Do these (microbial) genes make me look fat?
What germ-free mice can teach us about gut bacteria and weight.
By Daniel Sprockett
Illustration by Daniel Gray
This is part one of a four part on series on the significant but largely unknown impact that microbes have on our lives. Read the rest here.
Every last inch of your body is covered in a thin patina of microbial life. From the oily jungles of your scalp, all the way down to the damp marshes between your toes, nearly every conceivable niche is inhabited by a diverse array of bacteria, fungi, and viruses. At the time of your birth, these microbes rapidly colonize your skin, mouth, lungs, and most of all, your gut. All together, around 100 trillion microbes live on your body, mostly residing in your large intestine. Collectively, they outnumber our own human cells ten to one, and weigh about as much as your brain.
But fret not. While the thought being covered in germs may make you want to bathe, take a second to reflect on its implications. Our microbes remain relatively stable over time, yet also reflect our own unique set of exposures and experiences that occurs over the course of our lifetimes. The more we learn about the important role that our microbiome plays in our lives, the more fuzzy the line becomes that used to divide us from the rest of nature. From this perspective, your body isn’t covered in microbes; your body IS microbe…at least partly. Termed “the microbiome”, this complex community of microbes can even be thought of as a new organ.
This coalescence makes identifying exactly how our microbiome is affecting our health very difficult. For example, if you want to know how a given pathogen makes people sick, you can compare a group of infected people to a similar group of healthy people. If it’s a well-controlled study, you should be able to attribute differences in health to the presence or absence of the pathogen. However, since our microbiome is present in everyone essentially since birth, we don’t have a group lacking a microbiome for comparison. Researchers have gotten around this by breeding laboratory animals that are completely germ-free, named gnotobiotic animals (Greek for “known–life”) since their particular suite of microbes are completely known.
I had the good fortune to spend much of last spring working on a project in Stanford’s gnotobiotic facility, and it was a fascinating experience. Maintaining gnotobiotic mice is an extremely labor-intensive form of animal husbandry, so I was glad to have lots of help. The mice are housed in cages that are kept inside sterile plastic bubbles. These specially designed isolators come complete with an air filtration system and thick, long-sleeved gloves, inverted so researchers can handle the mice. Everything that goes into the isolators, from their food and bedding to whatever tools are needed for your particular experiment, must first be thoroughly sterilized in an autoclave, which is essentially an industrial pressure-cooker the size of a large closet. All of this must be done in strict accordance to detailed experimental guidelines laid out by the researchers and a committee of veterinarians and animal welfare officers.
In keeping these gnotobiotic mice sterile to see what happens when microbes are missing, we’ve found that the gut and immune systems of these animals don’t mature normally. These germ-free mice have a larger caecum (a pouch at the beginning of the large intestine), fewer immune cells, a thinner layer of mucus coating their intestines, and are more susceptible to pathogens. Gnotobiotic animals can also be purposely colonized with one or more bacterial strains or even whole microbial communities that have been isolated from a medically relevant environment. These types of studies show that your indigenous microbiota aren’t merely passive colonists, but rather they actively alter your physiology in ways that directly impact your health. One excellent example is how your microbiome can affect your likelihood of obesity.
The microbes that are commonly found in the gut of obese people are very different than those found in lean people. These microbes are correlated with obesity, but that doesn’t necessarily mean that they cause it. It could very well be that some other factor, like eating too many fatty foods, causes both obesity and the changes that we observe in microbial communities when people gain weight. However, experiments using gnotobiotic mice have enabled scientists to track how obesity changes as a direct result of manipulating which microbes are available to colonize their gut.
One study published last year in the journal Science is an excellent example. Researchers from Washington University in St. Louis first isolated complex microbial communities from stool samples. These samples were taken from sets of twins where one was obese and the other was not. As was previously shown, their microbial communities were quite different from each other. They then used those microbes to colonize different groups of germ-free mice. Mice that received microbes from the obese donor, we’ll call them the Ob mice, gained weight much quicker than those that received microbes from the lean donor, or the Ln mice. Even though both groups ate the same amount of food, the Ob mice gained more weight, and a larger proportion of that weight gain was fatty tissue. They also found that communities from the obese donor produced more enzymes involved in metabolizing branched-chain amino acids, like the those found in dietary protein, whereas microbes from the lean donor were enriched in enzymes that breakdown the fiber found in plant-rich diets. The differences weren’t limited to which microbes were present in the lean or obese twin, but also to the metabolic functions that those microbes perform for their human host.
They then went one step further, and saw what happens when you gave the mice a chance to mingle. Mice from both groups were co-housed in the same cage, and since mice are coprophagic (meaning they eat feces, both their housemates’ and their own), that meant there was quite a bit of microbial mixing going on. Would the microbes from the lean donor prevent weight gain in the Ob mice, or would the Ln mice start gaining weight too?
Amazingly, co-housing the mice actually prevented the weight gain in the Ob mice. The researchers found that the microbes from the Ln mice rapidly invaded the guts of the Ob mice and took over. However, the Ln mice didn’t experience any increases in weight gain, so the invasion wasn’t occurring in both directions. This seemed almost too good to be true. If obese people colonize their guts with microbes from their lean friends or family, will it help them lose weight?
As most things go, the story is actually more complex then that. In their final sets of experiments, the researchers looked to see these microbial interactions are affected by what the mice were eating. They again cohoused Ob and Ln mice, but fed half of them a healthy diet, low in saturated fats and high in fruits and vegetables, while the rest got a poor diet, high in fats and low in fruits and vegetables. They found that the ability of Ln microbes to prevent weight gain in Ob mice was completely diet-depended! The protective Ln microbes weren’t able to invade the guts of the Ob mice in the context of a poor diet that is high in fats and low in plant-based foods.
These experiments show that microbes living in your gut are capable of wielding enormous influence over your waistline. But besides the obvious applications this may have for helping obese people lose weight, these types of studies could also have a positive impact on the way we treat malnutrition, especially in food-insecure regions of the world. The microbiome has been implicated for its role in exacerbating a particularly severe from of malnutrition called kwashiorkor, which is instantly recognizable by the distended abdomens in young children. Researchers selected twins living in rural Malawi where one was afflicted by kwashiorkor while the other was not. Once one child was diagnosed, both children in the twin pair were put on a nutritional treatment regimen involving a peanut-based, ready-to-use therapeutic food. Since we know that gut microbes can affect how we metabolize different foods, researchers looked for differences between twin pairs both during and after treatment. They found that while microbial communities from well-nourished children progressively became more complex, as they normally do in healthy children, communities from children with kwashiorkor transiently changed during treatment, but then quickly relapsed into their previous dysbiotic state when they returned to their normal Malawian diets.
To investigate this further, the researchers then colonized germ-free gnotobiotic mice with microbes from the guts of either well-nourished children or children with kwashiorkor. As was expected, mice colonized by microbes from children with kwashiorkor lost a significant amount of weight compared to mice harboring microbiota from their healthy sibling. Following a switch to a diet of the same therapeutic food supplement, all mice gained weight. However, mice with kwashiorkor microbes gained less weight, and once they were switched back to a normal Malawian diet, rapidly became anorexic again. These studies show that the microbiome interacts with diet in complex ways, and suggest that food supplementation alone may not be the optimal treatment for kwashiorkor.
This is a very active field of research, and new tools are constantly being developed to unravel the elaborate network of interactions that tie microbes to their host. These complex communities interact with the foods you eat (and probably lots of other factors) in a myriad of ways, most of which are only beginning to be understood. What is clear, though, is that weight loss or gain is not always a simple matter of balancing food intake against energy output. Microbes can make all the difference.
Daniel Sprockett is a PhD student in the Department of Microbiology and Immunology in the Stanford University School of Medicine in Palo Alto, California, where he studies the ecology of the human microbiome.
Email: daniel.sprockett@stanford.edu
Twitter: @DanielSprockett
