Feature articles

Non-nutritive sweeteners and the microbiome

Gut health is proving to be a new frontier in health and a popular area for researchers. The more we look, the more we find about how our gut microbiota affects our health, including our mental health and mood (see Issue 33, April 2018, diet and mental health). Non-nutritive sweeteners (also known as artificial sweeteners) have also come under increasing scrutiny because of associations with metabolic diseases including diabetes and weight gain (see Issue 31, Feb 2018, do non-nutritive sweeteners help weight loss?). Emerging evidence suggests non-nutritive sweeteners can adversely affect gut bacteria, and this may be the mechanism through which they exert metabolic effects.

Image 1 SB 40.jpg

Terms defined

The terms microbiota and microbiome are often used interchangeably.

Gut microbiota: the trillion of cells including bacteria, viruses and fungi that live in our gut. Every person carries around about 1.8kg of biomass in their gut, and we all have a unique mix of species

Gut microbiome: the microbial cells and their genetic material that live in our gut

Aren’t non-nutritive sweeteners inert?

While most NNS are not metabolised by the human body and considered inert, they do affect the billions of bacteria we host in our gut. We know that our microbiome can be rapidly altered by diet, and some microbiome compositions may play a causal role in weight gain.

Research so far: In-vitro

The most recently published research used three strains of E.coli to represent gut bacteria and exposed them to six FDA-approved non-nutritive sweeteners (NNS): aspartame, sucralose, saccharine, neotame, advantame and acesulfame potassium (ace-k) (also approved in Australia and New Zealand and the EU). The study used bacteria engineered to show luminescence when exposed to toxicants.

Results showed toxicity was induced by all sweeteners to varying degrees, and the lowest response was for sucralose and the highest for saccharin. There was also variability in the type of toxicity induced by the different NNS. The authors concluded these findings correlate with previous toxicity studies.

Another in-vitro study conducted in Sydney Australia using E.coli and the bacterial phyla Bacteroidetes and Firmacutes found ace-k, saccharin and sucralose as well as rebaudioside A in stevia all exerted strong bacteriostatic effects. This study also found these NNS affected the intestinal microbiota in mice.

Animal studies

Reverse causality is a very real challenge in NNS research because so many people already experiencing metabolic disease are using NNS. Because of this, research has focussed on animal models.

There are numerous studies in mice showing adverse effects of NNS on the gut microbiome. One study found ace-k perturbed gut microbiome as measured by fecal metabolic profiles after a 4-week treatment, and increased body weight in males (but not females). The same research group also did another study in the same mouse model and showed adverse effects from neotame. Specifically, neotame reduced alpha-diversity (number of taxa) and altered beta-diversity (difference in taxanomic abundance from different samples) of the gut microbiome. Neotame affected the balance between Bacteroidetes and firmicutes phyla, which is implicated in gut health. It also changed functional genes and the metabolome.

Another study examined the effects of sucralose and ace-k in lower concentrations than typically administered to mice; no more than the maximum acceptable daily intake (ADI), or 15mg/kg body weight. Results showed sucralose but not ace-k affected the relative amount of clostridium cluster XIVa (a producer of beneficial butyrate) in the fecal microbiome and also affected cholesterol bile acid metabolism. To add further weight to a causal effect, a study that gave saccharin to mice found it caused glucose intolerance, which was then ameliorated with antibiotics. When the gut bacteria from the glucose intolerant mice given saccharin were transferred to germ-free mice, they too developed glucose intolerance.

These studies suggest NNS do change the microbiome in animal models. However, when it comes to demonstrating metabolic disturbances from NNS in animal models, results have been mixed, with different methodologies and models producing conflicting results. This suggests there is more work to be done to understand the effects of microbiome structure and function on metabolic health.

Human studies

Research into the effects of NNS on the human gut microbiome and resulting metabolic effects is in its infancy but suggests there is cause for concern. For example, Suez et al found many common NNS induce changes in the intestinal microbiota and glucose intolerance in healthy subjects. 

In an analysis of NNS consumption, microbiome composition and metabolic outcomes in 381 subjects, Suez and colleaguesfound NNS was associated with BMI, blood pressure, HbA1C and fasting glucose levels, but also the presence of certain taxa: Actinobacteria phylum, the Enterobacteriales order and various taxa from the Clostridiales order. They then tested causation in a small intervention (7 people) with the upper daily limit of saccharin and found elevated glycemic responses in 4 out of 7. The 4 responders had microbiome alterations, which when transplanted to germ-free mice created the same glycemic responses.

Perhaps most interesting of all was the finding that responders and non-responders shared distinct differences in gut microbiome composition even before the intervention, suggesting a personalised susceptibility and response to NNS.

In their review paper, Suez et al says only a few prospective intervention studies on the metabolic effects of NNS have been conducted in humans, presumably due to the difficulty of finding subjects who have not been exposed to NNS and the difficulty of overcoming confounding factors such as genetics and lifestyle.

Practical implications

Foods with NNS have been advocated for people who are overweight or who have diabetes as a way of limiting kilojoule intake and managing blood glucose levels. Non-nutritive sweeteners are increasingly being added to foods and promoted as healthier because they have less added sugar. While the amounts of NNS permitted are small and within currently accepted safety limits, the evidence so far suggests caution with their use because of adverse effects of NNS on the microbiome.

Individuals’ microbiomes may vary in their response to NNS, but we don’t yet know how to identify those who are more susceptible to adverse effects. While NNS may not exert metabolic effects directly on us, they appear to exert effects on our microbiome which then in turn exert effects on us as their host. When it comes to the bacteria in our gut, it may be a case of ‘biting the hand that feeds them’ when we eat things they don’t like.

NEXT:What are ultra-processed foods?




Portion control, sugars intakes and more 


Sugar and health

How much sugar are we recommended to eat?


Frequently asked questions

Natural versus added sugars - what's the difference?