Sugar Substitutes Surprise

 BY DEREK LOWE

Non-sugar sweeteners have a long and rather tangled history, both chemically and nutritionally. There are several chemical classes: sulfimides/sulfamates (saccharin, acesulfam-K, and cyclamate), small peptides (aspartame and neotame), (mostly) nonmetabolized carbohydrates (sucralose) or ones that are metabolized in a way that does not set off an insulin response (sugar alcohols such as xylitol and sorbitol), and some odd natural-product glycosides, such as steviol glycosides from stevia and mogosides from monk fruit. The sugar alcohols and glycosides are found in nature, but all the man-made types (the first three classes) were discovered accidentally when someone in a lab happened to ingest small amounts and noticed the taste. That last pathway is not as common as it used to be decades ago - who knows what artificial sweeteners we might have missed out on due to lack of sloppy lab technique? Legal disclaimer: I am not encouraging this method of discovery, although I’m assuming that if Barry Sharpless had run across something interesting he would have told the rest of us about it.

But what all of these have in common, especially the man-made ones, are doubts in some members of the public about their safety. Cyclamates were famously banned in the US in the late 1960s due to bladder cancer findings in high-dose rodent studies, but these have failed to reproduce in realistic rodent doses, in primates or in human epidemiology. Only the US and South Korea continue the ban at this point. Somewhat more recently, I cannot count the number of times I have come across claims that aspartame or sucralose are supposedly causing epidemics of cancer, liver failure, epilepsy, autism, and basically every bad thing that you can possibly imagine. There is no actual evidence for any of those assertions, and plenty of evidence against them by this point. Annoyingly and predictably, you almost never hear such claims made about stevia or monkfruit, since those are benevolent All Natural substances harvested under blue skies to a background of chirping birds and gentle breezes. Not, y’know, decanted from a fuming vat by faceless minions wearing protective gear, like those other toxic chemicals.

That doesn’t mean that there can’t be bad effects, though. You can discount the mowing-us-down-with-cancer-and-autism stuff, because effect sizes like that are actually pretty easy to pick up on, but looking for more subtle effects is certainly worth the effort, considering how much the consumption of these sweeteners has increased over the years. The belief has been that most of these are either barely absorbed or metabolized (sucralose) or broken down into harmless substances (aspartame, unless you have PKU), but it’s still worth checking on possible effects.

And that’s what this recent paper has done, with some rather counterintuitive findings. The authors conducted a randomized controlled trial in 120 patients in Israel, and it wasn’t easy. They were looking to enroll people who did not use such sweeteners at all, and they lost about 90% of their potential patients as they screened folks out. The participants (20 per group) were given sweetener packets of aspartame, sucralose, saccharin, or stevia, each bulked out with glucose to an equivalent size, with another group that got just glucose and another group that took no sweeteners at all. Participants took two packets of sweetener three times a day for fourteen days during the dosing period, and there was a one-week baseline evaluation beforehand and a further week after the dosing. Everyone wore a 24-hour glucose monitor, logged all food intake, and oral/stool/blood samples were collected at intervals as well as regular glucose tolerance tests.

It’s a pretty thorough look, and what happened was surprising. Exposure to saccharin and sucralose significantly impaired glycemic response, but this was not seen with the aspartame or stevia groups. None of the blood markers show real changes in any group except for insulin levels going up in the glucose and stevia groups (and since everyone was getting glucose as part of the dosing, that suggests a lowering of the glucose-driven insulin response overall). And what all of the sweetener patients showed, though, were significant changes in the gut microbiome, and these correlated with the glycemic response changes in the saccharin and sucralose groups. These changes were apparent at both the species distribution level and at the metabolome level. What’s more, transplantation of samples from the top and bottom responders in these groups into germ-free mice caused them to closely track the glycemic responses seen in the human patients. The authors:

Collectively, our study suggests that commonly consumed NNS may not be physiologically inert in humans as previously contemplated, with some of their effects mediated indirectly through impacts exerted on distinct configurations of the human microbiome.

They make sure to note that they’re not calling for consumption of sugar instead, because excess sugar is absolutely, positively linked to adverse health effects. But attempting to replace it with artificial sweeteners may not be a good way to go, either - and it’s worth remembering that while saccharin and sucralose showed the real effects in glucose tolerance, all four of them significantly altered the gut flora, and in ways whose further effects are difficult to predict. Literal food for thought. . .

https://www.science.org/content/blog-post/sugar-substitutes-surprise