DIET COKE is one of science’s great miracles. Ordinary Coca-Cola relies on lashings of sugar to achieve its trademark sickly sweetness—15.9 grams per can, or about a third of the total daily intake recommended for women by Britain’s National Health Service. A can of Diet Coke, by contrast, contains no sugar at all. It owes its sweetness to aspartame and acesulfame-K, a pair of chemicals that are far sweeter than ordinary sugar, but which provide the body with no energy at all.

In this section
Related topics

That magic combination of sweetness without calories has made artificial sweeteners among the most widely used food additives in a world that is struggling to keep its waistline in check. But people generally dislike the idea of “chemicals” in their food, and sweeteners have attracted their share of scare stories. The idea that they cause cancer has proved especially hard to shift, despite no evidence to suggest it is true. A lesser-known (though more respectable) worry is that consuming them might—ironically, and in defiance of common sense—be associated with obesity.

A paper just published in Nature bolsters that view. It provides a big dollop of evidence in support of an emerging idea that artificial sweeteners are not directly bad for people (humans cannot even digest most of them). Rather, they may be bad for the zillions of microbes that live in people’s guts—and this, in turn, may be bad for their human hosts.

Between three and ten times as many bacteria live on or in the typical human as there are cells in that person’s body. Researchers have only recently begun to appreciate the importance of these fellow travellers, known collectively as the “microbiome”. Gut bacteria, in particular, seem able to affect all kinds of bodily functions. Their actions and secretions have been implicated in everything from depression and arthritis to the regulation of the immune system.

Several previous studies have suggested that artificial sweeteners might affect intestinal bacteria. But the Nature paper, whose lead authors are Eran Elinav and Eran Segal, of the Weizmann Institute of Science, in Israel, is the most robust yet. Like much biomedical research, the initial work was done in mice. Three groups of rodents were given water containing aspartame, sucralose or saccharin, three common commercial sugar substitutes. Three control groups were given plain water or water laced with glucose or sucrose—sugars from which the body can extract energy.

After a week, Dr Elinav and Dr Segal gave their animals a hefty dose of glucose and measured how well they processed it (inability to do so properly is a risk factor for obesity, and is characteristic of diabetes). The mice drinking the artificial sweeteners had higher levels of glucose in their blood than did their confrères who had been sipping water or ordinary sugar.

To check whether the sweeteners were affecting the murine microbiome, the researchers dosed their mice with broad-spectrum antibiotics. Sure enough, killing off the gut bacteria reversed the metabolic changes. To make doubly sure, they transplanted faeces from mice that had been drinking artificial sweeteners into others that had been raised in sterile conditions, and which, therefore, had no gut bacteria of their own. Once the transplanted bacteria had colonised their new hosts, these too began showing signs of glucose intolerance. Gene sequencing confirmed that mice fed artificial sweeteners had a notably different set of bacteria living in their guts from those fed on the natural kind. Intriguingly, the microbiomes of the sweetener-fed mice looked a lot like those found, by other studies, in obese individuals.

Mice, of course, are not people, and it is here that things get trickier. To check whether their results applied to humans, Dr Elinav and Dr Segal looked at data from an ongoing human nutritional study that both are involved in. As with the mice, they found a correlation between consumption of artificial sweeteners and various risk factors for diabetes, including higher weight, a greater waist-to-hip ratio and higher levels of glucose in the blood.

But that result runs against the general weight of evidence. Artificial sweeteners are well-studied, and although different papers have come to different conclusions, meta-analyses (which combine lots of studies in a statistically meaningful way) have tended to discount the idea that they make people fat. Indeed, the most recent analysis, published in 2014, found that, when considering only the highest-quality evidence, food and drink sweetened with aspartame and the like were associated with a moderate reduction in weight.

But the epidemiological research is tricky to interpret, says Kristina Rother, a researcher at the National Institutes of Health, in Maryland. Many studies are of people who are trying actively to lose weight, which means that effects from sweeteners may be swamped by the effect of simply eating less. And finding a good control group is impossible, because artificial sweeteners are found in everything from toothpaste to painkillers.

Dr Elinav offers another theory that may reconcile his laboratory results with the epidemiological evidence. When he and Dr Segal conducted a small-scale human version of their mouse trial, dosing seven people with saccharin for a week, they found similar sorts of biochemical changes to the ones they had seen in their mice—but in only four of the participants.

Unlike those of mice—animals which are enthusiastic eaters of each others’ faeces, and which thereby regularly swap gut bacteria—the microbiomes of humans differ from one individual to the next, says Dr Elinav. It is a lot to hang on one small experiment, but if the unpleasant effects of artificial sweeteners affect only some people, that could explain why the large epidemiological studies have failed to find that they consistently make people fat.