FACTS BEHIND THE HEADLINES
Could low-calorie sweeteners be contributing
to the diabetes epidemic?
R. Miller
British Nutrition Foundation, London, UK
Low-calorie sweeteners are often recommended to help
lower total calorie intake or to help manage blood
glucose levels in diabetics and those with glucose intolerance.
Furthermore, low-calorie sweeteners are commonly
used and consumed as tabletop sweeteners and as
replacements for added sugars in manufactured products,
helping towards producing alternative products
and improving choice for the consumer. The safety of
low-calorie sweeteners has been intensely investigated
and reviewed over the past few decades and recent
reevaluations of pre-approved low-calorie sweeteners
have deemed their consumption, within speciﬁed intake
levels, to be safe. However, recent research by Suez et al.
(2014) entitled ‘Artiﬁcial sweeteners induce glucose
intolerance by altering the gut microbiota’ has again
ignited interest into the safety of low-calorie sweeteners.
The research prompted media headlines including ‘Artiﬁcial
food sweeteners linked to diabetes’ (Daily Express
2014), ‘Sweeteners could cause obesity scientists warn’
(The Telegraph 2014) and ‘Study raises doubts over
sweeteners’ (The Times 2014). This is likely to have
caused confusion among low-calorie sweetener consumers,
especially as there are concurrent health messages to
reduce the intake of ‘added sugars’.
The research behind these headlines was predominately
carried out in mice, starting with experiments that
supplemented the rodents’ drinking water with commercially
available low-calorie sweeteners, namely saccharin,
sucralose or aspartame (Suez et al. 2014). After 11
weeks, the glucose tolerance (a marker of diabetes risk) of
the mice consuming water supplemented with commercially
available low-calorie sweeteners (10% solution)
was impaired compared with mice consuming the control
drinks of either water or water supplemented with
glucose or sucrose. Saccharin demonstrated the largest
effect on glucose tolerance and, therefore, was subjected
to further investigation. Accordingly, commercial
saccharin and pure saccharin, at an estimated precalculated
dose of 5 mg/kg bodyweight per day for 11
weeks, were found to impair glucose tolerance in both
lean and obese mice, compared with mice consuming
glucose or water control drinks. The dose of pure saccharin
was equivalent to the Food and Drug Administration
(FDA) maximal acceptable daily intake (ADI) for
humans (adjusted to the bodyweights of the mice).
However, it is worth noting that this dose may well have
a different effect in mice compared with humans because
of the biological differences among species. The authors
suggested that the observed effect caused by the saccharin
may be mediated by changes in the gut microbiota of the
mice because antibiotic treatment, which decreases the
density of intestinal bacteria, resulted in glucose tolerance
proﬁles that were comparable with the control
groups. Furthermore, transferring the gut microﬂora
from mice consuming the commercial saccharin to
germ-free mice (bred in aseptic conditions to prevent
gut microﬂora development) led to impaired glucose
tolerance, compared with germ-free mice receiving
microbiota from control mice consuming glucose.
Sequencing of the faecal microﬂora 16S ribosomal RNA
(rRNA), which indicates the microbiota composition,
showed possible differences in the relative abundance
of some bacterial taxonomic groups between mice consuming
saccharin and those consuming the controls.
Therefore, the authors suggest the saccharin may be
modulating the microbe taxonomic entities residing in
the guts of the mice. In addition, metagenomic sequencing
and mapping of faecal samples collected from the
mice consuming commercial saccharin suggests the
enhancement of pathways involved in glycan degradation,
compared with control mice consuming glucose.
Suez and colleagues postulate that the enhancement of
these pathways may lead to increased energy harvest
and de novo glucose and lipid synthesis in the host.
Pathways involved in the metabolism of starch, sucrose,
fructose and mannose, and the biosynthesis of folate,
glycerolipids and fatty acids were also enhanced in the
microbiomes of the mice consuming commercial saccharin,
compared with control mice consuming glucose.
Similar pathways have been shown to be enhanced in the
Correspondence: Dr Ros Miller, Nutrition Scientist, British
Nutrition Foundation, Imperial House, 15-19 Kingsway, London
WC2B 6UN, UK.
E-mail: r.miller@nutrition.org.uk
bs_bs_banner
DOI: 10.1111/nbu.12129
© 2015 British Nutrition Foundation Nutrition Bulletin, 40, 33–35 33
gut microbiota of type 2 diabetics and overweight individuals
but this does not demonstrate causation. Therefore,
the mechanism behind the observed impairment
of glucose tolerance in the mice consuming saccharin
requires further exploration. This seemingly expansive
rodent study, which has its limitations with regard to the
conclusions that can be drawn, was then followed by
pilot investigations in humans.
A relatively small cross-sectional study that examined
the relationship between long-term low-calorie sweetener
consumption (assessed by a food frequency questionnaire)
and clinical parameters related to metabolic
syndrome in 381 non-diabetic participants, found signiﬁcant
positive correlations between low-calorie sweetener
consumption and measures of bodyweight and glucose
tolerance. However, this ﬁnding may be a result of reverse
causation, as individuals may consume low-calorie
sweeteners to help reduce bodyweight and/or risk of type
2 diabetes. The gut microﬂora 16S rRNA was characterised
in 172 participants and signiﬁcant positive correlations
between taxonomic entities (including the
Enterobacteriaceae family, the Deltaproteobacteria class
and the Actinobacteria phylum) and low-calorie sweetener
consumption were found. However, it is unclear
whether this was due to the consumption of low-calorie
sweeteners or another dietary or lifestyle factor associated
with their consumption. To follow this up, seven
healthy participants who did not normally consume lowcalorie
sweeteners were given commercial saccharin at
the FDA ADI (5 mg/kg bodyweight) for 5 days. This
underpowered and uncontrolled study showed that 4 out
7 participants had a reduction in glucose tolerance at the
end of the treatment period compared with baseline and
as such were termed ‘responders’ by the investigators.
Bodyweight change was not investigated in this shortterm
study. Among these ‘responders’ the faecal
microﬂora composition was found to be altered after
consumption of saccharin compared with baseline and
appeared to cluster differently from the ‘non-responders’
both before and after the consumption of saccharin. In
addition, transfer of faecal samples, collected from the
‘responders’ post-saccharin consumption, to germ-free
mice resulted in impaired glucose tolerance in the mice
compared with mice who received faecal samples collected
from the ‘responders’ before saccharin consumption.
The authors concluded that the results of these
experiments suggest that saccharin, at the levels consumed
(maximal ADI), reduced glucose tolerance
through modulation of the gut microbiota. However,
these studies were not without their limitations, which
reduce the transferability of the results to the general
population.
Unfortunately, the study authors and resulting media
coverage have generalised these ﬁndings to all lowcalorie
sweeteners. The initial rodent experiment was
the only experiment to include the three low-calorie
sweeteners (saccharin, sucralose and aspartame). Within
this experiment, the data for the groups of mice receiving
each low-calorie sweetener were combined together
for statistical comparison with the data from the combined
control groups (water, sucrose and glucose) and a
signiﬁcantly lower glucose tolerance in the combined
low-calorie sweetener group was found. However, it
appears the difference in glucose tolerance was primarily
driven by saccharin, with negligible differences
being found for one of the most commonly consumed
low-calorie sweeteners in the UK, aspartame. Furthermore,
the chemistry and metabolism of different lowcalorie
sweeteners can be quite diverse, with the only
common feature being the provision of sweet taste. For
example, the low-calorie sweetener aspartame is completely
metabolised into amino acids and methanol,
which are absorbed in the small intestine and so does
not reach the large intestine (Butchko et al. 2002),
whereas although sucralose does reach the large intestine,
studies have shown that it cannot be metabolised
by the gut microﬂora (Roberts et al. 2000). Therefore,
generalising the ﬁndings of this study to include all
low-calorie sweeteners is highly inappropriate. In
addition, translating the changes in mice or human
microbiota to health effects in humans is extremely
complex. The consequence of the microbiota changes
observed within this study to human health and whether
these changes are sustained over medium- or long-term
saccharin exposure remains unknown.
Moreover, although animal studies can prove useful
in generating theories or designing experiments that
cannot be performed in humans, the different biological
processes in animals compared with humans means that
the results obtained cannot be generalised to humans.
Furthermore, even though the experimental human
study was set out to help address this, it was severely
underpowered, uncontrolled and of short duration. The
cross-sectional human study was also relatively small,
involving only 381 participants. Indeed, a recent large
case cohort study, involving more than a quarter of a
million people, showed no association between consumption
of low-calorie sweeteners and incidence of
type 2 diabetes (Romaguera et al. 2013). Moreover, as
the experimental human study that involved only seven
participants did not have a control group, a speciﬁc
cause and effect relationship could not be established. In
addition, separating ‘responders’ from ‘non-responders’
is likely to have occurred during post hoc analysis, and
34 R. Miller
© 2015 British Nutrition Foundation Nutrition Bulletin, 40, 33–35
the lack of reporting from the whole group analysis can
only lead the reader to assume that no signiﬁcant results
were found. Furthermore, even within the ‘responders’
group, huge inter-individual variation in glycaemic
responses was evident, meaning that other factors were
likely to have been involved. Indeed, other dietary or
lifestyle factors were not considered during the analysis
of the study results and only a small part of the observational
study, in which the relationship between
glycosylated haemoglobin and the consumption of lowcalorie
sweeteners was analysed, corrected for BMI. In
addition, the saccharin dose given in both the rodent
and experimental human study equated to the FDA ADI
(equivalent to 250 individual portion packets of tabletop
sweetener for the average adult), a level which is
highly unlikely to be consumed, even for those regarded
as high consumers of saccharin (FDA 2014). Moreover,
blends of saccharin with other low-calorie sweeteners
are most often used within the food and drink industry,
to help deliver the best ﬂavour proﬁle. Therefore, investigations
into low-calorie sweeteners, at the doses and
blends commonly consumed, would be more applicable
to the general population.
The safety of low-calorie sweeteners has been extensively
scrutinised over the past couple of decades.
Several reevaluations have been performed by risk
assessment authorities including the FDA in the US and
the European Food Safety Authority (EFSA), which
have deemed sucralose, saccharin and aspartame to be
of no safety concern at intakes below the speciﬁed ADI
levels (Scientiﬁc Committee on Food 2000; EFSA 2013;
FDA 2014). Furthermore, there is a strong body of
evidence contradicting the ﬁndings of the study by Suez
et al. A recent meta-analysis of 15 randomised controlled
trials that investigated the substitution of added
sugars with low-calorie sweeteners found beneﬁcial
effects on bodyweight (Miller & Perez 2014), which
contradicts what was observed in the small crosssectional
human study. A review of the existing evidence
investigating low-calorie sweeteners and metabolic disorders
also found beneﬁcial effects and concluded that
the replacement of ‘added sugars’ with low-calorie
sweeteners (which is associated with energy reduction)
may help to reduce the risk of type 2 diabetes (Raben &
Richelsen 2012), again contradicting what was suggested
in both the animal and human studies.
On average, the UK population currently exceeds
the recommended maximum intake level for ‘added
sugars’ and for some individuals, substituting ‘added
sugars’ with low-calorie sweeteners may help towards
meeting these recommendations. Furthermore, lowcalorie
sweeteners may help to reduce calorie intake and
so reduce the risk of obesity and type 2 diabetes. In
conclusion, although emerging evidence from rodent
studies is of interest, the current study does not provide
convincing evidence that low-calorie sweeteners can
impair glucose tolerance or increase the risk of obesity in
humans. Larger randomised controlled human trials
using realistic doses are required before any conclusions
can be made.
Conﬂict of interest
The author has no conﬂict of interest to disclose.
References
Butchko HH, Stargel WW & Comer CP (2002) Aspartame: review
of safety. Regulatory Toxicology and Pharmacology 35: S1–93.
Daily Express (2014) Artiﬁcial food sweeteners linked to diabetes.
By J Willey, 17 September 2014. Available at: http://
www.express.co.uk/life-style/health/512131/food-sweetenersdiabetes-link
(accessed 8 December 2014).
EFSA (European Food Safety Authority) (2013) Scientiﬁc opinion
on the re-evaluation of aspartame (E 951) as a food additive.
EFSA Journal 11: 3496.
FDA (Food and Drug Administration) (2014) Additional information
about high-intensity sweeteners permitted for use in food in
the United States. Available at: http://www.fda.gov/Food/
IngredientsPackagingLabeling/FoodAdditivesIngredients/
ucm397725.htm (accessed 8 December 2014).
Miller PE & Perez V (2014) Low-calorie sweeteners and body
weight and composition: a meta-analysis of randomized controlled
trials and prospective cohort studies. American Journal of
Clinical Nutrition 100: 765–77.
Raben A & Richelsen B (2012) Artiﬁcial sweeteners: a place in the
ﬁeld of functional foods? Focus on obesity and related metabolic
disorders. Current Opinion in Clinical Nutrition and Metabolic
Care 15: 597–604.
Roberts A, Renwick AG, Sims J et al. (2000) Sucralose metabolism
and pharmacokinetics in man. Food Chemistry and Toxicology
38: S31–41.
Romaguera D, Norat T, Wark PA et al. (2013) Consumption of
sweet beverages and type 2 diabetes incidence in European adults:
results from EPIC-InterAct. Diabetologia 56: 1520–30.
Scientiﬁc Committee on Food (2000) Opinion of the scientiﬁc committee
on food on sucralose. European Commission. SCF/CS/
ADDS/EDUL/190 Final.
Suez J, Korem T, Zeevi D et al. (2014) Artiﬁcial sweeteners induce
glucose intolerance by altering the gut microbiota. Nature 514:
181–6.
The Telegraph (2014) Sweeteners could cause obesity scientists
warn. By Laura Donnelly, 17 September 2014. Available at:
http://www.telegraph.co.uk/health/healthnews/11101477/
Sweeteners-could-cause-obesity-scientists-warn.html (accessed 8
December 2014).
The Times (2014) Study raises doubts over sweeteners. By Hannah
Devlin, 18 September 2014. Available at: http://
www.thetimes.co.uk/tto/science/article4209668.ece (accessed 8
December 2014).
Low-calorie sweeteners 35
© 2015 British Nutrition Foundation Nutrition Bulletin, 40, 33–35