intake, fasting glucose, and fat pad measurements) back to WT levels.

To investigate whether the gut microbiome differences seen in T5KO mice were a cause or consequence of metabolic syndrome, the researchers then transplanted T5KO or WT microbiota into WT germ-free mice. The T5KO recipients showed many aspects of the T5KO phenotype including hyperphagia, obesity, hyperglycemia, insulin resistance, colomegaly, and elevated levels of proinflammatory cytokines. This suggests that the changes in the gut microbiome observed in T5KO mice are likely a contributing factor in the development of metabolic syndrome.

Obesity study

The Ridaura et al paper reports findings in "humanized mice". Genetically identical baby mice were raised germ free, so that their intestinal tracts contained no bacteria. The researchers then transplanted intestinal microbes collected via fecal sampling from 4 pairs of discordant human twins into the mice (each twin pair comprised one obese and one non-obese twin). UniFrac-based comparison of bacterial 16S rRNA datasets from the mice and human participants confirmed that the transplant recipients efficiently and reproducibly captured the taxonomic features of their donor's microbiota.

Mice were individually caged and fed a low-fat, high plant polysaccharide, sterilized commercial chow without restriction. At Day 15, the mice who had received microbes from the obese twin had significantly greater adipose mass and significantly higher epididymal fat pad weights compared with those who received microbes from the lean twin. Dietary intake and inflammatory responses (FACS analysis of CD4+ and CD8+ T-cell compartments) were not significantly different in the 2 groups.

Genetic analysis of fecal samples showed that the microbiomes in the obese-transplanted mice had reduced ability to break down and ferment polysaccharides compared with those in the lean-transplanted mice.

When pairs of mice, one obese-transplanted and one lean-transplanted, were housed together, all of the mice became lean. Mice are typically coprophagic, resulting in a mixing of gut bacterial species. This was confirmed by genetic analysis, notably the introduction of Bacteroidetes species from the lean-transplanted to the obese-transplanted mice. It's interesting to note that the lean-transplanted mice did not become obese.

When individually-caged transplanted mice were fed a diet with low saturated fat/high fruit and vegetable content, the obese-transplanted mice again became significantly fatter with increases in both adipose and lean body mass in comparison with lean-transplanted mice. Cohousing with lean-transplanted mice again prevented obesity from developing in the obese-transplanted mice, and again the greatest "invader" was a Bacteroidetes. Fecal microbial biomass was greater in the lean-transplanted and lean/obese cohoused mice than in the obese-transplanted mice; this was true with the regular chow diet or with the low saturated fat/high fruit and vegetable diet.

A high saturated fat/low fruit and vegetable diet was also evaluated. Again, significant differences in body mass were seen between the obese- and lean-transplanted mice. But this time, cohousing of the two types of mice did not prevent the increase in body mass in the obese-transplanted mice, and significant invasion of members of the "lean" microbiota did not occur. Fat mass and lean body mass were greater in all groups (lean or obese-transplanted, cohoused or not) with the high-fat diet than with the low-fat diet.

Translating the results of these and many other studies into useful clinical information for humans will require jumping some significant scientific and regulatory hurdles. In the meantime, better pass on the jelly doughnuts...

Note: Thanks to Sandoval and Seeley for inspiration for the title of this post.

Blogger: Ginny Fleming, Founder, Lucidize Medical & Scientific Editing. Chief capacities: medical, scientific, and technical writing and editing.