Early Antibiotics Change Gut Microbes, Fuel Obesity

What’s the short version?

There are tens of trillions of microbes in our guts, which are important for our digestion and our health. The antibiotics that we take to kill off disease-causing bacteria also indiscriminately nuke these beneficial bugs. Now, a new set of experiments in mice have shown that low, regular doses of antibiotics at an early age can disrupt these microbe communities, leading to weight gain later in life. The increase in body weight was small, but compounded by a high-fat diet. If the results apply to humans, they would add to the large body of evidence suggesting that antibiotics should be used more carefully in infants and children.

“I’m not saying people should never take antibiotics,” says Martin Blaser from the NYU Langone Medical Centre, who led the study. “But we need to be more judicious. Antibiotics can have long-term consequences. I hope that knowledge will enter the examining room, so that parents don’t demand antibiotics and doctors are more cautious about using them.”

This sounds familiar…

Two years ago, Blaser’s team showed that antibiotics can change the gut microbes of young mice, which then grow up fatter. This new study confirms and builds upon those earlier results. More generally, farmers have been fattening livestock for decades by giving them low doses of antibiotics in their food—it’s the microbe connection that’s new. And Blaser himself has been discussing these ideas a lot, in the wake of his recent book, Missing Microbes.

What did the new experiments show?

The new studies represent a huge amount of work, largely done by graduate student Laurie Cox. First, she exposed mice to low doses of penicillin at two points in time: either when they were being weaned at four weeks old, or right from the start of their lives (by dosing their mothers). By 20 weeks of age, the mice that experienced penicillin from birth were heavier and fatter, especially the males. They also had very different gut microbes.

A high-fat diet exacerbated this effect, especially in females. If female mice were raised on fatty chow and penicillin, they put on twice as much fat as those that ate the fatty chow alone. (It’s not clear why there’s a gender difference.)

In these experiments, the mice were on antibiotics for their entire lives. Next, Cox showed that much shorter bursts are enough. Four weeks of early exposure can change the rodents’ gut microbes. These communities returned to normal after eight weeks, but the mice still got fatter, and their immune systems were still weaker.  This suggests that there’s a critical window where microbial upheavals have long-lasting consequences. This makes sense: our gut microbes are an active part of our development. As we grow up, they help to set our metabolism and train our immune systems. Just as schooling and education can have lasting effects on our lives, microbial education might permanently affect a mammal’s health.

Finally, Cox transplanted the microbes from the antibiotic-treated mice into germ-free ones that had no microbes of their own. These recipients also became heavier and fatter. “The fact that you can transfer the [effects] by transferring the microbes is pretty solid evidence of causality,” says Jack Gilbert from the University of Chicago. “It is excellent confirmation of what [Blaser] has been expounding for a while.”

Which microbes were affected?

The penicillin depleted four particular groups, and they’re a slightly weird quartet. One of them, Lactobacillus, is very well known—common in our guts and in probiotics. But the others—two species called Allobaculum and Arthromitus, and a wider group called the Rikenellaceae—are more obscure. Allobaculum was only discovered ten years ago. Arthromitus seems to only live in mice and not humans. These microbes might help to prevent their hosts from putting on too much weight, but it’s too early to say.

How does antibiotic treatment lead to weight gain?

It’s clear that the penicillin did at least four things: it changed the gut microbes; it changed the rodents’ metabolism; it increased inflammation in the gut; and it increased the risk of obesity. But, as Les Dethlefsen from Stanford University told me, it’s hard to know how these effects are connected. The microbes could be responsible for everything else. Alternatively, the drugs themselves could cause some of the effects, and the microbes others. We’re probably looking at a tangled web of causality rather than a linear chain.

How big are the effects?

The antibiotic-exposed mice put on around 10 percent more weight than the unexposed ones. Assuming that the effect applies to humans (and see below), a 70 kilogram person would end up being 77 kilograms—around two body mass index (BMI) units. These figures went up if the mice ate a high-fat diet, and especially if they were female—then, the antibiotic group weighed around 25 percent more than their peers.

But, does this apply to humans?

Mice aren’t humans—that goes without saying. Still, there are many similarities in the ways both species interact with our respective gut microbes, and in how those microbes affect our metabolism. And a few studies that tracked the health of human babies over time found that early antibiotic exposure does affect body weight at later ages. Here’s the most recent one—it found that 32 percent of children who received antibiotics before their first birthday were overweight at age 12, compared to just 18 percent of those who didn’t.

But Blaser cautions that they’re “doing these experiments as models”. That is, they’re meant to tell us about the general sorts of effects that antibiotics produce. The details will probably differ in humans. For a start, our diets and lifestyles are far more varied than those of lab mice, and many of our choices will undoubtedly exacerbate or nullify the effects that Cox described.

Also, the pups in the study received regular doses of penicillin for at least four weeks, if not much longer. You can’t exactly map mouse years onto human years, but this is roughly akin to someone getting antibiotics constantly throughout infancy, or even until adolescence—which almost never happens. The team are now repeating their experiments with short high-dose pulses of antibiotics that more closely mimic the exposures human children get. “We’re giving them antibiotics as if we were treating an ear infection,” says Blaser.

So, should we stop giving antibiotics to kids?

Antibiotic exposure in early life doesn’t guarantee later obesity, but it does seem to increase the risk of it. The question then becomes: how do you weigh that risk against the obvious benefit of curing infections?

Rob Knight, who studies the human microbiome, thinks about this a lot, and especially when his young daughter came down with a Staphylococcus infectionStaphylococcus infection. When I spoke to him a few months ago, he told me, “With staph infections, you want antibiotics. On the one hand, this infection, which could be life-threatening and is causing her a whole lot of pain right now, could evaporate. On the other hand, she could be one BMI fatter at eight. We try to keep her off antibiotics in general but when they work, they’re amazing.”

Blaser too is calling for moderation rather than elimination. “Up to this point, we’ve been viewing antibiotics as just a positive. A doctor might say: It probably won’t help you but it won’t hurt,” he says. “But once you think that it might hurt, you have to recalculate things.” There are also other reasons to practice the caution that Blaser preaches. Foremost among them: the unnecessary overuse of antibiotics has also fuelled the rise of antibiotic-resistant superbugs, ironically hastening the onset of a world where these drugs no longer work.

Once antibiotics are given, can weight gain be avoided?

Maybe. Cox’s transplant experiments are encouraging, says Dethlefsen. If the microbes are causing the problems, it may be possible to manipulate them to prevent some of the long-term consequences of early antibiotic exposure. You could imagine doing that by adding in the right bacteria (through probiotics or faecal transplants), or by providing nutrients that nourish the desirable microbes (prebiotics).

Then again, Cox also found that metabolic problems caused by antibiotics stuck around after the drugs were withdrawn, and even after the microbes had largely normalised.  “That is somewhat discouraging,” says Dethlefsen, “in that such microbial manipulations might need to happen fairly early in life to have the desired effect.”

Either way, it’s unlikely that a simple pill will do the trick. This is medicine as ecosystem management, as restoring a broken habitat much like a conservationist might tend to a logged forest. It requires a different kind of science—one that is now being pieced together by the hundreds of scientists rushing to explore our resident microbes.

What do we still need to know?

Blaser’s team is actively looking into two of the areas I have already discussed: how exactly the disrupted microbes change body weight; and how to restore those disrupted communities. Dethlefsen has another suggestion. We know that not all mice (or humans) who get early antibiotics become overweight, so are there any important differences between those who do and those who don’t? Do these sub-groups also differ in the microbes that survive in (or disappear from) their guts?

Reference: Cox, Yamanishi, Sohn, Alekseyenko, Leung, Cho, Kim, Li, Gao, Mahana, Zarate Rodriguez, Rogers, Robine, Loke & Blaser. 2014. Altering the Intestinal Microbiota during a Critical Developmental Window Has Lasting Metabolic Consequences. Cell http://dx.doi.org/10.1016/j.cell.2014.05.052