Full mechanism, in text

How sweeteners change gut bacteria

1 · Baseline glucose tolerance

Before any sweetener, the trial measured glucose tolerance, microbiome and blood metabolites in every participant: tolerance was normal, but microbiomes differed from person to person.Starting point (Suez 2014 + 2022):

· Suez 2022 (Cell): a in 120 healthy adults, supplemented for 2 weeks
· Baseline measured first: a glucose tolerance test (how fast blood glucose comes back down after a dose of sugar) + stool and oral microbiome + blood metabolites

Baseline state:

· Glucose curve: normal glucose tolerance (2-hour glucose < 7.8 mmol/L)
· Microbiome: different in every person, which is where the later individual differences come from

Key background: the zero-glycemic-index reasoning:

· Classic nutrition reasoned: no sugar, so no effect on blood glucose
· The hidden premise is that sugar is the only direct route to blood glucose
· The two Suez studies challenge that premise — sweeteners can affect blood glucose in some people indirectly, through the microbiome
· This is not a defence of sugar — sugar's own problems are unchanged

The two studies differ in scale:

· 2014: mouse experiments + a small trial in 7 people for about a week
· 2022: a randomized trial in 120 people over 2 weeks — milder effects, and they vary from person to person

2 · 2-wk NNS exposure

120 healthy adults were randomized into 6 groups, 4 taking different sweeteners and 2 as controls, all at doses below the ADI, for 2 weeks.Suez 2022 Cell · a in 120 healthy adults:

4 sweetener groups + 2 control groups = 6 groups (20 people each):

· Saccharin
· Sucralose
· Aspartame
· Stevia
· Glucose-vehicle control: only the glucose carrier, in the same small sachet
· No-supplement control: nothing added

Supplemented for 2 weeks, all doses below the ADI (acceptable daily intake, the amount regulators consider safe every day for life) — not astronomical doses.

The 2014 groundwork (Suez 2014):

· Mice drinking commercial sucralose, aspartame or saccharin formulations → changed gut microbiome and worse glucose tolerance
· Wiping out the microbiome with antibiotics → the effect disappeared
· Transferring the microbiome of people whose glucose response worsened into germ-free mice → the mice also developed worse glucose tolerance — support, in mice, that the microbiome is the middle link

Why the microbiome might respond to sweeteners (hypotheses):

· Sweeteners the body does not absorb reach gut bacteria directly
· Zero calories for you can be something some bacteria use or have to deal with
· Sucralose carries chlorine atoms: one idea is that this makes its disturbance of the microbiome larger — not yet directly tested

Important detail: individual variation is large. At the same dose some people get clearly worse and others barely change — the lead-in to the later responders vs non-responders story.

3 · Microbiome shift + germ-free mice

All four sweeteners changed the microbiome, saccharin and sucralose worsened glucose responses at group level, and germ-free mice given those microbiomes largely reproduced the responses of their donors.Key findings of Suez 2022:

· Each of the four sweeteners changed the stool and oral microbiome, and blood metabolites, in its own way
· Saccharin and sucralose groups: at group level, glucose responses got significantly worse
· Aspartame and stevia groups: the microbiome changed too, but no worsening of blood glucose was seen at group level

Proposed mechanism (the second half is mainly inference):

· In the 2014 mouse experiments, some Bacteroidales rose and some Clostridiales fell
· Once the microbiome shifts, short-chain fatty acid () and bile acid metabolism shift with it, which could affect insulin sensitivity and glucose responses
· The second half of this chain has not yet been measured step by step in people

The full chain: sweetener → colon microbiome composition shifts → SCFA + bile acids shift (inferred) → insulin sensitivity + gut hormone signals (inferred) → altered post-meal glucose response.

Transferring the microbiome into germ-free mice:

· Suez 2022: the researchers moved the microbiomes of the strongest and weakest responders into germ-free mice, and the mice largely reproduced their donors' glucose responses
· Why this step matters most: seeing that people whose glucose worsened also have different microbiomes allows arrows in both directions; germ-free mice receive only the microbiome, so if the effect moves house with it, the arrow points from microbiome to glucose
· Limits: this bridge stands in mice; the human trial lasted only 2 weeks and was not large

WHO 2023 guideline:

· Based on a systematic review of 283 studies
· Does not recommend non-sugar sweeteners for weight control or for lowering the risk of noncommunicable disease
· Bans no sweetener — it simply does not recommend them for these two aims
· Three qualifiers that must stay attached: WHO rates it a conditional recommendation, not a strong one · it applies to everyone except people with diagnosed diabetes · it does not cover polyols — erythritol and xylitol are outside its scope

4 · Repeat glucose test

Repeat testing showed that sweeteners affect blood glucose in some people, indirectly through the microbiome, with large individual variation, so zero does not mean glucose-neutral.The 2014 small human trial (N = 7 healthy people who did not normally use sweeteners, saccharin for about a week):

· 4 of 7: glucose tolerance got worse, and their microbiome changed with it
· The other 3: almost no change
· The trial was very small and had no control group

Repeat testing in the 2022 randomized trial:

· Saccharin + sucralose groups: glucose responses significantly worse at group level
· Aspartame + stevia groups: no such change at group level
· Individual variation is large: at the same dose, some got clearly worse and some barely changed

Conclusions:

· Zero ≠ glucose-neutral — through the indirect microbiome route, some people's blood glucose is affected
· The transfer experiments support, in mice, that the microbiome is a cause and not merely a correlate
· Who is affected, and by how much, cannot yet be said

Sugar vs sweeteners:

· The evidence of harm from eating a lot of sugar is far larger and more certain
· Sweeteners: may affect blood glucose in some people indirectly through the microbiome; the evidence is still building
· Swapping sugary drinks for water, tea or black coffee: sidesteps both problems

In practice:

· A zero-sugar label ≠ healthy — do not treat it as permission to drink freely
· In Suez 2022 all four sweeteners changed the microbiome; none was shown to leave it untouched
· Erythritol and xylitol: observational studies show separate cardiovascular signals (Witkowski 2023 / 2024); causation is not established

Special groups:

· Children: there is little safety data for children; the WHO recommendation covers everyone except people with diagnosed diabetes, not only children
· Pregnancy: in the Canadian CHILD birth cohort (Azad 2016), mothers who drank artificially sweetened drinks daily in pregnancy had babies with higher at age 1 — an association that does not show the sweeteners caused it
· Diagnosed diabetes: outside the scope of this WHO recommendation; discuss use with a doctor

Full mechanism, in text

1 · Baseline glucose tolerance

2 · 2-wk NNS exposure

3 · Microbiome shift + germ-free mice

4 · Repeat glucose test

Suez RCT timeline — sweeteners shift the microbiome

Full mechanism, in text

1 · Baseline glucose tolerance

2 · 2-wk NNS exposure

3 · Microbiome shift + germ-free mice

4 · Repeat glucose test

Suez RCT timeline — sweeteners shift the microbiome