Food and Chemical Toxicology 106 (2017) 324e355
Contents lists available at ScienceDirect
Food and Chemical Toxicology
journal homepage: www.elsevier.com/locate/foodchemtox
Review
Critical review of the current literature on the safety of sucralose
Bernadene A. Magnuson a, *, Ashley Roberts b, Earle R. Nestmann c
a
Health Science Consultants, Inc, 7105 Branigan Gate, Unit 68, Mississauga, ON, L5N7S2, Canada
Intertek Scientific & Regulatory Consultancy, 2233 Argentia Road, Suite 201, Mississauga, Ontario, L5N 2X7, Canada
c
Health Science Consultants, Inc, 7105 Branigan Gate, Unit 68, Mississauga, ON, L5N 2X7, Canada
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 March 2017
Received in revised form
6 May 2017
Accepted 22 May 2017
Available online 27 May 2017
Sucralose is a non-caloric high intensity sweetener that is approved globally for use in foods and beverages. This review provides an updated summary of the literature addressing the safety of use of
sucralose. Studies reviewed include chemical characterization and stability, toxicokinetics in animals and
humans, assessment of genotoxicity, and animal and human feeding studies. Endpoints evaluated
include effects on growth, development, reproduction, neurotoxicity, immunotoxicity, carcinogenicity
and overall health status. Human clinical studies investigated potential effects of repeated consumption
in individuals with diabetes. Recent studies on the safety of sucralose focused on carcinogenic potential
and the effect of sucralose on the gut microflora are reviewed. Following the discovery of sweet taste
receptors in the gut and studies investigating the activation of these receptors by sucralose lead to
numerous human clinical studies assessing the effect of sucralose on overall glycemic control. Estimated
daily intakes of sucralose in different population subgroups, including recent studies on children with
special dietary needs, consistently find that the intakes of sucralose in all members of the population
remain well below the acceptable daily intake. Collectively, critical review of the extensive database of
research demonstrates that sucralose is safe for its intended use as a non-caloric sugar alternative.
© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords:
Sucralose
Safety
Glycemic control
Consumption
Low-calorie sweetener
Contents
1.
2.
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
1.1.
History, regulatory status and health agency positions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
1.2.
Chemistry and properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
1.3.
Sensory properties and nutritional value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
1.4.
Stability of sucralose in food products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
Safety assessment of sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
2.1.
Absorption, distribution, metabolism and excretion (ADME) of sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
2.1.1.
Human ADME studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
2.1.2.
Animal ADME studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
2.1.3.
In vitro metabolism studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
2.1.4.
Summary of ADME of sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
2.2.
Genotoxic potential of sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
2.3.
Animal toxicology studies on sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
2.3.1.
Acute studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
2.3.2.
Sub-acute and sub-chronic toxicity studies on sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
2.3.3.
Other short-term research studies on sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
2.3.4.
Reproduction and development studies on sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
2.3.5.
Long-term chronic toxicity and carcinogenicity of sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
2.3.6.
Specialized animal studies on sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
* Corresponding author.
E-mail addresses: berna@bernamagnuson.com (B.A. Magnuson),
roberts@intertek.com (A. Roberts), ern@enestmann.com (E.R. Nestmann).
ashley.
http://dx.doi.org/10.1016/j.fct.2017.05.047
0278-6915/© 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
325
2.4.
2.5.
2.6.
3.
4.
In vitro studies other than genotoxicity studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
Animal studies on the safety of hydrolysis products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
Human clinical studies on sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
2.6.1.
Repeated daily consumption of sucralose in human subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
2.6.2.
Acute or single administration of sucralose in human subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
2.7.
Case reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Estimates of consumption of sucralose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Summary and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350
Financial support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
Transparency document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
Sucralose is a non-caloric sweetener that is widely approved
globally for use in foods and beverages. It is derived from sucrose by
the selective replacement of three hydroxyl groups by chlorine
atoms. Sucralose has a sweetness potency of about 600 times that
of sucrose, thus the addition of very small amounts of sucralose can
be used to sweeten foods and beverages. Unlike sucrose, sucralose
is not digested or metabolized for energy, therefore, no calories are
obtained from sucralose, and sucralose does not affect blood
glucose levels. These properties result in the use of sucralose to
produce foods and beverages that are suitable for persons with
diabetes or those aiming to reduce calorie or carbohydrate intake.
Although several reviews have been published previously (Grice
and Goldsmith, 2000: Grotz and Munro, 2009; JECFA, 1989a,
1991a; SCF, 1989, 2000a), the purposes of this review are (1) to
provide an updated summary of the research investigating the
safety of sucralose in one publication including studies that have
been the genesis of new questions on sucralose safety, and (2) to
provide background on the regulatory process of testing and
approval of food additives for health professionals. Numerous
clinical investigations into the effect of sucralose on glycemic responses are a particular focus, following the discovery of gut sweet
taste receptors and academic studies investigating the potential
role that activation of these has on overall glycemic control. Also
reviewed are several recent studies that report on estimated daily
sucralose intakes in different population subgroups, including
children. Collectively, the data continue to demonstrate that
sucralose is safe for its intended use as a non-caloric sugar
alternative.
1. Background
1.1. History, regulatory status and health agency positions
The discovery and development of sucralose was the result of a
collaborative research project of the Tate & Lyle Company and the
Queen Elizabeth College of the University of London during the late
1980s (Knight, 1994). Extensive chemical characterization and
toxicology studies were undertaken as required for premarket
regulatory investigation into the safety of a proposed new food
additive.
The general principles for the premarket safety assessment of
new food additives, such as a non-caloric sweetener were first
established by the Joint FAO/WHO Expert Committee on Food Additives (JECFA) in 1958. Although these principles remain to be the
basis for approvals by regulatory agencies globally, recent revisions
incorporate current knowledge and advances in toxicological science (reviewed Magnuson et al., 2013). A detailed description of the
studies required is publically available (WHO, 2009) and is similar
to those required by other regulatory agencies including the Redbook by the US Food and Drug Administration (FDA), and the
Organisation for Economic Co-operation and Development (OECD)
(OECD, 1998; US FDA, 2000). These guidelines establish both the
types of studies to be conducted and the appropriate study protocols to be used in investigating the safety of a new food ingredient.
Regulatory
agencies
require
complete
chemical
characterization of the ingredient, studies that demonstrate its
intended functionality and stability in food, the method of manufacture, the detection method including data that validate the
analytical method development, and comprehensive toxicological
research. Toxicology testing requirements include evaluation of
genetic effects, pharmacokinetics and metabolism, toxicology
studies in rodents and in non-rodent species including life-time
exposures to ensure no evidence of adverse effects on growth
and development, organ function or structure, blood chemistry,
and/or potential to cause cancer. Multigenerational studies assess
possible effects on male or female reproduction, pregnancy, and
offspring health and development. In addition, clinical studies are
often conducted to compare the pharmacokinetics (absorption,
distribution, metabolism and excretion) determined in experimental animals to data from humans, to demonstrate the appropriateness of the animal models used in safety testing. All data from
the investigative studies must be submitted to the regulatory
agencies. In the U.S., research studies submitted to FDA are available
through a Freedom of Information Act request, and JECFA evaluations of the submitted toxicology studies, which form the basis for
approvals in the EU and numerous other countries, are published
and publicly available online. Manufacturers of new food ingredients may also further move to enable publications that
describe the core research studies. Such is the case with sucralose
(FCT, 2000).
During the regulatory review process, study designs and data
are critically reviewed by expert scientists, in their respective fields,
to determine if there is sufficient evidence to establish the safe level
of the food additive that can be consumed by the entire population
on a daily basis, which is called the acceptable daily intake (ADI).
The ADI is based on the No Observed Adverse Effect Level (NOAEL),
which is the highest dose that was fed to animals in long-term
studies with no toxicological effects. The NOAEL is then divided
by a safety factor to ensure the resulting ADI is safe for all potential
consumers, including subgroups such as children.
The JECFA first approved sucralose in 1989, after reviewing
extensive studies, establishing a temporary ADI of 0e3.5 mg/kg bw/
d based on a NOAEL of 750 mg/kg bw/d in a one year study in dogs
and a 200 fold safety factor. At that time, the dog study was
considered the most appropriate of all the studies to use to establish the ADI. However, further studies were requested, including
assessment of safety of long-term consumption by individuals with
diabetes. In 1991, following the evaluation of data from additional
studies in both animals and humans, the Committee allocated a
permanent ADI of 0e15 mg/kg bw/d based on the NOAEL of
326
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
1500 mg/kg bw/d from a long-term study in rats, and a safety factor
of 100 (JECFA, 1991b). As the rat study included in utero exposure, as
well as exposure during postnatal growth and development, this
study was the most comprehensive study for establishing the ADI.
Other major food safety and health regulatory agencies around
the world, including Health Canada (2016), US FDA (1998, 1999,
2015), the European Union (SCF, 2000a), Australia, New Zealand,
Japan (JMOHW e JFCRF, 1999) and others (Grotz and Munro, 2009)
conducted intensive reviews of all the data generated in safety
investigations. All reviewing agencies have approved sucralose for
its intended use as a sweetener in a variety of products. Roberts
(2016) discussed the approval process of low calorie sweeteners
in the United States, while Rulis and Levitt (2009) provide an
excellent overview of the food additive safety assessment process
and describe the review of the approval of sucralose as a case study.
In addition, major health associations, such as the American
Heart Association (Gardner et al., 2012), the American Diabetes
Association (Gardner et al., 2012) and the Academy of Nutrition and
Dietetics (Fitch and Keim, 2012) have conducted detailed reviews of
the available literature on low calorie sweeteners including sucralose, and confirm that approved no calorie sweeteners can be safely
used in nutritional strategies for lowering sugar intake.
1.2. Chemistry and properties
Structurally, sucralose is a substituted disaccharide, similar to
sucrose with 3 chlorine atoms replacing 3 hydroxyl groups. The
chemical name of sucralose is 1,6-dichloro-1,6-dideoxy-ß-D-fructofuranosyl-4-chloro-4-deoxy-a-D-galactopyranoside, which has
also been described as 4,10,6'-trichlorogalactosucrose (TGS) (JECFA,
1991b). Descriptive names for sucralose can include trichlorosucrose, or substituted chlorinated disaccharide. Like sucrose, sucralose is a relatively small molecule (MW 400) and
polyhydroxylated. The presence of the multiple hydroxyl groups in
sucralose makes it, like sugar, a hydrophilic, rather than a lipophilic
compound. The considerable water solubility of sucralose, >25% at
22 C and relatively poor lipid solubility is expected. Sucralose has a
low octanol/water partition coefficient (Jenner and Smithson,
1989). Although sucralose contains chlorine and can be described
as a chlorinated carbohydrate, it is not in the class of substances
called chlorinated hydrocarbons. Sucralose has very different
chemical and physicochemical properties from those of chlorinated
hydrocarbons, as shown in the comparisons in Table 1. Key differences are the numerous exposed hydroxyl groups in sucralose that
are generally unreactive but form hydrogen bonds with water
resulting in the overall low reactivity, high hydrophilicity and low
fat solubility. In contrast, chlorinated hydrocarbons typically have
few or no hydroxyl groups and include the hallmark feature of
carbons linked by double bonds. The latter explains the high fat
solubility of these different substances and their reactivity and fate
in the body. Like sugar, sucralose has no carbon-carbon double
bonds. Thus, the suggestion that sucralose is an ‘organochlorine’ or
a substance in the class of chlorinated hydrocarbons (which notably
includes the commonly known substance DDT) is not appropriate.
Additional details on the chemistry and properties of sucralose are
also available (Jenner and Smithson, 1989; Grice and Goldsmith,
2000).
1.3. Sensory properties and nutritional value
Average sweetness potency of sucralose is estimated at about
600 times that of sugar, on a weight for weight basis. Thus, a very
small amount of sucralose (1/600th) can replace sugar on a weight
basis to achieve the same level of sweet taste. The sweetness of
sucralose is dependent on both its concentration in the final
product and on the properties of the food or beverage system in
which it is used (Wiet and Beyts, 1992; Knight, 1994). Sucralose, by
itself, has no nutritional value. All major no-low calorie sweetener
products formulated for consumer use (such as table-top sweetener packets), including those using sucralose, however, contain
some type of bulk carrier such as maltodextrin, to provide volume
so that the low calorie sweetener product can be used, more like
sugar volumetrically. These carriers or fillers can have some
nutritive value, but calories per serving must remain low. For
example, in the U.S., retail “no calorie” sweeteners must have 5 or
less calories per serving, which is also considered a trivial amount
(US FDA, 2016).
1.4. Stability of sucralose in food products
Regulatory agencies require testing of the stability of food additives during typical food processing and in the matrices of the
foods and beverages that will contain them. If the food additive is
found to be not stable during processing, the identification and
safety of any formed compounds must be assessed. Based on a
chemistry modeling approach, Pariza et al. (1998) illustrated that
the critical chemical and physical factors that affect the stability of a
food additive during processing are water content, pH, and temperature. The testing of the stability of the food additive in a few
food and beverage matrices e carbonated soft drinks, hot pack still
beverages, yogurt, yellow cake and powdered dry mixes - represent
the extremes of these conditions that are likely to be encountered
in all processed food applications for a low calorie sweetener
(Pariza et al., 1998).
Stability experiments conducted for sucralose added to
carbonated soft drinks, still drinks, dry mixes and strawberry milk
Table 1
Chemical and physiologic properties of chlorinated hydrocarbons (“organochlorines”) and sucralose.
Parameter
Chlorinated hydrocarbons
Sucralose
Reference
Lipophilicity/Hydrophilicity
Lipophilic
Hydrophilic
Gastrointestinal absorption
Most are efficiently absorbed
Poorly absorbed.
Distribution
Toxicity
Stored in adipose fat; prolonged
retention time
Dechlorinated, oxidized and
then conjugated
Primarily neurotoxic
Not accumulated in fat;
readily eliminated
No dechlorination,
minor conjugations
No neurotoxicity.
Tollefsen et al. (2012); Reigart and
Roberts (2013); Jenner and Smithson (1989)
Tollefsen et al. (2012); Reigart and
Roberts (2013); Tordoir and van Sittert (1994)
Ecobichon (1996); Reigart and Roberts (2013);
Tordoir and van Sittert (1994)
Reigart and Roberts (2013); Tordoir and van Sittert (1994)
Persistence in animals,
invertebrate and plant species
Wide spread evidence
of accumulation
Absence of accumulation
in animals, invertebrates
and plants
Metabolism
Ecobichon (1996); Reigart and Roberts (2013);
Tordoir and van Sittert (1994)
Tollefsen et al. (2012); Panseri et al. (2013)
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
illustrated extremely high stability even during extended storage
(see review Goldsmith and Merkel, 2001). For example, there was
no detectable loss of sucralose in a gelatin mix stored at 30 C for 6
months followed by 2 years storage at room temperature
(Goldsmith and Merkel, 2001) or in a cola beverage stored at 20 C
(Grotz et al., 2012).
Under acidic conditions, sucralose will very slowly hydrolyze to
its two constituent chlorinated monosaccharides, 1,6-dichloro-1,6dideoxy-D-fructose (1,6-DCF) and 4-chloro-4-deoxy-D-galactose (4CG) (see review Goldsmith and Merkel, 2001). There was no
breakdown of sucralose in solution stored for 1 year at 30 C when
the pH was 4.0, 6.0 or 7.5, but a 4% loss of sucralose occurred at pH
3.0. Extensive investigation into the fate of sucralose in the body
following consumption has demonstrated that these hydrolysis
products are not formed in the body (see below). However, there is
the potential for a small amount of them to be formed in acidic food
products during long term storage. For this reason, extensive
toxicology studies were conducted on these hydrolysis products as
well on sucralose, and reviewed prior to approval. These will be
discussed below in the appropriate toxicology sections.
The stability of sucralose during baking was tested in a
commonly-used yellow cake matrix, cookies and graham crackers
to ensure testing included a broad spectrum of different ingredients, water content, pH, and temperatures that are likely to be
encountered in various baked goods (Barndt and Jackson, 1990). In
this study, radioactive sucralose (14C sucralose) was used to accurately recover and detect the low levels of sucralose that would
commonly be used in these foods. Complete recovery of the
sucralose was achieved and thin layer chromatography designed to
separate any breakdown products from sucralose demonstrated
excellent stability, while the absence of any peaks other than
sucralose confirmed the lack of formation of breakdown products
(Barndt and Jackson, 1990). Based on the results of this and other
stability experiments, sucralose was determined to be suitable and
safe for use as a general purpose sweetener in heated beverages
and foods that require cooking, such as in baked goods (US FDA,
1999; 2015).
It should be noted that the sensitivity of dry pure sucralose to
high temperatures was also assessed during the initial stability
testing conducted for regulatory approval (Goldsmith and Merkel,
2001; Grotz et al., 2012). Heating solutions of pure sucralose or
pure sucralose plus glycine for 60 min at 180 C resulted in the
formation of several furan derivative volatile compounds and a
dramatic drop in pH to less than 2 (Hutchinson et al., 1999). Furans
are also formed from breakdown products of monosaccharides, and
are also found in many foods providing flavor notes (Hutchinson
et al., 1999).
Several recent investigations have reported the formation of low
levels of various chlorinated compounds and furans during high
temperature heating of dry pure sucralose (Bannach et al., 2009; de
Oliveira et al., 2015); pure sucralose in the presence of glycerol
(Rahn &Yaylayan, 2010) or metal oxides (Dong et al., 2013a) or high
concentrations of sucralose added to oil (Dong et al., 2013b) or oil
plus meat mixtures (Dong et al., 2011). Despite allegations by some
authors (Schiffman, 2012; Schiffman and Rother, 2013; de Oliveira
et al., 2015) that the formation of these compounds is a safety
concern for consumer use of sucralose, such allegations are not
appropriate. As noted by the authors (Dong et al., 2011, 2013a),
these conditions do not represent real-life uses of sucralose in
terms of the foods that sucralose is added to, the concentrations of
sucralose added to foods and beverages, or temperatures that are
typical for food processing, home cooking and baking, or storage.
In summary, the properties of sucralose contribute to its suitability as a non-caloric sweetener for a wide variety of products
including beverages, baked goods, breakfast cereals, desserts, jams
327
and jellies and more. These include high sweetening potency, nonreactivity in food systems, low viscosity, and no effect on food color
or surface tension (Goldsmith and Merkel, 2001). Sucralose is
highly stable in the ranges of temperatures and pH that are
encountered in food and beverage processing and, to date, no study
has demonstrated development of significant levels of either
breakdown or thermal byproducts in a matrix that compares to
food and beverage applications (Jenner and Smithson, 1989). These
functional properties, in combination with the totality of data on
safety and intended technical effects, led to the approval of sucralose as a general purpose sweetener, which can be used in any
application where the intended technical effect is to impart
sweetness (US FDA, 1999).
2. Safety assessment of sucralose
2.1. Absorption, distribution, metabolism and excretion (ADME) of
sucralose
An important aspect of the safety assessment of a potential new
food additive is consideration of whether the experimental animals
used in the toxicology studies represent a good model for the safety
assessment in humans. Comparison of the toxicokinetics (i.e. how
the test compound is absorbed, distributed, metabolized and
excreted) in different species to that in humans is used to identify
the experimental animal species that most closely represents
humans. Potential differences between males and females are also
evaluated. The toxicokinetics of sucralose were evaluated in mice
(John et al., 2000a), rats (Sims et al., 2000), dogs (Wood et al., 2000)
rabbits (John et al., 2000b) and humans (Roberts et al., 2000).
Numerous studies were conducted with varying doses of radiolabelled sucralose, and administration through oral, intravenous or
bile duct delivery to further understand the toxicokinetics of
sucralose. These studies are summarized in Tables 2 and 3. The
metabolic fate of sucralose and other low calorie sweeteners has
also recently been reviewed and compared (Magnuson et al., 2016).
2.1.1. Human ADME studies
To determine the fate of sucralose in humans, 8 healthy men
were administered 1 mg/kg [14C] sucralose in drinking water after
an overnight fast (Table 2) (Roberts et al., 2000). Two of these men
received 10 mg/kg in a second experiment. Radioactive sucralose
facilitated the detection of low levels of sucralose in blood, urine
and feces and determination of total recovery of administered
sucralose. After 5 days, 93% of the sucralose administered at 1 mg/
kg was recovered with fecal and urinary excretion representing 78%
and 14.5% of the total dose, respectively. The low absorption of
sucralose (ranging from 9 to 22%, mean of ~15%) was supported by
low peak plasma concentrations (Cmax). In subjects receiving
10 mg/kg, higher fecal and lower urinary excretion indicated lower
absorption at higher consumption levels (Roberts et al., 2000).
The structure of the excreted radiolabeled compounds was
determined to assess if and how sucralose was metabolized in the
human body. Thin layer and gas chromatography, in combination
with mass spectrometry analysis demonstrated that essentially all
(>99%) of the radioactivity found in feces was unchanged sucralose.
The major component of radioactivity found in urine was also unchanged sucralose, with two minor metabolites, representing ~2%
of the total dose. These metabolites were identified as glucuronide
conjugates of sucralose (Roberts et al., 2000). Thus, in humans, the
fate of sucralose is primarily no absorption and simple fecal
excretion with no evidence of digestion or breakdown of sucralose,
loss of chlorine or metabolism by fecal microflora. This is a critically
important point in light of recent allegations that sucralose may
affect gut microflora (Suez et al., 2015; Abou-Donia et al., 2008)
328
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 2
Summary of Sucralose ADME studies following oral administration.
Species
Route of Administration,
Test material
Dose mg/kg
N
Recovery of radioactivity
(% of administered dose)
Metabolites
Reference
Human, adult, (30
e48 yr)
Oral, drank sucralose in
water following overnight
fast;
[14C]sucralose
Oral, drank sucralose in
water following overnight
fast; [14C]sucralose
1
8M
Urine 14% (range 9e22%).
Feces 78% (range 69e90%);
Recovery 93%
Roberts
et al. (2000)
10
2M
Urine 11% (range 10e13%).
Feces 86% (range 84e87%);
Recovery 97%
Mouse, CD-1, adult
Oral, gavage;
[14C]sucralose
100,
1500
3000
8 (4M, 4F)
4 (2M, 2F)
4 (2M, 2F)
Rats, SD adult
Oral gavage; [36Cl]sucralose
100
1000
6 (3F, 3M)
6 (3F, 3M)
Not determined
Sims
et al. (2000)
Rats, SD,
Bile duct cannulated
Rats, Sprague Dawley,
3 mo
Oral gavage; [36Cl]sucralose
50,100
2 (1F, 1M)
2 (1F, 1M)
6 (3F, 3M)
Urine 23%, Feces 70%; Total 96%
Urine 15%, Feces 74%; Total 92%
Urine 16%, Feces 72%; Total 94%
Calculated 20e30% absorbed
Urine 4e13% after 5d
Feces 85e95% after 5d
Total 94e99%
Bile 3e9% after 32h
Bile 1e2% after 32h
Urine 24h, 5%; 5 d, 5.5%
Feces 24 h 80%; 5 d 93%; Total
97e101%
Urine ~7% dose
Feces ~ 90%
No difference with long term
dietary exposure to sucralose
Urine 24h, 8%; 5d, 22%.
Feces 24h, 22%; 5d, 60%. After.
Total recovery 81e87%
No differences between
pregnant and non-pregnant
Plasma concentrations:
5.8e6.8 mg/ml at 1h;
1.0e1.3 mg/ml at 24h
Plasma concentrations:
28.6e32.6 mg/ml at 1h;
8.0e9.4 mg/ml at 24h
Excretion not measured
Urine 24h, 26.5%; 5d, 28%.
Feces 24h, 65.9%; 5d, 68%.
Total recovery 98%
Calculated 35% absorbed
Primarily unchanged
sucralose.
Glucuronide conjugates of
sucralose 2.6%
Primarily unchanged
sucralose.
Glucuronide conjugates of
sucralose 1.6e1.9%
Primarily unchanged
sucralose. Other ~2%. Too
low for identification
Not determined
Sims
et al. (2000)
Sims
et al. (2000)
Human, adult
14
Oral gavage; [ C]sucralose
10
Rats, Sprague Dawley,
adult
Oral, in diet for 26, 52 or
85 wk, then acute oral
gavage; [14C]sucralose
0 or 3%
diet þ single
oral dose 100
34 (17F, 17M)
Rabbit, New Zealand,
pregnant and nonpregnant
Oral, gavage;
[14C]sucralose
10
3 F non-pregnant
3 F pregnant
Rabbit, New Zealand,
pregnant
Oral, unlabeled in diet for
first 18 days, labeled [14C]
sucralose by gavage on day
19 of gestation
100
750
2F
2F
Dog, 6 month
Oral, gavage;
[14C]sucralose
10
2M, 2F
Primarily unchanged
sucralose. Other <1%
Roberts
et al. (2000)
John
et al. (2000a)
Unchanged sucralose. Other
<1% Too low for
identification
Sims
et al. (2000)
Unchanged sucralose. No
other metabolites.
John
et al. (2000b)
Not determined.
Kille
et al. (2000b)
Primarily unchanged
sucralose.
Glucuronide conjugates of
sucralose ~ 2e8%
Wood
et al. (2000)
Table 3
Summary of Sucralose ADME studies following intravenous administration.
Species
Route of Administration,
Test material
Dose
mg/kg
N
Recovery of radioactivity
(% of administered dose)
Metabolites
Reference
Mouse, CD-1, adult
Intravenous;
[14C]sucralose
Intravenous [14C]sucralose
20
8 (4M, 4F)
Urine 80%, Feces 2%; Total 104%
John et al. (2000a)
2
6 (3F, 3M)
Intravenous [36Cl]sucralose
20
3M
2
2M, 2F
Urine 72e80% Feces 16% after 2d
Total 95e100%
Urine 83% after 2 d
Feces 8% after 2 d, Total 91%
Urine 81% after 5d.
Feces 12% after 5d
Total recovery 96%
Primarily unchanged sucralose.
Other ~8% of total in urine
Primarily unchanged sucralose.
Other ~3%
Not determined
Rats, Sprague
Dawley, 3 mo
Rats, Sprague
Dawley adult
Dog, 6 month
14
Intravenous [ C]sucralose
discussed later in this review.
Sylvetsky et al. (2016) reported plasma concentrations of
sucralose measured before and at various time points for up to 2 h
following oral administration of sucralose in water, soda or seltzer
in adults and children. No other ADME parameters were reported.
In the first experiment, adults (n ¼ 11) consumed 0, 68, 170 or
250 mg sucralose in water (resulting in doses of 0.66e1.31,
1.65e3.28, and 2.51e4.82 mg/kg, respectively). In a second experiment, adults (n ¼ 11) consumed 68 mg sucralose and 41 mg
acesulfame-potassium in either cola or seltzer. In the third experiment, children (n ¼ 11, ages 6e11 y) consumed 0 or 68 mg
Primarily unchanged sucralose.
Glucuronide conjugates of
sucralose 15e20%
Sims et al. (2000)
Sims et al. (2000)
Wood et al. (2000)
sucralose in water, (resulting doses 1.25e2.78 mg/kg). Peak plasma
concentrations (adjusted for body weight) increased with dose as
expected, were similar in adults and children, and were not affected
by delivery vehicle or BMI of the subject. Peak plasma concentrations in subjects administered 68 mg sucralose (0.66e1.31 mg/kg)
were similar to those reported in the comprehensive ADME study
by Roberts et al. (2000) following administration of 1 mg/kg
sucralose, providing further evidence of low absorption of sucralose
in humans, including children.
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
2.1.2. Animal ADME studies
Mice (John et al., 2000a), rats (Sims et al., 2000) and dogs (Wood
et al., 2000) have similar toxicokinetic patterns to that of humans there is little absorption of orally administered sucralose as the vast
majority is excreted unchanged in the feces with urinary excretion
as a minor route (Table 2). Rates of excretion vary somewhat from
species to species based on the percentage of total dose, but no
differences were observed between males and females. Rabbits had
a more prolonged time of excretion than other species (Table 2),
although the metabolic fate was similar with the majority of
sucralose being excreted unchanged in the feces (John et al.,
2000b). It is important to note that the unique extensive
coprophagy in rabbits facilitates oral recycling of compounds
excreted in the feces, including sucralose, back into the digestive
system, slowing excretion (John et al., 2000b). There were no differences in excretion rates between pregnant and non-pregnant
rabbits (John et al., 2000b).
Chronic exposure to high levels of sucralose (3% of diet) for 18
months in rats did not alter the percent of sucralose excreted in the
urine or feces following acute oral exposure to [14C] sucralose (Sims
et al., 2000). There was also no evidence of new or increased levels
of sucralose metabolites in urine or feces. Only unchanged sucralose was detected in fecal extracts, and the same 2 minor metabolites (less than 1% of total dose) observed in the urine were
chromatographically similar to those in rats receiving a single dose
of sucralose without prior dietary exposure. The level of the metabolites was too low for structural determination, but the metabolites are suspected to be the same glucuronide metabolites as
found in dogs (Wood et al., 2000) and humans (Roberts et al.,
2000). This study demonstrated that long-term exposure to
sucralose did not induce mammalian metabolizing enzymes or
change the metabolism of sucralose, and that there was no metabolic adaptation to sucralose by gut microflora (Sims et al., 2000).
Similarly, rats fed 1500 mg sucralose/kg/d for 21 days had no
change in hepatic microsomal enzyme activity (Hawkins et al.,
1987).
Intravenous sucralose dosing was used to further explore
metabolism pathways (Table 3) to obtain higher blood levels of
sucralose than can be achieved with oral dosing due to very low
absorption. Although intravenous administration does not represent human exposures, which will always be oral, the higher blood
levels allow for improved detection of minor metabolites potentially formed in body tissues such as the liver. Metabolites were
characterized using a variety of methods including thin layer
chromatography, enzyme hydrolysis studies, HPLC and GC followed
by mass spectrometry. These additional studies found no new
metabolites formed in any species, confirming results obtained
with oral studies (John et al., 2000a, 2000b; Sims et al., 2000; Wood
et al., 2000). Intravenous administration of radioactive sucralose
was also utilized to evaluate sucralose distribution in the body in
rats and demonstrated no evidence of uptake into the central
nervous system (Sims et al., 2000) or selective or active transport
across the placenta (JECFA, 1989a, 1989b). Further, the glucuronide
conjugates metabolites found in the rat, mouse and dog toxicology
studies, encompassed all types of metabolites found in humans.
These findings demonstrated that rats, mice and dogs are the
appropriate experimental models for toxicology studies of sucralose as the rate of absorption, products of metabolism, and routes of
excretion in these species mimic those observed in humans.
Although rabbits are often the species of choice for reproductive
toxicology studies, it must be recognized that the fate of sucralose
in rabbits is not considered representative of that observed in
humans.
In 2008, Abou-Donia et al. (2008) reported that sucralose could
affect P-glycoprotein transporter and/or p450 enzymes in the gut,
329
and hypothesized that such effects could, in turn, affect the ADME
of sucralose. Certain other papers have referred to the Abou-Donia
et al. publication, in considering potential responses to sucralose
(Rother et al., 2015; Sylvetsky et al., 2015, 2016); however, there is
no evidence from any of the above-discussed ADME studies
(Tables 2 and 3), or subsequent studies, to support this hypothesis.
As will be further discussed below, the exploratory studies conducted by Abou-Donia et al. (2008), had serious problems with
study design and conduct; there was no actual study of the effects
of sucralose alone, on effects or interaction with either P-glycoprotein or p450 enzymes; and there was no actual investigation of
the metabolism of sucralose (Brusick et al., 2009).
2.1.3. In vitro metabolism studies
To further investigate the possible metabolism of sucralose,
in vitro studies evaluated sucralose as substrate for a range of glycosidases from microbial and plant sources, and mammalian intestinal extracts, including a-galactosidase and amyloglucosidase
from Aspergillus niger, yeast a-glucosidase, almond b-glucosidase,
b-galactosidase from Escherichia coli, bakers yeast and Candida
utilis, invertases, and extracts of porcine and calf intestines. There
was no evidence of hydrolysis of sucralose by any enzymes
(Rodgers et al., 1986).
Based on the structure of sucralose, there is a theoretical potential for the formation of mono-chlorinated monosaccharides,
but as illustrated in the discussion above, extensive metabolic
studies of sucralose found no evidence of either dechlorination or
hydrolysis of sucralose in vivo in any species or in extensive food
stability and processing experiments. In addition, numerous in vitro
incubation studies exploring the potential for formation of these
metabolites were conducted (JECFA, 1991b). Specifically, exposure
to 6-chlorofructose from sucralose was concluded to be virtually nil
under all foreseeable storage or physiological conditions (JECFA,
1991b).
2.1.4. Summary of ADME of sucralose
In summary, the fate of sucralose has been shown to be similar
in all species evaluated, with very low levels of absorption and its
primary route of excretion being unchanged sucralose in the feces.
There is no retention or build-up of sucralose in the body. Following
limited absorption from the gastrointestinal tract, minor amounts
of sucralose glucuronide conjugates, which are eliminated readily
in the urine, were detected in all species. Glucuronidation is
commonly known as a metabolic pathway to enable urinary
excretion of xenobiotics. Furthermore, the similarity of the ADME of
sucralose in humans to that observed in rats, dogs and mice, confirms that these species are good models for assessing the safety of
sucralose in man. There is no evidence of either dechlorination or
hydrolysis of sucralose in any species. There is also no catabolic
(break-down) process, confirming that sucralose is not a source of
energy.
2.2. Genotoxic potential of sucralose
The toxicological assessment of a food additive usually begins
with evaluation of genetic toxicology as these studies are considered screening tests both for potential cancer development and
certain adverse reproductive effects. The genetic toxicology, or
genetox, studies are rapid and less costly than long-term animal
experiments; thus obtaining definitive evidence of lack of genetic
toxicology is advised before investment in additional toxicology
studies. At the outset, it is instructive to note that sucralose is not
electrophilic nor does it contain structural alerts for genotoxic or
carcinogenic activity (Berry et al., 2016).
Sucralose was subjected to a full battery of in vitro and in vivo
330
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
mutagenicity and clastogenicity studies, and results were submitted to regulatory agencies around the world (Grice and Goldsmith,
2000). These studies, summarized in Table 4, were subsequently
published in a comprehensive report by Brusick et al. (2010). Four
additional studies by independent investigators prior to 2010 were
also discussed by Brusick et al. (2010) and are included in Table 4.
Thus, none of these studies will be discussed further in this review.
In a more recent study, 5 different artificial sweeteners
including sucralose were reported to have abnormal and lethal
effects in 2 colon cancer cell lines (Caco-2 and HT-29) and a kidney
cancer cell line (HEK-293), treated in vitro with very high concentrations (van Eyk, 2015; study 6 in Table 4). The protocol employed
is one usually used to assess cancer preventive properties of compounds, and the reported results would normally be characterized
as evidence of anti-carcinogenic potential. The findings are questionable since these immortalized cell lines are of limited relevance
to the in vivo or human situation. Furthermore, as the author stated,
these results “cannot be directly extrapolated to the in vivo situation since damage could only be observed at concentration levels in
excess of 1 mM of some of the sweeteners which is unachievable
in vivo after ingesting them orally”. Another phase of this study
involved the in vitro comet assay with positive effects at
0.1e10 mM. A major limitation of the in vitro comet assay is its
proclivity for false positives, in the sense that non-mutagens and
non-carcinogens, i.e., compounds known to be non-genotoxic, are
often found to be positive. In fact, one study showed that of 10
compounds shown to be positive in the in vitro comet assay, 8 were
negative in the in vivo comet assay (Hartmann et al., 2004). These
authors pointed out that cell death, leading to DNA degradation and
migration, is a common cause of these false positives that are not
relevant to the whole animal. Consequently, reported results from
van Eyk (2015) not only are inconsistent with a wide range of other
genotoxicity screening studies for sucralose but are likely false
positive findings in a non-validated in vitro test system.
In summary, the weight of evidence from the genetox studies
described above is that sucralose does not have genotoxic potential
to induce genetic effects or cancer in humans. The Scientific Committee on Food (SCF) stated in its opinion that sucralose as such was
considered to have no genotoxic potential (SCF, 1989).
The US FDA (1998) concluded that any possible concern from
studies on sucralose and its hydrolysis products was outweighed by
the results of carcinogenicity studies that have shown sucralose to
be non-carcinogenic in the mouse and rat (Mann et al., 2000a,
2000b) and have shown an equimolar mixture of 4-CG and 1,6DCF to have no carcinogenic activity in the rat (Amyes et al.,
1986). These carcinogenicity studies (Mann et al., 2000a, 2000b;
Amyes et al., 1986) and a recent study (Soffritti et al., 2016) will
be discussed below.
2.3. Animal toxicology studies on sucralose
It is important to point out that there are specific regulatory
guidelines for the conduct of toxicology studies that are required
for safety assessments for approval of proposed new food additives.
These standardized testing protocols, such as the US FDA Redbook
guidelines and OECD testing guidelines, are considered to be most
reliable for safety assessment because they are designed to facilitate interpretation of results, and to distinguish true compoundrelated outcomes from observations due to normal variability,
background incidences, aging or other factors (OECD, 1998; US FDA,
2000). Critically important is inclusion of sufficient doses to assess
dose response and an appropriate control group. A complete
assessment of the overall animal is conducted including evaluation
of body weight and food consumption, growth and development,
appearance and behavior, hematology, clinical chemistry, urinalysis, ophthalmic changes, and macro- and microscopic (histopathology) examination of over 40 tissue types upon termination. The
parameters evaluated in each study are described below and are
summarized in Tables 6e12.
Another important factor that affects the reliability of the study
Table 4
Genetic toxicology studies conducted with sucralose.
Study number, Test name
Method
Results
Comments
Reference
1) Salmonella mammalianmicrosome test
2) E. coli Pol A ± DNA damage
assay
3) Mouse lymphoma L5178Y TK
±
mutagenesis test
Ames et al. (1975)
Negative in 5 strains, þ/ S9; nonmutagenic and non-toxic
Negative in differential survival
assay; does not induce DNA damage
Non-mutagenic; toxic to
mammalian cells at high doses
Satisfies all criteria for a valid test
Brusick et al. (2010)
Satisfies all criteria for a valid test
Brusick et al. (2010)
Brusick et al. (2010)
4) Human peripheral
lymphocyte assay
5) Rat hepatocyte DNA repair
assay
6) Comet assay in 3 mammalian
cell lines
Preston et al. (1987);
Brusick et al. (2008)
Williams (1977)
Non-clastogenic; toxic to human
cells at high doses
No induction of DNA damage
Satisfies all criteria for a valid test;
negative using current criteria
(Moore et al., 2007); also indicates
non-clastogenicity (Clive et al.,
1983)
Satisfies all criteria for a valid test
Satisfies all criteria for a valid test
Jeffrey and Williams (2000)
Tice et al. (2000)
Induction of DNA damage
van Eyk (2015)
Schmid (1975);
Salamone et al. (1980)
Schmid (1975);
Salamone et al. (1980)
Heddle and
Salamone (1981)
Unknown
Non-clastogenic in 3 trials at
multiple doses
Non-clastogenic at 2 doses tested at
3 time periods
Non-clastogenic at 3 dose levels
after 5 days of oral administration
Non-clastogenic at 2 dose levels
after 5 days oral administration
Non-clastogenic at 3 dose levels in
the diet after 60 days
Induction of strand breaks in
multiple organs of ddY mouse
Non-GLP study using an
unvalidated assay that is prone to
false positive results; see text for
detailed comments
Non-GLP study
Satisfies all criteria for a valid test
Brusick et al. (2010)
Satisfies all criteria for a valid test
Brusick et al. (2010)
Article in Russian; reported in
Brusick et al. (2010)
Satisfies all criteria for a valid test
Durnev et al. (1995)
7) Mouse bone marrow
micronucleus assay
8) Mouse bone marrow
micronucleus assay
9) Rat bone marrow
micronucleus assay
10) Mouse bone marrow
clastogenicity assay
11) Mouse bone marrow
clastogenicity assay
12) Comet assay for DNA strand
breaks
Rosenkranz and Leifer
(1980)
Clive and Spector (1975)
Modified method of Ford
and Hamerton (1956)
Tice et al. (2000)
Unvalidated test system; see text
for detailed comments
Brusick et al. (2010)
Brusick et al. (2010)
Sharma et al. (2007)
Sasaki et al. (2002)
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
331
Table 5
Acute studies conducted in animals on the safety of sucralose.
Study
Doses (mg/kg) and adminstration
Parameters assessed
Reported findings
Comments
Acute, Mice
(ICI Alderley Park strain)
Goldsmith (2000)
Sucralose 16,000;
M and F N ¼ 10/sex/group
Observed 14 d after dosing.
GLP study. Standard
acute toxicity protocol.
Acute, Rats
CD1 (SD)
Goldsmith (2000)
Sucralose 10,000; M, N ¼ 10
Observed 14 d after dosing.
Gross pathology.
Organ weights.
Acute, Rats, Wistar,
Rocha et al. (2011)
0, 0.5, 5 and 50 Commerical
sucralose sweetener;
(composition not reported)
N ¼ 10 to 12 male/group.
Method of exposure not reported.
Radiolabeling and morphology
of red blood cells (RBC) at 0, 15,
60 or 120 min after exposure.
Biodistribution of radiopharmaceuticals
60 min following exposure to
0.5 or 50 mg sweetener formulation/kg.
LD50 of
orally administered
sucralose is
>16 g/kg
LD50 of
orally administered
sucralose is
>10 g/kg
No effect on radiolabelling
or morphology of RBC.
Reported change in
biodistribution of
radiopharmaceuticals;
but no dose response.
GLP study. Standard
acute toxicity protocol.
Composition of test
compound not defined.
Dose response not
observed although
multiple doses used.
Table 6
Short-term standardized animal toxicology studies evaluating sucralose.
Study
Doses (mg/kg) and adminstration
Parameters assessed
Reported findings
Subacute,
4 wk.
Rats CD1 (SD)
Goldsmith (2000)
Sucralose 0, 1.0, 2.5 or 5.0% of diet,
resulting in doses of 0, 737e1287,
1865e3218 and 2794e6406 (M-F)
N ¼ 15/sex/group
Weekly body weight, food
consumption, haematological and
clinical chemistry Urinalysis. Gross
pathology.
Organ weights. Histology of all tissues
from control and high dose groups.
Weekly body weight, food
consumption, Serum ALT, Ca and Mg.
Urinalysis. Organ weights.
GLP study. Reduced palatability
and digestibility of diets
containing high conc. sucralose
seen as cause for decreased
food consumption and other
alterations.
GLP study. Reduced palatability
Decreased food consumption
and body weight and change in and digestibility of diets
containing high conc. sucralose
some relative organ weights
seen as cause for decreased
reported in 5% group.
food consumption and other
alterations.
No other adverse effects.
Increased cecal weight in rats
Increased cecal weight
receiving sucralose. No other
attributed to high amount
treatment-related effects.
unabsorbed sucralose.
Sucralose 0 or 5.0% of diet
Subacute,
N ¼ 6/sex/group
8 wk.
Rats CD1 (SD)
Goldsmith (2000)
Subacute,
4, 9 and 13-week
end points.
Rats SD
SCF (2000a)
Subchronic,
26 wk.
Rats CD1 (SD)
Goldsmith (2000)
Sucralose 0 or 4000 by gavage.
N ¼ 15/sex/group/time point
Body weights, food and water
consumption, clinical signs of toxicity,
urinalysis, gross and histopathology.
Daily body weight, weekly food &
monthly water consumption.
Ophthalmoscopic exams,
haematological and clinical chemistry
Urinalysis. Gross pathology. Organ
weights. Histology of all tissues from
control and high dose groups.
Sucralose 0, 0.3, 1.0, or 3.0% of certified Weekly physical exams, food
Subchronic,
consumption and body weight.
canine diet, resulting in doses of 0,
12 mo.
Ophthalmoscopic exams at 6 and
91-89, 296-274, and 889-858 (F-M)
Dogs, beagle.
12 mo.
Goldsmith (2000) N ¼ 4/sex/group
Baseline urinalysis, haematological and
clinical chemistry. Clinical chemistry &
haematological analyses at 3,6,9
&12 mo. Urinalysis at 3 & 12 mo. Organ
weights, tissue histology on all animals.
Sucralose 0, 750, 1500 or 3000 by
gavage; M and F, N ¼ 20/sex/group
Comments
No adverse effects in 1% or 2.5%
diet groups. Decreased food
consumption, reduced body
weight and some biochemical
changes reported in 5% group.
Increased cecum weight as
consequence of increased
osmolarity of caecal contents.
Not considered a toxicological
response.
GLP study. Use of gavage
eliminated effect on diet
palatability and food
consumption.
NOEL: 1500 mg/kg.
LOEL: 3000 mg/kg.
Increased food consumption in
dogs fed sucralose. No adverse
effects observed.
GLP study. NOAEL at highest
doses of 889 mg/kg (F) and
858 mg/kg (M).
No Observed Effect Level (NOEL); Lowest Observed Effect Level (LOEL).
results, is whether the study was conducted under the principles of
Good Laboratory Practice (GLP), which outlines the conditions for
the planning, performance, monitoring, recording and reporting of
non-clinical safety testing of test items contained in pharmaceutical products, pesticide products, cosmetic products, veterinary
drugs as well as food additives, feed additives, and industrial
chemicals (OECD guidelines, 1998). For example, GLP requires
documentation of purity and stability of the test item, validation of
methods used, specifications of experimental conditions, monitoring, recording and reporting of data, including pathology data
and interpretation of results (Morton et al., 2006). Studies that are
to be submitted to national authorities for the purpose of safety
assessment are required to be conducted using GLP so that authorities can inspect the final reports to confirm that the methods,
procedures, and observations are accurately and completely
described, and that the reported results accurately and completely
reflect the raw data of the studies. Thus, studies conducted under
GLP are considered to be more reliable. The role of GLP in safety
studies for regulatory submissions has been further discussed in a
recent systematic review of the carcinogenicity of sucralose (Berry
et al., 2016).
2.3.1. Acute studies
Most often, following determination of negative findings in
genetox studies, the next step in toxicological assessment of a food
additive is determination of acute toxicity following a one-time
exposure to a high dose in animal models. A commonly reported
value of acute toxicity, the LD50, (i.e. the dose that results in death
332
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 7
Neurotoxicity studies on sucralose and sucralose hydrolysis products in animals.
Study
Doses (mg/kg) and adminstration
Parameters assessed
Reported findings
Comments
Subacute Neurotoxicity
study, 21 d
CD-1 mice
Finn and Lord (2000)
Daily by gavage
Sucralose 1000;
Negative control: water;
Positive control: 6-CG, 500;
Equimolar mixture of sucralose
HP, 150, 500 or 1000
N ¼ 5/sex/group
Equimolar mixture of sucralose
HP, 50, N ¼ 5M
No evidence of neurotoxicity in
mice administered sucralose.
No histological changes in
sucralose HP mice. Mild
behavior changes in high dose
sucralose-HP group.
Clinical and histologic evidence
of neurotoxicity in 6-CG
positive control mice.
Neither sucralose nor its
hydrolysis products
represent a neurotoxic risk.
Subacute neurotoxicity
study, 28 d
Monkeys
Finn and Lord (2000)
Daily by gavage.
Sucralose 1000;
Negative control: water;
Positive control: 6-CG, 500;
Equimolar mixture of
sucralose HP, 1000
N ¼ 3/group
No evidence of neurotoxicity in
monkeys administered distilled
water, sucralose or sucralose
HP.
Clinical and histologic evidence
of neurotoxicity in 6-CG
positive control monkeys
Neither sucralose nor its
hydrolysis products
represent a neurotoxic risk
to man.
Subacute. Developmental
Neurotoxicity, Mice
Viberg and Fredriksson (2011)
Sucralose 0 or 125 by gavage
during postnatal days 8e12.
N ¼ 10e12 male/group
No difference in expression of
proteins 24 h after last dose.
No evidence of
neurotoxicity in mice
administered sucralose
during periods of brain
development.
Subacute, Adult Behavior
following neonatal exposure,
Mice.
Viberg and Fredriksson (2011)
Sucralose 0, 5, 15 or 125 by
gavage during postnatal days 8e12.
N ¼ 12 male/group
Muscle tone, activity, gait,
stance, tremors, righting and
placement reflexes observed
3 daily.
Weekly body weight.
Light and electron microscopy
tissue examinations: brain (7
areas), spinal cord (5 areas),
both sciatic nerves and skeletal
muscle.
Neurological examinations on
day 13 and 27. Food and water
consumption.
Biweekly body weights.
Histopathological and
ultrastructural changes in
neural tissues: and kidneys,
liver, lungs, spleen, eyes,
salivary glands and striated
muscle.
Levels of brain proteins:
calcium/calmodulin-dependent
protein kinase II (CaMKII),
growth-associated protein-43
(GAP-43), synaptophysin, and
tau.
Spontaneous behavior
(locomotion, rearing, and total
activity) at 2 months. Body
weights.
No differences in animal
behavior in adulthood
following neonatal exposure to
sucralose.
Not reported as GLP. No
evidence of neurotoxicity in
mice administered
sucralose during periods of
brain development.
Sucralose hydrolysis products (HP): 6- CG: 6-chloro-6-deoxyglucose.
of 50% of the animals) is based on a standardized GLP experimental
protocol evaluating animal health for 14 days after dosing. The
acute toxicity of sucralose in mice and rats was evaluated at doses
of 16,000 and 10,000 mg/kg body weight, respectively, administered by oral gavage (Table 5; Goldsmith, 2000). No deaths occurred
in either species and no evidence of toxicity was observed during
the following 14 days based on measurement of body weights,
animal behavior and gross evaluation of animal tissues through
necropsy at the end of the experiment. Therefore, the LD50s for
these experiments are actually unknown, and are simply reported
as greater than the doses tested (i.e. >16 g/kg in mice and >10 g/kg
in rats). Based on this study, sucralose has extremely low acute
toxicity potential.
Although not a traditional toxicology study, Rocha et al. (2011
evaluated the acute effects (within 24 h) of a sweetener containing sucralose in rats; however the composition of the sweetener
formulation and amount of sucralose in the formulation was not
reported. The authors found no effect of the sweetener on radiolabelling and morphology of red blood cells. The distribution of 2
radiolabelled pharmaceuticals to the kidney were reported to be
affected by administration of low but not high doses of the
sweetener; however this observation is of no biological significance, given that the test material is unknown and the lack of dose
response.
2.3.2. Sub-acute and sub-chronic toxicity studies on sucralose
Sub-acute and sub-chronic toxicity studies typically range in
duration from more than 24 h up to 10% of the lifespan of the
species. Experimental animals are given repeated oral doses of the
food additive by gavage or it is incorporated into diet to result in
daily exposures. These studies are conducted prior to long term
chronic studies for several purposes, including establishing palatability of the diet containing the food additive, determination of
upper tolerable doses, and identification of potential toxicity due to
repeated exposures. The recommended maximum amount of a
food additive to be added to the diet is 5%, as higher levels may
affect the nutrient content of the diet and potentially lead to
nutritional deficiencies. The sub-acute and sub-chronic toxicity of
sucralose was investigated in rats and dogs according to regulatory
guideline protocols as summarized in Table 6.
In the first rat studies, sucralose was added to the diet at levels of
0, 1.0, 2.5 or 5.0% of diet for 4 weeks or 0 and 5% for 8 weeks
(Goldsmith, 2000). Parameters assessed are shown in Table 6. This
study demonstrated that dietary sucralose levels of up to 2.5% had
no adverse effects, but reductions in food intake in animals fed the
diet containing 5% sucralose indicated reduced palatability and was
associated with reductions in body weight, changes in relative
weights of several organs and some biochemical parameters.
Based on the findings in the 4- and 8-week studies, sucralose
was administered by gavage, to avoid reduced food intake due to
palatability effects, in the subsequent 26-week rat study and doses
of 0, 750, 1500 and 3000 mg/kg bw/d. Although administration of a
substance by gavage is often used to eliminate food refusal due to
poor palatability of a diet containing a test compound, it should be
noted that this results in significant differences in the metabolism
and kinetics of the test agent. Compared to incorporation in the diet
that is consumed over a period of time each day, gavage administration provided the daily dose in one bolus at one time. The NOAEL
in this study was 1500 mg/kg bw/d as an increase in the weight of
cecal contents and an enlargement of the cecal tissues was
observed in the high dose group, attributed to the high osmolarity
of sucralose in the intestinal contents. Enlargement of the cecum is
a common occurrence following repeated consumption of high
doses of unabsorbed substances (Leegwater et al., 1974; Lord and
Newberne, 1990). The high dose group also had higher water
consumption and increased kidney weights relative to body
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
333
Table 8
Short-term studies assessing the effect of sucralose on limited specific endpoints.
Study
Doses (mg/kg) and
adminstration
Reported findings
Parameters assessed
Liver microsomal and cytosolic
Sucralose, 0, 500 or 1500
Liver enzyme induction.
protein, cytochrome P450, 7gavage.
3 weeks. Rats,
Reviewed in JECFA (1989a) Positive control Aroclor 500 ip. ethoxyresorufin-O-deethylase,
5/sex/group
p-nitrophenol-glucuronyl
transferase and glutathione-Stransferase.
Mineral utilization, Rats; 8 Sucralose in diet, 0, 1, 2, 4, or 8%. Health status, body weight,
N:NR
food and water intake, urine
weeks (59 d);
and feces output, mineral
Reviewed in JECFA (1989a);
excretion, urine corticosterone.
SCF (2000a,b)
Sucralose in diet, 0 or 3%;
control diet pair-fed group;
sucralose gavage, control
gavage.
N¼F, 20
Fecal bacteria and intestinal Splenda®
enzymes.
0 & 100, 300, 500, or 1000
12 week study, Adult SD
gavage.
rats. Abou-Donia et al.
M; N ¼ 10/group for fecal
(2008)
analysis. N ¼ 5/group for
protein analysis
11 mg/kg sucralose, in Splenda.
Blood chemistry
6 weeks, weanling Sprague Control-distilled water.
Gavage.
Dawley rats, with
M; N ¼ 10/group
induced diabetes.
Saada et al. (2013)
Palatability;
Rats, 8 weeks.
Reviewed in JECFA (1989a);
SCF (2000a,b)
Histology of pancreas
30 day study, Adult Wistar
rats,
Gupta et al. (2014)
Sucralose: source and purity
undefined.
0 & 3000 mg/kg, by gavage.
M; N ¼ 6/group
Food intake, body weight, feed
efficiency, water intake.
Fecal bacteria, pH.
Intestinal expression of pglycoprotein and CypP450
% body weight relative to initial
weight.
Missing: food intake
Blood glucose, insulin,
triglycerides, cholesterol (Total,
HDL, LDL),
Antioxidant biomarkers in
brain and testes.
Histology of pancreas only.
Final body weight. Missing
data: incidence and severity of
histological changes, food
intake,
insulin and blood glucose
measurements
Comments
Positive control increased liver weight Sucralose had no effect on liver
and all enzyme activities. Reduced P450 enzyme levels; all values within
historical controls.
levels in female rats at 1500 dose
compared to control but within normal
range.
Decreased food consumption & body
weight at high doses. Increased fecal
output, fecal water content and cecal
weights. No dose related effects on
urine, caloric efficiency, corticosterone,
excretion of Ca, Mg, Zn or Cu.
Reduced food intake and body weight
in sucralose-fed and pair-fed compared
to ad libitum control. No effect due to
sucralose gavage.
Reduction in number of bacteria
compared to water group; No dose
response in treatment groups.
Increased p-glycoprotein and CypP450
expression; no dose response.
No effect of sucralose on
mineral utilization.
NOEL ¼ 1000 mg/kg/d;
LOEL ¼ 2000 mg/kg/d
Sucralose in diet impairs
palatability and affects growth
due to reduced food
consumption.
Not GLP; Splenda stated to
contain 1.1% sucralose, 1.1%
glucose, 94% maltodextrin, 4%
water.
No maltodextrin control.
Unconventional protocol.
Missing critical data including:
Stability of diabetic condition,
Food consumption,
Body weight.
Photos of altered histology of pancreas Not GLP; Unconventional
protocol. Missing critical data
in sucralose group.
No change in body weight.
Decrease serum glucose
and triglycerides in normal and diabetic
rats; Increase in Total, HDL and LDL
cholesterol. No effect on insulin or
antioxidant markers.
No Observed Effect Level (NOEL); Lowest Observed Effect Level (LOEL).
weights. No microscopic cellular changes were observed in association with the higher cecum or kidney organ weights (Goldsmith,
2000).
In a 12-month study in dogs (Table 6), sucralose was added to
the canine diet at levels of 0, 0.3, 1.0, or 3.0% (Goldsmith, 2000). No
adverse effects were observed on the many parameters assessed
(Table 6) throughout the study. In contrast to the reduced food
consumption and weight gain in the rat study when sucralose was
added to the diet, dogs fed diets containing sucralose consumed
more food and had significantly higher body weights by the end of
the 52-week study. Thus, the NOAEL was at the highest concentration of 3% sucralose in the diet. Based on food consumption and
body weights, the NOAELs were 889 mg/kg bw/d in female and
858 mg/kg bw/d in male dogs.
2.3.3. Other short-term research studies on sucralose
Several publications have reported short-term research studies
on sucralose or a retail low-calorie sweetener product containing
sucralose, but with limited endpoints and usefulness for assessment of overall toxicity and total body response as compared to
studies following protocols recommended by FDA's Redbook or the
OECD guidelines (OECD, 1998; US FDA, 2000). Table 7 illustrates the
limited nature of these studies. Some of these have been the genesis
of a number of claims of adverse effects (Abou-Donia et al., 2008;
Saada et al., 2013; Gupta et al., 2014).
In the study by Abou-Donia et al. (2008), male rats were gavaged
daily for 12 weeks with aqueous solutions of “Splenda” (a retail low
calorie sweetening product comprised of 99% maltodextrin and 1%
sucralose) and provided rat chow ad libitum. Parameters evaluated
included fecal microflora, fecal pH, the protein level of 3 specific
proteins in the small intestine, and the histology of only the colon.
No other biochemical parameters or tissues were assessed. The
authors reported changes in levels of various fecal microflora, a
change in fecal pH and increased protein levels of P-glycoprotein
and the intestinal cytochrome P-450 enzymes, CYP3A4 and
CYP2D1.
An expert panel review (Brusick et al., 2009) of the study by
Abou-Donia et al., (2008), details serious problems in the experimental design, statistical analysis, and interpretations which
illustrate that allegations of adverse effects attributed to sucralose
were not valid for a number of reasons: (1) 99% of the test agent
was maltodextrin not sucralose; (2) no isocaloric control for
maltodextrin; (3) no dose-response in the changes in protein
expression; and (4) changes in fecal pH were within the range of
normal variation. A subsequent response from the authors
(Schiffman and Abou-Donia, 2012) to the expert panel comments
did not provide any additional experimental data. Abou-Donia et al.
(2008) also asserted that their rat study data supported that
sucralose would adversely affect nutrient bioavailability, however,
numerous larger and longer studies in rats at higher doses (Mann
et al., 2000a; discussed herein) do not support this assertion.
Saada et al. (2013) conducted a short-term study to assess the
effect of sucralose in diabetic rats; however the test agent was not
neat sucralose, but, a product containing primarily maltodextrin
(94%), glucose (1%), moisture (4%) and sucralose (1%). Diabetes was
induced in Sprague Dawley rats with intraperitoneal injections of
334
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 9
Summary of studies assess safety of sucralose on reproduction and development.
Study
Doses (mg/kg)and
adminstration
Teratogenicity Rats (SD)
Kille et al. (2000a)
Sucralose, 0, 500, 1000 or
2000 by gavage during Days
6e15 of gestation
N ¼ 20/sex/group
Teratogenicity Rabbits,
New Zealand
Kille et al. (2000a)
Reproductive,
2-generation
Rats (SD)
Kille et al. (2000b)
Reported findings
Parameters assessed
Food and water intakes, body weight of
mothers. Number of corpora lutea,
implantation sites, resorption sites,
number and distribution of fetuses, fetal
weight, soft tissue or skeletal
alterations, histological examination.
Food and water intakes, body weight of
Sucralose, 0, 175, 350 or
700 by gavage during Days mothers. Number of corpora lutea,
implantation sites, resorption sites,
6e19 of gestation
number and distribution of fetuses, fetal
N ¼ 16e18 inseminated
weight, soft tissue or skeletal
females/group
alterations, histological examination.
Sucralose, 0, 0.3, 1.0 and
3.0% of diet
N ¼ 30/sex/dose 10 weeks
before and during mating,
and through 2 generations.
Reproductive: male fertility Sucralose, 500;
Distilled water (negative
Sperm glycolysis
control);
28-day
6-CG, 24 (þcontrol);
Kille et al. (2000b)
TCDS, 100.
By gavage for 28 days.
N ¼ 10 male rats/group
Reproductive measures (oestrous
cycles, mating behavior, fertility,
gestation, maternal and foetal viability,
foetal development, parturition, pup
maturation and lactation. Food intake,
body weight, organ weights,
histological examination.
Glycolytic ability of epididymal
spermatozoa recovered from rats,
Sperm count.
Comments
GLP study.
No effects in mothers or fetuses in all
groups. No evidence of teratogenicity or NOEL highest dose
>2000
developmental toxicity.
No teratogenic or developmental effects
observed in fetuses at any dose.
Maternal gastrointestinal disturbance
at high dose responsible for 2 maternal
deaths and 4 abortions.
No effect on male or female
reproductive performance. Nonreproductive changes included caecal
enlargement due to large dose of
indigestible material. Reduced body
weight due to reduced food
consumption at high doses, reduced
thymus weight attributed to stress of
reduced weight gain.
Sucralose and TCDS had no effect on
sperm number or glycolysis.
Spermatozoa from rats treated with 6CG exhibited inhibition of glycolysis.
GLP study. NOEL ¼ 350
LOEL ¼ 700 Note: Rabbits
uniquely susceptible to
gastrointestinal effects
following ingestion of large
amounts of osmotically
active, poorly absorbed
compounds.
GLP study. Sucralose at
consumption levels up to
1150 times the expected
daily human intake level
produces no effect on male
or female reproductive
performance in the rat.
Unlike mono- or
disaccharides chlorinated
at only the 6 or 60 position,
neither sucralose or TCDS
affected sperm.
GLP study.
Note - suggest fetal exposure to rabbit maternal of dose of 350 mg/kg is equivalent to fetal exposure of rat maternal dose of 2000 mg/kg (Kille et al., 2000a); 6-CG: 6chloroglucose; TCDS: 6'- substituted isomer of sucralose, trichloro de-oxy sucrose (TCDS, 100 mg/kg/day, a potential trace impurity in commercial sucralose.
streptozotocin (STZ). The treatment was described as 11 mg/kg of
sucralose, administered daily by gavage for 6 weeks, however based
on test product composition, rats also received 1045 mg/kg
maltodextrin and glucose, daily. Critical missing data include food
consumption and body weight, and evidence that the diabetic
condition induced in the rats with STZ had reached a steady-state.
The authors reported increased cholesterol levels diabetic rats
provided the sweetener, and advised lipid monitoring for persons
with diabetes who consume a high amount of sucralose. This
recommendation is not warranted for several reasons. There is no
proposed mechanism for an effect on cholesterol levels. Other reported outcomes also do not indicate other effects on health status
that could be potentially related as the authors state “consumption
of sucralose didn't induce oxidative stress, has no effect on insulin,
[and] reduce[d] glucose absorption”. As only one dose was used, a
dose-response cannot be determined. With these limitations, it is
not possible to attribute any findings of the study to the consumption of sucralose, per se. Furthermore, cholesterol kinetics and
metabolism in rodents differ considerably from humans, as does
response to drug and dietary interventions, illustrated by the lack of
a lowering of cholesterol by statins in rats (Yin et al., 2012).
Another study published with insufficient details and data to
allow interpretation is one by Gupta et al. (2014). Adult male Wistar
rats (n ¼ 6/group) were gavaged daily with 3000 mg/kg sucralose
(source and purity not provided) in water or water alone for 30
days. The only data reported are photographs of altered histology of
the pancreas from a sucralose-treated animal. Changes reported are
only descriptive. No data are provided on incidence or severity of
lesions in either the control or treated group. The authors did not
report food consumption or any relevant confirmatory biochemical
measures that would be expected (such as blood glucose, blood
amylase or insulin levels and insulin granules) in the light of
assertions being made. The use of one dose eliminating evidence of
a dose-response and lack of adequate reporting of critical data
prohibit interpretation of the reported findings. Longer and larger
studies in rats of both sexes also do not report adverse effects of
sucralose on the pancreas (Mann et al., 2000a; Goldsmith, 2000).
Furthermore, the lack of human relevance of compound-induced
pancreatic changes in the rat is well established and shown to be
due to species-related differences in response to cholecystokinin
and expression of cholecystokinin receptor expression in acinar
cells (Myer et al., 2014).
Berry et al. (2016) further discuss the differences in hypothesisdriven research studies such as those described above, as compared
to holistic safety testing studies which require complete assessments and are highly standardized to reduce variability that
confound data interpretation.
2.3.4. Reproduction and development studies on sucralose
The effect of sucralose on fertility and reproduction has been
assessed in several studies reported by Kille et al. (2000a, 2000b)
(Table 8). A 2-generation rat study was conducted to investigate
thoroughly the potential of sucralose to affect the reproductive
performance and/or fertility of both male and female rats (Kille
et al., 2000a, 2000b). Exposure occurred from gametogenesis
through gestation and lactation, with evaluation of a broad range of
endpoints including estrous cycles, mating behavior, fertility,
gestation, maternal health, fetal viability and development, and
pup maturation and lactation. Sucralose, added to the diet at levels
of 0, 0.3, 1.0 and 3.0%, was found to have no effects on male or female reproduction or offspring development. There was no effect
on F0 generation mating performance or outcome following 10
weeks exposure to sucralose prior to mating; no effect on F1
offspring growth or development, including mating performance or
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
335
Table 10
Long-term animal studies on the safety of consumption of sucralose.
Reported findings
Study
Doses (mg/kg) and
adminstration
Chronic toxicity, Rats (SD)
Mann et al. (2000a); JECFA
(1989a).
No effect on reproductive
parameters or offspring.
Lower food consumption and body
weights in groups fed sucralose. No
effect on ophthalmology,
hematology, reduced thyroxine
levels in males but not females,
trend for lower blood glucose, and
higher urinary Mg and P but not
dose-dependent. Higher cecal
weights and incidence of renal
mineralization.
Lower food consumption and body
Sucralose 0, 0.3, 1.0, or 3.0% Body weights, food and water
weight in groups fed sucralose
in the diets, for 104 weeks. consumption, survival, behavior,
Survival rates higher in sucralose
clinical condition, organ weights,
exposure began in utero
and histological examination of all groups. Higher cecal weights and
N ¼ 50/sex/group
incidence of renal mineralization.
tissues in all groups.
4500 mg/kg
No increases in incidence of
neoplasms, no difference in types of
tumors observed due to sucralose.
High dose group had lower body
Sucralose 0, 0.3, 1.0, or 3.0% Body weights, food and water
weight and slight decrease in
consumption, survival, behavior,
of diet
clinical condition, organ weights for erythrocyte counts. No dose-related
1500 mg/kg
all groups. Histological examination response to non-neoplastic lesions.
104 weeks;
No effect on survival or tumor
of all tissues in control and high
N ¼ 52/sex/sucralose test
development.
dose groups
groups
N ¼ 72/sex control group
No effect on food consumption or
Drinking water and feed
Sucralose 0, 500, 2,000,
body weight. Reported decreased
consumption by cage, (10 mice/
8,000, and 16,000 ppm in
cage). Body weights, clinical checks. survival in males, although data
“Corticella”
Diet, from 12th d fetal life Mice allowed to die, stored and
show no dose response. State
necropsied 16e19 h following
until death
increased incidence of brain tumors
death. Survival, tissues preserved in only in males, but data not given.
N (M) ¼ 117, 114, 90, 66,
solvanol Histological examination Higher incidence of some tumor
and 70,
N (F) ¼ 102, 105, 60, 65, and of all tissues.
types in each sex, but not consistent
64.
across both sexes and no clear dose
response. Also lower incidence of
some tumor types in sucralosetreated.
Chronic, carcinogenicity,
Rats (SD)
Mann et al. (2000a)
Carcinogenicity,
Mice, CD-1
Mann et al. (2000b)
Carcinogenicity,
Mice, Swiss
Soffritti et al. (2016)
Parameters assessed
Sucralose 0, 0.3, 1.0, or 3.0%
in the diets, for 78 weeks.
Exposure began in utero as
parental rats fed sucralose
for 4 weeks prior to mating,
and during gestation.
N ¼ 30/sex/group
Mating performance, fertility, litter
size, pup viability of parental rats.
Body weights, food and water
consumption, survival, behavior,
clinical condition, organ weights,
ophthalmology,
hematology, blood and urine
chemistry for all groups.
Histological examination of all
tissues in all groups.
Comments
GLP study. Breeding males and
females fed sucralose diets before
and during mating, gestation and
lactation. Highest diet level during
lactation reduced to 1%.
Body weight, cecal weight and renal
mineralization attributed to lower
palatability and gastrointestinal
effects due osmatic effect of poorly
absorbed compound at high dietary
levels. Not toxicologically relevant
at human consumption levels.
GLP study. Same comments as in
above study. No evidence of
carcinogenic potential of sucralose.
Sucralose is not carcinogenic at
maximum tolerated dose. GLP
study.
NOEL ¼ 1500 mg/kg
LOEL ¼ 4500 mg/kg
Authors state evidence of
carcinogencity, but data do not
support conclusion. Methods
unconventional and pathological
analysis of this laboratory
determined to be unreliable for
certain tumors as discussed in text.
Proposal that sucralose alters gut
microflora is unsubstantiated and is
not recognized as carcinogenic
mechanism.
No Observed Effect Level (NOEL); Lowest Observed Effect Level (LOEL).
Table 11
Additional safety studies on sucralose cited in regulatory agency reviews.
Study
Adminstration
Findings
Conclusion
Immunotoxicity
Tier-I US National
Toxicology Guidelines
28 d, Rats
SCF (2000b)
Utilization of glucose
and lactate, rats
and rat tissue
JECFA (1989a)
Glycolysis in tissues,
rat JECFA (1989a)
Teratology, Rabbit
SCF (2000b)
Insulin secretion, rat
JECFA (1989a)
26-week dietary study
and dietary restriction, rats
SCF (2000b)
Highest dose
3000 mg/kg/d
Both gavage
and dietary
exposure.
In vitro, 5 mM;
In vivo, 500 IP
No effect on immune function, including serum
immunoglobin concentration, spleen lymphocytes and
natural killer cells, bone marrow counts and pathology,
spleen and thymus weight and histology. Lower body
weight at high dose.
No effect in rat brain, liver, kidney, diaphragm or ilium
tissues.
NOEL >3000 for effects on lymphoid organs and immune
system.
In vitro, 5 and
20 mM
Highest dose,
1000 mg/kg/d
In vivo, IV, IP
and oral
Dietary exposure
No effect on respiration or phosphorylation of hepatic or
renal mitochondria
Maternal gastrointestinal effects at high dose.
Sucralose has no effect on regulation of carbohydrate
metabolism.
No evidence of teratogenic potential.
No effect on insulin at 100, 500 and 1000 mg/kg.
Sucralose has no effect on insulin secretion.
Sucralose has no effect on nutrient utilization.
Decreased body weight and food consumption at high doses NOEL 628-787 mg/kg/d
LOEL 1973-2455 mg/kg/d
No Observed Effect Level (NOEL); Lowest Observed Effect Level (LOEL).
outcome, which included in utero and neonatal exposure and
consumption of sucralose in diet from weaning through to adulthood and mating; and no effect on F2 generation health, growth or
development through to adulthood.
An examination of potential adverse reproductive effects in the
male was conducted by assessing glycolytic activity of epididymal
spermatozoa (Kille et al., 2000a, 2000b). In the experiment (Kille
et al., 2000a, 2000b), 6-chloroglucose (24 mg/kg/d) was the
336
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 12
Genetic toxicology studies conducted with degradation products.
Study number, Test name
Method
Results
Comments
Reference
1) Salmonella mammalian-microsome test on 4CG
2) Mouse lymphoma L5178Y TK ± mutagenesis
test on 4-CG
3) Mammalian cell chromosome aberration
assay on 4-CG
4) Human peripheral lymphocyte assay on 4-CG
5) Rat bone marrow cytogenetics assay on 4-CG
6) Salmonella mammalian- microsome test
on 1,6-DCF
Ames et al. (1975)
Non-mutagenic
Satisfies all criteria for a valid test
US FDA (1998)
Clive and Spector (1975)
Non-mutagenic
Satisfies all criteria for a valid test
US FDA (1998)
Standard protocol
Non-clastogenic
Satisfies all criteria for a valid test
US FDA (1998)
Standard protocol
Standard protocol
Ames et al. (1975)
Inconclusive
Inconclusive
Dose-related increase in
revertants in one strain; nonmutagenic in 4 other strains
Dose-related increase in
revertants in one strain; nonmutagenic in 4 other strains
Dose-related increase in
revertants in one strain; nonmutagenic in 4 other strains
Positive response in absence of
S9 greatly reduced with S9
Positive response in absence of
S9 greatly reduced with S9
Negative for induction of
chromosome aberrations
Satisfies all criteria for a valid test
Satisfies all criteria for a valid test
Satisfies all criteria for a valid test
US FDA (1998)
US FDA (1998)
Bootman and
Lodge (1980)
Satisfies all criteria for a valid test
Bootman and
May (1981)
Satisfies all criteria for a valid test
Haworth et al.,
1981
Satisfies all criteria for a valid test
Kirby et al.
(1981a)
Kirby et al.
(1981b)
Bootman and
Rees (1981); US
FDA (1998)
Molecular
Toxicology
(Microtest)
(1994)
Bootman and
Lodge (1981)
Bootman and
Whalley (1981)
Hazleton
Laboratories
America
(1988a)
Hazleton
Laboratories
America
(1988b)
Microtest
Research
Limited (1989)
7) Salmonella mammalian- microsome test
on 1,6-DCF
Ames et al. (1975)
8) Salmonella mammalian- microsome test
on 1,6-DCF
Ames et al. (1975)
9) Mouse lymphoma L5178Y TK ± mutagenesis
test on 1,6-DCF
10) Mouse lymphoma L5178Y TK ± mutagenesis
test on 1,6-DCF
11) Human peripheral lymphocyte assay on 1,6DCF
Clive and Spector (1975)
Clive and Spector (1975)
Standard protocol
Satisfies all criteria for a valid test
Satisfies all criteria for a valid test
12) Rat hepatocyte DNA repair assay on 1,6-DCF
Williams (1977)
No induction of DNA damage
Satisfies all criteria for a valid test
13) Sex-linked recessive mutation in Drosophila
melanogaster on 1,6-DCF
14) In vivo cytogenetic study in rat bone
marrow on 1,6-DCF
15) Sister chromatid exchange in the mouse for
1,6-DCF
Standard protocol
Not mutagenic in this in vivo
insect assay
Chromosome aberrations not
induced
Chromosome damage not
induced
Satisfies all criteria for a valid test
Standard protocol
Standard protocol
Satisfies all criteria for a valid test
Satisfies all criteria for a valid test
16) Bone marrow micronucleus assay in the
mouse for 1,6-DCF
Standard protocol
Chromosome aberrations not
induced
Satisfies all criteria for a valid test
17) In vivo DNA binding study in the rat for 1,6DCF
Research study
Not a validated test method
18) Sister chromatid exchange in the mouse for
equimolar mixture of 4-CG and 1,6-DCF
19) Dominant lethal test in the mouse for
equimolar mixture of 4-CG and 1,6-DCF
Standard protocol
Inconclusive; an association of
test compound with isolated
tissue samples was not shown
to be covalent binding to DNA
Chromosome damage not
induced
Inconclusive
Standard protocol
Satisfies all criteria for a valid test
US FDA (1998)
Satisfies all criteria for a valid test
US FDA (1998)
4-chloro-4-deoxy-D-galactose (4-CG); 1,6-dichloro-1,6-dideoxy-D-fructose (1,6-DCF).
positive control, and purified sucralose (500 mg/kg/day) and a
potential trace impurity of sucralose, trichloro de-oxy sucrose
(TCDS, 100 mg/kg/day), were administered by gavage for 28 days.
No effects on sperm count, sperm glycolysis and sperm ATP production were observed with either sucralose or TCDS exposure, in
contrast to the positive control. These findings further confirm the
lack of effect of sucralose on male reproduction.
In addition, other studies have been published on the effect of
sucralose consumption during the critical periods of development
(Table 8). These include detailed teratogenicity studies conducted in
rabbits and rats (Kille et al., 2000a, 2000b), and a neonatal neurotoxicity study conducted in mice (Viberg and Fredriksson, 2011). No
evidence of an effect of sucralose on fetal development or growth
was observed in rats administered doses up to 2000 mg/kg/d. Preliminary studies in rabbits at doses up to 1000 mg/kg/d demonstrated normal fetal development, but caused maternal
gastrointestinal disturbances at the high dose. (Kille et al., 2000a).
This study confirmed a lack of effect of sucralose on fetal development (Kille et al., 2000a). A NOAEL of 350 mg sucralose/kg bw/d in
rabbits was established by SCF (2000a), based on the observed effects in the dams, considered a consequence of the unique susceptibility of rabbits to gastrointestinal effects due to unabsorbed
material. In contrast, the rat teratology study (Kille et al., 2000a)
established a NOAEL of greater than 2000 mg/kg bw/d (the highest
dose tested) for both maternal and offspring endpoints and rats are
among the species considered appropriate surrogates for humans,
based on the collective evidence.
To assess the potential effects of sucralose on neuronal development, Viberg and Fredriksson (2011) examined the effects of
sucralose consumption during critical periods of brain development on proteins involved in neuronal growth, and the effect of
early exposure to sucralose on behavior during adulthood. The
mouse brain growth spurt is neonatal, occurring during the first
3e4 wk of life, reaching its peak around postnatal day 10. Proteins
that have been shown to be important to neuronal survival, growth,
and synaptogenesis include calcium/calmodulin-dependent protein kinase II (CaMKII), growth-associated protein-43 (GAP-43),
synaptophysin, and tau. Modification of the expression of these
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
proteins resulting from exposure to chemicals during brain development has been shown to result in alterations in behavior,
learning, and memory in adulthood (Viberg and Fredriksson, 2011),
therefore, mice were evaluated both immediately following
administration of sucralose, and subsequently as adults.
In the study, male mice were orally administered 0, 5, 25 or
125 mg sucralose/kg bw/d during postnatal days 8e12 by gavage.
There was no effect of sucralose on brain protein expression levels
assessed after 24 h. Remaining mice were maintained and evaluated 2 months later (adulthood) for spontaneous behavior (locomotion, rearing, and total activity). No significant effect on behavior
was observed following repeated neonatal exposure of up to
125 mg sucralose/kg bw/d. This study provides additional support
for the lack of adverse effects of sucralose on development and
safety of consumption during developmental, including neurodevelopmental stages.
Exposure to sucralose in utero was included in the chronic
toxicity and carcinogenicity study of sucralose in Sprague Dawley
rats (Mann et al., 2000a), which is discussed below. Breeding males
and females were administered sucralose in the diet at levels of 0,
0.3, 1 and 3% for 4 weeks before pairing, throughout pairing, mating, and gestation. During lactation, the level of sucralose in the 3%
group was reduced to 1%, due to higher food intakes during lactation. Sucralose did not affect any reproductive (mating performance, fertility, pup survival) or developmental parameters. A
reduction in body weight in offspring fed the sucralose diets was
observed prior to weaning when pups began eating diets, however,
this was expected with known palatability effects of diet containing
high concentrations of sucralose in rats, and the difference led to no
adverse outcomes. Following weaning, offspring were administered the sucralose diets for either 78 (chronic toxicity) or 104
(carcinogenicity) weeks. No adverse effects due to in utero exposure
were observed.
2.3.5. Long-term chronic toxicity and carcinogenicity of sucralose
The scientific data that are collectively considered in assessing
the carcinogenic potential of a chemical compound include characteristics of the compound, mutagenicity data, studies on the
metabolic fate and toxicokinetics of the compound, all growth and
physiological data from short and long-term bioassays and the
appropriate evaluation of preneoplastic and neoplastic findings.
Berry et al. (2016) have provided an excellent systematic review of
the totality of data specific to the carcinogenic potential of sucralose. These long-term and carcinogenicity studies are summarized
below and in Table 9.
The key studies include 2-year GLP-compliant carcinogenicity
studies in the mouse and rat (Mann et al., 2000a,b) with 0, 3,000,
10,000 and 30,000 ppm sucralose in the diet. The rat study included
an in utero phase and a 52-week chronic toxicity evaluation. In
mice, mean body weight gain in both sexes administered the
highest dose (30,000 ppm) was decreased to about 80% that of the
controls (p < 0.01). Terminal body weights in the high-dose groups
were also lower (p < 0.01) than the controls by about 10%. In rats,
there was a consistent finding of decreased body weight gain in all
groups treated with sucralose (p < 0.001). These lower body
weights were attributable to both decreased food consumption and
a physiological response to a high concentration of a non-digestible
substance in the diet (WHO, 1987; Lu, 1988). Similar effects have
been seen in studies of other non-nutritive sweeteners in prolonged dietary rodent studies (Flamm et al., 2003; Mayhew et al.,
2003). In both rats and mice, no other treatment-related adverse
effects observed, including no neoplastic or non-neoplastic histopathological changes in tissues. In the rat, the only notable effect of
sucralose was the finding of increased absolute cecal weights
(relative weights not reported) in high-dose group animals. As
337
previously discussed, increased cecal weights are commonly seen
when feeding large dietary concentrations of non-digestible carbohydrates to rodent species (Leegwater et al., 1974; Lord and
Newberne, 1990). A NOAEL of 10,000 ppm in the diet, or approximately 1500 mg/kg wt/d was identified on the basis of decreased
overall body weight gains and terminal body weights in mice of
both sexes. In both rats and mice, 2-year GLP-compliant studies
using standard carcinogenicity testing protocols demonstrated that
sucralose is without carcinogenic potential.
Recently, Soffritti et al. of the Ramazzini Institute (RI) have
published a study in mice purporting to show that sucralose is
“carcinogenic” (Soffritti et al., 2016). There are several critical
problems with the RI study design, conduct and data interpretation,
including the inappropriate diagnoses of the reported key finding
of hematopoietic neoplasias.
First of all, the lifetime study design where animals are dosed
until death is recognized as having significant disadvantages
compared with recommended designs for rodent carcinogenicity
testing (Hayes et al., 2011). FDA and OECD standard protocols have a
defined termination date of up to 2 years, or 18 months in some
mouse strains (Gart et al., 1986; Contrera et al., 1996; Williams and
Iatropoulos, 2001; Hayes et al., 2011). The use of the 2-year limit, or
18 months in certain mouse strains, is to reduce potential confounding factors that may be introduced by variations in intercurrent mortality between the control and treated groups (Hayes
et al., 2011). The use of lifetime dosing has been criticized by the
European Food Safety Authority (EFSA, 2006, 2011), as confounders
related to this type of study design “can lead to erroneous conclusions”. The study design and animal model(s) used at the RI also
appear to result in very high, but variable, spontaneous incidence
rates of a number neoplasms in both rats and mice, including
various forms of leukemia and lymphoma, thus confounding data
interpretation.
Other considerations of the study conduct include: (1) there is
no indication that the study was conducted following the principles
of GLP; (2) lack of randomization of animals, an observation that
has been discussed previously as a major flaw of the RI protocol
(Gift et al., 2013); (3) the tumor data for the controls appears to be
identical to that for the controls in a similar study on aspartame
(Soffritti et al., 2010); hence, it is not clear if the controls were
“concurrent” to the study of sucralose, or if the control animal cages
were in the same room as the treatment group cages; and (4) the
use of a diet which is unique to the RI, with unknown nutrient
adequacy when diluted with high amounts of test compound.
These uncertainties raise questions regarding the validity of the
Soffritti et al. (2016) study.
There appears also to be several errors and misreporting and
over-interpretation of the data within the publication. Soffritti et al.
(2016) reported significantly increased incidences of ‘all hematopoietic neoplasias’ in male mice in the highest dose group, which
resulted from combining several hematopoietic neoplasms (lymphomas, leukemias, histiocytic sarcomas) to a total incidence. It is
not clear if it was appropriate to combine tumor types as tumor
types with differing cellular origins should not be combined for
assessment of carcinogenic potential (McConnell et al., 1988; Brix
et al., 2010). This is further complicated by the fact that concerns
have been raised regarding the reliability of RI lymphoma and
leukemia diagnoses (Cruzan, 2009; Gift et al., 2013) and the U.S. EPA
has publicly stated that they will not rely on lymphoma and leukemia data from RI studies (US EPA, 2012). Beyond these issues, in
many lifetime carcinogenicity studies at the RI, the animals have
had a high incidence of respiratory infections a condition which can
lead to the development of lymphoproliferative disorders including
lymphomas and which can also confound the pathology diagnosis.
This was noted in peer reviews conducted on RI carcinogenicity
338
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
studies (NTP, 2011; Gift et al., 2013) where a large number of misdiagnoses were identified.
Soffritti et al. (2016) infer that in male mice sucralose is associated with a significant increase in total malignant neoplasms in
high dose animals with a significant dose-related trend, although
no statistical analyses were presented. The actual incidence rates
for mice with malignant neoplasms in male mice were 56.4%, 58.8%,
58.8%, 53.0%, and 62.9% in the control, 500, 2,000, 8,000, and
16,000 ppm dose groups, respectively. As can be seen, there is an
absence of dose-response and only minor differences in the actual
rate values. No effect is seen in females (percent of females with
malignant neoplasms was reported as 67.6, 64.8, 63.3, 55.4 and 59.4
in the control, 500, 2,000, 8000 and 16,000 ppm dose groups,
respectively. These data do not lend support for a carcinogenic effect of sucralose.
The RI sucralose study has recently been reviewed by EFSA
(2017). The Panel noted that: a) the lifetime dosing protocol utilized is subject to increased background pathology that can
confound study interpretation, b) the study lacked dose-response
relationships for the combined incidence of lymphomas and leukemias, c) there was no plausible mode of action, and d) there was
no evidence of a carcinogenic or genotoxic effect in the extensive
database available for sucralose. As a result, the EFSA Panel
concluded that the conclusions of Soffritti et al. (2016) were not
supported by the available data.
The main inferred biological mechanisms of the alleged carcinogenetic potential of sucralose proposed by Soffritti is alteration of
the gut microflora, as well as allegation of weak mutagenic activity.
However, regulatory and genotoxicity experts concur that sucralose
is not mutagenic and there is no evidence that sucralose alters the
gut microflora. Moreover, as noted previously, radiolabeled sucralose show that gut microflora do not use sucralose as an energy
source or otherwise metabolize sucralose. Indeed, sucralose does
not exhibit any of the ten recognized characteristics of carcinogenic
compounds (Smith et al., 2016). Therefore the credible science and
overall weight of evidence confirms the lack of carcinogenic potential of sucralose.
2.3.6. Specialized animal studies on sucralose
2.3.6.1. Neurotoxicity. Although no indication of neurotoxicity was
observed in the comprehensive toxicology studies conducted with
sucralose described above, additional specialized neurotoxicity
studies were undertaken due to the structural similarity of the
hydrolysis products of sucralose and certain monochlorinated
monosaccharides known to have neurotoxicity potential (Finn and
Lord, 2000). The chlorosugar, 6-chloro-6-deoxyglucose (6-CG),
which is monochlorinated in the C6 position, has been reported to
induce neurotoxicity in mice and monkeys, but not in rats
(reviewed in Finn and Lord, 2000). As chlorination of monosaccharides in positions other than C6 results in loss of neurotoxic
activity, neurotoxicity of the sucralose hydrolysis products, 1,6-DCF
and 4-CG, was unlikely as they are not monochlorinated in the C6
position. Specific studies summarized in Table 10 were conducted
in the 2 most sensitive species, mice and monkeys, to confirm the
lack of neurotoxicity of sucralose and the potential hydrolysis
products.
Male and female mice were gavaged with water (negative
control), 6-CG (positive control, 500 mg/kg), sucralose (1000 mg/
kg), and 4 doses of an equimolar mixture of 1,6-DCF and 4-CG (50,
150, 500 and 1000 mg/kg) for 21 days (see details in Table 10).
Similarly, male marmoset monkeys were administered water
(negative control), 6-CG (positive control, 500 mg/kg), sucralose
(1000 mg/kg), or an equimolar mixture of 1,6-DCF and 4-CG
(1000 mg/kg) by gavage for 28 days. Mice and monkeys were
evaluated for clinical symptoms of neurotoxicity multiple times
daily during life, and evidence of histological changes in the CNS
using light and electron microscopy upon termination (Finn and
Lord, 2000). Clinical and histological changes indicative of neurotoxicity were observed in groups administered 6-CG in both species, but no changes were detected in mice or monkeys receiving
sucralose or sucralose hydrolysis products (Finn and Lord, 2000),
confirming lack of potential risk of neurotoxicity by sucralose or its
hydrolysis products in man.
The clear evidence of lack of neurotoxicity of sucralose is also an
important point with respect to the frequent allegations that
sucralose may induce similar toxicity to organochlorine compounds. The main site of toxicity observed with exposure to organochlorines, such as cyclodienes, mirex and lindane, is the
central nervous system, presenting with symptoms of neurotoxicity (Ecobichon, 1996; Reigart and Roberts, 2013; Tordoir and van
Sittert, 1994). Clearly, unlike organochlorines, the results of
extensive investigations demonstrate that sucralose does not cause
neurotoxicity and is dissimilar from organochlorines in many other
aspects as discussed earlier (see Table 1).
2.3.6.2. Effect on gastrointestinal microflora. The role of the gut
microflora on health has become an intensive area of research, and
a wide array of dietary compounds have now been reported to
affect the gut microflora composition. These include protein, fat,
carbohydrate, fiber and many other dietary components (Fava et al.,
2013; Xu and Knight, 2015). Whether an overall change is beneficial
or likely to lead to adverse health effects is often not clear. In
addition, with any study on effects of diet on specific gut microflora
population density or prevalence relative to one another, it is critically important to assess that the study design will control for the
wide range of factors that can affect these outcomes such as other
dietary components, total food intake, fluid intake, housing conditions which can induce stress factors, and potentially alter study
outcome.
Two animal research studies, one by Abou-Donia et al. (2008),
another by Suez et al. (2014), have reported the potential for
sucralose to alter gut microflora. The study by Abou-Donia et al. is
discussed previously and is the subject of a rigorous critical review
(Brusick et al., 2009) which noted that study did not control for the
maltodextrin in the source of sucralose. Dietary intake of solid food
(Purina rat chow) was not measured. As pointed out by Brusick
et al. (2009), microflora population density is readily impacted by
dietary differences, including and specifically with differences in
type of carbohydrate exposure. Thus, the design confounds interpretation of results with respect to investigating effects of sucralose. Moreover, while the study reports changes in certain gut
microflora populations, the magnitude of the change is within
normal variation and colonic microflora concentration was based
on the wet, rather than dry weight of feces, so differences in
number of microflora could be due to differences in water content
of feces (Brusick et al., 2009).
The study by Suez et al. (2014) reported that “artificial sweeteners” affected gut microflora and that this change in microbiome
led to glucose intolerance. Several points should be made with
respect to the applicability of the authors’ general conclusions to
sucralose. Most importantly, the effect of consumption of sucralose
on gut microflora was actually not assessed in this study but rather
was inferred by assessing other parameters. Changes observed in
the microflora of animals fed saccharin were assumed to be
representative of all artificial sweeteners, despite that sucralose
and other sweeteners have very different chemical structures and
metabolic profiles (Magnuson et al., 2016).
In this study (Suez et al., 2014), groups of mice consumed, for 15
weeks, plain drinking water or drinking water with an added
sweetener (10% w/w concentration): either glucose, sucrose or a
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
retail low calorie sweetener product containing either 5% aspartame, sucralose or saccharin and 95% carbohydrate. Standard rodent chow (Harlan-Teklad) was available ad libitum. After 11
weeks, antibiotics were added to the drinking water and were
present until the end of the 15-week test. Some mice in each group
received a combination of ciprofloxacin and metronidazole, and the
remaining received vancomycin. An oral glucose tolerance test
(OGTT) was performed at 11 and 15 weeks, or just before and at the
end of the antibiotic treatment. The mean blood glucose AUC for all
mice that consumed water with a retail low-calorie sweetener
product (data pooled for the 3 groups of mice; single data point)
was reported as statistically significantly increased compared to the
mean blood glucose AUC for all mice that consumed water with
either no sweetener or glucose or sucrose (data pooled for the 3
groups of mice; single data point). All groups (data pooled for 6
sweetener groups and both antibiotic treatment groups) had a
mean blood glucose AUC that was lower after the 4 weeks of the
antibiotic treatment compared to results obtained just before
antibiotic treatment. The investigators reported that the combined
results of these investigations suggest that exposure to any of the
non-nutritive sweeteners consumed during these tests induced
“marked glucose intolerance” “mediated through alterations to the
commensal microbiota”.
There are serious issues with the data interpretations of this
study. There was no investigation of effects of either pure sucralose,
or a retail product containing sucralose, on gut microflora population types or levels. Lower blood glucose AUC after antibiotic
treatment in sucralose-treated mice cannot be presumed to be
evidence of sucralose inducing glucose effects via an effect on the
gut microflora. Indeed, a lower blood glucose AUC was found for all
groups of mice, including those who had only plain drinking water,
after the 4-week exposure to antibiotics. The statistically significant
“increase” reported applies to a comparison of pooled data for
different groups. Importantly, there is no baseline or mid-point
blood glucose control measures for comparison to understand
either within-test variability or treatment response. Additionally,
there were significant differences between test and control groups
in food and water intakes, which may alone have affected gut
microflora making it difficult to interpret any of the results. Notably,
differences in water intake could result in dissimilar total antibiotic
intakes. Thus, the design methodology does not make it possible to
understand the impact of any one sweetener on either microflora or
glucose control. There can be no meaning ascribed to the results
reported with the significant limitations to this study.
In contrast, both subchronic and long-term studies in rodents,
discussed previously, support no clinically meaningful, if any, effect
on gut microfloral populations or profiles. As noted previously,
sucralose is not an energy source for, and is not otherwise metabolized by, gut microflora. Additionally, as discussed previously,
toxicokinetic metabolic studies in both rodents and humans reported that all sucralose in feces is unchanged, indicating that
sucralose is not a substrate for colonic microflora (Roberts et al.,
2000; Sims et al., 2000; John et al., 2000a, 2000b). The potential
for effects on gut microflora was also assessed by the SCF (2000a)
who concluded there was no evidence supporting metabolic
adaptation to sucralose due to the high stability and resistance to
hydrolysis of sucralose (SCF, 2000a). This is further supported by
studies conducted with sucralose and other sources of microflora
including oral cavity pathogens (Young and Bowen, 1990) and
environmental microflora (Omran et al., 2013) confirming that
sucralose is nonnutritive to bacteria and resistant to degradation.
It's also important to remember that gut microflora populations
change readily with dietary variations and there is significant interindividual variability in the gut microbiome profile for humans.
Minor changes cannot be assumed to result in adverse
339
consequences.
2.3.6.3. Additional animal studies on sucralose safety cited in regulatory reviews. Other unpublished sucralose safety studies summarized in Table 11 were submitted to and reviewed by regulatory
agencies. These addressed potential questions on possible effects of
sucralose on palatability, immunotoxicity, mineral bioavailability,
insulin secretion, tissue glycolysis and liver enzyme induction.
Comments on these studies are also available in JECFA (1989b,
1991b) and SCF (2000a) reviews. The studies were also submitted
in other regulatory petitions, and, in the U.S., are subject to review
through the U.S. Federal Freedom of Information Act. The available
regulatory comments on these studies indicate that they provide
additional evidence that sucralose is without adverse effects with
intended uses.
2.4. In vitro studies other than genotoxicity studies
It must be recognized that there are both limitations and advantages to in vitro toxicity studies conducted with the aim of
assessing potential safety effects of oral exposure to food ingredients. One important limitation is that the direct addition of
the food ingredient to tissues or cells eliminates the critical processes of digestion, absorption and metabolism that may greatly
affect the true exposure of the tissue cells to the test compound. In
the case of sucralose, the low levels of absorption of sucralose into
the blood will result in very low tissue level exposure concentrations. For example, even for consumers with high intakes, average
daily maximum intake is estimated at less than 150 mg, of which
only about 15% is absorbed, or about 7.5 mg. Since sucralose is
distributed to essentially total body water, maximum blood levels
are exceedingly low, thus limiting the potential for effects.
Furthermore, sucralose ingestion will commonly occur with the
simultaneous ingestion of foods and/or beverages, the digestion
and absorption of which can also affect exposure of luminal
gastrointestinal tissues to sucralose. Thus, the exposure scenario of
sucralose added to isolated tissues or cells, as occurs with in vitro
studies, does not represent the exposure scenario that would result
from consumption of foods sweetened with sucralose.
There have been a few in vitro studies conducted that report a
potential adverse effect of sucralose, however, the interpretation of
these studies must be made with consideration of the limitations of
such test systems, and the low absorption of sucralose and system
exposure as discussed above.
Rocha et al. (2012) used a commercial formulation of sucralose
(concentration of sucralose undefined) with lactose as the dispersant in an in vitro assay to evaluate effect of sucralose on red blood
cells and to assess whether sucralose interferes with the radiolabeling of blood constituents with 99Tc. Blood from Wistar rats was
incubated for 60 min with concentrations of up to 4 mg/ml of the
commercial formulation. Even at these high concentrations,
sucralose caused no interference with 99Tc labeling of blood constituents (red blood cells, plasma proteins) or morphology of red
blood cells. This study followed earlier investigations by this same
laboratory that also reported no effect on these parameters in rats
dosed with up to 50 mg/kg commercial formulation of sucralose
(Rocha et al., 2011).
Rahiman and Pool (2014) examined the effect of various low
calorie and caloric sweeteners on human whole blood cultures
treated with the B-cell mitogen, lipopolysaccharide (LPS), and the T
cell mitogen, phytohemagglutinin (PHA) as a model system of immune function. Various sugars and commercial formulations of
aspartame, saccharin and sucralose were added to the whole blood
cultures. The study did not include tests for either sucralose alone
or carrier alone. No evidence of cytotoxicity was reported for any
340
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
test agent. The authors suggest that sucralose-containing sweeteners may potentially reduce humoral immunity and affect susceptibility of the defense of the host against extracellular
pathogens. The conclusions are not well justified, however,
considering the very low levels of absorption of sucralose and lack
of support by in vivo studies, demonstrating no effect on immune
function in rats in a 28-day oral immunotoxicolgy study with doses
up to 3000 mg/kg (SCF, 2000a).
There are other published in vitro studies where sucralose has
been used to study specific cellular mechanisms. One such example
includes observations that sucralose and other artificial sweeteners, added to specific cells isolated from the pancreas, can activate the glucose-sensing receptor called TIR3, and lead to elevation
of adenosine triphosphate (ATP) within the cells (reviewed in
Kojima et al., 2015). It is not possible to utilize such studies in the
safety assessment of sucralose as a food ingredient, for several
reasons. The concentrations used in cell studies are not reflective of
in vivo tissue and/or blood level concentrations that result from oral
consumption of the amount of sucralose found in foods and
beverage. There is no evidence that the observed biological
response is an adverse effect, and most importantly, whether such
an effect would be observed in a whole animal where many other
biological systems are active. For example, while increases have
been reported in the secretion of certain gut incretins in some
in vitro studies, the in vivo, including clinical studies, do not provide
support for such changes nor any clinically meaningful effect of
nonnutritive sweeteners on gut function (Bryant and Mclaughlin,
2016). Thus, these mechanistic studies are not further discussed
in this safety review.
In summary, potential adverse effects proposed based on in vitro
studies with sucralose have not been observed or confirmed in
extensive in vivo animal and human oral exposure studies.
2.5. Animal studies on the safety of hydrolysis products
In addition to safety studies on sucralose, extensive safety
testing of potential breakdown products resulting from hydrolysis
was conducted, and the studies were submitted to regulatory
agencies prior to approval of use of sucralose. As discussed in
Section 1.4, under acidic storage conditions, sucralose can be hydrolyzed to result in the two chlorinated monosaccharides, 1,6dichloro-1,6-dideoxy-D-fructose (1,6-DCF) and 4-chloro-4-deoxyD-galactose (4-CG). This is a pH and temperature dependent process, which is very slow under normal storage conditions; for
example, less than 1% of sucralose is expected to be hydrolyzed
after one year exposure at 25 C if the pH is 3, with no detectable
hydrolysis at pH 4 and higher (Grice and Goldsmith, 2000).
It should be restated that these are not formed under physiological conditions in the body, and are not metabolites of sucralose
(Grice and Goldsmith, 2000). However, as there is potential for
exposure to these compounds when consuming sucralose with
extreme storage conditions, extensive toxicology testing of these
compounds was conducted, submitted to regulatory agencies, and
reviewed to ensure they did not present any safety concerns.
Among the toxicology tests performed was a battery of genotoxicity assays. Unlike the regulatory studies on sucralose that
appeared in the peer-reviewed literature (Brusick et al., 2010), the
tests on the hydrolysis products have not been published in a
consolidated form. Rather, the test results were discussed in reviews by various agencies and regulatory authorities. Thus, they are
included in the present review (Table 12) and discussed briefly as
follows.
4-CG was non-mutagenic in the Ames bacterial assay (US FDA,
1998; study 1 in Table 12) and in the mouse lymphoma mammalian mutation test (study 2 in Table 12), and was non-clastogenic in
the in vitro chromosome aberration test (study 3 in Table 12). It was
considered to be inconclusive in a human lymphocyte chromosomal aberration study (study 4 in Table 12) and in an in vivo cytogenetics assay in rat bone marrow (study 5 in Table 12). These
studies on 4-CG are summarized in a Federal Register notice by the
US FDA (1998), with the concluding comment that no genotoxicity
test showed a genotoxic response for 4-CG.
1,6-DCF was found to be positive in the Ames test (study 6 in
Table 12), showing a dose-related response in strain TA1535, plus
and minus mammalian activation (Bootman and Lodge, 1980);
however, no effect was found in other standard tester strains.
Similar results were found in repeat studies in the same laboratory
(Bootman & May 1981; study 7 in Table 12) and in a different
laboratory (Haworth et al., 1981; study 8 in Table 5). Two mouse
lymphoma mutation assays showed that positive results in the
absence of metabolic activation were greatly reduced in the presence of mammalian metabolic enzymes (Kirby et al., 1981a,b;
studies 9 and 10 in Table 12). 1,6-DCF did not induce chromosome
aberrations in cultured human lymphocytes (Bootman and Rees,
1981; study 11 in Table 12). The finding that 1,6-DCF did not
induce DNA damage in isolated rat hepatocytes (Molecular Toxicology (Microtest) 1994; study 12 in Table 12) was confirmed as
negative by the US FDA (1998) and the UK Committee on Mutagenicity (COM, 1995). 1,6-DCF did not induce sex-linked recessive
mutations in the fruit fly Drosophila melanogaster (Bootman and
Lodge, 1981; study 13 in Table 12), chromosome damage in an
in vivo rat bone marrow assay (Bootman and Whalley, 1981; study
14 in Table 12), sister chromatid exchange (Hazleton Laboratories
America, 1988a; study 15 in Table 12) or micronucleus production
(Hazleton Laboratories America, 1988b; study 16 in Table 12) in the
mouse. These in vivo studies show that the weakly genotoxic in vitro
effects of 1,6-DCF were not found in the whole animal. Lastly, a rat
study to investigate possible DNA binding was performed
(Microtest Research Limited, 1989; study 17 in Table 12), and the
results were inconclusive (US FDA, 1998). Two further in vivo
studies were performed using an equimolar mixture of 4-CG and
1,6-DCF. As reported by the US FDA (1998), the mixture was not
genotoxic in the sister chromatid exchange assay in mice (study 18
in Table 12), and the dominant lethal test in the mouse (study 19 in
Table 12) was inconclusive. Given the absence of genotoxicity in 3
in vivo studies using standard, validated protocols, 1,6-DCF was
considered not to be a genotoxic hazard in the whole animal, as
summarized by SCF (2000a).
2.6. Human clinical studies on sucralose
International food agencies around the globe conduct safety
assessment of food ingredients, such as low calorie sweeteners,
using similar assessment approaches (reviewed in Magnuson et al.,
2013). In all cases, safety related to exposure to a food ingredient is
based on assessment of the chemistry and structure of the ingredient, results from extensive genetic and animal toxicology studies,
and the predicted levels of consumption of that ingredient as discussed in Section 1. Typically, human clinical studies are not a
requirement for novel food ingredient or food additive approvals,
although human studies are often done to establish the appropriateness of animal models used/to be used in safety investigations.
In a review of the FDA safety assessment process, Rulis and Levitt
(2009) note that a 100-fold ‘‘safety factor” is applied when establishing safe intake levels, to account for both inter-species variations and “normal genetic variations and the range of
susceptibilities that is possible across the human population”, and
that this approach has been used by regulatory agencies including
FDA “for many decades, and has proven to be reliably protective of
public health.”
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
The human studies with sucralose initially conducted were
specifically designed to assess overall safety and tolerance of consumption of sucralose, and thus included many physiological and
biochemical measurements. Most important, from a safety assessment perspective, are those involving repeated daily consumption
of sucralose (Baird et al., 2000; Binns, 2003) (summarized in
Table 13). As sucralose was intended to be used as a sugar
replacement by individuals with diabetes, studies with both Type 1
insulin-dependent (T1D) and Type 2 non-insulin dependent (T2D)
diabetic subjects were conducted to determine if sucralose consumption had any effect on blood glucose control and insulin prior
to approval. These included both acute (single dose) (Mezitis et al.,
1996) and repeated daily exposure (3 month) studies in persons
with diabetes (Grotz et al., 2003). These studies are summarized in
Tables 13 and 14, and will be discussed in detail below. In addition,
there have been numerous single dose acute response studies in
humans, which have focused on the potential for sucralose to have
an effect on glycemic response and incretin secretion. The details
are summarized in Tables 15 and 16. The usefulness of these studies
from an overall safety assessment will be discussed.
2.6.1. Repeated daily consumption of sucralose in human subjects
2.6.1.1. Healthy subjects. To assess the effect of repeated doses of
sucralose, Baird et al. (2000) conducted two single-blind randomized studies in healthy volunteers. In the first study (“17-Day
ascending dose study”), a pilot tolerance study in eight subjects,
ascending doses of sucralose were given up to 10 mg/kg every other
day for the first 10 days, up to 5 mg/kg for another 7 days, followed
by an 8- day washout period. Endpoints included physical examinations blood biochemistry, hematology, urinalysis, blood insulin
and blood sucralose concentrations. In the second study (Baird
et al., 2000), a larger and longer follow-up tolerance study, participants consumed either sucralose (up to 500 mg/d) (n ¼ 77) or
fructose (50 g/d) (n ¼ 31) for 13 weeks. Based on body weights of
the participants, intakes of sucralose varied between 4.8 and
8.0 mg/kg bw/d for males and 6.4 and 10.1 mg/kg bw/d for females.
Prior to the study, ECG, hematology, standard serum biochemistry
and urinalysis were conducted. Blood and urine parameters were
analyzed at the end of weeks 3, 7 and 13. Sucralose blood levels
were also determined in 5 male and 5 female participants on 5 days
before and after dosing during week 12.
The results demonstrated that daily sucralose consumption of
up to 10 mg/kg bw for 13 weeks had no adverse effect on any health
parameters measured. No evidence of sucralose accumulation in
the blood confirmed the results of toxicokinetic studies in animals.
Consumption of sucralose had no effect on blood glucose or insulin
levels in either the pilot or follow-up tolerance studies, and did not
affect the normal insulin response to sucrose. Thus, these two
studies resulted in a very comprehensive assessment of the tolerability to sucralose, and demonstrated no adverse effect of repeated
sucralose consumption at levels much higher than the EDI (estimated daily intake) of 1.1 mg sucralose/kg bw/d (Baird et al., 2000).
Other clinical trials were conducted in the pre-approval investigations of sucralose safety, and the resulting data submitted to
and reviewed by regulatory agencies (SCF, 2000a; Binns, 2003). The
details and results available from such reviews are summarized in
Table 13. In a double-blind randomized placebo-controlled study,
healthy males consumed 1000 mg of either sucralose (n ¼ 22) or a
placebo of cellulose (n ¼ 23) daily for 12 weeks. The average dose
based on body weight was 13.22 mg/kg bw/d. HbA1c and fasting
glucose, insulin and C-peptide were measured weekly. OGTTs were
conducted at baseline and after 6 and 12 weeks, within 15 min of
consuming sucralose or the placebo. All measures of glucose control were within normal range throughout the study, and there
were no differences between groups in changes from baseline in
341
any measure. This study thus demonstrated no effect of daily
consumption of 13 mg/kg bw/d sucralose on either glucose control
or measures of insulin sensitivity (Binns, 2003).
Although not designed as a safety study, for sake of completeness, we also note a 6-month lifestyle change study by Rodearmel
et al. (2007), that included sucralose consumption, designed for use
by families with at least one overweight child. The goals were to
both increase daily energy expenditure by 100 calories, through
modest changes in physical activity, and eliminate at least 100
calories from the diet by replacing dietary sugar with sucralosesweetened products for 6 months, with the aim of helping to prevent excessive weight gain in the overweight child. Compared to
the control group, who was asked to monitor physical activity, but
not to change diet or activity, the experimental group consuming
sucralose reported a reduction in sugar intake, increased physical
activity and improvement in BMI for age. No other health parameters were reported.
2.6.1.2. Diabetic subjects. A double-blind, placebo-controlled, randomized multi-center repeated-dose study was conducted in subjects with T2D for 3 months (Grotz et al., 2003). A 6-week screening
established eligibility and baseline glucose homeostasis. Subjects
received 667 mg encapsulated sucralose (n ¼ 67) or placebo (cellulose) capsules (n ¼ 69) daily for 13 weeks. Fasting HbA1c, plasma
glucose and serum C-peptide levels were measured every 2 weeks.
Subjects were also monitored during a 4-week follow-up phase
during which all subjects received placebo capsules. There were no
significant differences in the primary measure of blood glucose
control, HbA1c, between groups at baseline or throughout the
experiment. Similarly, sucralose had no effect on fasting glucose, Cpeptide levels or diabetic therapeutic regime (i.e. medications)
during the 3-month study. Four weeks after the treatments ended,
there was a significant reduction in fasting plasma glucose in the
sucralose-treated group relative to baseline and compared to the
placebo-treated group, with no differences in serum C-peptide.
This study where daily sucralose consumption was approximately
7.5 mg/kg bw/d, or about 6e7 times the estimated average daily
intake, demonstrates no effect of repeated daily consumption of
sucralose in persons with T2D and further supports the safety of
human sucralose consumption.
Reyna et al. (2003) conducted a study with 16 well-controlled
T2D male participants to compare possible metabolic and anthropometric benefits resulting from consumption of a diet consistent
with American Diabetes Association (ADA) guidelines for diabetic
patients (control, n ¼ 8) or the ADA diet modified to incorporate a
fat replacer and non-sucrose sweeteners (sucralose and fructose)
(experiment, n ¼ 8). Participants in the experimental diet group
consumed bread prepared with a fat replacer and cookies sweetened with fructose and sucralose (30:70 ratio), replacing sucrose.
After 4 weeks, participants in both groups had lower body weight,
BMI, fasting blood glucose, HbA1, total cholesterol, LDL cholesterol,
and triglycerides and higher HDL-cholesterol; however, greater
improvements were observed in the experimental diet group
(Reyna et al., 2003). There are several limitations of this study, as
the amount of sucralose consumed per day is not reported and
observed benefits cannot be attributed to sucralose alone as there
were numerous differences between the two diets. The lack of
evidence of adverse effects on the measured glycemic parameters,
however, supports the conclusion that sucralose in the diet would
be well-tolerated by persons with diabetes.
2.6.1.3. Summary. In summary, all of the clinical studies that have
evaluated the effects of long-term ingestion of sucralose support its
safety under the intended conditions of use. No adverse effects
were observed on glucose homeostasis, as determined by fasting
342
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 13
Summary of studies evaluating effects of repeated consumption of sucralose in healthy subjects.
Reference
Dosing
Measurements
Results
Conclusions
Baird et al. (2000) Study 1:
Duration: 17 d
Healthy subjects
(n ¼ 8; 4M/4F)
Ave wt.: 70 kg
Ave. Age: 32 yrs
Single-blind randomized
ascending dose. Subjects
fasted prior to dosing.
Sucralose provided in
aqueous drink.
Phase 1
Day 1 ¼ 0 mg/kg
Day 3 ¼ 1 mg/kg
Day 5 ¼ 2.5 mg/kg
Day 7 ¼ 5 mg/kg
Day 9 ¼ 10 mg/kg
Phase 2
Days 11e13 ¼ 2 mg/kg/d
Days 14e17 ¼ 5 mg/kg/d
Day 18e25 ¼ 0 mg/kg
- No effect on fasting insulin
or blood glucose over
course of study.
- No abnormal variation in
body temperatures, pulse,
BP or respiratory rates.
- No effect on
haematological,
biochemical, urine values.
All within normal range.
- Peak sucralose
concentration detected 1 h
after 10 mg/kg dose, at
500 mg/ml. Decreased to
barely detectable by 24 h.
- Normal insulin response
to sucrose 1 wk after the
last dose of sucralose (Day
25).
Consumption of sucralose
had no effect on blood
glucose or insulin levels. No
evidence of adverse effects.
Demonstrates no
bioaccumulation of
sucralose in blood.
Baird et al. (2000) Study 2:
Duration: 13 wk
Healthy subjects
Control group:
n ¼ 31 (17 M/14 F);
Ave. wt.: 69.3 kg;
Ave. Age: 33.9 yrs.
Sucralose group:
n ¼ 77 (47 M/30F);
Ave. wt.: 71.5 kg;
Ave. Age: 34.6 yrs
Single-blinded,
randomized, controlled.
Aqueous drinks consumed
2/d for total daily dose:
Control group: Fructose
Weeks 1e13: 50 g/d
Sucralose group:
Weeks 1e3; 125 mg/d
Weeks 4e7: 250 mg/d
Weeks 5e12: 500 mg/d
Phase 1
Physical examinations before
and at 0, 2, 4, 8, 12, 16, 20, and
24 h post-dosing
.. Insulin 0.5 h after each dose.
24 h after every dose:
- ECGs, Hematologya
- Blood biochemistryb
Daily urine analysis first 9 daysc
Day 9, Blood sucralose levels 0,
0.5, 1, 2, 3, 4, 6, 8, and 24 h post
dosing.
Phase 2
Physical examinations after
each dose, as in Phase 1.
Daily ECG.
Daily insulin, 0.5 h post-dosing.
Daily urinalysisc
Days 14, 18, and 25, blood
collected 0.5 h after each dose
for haematologya and
biochemistryb
Day 25: Blood insulin level 0.5 h
after 50 g standard sucrose.
Physical examination before
and at end of study.
Blood and urine analyses at end
of week 3, 7 and 13.
Fasting and 2-h post-dosing
blood sucralose concentrations
daily during mornings of week
12 (for 10 subjects taking
sucralose).
Ophthalmological
examinations on 24 subjects
(18 sucralose, 6 control).
No changes in weight, body
temperature, pulse, blood
pressure, respiratory or
cardiovascular values. No
change in ECG parameters.
No change in
haematological,
biochemical, or urine
analysis parameters. No
effects observed in
ophthalmology.
No evidence of sucralose
bioaccumulation in blood.
No differences in fasting or
post-prandial (post-OGTT)
blood glucose, insulin or Cpeptide or in HbA1c
between the placebo and
sucralose at any time point
or over the entire 12-week
period.
Consumption of sucralose
had no effect on blood
glucose or insulin levels. No
evidence of adverse effects.
Demonstrates no
bioaccumulation of
sucralose in blood after
prolonged high doses.
Binns (2003).
Reviewed in
SCF (2000b)
Duration: 12 wk
Subjects
Double-blind, randomized
Healthy males;
placebo-controlled:
Placebo group:
Placebo group: Ingested
n ¼ 23;
333.3 mg cellulose
Ave. wt. 74.07 kg;
3 times/d(1000 mg/d);
Ave. Age: 29.04 yr
Sucralose group: n ¼ 22; Sucralose group: Ingested
333.3 mg sucralose 3 times/
Ave. wt. 75.64;
d (1000 mg/d)
Ave. Age: 28.92 yr
Dose ¼ 13.22 mg/kg/d
Baseline OGTT.
After 12 weeks treatment,
OGTTs performed 15 min after
consuming placebo or
sucralose.
Consumption of 1000 mg
sucralose per day for 12
weeks had no effect on
blood glucose, insulin, cpeptide or HbA1c levels in
healthy adult males.
a
Hematology included: haemoglobin, RBC, PCV, MCV, MCH, MCHC, WBC, ESR and platelets.
Blood biochemistry included: thyroxin, total protein, albumin, globulin, calcium, phosphorus, urea, uric acid, creatinine, total bilirubin, alkaline phosphatase, SGOT, SGPT,
gamma GT, glucose, cholesterol and triglycerides.
c
Urinalysis included pH, ketones, blood, glucose, bilirubin, protein, urobilinogen, specific gravity, white cells, red cells, squamous cells, crystals and organisms.
b
and post-prandial blood glucose, C-peptide, and insulin, and on
HbA1c, the latter being a biomarker that reflects average blood
glucose levels over extended periods of time.
2.6.2. Acute or single administration of sucralose in human subjects
The question of whether sucralose may affect glucose absorption in humans and glycemic response, has been raised based on
reports of sucralose stimulation of secretion of glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide
(GIP) in cell lines (Jang et al., 2007; Margolskee et al., 2007) and
enhancement of expression of GLUT transporter protein and
glucose absorption in animal studies (Mace et al., 2007). GIP and
GLP-1 are gastrointestinal hormones known as “incretin” hormones, as they are released from the gut in response to food
components. The number and variety of compounds that can
induce the release of incretins is growing as research continues, and
include carbohydrates, fats, proteins and some non-nutritive
compounds such as phenolic compounds, dietary fibers and
capsaicin (reviewed in Moran-Ramos et al., 2012). The release of
incretins stimulates synthesis and release of insulin from pancreatic cells, slows gastric emptying and other biological effects
(reviewed in Meier and Nauck, 2005). As a result of observed
possible beneficial effects on appetite, gastric emptying, insulin
secretion, and glucagon suppression GLP-1 administration has been
investigated as a possible treatment for obesity and diabetes (Meier
and Nauck, 2005).
The toxicology animal studies and multiple-dose exposure in
humans studies with sucralose described above do not provide any
suggestion of long-term consequences of possible enhanced
glucose absorption and postprandial blood glucose concentrations
e and in vivo studies in rats reported no adverse effect of sucralose
on blood glucose or incretin levels (Fujita et al., 2009). However,
numerous human clinical studies have been undertaken to specifically investigate a possible effect of acute sucralose administration
on glycemic parameters and incretins. To date, the effects of a single
dose of sucralose on either fasting blood parameters or following an
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
343
Table 14
Summary of studies evaluating effects of repeated daily consumption of sucralose by diabetic subjects.
Reference
Subjects
Dosing
Measurements
Results
Unpublished,
summarized
in SCF
(2000b)
Duration: 6 mo
T2Ds
No further details
reported.
Sucralose: 667 mg/d.
Placebo control
Fasting HbA1c, plasma
glucose, insulin and
serum C-peptide levels
measured every 2
weeks.
Clinical chemistry but
parameters not listed.
Grotz et al.
(2003)
Duration:
13 wk,
Follow-up:
4 wk
T2Ds
Placebo group:
n ¼ 69 (46 M/23 F);
Ave. BMI: 31.6 kg/m2;
Ave. Age: 58.0 yr
Sucralose group:
n ¼ 67 (42 M/25 F);
Ave. BMI: 31.6 kg/m2;
Ave. Age: 57.2 yr
*Approximately 50%
subjects took oral
hypoglycemic agents,
similar distribution
among t groups.
Remaining subjects
took insulin.
Double-blind,
randomized placebocontrolled trial:
Placebo group:
Ingested 2 capsules/
d containing cellulose;
Sucralose group:
Ingested 2 capsules/day
containing sucralose.
Total daily intake was
667 mg.
Follow-up phase:
2 capsules/d containing
cellulose.
Fasting HbA1c, plasma
glucose and serum Cpeptide levels
measured every 2
weeks.
Reyna et al.
(2003)
Duration:
4 weeks
T2D males, 45e55 yr
American Diabetic
Association (ADA) diet
group (control): n ¼ 8;
Ave. BMI: 28.9 ± 2.0 kg/
m2
Modified diet group:
n ¼ 8;
Ave. BMI: 28.5 ± 1.7 kg/
m2
ADA diet group:
Components not
specified.
Modified low-calorie
diet group:
Included bread
prepared with no
added fat, but with 8%
fat replacer with bglucans and oats and
cookies prepared with
50% fat replacer with bglucans, and fructose
and sucralose in a ratio
of 30:70 were used as
sweeteners.
Each patient consumed
two daily bread units of
60 g each and three
cookies (20 g/unit).
Baseline and after 4
weeks:
Body weight, BMI,
blood glucose,
hemoglobin HbA1C,
and lipid profile.
Cause of change in HbA1C
unclear as no changes in other
indicators of glucose
homeostasis. May be placebo
group diabetes better
controlled after start of
experiment. Small sample size
noted by SCF.
Sucralose had no effect on
No significant differences in fasting
glucose homeostasis.
HbA1c, plasma glucose or serum CNo significant differences in
peptide levels between sucralose and
placebo groups over the course of study. adverse events attributable to
treatment.
Sucralose group: Relative to baseline
HbA1c levels decreased at 2, 8, 10, 12, No clinically meaningful
differences between groups in
and 13 wks during, and 2 and 4 wks
after treatment. Overall HbA1c change safety measures or concomitant
medications.
significantly decreased (P ¼ 0.01)
Eight subjects (4 each in the
compared to baseline.
sucralose and placebo groups)
Placebo group: Overall change in
discontinued after
HbA1c from baseline not significantly
randomization to the test
different. 4 wks after treatment
phase, none as a result of an
cessation, significantly lower fasting
adverse event.
plasma glucose for sucralose group
compared to control, and relative to
baseline.
No differences in proportion of subjects
who maintained, increased, or
decreased their insulin or oral
hypoglycemic dose regimen.
No adverse effects attributable
Significant improvements in weight,
to the prescribed sucraloseBMI, lipid profile, basal glucose, and
containing diet, including
HbA1C from baseline after 4 weeks in
effects on measured
both groups.
anthropometric and metabolic
Greater improvement with modified
sucralose-containing diet compared to variables. Effects of sucralose
alone on specific parameters
ADA diet: greater increase in HDL
could not be determined
cholesterol and larger decreases in
because consumption of
HbA1C, body weight, and BMI
sucralose was not the only
difference between the two
diets.
oral glucose tolerance test have been assessed in twelve studies
with healthy subjects (Table 15), and four studies with diabetic or
pre-diabetic obese subjects (Table 16). There are additional reports
in which glycemic parameters in response to various sweetener
treatments have been assessed using sucralose as a negative noncaloric control but as no appropriate no sweetener negative controls that would allow for assessment of the effect of sucralose were
included in the design, these studies are not included in this review
(for example, Kwak et al., 2013; Wu et al., 2013a; Wang et al., 2014).
2.6.2.1. Healthy subjects. Studies of acute administration of sucralose conducted with healthy subjects are summarized in Table 15.
Three studies measured the effects of sucralose on fasting levels of
blood glucose, insulin, and incretin release (Ma et al., 2009; Ford
et al., 2011; Steinert et al., 2011). Ma et al. (2009) measured blood
levels of glucose, insulin, GLP-1 and GIP in subjects following
intragastric infusions of sucrose (50 g), sucralose (80 or 800 mg), or
saline. Unlike the administration of sucrose (positive control),
which led to increases in blood glucose, insulin, GLP-1, and GIP,
Conclusions
No effect of sucralose on blood glucose,
insulin and serum C-peptide levels.
Compared to placebo, HbA1c levels
increased in sucralose group after start
of experiment.
administration of both the sucralose and saline solutions had no
effect on glucose, insulin, GLP-1 or GIP levels. Ford et al. (2011)
evaluated the effects of sucralose ingestion or sucralose shamfeeding, which involved drawing sucralose into the mouth and
spitting it out, in fasted healthy subjects. Neither sucralose
(41.5 mg) ingestion nor sucralose sham-feeding affected glucose,
insulin, GLP-1 or plasma tyrosine tyrosine (PYY) levels or feelings of
appetite for up to 120 min. Steinert et al. (2011) studied the effect of
intragastric infusion of solutions containing water, 62 mg sucralose,
50 g glucose, or 25 g fructose. Compared to the infusion of the
positive control glucose, sucralose administration had no affect on
blood glucose, insulin, GLP-1, PPY or ghrelin levels or appetite.
Ibero-Baraibar et al. (2014) investigated the acute glycemic
response in fasting subjects following consumption of 60 g strawberry jams sweetened with either sugar (41.8 g sugar/100 g) or
sucralose (253e282 mg sucralose/100 g). Sucralose-containing
jams did not affect blood glucose or insulin levels or insulin resistance above baseline levels, whereas both parameters were
significantly increased following consumption of sugar-sweetened
344
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 15
Single dose acute studies in healthy adults.
Reference/
Condition of
study
Subjects
Ma et al. (2009)
Fasting state
Healthy subjects
(n ¼ 7); Sex NR.
Ave. BMI: 21.6 kg/
m2;
Ave. Age: 24 yrs.
Dosing
Measurements
Results
Conclusions
Blood glucose,
plasma levels of
insulin, GLP-1, and
GIP at 0, 5, 15, 30,
60, 90, 120, 150,
180, and 240 min
after infusion.
Gastric emptying
(by breath test).
Sucralose had no effect on blood
glucose, insulin, plasma GLP-1 or GIP
levels when given alone. No effect on
gastric emptying time.
Effects observed only in sucrose group
(positive control).
Sucralose does not acutely
affect glucose homeostasis. No
effect on insulin, GLP-1 or GIP
release or gastric emptying in
health subjects.
Glucose, insulin,
and GLP-1 were
measured for
180 min after the
glucose load.
Preload treatments did not affect peak
concentrations or AUC of glucose and
insulin following the OGTT.
Peak GLP-1 levels following OGTT were
higher in diet cola preload (~5.5 pmol/L)
group than the water preload (~4 pmol/
L) group: same trend with AUC GLP-1.
No adverse effects observed
due to of consumption of
preload of sucralose-sweetened
soda.
An appropriate negative control
was not included. Diet cola
contains other ingredients in
addition to sweeteners not in
the carbonated water control.
Blood glucose and
GLP-1 levels at
t ¼ 30, 15, 0, 5,
10, 20, 30, 40, 60,
90 and 120 min
Serum 3-OMG
levels at T ¼ 0, 5, 10,
20, 30, 40, 60, 90
and 120 min.
Sucralose alone had no effect on glucose
or GLP-1.
Blood glucose and GLP-1 levels
increased in both treatment groups
only after glucose and 3-OMG were
infused.
Blood glucose, 3-OMG and GLP-1 levels
were similar in the saline- and
sucralose-treated groups.
Sucralose did not modify
glucose or GLP-1 levels.
Sucralose did not modify the
rate of glucose absorption or
the glycaemic and incretin
response to glucose in healthy
human subjects.
Blood glucose,
insulin, and
glucagon; plasma
GLP-1, PYY and
ghrelin levels; and
appetite profile:
(hunger, satiety,
fullness); and vital
signs (blood
pressure and heart
rate) and reported
side effects over
120 min.
Ford et al. (2011) Healthy men and Randomized, single-blinded,
Plasma glucose,
Fasting
women (n ¼ 8; 7 crossover, Treatment by ingestion. insulin, GLP-1, PYY
females and 1
levels and appetite
*All subjects were fasted.
male);
ratings (VAS) at -15,
1: 50 ml water
BMI: 18.8
2: 50 ml sucralose solution (0.083% 0, 15, 30, 45, 60, 90
2
e23.9 kg/m ;
and 120 min.
w/v; equivalent to 41.5 mg
For analysis of the
sucralose);
Age: 22e27 yrs
cephalic phase
3: 50 ml maltodextrin (50% w/
insulin and GLP-1,
v) þ sucralose (0.083% w/v)
blood samples also
4: 50 ml water
at 2, 4, 6, 8, 10 min.
Treatments 1e3 followed 1 min
later by modified sham feed (MSF) At 120 min, energy
intake at a buffet
of same solution that was
swallowed. Treatment 4 followed meal was
1 min later by MSF with sucralose measured.
(0.083%; equivalent to 41.5 mg
sucralose).
*MSF protocol involved drawing
solution into mouth and spitting it
out until 200 ml of the solution was
finished.
Blood glucose,
Healthy females Randomized cross-over study:
Brown et al.
insulin, glucagon,
Treatment by ingestion.
(n ¼ 8);
(2011)
TG, and acylated
*All subjects were initially fasted.
Fasting and Post- BMI:
ghrelin at fasting,
1: 355 ml water
prandial
Sucralose (Group 4): g Glucose, insulin,
glucagon, GLP-1, PYY and ghrelin levels
not different vs following infusion of
water No effect on appetite.
No side effects reported.
Glucose (Group 6): blood glucose,
insulin and GLP-1, and PYY levels
increased compared to water. Plasma
ghrelin levels declined then to baseline
by 120 min.
Sucralose alone did not modify
blood glucose, insulin,
glucagon, GLP-1, PYY or ghrelin
levels.
Sucralose did not increase
appetite. No side effects
associated with sucralose
consumption.
Sucralose ingestion (Treatment 2) or
sucralose administered via the MSF
(Treatment 4) had no effect on plasma
glucose and insulin, or GLP-1 or PYY
AUCs compared to water.
Maltodextrin and sucralose (Treatment
3) increased plasma glucose and insulin
AUCs, with no effect on plasma GLP-1 or
PYY AUCs.
There was no difference due to
treatment in appetite scores for up to
120 min, or energy or water intake at
the buffet meal after 120 min.
Sucralose does not increase
plasma glucose, insulin, GLP-1
or PYY nor does it affect
subjective feelings of appetite
or energy intake at the next
meal in healthy volunteers.
Randomized, single-blinded,
crossover: Treatment by
intragastric infusion.
*All subjects were fasted.
1: Saline;
2: 50 g sucrose;
3: 80 mg sucralose;
4: 800 mg sucralose
*Infusates labeled with 150 mg 13Cacetate.
Healthy young
Brown et al.
Randomized cross-over study:
males and
(2009)
Treatment by ingestion.
females (n ¼ 22; *All subjects were initially fasted.
Post-prandial
10 males/12
(OGTT)
1: 240 ml carbonated water;
(Also reported in females);
2: 240 ml diet cola caffeine-free
Ave. BMI:
Brown et al.,
soda (containing sucralose
2
25.6 ± 4.6 kg/m ; (concentration not specified)
2012)
Age: 12e25 yrs. Acesulfame-K and other
ingredients.
*10 min after treatments were
administered, an OGTT with 75 g
glucose was performed.
Ma et al. (2010) Healthy males
Randomized, single-blinded,
and females
Intraduodenal
crossover. Treatment by
(n ¼ 10; 8M and intraduodenal infusion.
infusion
2F)
*All subjects were initially fasted.
Ave.
1: Saline;
BMI ¼ 23.4 kg/m2 2: 4mM sucralose (total 960 mg
Ave. Age ¼ 27 yrs infused over 150 min).
*30 min after start of control or
sucralose infusion glucose and its
non-metabolized analogue 3-Omethylglucose (3-OMG) infused
intraduodenally.
Healthy males
Steinert et al.
Double-blind, placebo-controlled,
and females
(2011)
crossover study: Treatment by
(n ¼ 12; 6 males Intragastric infusion;
Fasting
and 6 females)
*All subjects were fasted.
Ave. BMI:
1: water
2
23 ± 0.5 kg/m
2: 169 mg aspartame in water;
3: 220 mg Acesulfame K in water;
Ave. Age:
4: 62 mg sucralose in water;
23.3 ± 0.7 yrs
5: 25 g fructose in water;
6: 50 g glucose in water.
No significant interaction effect of retail Both fasting and post-prandial
data support that sucralose has
formulation of sucralose (“Splenda”)
treatment vs. sucrose vs. time point on effects similar to water with
glucose, insulin, glucagon, TG, and
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
345
Table 15 (continued )
Reference/
Condition of
study
(Standardized
Breakfast)
Wu et al. (2012)
Meal
Subjects
Dosing
22.1 ± 1.7 kg/m2; 2. 50 g sucrose in 355 ml water
Age: 21.8 ± 2.5 yr 3. 6 g Splenda in 355 ml water
4. 50 g sucrose and 6 g Splenda in
355 ml water.
*60 min after preload, breakfast
consisting of eggs cheese, orange
juice, and toast was administered.
Measurements
30 and 60 min after
preload, and 30, 60,
90 and 120 min
after breakfast.
Hunger, tiredness,
gastrointestinal
well-being, and
overall well-being
by VAS
Healthy men and Randomized, single-blinded:
Blood glucose and
women (n ¼ 10; 7 *All subjects were fasted. Treatment serum insulin;
men/3 women)
plasma GLP-1 and
by ingestion of drink containing:
BMI:
GIP at t ¼ 25
1: 40 g glucose;
25.5 ± 1.5 kg/m2 2: 40 g 3-OMG;
(before
preload), 10, 0, 15,
Ave. Age:
3: 40 g tagatose/isomalt mixture
30, 60, 90, 120 and
28.8 ± 4.0 yrs
(TIM)
240 min.
Duration: prior to 4: 60 mg sucralose.
Gastrointestinal
Fifteen minutes later (t ¼ 5 to
and over the
0 min), all subjects ingested a meal sensations hunger,
course of
fullness, desire to
consisting of 65 g powdered
240 min.
eat, and projected
potatoes and 20 g glucose, with
consumption) by
200 mL water and egg yolk
containing 100 mL 13C octanoic acid. VAS.
Wu et al. (2013b)
Fasting and Postprandial
(OGTT)
Healthy males
(n ¼ 10)
Ave. BMI:
25.5 ± 1.0 kg/m2
Ave. Age:
33.6 ± 5.9 yrs
Stellingwerff
et al. (2013)
Pre- and postprandial
(Maltodextrin
drink) and in
absence and
presence of
exercise
Healthy males
(n ¼ 23);
Ave. BMI:
23.1 ± 1.9 kg/m2;
Ave. Age: 29 ± 7
yrs.
Duration: prior
to, during, and
after 2 h exercise.
Temizkan et al.
(2015)
Post-prandial
(OGTT)
*This publication
described two
studies, The
healthy subject
study is
described here.
Healthy males
and females
(n ¼ 8; 4 females/
4 males)
Ave. BMI:
30.3 ± 4.5 kg/m2
Ave. Age
45.0 ± 4.1 yrs.
Ibero-Baraibar
et al. (2014)
Post-prandial
Healthy males
and females
(n ¼ 16; 10
females/6 males)
Results
Conclusions
acylated ghrelin.
Significant differences due to sucrose
only.
No differences due to treatment on
hunger or well-being.
regard to glycemic and appetite
regulation.
Sucralose and TIM had no effect on
blood glucose, insulin, GLP-1 or GIP
levels after preload, prior to ingestion of
the meal. Both glucose and 3-OMG
preloads stimulated GLP-1 and GIP.
Postprandial blood glucose AUC were
greater for glucose than sucralose, less
for 3-OMG than sucralose. There were
no differences among sucralose, 3-OMG
or TIM in postprandial GLP-1 or insulin
responses, all lower than glucose.
Postprandial GIP was greater after
glucose and 3OMG, than after sucralose
and TIM.
No differences in hunger, desire to eat,
or prospective consumption between
any of the preloads.
Prior to OGTT: Blood glucose, plasma
Blood glucose,
Randomized, single binded,
insulin and GLP-1 levels did not change
insulin, GLP-1
crossover study; Treatment by
after either water or sucralose or
injection. *All subjects were initially concentrations
prior to OGTT and Acesulfame-K sweetened drinks.
fasted
after, over 240 min. After OGTT: Blood glucose, plasma
1: water
insulin and GLP-1 levels increased and
Gastric emptying
2: 52 mg sucralose
were similar across all the treatment
over 240 min.
3: 200 mg Acesulfame-K
groups.
4: 46 mg sucralose þ26 mg
No effects observed on gastric
Acesulfame-K,
emptying.
*10 min after treatments were
administered, an OGTT with 75 g
glucose and 150 mg 13C-acetate was
performed.
Consumption of sucralose had no effect
Randomized, double-blind, cross- Plasma glucose,
insulin, and lactate on plasma glucose, insulin, or lactate
over study: All subjects were
initially fasted. Treatment as below. levels at t ¼ 120 levels throughout the study.
No difference in carbohydrate or fat
(baseline0, 0, 10,
1 (Placebo): Ingested 50 ml of
substrate utilization, heart rate,
20, 30, 40, 50, 60,
sucralose mouthwash and spit it
perceived exertion, and gastrointestinal
out followed by a ingestion 50 ml of 80, 100 and
symptoms between placebo and
120 min.
water
Carbohydrate or fat sucralose group.
2: Ingested 50 ml of solution
containing 1 mM sucralose (20 mg) substrate
utilization, heart
every 15 min for 2 h.
rate, perceived
All subjects were given a
exertion, and
maltodextrin drink over 2 h
exercise period. 34 g at the onset of gastrointestinal
symptoms.
exercise; 10 g every 10 min until
the end of exercise.
Sucralose tabletop: blood glucose level
Blood glucose,
Randomized, single-blinded
randomized cross over: All subjects insulin, C-peptide, peaks (~7.3 mmol/L) similar to water
were initially fasted. Treatment by GLP-1 levels taken group, peak occurred earlier than water
at 15, 0, 15, 30, 45, (30e45 vs 60 min).
ingestion.
60, 75, 90, 105 and Total AUC of glucose was significantly
1: 200 ml water
lower compared to water.
120 min.
2: 24 mg sucralose (in tabletop
There were no differences in insulin,
formulation) in 200 ml water
GLP-1 and C-peptide levels between the
3: 72 mg aspartame (in tabletop
water and sucralose tabletop groups
formulation) in 200 ml water
during the OGTT.
*15 min after treatments were
AUC for insulin and C-peptide were
administered, an OGTT with 75 g
similar. AUC for GLP-1 was higher in the
glucose was performed.
sucralose tabletop-treated group
compared to water-treated group.
Blood samples were Blood glucose concentrations were
Randomized, crossover, doublecollected at fasting maintained at normal values (<100 mg/
blind study with three arms. All
dl) and without peaks for 2 h after
and at 30, 60, 90,
subjects were initially fasted.*
consumption of sucralose jams, but
Subjects consumed 60 g of different and
Sucralose had no effect on
blood glucose, insulin, GLP-1 or
GIP levels after preload, prior to
ingestion of the meal.
No evidence of significant effect
of sucralose on postprandial
glycemic responses but lack of a
negative control preload (e.g.
water) limits interpretation.
No effect on appetite.
Sucralose did not affect pre- or
post-prandial glucose, insulin
or GLP-1 levels, and had no
effect on gastric emptying.
Consumption of sucralose in the
immediate period prior to
exercise had no effect on
glucose, insulin or carbohydrate
or fat substrate utilization
during exercise. Results
additionally suggest no effect of
sucralose on glucose
transporter function or density.
Sucralose consumed in a
tabletop sweetener formulation
was associated with a
decreased blood glucose and
increased GLP-1 response
(AUC). No effect on blood
insulin, or C-peptide levels.
Carrier ingredients in tabletop
sweeteners not included in
water control.
Sucralose-containing jams did
not affect blood glucose or
insulin levels or insulin
resistance in an acute study in
(continued on next page)
346
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 15 (continued )
Subjects
Dosing
(Strawberry
Jam)
Ave. BMI:
23.99 ± 3.05 kg/
m2
Ave. age
25.9 ± 3.0 yr)
strawberry jams containing
1) Control jam:
(41.8 g sugar, 480 mmol catechin/
100 g).
2) Low sugar, low polyphenol jam
(2.6 g sugar; 370 mmol l
catechin, 253 mg sucralose/100 g or
152 mg sucralose)
3) 60 g Low sugar, high polyphenol
jam
(2.7 g sugar, 849 mmol catechin,
282 mg sucralose/100 g or 169 mg
sucralose)
*Different strawberry jams given in
randomized order on days 0, 7, and
14.
Sylvetsky et al.
(2016)
Post-prandial
(OGTT)
Study arm 1:
Healthy males
and females
(n ¼ 30; 16
females/14
males)
Ave. BMI:
25.8 ± 4.2 kg/m2
Ave. age
29.7 ± 7.6 yr)
Study arm 2:
Healthy males
and females
(n ¼ 31; 17
females/14
males)
Ave. BMI:
26.3 ± 7.5 kg/m2
Ave. age
27.4 ± 6.7 yr)
Reference/
Condition of
study
Measurements
120 min after jam
ingestion.
Blood glucose,
insulin levels, lipid
metabolism,a uric
acid, and oxidative
stress profiles.b
Insulin resistance
was calculated by
the homeostasis
model assessmentestimated insulin
resistance (HOMAIR) index.
Appetite scores
(VAS) before and
30, 60, 90, and
120 min.
Blood pressure,
anthropometry,
and body
composition before
and end of trial.
Study arms 1 and
Randomized cross-over study:
2:
Treatment by ingestion.
Blood samples at
*All subjects were initially fasted.
t ¼ 10, 0, 10. 20
Study arm 1:
30, 60, 90, and
1: 355 ml water
2. 68 mg sucrose in 355 ml water 120 min.
3. 170 mg sucrose in 355 ml water Concentrations of
4. 250 mg sucrose in 355 ml water. glucose, insulin, Cpeptide, GLP-1, 3Study arm 2:
OMG (proxy for
1: 355 ml carbonated water;
intestinal glucose
2: 355 ml diet cola caffeine-free
soda (containing 68 mg sucralose absorption),
acetaminophen
41, mg Acesulfame-K and other
(proxy for gastric
ingredients.
3: 355 ml diet lemon-lime flavored emptying)
measured in blood
soda containing 18 mg sucralose,
samples.
18 mg Acesulfame-K, 57 mg
Hunger and satiety
aspartame and other ingredients.
ratings.
4. 355 ml carbonated water with
68 mg sucralose and 41 mg
Acesulfame-K
*10 min after treatments were
administered, an OGTT with 75 g
glucose, 1450 mg acetaminophen,
and 7.5 3OMG performed.
Results
Conclusions
increased as expected with high sugar
jam.
No change in insulin or HOMA-IR index
after ingestion of sucralose jams; but
increased as expected with high sugar
jam. Lower plasma levels of free fatty
acids were observed following high
sugar jam, likely due to insulin release.
No differences in lipid metabolism or
oxidative stress measures due to jam
type.
Sucralose-containing jams did not
affect satiety scores for up to 2 h.
healthy adults.
Sucralose-containing jam has
no effect on appetite or any
other parameter assessed.
Study arm 1:
Sucralose preload, regardless of dose,
had no effect on measured parameters blood glucose, insulin, C-peptide and
GLP-1 peaks and AUCs, glucose
absorption, gastric emptying and
hunger and satiety measures.
Study arm 2:
Carbonated water containing the same
sweetener concentrations as sodas had
no effect on parameters as compared to
carbonated water alone. The only
statistically significant difference
observed was a higher GLP-1 AUC
following the OGTT with diet cola
preload compared to carbonated water.
Study arm 1:
Sucralose preload up to 250 mg
has no effect on metabolic
outcomes following an OGTT in
healthy adults.
Study arm 2:
A preload of sucralose in
combination with other low
calorie sweeteners has no effect
on metabolic outcomes
following an OGTT in healthy
adults.
An observed increase in GLP-1
AUC with diet cola was not
replicated in control
carbonated drink with
sucralose. Other ingredients in
diet cola may be responsible for
differences in GLP-1 AUC
observed in this and previous
studies with diet cola.
a
Total cholesterol, HDL-c, LDL-c, TG.
Malondialdehyde (MDA), glutathione peroxidase (GPx), and total antioxidant capacity (TAC).Area Under the Curve (AUC); Glucagon-like peptide 1 (GLP-1); glucosedependent insulinotropic polypeptide (GIP); Modified sham feeding (MSF); 3-O-methylglucose (3-OMG); Oral glucose tolerance test (OGTT); Plasma Tyrosine Tyrosine
(PYY); Triglycerides (TG); Visual Analogue Scale (VAS).
b
jams. Taken together, these studies demonstrate that sucralose,
when administered in the absence of glucose or sucrose has no
effect on blood glucose, insulin or incretin release in healthy
subjects.
Ten studies in healthy subjects measured the effects of sucralose
on blood glucose, insulin and incretin levels following the administration of a carbohydrate bolus delivered either as an intraduodenal glucose infusion, oral glucose tolerance test (OGTT), meal,
or a bolus of maltodextrin (Ma et al., 2010; Brown et al., 2009, 2011,
2012; Temizkan et al., 2015; Stellingwerff et al., 2013; Sylvetsky
et al., 2016; Wu et al., 2012, 2013a). Ma et al. (2010) infused
either saline or 960 mg sucralose intraduodenally 30 min prior to
the co-administration of a glucose solution (25%) mixed with a
non-metabolizable form of glucose, 3-O-methylglucose (3-OMG).
Glucose, GLP-1 and 3-OMG were elevated after the infusion of
glucose, irrespective of treatment. There were no significant differences in blood glucose, GLP-1 or serum 3-OMG concentrations
between the sucralose and saline infusions. Wu et al. (2012) reported no effect of a sucralose preload on blood glucose, insulin,
GLP-1 or GIP levels prior to ingestion of a meal. Postprandial insulin
and GLP-1 levels were similar to those following preload of nonmetabolizable 3-OMG; blood glucose levels with sucralose were
lower than 3-OMG, which stimulated GLP-1 and GIP. The lack of a
water-only negative control in this study limits postprandial
interpretation of sucralose specific effects. Brown et al. (2011)
assessed glucose, insulin, glucagon and acylated ghrelin levels
due to administration of sucralose-containing tabletop sweetener,
water or sucrose prior to breakfast and found no effects due to
sucralose tabletop sweetener consumption. Stellingwerff et al.
(2013) evaluated the effects of water and sucralose (20 mg)
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
ingestion on maltodextrin-induced glucose and insulin responses
during exercise. Plasma glucose and insulin levels were similar
following the water and sucralose treatments. No differences were
observed in carbohydrate or fat utilization between treatments.
Temizkan et al. (2015) and Wu et al. (2013a) assessed the effects of
oral administration of sucralose on the glycemic response following
an OGTT. Wu et al. (2013a) reported that there were no differences
in OGTT-induced blood glucose, serum insulin or GLP-1 responses
between the water and sucralose treatments. In contrast, Temizkan
et al. (2015) reported that consumption of a sucralose-sweetened
drink (containing 24 mg of sucralose) 15 min before an OGTT
resulted in a significant decrease in blood glucose area-under-thecurve (AUC), with an earlier time-to-peak blood glucose level
compared to the water control in healthy subjects. There were no
differences at any time point in insulin, GLP-1 and C-peptide levels
between the water and sucralose treatments; however, ingestion of
the sucralose solution was associated with a significant increase in
the plasma GLP-1 AUC over 120 min (Temizkan et al., 2015).
In two studies (Brown et al., 2009, 2012), the effect of consumption of either carbonated water or a commercially available
diet cola sweetened with sucralose and Acesulfame K followed by
an OGTT on glucose, insulin, C-peptide, GLP-1 and GIP and PYY
levels were assessed. No effects were observed on glucose, insulin,
C-peptide, GIP or PYY. Peak GLP-1 levels and AUC GLP-1 following
OGTT were higher in diet cola preload group than the water preload
(Brown et al., 2009); (diet cola ~5.5 pmol/L versus water ~4 pmol/L).
The physiological significance of this difference is not clear as both
values are within the normal fasting range and somewhat low for
postprandial conditions. In a review by Meier and Nauck (2005),
typical concentration ranges reported for GLP-1 and GIP in normal
healthy humans are 5e10 pmol GLP-1/l in fasting state, with peaks
of 15e 50 pmol GLP-1/L following consumption of carbohydrate
loads, and 9 ± 2 pmol GIP/l fasting and 100 ± 11 pmol GIP/l in
postprandial conditions. Furthermore, the potential effect of other
ingredients in the diet cola that were not present in the carbonated
water negative control limit the interpretation to assigning the
observation as due to sucralose alone.
Recently, Sylvetsky et al., 2016 reported the results of two
studies in healthy subjects; one with sucralose added to water and
one with diet sodas similar to the studies previously reported by
Brown et al. (2009, 2012). In this study, preloads of 68, 170 and
250 mg sucralose added to water had no effect on glycemic and
incretin parameters measured following an OGTT, including blood
glucose, insulin, C-peptide and GLP-1 peaks and AUCs, glucose
absorption, gastric emptying and hunger and satiety measures. The
second study was very similar in design to previous studies with
diet sodas, with the advantage that additional negative controls
were included, improving interpretation of the results. Subjects
consumed a 355 ml preload of carbonated water, carbonated water
with 68 mg sucralose and 41 mg Acesulfame-K; diet cola caffeinefree soda (containing 68 mg sucralose 41, mg Acesulfame-K and
other ingredients) or a diet lemon-lime flavored soda containing
18 mg sucralose, 18 mg Acesulfame-K, 57 mg aspartame and other
ingredients followed by an OGTT. The only statistically significant
difference observed was a higher GLP-1 AUC following the OGTT
with diet cola preload compared to carbonated water. This study
clearly demonstrates that sucralose, either alone or in combination
with other low calorie sweeteners, has no acute effect on glycemic
and incretin parameters, glucose absorption, gastric emptying or
hunger and satiety ratings.
In summary, numerous studies found no acute effect of consumption of sucralose prior to or in combination with carbohydrates on blood glucose, insulin or GLP-1 and GIP levels in healthy
human subjects. The inconsistent observation of a higher GLP-1
level or GLP-1 AUC in association with a sucralose exposure is
347
likely a reflection of the variability of experimental conditions and
possible influence of the many other factors that can modify
incretin response.
2.6.2.2. Diabetic and prediabetic obese subjects. Four studies
assessed the acute effect of sucralose on glucose, insulin and
incretin release in diabetic or prediabetic obese subjects (Mezitis
et al., 1996; Brown et al., 2012; Pepino et al., 2013; Temizkan
et al., 2015). All studies administered sucralose prior to a meal or
OGTT.
In the first study, Mezitis et al. (1996) examined the effect of
acute administration of sucralose on plasma glucose and serum Cpeptide in diabetic subjects. This was a randomized double-blind
cross-over design. Participants included male and female insulindependent (n ¼ 13) and non-insulin dependent (n ¼ 13) diabetics. Adequate control of diabetes prior to the sucralose test was
confirmed by glycosylated hemoglobin levels of <10% and fasting
glucose levels of <9.7 mmol/l during the prescreening phase.
Following an overnight fast, participants consumed 3 opaque tablets containing either 1000 mg sucralose (10e13 mg/kg bw) or
cellulose (placebo), and then consumed a 360-kcal liquid breakfast.
Consumption of this high dose of sucralose had no significant effect
on glucose and serum C-peptide AUC measured over the next 4 h in
either group as compared to placebo.
Consistent with results of Mezitis et al. (1996), Temizkan et al.
(2015) also found that administration of sucralose (24 mg) to T2D
participants had no effect on blood glucose, insulin and C-peptide
levels during the OGTTs. Temizkan et al. (2015) also showed that
sucralose had no effect on incretin hormone release (GLP-1, GIP and
PYY) during the OGTT in T2D.
Brown et al. (2012) administered carbonated water or diet cola
sweetened with approximately 46 mg of sucralose and 26 mg
acesulfame K, 10 min prior to an OGTT in both T1D subjects (n ¼ 9)
and T2D subjects (n ¼ 10). Compared to carbonated water, sucralose (combined with Acesulfame-K) had no effect on blood glucose,
C-peptide, GIP or PYY levels during the OGTT in both groups. In T1D,
but not T2D subjects, GLP-1 AUC was higher when diet cola was
administered prior to the OGTT as compared to when carbonated
water was administered. GLP-1 levels were not reported. As
mentioned above, the inclusion of other ingredients in addition to
sucralose present in the diet cola and lack of appropriate negative
control for sucralose only (i.e., cola soda devoid of sucralose) limits
interpretation specific for sucralose.
One study evaluated the effects of sucralose in 17 morbidlyobese, insulin-sensitive subjects (BMI>42) who were considered
non-nutritive sweetener (NNS)-naïve because they consumed less
than one can of a diet beverage or one spoonful of NNSs per day
(Pepino et al., 2013). Blood glucose measurements taken during the
study indicate that the subjects were pre-diabetic, based on criteria
published by the American Diabetes Association (2014). On
different days, water (50 ml) with 0 or 48 mg sucralose was
administered prior to an OGTT. Blood levels of glucose, insulin, Cpeptide, GIP or GLP-1 were evaluated over 5 h (300 min). OGTT
results following ingestion of the sucralose-containing drink
showed a small, but statistically significant increase in the peak
blood glucose, plasma insulin and C-peptide levels post-OGTT, and
insulin AUC compared to the results following ingestion of only
water. The clinical significance of the reported differences is
questionable. The mean peak blood glucose level found in the OGTT
done post-consumption of the sucralose drink was within the
normal range for this test (ADA, 2014). There was no betweengroup difference in the blood glucose AUC, which is not consistent with a true effect on insulin sensitivity or secretion. Differences
reported for the other measures were small and likely result from a
small study in pre-diabetic subjects, where minor changes in
348
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Table 16
Summary of studies assessing the effects of a single acute dose of sucralose in diabetic and prediabetic obese adults.
Mezitis et al. (1996)
Post-prandial
(standardized
meal)
T1D (n ¼ 13; 8M, 5F); BMI:
32.0 ± 1.9 kg/m2; Age:
37.8 ± 2.6 yrs.
T2D (n ¼ 13; 8M, 5F); BMI:
23.7 ± 0.9 kg/m2; Age:
54.3 ± 1.7 yrs.
*9 T2D patients treated
with sulfonylureas,
2 with diet alone,
2 with insulin
Brown et al. (2012)
Post-prandial
(OGTT)
*This publication
described
three studies.
Only studies
in T1Ds and T2Ds
are
described here.
Type 1 Diabetics (T1Ds;
n ¼ 9);
Ave. BMI: 21.7 ± 2.4 kg/m2;
Age: 13e24 yrs
Type 2 Diabetics (T2Ds;
n ¼ 10);
Ave. BMI: 35.0 ± 6.8 kg/m2;
Age: 13e24 yrs
Temizkan et al.
(2015)
Post-prandial
(OGTT)
*This publication
described
two studies. Only
T2D
study is
described here.
Newly diagnosed T2Ds
(n ¼ 8; 4 males/4 females)
Ave. BMI: 33.7 ± 5.4 kg/m2
Ave. Age: 51.5 ± 9.2 yrs
Pepino et al. (2013)
Post-prandial
(OGTT)
Obese males and females
(n ¼ 17; 15 females/2
males);
Ave. BMI: 41.0 ± 1.5 kg/m2;
Ave. Age: 35.1 ± 1.0.
Randomized, double-blind
crossover, single acute dose.
*All subjects were initially
fasted.
Control: Ingested cellulose;
Sucralose: Ingested 1000 mg
sucralose.
*Treatments were immediately
followed by standardized 360kcal liquid breakfast.
Patients received usual insulin
or sulfonylurea dose 30 min
before consuming the
treatments.
Randomized cross-over study:
Treatment by ingestion. All
subjects were initially fasted.*
1: (Placebo): 240 ml carbonated
water;
2: (Diet Soda): 240 ml diet soda
containing sucralose and
Acesulfame-K
(190 ± 38 mg/ml and
108 ± 0.6 mg/ml, respectively)
*10 min after treatments
administered, OGTT with 75 g
glucose performed.
Randomized, single-blinded
randomized cross over:
Treatment by ingestion. All
subjects were initially fasted;
1: 200 ml water;
2: 24 mg sucralose (tabletop
formulation) in 200 ml water;
3: 72 mg aspartame (tabletop
formulation) in 200 ml water.
*15 min after treatments
administered, OGTT with 75 g
glucose performed.
Randomized crossover:
All treatments by ingestion
*All subjects were initially
fasted.
1: 60 ml distilled water;
2: 60 ml sucralose solution
(2 mmol/L; 48 mg sucralose)
*After treatments were
administered, an OGTT with
75 g glucose was performed.
Blood glucose and serum Cpeptide before and at 30, 60
90, 120, 180 and 240 min
after the breakfast.
Blood glucose and serum Cpeptide AUC change from
baseline.
No difference in glucose
and serum C-peptide
response to sucralose
vs. water in T2Ds and
T1Ds ingesting
sucralose or placebo.
Sucralose does not
adversely affect blood
glucose control in T1D
or T2D subjects.
Glucose, C-peptide, GLP-1,
GIP, and PYY at t ¼ 10, -5,
0, 5, 10, 15, 20, 25, 30, 45,
60, 90, 120, 150, and
180 min.
Glucose, C-peptide, GIP,
and PYY AUC were not
statistically different
between the two
conditions in both
groups
GLP-1 AUC was higher
after preload of diet
cola vs. carbonated
water in individuals
with T1D but no
difference observed in
individuals T2D.
Blood glucose, insulin,
C-peptide, and GLP-1
AUCs were similar
following treatment
with water and
sucralose.
No adverse effects on
glucose, insulin, GIP or
PYY due to
consumption of preload
of sucralose-sweetened
soda in individuals with
T1D or T2D. Higher
GLP-1 in subjects with
T1D. Other ingredients
in diet cola may be
responsible for
observed difference.
Blood glucose and Cpeptide AUCs in the
water- and sucralose
treated groups were
similar; however, peak
values were higher
following the OGTT
with a sucralose
preload than observed
with water preload.
Peak insulin and insulin
AUC was higher
following the OGTT in
the sucralose-preload
group compared to the
water-controls.
Glucagon, GLP-1, and
GIP levels and AUCs
were similar in the
water- and sucralose
treated groups.
Compared to water
preload, higher peak
glucose, insulin and Cpeptide levels following
OGTT with sucralose
compared to water
preload in obese
individuals. Absence of
effect on overall
glucose AUC. Peak
glucose levels within
normal range and
insulin changes were
small. With both
preloads, blood glucose
following OGTT, above
7.8 mmol/l indicating
subjects pre-diabetic.
No differences in
glucagon, GLP-1, and
GIP.
Blood glucose, insulin, Cpeptide, GLP-1 levels
at 15, 0, 15, 30, 45, 60, 75,
90, 105 and 120 min.
Plasma glucose, insulin, Cpeptide, glucagon, GIP, GLP1 concentrations
at 20, 15, 10, 6,2,
10, 20, 30, 40, 60, 90, 120,
150, 180, 240, and 300 min.
No effects of sucralose
tabletop sweetener on
glucose, insulin, Cpeptide or GLP-1 in
individuals with T2D.
Carrier ingredients in
tabletop sweeteners
not included in water
control.
Area Under the Curve (AUC); Glucagon-like peptide 1 (GLP-1); glucose-dependent insulinotropic polypeptide (GIP); Oral glucose tolerance test (OGTT); Plasma Tyrosine
Tyrosine (PYY); Type 1 Diabetic (T1D); Type 2 Diabetic (T2D).
insulin secretion can result from many factors. There also were no
significant treatment group differences in serum GIP and GLP-1
levels.
In summary, the totality of evidence from acute clinical trials
further supports that sucralose has no effect on blood glucose
control in either normoglycemic or hyperglycemic individuals. Only
a few studies have reported that sucralose increases blood levels of
GLP-1 or GLP-1 AUC following the administration of a glucose load
and others report no effect of sucralose on GLP-1. It should also be
noted that the parameters of these studies were highly variable
with differences in the composition of test meals, the time between
the preload and the meal, the studied subject groups (age, sex,
healthy, diabetic or obese subjects) and the preload form (solid or
liquid). Although there are both in vitro and in vivo reports that
sucralose may induce GLP-1 release by activating sweet receptors
in the gut (Brown et al., 2012 review), other in vivo studies in mice
(Fujita et al., 2009) and human subject trials described above have
failed to support this hypothesis. As has been previously discussed
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
(Ma et al., 2010), species differences in expression of transporter
proteins may account for the lack of effect of sucralose on glucose
absorption in human subjects. Collectively, the available research
from the acute human clinical studies in healthy and diabetic
subjects does not indicate an adverse effect of sucralose on blood
glucose control. This is also the conclusion of a recent review of
studies by the French Agency for Food, Environmental and Occupational Health & Safety (ANSES, 2015). Furthermore, this is
consistent with the lack of evidence of adverse effects on glycemic
responses in long-term studies of sucralose administration in both
normal, healthy individuals and persons with diabetes.
2.7. Case reports
Three case reports have appeared in the literature that suggested sucralose may be associated with the onset of migraine
headaches in female adults (Patel et al., 2006; Bigal and
Krymchantowski, 2006; Hirsch, 2007), each involving only one
individual. These observations have not been followed up with a
hypothesis-based, properly controlled clinical trial. The lack of
biological plausibility for such an effect was discussed in a subsequent letter to the editor (Grotz, 2008).
3. Estimates of consumption of sucralose
The consumption of any food ingredient, including essential
nutrients such vitamins and minerals, will result in toxicity at some
level of consumption, either due to toxicity of the compound itself
or due to displacement or dilution of other nutrients in the diet.
Thus, to assess the safety of use of any food ingredient, the expected
amount of the ingredient that will be consumed must be determined. This process can range from use of very simple general estimates that are highly conservative and overestimate intakes
requiring few resources, to very refined estimates for specific
unique populations requiring much greater time and resources
(reviewed in Kroes et al., 2002). In most cases, a tiered approach is
used to identify when more refined data are needed (EFSA, 2012a).
The first tier uses theoretical food consumption data combined
with the Maximum Permitted Levels (MPLs) of the ingredient; the
second tier utilizes survey data of food consumption combined
with MPL; and the third tier refines estimates further by using data
on the actual levels of the ingredient in the foods.
Interpretation of exposure estimates must consider the data
used to generate the estimates, recognizing that the goal of the first
tiers is to optimize use of available resources and to provide ‘‘worst
case’’ estimates; thus if the first tiers do not suggest a toxicologically significant exposure (i.e. exceeding the ADI), more sophisticated estimates may not be conducted although it is recognized
that exposure is likely overestimated. An example of a first tier
intake assessment based on Maximum Permitted Levels (MPLs) or
on maximum actual usage data is the Food Additives Intake Model
(FAIM) developed by the European Food Safety Authority (EFSA,
2012b; 2013).
As discussed by Kettler et al. (2015), there are many sources of
uncertainties that are inherent in all exposure assessments,
including “refined Tier 3” estimates. Uncertainties include the estimates of population food consumption using different methodologies and databases, population characteristics (age, body
weight), estimates of the concentration of the food ingredient in
the food or beverages, predictions of brand specific differences and
market share, and uncertainties in the various models used to link
food concentration data to the food consumption data. Several
studies described below illustrate the difference in the estimates of
exposure generated depending on the method and assumptions
used in the analysis.
349
Renwick (2006) provided an excellent overview of the methods
used to estimate low calorie sweetener intake, and the results of
previously published exposure assessments of several low calorie
sweeteners. For sucralose with an ADI of 0e15 mg/kg bw/d, the
estimated exposure ranged from 1% to 3% of the ADI for average
consumers, and 6%e15% of the ADI for high consumers in the 95%
percentile of intakes. The highest reported intake for sucralose
(2.25 mg/kg/d) was reported in a food diary study by the Food
Standards Australia New Zealand in 2004 (FSANZ, 2004).
Ng et al. (2012) reported the frequency of the presence of caloric
and non-caloric sweeteners on food labels of 85,451 consumer
products in the US. As expected, most products are sweetened with
caloric sweeteners, with only 1% sweetened with non-caloric
sweeteners. The fraction of this 1% that contained sucralose was
not reported. No estimates of sucralose consumption were
reported.
In a study of the Belgian population older than 15 years
(Huvaere et al., 2012), two approaches were used to estimate low
calorie sweetener intake. In the first approach (Tier 2) national food
consumption data were used in combination with the MPLs of
sweeteners in food and beverages. Tier 3 utilized the actual
measured concentrations of low calorie sweeteners, rather than the
theoretical maximum level. The highest users were in the 95th
percentile intake of the diabetic subset of the survey population,
with an intake of 10% of the ADI, or 1.53 mg sucralose/kg/d using a
Tier 3 refinement (Huvaere et al., 2012).
Although no specific values for intake were reported, the French
Food Authority (ANSES, 2015) estimated exposure to sucralose
based on use levels and food consumption surveys, and concluded
that sucralose consumption is below the ADI for all consumers.
The consumption of sucralose by Korean consumers was estimated by Ha et al. (2013) using 24-h recall food consumption data
from the 2009 Korean National Health and Nutrition Survey (2009)
for 8081 consumers aged 1e65 years. In the refined estimates, the
concentration of low calorie sweeteners (sucralose and acesulfame
K) was determined for 605 food samples. Estimated daily intakes
(EDI; mg/kg bw/d) were calculated using concentrations of sucralose in the foods, reported food consumption and body weights of
consumers. In the first scenario assuming that consumers
randomly chose food products with and without sweeteners, the
average intake for sucralose was 1.2 mg/kg/d and high consumer
intake was 3.4 mg/kg/d. In the second scenario designed to estimate consumption of brand-loyal consumers selectively choosing
foods sweetened with sucralose, the average and high consumer
intakes were 6.5 mg/kg/d and 17.7 mg/kg/d, respectively. The large
difference in the two scenarios is the result of high concentrations
of sucralose in a few specific products, representing a worst-case
estimate for high consumers. As the food category contributing
most to the EDI of sucralose was soju, an alcoholic beverage popular
in Korea, these data may not be relevant to populations not
consuming this beverage.
Dietary exposure to 70 food additives, including sucralose, was
determined for adults in Belgium using a FAIM followed by a Tier 2
assessment (Van Loco et al., 2015). The FAIM estimates were based
on theoretical Belgian consumption data for 19 food groups (such
as dairy products) and the MPL for sucralose in those food groups.
The resulting intake estimate was unrealistically high, at 1475 mg/
kg/d for the average consumer and nearly the same for the 95th
percentile consumer. In contrast, the Tier 2 estimates, which also
used MPLs but were combined with actual consumption data from
the Belgian Food Consumption Survey, were 0.8 mg/kg/d for
average and 3.1 mg/kg/d for high consumers. This example clearly
illustrates the extreme overestimation by the Tier 1 FAIM. A further
refinement in intake estimates, utilizing actual sucralose concentration data for food products would have further reduced and
350
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
more closely estimated actual intakes as illustrated by (Huvaere
et al., 2012), but was not needed as the Tier 2 assessment determined that the exposures to sucralose are well below the ADI.
Three recent studies (Martyn et al., 2016; O'Sullivan et al., 2016;
Dewinter et al., 2016) have reported estimates of exposure of low
calorie sweeteners in children, and specifically children with special dietary requirements. Dietary intakes of specific food ingredients are often highest in children due to their high food intake
per kg body weight to support growth and distinct food intake
patterns such that specific foods may represent a higher portion of
their diet.
Martyn et al. (2016) estimated dietary intake of 4 sweeteners
including sucralose in Irish children aged 1e4 years using food
consumption data from the Irish National Pre-school Nutrition
Survey (2010-11) and analytical data for sweetener concentration
in foods and beverages. Four different methods to assess intake
were used, and all found that the average intake of sucralose was
below the ADI. The most realistic estimates generated using
sweetener occurrence and concentration data, were 0.65 mg and
1.97 mg sucralose/kg/d for average and high consumer respectively; well below the ADI. The authors concluded that there is no
health risk to Irish pre-school children at current dietary intake
levels of sucralose (Martyn et al., 2016).
Children with specific dietary needs that may affect intake
include those with allergies or food intolerances. O'Sullivan et al.
(2016) estimated the intake of artificial sweeteners, including
sucralose by children aged 1e3 years with dietary restrictions due
to phenylketonuria or cows milk allergy, using a tiered model
approach. Although FAIM estimates for sucralose exposure in these
children did not exceed the ADI for the average consumption (50th
percentile), estimates for high consumers were higher than the ADI,
thus refinements using Tier 2 and Tier 3 analyses were conducted.
For children adhering to a diet for management of phenylketonuria,
the mean exposure estimates ranged from 2.9 to 4.7 mg/kg/d and
95th percentile estimates ranged from 6.5 to 9.9 depending on the
predicted adherence to the diet. Similarly, children adhering to a
diet for management of cow's milk allergy had mean exposure
estimates of 2.5e3.6 mg/kg/d, and 5.8e7.8 mg/kg/d for children at
the 95th percentile. As discussed by the authors, these estimates
are also conservative and likely overestimate true consumption as
various assumptions were still required due to lack of specific data.
Therefore, there is little likelihood that sucralose consumption will
exceed the ADI in children adhering to these special diets. This is in
agreement with the recent opinion by the EFSA Panel on Food
Additives and Nutrient Sources added to Food (ANS) on the safety
of use of sucralose in dietary food for special medical purpose
intended for young children aged from 1 to 3 years (EFSA, 2016).
Dewinter et al. (2016) assessed intakes of sweeteners including
sucralose in Belgian children with T1D using Tier 2 and Tier 3 approaches. Food intake of children was estimated using a food frequency questionnaire specifically designed to assess intake of foods
sweetened with low calorie sweeteners. Although Tier 2 estimates
for consumption of sucralose were below the ADI for all consumers,
a Tier 3 analysis was also conducted and provides estimates more
closely predicting consumer intakes. In children with T1D, the
mean and 95th percentile estimated intakes of sucralose were: 2.6
and 8.6 mg/kg/d, 2.0 and 5.1 mg/kg/d, and 1.1 and 4.9 mg/kg/d, for
children ages 4e6, 7e12 and 13e18 years, respectively. In all cases,
the mean and high intake estimates were well below the ADI for
sucralose, indicating that there is no safety concern regarding
consumption of sucralose-containing foods by this special population of children with T1D, who are expected to be high
consumers.
The results of well-conducted consumption estimates, even
using conservative approaches such as use of maximum use levels,
consistently find that the intakes of sucralose in all members of the
population remain well below the ADI. These results support the
initial estimates by regulatory agencies (US FDA, 1998, 1999) that
expected intakes would be below the ADI, and even for high intake
consumers, average daily intakes are likely less than 3 mg/kg/d.
Recent studies specifically in children and those with special dietary needs have provided additional confidence that consumption
of sucralose is at safe levels in these unique population subgroups.
4. Summary and conclusions
The safety of sucralose has been extensively evaluated by regulatory agencies around the world, and it is approved globally for
use in foods and beverages as a non-caloric sweetener. Detailed
metabolism studies using radiolabelled sucralose to allow accurate
determination of its fate demonstrate very little absorption from
the gastrointestinal tract, and most ingested sucralose is excreted
unchanged in the feces. There is no retention or build-up in the
body with long-term use, and no evidence of either dechlorination
or hydrolysis of sucralose to metabolites in any species. Unlike
sucrose, there is no digestion or breakdown, confirming that
sucralose is not a source of energy or calories.
The toxicology studies required for approval of sucralose require
use of multiple doses, measurement and reporting of an extensive
list of endpoints, including growth, food consumption, blood
chemistry and enzyme levels, hematology, clinical analyses of
urine, eye examinations, changes in animal behavior and ultimately, tissue weights, histological examinations and pathology
findings. Studies to assess long-term exposure, reproduction and
development, neurotoxicity, genetic toxicity and cancer development have repeatedly demonstrated that sucralose has no safety
concerns. When sucralose is added to rodent feed at very high
levels (over one hundred times expected human consumption),
palatability of the feed is affected, and palatability-related effects
have been found in such circumstances, e.g., reduced food consumption, and expected lower body weight gain and lower weights
of some organs. These effects are not found when sucralose is
administered by gavage, also at very high levels. The gavage studies
thus confirm no direct/toxicity-related effect of sucralose on food
consumption, body weight gain and/or organ weight.
Several research studies in animals or cell culture systems have
reported various changes allegedly due to sucralose; however, inspection of these studies shows significant limitations, including
issues such as an absence of dose-response; inadequate assays and/
or number of endpoints, insufficient and/or inappropriate statistical analyses; and lack of appropriate controls. Such deficiencies
limit data interpretation.
The collective evidence supports that sucralose is noncarcinogenic, based on carcinogenicity studies that comply with
regulatory standards for appropriate design and conduct and no
evidence of genotoxicity.
Previous and numerous more recent clinical trials, in both
healthy and diabetic subjects and using a variety of approaches and
conditions, and including measurement of gastrointestinal incretins, collectively provides evidence of a lack of effect of sucralose on
both glycemic control and gut hormones and/or gut function.
The results of well-conducted consumption estimates, even
using conservative approaches such as use of maximum use levels,
consistently find that the intakes of sucralose in all members of the
population, including children and diabetics, remain well below the
ADI. The current average sucralose consumption level is also less
than 3 mg/kg bw/d, even at the 95th percentile of use, and
including special populations such as children and persons with
diabetes.
In conclusion, the extensive database of studies assessing
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
genetic toxicology, short and long term safety, animal and human
absorption, distribution, metabolism and excretion, reproductive,
development, and neurological effects and, most recently human
clinical trials in healthy and diabetic subjects by numerous researchers provide a clear demonstration of safety of use of sucralose as a non-caloric sweetener in foods and beverages.
Financial support
Financial support was provided by the Calorie Control Council,
Atlanta GA, to the employers of the authors for the preparation and
publication of this review.
Transparency document
Transparency document related to this article can be found
online at http://dx.doi.org/10.1016/j.fct.2017.05.047.
References
Abou-Donia, M.B., El-Masry, E.M., Abdel-Rahman, A.A., McLendon, R.E.,
Schiffman, S.S., 2008. Splenda alters gut microflora and increases intestinal Pglycoprotein and cytochrome P-450 in male rats. J. Toxicol. Environ. Health A 71,
1415e1429.
American Diabetes Association, 2014. Standards of medical care in diabetes e 2014.
Diabetes Care 37, S14eS80.
Ames, B.N., McCann, J., Yamasaki, E., 1975. Methods for detecting carcinogens and
mutagens with the salmonella/mammalian-microsome mutagenicity test.
Mutat. Res. 31, 347e363.
Amyes, S.J., Lee, P., Ashby, R., Finn, J.P., Fowler, J.S.L., Aughton, P., 1986. An Equimolar
Mixture of 4-chloro-4-deoxygalactose and 1,6-dichloro-1,6-dideoxyfructose:
104-week Oncogenicity Study in Rats. Unpublished report. Life Science
Research Ltd., Essex (UK). Submitted to the World Health Organization by Tate
& Lyle. Cited In: JECFA 1989b [As Amyes et al., 1986c].
curite
Sanitaire de Alimentation de Environment et
[ANSES] Agence Nationale de Se
du Travail, 2015. Opinion of the French Agency for Food, Environmental and
Occupational Health & Safety of 19 November 2014, Revised on 9 January 2015,
on the Assessment of the Nutritional Benefits and Risks Related to Intense
curite
Sanitaire de Alimentation
Sweeteners [Internet]. Agence Nationale de Se
de Environment et du Travail (ANSES), Maisons-Alfort (France) [cited 2017 Jan
20]. (ANSES Opinion - Request no. 2011-SA-0161). Available from: https://www.
anses.fr/en/content/opinion-french-agency-food-environmental-andoccupational-health-safety-19-november-2014.
Baird, I.M., Shephard, N.W., Merritt, R.J., Hildick-Smith, G., 2000. Repeated dose
study of sucralose tolerance in human subjects. Food Chem. Toxicol. 38
(Suppl. 2), S123eS129.
Bannach, G., Almeida, R.R., Lacerda, L.G., Schnitzler, E., Ionashiro, M., 2009. Thermal
stability and thermal decomposition of sucralose. Eclet Quim (Sao Paulo) 34,
21e26.
Barndt, R.L., Jackson, G., 1990. Stability of sucralose in baked goods. Food Technol.
44, 62e66.
Berry, C., Brusick, D., Cohen, S.M., Hardisty, J.F., Grotz, V.L., Williams, G.M., 2016.
Sucralose non-carcinogenicity: a review of the scientific and regulatory rationale. Nutr. Cancer 68, 1247e1261.
Bigal, M.E., Krymchantowski, A.V., 2006. Migraine triggered by sucralose-a case
report. Headache 46, 515e517.
Binns, N.M., 2003. Sucralose e all sweeteners and light. Nutr. Bull. BNF 28, 53e58.
Bootman, J., Lodge, D.C., 1980. 1,6-dichloro-1,6-dideoxyfructose: Assessment of its
Mutagenic Potential in Histidine Auxotrophs of Salmonella typhimurium. Unpublished report. Life Science Research, UK. Reading (UK): Submitted by
Reading (UK): Tate & Lyle Speciality Sweeteners. (Study No. 80/TYL057/308).
Cited In: SCF 2000b [Ref. #32].
Bootman, J., Lodge, D.C., 1981. 1,6-dichloro-1,6-dideoxyfructose: Assessment of its
Mutagenic Potential in Drosophila melanogaster, Using Sex-linked Recessive
Lethal Test. Unpublished report. Life Science Research. Reading (UK): Submitted
by Reading (UK): Tate & Lyle Speciality Sweeteners. (Study No. 81/MSP006/
089). Cited In: SCF 2000b [Ref. #37].
Bootman, J., May, K., 1981. The Detection of Mutagenic Activity by Increased
Reversion of Histidine Auxotrophs of Salmonella typhimurium. Unpublished
report. Life Science Research, UK. Reading (UK): Submitted by Reading (UK):
Tate & Lyle Speciality Sweeteners. (Study No. 81/MSP011/164). Cited In: SCF
2000b [Ref. #33].
Bootman, J., Rees, R., 1981. In Vitro Assessment of the Clastogenic Activity of 1,6DCF, 4-CG and TGS in Cultured Human Peripheral Lymphocytes. Unpublished
report. Life Science Research, UK. Reading (UK): Submitted by Reading (UK):
Tate & Lyle Speciality Sweeteners. (Study No. 81/MSP005/095). Cited In: SCF
2000b [Ref. #38].
Bootman, J., Whalley, H.E., 1981. 1,6-dichloro-1,6-dideoxyfructose: Investigation of
Effects on Bone Marrow Chromosomes of the Rat after Acute and Subacute Oral
351
Administration. Unpublished report. Life Science Research, UK. Reading (UK):
Submitted by Reading (UK): Tate & Lyle Speciality Sweeteners. (Study No. 81/
MSP012/257). Cited In: SCF 2000b [Ref. #39].
Brix, A.E., Hardisty, J.F., McConnell, E.E., 2010. Combining neoplasms for evaluation
of rodent carcinogenesis studies. In: Hsu, C.-H., Stedeford, T. (Eds.), Cancer Risk
Assessment: Chemical Carcinogenesis, Hazard Evaluation, and Risk Quantification. John Wiley & Sons, Hoboken (NJ), pp. 699e715.
Brown, R.J., Walter, M., Rother, K.I., 2009. Ingestion of diet soda before a glucose
load augments glucagon-like peptide-1 secretion. Diabetes Care 32, 2184e2186.
Brown, A., Brown, M.M.B., Onken, K.L., Beitz, D.C., 2011. Short term consumption of
sucralose, a nonnutritive sweetener, is similar to water with regard to select
markers of hunger signalling and short term glucose homeostasis in women.
Nutr. Res. 31, 882e888.
Brown, R.J., Walter, M., Rother, K.I., 2012. Effects of diet soda on gut hormones in
youths with diabetes. Diabetes Care 2012 (35), 959e964.
Brusick, D.J., Fields, W.R., Myhr, B.C., Doolittle, D.J., 2008. Genetic toxicology. In:
Hayes, A.W. (Ed.), PrInciples and Methods of Toxicology, fifth ed. CRC Press,
Boca Raton (FL), pp. 1179e1221.
Brusick, D., Borzelleca, J.F., Gallo, M., Williams, G., Kille, J., Hayes, A.W., PiSunyer, F.X., Williams, C., Burks, W., 2009. Expert panel report on a study of
Splenda in male rats. Regul. Toxicol. Pharmacol. 55, 6e12.
Brusick, D., Grotz, V.L., Slesinski, R., Kruger, C.L., Hayes, A.W., 2010. The absence of
genotoxicity of sucralose. Food Chem. Toxicol. 48, 3067e3072.
Bryant, C., Mclaughlin, J., 2016. Low calorie sweeteners: evidence remains lacking
for effects on human gut function. Physiol. Behav. 164 (Pt B), 482e485.
Clive, D., Spector, J.F.S., 1975. Laboratory procedure for assessing specific locus
mutations at the TK locus in L5178Y mouse lymphoma cells. Mutat. Res. 31,
17e29.
Clive, D., McCuen, R., Spector, J.F., Piper, C., Mavournin, K.H., 1983. Specific gene
mutations in L5178Y cells in culture. Mutat. Res. 115, 225e251.
COM, Committees on Toxicity, Mutagenicity, Carcinogenicity of Chemicals in Food,
1995. 1994 Annual Report: Committees on Toxicity, Mutagenicity, Carcinogenicity of Chemicals in Food, Consumer Products and the Environment.
Department of Health/Her Majesty's Stationary Office (HMSO), London (UK).
Cited In: SCF 2000b [Ref. #44].
Contrera, J.F., Jacobs, A.C., DeGeorge, J.J., Chen, C., Choudary, J., DeFelice, A.,
Fairweather, W., Farrelly, J., Fitzgerald, G., Goheer, A., et al., 1996. Carcinogenicity
Testing and the Evaluation of Regulatory Requirements for Pharmaceuticals.
Federal Register, U.S. Department of Health and Human Services, Public Health
Service (Docket No. 96D-0235). Cited In: Williams & Iatropoulos 2001
[Ref. #33].
Cruzan, G., 2009. Assessment of the cancer potential of methanol. Crit. Rev. Toxicol.
39, 347e363.
de Oliveira, D.N., de Menezes, M., Catharino, R.R., 2015. Thermal degradation of
sucralose: a combination of analytical methods to determine stability and
chlorinated byproducts. Sci. Rep. 5 http://dx.doi.org/10.1038/srep09598, 9598
[5pp].
Dewinter, L., Casteels, K., Corthouts, K., Van de Kerckhove, K., Van der Vaerent, K.,
Vanmeerbeeck, K., Matthys, C., 2016. Dietary intake of non-nutritive sweeteners
in type 1 diabetes mellitus children. Food Addit. Contam. Part A Chem. Anal.
Control Expo. Risk Assess. 33, 19e26.
Dong, S., Wu, J., Liu, G., Zhang, B., Zheng, M., 2011. Unintentionally produced
dioxinlike polychlorinated biphenyls during cooking. Food Control 22,
1797e1802.
Dong, S., Liu, G., Hu, J., Zheng, M., 2013a. Polychlorinated dibenzo-p-dioxins and
dibenzofurans formed from sucralose at high temperatures. Sci. Rep. 3 http://
dx.doi.org/10.1038/srep02946 [4pp].
Dong, S., Liu, G., Zhang, B., Gao, L., Zheng, M., 2013b. Formation of polychlorinated
naphthalenes during the heating of cooking oil in the presence of high amounts
of sucralose. Food Control 32, 1e5.
Durnev, A.D., Oreshchenko, A.V., Kulakova, A.V., Beresten, N.F., Seredenin, S.B., 1995.
[Study into the clastogenic activity of dietary sugar substitutes]. Vopr. Med.
Khim 41, 31e33.
Ecobichon, D.J., 1996. Toxic effects of pesticides. In: Klaassen, C.D., Amdur, M.O.,
Doull, J. (Eds.), Casarett and Doull's Toxicology: the Basic Science of Poisons,
fifth ed. McGraw-Hill Health Professions Division, New York (NY)/Toronto (ON),
pp. 643e689.
EFSA, European Food Safety Authority, 2006. Opinion of the Scientific Panel on food
additives, flavourings, processing aids and materials in contact with food (AFC)
related to a new long-term carcinogenicity study on aspartame. EFSA J. 356,
1e44. http://dx.doi.org/10.2903/j.efsa.2006.356 [Internet]. [cited 2017 Jan 20].
Available from: http://www.efsa.europa.eu/en/efsajournal/pub/356.
EFSA, European Food Safety Authority, 2011. EFSA Statement on the scientific
evaluation of two studies related to the safety of artificial sweeteners. EFSA J. 9
(2) http://dx.doi.org/10.2903/j.efsa.2011.2089, 2089 [16 pp.]. [Internet]. [cited
2017 Jan 20]. Available from: http://www.efsa.europa.eu/en/efsajournal/pub/
2089.
EFSA, European Food Safety Authority, 2012a. Guidance on the use of probabilistic
methodology for modelling dietary exposure to pesticide residues. EFSA J. 10
(10) http://dx.doi.org/10.2903/j.efsa.2012.2839, 2839 [95pp.]. [Internet]. [cited
2017 Jan 20]. Available from: http://www.efsa.europa.eu/en/efsajournal/pub/
2839.
EFSA, European Food Safety Authority, 2012b. Food Additives Intake Model (FAIM) December 2012 [Internet]. European Food Safety Authority (EFSA), Parma
(Italy). FIP Unit [cited 2017 Jan 20]. Available from: http://www.efsa.europa.eu/
352
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
sites/default/files/efsa_rep/blobserver_assets/faimtemplateinstructions.pdf.
EFSA, European Food Safety Authority, 2013. Food Additives Intake Model (FAIM) Version 1.1-July 2013 [Internet]. European Food Safety Authority (EFSA), Parma
(Italy). FIP Unit [cited 2017 Jan 20]. Available from: http://www.efsa.europa.eu/
sites/default/files/assets/faimtemplate.xls.
EFSA, European Food Safety Authority, 2016. Safety of the proposed extension of use
of sucralose (E 955) in foods for special medical purposes in young children.
EFSA J. 14 (1) http://dx.doi.org/10.2903/j.efsa.2016.4361, 4361 [11 pp.].
[Internet]. [cited 2017 Jan 20]. Available from: http://www.efsa.europa.eu/en/
efsajournal/pub/4361.
EFSA, 2017. Statement on the validity of the conclusions of a mouse carcinogenicity
study on sucralose (E 955) performed by the Ramazzini Institute. European
Food Safety Authority. EFSA J. 15 (5), 4784. Available at. http://onlinelibrary.
wiley.com/doi/10.2903/j.efsa.2017.4784/epdf.
Fava, F., Gitau, R., Griffin, B.A., Gibson, G.R., Tuohy, K.M., Lovegrove, J.A., 2013. The
type and quantity of dietary fat and carbohydrate alter faecal microbiome and
short-chain fatty acid excretion in a metabolic syndrome 'at-risk' population.
Int. J. Obes. 37, 216e223.
FCT, 2000. Sucralose safety assessment. Food Chem. Toxicol. 38 (Suppl. 2) iii-129.
Finn, J.P., Lord, G.H., 2000. Neurotoxicity studies on sucralose and its hydrolysis
products with special reference to histopathologic and ultrastructural changes.
Food Chem. Toxicol. 38 (Suppl. 2), S7eS17.
Fitch, C., Keim, K.S., 2012. Position of the Academy of Nutrition and Dietetics: use of
nutritive and nonnutritive sweeteners. J. Acad. Nutr. Diet. 112, 739e758.
Flamm, W.G., Blackburn, G.L., Comer, C.P., Mayhew, D.A., Stargel, W.W., 2003. Longterm food consumption and body weight changes in neotame safety studies are
consistent with the allometric relationship observed for other sweeteners and
during dietary restrictions. Regul. Toxicol. Pharmacol. 38, 144e156.
FSANZ, Food Standards Australia New Zealand, 2004. Consumption of Intense
Sweeteners in Australia and New Zealand: Benchmark Survey 2003 [Internet].
Food Standards Australia New Zealand (FSANZ), Canberra (Australia)/
Wellington (NZ) [cited 2017 Jan 20]. (Evaluation Report Series no 8). Available
from:
http://www.foodstandards.gov.au/publications/documents/Intense_
sweetener_Report_feb04.pdf.
Ford, C.D., Hamerton, J.L., 1956. The chromosomes of man. Nature 178, 1020e1023.
Ford, H.E., Peters, V., Martin, N.M., Sleeth, M.L., Ghatei, M.A., Frost, G.S., Bloom, S.R.,
2011. Effects of oral ingestion of sucralose on gut hormone response and
appetite in healthy normal-weight subjects. Eur. J. Clin. Nutr. 65, 508e513.
Fujita, Y., Wideman, R.D., Speck, M., Asadi, A., King, D.S., Webber, T.D., Haneda, M.,
Kieffer, T.J., 2009. Incretin release from gut is acutely enhanced by sugar but not
by sweeteners in vivo. Am. J. Physiol. Endocrinol. Metab. 296, E473eE479.
Gardner, C., Wylie-Rosett, J., Gidding, S.S., Steffen, L.M., Johnson, R.K., Reader, D.,
Lichtenstein, A.H., 2012. Nonnutritive sweeteners: current use and health perspectives: a scientific statement from the American Heart association and the
American diabetes association [American Heart association nutrition committee of the Council on nutrition, physical activity and metabolism, Council on
arteriosclerosis, thrombosis and vascular biology, Council on cardiovascular
disease in the young, and the American diabetes association]. Circulation 126,
509e519.
Gart, J.J., Krewski, D., Tarone, R.E., Wahrendorf, J., 1986. Statistical Methods in Cancer
Research. Volume 3-the Design and Analysis of Long-term Animal Experiments.
World Health Organization (WHO). Lyon (France): International Agency for
Research on Cancer (IARC). (IARC Scientific Publications, no 79).
Gift, J.S., Caldwell, J.C., Jinot, J., Evans, M.V., Cote, I., Vandenberg, J.J., 2013. Scientific
considerations for evaluating cancer bioassays conducted by the Ramazzini
Institute. Environ. Health Perspect. 121, 1253e1263.
Goldsmith, L.A., 2000. Acute and subchronic toxicity of sucralose. Food Chem.
Toxicol. 38 (Suppl. 2), S53eS69.
Goldsmith, L.A., Merkel, C.M., 2001. Sucralose. In: O'Brien Nabors, L., Gelardi, R.C.
(Eds.), Alternative Sweeteners. 3rd ed., rev., Exp. Marcel Dekker, Inc. (Food
Science and Technology, no 112), New York (NY), pp. 185e207 [See p 192].
Grice, H.C., Goldsmith, L.A., 2000. Sucralose-An overview of the toxicity data. Food
Chem. Toxicol. 38 (Suppl. 2), S1eS6.
Grotz, V.L., 2008. Sucralose and migraine. Headache 48, 164e165.
Grotz, V.L., Munro, I.C., 2009. An overview of the safety of sucralose. Regul. Toxicol.
Pharmacol. 55, 1e5.
Grotz, V.L., Henry, R.R., McGill, J.B., Prince, M.J., Shamoon, H., Trout, J.R., PiSunyer, F.X., 2003. Lack of effect of sucralose on glucose homeostasis in subjects
with type 2 diabetes. J. Am. Diet. Assoc. 103, 1607e1612.
Grotz, V.L., Molinary, S., Molinary, R.C., Peterson, R.C., Quinlan, M.E., Reo, R., 2012.
Chapter 12. Sucralose. In: O'Brien Nabors, L., Gelardi, R.C. (Eds.), Alternative
Sweeteners. 4th ed., rev., Exp. CRC Press, Taylor and Francis Group (Food Science
and Technology, no 112), Boca Raton (FL), pp. 181e196 [See p 1982].
Gupta, S., Kalra, S., Bharihoke, V., Dhurandhar, D., 2014. Sucralose induced
pancreatic toxicity in albino rats: histomorphological evidence. J. Morphol. Sci.
31, 123e127.
Ha, M.S., Ha, S.D., Choi, S.H., Bae, D.H., 2013. Assessment of exposure of Korean
consumers to acesulfame K and sucralose using a stepwise approach. Int. J. Food
Sci. Nutr. 64, 715e723.
Hartmann, A., Schumacher, M., Plappert-Helbig, U., Lowe, P., Suter, W., Mueller, L.,
2004. Use of the alkaline in vivo Comet assay for mechanistic genotoxicity investigations. Mutagenesis 19, 51e59.
Hawkins, D.R., Wood, S.W., Waller, A.R., Jordan, M.C., 1987. Enzyme Induction
Studies of TGS and TGS-HP in the Rat. Unpublished report. Huntingdon
Research Centre, Huntingdon (UK). Submitted to the World Health Organization
by Tate & Lyle. Cited In: JECFA 1989b [As Hawkins et al. 1987d].
Haworth, S.R., Lawlor, T.E., Smith, T.K., Williams, N.A., Simmons, R.T., Hans, L.T.,
Reichard, G.L., 1981. Salmonella/mammalian-microsome Plate Incorporation
Mutagenesis Assays: Test Article 1,6-dichlorofructose. Unpublished report. EG
and G Mason Research Institute, Rockville (MD). Reading (UK): Submitted by
Reading (UK): Tate & Lyle Speciality Sweeteners. (Study No. 45-280-536-1; MRI
#536). Cited In: SCF 2000b [Ref. #34].
Hayes, A.W., Dayan, A.D., Hall, W.C., Kodell, R.L., Williams, G.M., Waddell, W.D.,
Slesinski, R.S., Kruger, C.L., 2011. A review of mammalian carcinogenicity study
design and potential effects of alternate test procedures on the safety evaluation of food ingredients. Regul. Toxicol. Pharmacol. 59, 142e175.
Hazleton Laboratories America, 1988a. 1,6-dichloro-1,6-dideoxy-D-fructose: in Vivo
Sister Chromatid Exchange Assay in the Mouse. Unpublished report. Hazleton
Laboratories America, Inc., Kensington (MD). Submitted by Reading (UK): Tate
& Lyle Speciality Sweeteners. (Study No. 10221-0-458. November 4, 1988). Cited
In: SCF 2000b [Ref. #40].
Hazleton Laboratories America, 1988b. 1,6-dichloro-1,6-dideoxy-D-fructose: in Vivo
Bone Marrow Micronucleus Assay in the Mouse. Unpublished report. Hazleton
Laboratories America, Inc., Kensington (MD). Reading (UK): Submitted by
Reading (UK): Tate & Lyle Speciality Sweeteners. (Study No. 10221-0-455.
November 4, 1988). Cited In: SCF 2000b [Ref. #41].
Health Canada, 2016. List of Permitted Sweeteners [Internet]. Health Canada,
Ottawa (ON) [cited 2017 Jan 20]. (Lists of Permitted Food Additives, No. 9).
Available from: http://www.hc-sc.gc.ca/fn-an/securit/addit/list/9-sweeteneredulcorant-eng.php [Date Modified: 2016-11-25].
Heddle, J.A., Salamone, M.F., 1981. Chromosomal aberrations and bone marrow
toxicity. Environ. Health Perspect. 39, 23e27.
Hirsch, A.R., 2007. Migraine triggered by sucraloseea case report. Headache 47, 447.
Hutchinson, S.A., Ho, G.S., Ho, C.T., 1999. Stability and degradation of the high intensity sweeteners: aspartame, alitame, and sucralose. Food Rev. Int. 15,
249e261.
Huvaere, K., Vandevijvere, S., Hasni, M., Vinkx, C., Van Loco, J., 2012. Dietary intake
of artificial sweeteners by the Belgian population. Food Addit. Contam. Part A
Chem. Anal. Control Expo. Risk Assess. 2012 (29), 54e65.
Ibero-Baraibar, I., Cuervo, M., Navas-Carretero, S., Abete, I., Zulet, M.A.,
Martinez, J.A., 2014. Different postprandial acute response in healthy subjects to
three strawberry jams varying in carbohydrate and antioxidant content: a
randomized, crossover trial. Eur. J. Nutr. 53, 201e210.
Jang, H.-J., Kokrashvili, Z., Theodorakis, M.J., Carlson, O.D., Kim, B.-J., Zhou, J.,
Kim, H.H., Xu, X., Chan, S.L., Juhaszova, M., et al., 2007. Gut-expressed gustducin
and taste receptors regulate secretion of glucagon-like peptide-1. Proc. Natl.
Acad. Sci. U. S. A. 104, 15069e15074.
JFCRF, Japan Food Chemical Research Foundation, 1999. A Report on Sucralose from
the Food Sanitation Council [Internet]. Ministry of Health, Labour and Welfare,
Japan (MHLW) and Japan Food Chemical Research Foundation (JFCRF), Tokyo
(Japan) [cited 2017 Jan 20]. Available from: http://www.ffcr.or.jp/zaidan/
ffcrhome.nsf/pages/e-kousei-sucra [January 6, 1999].
JECFA, Joint FAO/WHO Expert Committee on Food Additives, 1989a. Trichorogalactosucrose [sucralose]. In: Evaluation of Certain Food Additives and Contaminants. Thirty-third Report of the Joint FAO/WHO Expert Committee on
Food Additives Joint FAO/WHO Expert Committee on Food Additives (JECFA),
Mar. 21-30, 1988, Geneva [Internet]. World Health Organization (WHO), Geneva
[cited 2017 Jan 20]. (WHO Technical Report Series, no 776), pp. 20e22. Available
from: http://whqlibdoc.who.int/trs/WHO_TRS_776.pdf.
JECFA, Joint FAO/WHO Expert Committee on Food Additives, 1989b. Trichorogalactosucrose [sucralose]. In: Toxicological Evaluation of Certain Food Additives and Contaminants. 33rd Meeting of the Joint FAO/WHO Expert Committee
on Food Additives (JECFA), Mar. 21-30, 1989, Geneva [Internet]. World Health
Organization (WHO), Geneva [cited 2017 Jan 20]. (WHO Food Additives Series,
no 24), pp. 45e94. Available from: http://www.inchem.org/documents/jecfa/
jecmono/v024je05.htm.
JECFA, Joint FAO/WHO Expert Committee on Food Additives, 1991a. 3.1.5. Sweetening agents. Trichlorogalactosucrose (TGS) [sucralose]. In: Evaluation of
certain food additives and contaminants. Thirty-seventh meeting of the Joint
FAO/WHO Expert Committee on Food Additives Joint FAO/WHO Expert Committee on Food Additives (JECFA), Jun. 5-14, 1990, Geneva [Internet]. Geneva:
World Health Organization (WHO) [cited 2017 Jan 20]. (WHO Technical Report
Series, no 806), p. 21e23, 42. Available from: http://whqlibdoc.who.int/trs/
WHO_TRS_806.pdf.
JECFA, Joint FAO/WHO Expert Committee on Food Additives, 1991b. Trichlorogalactosucrose (TGS) [sucralose]. In: Toxicological Evaluation of Certain
Food Additives and Contaminants. Thirty-seventh Meeting of the Joint FAO/
WHO Expert Committee on Food Additives (JECFA), June 5-14, 1990, Geneva
[Internet]. World Health Organization (WHO)/International Programme on
Chemical Safety (IPCS), Geneva [cited 2017 Jan 20]. (WHO Food Additives Series,
no 28). Available from: http://www.inchem.org/documents/jecfa/jecmono/
v28je14.htm.
Jeffrey, A.M., Williams, G.M., 2000. Lack of DNA-damaging activity of five nonnutritive sweeteners in the rat hepatocyte/DNA repair assay. Food Chem. Toxicol. 38, 335e338.
Jenner, M.R., Smithson, A., 1989. Physicochemical properties of the sweetener
sucralose. J. Food Sci. 54, 1646e1649.
John, B.A., Wood, S.G., Hawkins, D.R., 2000a. The pharmacokinetics and metabolism
of sucralose in the mouse. Food Chem. Toxicol. 38 (Suppl. 2), S107eS110.
John, B.A., Wood, S.G., Hawkins, D.R., 2000b. The pharmacokinetics and metabolism
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
of sucralose in the rabbit. Food Chem. Toxicol. 38 (Suppl. 2), S111eS113.
€ rfer, R., O'Mahony, C., Schnabel, J.,
Kettler, S., Kennedy, M., McNamara, C., Oberdo
Smith, B., Sprong, C., Faludi, R., Tennant, D., 2015. Assessing and reporting uncertainties in dietary exposure analysis: mapping of uncertainties in a tiered
approach. Food Chem. Toxicol. 82, 79e95.
Kille, J.W., Tesh, J.M., McAnulty, P.A., Ross, F.W., Willoughby, C.R., Bailey, G.P.,
Wilby, O.K., Tesh, S.A., 2000a. Sucralose: assessment of teratogenic potential in
the rat and the rabbit. Food Chem. Toxicol. 38 (Suppl. 2), S43eS52.
Kille, J.W., Ford, W.C., McAnulty, P., Tesh, J.M., Ross, F.W., Willoughby, C.R., 2000b.
Sucralose: lack of effects on sperm glycolysis and reproduction in the rat. Food
Chem. Toxicol. 38 (Suppl. 2), S19eS29.
Kirby, P.E., Pizzarello, R.F., Brauninger, R.M., Vega, R.A., Reichard, G.L., 1981a. Evaluation of Test Article 1,6-dichlorofructose (MRI #536) for Mutagenic Potential
Employing the L5178Y TK þ/- Mutagenesis Assay. Unpublished report. EG and G
Mason Research Institute, Rockville (MD). Reading (UK): Submitted by Reading
(UK): Tate & Lyle Speciality Sweeteners. (Study No. 045-280-536-7). Cited In:
SCF 2000b [Ref. #35].
Kirby, P.E., Pizzarello, R.F., Cohen, A., Williams, P.E., Reichard, G.L., Johnson, J.L.,
Clarke, J.J., Condon, B., Wattam, R.E., 1981b. Evaluation of Test Article 1,6dichlorofructose (MRI #629) for Mutagenic Potential Employing the L5178Y
TK þ/- Mutagenesis Assay. Unpublished report. EG and G Mason Research
Institute, Rockville (MD). Reading (UK): Submitted by Reading (UK): Tate & Lyle
Speciality Sweeteners. (Study No. 045-405-629-7). Cited In: SCF 2000b
[Ref. #36].
Knight, I., 1994. The development and applications of sucralose, a new highintensity sweetener. Can. J. Physiol. Pharmacol. 72, 435e439.
Kojima, I., Nakagawa, Y., Ohtsu, Y., Hamano, K., Medina, J., Nagasawa, M., 2015.
Return of the glucoreceptor: glucose activates the glucose-sensing receptor
T1R3 and facilitates metabolism in pancreatic b-cells. J. Diabetes Investig. 6,
256e263.
€wik, M.R.H., van Klaveren, J., Kleiner, J., Massey, R.,
Kroes, R., Müller, D., Lambe, J., Lo
Mayer, S., Urieta, I., Verger, P., Visconti, A., 2002. Assessment of intake from the
diet. Food Chem. Toxicol. 40 (2/3), 327e385.
Kwak, J.H., Kim, M.S., Lee, J.H., Yang, Y.J., Lee, K.H., Kim, O.Y., Lee, J.H., 2013. Beneficial
effect of tagatose consumption on postprandial hyperglycemia in Koreans: a
double-blind crossover designed study. Food Funct. 4, 1223e12228.
Leegwater, D.C., de Groot, A.P., van Kalmthout-Kuyper, M., 1974. The aetiology of
cacecal enlargement in the rat. Food Cosmet. Toxicol. 12, 687e698.
Lord, G.H., Newberne, P.M., 1990. Renal mineralization-A ubiquitous lesion in
chronic rat studies. Food Chem. Toxicol. 28, 449e455.
Lu, F.C., 1988. Acceptable daily intake: inception, evolution, and application. Regul.
Toxicol. Pharmacol. 8, 45e60.
Ma, J., Bellon, M., Wishart, J.M., Young, R., Blackshaw, L.A., Jones, K.L., Horowitz, M.,
Rayner, C.K., 2009. Effect of the artificial sweetener, sucralose, on gastric
emptying and incretin hormone release in healthy subjects. Am. J. Physiol.
Gastrointest. Liver Physiol. 296, G735eG739.
Ma, J., Chang, J., Checklin, H.L., Young, R.L., Jones, K.L., Horowitz, M., Rayner, C.K.,
2010. Effect of the artificial sweetener, sucralose, on small intestinal glucose
absorption in healthy human subjects. Br. J. Nutr. 104, 803e806.
Mace, O.J., Affleck, J., Patel, N., Kellett, G.L., 2007. Sweet taste receptors in rat small
intestine stimulate glucose absorption through apical GLUT2. J. Physiol. 582,
397e392. erratum, 583:411.
Magnuson, B., Munro, I.C., Abbott, P.J., Baldwin, N., Lopez-Garcia, R., Ly, K.,
McGirr, L.G., Roberts, A., Socolovsky, S., 2013. Review of the regulation and
safety assessment of food substances in various countries and jurisdictions.
Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk Assess. 30,
1147e1220.
Magnuson, B.A., Carakostas, M.C., Moore, N.H., Poulos, S.P., Renwick, A.G., 2016.
Biological fate of low-calorie sweeteners. Nutr. Rev. 74, 670e689.
Mann, S.W., Yuschak, M.M., Amyes, S.J., Aughton, P., Finn, J.P., 2000a. A combined
chronic toxicity/carcinogenicity study of sucralose in sprague-dawley rats. Food
Chem. Toxicol. 38 (Suppl. 2), S71eS89.
Mann, S.W., Yuschak, M.M., Amyes, S.J., Aughton, P., Finn, J.P., 2000b.
A carcinogenicity study of sucralose in the CD-1 mouse. Food Chem. Toxicol. 38
(Suppl. 2), S91eS98.
Margolskee, R.F., Dyer, J., Kokrashvili, Z., Salmon, K.S., Ilegems, E., Daly, K.,
Maillet, E.L., Ninomiya, Y., Mosinger, B., Shirazi-Beechey, S.P., 2007. T1R3 and
gustducin in gut sense sugars to regulate expression of Naþ-glucose cotransporter 1. Proc. Natl. Acad. Sci. U. S. A. 104, 15075e15080.
Martyn, D.M., Nugent, A.P., McNulty, B.A., O'Reilly, E., Tlustos, C., Walton, J., Flynn, A.,
Gibney, M.J., 2016. Dietary intake of four artificial sweeteners by Irish preschool children. Food Addit. Contam. Part A Chem. Anal. Control Expo. Risk
Assess. 33, 592e602.
Mayhew, D.A., Comer, C.P., Stargel, W.W., 2003. Food consumption and body weight
changes with neotame, a new sweetener with intense taste: differentiating
effects of palatability from toxicity in dietary safety studies. Regul. Toxicol.
Pharmacol. 38, 124e143.
McConnell, E.E., Solleveld, H.A., Swenberg, J.A., Boorman, G.A., 1988. Guidelines for
combining neoplasms for evaluation of rodent carcinogenesis studies. In:
Grice, H.C., Ciminera, J.L. (Eds.), Carcinogenicity: the Design, Analysis, and
Interpretation of Long-term Animal Studies. Interdisciplinary Discussion Group
on Carcinogenicity Studies - Proceedings, June 23-26, 1986. Springer Verlag/
Washington (DC): International Life Sciences Institute (ILSI), Chapel Hill, NC
(ILSI Monographs). New York (NY), pp. 183e196.
Meier, J.J., Nauck, M.A., 2005. Glucagon-like peptide 1(GLP-1) in biology and
353
pathology. Diabetes Metab. Res. Rev. 21, 91e117.
Myer, J.R., Romach, E.H., Elangbam, C.S., 2014. Species- and dose-specific pancreatic
responses and progression in single- and repeat-dose studies with GI181771X:
a novel cholecystokinin 1 receptor agonist in mice, rats, and monkeys. Toxicol.
Pathol. 42, 260e274.
Mezitis, N.H., Maggio, C.A., Koch, P., Quddoos, A., Allison, D.B., Pi-Sunyer, F.X., 1996.
Glycemic effect of a single high oral dose of the novel sweetener sucralose in
patients with diabetes. Diabetes Care 19, 1004e1005.
Microtest Research Limited, 1989. Study to Evaluate the Covalent DNA Binding
Potential of 1,6-dichlorofructose in Organs of Treated Rats. Microtest Research
Limited, York (UK) (Study Number TAL 1/DBS; March 23, 1989). Cited In: SCF
2000b [Ref. #42].
Molecular Toxicology (Microtest), 1994. Study to Evaluate the Potential of 1,6dichlorofructose to Induce Unscheduled DNA Synthesis in Isolated Rat Hepatocytes in Vitro. Unpublished report. Reading (UK): Submitted by Reading (UK):
Tate & Lyle Speciality Sweeteners. (Hazleton Europe Study Number 178/2,
report reference RHURETAL.001). Cited In: SCF 2000b [Ref. #31].
Moore, M.M., Honma, M., Clements, J., Bolcsfoldi, G., Burlinson, B., Cifone, M.,
Clarke, J., Clay, P., Doppalapudi, R., Fellows, M., et al., 2007. Mouse lymphoma
thymidine kinase gene mutation assay: meeting of the international workshop
on genotoxicity testing, san Francisco, 2005, recommendations for 24-h treatment [mouse lymphoma assay workgroup, IWGT]. Mutat. Res. 627, 36e40.
Morton, D., Kemp, R.K., Francke-Carroll, S., Jensen, K., McCartney, J., Monticello, T.M.,
Perry, R., Pulido, O., Roome, N., Schafer, K., et al., 2006. Best Practices for
Reporting Pathology Interpretations within GLP Toxicology Studies [Internet].
Society of Toxicologic Pathology (SOTP), Reston (VA) [cited 2017 Jan 20]. (STP
Position Papers). Available from: http://www.toxpath.org/Position_Papers/GLP.
pdf.
Moran-Ramos, S., Tovar, A.R., Torres, N., 2012. Diet: friend or foe of enteroendocrine
cellsehow it interacts with enteroendocrine cells. Adv. Nutr. 3, 8e20.
NTP, 2011. Summary Report of the National Toxicology Program and Environmental
Protection Agency Sponsored Review of Pathology Materials from Selected
Ramazzini Institute Rodent Cancer Bioassays. National Toxicology Program,
Research Triangle Park. Available at. http://ntp.niehs.nih.gov/NTP/About_NTP/
Partnerships/International/SummaryPWG_Report_RI_Bioassays.pdf.
Ng, S.W., Slining, M.M., Popkin, B.M., 2012. Use of caloric and noncaloric sweeteners
in US consumer packaged foods, 2005-2009. J. Acad. Nutr. Diet. 112, 1828e1834
e1-6.
O'Sullivan, A.J., Pigat, S., O'Mahony, C., Gibney, M.J., McKevitt, A.I., 2016. Probabilistic
modelling to assess exposure to three artificial sweeteners of young Irish patients aged 1-3 years with PKU and CMPA. Food Addit. Contam. Part A Chem.
Anal. Control Expo. Risk Assess. 2016 (33), 1660e1671.
OECD, Organisation for Economic Co-operation and Development, 1998. OECD
Principles of Good Laboratory Practice [Internet]. Organisation for Economic
Co-Operation & Development (OECD), Environment Directorate, Chemicals
Group and Management Committee, OECD Environmental Health and Safety
Publications, Paris (France) [cited 2017 Jan 20]. (Series on Principles of Good
Laboratory Practice and Compliance Monitoring, no. 1 [ENV/MC/CHEM(98)17]).
Available from: http://www.oecd-ilibrary.org/environment/oecd-principles-ongood-laboratory-practice_9789264078536-en [As revised in 1997].
Omran, A., Ahearn, G., Bowers, D., Swenson, J., Coughlin, C., 2013. Metabolic effects
of sucralose on environmental bacteria. J. Toxicol. 2013 http://dx.doi.org/
10.1155/2013/372986. Article ID 372986 [7pp]. Cited In: Magnuson et al. 2016.
Panseri, S., Biondi, P.A., Vigo, D., Communod, R., Chiesa, L.M., 2013. Occurrence of
organochlorine pesticides residues in animal feed and fatty bovine tissue. In:
Muzzalupo, I. (Ed.), Food Industry [Internet]. InTech, Rijeka (Croatia),
pp. 261e283 [cited 2017 Jan 20]. Available from: http://www.intechopen.com/
books/food-industry/occurrence-of-organochlorine-pesticides-residues-inanimal-feed-and-fatty-bovine-tissue.
Pariza, M.W., Ponakala, S.V., Gerlat, P.A., Andress, S., 1998. Predicting the functionality of direct food additives. Food Technol. 52, 56e60.
Patel, R.M., Sarma, R., Grimsley, E., 2006. Popular sweetener sucralose as a migraine
trigger. Headache 46, 1303e1304.
Pepino, M.Y., Tiemann, C.D., Patterson, B.W., Wice, B.M., Klein, S., 2013. Sucralose
affects glycemic and hormonal responses to an oral glucose load. Diabetes Care
36, 2530e2535.
Preston, R.J., Dean, B.J., Galloway, S., Holden, H., McFee, A.F., Shelby, M., 1987.
Mammalian in vivo cytogenetic assays. Analysis of chromosome aberrations in
bone marrow cells. Mutat. Res. 189, 157e165.
Rahiman, F., Pool, E.J., 2014. The in vitro effects of artificial and natural sweeteners
on the immune system using whole blood culture assays. J. Immunoass.
Immunochem. 35, 26e36.
Rahn, A., Yaylayan, V.A., 2010. Thermal degradation of sucralose and its potential in
generating chloropropanols in the presence of glycerol. Food Chem. 118, 56e61.
Reigart, J.R., Roberts, J.R., 2013. Organochlorines [Internet]. In: Recognition and
Management of Pesticide Poisonings, sixth ed. National Pesticide Information
Center (NPIC). Washington (DC): U.S. Environmental Protection Agency (US
EPA), Office of Pesticide Programs, Corvallis (OR), pp. 63e69 [cited 2017 Jan 20].
Available from: http://npic.orst.edu/rmpp.htm. [Select (chapter 7)].
Renwick, A.G., 2006. The intake of intense sweeteners - an update review. Food
Addit. Contam. 23, 327e338.
Reyna, N.Y., Cano, C., Bermúdez, V.J., Medina, M.T., Souki, A.J., Ambard, M.,
~ ez, M., Ferrer, M.A., Inglett, G.E., 2003. Sweeteners and beta-glucans
Nun
improve metabolic and anthropometrics variables in well controlled type 2
diabetic patients. Am. J. Ther. 10, 438e443.
354
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
Roberts, A., 2016. The safety and regulatory process for low calorie sweeteners in
the United States. Physiol. Behav. 164, 439e444.
Roberts, A., Renwick, A.G., Sims, J., Snodin, D.J., 2000. Sucralose metabolism and
pharmacokinetics in man. Food Chem. Toxicol. 38 (Suppl. 2), S31eS41.
Rocha, G.S., Pereira, M.O., Benarroz, M.O., Frydman, J.N., Rocha, V.C., Pereira, M.J.,
Fonseca, A.S., Medeiros, A.C., Bernardo-Filho, M., 2011. Sucralose sweetener
in vivo effects on blood constituents radiolabeling, red blood cell morphology
and radiopharmaceutical biodistribution in rats. Appl. Radiat. Isot. 69, 46e51.
Rocha, G.S., Pereira, M.O., Benarroz, M.O., Frydman, J.N., Rocha, V.C., Pereira, M.J.,
Fonseca, A.S., Medeiros, A.C., Bernardo-Filho, M., 2012. Sucralose sweetener
does not modify radiolabeling of blood constituents and morphology of red
blood cells. Med. Chem. Res. 21, 1084e1089. Cited In: van Eyk 2015.
Rodearmel, S.J., Wyatt, H.R., Stroebele, N., Smith, S.M., Ogden, L.G., Hill, J.O., 2007.
Small changes in dietary sugar and physical activity as an approach to preventing excessive weight gain: the America on the Move family study. Pediatrics 120, e869ee879. Cited In: Grotz & Munro 2009.
Rodgers, P.B., Jenner, M.R., Jones, H.F., 1986. Stability of Chlorinated Disaccharides to
Hydrolysis by Microbial Plant and Mammalian Glycosidases. Unpublished
report. Reading (UK): Tate & Lyle Group Research & Development. Submitted to
the World Health Organization by Tate & Lyle. Cited In: JECFA 1989b.
Rosenkranz, H.S., Leifer, Z., 1980. Determining the DNA-modifying activity of
chemicals using DNA- polymerase-deficient Escherichia coli. In: de Serres, F.J.,
Hollaender, A. (Eds.), Chemical Mutagens: Principles and Methods for Their
Detection, vol. 6. Plenum Press, New York (NY), pp. 109e147.
Rother, K.I., Sylvetsky, A.C., Schiffman, S.S., 2015. Non-nutritive sweeteners in breast
milk: perspective on potential implications of recent findings. Arch. Toxicol. 89,
2169e2171.
Rulis, A.M., Levitt, J.A., 2009. FDA'S food ingredient approval process: safety
assurance based on scientific assessment. Regul. Toxicol. Pharmacol. 53, 20e31.
Saada, H.N., Mekky, N.H., Eldawy, H.A., Abdelaal, A.F., 2013. Biological effect of
sucralose in diabetic rats: special issue on advanced research in food safety.
Food Nutr. Sci. 4, 82e89.
Salamone, M., Heddle, J., Stuart, E., Katz, M., 1980. Towards an improved micronucleus test. Studies on 3 model agents, mitomycin C, cyclophosphamide and
dimethylbenzanthracene. Mutat. Res. 74, 347e356.
Sasaki, Y.F., Kawaguchi, S., Kamaya, A., Ohshita, M., Kabasawa, K., Iwama, K.,
Taniguchi, K., Tsuda, S., 2002. The comet assay with 8 mouse organs: results
with 39 currently used food additives. Mutat. Res. 519, 103e119.
Schiffman, S.S., 2012. Rationale for further medical and health research on highpotency sweeteners. Chem. Senses 37, 671e679.
Schiffman, S.S., Abou-Donia, M.B., 2012. Sucralose revisited: rebuttal of two papers
about Splenda safety. Regul. Toxicol. Pharmacol. 63, 505e508, 509-513 (reply).
Schiffman, S.S., Rother, K.I., 2013. Sucralose, a synthetic organochlorine sweetener:
overview of biological issues. J. Toxicol. Environ. Health B Crit. Rev. 16, 399e451.
Schmid, W., 1975. The micronucleus test. Mutat. Res. 31, 9e15.
SCF, Scientific Committee on Food, 1989. Report of the scientific committee for food
on sweeteners (opinion expressed on 11 december 1987 and 10 November
1988). In: Food Science and Techniques [Internet]. Commission of the European
Communities (EEC), Scientific Committee for Food (SCF), Brussels (Belgium),
pp. 19e37 (Reports of the Scientific Committee for Food, 21st Series; EUR 11617
EN), Available from: https://ec.europa.eu/food/sites/food/files/safety/docs/scicom_scf_reports_21.pdf.
SCF, Scientific Committee on Food, 2000a. Opinion of the Scientific Committee on
Food on Sucralose (Opinion Expressed by the SCF on 7 September 2000)
[Internet]. European Commission, Health & Consumer Protection DirectorateGeneral, Scientific Committee on Food (SCF), Brussels (Belgium) [cited 2017
Jan 20]. (SCF/CS/ADD/EDUL/190 Final). Available from: https://ec.europa.eu/
food/sites/food/files/safety/docs/sci-com_scf_out68_en.pdf.
SCF, Scientific Committee on Food, 2000b. Annex I. Studies on glucose homeostasis
in healthy humans and diabetic volunteers. In: Opinion of the Scientific Committee on Food on Sucralose (Opinion Expressed by the SCF on 7 September
2000) [Internet]. European Commission, Health & Consumer Protection
Directorate-General, Scientific Committee on Food (SCF), Brussels (Belgium),
pp. 21e22 [cited 2017 Jan 20]. (SCF/CS/ADD/EDUL/190 Final). Available from:
https://ec.europa.eu/food/sites/food/files/safety/docs/sci-com_scf_out68_en.
pdf.
Sharma, A., Panwar, S., Singh, A.K., Jakhar, K.K., 2007. Studies on the genotoxic effects of sucralose in laboratory mice. Indian J. Anim. Res. 41, 1e8.
Sims, J., Roberts, A., Daniel, J.W., Renwick, A.G., 2000. The metabolic fate of sucralose
in rats. Food Chem. Toxicol. 38 (Suppl. 2), S115eS121.
Smith, M.T., Guyton, K.Z., Gibbons, C.F., Fritz, J.M., Portier, C.J., Rusyn, I.,
DeMarini, D.M., Caldwell, J.C., Kavlock, R.J., Lambert, P., et al., 2016. Key characteristics of carcinogens as a basis for organizing data on mechanisms of
carcinogenesis. Environ. Health Perspect. 124, 713e721.
Soffritti, M., Belpoggi, F., Manservigi, M., Tibaldi, E., Lauriola, M., Falcioni, L., Bua, L.,
2010. Aspartame administered in feed, beginning prenatally through life span,
induces cancers of the liver and lung in male Swiss mice. Am. J. Ind. Med. 53,
1197e1206.
Soffritti, M., Padovani, M., Tibaldi, E., Falconi, L., Manservisi, F., Lauriola, M., Bua, L.,
Manservigi, M., Belpoggi, F., 2016. Sucralose administered in feed beginning
prenatally through lifespan, induces hematopoietic neoplasias in male Swiss
mice. Int. J. Occup. Environ. Health 22, 7e17.
€ pfer, A., Drewe, J., Beglinger, C., 2011. Effects of carbohydrate
Steinert, R.E., Frey, F., To
sugars and artificial sweeteners on appetite and the secretion of gastrointestinal satiety peptides. Br. J. Nutr. 105, 1320e1328.
Stellingwerff, T., Godin, J.P., Beaumont, M., Tavenard, A., Grathwohl, D., van
Bladeren, P.J., Kapp, A.F., le Coutre, J., Damak, S., 2013. Effects of pre-exercise
sucralose ingestion on carbohydrate oxidation during exercise. Int. J. Sport
Nutr. Exerc Metab. 23, 584e592.
Suez, J., Korem, T., Zeevi, D., Zilberman-Schapira, G., Thaiss, C.A., Maza, O., Israeli, D.,
Zmora, N., Gilad, S., Weinberger, A., et al., 2014. Artificial sweeteners induce
glucose intolerance by altering the gut microbiota. Nature 514, 181e186.
Suez, J., Korem, T., Zilberman-Schapira, G., Segal, E., Elinav, E., 2015. Non-caloric
artificial sweeteners and the microbiome: findings and challenges. Gut Microbes 6, 149e155.
Sylvetsky, A.C., Gardner, A.L., Bauman, V., Blau, J.E., Garraffo, H.M., Walter, P.J.,
Rother, K.I., 2015. Nonnutritive sweeteners in breast milk. J. Toxicol. Environ.
Health A 78, 1029e1032.
Sylvetsky, A.C., Brown, R.J., Blau, J.E., Walter, M., Rother, K.I., 2016. Hormonal responses to non-nutritive sweeteners in water and diet soda. Nutr. Metab.
(Lond.) 13. http://dx.doi.org/10.1186/s12986-016-0129-3, 71 [8pp].
Temizkan, S., Deyneli, O., Yasar, M., Arpa, M., Gunes, M., Yazici, D., Sirikci, O.,
Haklar, G., Imeryuz, N., Yavuz, D.G., 2015. Sucralose enhances GLP-1 release and
lowers blood glucose in the presence of carbohydrate in healthy subjects but
not in patients with type 2 diabetes. Eur. J. Clin. Nutr. 69, 162e166.
Tice, R.R., Agurell, E., Anderson, D., Burlinson, B., Hartmann, A., Kobayashi, H.,
Miyamae, Y., Rojas, E., Ryu, J.-C., Sasaki, F., 2000. Single cell gel/comet assay:
guidelines for in vitro and in vivo genetic toxicology testing. Environ. Mol.
Mutagen. 35, 206e221.
Tollefsen, K.E., Nizzetto, L., Huggett, D.B., 2012. Presence, fate and effects of the
intense sweetener sucralose in the aquatic environment. Sci. Total Environ. 438,
510e516.
Tordoir, W.F., van Sittert, N.J., 1994. Organochlorines [chapter 7]. Toxicology 91,
51e57.
US EPA, U.S. Environmental Protection Agency, 2012. Update on Ramazzini Institute
Data in IRIS Assessments [Internet]. U.S. Environmental Protection Agency (US
EPA), Washington (DC) [cited 2017 Jan 20]. Available from: https://www.epa.
gov/iris/update-ramazzini-institute-data-iris-assessments [Mar. 8, 2012].
US FDA, U.S. Food and Drug Administration, 1998. Food additives permitted for
direct addition to food for human consumption: sucralose; final rule [21CFR
Part 172; Docket No. 87F-0086]. Fed. Regist. (US) 63, 16417e16433.
US FDA, U.S. Food and Drug Administration, 1999. Food additives permitted for
direct addition to food for human consumption: sucralose; final rule [21CFR
Part 172; Docket No. 99F-0001]. Fed. Regist. (US) 64, 43908e43909.
US FDA, U.S. Food and Drug Administration, 2000. Guidance for Industry and Other
Stakeholders: Toxicological Principles for the Safety Assessment of Food Ingredients: Redbook 2000 [Updated to July, 2007] [Internet]. U.S. Food and Drug
Administration (US FDA), Center for Food Safety and Applied Nutrition (CFSAN),
Silver Spring (MD) [cited 2017 Jan 20]. Available from: http://www.fda.gov/
Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/
IngredientsAdditivesGRASPackaging/ucm2006826.htm.
US FDA, U.S. Food and Drug Administration, 2015. Additional Information about
High-intensity Sweeteners Permitted for Use in Food in the United States
[Internet] U.S. Food and Drug Administration (US FDA). Center for Food Safety &
Applied Nutrition (CFSAN) [cited 2017 Jan 20]. Available from: http://www.fda.
gov/Food/IngredientsPackagingLabeling/FoodAdditivesIngredients/ucm397725.
htm [Page Last Updated: 05/26/2015].
US FDA, U.S. Food and Drug Administration, 2016. Part 101dfood Labeling.
x101.60dnutrient content claims for the calorie content of foods. In: U.S. Code
of Federal Regulations (CFR). Title 21: Food and Drugs (Food and Drug
Administration) [Internet]. U.S. Government Printing Office (GPO), Washington
(DC) [cited 2017 Jan 20]. Available from: https://www.accessdata.fda.gov/
scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr¼101.60.
van Eyk, E.D., 2015. The effect of five artificial sweeteners on Caco-2, HT-29 and
HEK-293 cells. Drug Chem. Toxicol. 38, 318e327.
Van Loco, J., Vandevijvere, S., Cimenci, O., Vinkx, C., Goscinny, S., 2015. Dietary
exposure of the Belgian adult population to 70 food additives with numerical
ADI. Food Control 54, 86e94.
Viberg, H., Fredriksson, A., 2011. Neonatal exposure to sucralose does not alter
biochemical markers of neuronal development or adult behavior. Nutrition
2011 (27), 81e85.
Wang, G.J., Tomasi, D., Convit, A., Logan, J., Wong, C.T., Shumay, E., Fowler, J.S.,
Volkow, N.D., 2014. BMI modulates calorie-dependent dopamine changes in
accumbens from glucose intake. PLoS One 9. http://dx.doi.org/10.1371/journal.pone.0101585 e101585.
WHO, World Health Organization, 1987. Principles for the Safety Assessment of
Food Additives and Contaminants in Food [Internet]. World Health Organization (WHO)/International Programme on Chemical Safety (IPCS)/United Nations
Environment Programme (UNEP), Geneva [cited 2017 Jan 20]. (Environmental
Health Criteria, no 70). Available from: http://www.inchem.org/documents/
ehc/ehc/ehc70.htm.
WHO, World Health Organization, 2009. Principles and Methods for the Risk
Assessment of Chemicals in Food [Internet]. Food and Agriculture Organization
of the United Nations (FAO)/Geneva: World Health Organization (WHO)/International Programme on Chemical Safety (IPCS) (Environmental Health Criteria,
no 240), Rome [cited 2017 Jan 20]. Available from: http://www.who.int/
foodsafety/publications/chemical-food/en/.
Wiet, S.G., Beyts, P.K., 1992. Sensory characteristics of sucralose and other high
intensity sweeteners. J. Food Sci. 57, 1014e1019. Cited In: Grotz & Munro 2009.
Williams, G.M., 1977. Detection of chemical carcinogens by unscheduled DNA
B.A. Magnuson et al. / Food and Chemical Toxicology 106 (2017) 324e355
synthesis in rat liver primary cell cultures. Cancer Res. 37, 1845e1851.
Williams, G.M., Iatropoulos, M.J., 2001. Principles of testing for carcinogenic activity.
In: Hayes, W. (Ed.), Principles and Methods of Toxicology, fourth ed. Taylor &
Francis, Philadelphia (PA), pp. 959e1000.
Wood, S.G., John, B.A., Hawkins, D.R., 2000. The pharmacokinetics and metabolism
of sucralose in the dog. Food Chem. Toxicol. 38 (Suppl. 2), S99eS106.
Wu, T., Zhao, B.R., Bound, M.J., Checklin, H.L., Bellon, M., Little, T.J., Young, R.L.,
Jones, K.L., Horowitz, M., Rayner, C.K., 2012. Effects of different sweet preloads
on incretin hormone secretion, gastric emptying, and postprandial glycemia in
healthy humans. Am. J. Clin. Nutr. 95, 78e83.
Wu, T., Bound, M.J., Zhao, B.R., Standfield, S.D., Bellon, M., Jones, K.L., Horowitz, M.,
Rayner, C.K., 2013a. Effects of a D-xylose preload with or without sitagliptin on
gastric emptying, glucagon-like peptide-1, and postprandial glycemia in type 2
355
diabetes. Diabetes Care 36, 1913e1918.
Wu, T., Bound, M.J., Standfield, S.D., Bellon, M., Young, R.L., Jones, K.L., Horowitz, M.,
Rayner, C.K., 2013b. Artificial sweeteners have no effect on gastric emptying,
glucagon-like peptide-1, or glycemia after oral glucose in healthy humans.
Diabetes Care 36, e202ee203.
Xu, Z., Knight, R., 2015. Dietary effects on human gut microbiome diversity. Br. J.
Nutr. 113 (Suppl. l), S1eS5.
Yin, W., Carballo-Jane, E., McLaren, D.G., Mendoza, V.H., Gagen, K., Geoghagen, N.S.,
et al., 2012. Plasma lipid profiling across species for the identification of optimal
animal models of human dyslipidemia. J. Lipid Res. 53, 51e65.
Young, D.A., Bowen, W.H., 1990 Aug. The influence of sucralose on bacterial metabolism. J. Dent. Res. 69 (8), 1480e1484. PubMed PMID: 2143512.