Trehalose Produced by A Novel Enzymatic Process

TREHALOSE
produced by a novel enzymatic process
Dossier prepared and submitted on behalf of

Hayashibara Co., Ltd., 2-3 Shimoishii 1-chome, Okayama 700,
Japan for evaluation pursuant to the EU Novel Foods Regulation
(258/97) by the UK Advisory Committee on
Novel Foods and Processes
Under the "UK Novel Foods and Novel Food Ingredients (Amendment) Regulations 1999"
and the corresponding "Guidelines on the disclosure of information and confidential-
ity in respect of applications made to the ACNFP", this dossier and the pertinent
parts of the pivotal safety studies are to be made public. Certain information re-
lated to marketing and the production process is confidential and has been removed
from this version. However, it is contained in the original dossier and has been
given to the Committee for their consideration.
The animal studies which were commissioned by the applicant in order to fulfill the
toxicological requirements, were conducted with the approval and under the supervi-
sion of the competent Animal Welfare Committee.
Author: Albert Bär PhD
Date: May 18, 2000
Bioresco Ltd. Bundesstrasse 29 CH-4054 Basel Telephone ++41 61 273 77 00 Telefax ++41 61 273 77 03
e-mail: [email protected] Internet: http://www.bioresco.ch
CONTENTS
1. Introduction ........................................... 5
2. "Specifications of the NF" .............................. 8
2.1. Specifications ................................. 8
2.2. Physico-chemical properties ..................... 9
2.3. Other properties relevant to the use of treha-
lose as a food ingredient ....................... 9
2.3.1. Sweetness ............................. 9
2.3.2. Hygroscopicity ........................ 10
2.3.3. Chemical stability .................... 10
2.3.4. Thermal stability...................... 10
3. "Effect of the production process applied to the NF" .... 11
3.1. Novelty of the process .......................... 11
3.2. General description of the process .............. 12
3.3. Detailed description of the process ............. 13
3.3.1. Starch liquefaction .................... 13
3.3.2. Saccharification, production of trehalose 13
3.3.3. Purification ........................... 13
3.3.4. Concentration and crystallization ...... 13
3.4. Safety of raw material, chemicals, and enzymes
used in the process ............................. 14
3.5. Potential impurities resulting from the
production process .............................. 19
3.5.1. Impurities from raw material, process
chemicals and enzymes ................... 19
3.5.2. By-products from enzymatic saccharifi-
cation .................................. 20
3.5.3. Purification steps ...................... 21
4. "History of the organism used as the source of the NF" .. 23
5. "Anticipated intake / extent of use of the NF" .......... 24
5.1. Intended uses in food ........................... 24
5.2. Current food applications of trehalose in Japan . 26
TREHALOSE_EU_MANU1_PUBLIC_120500.doc 2/79
5.3. Estimated daily intake .......................... 26
6. "Information from previous human exposure to the NF or
its source ............................................. 30
6.1. Occurrence in nature ............................ 30
6.1.1. Occurrence in plants .................... 30
6.1.2. Occurrence in animals ................... 31
6.2. Consumption of trehalose added to food in Japan . 32
6.3. Intake of high single doses in trehalose tole-
rance tests ..................................... 33
7. "Nutritional information on the NF" ..................... 34
8. "Microbiological information on the NF" ................. 35
9. "Toxicological information on the NF" ................... 36
9.1. General remarks ................................. 36
9.2. Metabolic studies on trehalose .................. 37
9.2.1. Digestion and absorption of ingested
trehalose ............................... 37
9.2.2. Metabolism of parenterally administered
trehalose ............................... 41
9.3. Toxicological studies ........................... 43
9.3.1. Acute toxicity .......................... 43
9.3.2. Short-term studies of toxicity .......... 44
9.3.3. Genotoxicity ............................ 47
9.3.4. Developmental toxicity .................. 48
9.3.5. Special studies ......................... 51
9.3.5.1. Ocular irritation ............ 51
9.4. Tolerance of trehalose in humans ................ 51
9.4.1. Intestinal tolerance in healthy volunteers 51
9.4.2. Impaired intestinal tolerance due to
intestinal trehalase deficiency.......... 53
10. Summary ................................................ 57

11. References ............................................. 61
FIGURES 1 – 3
TABLES 1 – 4
Annex 1 Data requirements for the evaluation of trehalose produced by

a novel process (according to Commission Recommendation
97/618/EC and ACNFP decision tree)
Annex 2 Draft specifications
Annex 3 Analytical data of representative batches of trehalose
Annex 4 Critical Control Points and Control Standards of the produc-

tion process
Specifications of applied raw material, process chemicals,
enzymes and ion exchange resins
Annex 5 Current applications and use levels of trehalose in Japanese

foods
Annex 6 Projection of trehalose intake by the dietary survey approach
1. Introduction
Trehalose is a naturally occurring disaccharide which consists of

two glucose molecules linked in a 1,1-position by an α-glycosidic
bond (Figure 1)1. It is produced in bacteria and yeast cells, fungi,
algae, and a few higher plants. In many insects it is produced as a
major blood sugar and as a reserve carbohydrate during periods of
dehydration and freezing. Trehalose also occurs at low concentra-
tions in a number of foods which are consumed as part of a regular
diet (mushrooms, bread, fermented beverages, honey).
The Advisory Committee on Novel Foods and Processes (ACNFP) assessed

the safety of trehalose, extracted from yeast, in 1990. The intended
food applications that were considered by the Committee at that
time, included the use of trehalose for the stabilization of certain
foodstuffs during the drying process and upon rehydration (e.g.,
milk powder, dry soups). On advice of the ACNFP, the Committee on
Toxicity (COT), and the Food Advisory Committee (FAC), trehalose was
accepted for use in foods (except infant formulae and follow-on for-
mulae) in April 1991 (MAFF, 1991; ACNFP, 1990, 1991).
Hayashibara Co., Ltd., Okayama (Japan) has developed a novel enzy-

matic production process. In this process, liquefied starch serves
as raw material. Four enzymes which have not hitherto been used in
food or food manufacturing in the EU, are applied in this process.
The two pivotal enzymes are produced by a strain of Arthrobacter
1
Synonyms: α,α-trehalose, α-D-trehalose, D-(+)–trehalose, α-trehalose, D-trehalose, mycose
("mushrooms sugar"). Brand names: TrehaoseTM, TrehaTM (Hayashibara Co., Ltd.)
ramosus. One enzyme converts the terminal (reducing) maltosyl unit
of maltooligosaccharides (DP >3) to a trehalose unit. The other en-
zyme hydrolyzes the α-1,4 glycosidic bond adjacent to the trehalose
unit thereby liberating trehalose (Figure 2)2. In order to increase
the yield of the process, isoamylase from Pseudomonas amyloderamosa
and cyclodextrin glucosyltransferase (CGTase) from Bacillus stearo-
thermophilus are used as ancillary enzymes. None of the source mi-
croorganisms of these enzymes is genetically modified.
The present application for authorization of trehalose produced by

a novel process was prepared according to the EU Commission's
guidelines on the scientific information necessary to support ap-
plications for placing on the market of novel foods or novel food
ingredients (Commission Recommendation (97/618/EC) and the Guidance
Notes on Novel Foods and Novel Food Ingredients Legislation of the
UK Ministry of Agriculture, Fisheries and Food (June 1999). Treha-
lose was identified as belonging to class 6 (foods produced using a
novel process. Using the ACNFP's Computerized Decision Tree pack-
age, the data requirement was identified. The outcome of this
evaluation is shown in Annex 1.
For evaluation of a class 6 product the following information is

required (numbering according to Commission Guideline, Table II):
I. Specification of the Novel Food (NF)
II. Effect of the production process applied to the NF
III. History of the organism used as the source of the NF
IX. Anticipated intake/extent of use of the NF
2
The biosynthesis of trehalose in Rhizobia and certain other microorganisms relies on
the same enzymatic steps (Kato, 1999; Streeter & Bhagwat, 1999).
X. Information from previous human exposure to the NF or its
source
XI. Nutritional information on the NF
XII. Microbiological information on the NF
XIII. Toxicological information on the NF
The information presented in this dossier is structured according to

this scheme.
2. "Specifications of the NF"
2.1. Specifications
Specifications as submitted to JECFA for evaluation at its 55th meeting

(June 6 – 15, 2000) are shown in Annex 2 of this document. Unless
mentioned otherwise, the applied test methods are those specified in
the JECFA Guide to Specifications (FAO Food and Nutrition Paper 5 Rev.
2).
Analyses of representative batches of trehalose demonstrate that

trehalose produced at commercial scale complies with these specifica-
tions (Annex 3). Every batch of trehalose is tested for compliance
with these specifications and is released for use in food only if the
product meets all specified criteria.
In the pivotal safety studies (genotoxicity tests, embryotoxic-

ity/teratogenicity studies in rats and rabbits, 2-generation reprotox
study in rats, 3-month toxicity study in mice) trehalose was used
which has been produced by the enzymatic process described in this
dossier and complies with the specifications shown in Annex 2.
2.2. Physico-chemical properties
Solubility in water
(g anhydrous trehalose/100ml solution at 20°C): 43.03
(g trehalose dihydrate/100ml solution at 20°C): 46.63

20
Specific rotation [α] D
(5% trehalose dihydrate, H2O): +199°4
Melting point (°C): 210.5 (anhydrous)5; 97.0 (dihydrate)6
Heat of solution (kJ/mol): + 20.7 (trehalose dihydrate)
2.3. Other properties relevant to the use of trehalose as a

food ingredient
2.3.1. Sweetness
Trehalose in aqueous solution (at concentrations from about 10-34%

anhydrous trehalose) has a sweetness of about 40-45% relative to
that of sucrose (Birch et al., 1970; Miyake, unpublished report).
Correspondingly, the concentration at which a solution of trehalose
is just perceived as sweet, is about two-times higher than that of
sucrose (43.6 and 19 mM, respectively) (Lee & Birch, 1975). On the
3
Lammert et al., 1998. A slightly lower solubility was reported by Sugimoto, 1995 and
Miller et al., 1997.
4
Values of +177° to +199° have been reported in the literature (Birch, 1963; Koch & Koch,
1925).
5
Values of 203–216°C are given in the literature (Birch, 1965; Koch & Koch, 1925). More
recently, the existence of three different polymorphs of anhydrous trehalose with dif-
ferent melting points has been reported, namely trehalose-β, trehalose-α, and trehalose-
γ with m.p. 215, 125 and 120-130°C, respectively (Sussich et al., 1998).
6
Values of 94-100°C have been reported in the literature (Birch, 1963).
other hand, the sweetness of trehalose has a longer persistence
than that of sucrose (Portmann & Birch, 1995).
2.3.2. Hygroscopicity
Trehalose (dihydrate) has a very low hygroscopicity, similar to

that of lactose (Miyake, unpublished report). Anhydrous (amorphous)
trehalose absorbs moisture avidly until moisture content reaches a
plateau at a concentration which is very similar to that of water
in the dihydrate (10.5%) (Iglesias et al., 1997).
2.3.3. Chemical stability
Trehalose (dihydrate) is stable on storage at ambient temperature.

In aqueous solution, trehalose is stable (i.e., is not hydrolyzed)
in a pH range from 2 to 10 (4% solution, 100°C). As a non-reducing
sugar, trehalose does not undergo Maillard reactions (Miyake, un-
published report).
2.3.4. Thermal stability
Under the temperature conditions applied typically during the proc-

essing and storage of food, trehalose is stable. As a non-reducing
sugar, trehalose does not caramelize at elevated temperature.
3. "Effect of the production process applied to the NF"
3.1. Novelty of the process
Trehalose may be obtained by extraction from yeast. Trehalose ob-

tained in this way was the subject of the evaluation by the ACNFP
in 1990 (ACNFP 1990; 1991).
The subject of the present submission is a novel enzymatic process

by which trehalose is produced from food-grade starch. Four enzymes
are used in this process which have not been used hitherto in the
EU in food or a food manufacturing process.
Trehalose produced by this enzymatic process is chemically identi-

cal to trehalose extracted from yeast as demonstrated by identical
melting points and specific rotation7.
7
There exist three different isomers of trehalose: α,α-trehalose (trehalose) which is the
subject of the present application; α,β-trehalose (neo-trehalose); and β,β-trehalose (iso-
trehalose). The three isomers have distinctly different specific rotations and melting
points (Birch, 1963).
3.2. General description of the process
In a first step, starch is liquefied by treatment with a thermo-

philic α-amylase. In a second step, the obtained maltooligosaccha-
rides are treated concurrently with maltooligosyl trehalose syn-
thase (acronym: "MTSase")8, maltooligosyl trehalose trehalohydro-
lase (acronym: "MTHase")9, isoamylase, and cyclodextrin glucano-
transferase (CGTase). The reactions catalyzed by MTSase and MTHase
are depicted in Figure 2. Isoamylase is used as "debranching" en-
zyme, i.e., for cleaving α-1,6 glycosidic bonds of the starch mole-
cule. CGTase is added in order to recycle maltose back into the
process (Mandai et al., 1999). CGTase catalyzes intermolecular
transglycosylation reactions in which maltose acts as an acceptor
molecule (Tonkova, 1998; Brunet et al., 1998). Glucoamylase and α-
amylase are added to release any remaining trehalose moieties and
to degrade any remaining oligosaccharides and maltose to glucose.
After completion of the trehalose forming enzymatic step (sacchari-
fication), the reaction mixture is decolorized with activated car-
bon, filtered using diatomaceous earth and perlite as filtering
aid, de-ionized with ion exchange resins, and concentrated by
evaporation. Trehalose is obtained by crystallization.
8
Categorized as α-glucanotransferase in Japan. The systematic name is (1à4)-α-D-glucan
1-α-D-glucosylmutase (EC 5.4.99.15).
9
Categorized as α-amylase in Japan. The systematic name is 4-α-D [(1à4)-α-D-glucano]
trehalose glucanohydrolase (EC 3.2.1.141).
3.3. Detailed description of the process
The flow scheme of the trehalose production process is presented in

Figure 3. The critical control points and control standards are
shown in Annex 4.
3.3.1. Starch liquefaction
[This chapter reports on details of the production process for

which the applicant claims confidentiality]
3.3.2. Saccharification, production of trehalose

3.3.3. Purification

3.3.4. Concentration and crystallization

3.4. Safety of raw material, chemicals, and enzymes used in the
process
Food-grade starch is used as the starting material for the treha-

lose production process.
The chemicals used as processing aids in the manufacturing process

(calcium carbonate, calcium chloride, hydrochloric acid, sodium hy-
droxide, sodium carbonate, activated carbon, perlite, diatomaceous
earth) have a purity that makes them suitable for use in the pres-
ent process (specifications are presented in Annex 4).
The applied cation and anion exchange resins comply with pertinent
US regulations [21 CFR § 173.25 (ion exchange resins)] and the
Council of Europe Resolution AP (97)1 (specifications are presented
in Annex 4).
The thermophilic α-amylase (EC 3.2.1.1), which is used for starch

liquefaction, is obtained from Bacillus licheniformis. α-Amylase
from this source microorganism has an "ADI not specified" (WHO,
1987). The enzyme applied in the present process complies with
JECFA specifications (FAO, 1992b). In the US, a mixed carbohydrase
and protease enzyme preparation from B. licheniformis has been af-
firmed as generally recognized as safe (GRAS) for use in the pro-
duction of certain foods (21 CFR § 184.1027).
The MTSase (EC 5.4.99.15) and MTHase (EC 3.2.1.141), which are the
crucial enzymes for the enzymatic synthesis of trehalose, are ob-
tained from Arthrobacter ramosus (strain S34). The genus Arthrobac-
ter is widely distributed in nature. Bacteria of this genus are
predominant members of the soil microbiota (for taxonomic charac-
teristics see Kvasnikov et al., 1986; Masiak et al., 1991). A few
species of Arthrobacter have been identified in a small number of
clinical samples but in no instance was A. ramosus isolated (Funke
et al., 1996). Arthrobacter spp. is generally considered to be non-
pathogenic and belongs consequently to risk group 1 of the German
"Gentechnik-Sicherheitsverordnung" (GenTSV), i.e., Regulation on
the Safety of Gentechnology. Group 1 comprises microorganisms which
present no risk for humans and vertebrates. MTSase and MTHase are
produced by cultivating A. ramosus (strain S34) in a standard me-
dium containing food-grade maltose, food-grade mineral salts, a ni-
trogen source (peptone), and yeast extract. When the desired enzyme
activity has been reached in the fermentation broth, albumen lyso-
zyme is added in order to lyse the bacterial cells. The cells and
particles are then removed by membrane filtration (pore size 0.04
µm). The filtrate is purified by cross-flow ultra-filtration (cut-
off molecular weight of 10,000 daltons).
MTSase and MTHase have been purified and characterized. Their

amino-acid composition has been determined (Nakada et al., 1995
a,b). The genetic sequence which encodes for the two enzymes has
been identified as well (Maruta et al., 1996). Considering the gen-
erally recognized safety of Arthrobacter spp. and the efficacious
purification of trehalose which removes proteinaceous, glycidic,
and ionic impurities almost quantitatively10, the use of MTSase and
MTHase as processing aids in the production of trehalose appears to
be acceptable even in the absence of specific toxicological studies
on these two enzymes (see Annex 3 for data demonstrating the ab-
sence of residual enzymatic activity in trehalose).
Isoamylase (EC 3.2.1.68) is used in the trehalose production proc-

ess as debranching enzyme. The enzyme is obtained from a mutant
strain of Pseudomonas amyloderamosa11. The extra-cellular enzyme is
inducibly produced by the presence of maltose, maltodextrin or
starch (Yokobayashi, 1988). For production of the enzyme at an ap-
propriate scale, P. amyloderamosa is cultivated at about 30°C in a
standard medium containing food-grade maltose, food-grade mineral
salts, a nitrogen source (peptone), and yeast extract. When the de-
sired enzyme activity has been reached, the fermentation broth is
filtered (pore size 0.04 µm) and purified by cross-flow ultra-
filtration (cut-off molecular weight of 10,000 daltons).
In anticipation of a direct food use of isoamylase which is unre-

lated to the subject of this application but which would require a
full safety assessment of the enzyme, the safety of the isoamylase
preparation was examined in standard Ames tests, an acute toxicity
test in mice, and a 90-day oral toxicity study in rats. The Ames
tests were conducted in S. typhimurium strains TA 1535, TA 1537, TA
10
Analyses for protein of 5 different batches of trehalose using the Kjeldahl method gave
values of 7-8 mg protein/100 g trehalose. For comparison, the starch used as a starting
material has a protein content of <0.35%. The content of glycosidic impurities is de-
scribed in chapter 3.5.2. The content of inorganic substances, as measured by the
method for total ash, is not more than 50 mg/100 g trehalose (see chapter 3.5.1).
11
[This foot-note reports on details of the production process for which the applicant
claims confidentiality]
98, and TA 100, as well as in E. coli WP2 uvrA with and without
metabolic activation. At levels of up to 5000 µg/plate, the test
preparation was not toxic to the tester strains and was not muta-
genic (van Delft, 1999).
The acute toxicity of P. amyloderamosa (wet cells), culture fil-

trate, and purified isoamylase was tested in mice. Groups of 10
male and 10 female mice received, by gavage, the wet cells at doses
of 38.2 and 66.0 g/kg bw, the culture filtrate at doses of 34.7 and
60 ml/kg bw, and the isoamylase at doses of 12.0, 13.2, 14.5, 16.0,
17.6, 19.3 and 21.3 g/kg bw. The purified isoamylase was adminis-
tered as a 40% suspension in water. No mortalities or clinical
signs were observed after administration of the wet cells and the
culture filtrate (Morimoto et al., 1979 a,b). In mice receiving the
isoamylase suspension, mortalities occurred in a dose-dependent way
starting from a dose of 14.5 g/kg bw mostly within the first 5
hours after gavage. The LD50 of purified isoamylase was about 17
g/kg bw. It is conceivable that the mortalities were the unspecific
consequence of the forced ingestion of a thick suspension of pro-
tein (Morimoto et al., 1979 c).
In a subchronic (90-day) toxicity study 4 groups of rats

(20/sex/group) received stabilized isoamylase preparation by gavage
at levels of 0, 2.5, 5 and 10 ml/kg bw. The preparation had a pro-
tein content of 19 mg/ml. In order to administer the same volume of
liquid (10 ml/kg bw) to all groups, the isoamylase preparation was
diluted for the low and mid-dose group with control solution. The
control group received 10 ml/kg bw of control solution. The general
condition and behavior of the animals were not affected by the
treatment. Body weights did not differ between treatment groups and
controls. Six rats died during the study. In 4 cases, the death re-
sulted from a gavaging error (confirmed on autopsy). In the other
two cases (one male rat of the mid and high dose group each) the
cause of death could not be determined because of cannibalism
and/or autolysis. Food intake, hematological parameters, clinico-
chemical parameters, semi-quantitative urinalysis, organ weights at
termination, and histopathological examination of organs and tissue
did not reveal changes that could be attributed to the treatment.
It was concluded the high-dose level tested (10 ml/kg bw) was a no-
observed-adverse-effect level (NOAEL) (Lina, 1999).
On the basis of these results, the isoamylase preparation from

Pseudomonas amyloderamosa appears to be fit for use in the treha-
lose production process.
CGTase (EC 2.4.1.19) is used in the trehalose production to in-

crease the yield. The applied enzyme is obtained from a strain of
Bacillus stearothermophilus. The safety of CGTase (from other
source organisms) has been evaluated by JECFA in the context of the
safety assessments of β- and γ-cyclodextrin (WHO, 1999). The safety
of B. stearothermophilus as a source organism has been considered
when the safety of α-amylase from this organism was assessed (WHO,
1991a). The non-pathogenic and non-toxicogenic properties of this
microorganism were also appreciated in the evaluation of "malto-
genic" amylase (WHO, 1998) and α-amylase (21 CFR 184.1012). It fol-
lows from these evaluations that CGTase from B. stearothermophilus
can be safely used in the production of trehalose.
Glucoamylase (EC 3.2.1.3) from Aspergillus niger and α-amylase (EC
3.2.1.1) from Bacillus subtilis are used in the last step of the
trehalose production process to degrade remaining oligosaccharides
and maltose. The safety of these enzymes and source organisms has
been evaluated by JECFA (WHO, 1972, 1988, 1991b) and tentative or
final specifications have been laid down (FAO, 1992a, 1993). In the
US, both enzymes are authorized for the treatment of wine (27 CFR §
24.246).
3.5. Potential impurities resulting from the production process
1.5.1. Impurities from raw material, process chemicals and en-

zymes
The food-grade starch contains small amounts of protein and inor-

ganic salts. These natural impurities are removed almost completely
during the different purification steps10.
All applied chemicals are food-grade and/or have otherwise a high

purity (see Annex 4). Residues of inorganic material would be de-
tected in the tests for total ash. Lead would be detected in the
specific test for lead. Specifications for arsenic are not included
because there is no reason to expect the presence of this heavy
metal in the product [JECFA, 53rd meeting, June 1999, Summary and
Conclusions (item 5)].
The different enzyme preparations contain residues of the fermenta-
tion broth and by-products of the microbial metabolism. All ingre-
dients of the fermentation broth are ordinary nutritional sub-
stances (carbohydrates, peptides, minerals). The bacterial fermen-
tation of the carbohydrates which are added as a carbon source,
yields typically short chain fatty acids. Toxic by-products are not
known to be produced by the employed microorganisms. All substances
carrying cationic or anionic functional groups are expected to be
removed by the de-ionization step using ion-exchange resins. An ad-
ditional purification is obtained by the treatment with activated
carbon. Any remaining organic impurities are likely to be removed
during the crystallization of trehalose. Inorganic residues would
be detected by the test for total ash.
3.5.2. By-products from enzymatic saccharification
The treatment of the enzymatic depolymerisation products of starch

(maltooligosaccharides) with MTSase, MTHase, isoamylase and CGTase
results in the formation of trehalose, and smaller amounts of glu-
cose, maltose and some tri- and higher maltooligosaccharides. The
final treatment of this reaction mixture with glucoamylase and α-
amylase removes maltose and the oligosaccharides almost completely.
The resulting glucose and any remaining maltose and oligosaccha-
rides are removed during the crystallization of trehalose. On HPLC
analysis of crystalline trehalose produced according to this proc-
ess, only three glycosidic impurities could be detected, namely
glucose, glucosyltrehalose (syn: α-D-maltosyl- α-D-glucoside)12, and
12
The physicochemical properties of glucosyltrehalose have been determined. The LD50 of
this trisaccharide is > 50 g/kg bw in mice (Tabuchi et al., 1999). An enzyme prepara-
tion from rat small intestine hydrolyzed the substance to glucose and trehalose (Kuri-
moto et al., 1997b; Mandai et al., 1999).
α-D-isomaltosyl- α-D-glucoside13. In a trehalose batch with a purity
of 99.1% (as determined by HPLC) these three by-products occurred
at levels of 0.5, 0.3 and 0.1%, respectively. Other mono-, di- and
oligosaccharides were not detected.
3.5.3. Purification steps
The proteinaceous impurities from the raw-material (starch) and the

different enzyme preparations are removed by heat denaturation fol-
lowed by treatment with activated carbon and filtration10, 14. The 16,
treatment with activated carbon will also remove the main part of
other organic, non-ionic impurities.
The ionic impurities are removed in the two-step demineralization

procedure using cation and anion exchange resins. Any remaining in-
organic salts would be detected by the test for total ash. Lead
would be detected by the specific test for lead.
The glycosidic impurities and extractives of the ion-exchange res-

ins are removed during the crystallization of trehalose. Excessive
amounts of glycosidic impurities would be detected in the HPLC
analysis which forms the basis for the Assay.
13
α-Isomaltosyl-α-D-glucoside is digested by an enzyme preparation from rat small intes-
tine. Non-digested product is fermented by intestinal bacteria (e.g., Bifidobacteria)
(Kurimoto et al., 1997a).
14
Five batches of trehalose were analyzed for residual enzymatic activity of α-amylase,
CGTase, MTSase, isoamylase, and glucoamylase using ELISA assays (Nagaoka et al., 1997).
(The enzymes used for the immunization were the ones used in trehalose production and
had an identical purity). No enzymatic activity was detected (limit of detection 1 mU/g
trehalose for all enzymes except MTSase (10 mU/g trehalose) (Annex 4).
It is unlikely that microorganisms could be carried into the proc-
ess (mainly from the raw material) and survive the production proc-
ess which involves several heat treatment steps. Any surviving mi-
croorganisms would be detected in the tests for total (aerobic)
plate counts, E. coli, Salmonella, and yeast and molds.
4. "History of the organism used as the source of the NF"
In the novel production process which is the subject of this appli-

cation, trehalose is produced enzymatically from food-grade starch,
i.e., it is not obtained from a plant, animal, or microorganism.
Consequently, no information is required under this heading.
5. "Anticipated intake / extent of use of the NF"
5.1. Intended uses in food
Being about 40-45% as sweet as sucrose (Birch et al., 1970; Sugi-

moto, 1995), trehalose may be used to replace some of the sucrose in
those types of food which require a certain amount of sucrose for
technological reasons but would have a more balanced taste profile
if their sweetness was somewhat reduced.
Trehalose can be used to make fruit fillings and toppings, cream

fillings, etc. which are microbiologically and physically as stable
as those produced with sucrose but which have a richer flavor be-
cause trehalose has a lower sweetness. In fruit preparations with a
naturally high acid content there is less browning with trehalose
than with sucrose because trehalose is more resistant to acid-
catalyzed hydrolysis and does not participate in Maillard-type reac-
tions (O'Brien, 1996; Miyake, unpublished report).
Trehalose acts as a stabilizer for proteins during freezing and dry-

ing. It has, for example, been found that enzymes retained a higher
activity if they were dried in the presence of trehalose (Colaço et
al., 1992, 1994). Trehalose also stabilizes phospholipid bilayers
(e.g., liposomes) and more complex biological structures (Crowe et
al., 1988; Leslie et al, 1995; Aguilera & Karel, 1997). In food pro-
cessing, trehalose has, therefore, been used in the preparation of
egg powder and dried fruits (Roser, 1991; Colaço & Roser, 1995; Mi-
yake, unpublished report). As a cryoprotecant it is used in the pro-
duction of surimi, other restructured sea food, frozen gelatin des-
serts, and premium ice creams. Trehalose appears to have a stabiliz-
ing effect not only during dehydration but also during the rehydra-
tion of susceptible molecules. Incorporated in instant noodles or
precooked rice, it accelerates the rehydration of the product.
Sugars may inhibit the retrogradation of starch in bread, cake, or

other starch-based foods. In other words, sugars may stabilize the
quality and prevent staleness of such foods. In a model system with
gelatinized starch as well as in sponge cake and baked rolls, treha-
lose was found to be more efficient in this respect than sucrose
(Miyake, unpublished report). In cookies, a portion of the fat can
be replaced by trehalose without compromising the texture and shelf-
life of the product.
Trehalose possesses a significantly higher glass transition tempera-

ture than maltose, sucrose or glucose (Green & Angell, 1989; Mehl,
1997). It is, therefore, well suited for the production of transpar-
ent, hard-boiled candies and for use in coatings.
Because of its high chemical stability and low hygroscopicity, tre-

halose is well-suited for the production of compressed tablets. In
some cases, the stabilizing effect of trehalose on sensitive ingre-
dients of such tablets offers an additional benefit.
The low-hygroscopicity of trehalose makes it potential substitute

for other sugars which are presently used for icings, dusting, and
coatings.
5.2. Current food applications of trehalose in Japan
In Japan, trehalose is included under No 303 in the "List of Exist-

ing Food Additives". This list comprises naturally occurring addi-
tives which may not be produced, however, by chemical processes. The
listed products may be used in foods generally, unless restrictions
are specified (List of Existing Food Additives, published by the
Ministry of Health and Welfare, April 16, 1996).
Trehalose is used mainly in bakery products (cakes, frozen bread

dough, cream fillings, toppings, etc.), beverages (sports drinks,
fruit drinks), hard and soft confectionery, fruit jam, breakfast ce-
reals, rice, and noodles. The main purposes of use are the reduction
of sweetness (in bakery products and confectionery), the reduction
of moisture absorption (in breakfast cereals and certain types of
confectionery), the reduction of browning reactions (in beverages
and certain types of confectionery) and the prevention of starch
retrogradation (in bakery products and noodles). A list of typical
applications and use levels in Japanese foods is presented in Annex
5.
5.3. Estimated daily intake
The estimated daily intake (EDI) of trehalose from its different

projected uses in food (as specified in Table 1) but excluding
chewing gum, has been calculated for the US population by ENVIRON
(Arlington, VA) using the dietary survey approach (Annex 6). Al-
though dietary habits of US and European consumers differ somewhat,
an EDI calculation on basis of US data was considered adequate be-
cause the consumption of processed food is rather higher in the US
than in the EU and because European food intake data are not avail-
able for conducting EDI calculations of a similar degree of refine-
ment. This calculation model relies on food consumption data from
the 1994-96 Continuing Survey of Food Intakes by Individuals (94-96
CSFII) in which data were collected from a representative sample of
individuals residing in households in the US (approximately 15,000
individuals). Each individual was surveyed for two non-consecutive
days using 24-hour recall interviews. The foods consumed were coded
according to a system which contains about 6,000 different catego-
ries.
For the purpose of the present EDI calculation it was assumed that
each food (or food component) which may contain trehalose, indeed
contained trehalose at the highest, feasible concentration (as
specified in Table 1). Where trehalose was used in a component of
the food (e.g., in fruit-fillings), the intake of that component
was calculated from data on food composition.
The EDI of trehalose was calculated for each food category in which
trehalose may be used, and for all these food categories combined
(except for chewing gum for which such data are not available).
Mean and 90th percentile intakes were calculated on per-user basis
for the following age groups: children 2-12 years of age; teenagers
13-19 years of age; and the adult U.S. population 20+ years of age.
"Users" were defined as individuals who consumed food in the par-
ticular category on at least one occasion. Since food intake was
recorded by time of day and by eating occasion [breakfast, brunch,
lunch, dinner, supper, snack, and other (extended) eating occa-
sion], the intake of trehalose could be calculated per eating occa-
sion.
The main results of these calculations are presented in Table 2. De-

tails are shown in Annex 6. For adults, the estimated exposure to
trehalose from all proposed uses, excluding chewing gum, is 5.6 and
13.0 g/day at the mean and 90th percentile, respectively. Mean intake
by eating occasion (excluding extended eating occasions) ranged from
3.9 to 8.1 g/occasion, while intake at the 90th percentile ranged
from 7.6 to 18.6 g/occasion. The highest estimated exposure to tre-
halose results from the intake of ice cream. In teenagers, this
product results in an average intake of 16.7 g/day (intake may occur
at more than one eating occasion) (see Annex 6).
Trehalose may be used as a non-hygroscopic sweetener in the coating

of chewing gum. In this application, trehalose would account for not
more than 10% of the total weight of the chewing gum. The estimation
of trehalose intake from chewing gum is based on a separate survey
in which approximately 1,500 individuals reported their one-day in-
take of regular and sugarfree gum by mail. The survey, which was
conducted in 1995, distinguished between 2 age groups (children and
adults) but the amount of coated chewing gum consumption did not
differ between these two groups. From the results of this survey it
is concluded that the average user of coated chewing gum would in-
gest 0.4 g/d trehalose from this product (90th percentile consumer:
0.8 g/d) (Annex 6).
In assessing the total daily intake of trehalose from all dietary

sources it is important to note that, with regard to gastrointesti-
nal tolerance, the intake per eating occasion is a more important
parameter than the total daily intake from all dietary sources com-
bined. Considering the different anticipated uses of trehalose (as
described in chapter 5.1), it becomes apparent that the intake of
trehalose is not concentrated to certain eating occasions (such as
main meals) but is spread evenly over the day by the mere nature of
the products in which it is used. This is also reflected by the data
on estimated daily intakes. The figures shown in Table 2 demonstrate
that the mean and 90th percentile intake per eating occasion do not
exceed 8.1 and 18.6 g, respectively, in any age group. A comparison
between the intakes from the various food categories and the total
intake from all sources demonstrates that many uses are mutually ex-
clusive. In other words, a consumer of a sponge cake is unlikely to
eat a fruit pie at the same eating occasion. Consequently, trehalose
intakes per eating occasion are similar to the intake from one spe-
cific food category in a given age group (Annex 6).
Estimated intakes per eating occasion are far below the 50-g intake
which is typically used in trehalose loading tests and which is gen-
erally well tolerated (Bolte et al., 1973). The intakes per eating
occasion are also below the threshold dose for abdominal effects in
particularly sensitive subjects [>30 g per eating occasion (Oku &
Okazaki, 1998)]. Adverse gastrointestinal side-effects from the in-
tended uses of trehalose are, therefore, not to be expected. Since
in some applications trehalose may substitute for polyols, the total
intake of low-digestible carbohydrates could even slightly decrease.
6. "Information from previous human exposure to the NF or its
source"
6.1. Occurrence in nature
Trehalose occurs widespread in nature. Except for a few rare cases,

only the α,α-isomer has been found (Elbein, 1974). Trehalose occurs
as such in the cells of a variety of plants and in the blood (hemo-
lymph) of arthropods. In Mycobacteriae (e.g., M. tuberculosis), it
occurs in the form of a glycolipid, namely the 6,6'-di-ester of my-
colic acid (for references see Birch, 1963). Esters of other fatty
acids with trehalose have been identified in other bacteria (for
references see Birch, 1963).
6.1.1. Occurrence in plants
Trehalose occurs in bacteria15, yeast (e.g., S. cerevisiae)16, a

wide variety of fungi17, algae18, and a few higher plants19 (reviewed
by Elbein, 1974).
Intracellular trehalose appears to play an important role in the

protection of the cells from dehydration and freezing, as well as
15
Nicolaus et al., 1988; Stjernholm, 1958 cited in Birch, 1963; Streeter & Bhagwat, 1999.
16
Koch & Koch, 1925; Oku et al., 1998.
17
Elbein, 1974; Thevelein, 1984; Oku et al., 1998.
18
Myrbäck, 1949; Elander & Myrbäck, 1949 cited in Birch, 1963; Lindberg, 1955 cited in
Birch, 1963.
19
Oesch & Meier, 1967 cited in Elbein, 1974; Goddijn & Smeekens, 1998; Oku et al., 1998;
Ghasempour et al., 1998; Müller et al., 1999.
from other adverse environmental conditions (e.g., heat shock,
toxic levels of ethanol, osmostress) (Singer & Lindquist, 1998a,
b). In addition, trehalose may serve as reserve carbohydrate during
periods of carbon starvation (Silljé et al., 1999).
Because of its presence in baker's and brewer's yeast, in which it

reaches concentrations of up to 23% on a dry weight basis20, small
amounts of trehalose have been found in bread (1.2-1.5 g/kg dry
substance)21, beer (45-240 mg/l)21, wine (44-129 mg/l)21, and honey
(0.1-2.3 g/100g)22.
Mushrooms, including many edible species, contain trehalose at lev-

els of about 2-12 g/100g dry weight, but contents of up to 22% have
also been reported (Oku et al., 1998; for additional references see
Elbein, 1974).
6.1.2. Occurrence in animals
Trehalose has been found in many species of insects (Elbein, 1974;

Becker et al., 1996). It has also been detected in a variety of
other invertebrates, including nematodes23, crustaceans, and anne-
lids (reviewed by Elbein, 1974). The blood of lobsters (Homarus
20
Myrbäck & Örtenblad, 1936 cited in Birch, 1963; Steiner & Cori, 1935; Stewart et al.,
1950.
21
Oku et al., 1998.
22
Mendes et al., 1998.
23
Behm, 1997.
22 Mendes et al., 1998.
americanus) and crabs (Carcinus macenus) contains trehalose at con-
centrations of 2.5 and 1.5 mg/100ml, respectively.
In many insects, trehalose is the major sugar found in the blood

(hemolymph) at concentrations of between 1 to 2% (Wyatt & Kalf,
1957; Thompson, 1999). It occurs at all stages of the development
(pupal, larval, and adult stage) but concentrations may differ. In
flying insects, trehalose functions as the energy source to sustain
flight (Becker et al., 1996; Candy et al., 1997). Trehalose may
also play a role in the frost resistance of some insects (overwin-
tering prepupal larvae of the sawfly contain 5-9% trehalose) and
nematodes [juveniles of an entomopathogenic nematode increase their
trehalose levels with decreasing temperature [6% trehalose (% dry
weight) at 5-8°C] (Asahino & Tanno, 1964 cited in Elbein, 1974; Qiu
& Bedding, 1999).
High concentrations of trehalose (7%) were found in a type of manna

which was used by Bedouins as a sweetener for coffee in the North
Iraqian desert (Leibowitz, 1943, 1944). The manna is probably the
excretion product of some scale insects.
6.2. Consumption of trehalose added to food in Japan
Trehalose produced by the enzymatic process described in this dos-

sier became available in Japan for food use in 1995. By mid 1997
monthly sales reached about 500 tons.
6.3. Intake of high single doses in trehalose tolerance tests
The absorptive capacity of the small intestine may be examined in

patients suffering from malabsorption syndrom by measuring the ab-
sorption of ingested xylose or trehalose (Dominick & Anspach,
1975). Typically, rather high doses are administered (e.g., 50
g/person or 1 g/kg bw). The results that were obtained from such
tests with trehalose are described in chapters 9.2.1 and 9.4.1.
7. "Nutritional information on the NF"
Ingested trehalose is digested by trehalase in the small intestine

to glucose which is readily absorbed (see chapter 9.2.1). Trehalose
has, therefore, the same physiological energy value as glucose or
maltose. The metabolic pathways of trehalose and maltose (or starch
after digestion by amylase) merge after digestion to glucose at the
brush border of the intestinal mucosa.
It follows that trehalose is nutritionally equivalent to glucose

and maltose, as well as maltodextrin and starch. (Maltose is the
main product of the digestion of the latter two products by pancre-
atic amylase).
Trehalose may substitute for some of the sucrose in certain foods

(see chapter 5.1 for examples). In comparison to sucrose, trehalose
provides on digestion only glucose but no fructose. In terms of
physiological energy (caloric value), this does not make a differ-
ence. In terms of nutritional and metabolic properties, no disad-
vantage is associated with a reduced intake of fructose.
8. "Microbiological information on the NF"
The enzymes which are used in the novel production process of tre-
halose, are obtained from non-genetically modified strains of Ar-
throbacter ramosus, Pseudomonas amyloderamosa, Bacillus stearother-
mophilus, Aspergillus niger, and Bacillus subtilis.
The enzyme-containing fermentation broths are separated from the

source microorganisms by filtration. Moreover, the trehalose pro-
duction process comprises several steps with heat-treatment. There-
fore, it is unlikely that any of the source microorganisms would be
present in the final product. An inadvertent presence of micro-
organisms would be detected by the applied quality control proce-
dures. The trehalose solutions obtained from the de-ionization step
and the subsequent concentration step are controlled regularly in
process for microbiological contamination. The specifications of
the end-product also include microbiological parameters.
The safety of the applied enzymes and their source microorganisms

has been discussed in chapter 3.4.
9. "Toxicological information on the NF"
9.1. General remarks
Ingested trehalose enters the animal or human body after digestion

in the form of glucose. From a metabolic and toxicological perspec-
tive, glucose, maltose, maltodextrin and starch represent, there-
fore, counterparts which may be used as baseline to facilitate the
safety assessment.
The novel process by which trehalose is produced and which is the

subject of the present application, does not introduce by-products
that are potentially toxic as explained in chapters 3.4 and 3.5.
Allergic reactions to the intake of trehalose produced by the enzy-

matic process are not to be expected. Sugars generally have a very
low allergenic potential. The by-products which are present in tre-
halose in very small concentrations, are glucose, maltose and
maltooligosaccharides which are not known to be allergenic. Resid-
ual activity of the enzymes used in the production process was not
detected (see chapters 3.5.2 and 3.5.3). Proteinaceous impurities
which are present in trehalose at very low concentrations (70 – 80
ppm, see chapter 3.4) stem most likely from the applied raw mate-
rial, i.e. food grade starch, which has a protein content of < 3500
ppm).
9.2. Metabolic studies on trehalose
In humans and many animals, the fate of ingested or parenterally ad-

ministered trehalose corresponds to that of glucose because treha-
lose is rapidly hydrolyzed to glucose by trehalase (EC 3.2.1.28).
Trehalase occurs in humans and most animals at the brush-border of
the intestinal mucosa and, depending upon the species, also in the
proximal tubular cells of the kidney, as well as in the liver and
the blood plasma (Courtois & Demelier, 1966; Hore & Messer, 1968;
Berg et al., 1973; Demelier et al., 1975; Vázquez et al., 1975; La-
bat-Robert, 1982; van Handel, 1970; Rodeck & Dominick, 1983;
Niederst & Dauça, 1985; Jönsson et al., 1986; Eze, 1989; Riby et
al., 1990; Yoshida, 1993; Isichei & Gorecki, 1993).
9.2.1. Digestion and absorption of ingested trehalose
Intestinal trehalase is a membrane-bound enzyme. It is localized in

the brush border of the mucosal cells of the duodenum, jejunum, and
– at lower levels – the ileum (Hietanen, 1973). Trehalase accounts
for all the trehalase activity of the small intestinal mucosa (i.e.,
there are no other glycosidases of mammalian origin which could hy-
drolyze trehalose). Trehalase activity was found in the small intes-
tine of many animals (e.g., mouse, rat, guinea pig, rabbit, pig, ba-
boon) and humans (Galand, 1989; Maestracci, 1976; Ruppin et al.,
1974; Hietanen, 1973; Cerda et al., 1972). Only a few species appear
to lack intestinal trehalase (e.g., cat) (Hore & Messer, 1968; Ga-
land, 1989; Hietanen, 1973).
A very small fraction of the healthy human population exhibits a
primary (hereditary) or secondary (acquired) trehalase deficiency
(see chapter 9.4.2). People with this condition may experience in-
testinal side-effects (flatulence, laxation) after the ingestion of
excessive amounts of trehalose, like lactase-deficient people after
the intake of excessive amounts of lactose. In this regard it is
noteworthy, however, that the incidence of lactase deficiency is
higher than that of trehalase deficiency (Gudmand-Høyer & Skovbjerg,
1996).
If, under conditions of a relative or absolute deficiency of intes-

tinal trehalase, ingested trehalose is incompletely digested to glu-
cose at the small intestinal mucosa, a small fraction (about 0.5%)
may be absorbed by passive diffusion as shown for other disaccha-
rides (e.g., lactulose) (van Elburg et al., 1995)24. Any trehalose
not digested but absorbed would then be split to glucose by tre-
halase in the liver, plasma and/or kidney, or would be excreted un-
changed with the urine (Demelier et al., 1975). The not digested and
not absorbed portion of ingested trehalose will likely be fermented
by the intestinal microflora to short-chain fatty acids (mainly ace-
tate, propionate, butyrate) as has been shown for many other non-
digestible carbohydrates (e.g., resistant starch, inulin, etc.).
The following experimental data demonstrate that ingested trehalose

is efficiently digested to glucose which is then absorbed and me-
tabolized in animals and humans.
24
Since trehalose appears to have a higher hydrated volume than other sugars, the frac-
tional absorption by passive diffusion may actually be lower for trehalose than for
other disaccharides (Sola-Penna & Meyer-Fernandes, 1998).
In an early study on the metabolism of different rare sugars, groups
of fasted rats received about 0.5 g glucose, trehalose, and other
sugars by gavage. An additional group served as control. The animals
were killed 8-9 hours after dosing and liver glycogen was deter-
mined. In comparison to the untreated controls, the average liver
glycogen content was increased after trehalose ingestion suggesting
that trehalose was efficiently converted to glucose (Clarke et al.,
1939).
In the context of an acute toxicity test, four male Beagle dogs re-
ceived trehalose per os (5 g/kg bw). Blood and urine were collected
before and in regular intervals after dosing for analysis of serum
glucose and trehalose. Serum glucose increased markedly reaching
maximum levels after 1 hour. Urinary glucose levels increased tran-
siently above baseline levels (presumably due to the hydrolysis of
some absorbed trehalose in the kidney). Urinary trehalose was not
detected (Atkinson & Thomas, 1994c)25.
Fifty healthy subjects ingested solutions with 50 g trehalose and 50

g glucose on different days after an overnight fast. Blood was col-
lected before and 15, 30, 60, 90, and 120 min after dosing. The
fractional intestinal digestion of trehalose to absorbed glucose was
calculated from the ratio of the area under the blood glucose curve
(0-60 min) of trehalose relative to that of glucose. The observed
values ranged from 0.3 – 1.5 with a median of 0.7 (Bergoz, 1971,
Bergoz et al., 1973).
25
Similar measurements were made in the context of acute toxicity studies in mice and
rats. However, the number of urine samples available for analysis was too small to ob-
tain meaningful results (Atkinson & Thomas, 1994a, b).
Using an identical experimental protocol, the trehalose absorption
was studied in 9 uremic patients suffering from chronic renal insuf-
ficiency. The absorption of trehalose relative to that of glucose
was 0.83 + 0.5 (Pointner et al., 1974).
The digestion of trehalose in children is similarly high. Sixteen

healthy children (age 1 – 12 years) ingested trehalose and glucose
at a dose of 1 g/kg bw. The ratio of the area under the curve of
blood glucose (0 – 60 min) was 0.91 (mean). Values below 0.6 were
found only in children suffering from different forms of intestinal
disease (Dominick & Anspach, 1975).
Sixty healthy fasted volunteers received on different days 50 g tre-

halose, glucose, lactose and sucrose (dissolved in 400 ml water).
Blood was collected for determination of glucose before and 30, 60,
90 and 120 min after dosing. Symptoms of gastrointestinal intoler-
ance were not reported after intake of trehalose. An evaluation of
the blood glucose profiles indicated that none of the volunteers
suffered from trehalose malabsorption [ratio of AUC (trehalose)/AUC
(glucose) > 0.3] (Bolte et al., 1973). The reported digestion and
absorption rate (median 0.7) is similar to that reported by Bergoz
(1971, 1973) but is slightly lower than that observed. In this re-
gard, it must be noted, however, that the ingestion of 400 ml water
with 50 g trehalose on an empty stomach represents an extreme situa-
tion which is highly unlikely to occur under ordinary conditions of
use of trehalose as a food ingredient (see chapter 5.3). Typically,
trehalose will be ingested at lower doses and in most instances to-
gether with other foods. Under these conditions, the small intesti-
nal transit-time will be longer and hence digestion of trehalose to
absorbable glucose will be more complete (Heine et al., 1996).
9.2.2. Metabolism of parenterally administered trehalose
Trehalose which enters the circulating blood either through passive

absorption from the gut or by systemic administration, will be con-
verted to glucose by trehalase which, depending upon species, occurs
in the serum, kidney liver, and bile (Van Handel, 1969; Labat-
Robert, 1982; Arola et al., 1999). Any trehalose that would escape
hydrolysis by plasma and liver trehalase would be excreted in the
primary urine. In mice, rabbits, dogs, pigs and humans, excreted
trehalose would be cleaved by renal trehalase which is located in
the brush border of the proximal tubular cells (Demelier et al.,
1975; Niederst & Dauça, 1985; Riby et al., 1990)26. In rats, guinea
pigs, birds, and a few other species which appear to lack renal tre-
halase (van Handel, 1969; Demelier et al., 1975; Riby et al., 1990),
ultrafiltered trehalose would be excreted with the urine.
The following experimental data illustrate the efficacy of trehalose

elimination from the circulating blood.
Rats, guinea pigs and rabbits were given trehalose at a dose of 0.5
or 1 g/kg bw by intravenous injection. In rats, about 87% (range 75-
98%) of the administered dose was recovered from the urine indicat-
ing that trehalase activity in the liver and kidneys was low or ab-
sent. In the guinea pigs and rabbits, only 7-9% of the dose were ex-
26
Under physiological conditions, small amounts of trehalase are excreted with the human
urine (Maruhn et al., 1976). Increased levels of urinary trehalase are indicative of
proximal tubular damage (Sasai-Takedatsu et al., 1996).
creted with the urine. Examination of the trehalose levels drawn
from the renal artery and vein of rabbits indicated that trehalose
was hydrolyzed efficiently in the kidneys (Demelier et al., 1975).
Rabbits given intravenous injections of 500 mg trehalose (equivalent

to 200-300 mg/kg) were able to clear the sugar from their plasma
within 60 minutes. No trehalose was found in the urine. Only a small
amount of trehalose was detected in the urine of rabbits receiving
an intravenous dose of 1,000 mg. By contrast, rats given 100 mg (300
mg/kg) removed the sugar from the plasma at approximately the same
rate as the rabbits, but trehalose was recovered from the urine in
proportion to the plasma concentrations. These results indicated
that there was no renal metabolism of trehalose in the rat (Riby et
al., 1990).
Male rabbits received infusions of trehalose solution (10%), maltose

solution (10%), glucose solution (5%), or saline for 90 min at a
rate of 6.7 ml/kg bw/h (4 rabbits per treatment). Blood was col-
lected before and 30, 60, 90, 120 and 180 min after start of the in-
fusion. Urine was collected over 24 hours from start of the infu-
sion. Serum glucose, trehalose, maltose, and insulin were deter-
mined. The urine was analyzed for glucose, trehalose, and maltose.
The infusion of trehalose led to a rapid increase of serum glucose
levels which peaked at 90 min (end of infusion) and returned back to
baseline values within 90 min after end of the infusion. Serum insu-
lin levels exhibited a corresponding time course. The urinary glu-
cose excretion was slightly higher in trehalose infused rabbits than
in glucose or saline infused rabbits (difference not significant
with the low number of animals used). About 1% of the infused treha-
lose was recovered in the urine. The data on serum and urinary sug-
ars, and on serum insulin demonstrated that infused trehalose is me-
tabolized more rapidly and more completely than infused maltose at
the applied infusion rate. The better utilization of trehalose was
confirmed in an experiment with infusion of trehalose and maltose to
alloxan-diabetic rabbits. In a further experiment, the continuous
infusion of trehalose or glucose (supplemented with amino acids) for
5 days at a rate of 8.23 g/kg bw/d was not associated with any signs
of toxicity or adverse effects on serum biochemical parameters of
male rabbits (Sato et al., 1999).
In the context of an acute toxicity test, four male Beagle dogs re-
ceived trehalose by intravenous injection (1 g/kg bw). Blood and
urine were collected before and in regular intervals after dosing
for analysis of serum glucose and trehalose. Serum glucose raised
slightly reaching highest values after about 30-60 minutes of dos-
ing. Urinary glucose was increased markedly probably reflecting hy-
drolysis of trehalose by renal trehalose in the proximal tubuli. The
trehalose concentration in the urine 5 min after dosing was markedly
higher than that found in the plasma which suggests that trehalose
is rapidly cleared by the kidneys (Atkinson & Thomas, 1994c).
9.3. Toxicological studies
9.3.1. Acute toxicity
The acute toxicity of trehalose was examined in mice, rats, and

dogs. The results are summarized in Table 3.
In the rodent studies conducted by the Frederick Research Center,
groups of 5 animals of each sex were tested at one dose level (i.v.,
1 g/kg bw; p.o., 5 g/kg bw) or served as controls. In the dog study
conducted by the same contract institute, 4 male Beagles received a
single i.v. dose of 1 g/kg bw, followed 6 days later by an oral dose
of 5 g/kg bw. There were no mortalities due to the trehalose treat-
ments in any species. Signs of systemic toxicity were not observed.
Diarrhoea did not occur in response to the oral administration of
trehalose. Body weight gains were not impaired by the treatment dur-
ing the 7-day (dogs) or 14-day (rodents) post-treatment period (At-
kinson & Thomas, 1994 a, b, c).
In a subsequent acute toxicity test in rats, trehalose was adminis-

tered by gavage at a dose of 16 g/kg bw to five fasted Sprague-
Dawley rats of each sex. The dose was given in two equal portions
with a 1-hour period between administrations. No deaths occurred in
response to the treatment. Soft to liquid feces occurred in one male
on day 1 of the study. Piloerection was observed in all rats during
day 1. Body weight gains were within normal limits during the 14-day
post-treatment period. Gross necropsy on day 15 revealed no abnor-
malities (Mc Rae, 1995).
9.3.2. Short-term studies of toxicity
Mice
Groups of ten CD-1 mice of each sex were given trehalose at doses of
1 g/kg bw/d by intravenous injection, 5 g/kg bw/d by gavage, or 2.5
g/kg bw/d by subcutaneous injection for 14 consecutive days. An ad-
ditional group (control group) received sterile saline (same dose
volume as the oral treatment group). Body weights, weight gains, and
food consumption did not differ between the treatment groups over
the 2-week treatment period. One female of the i.v. group died after
dosing on day 6 (necropsy findings were unremarkable). No clinical
signs of toxicity were seen throughout the study. Mild necrosis
and/or local irritation of the tail in some animals of the i.v.
group was ascribed to the repeated administration of a hypertonic
solution. Blood samples were collected at termination for analysis
of standard hematological and biochemical parameters (5 animals/sex
for each set of parameters). The white blood cell counts were below
control levels in males but not females of the p.o. and s.c. groups.
Differences between treated groups and controls of some biochemical
parameters were quantitatively small and also limited to one sex
(males). Macroscopic and histopathological examination of all tis-
sues of p.o. treated mice did not reveal any changes that could be
attributed to the exposure to trehalose (Atkinson & Thomas, 1994d).
Four groups of HanIbm:NMRI mice (20 mice/sex/group) received diets

with admixture of 0 (control), 0.5, 1.5 and 5% trehalose for 13
weeks. Animals were examined for clinical signs of toxicity; body
weights and food consumption were monitored throughout the study.
Ophthalmoscopic examinations were performed on 10 mice/sex of the
control and high-dose group before start and in the final week of
the study. Blood samples were collected for analysis of standard he-
matological and biochemical parameters from fasted animals (10
mice/sex/group) in weeks 5, 9, and 13 (termination). Urine was col-
lected during the 18-hour fasting period for measurement of pH,
standard semi-quantitative urinary parameters, and microscopic ex-
amination of the sediment. After at least 90 days of treatment, the
animals were killed and subjected to gross necropsy. The absolute
and relative weights of main organs were determined. Tissues and or-
gans of the control and high dose group (and those exhibiting macro-
scopic abnormalities of the low and mid-dose group) were examined
microscopically for pathological changes.
There were four deaths during the study (1 spontaneous death in the
control group, 1 male of the high-dose group killed in extremis in
week 12, and two deaths in connection with blood sampling). The ani-
mal of the high-dose group exhibited severe pyelonephritis on histo-
pathologic examination which was considered not to be related to the
treatment. Food consumption and body weight gains were slightly be-
low controls in males of the mid- and high-dose groups. Treatment
related clinical signs were not observed. Ophthalmoscopic examina-
tion revealed no treatment-related effects. Hematological parameters
remained unaffected by the treatment. Plasma glucose levels tended
to be slightly increased in males and females of the high dose group
(significant in females in week 5 and 9). In the females, plasma
phosphate was increased at the same time. In week 13, plasma potas-
sium was decreased in a dose-related manner in males and females,
reaching statistical significance in the high-dose group. However,
the observed values were still well within the limits of historical
controls. Significant changes of a few other parameters were consid-
ered coincidental since they occurred only at one occasion and/or in
one sex. Urinary parameters did not reveal changes that could be at-
tributed to the treatment. Organ weights were unaffected by the
treatment. There were no macroscopic or microscopic findings in or-
gans and tissues which were considered to be related to the treat-
ment. The high-dose level [equivalent to 7.3 (males) and 9.3 (fe-
males) g/kg bw/d] was considered to be tolerated without toxicologi-
cal effects (Schmid et al., 1998).
Dogs
Groups of three male and three female Beagle dogs received trehalose
at doses of 1 g/kg bw/d by intravenous injection in the cephalic
vein, 5 g/kg/d by oral capsule, or 0.25 g/kg/d by subcutaneous in-
jection for 14 consecutive days. A control group was given empty
capsules at the same number of capsules/kg body weight as those
treated with the test article. All animals survived till the end of
the study. Body weights and food consumption did not differ between
controls and the different treatment groups. All dogs of the p.o.
group experienced diarrhoea on some days of the study (range: 4-8
days out of 14 days). Vomiting was observed in a few animals of the
control, p.o. and s.c. groups. Measurement of serum glucose after
dosing at the last treatment day revealed significant transient in-
creases in i.v. and p.o. dosed dogs. Analysis of blood samples col-
lected at termination did not reveal toxicologically relevant
changes of hematological or biochemical parameters. Macroscopic ex-
amination at necropsy and microscopic examination of standard organs
and tissue did not reveal changes which could be attributed to the
treatment. It was concluded that trehalose was not toxic at the ap-
plied doses (Atkinson & Thomas, 1994e).
9.3.3. Genotoxicity
The results of assays for genotoxicity are shown in Table 4.
9.3.4. Developmental toxicity
Rats
In a study of embryotoxicity and teratogenicity, groups of 28 pre-

sumed pregnant Wistar Crl:(WI)WU BR rats were fed diets containing
trehalose (purity: 99%) at concentrations of 0, 2.5, 5 and 10% on
days 0-21 of gestation. Trehalose was added to the test diets in
lieu of pre-gelatinized starch. The animals were examined throughout
the study, and body weight and food consumption were recorded. The
rats were killed on day 21 and examined for parameters of reproduc-
tive performance. Fetuses were examined for signs of toxicity, ex-
ternal malformations, and soft-tissue defects and were stained for
detection of skeletal anomalies.
No deaths occurred during the study. Food consumption and maternal

body-weight gain were similar in all groups. The trehalose intake
during the gestation period ranged from 1.4-2.0, 2.8-3.9 and 5.5-7.8
g trehalose/kg bw/d for the low, mid and high-dose group, respec-
tively. Early delivery was observed in 1 female of the control group
and 2 females of the low-dose group. Necropsy of the dams showed no
adverse effects that could be related to the treatment. The number
of viable litters, the number of corpora lutea, and the mean number
of implantation sites were similar in all groups. Fetal length and
body weight were also similar in all groups. Examination of the fe-
tuses of the control and high-dose group revealed no treatment-
related increase in gross, skeletal, or visceral abnormalities. Un-
der the conditions of this assay, trehalose was not teratogenic and
did not induce any maternal or developmental toxicity (Waalkens-
Berendsen, 1998).
In a 2-geneneration reproductive toxicity study, groups of Wistar
rats (Crl:(WI)WU BR) (28 rats/sex/group) were administered a diet
containing trehalose (purity 99%) at concentrations of 0, 2.5, 5 or
10%. After a period of 10 weeks (pre-mating period), each female was
caged with one male of the same treatment group until pregnancy oc-
curred. Upon evidence of copulation, the females were housed indi-
vidually. Trehalose treatment was continued during pregnancy and
lactation. On day 4 after delivery, the litters were culled to 8
pups. After weaning, 28 males and 28 females were selected at random
from as many litters as possible in each group and raised to matur-
ity while continuing the trehalose treatment. These animals (F1 gen-
eration) were then used to rear the next generation (F2 generation)
avoiding the mating of siblings and following an identical procedure
as for the F0 generation. The F0 and F1 males and females were killed
and necropsied after weaning of their litters. There were no clini-
cal signs or changes in behavior due to treatment. Some differences
between controls and treatment groups with regard to body weights
and food intake showed no consistent association with the dose or
duration of the treatment. In the high-dose group, males of the F0
and F1 generation consumed between 4.9-12.4 g trehalose per kg body
weight per day during the pre-mating period. F0 and F1 females of the
high-dose group consumed 5.9-11.7 g/kg bw/d during the same period.
During the gestation and lactation period, their trehalose intake
ranged from 4.4-7.2 and 10.1-19.9 g/kg bw/d, respectively. No ef-
fects of the trehalose treatment were observed on standard parame-
ters of reproductive performance. The number of pups delivered, pup
mortality, body weights and growth did not differ between treated
groups and controls. No treatment-related macroscopic changes were
observed at autopsy. Microscopic examination of reproductive organs
and accessory glands, as well as the pituitary and spleen did not
reveal any changes that could be attributed to the treatment. It was
concluded that ingestion of trehalose at dietary concentrations of
up to 10% had no adverse effects on reproduction and development of
the pups (Wolterbeek & Waalkens-Berendsen, 1999a).
Rabbits
In a study of embryotoxicity and teratogenicity, groups of 16 pre-

sumed pregnant New Zealand white rabbits were fed diets containing
trehalose (purity: 99%) at concentrations of 0, 2.5, 5, or 10% on
days 0-29 of gestation. The animals were examined throughout the
study, and body weights and food consumption were recorded regu-
larly. The animals were killed on day 29 of pregnancy, and the fe-
tuses were examined for signs of toxicity, external malformations,
and soft-tissue defects and were stained for detection of skeletal
anomalies.
Body weights, weight gains and food consumption did not differ be-
tween controls and treated groups. However, there were a few animals
in each of the three dose groups, which consumed only small amounts
of food during the third and/or fourth week of pregnancy. One of
these animals was killed in extremis, another died spontaneously
(both belonging to the high-dose group). At necropsy, a hairball was
found in the stomach of both animals. The trehalose intake ranged
from 0.21-0.77, 0.48-1.34, and 1.04-2.82 g/kg bw/d for the low-,
mid- and high-dose groups, respectively. Necropsy on the does showed
no adverse effects that could be related to the treatment. The num-
ber of viable litters, the number of corpora lutea, and the mean
number of implantation sites were similar in all groups. One early
delivery occurred in the high-dose. Fetal length and body weight
were similar in all groups. Examination of the fetuses revealed no
treatment-related increase in gross, skeletal, or visceral abnor-
malities. Under the conditions of this assay, trehalose was not
teratogenic and did not induce and maternal or developmental toxic-
ity (Wolterbeek & Waalkens-Berendsen, 1999b).
9.3.5. Special studies
9.3.5.1. Ocular irritation
A single ocular application of a 10% aqueous solution of trehalose

was not irritating or corrosive to the eyes of albino rabbits (At-
kinson & Thomas, 1994f).
9.4. Tolerance of trehalose in humans
9.4.1. Intestinal tolerance in healthy volunteers
Being rapidly digested to glucose, trehalose is expected to be tol-

erated at even high oral doses. Indeed, abdominal symptoms or diar-
rhea were not reported in a study in which 60 healthy subjects in-
gested a solution of 50 g trehalose in 400 ml water after an over-
night fast (Bolte et al., 1973). The ingestion of a single dose of
25 g trehalose dissolved in 200 ml water 1 hour after breakfast also
did not provoke diarrhoea or other noteworthy abdominal symptoms in
any of the ten test subjects (Heine et al., 1996).
Clinical signs of intolerance were not observed in any of 27 infants
and children (10 healthy and 17 with malabsorption) after ingestion
of a bolus trehalose dose of 2 g/kg bw (Fiehring et al., 1976). This
observation is consistent with fact that the intestinal trehalase
activity is fully developed already during the first year of life
(Niessen et al., 1975).
In order to determine the laxative threshold of trehalose, 20 Japa-

nese female students received trehalose solutions (200 ml) about 2-3
hours after a meal. The trehalose content of the solution was in-
creased gradually from 10 to 20, 30, 40, 50 and 60 g. Lactulose was
administered according to the same procedure as a positive control.
A negative control (e.g., glucose) was not included in the study. A
trehalose intake of 0.65 g/kg bw represented the maximum dose at
which there was no occurrence of very soft ("muddy") or watery
stool. Other signs of malabsorption such as flatulence, borborygmi
and distention were reported to occur at doses of > 30 g. Surpris-
ingly, the incidence of these symptoms was only slightly smaller af-
ter trehalose than after lactulose ingestion. In the absence of a
negative control treatment, it cannot be decided to what extent an
over-reporting of intestinal symptoms may have contributed to this
result (Oku & Okazaki, 1998).
An intake of 10 g trehalose did not increase breath hydrogen expira-

tion and did not elicit any gastrointestinal symptoms in 30 healthy
volunteers (10 Japanese, 10 Caucasian, 8 Black, and 2 others). At
doses of 20 g or more, mild gastrointestinal symptoms (type not
specified) were noted and hydrogen expiration was increased. When
the subjects received a single trehalose dose of 0.6 g/kg bw, 90% of
the Japanese volunteers reported gastrointestinal symptoms while 11%
of the Caucasians and none of the Blacks appeared to be affected.
Blood glucose levels rose to a significantly smaller degree in Japa-
nese subjects than in Caucasians or Black. This result suggests that
the rate of trehalose digestion is lower in Japanese people than in
Caucasians or Black (Ushijima et al., 1995).
9.4.2. Impaired intestinal tolerance due to intestinal trehalase

deficiency
A relative or absolute deficiency of intestinal trehalase results in

an intolerance to trehalose ingested in high amounts. Only a few
cases of primary (i.e., hereditary) or secondary (i.e., acquired)27
trehalase deficiency have been reported in the scientific litera-
ture.
A first case of trehalose deficiency was observed in Switzerland by

Bergoz in 1971. The patient, a 71-year old women, had noticed that
the intake of mushrooms ("Champignons de Paris") provoked regularly
diarrhoea before the end of the meal. When the patient was subjected
to a trehalose tolerance test (ingestion of 50 g trehalose), the
typical abdominal symptoms of carbohydrate malabsorption, including
diarrhoea, developed promptly. Blood glucose levels did not rise af-
ter trehalose ingestion. It was concluded that a relative or abso-
lute deficiency of intestinal trehalase was the cause of the ob-
served symptoms.
27
A generalized deficiency of disaccharidases (including trehalase) occurs not infrequently
in patients who suffer from various disorders of the digestive tract including Morbus
Crohn, Colitis ulcerosa, Sprue, etc. (Mališ et al., 1972; Bolte et al., 1973; Berg et al.,
1973; Lutkic et al., 1985; Rodeck & Dominick, 1983; Gupta et al., 1999). The reason is
that in these subjects the integrity of the intestinal mucosa is impaired generally.
A second, similar case was reported two years later from a clinic in
Czechoslovakia (Madzarovová-Nohejlova, 1973). In this case, the ab-
sence of intestinal trehalase could be confirmed in a biopsy taken
from the upper jejunum. In addition, the patient's father was also
found to be trehalase-deficient.
Two more subjects with a relative trehalase deficiency were identi-

fied in a survey of the disaccharidase activities of 100 consecu-
tive, morphologically normal intestinal biopsies (Bergoz et al.,
1982).
Since trehalase intolerance may be experienced as intestinal intol-

erance to mushrooms, the trehalose tolerance was examined in 30 Fin-
nish subjects who were selected for self-proclaimed intestinal mush-
room intolerance. For comparison, 34 mushroom-tolerant subjects were
included in the study as well. A trehalose tolerance test was con-
ducted with ingestion of a 25-g dose of trehalose (consumed with 400
ml water after an overnight fast) followed by measurement of the
rise in blood glucose levels and determination of breath hydrogen
and methane. In addition, trehalase activity was measured in duode-
nal biopsies. A placebo treatment (e.g., with glucose) which would
have allowed for a double-blind study design, was not included. In
the mushroom intolerant group, 13 subjects reported flatulence and
abdominal distention, and 6 experienced diarrhoea. However, a rela-
tive trehalase deficiency (by biopsy measurement) was detected in 2
subjects only. In the mushroom-tolerant group, no abdominal symptoms
were reported after trehalose intake. The duodenal trehalase-to-
sucrase ratio was significantly (p = 0.03) lower in the 19 subjects
who reported intestinal symptoms than in those who did not experi-
ence side-effects (43 subjects). However, the duodenal trehalase ac-
tivity (expressed in IU/g protein) and the rise in blood glucose af-
ter trehalose intake did not differ between the subjects with and
without intestinal symptoms (Arola et al., 1999). In the absence of
a double-blind, placebo-controlled design it cannot be excluded that
an over-reporting of intestinal side effects in the mushroom intol-
erant group has contributed to the seemingly more frequent occur-
rence of trehalose maldigestion.
A relative high incidence of trehalase deficiency has been observed

in Greenland Eskimos (2 out of 19 subjects) (Dahlqvist, 1974). In
this group of people, deficiencies of lactase (17/19), sucrase
(3/19), and isomaltase (3/19) occurred with a high incidence as
well. A similarly high prevalence of relative trehalase deficiency
(in association with deficiencies of all the other intestinal disac-
charidases) was found in an examination of 97 surgical Greenlandic
patients. Trehalase activities below the normal range were detected
in the small intestinal biopsies of 14 out of 97 subjects. In three
of these subjects, the trehalase deficiency was confirmed by a tre-
halose tolerance test (no increase of blood glucose levels following
ingestion of 50 g trehalose) (Gudmand-Høyer et al., 1988, 1996).
It is difficult to derive an estimate on the incidence of clinically

relevant trehalase deficiency in the human population from these
studies. Disregarding the possible higher incidence of trehalase de-
ficiency in specific populations (Eskimos), the incidence of 2% re-
ported by Bergoz (1982) appears high in view of the data by Welsh
(1978) and Gudmand-Høyer (1988) who did not observe a single case of
intolerance in 123 American and 248 Danish subjects, respectively.
In a more recent investigation in the U.K. of 369 patients with sus-
pected malabsorption, one person had a intestinal trehalase activity
below the normal range which was suggestive of primary trehalase de-
ficiency (Murray et al., 2000). In the tolerance studies summarized
in chapter 9.4.1, more than 100 healthy subjects, including chil-
dren, ingested trehalose at doses of up to 25-30 g without the oc-
currence of gastrointestinal symptoms (Bolte et al., 1973; Arola et
al., 1999; Heine et al., 1996; Pointner et al., 1974; Fiehring et
al., 1976).
10. Summary
Trehalose is a naturally occurring disaccharide which consists of

two glucose molecules linked in 1,1-position by an α-glycosidic
bond.
Its sweetness relative to that of sucrose is about 40-45%. Trehalose

is stable under all pH and temperature conditions that are typically
encountered in food processing. Being a non-reducing sugar, treha-
lose does not undergo Maillard reactions. The nutritional properties
of trehalose correspond to those of maltose. Both sugars are hydro-
lyzed to glucose by glucosidases of the small intestinal brush-
border membrane. The resulting glucose is absorbed.
Trehalose occurs widely in nature. It is produced in bacteria and

yeast cells, fungi, algae, and a few higher plants. In many insects
it is produced as a major blood sugar and as a reserve carbohydrate
during periods of dehydration and freezing. Trehalose also occurs at
low concentrations in a number of foods which are consumed as part
of a regular diet (mushrooms, bread, fermented beverages, honey).
The Advisory Committee on Novel Foods and Processes (ACNFP) assessed

the safety of trehalose, extracted from yeast, in 1990. The intended
food applications that were considered by the Committee at that
time, included the use of trehalose for the stabilization of certain
foodstuffs during the drying process and upon rehydration (e.g.,
milk powder, dry soups). On advice of the ACNFP, the Committee on
Toxicity (COT), and the Food Advisory Committee (FAC), trehalose was
accepted for use in foods (except infant formulae and follow-up
milks) in April 1991 (MAFF, 1991; ACNFP, 1990, 1991).
Hayashibara Co., Ltd., Okayama (Japan) has developed a novel enzy-

matic process for producing trehalose from food-grade starch on a
commercial scale. The two crucial enzymes in the process are maltoo-
ligosyl trehalose synthase and maltooligosyl trehalose trehalohydro-
lase. Both are obtained from a strain of Arthrobacter ramosus. This
microorganism is generally regarded as nonpathogenic and non-
toxigenic. Additional enzymes used in the process are isoamylase
from a strain of Pseudomonas amyloderamosa, cyclodextrin glucano-
transferase (CGTase) from a strain of Bacillus stearothermophilus,
glucoamylase from a strain of Aspergillus niger and α-amylase from a
strain of Bacillus subtilis. The isoamylase preparation has been
subjected to standard Ames tests. An acute toxicity test and a sub-
chronic (90-day) toxicity study in rats were conducted as well. The
isoamylase preparation was found to not be mutagenic or toxic under
the conditions of these experiments. The safety of all other enzymes
and source organisms has been evaluated by JECFA earlier already.
Trehalose produced according to this process, which includes differ-

ent purification steps, has a purity of > 98%. HPLC analysis re-
vealed the presence of three glycosidic by-products (glucose, glyco-
syltrehalose, α-D-isomaltosyl-α-D-glucoside).
Trehalose has been used in foods in Japan (since 1995), South Korea
and Taiwan. It is used for moisture retention, lower sweetness (in
comparison to sucrose), and advantageous properties during drying,
freezing and dehydration. Japanese foods formulated with trehalose
include certain types of confectionery, cakes, cookies, fruit pu-
rees, desserts, health or sport drinks, as well as rice and noodles.
The intake of trehalose with these foods varies between 1-10 g per
serving. In a few cases (e.g., candies, certain Japanese confection-
ery) it may reach 20 g per serving. It is estimated that trehalose
will be used in Western foods for the same technological reasons but
that the type of foods may differ somewhat. Calculation of the in-
takes which may result from the anticipated uses in Western food in-
dicates that not more than 8.1 g trehalose will be consumed per eat-
ing occasion by the average user of such products (18.6 g by the 90th
percentile consumer). The estimated average daily intake of treha-
lose from all its uses combined is less than 8 g.
The metabolism of ingested trehalose resembles that of maltose (or

starch) in that both products are absorbed in the form of glucose.
Only a very small fraction of trehalose, if any, may be absorbed as
such. In humans, trehalose entering the blood circulation would be
eliminated by trehalase in the liver, kidneys and plasma.
The safety of trehalose has been examined in acute and subchronic

toxicity tests in mice, rats and dogs using oral and intravenous ad-
ministration of the test substance. Trehalose produced by the enzy-
matic process was examined in tests for mutagenicity, chromosomal
aberration, and micronucleus formation, embryotoxicity/teratogeni-
city studies in rats and rabbits, a 2-generation reproduction study
in rats, and a 3-month toxicity study in mice. The results of these
studies show that trehalose is not genotoxic. Trehalose is generally
well tolerated even if ingested at high doses, and is not toxic.
Trehalose does not impair reproductive performance and is not em-
bryotoxic or teratogenic. Long-term toxicity/carcinogenicity studies
were not conducted with trehalose. However, considering the known
metabolism of trehalose, the high purity of the substance, and the
available toxicological data, it would appear that the safety as-
sessment can be concluded even in the absence of chronic studies.
There are numerous data on the intestinal tolerance of trehalose in

human volunteers. Depending upon individual sensitivity and possibly
race, single doses of 20-50 g are usually tolerated without abdomi-
nal side-effects. At higher levels, the presence in the gut of os-
motically active, incompletely absorbed carbohydrate manifests it-
self by abdominal symptoms, such as nausea, borborygmi, colic, and
laxation. Qualitatively these effects are similar to those which are
known for polyols. However, substantially higher amounts of treha-
lose are required to elicit such effects except in subjects who are
deficient in intestinal trehalase. The incidence of hereditary and
acquired trehalase deficiency appears to be very low. In the Western
population it is likely to be far below 1%28.
In conclusion, there is a substantial body of evidence to support

the safety of trehalose produced by a novel enzymatic process as a
food ingredient. On basis of the available toxicological data and in
consideration of the close similarity between the metabolism of tre-
halose and maltose, it is concluded that trehalose does not present
a significant risk for human health at the intake which would result
from its intended uses in food.
28
A higher incidence may be found in subjects who suffer from intestinal disease and in
Eskimos. However, these people are likely to be aware of their intolerance to sweet
foods because they exhibit a more general disaccharidase deficiency, i.e., they are
also deficient in lactase and sucrase/isomaltase.
11. REFERENCES
Note. Reports printed in italics have been submitted to the

ACNFP but are not made public because they either do not contain
data that are pivotal for the safety assessment of trehalose pro-
duced by the enzymatic process and/or are not required by the
guidelines accompanying the Novel Foods Regulation.
ACNFP (1990). Advisory Committee on Novel Food and Processes, An-

nual Report 1990, p.4.
ACNFP (1991). Advisory Committee on Novel Food and Processes, An-

nual Report 1991, p. 33-34.
Aguilera J.M. and Karel M. (1997). Preservation of biological mate-

rials under desiccation. Crit. Rev. Fd. Sci. Nutr. 37 (3): 287-
309.
Arola H., Koivula T., Karvonen A.-L., Jokela H., Ahola T. and
Isokoski M. (1999). Low trehalase activity is associated with
abdominal symptoms caused by edible mushrooms. Scand. J. Gas-
troenterol. 34: 898-903.
Atkinson J.E. and Thomas B.J. (1994a). An acute toxicity study of

trehalose in the albino mouse (Study No. 434), Unpublished re-
port of Frederick Research Centre, Frederick (MD) for Quadrant
Holdings, Cambridge, UK.
Atkinson J.E. and Thomas B.J. (1994b). An acute toxicity study of
trehalose in the albino rat (Study No. 435), Unpublished report
of Frederick Research Centre, Frederick (MD) for Quadrant Hold-
ings, Cambridge, UK.
Atkinson J.E. and Thomas B.J. (1994c). An acute toxicity study of

trehalose in the Beagle dog (Study No. 436), Unpublished report
Atkinson J.E. and Thomas B.J. (1994d). A 14-day toxicity study of

trehalose in the albino mouse (Study No. 459), Unpublished re-
port of Frederick Research Centre, Frederick (MD) for Quadrant
Holdings, Cambridge, UK.
Atkinson J.E. and Thomas B.J. (1994e). A 14-day toxicity study of

trehalose in the Beagle dog (Study No. 460), Unpublished report
Atkinson J.E. and Thomas B.J. (1994f). An acute eye irritation

study of trehalose in New Zealand white rabbits (Study No.
458), Unpublished report of Frederick Research Centre, Freder-
ick (MD) for Quadrant Holdings, Cambridge, UK.
Becker A., Schlöder P., Steele J.E. and Wegener G. (1996). The
regulation of trehalose metabolism in insects. Experientia 52:
433-439.
Behm C.A. (1997). The role of trehalose in the physiology of nema-

todes. Int. J. Parasitol. 27 (2): 215-229.
Berg N.O., Dahlqvist A., Lindberg T. and Nordén Å. (1973). Correla-
tion between morphological alterations and enzyme activities in
the mucosa of the small intestine. Scand. J. Gastroenterol. 8:
703-712.
Bergoz R. (1971). Trehalose malabsorption causing intolerance to

mushrooms. Gastroenterol. 60 (5): 909-912.
Bergoz R., Bolte J.P. and Meyer zum Bueschenfelde K.-H. (1973).
Trehalose tolerance test. Scand. J. Gastroenterol. 8: 657-663.
Bergoz R., Vallotton M.-C. and Loizeau E. (1982). Trehalase defi-

ciency. Ann. Nutr. Metabol. 26: 191-195.
Birch G.G. (1963). Trehaloses. Advances in Carbohydrate Chemistry

18: 201-225.
Birch G. (1965). A method of obtaining crystalline anhydrous α,α-

trehalose. J. Chem. Soc. 3: 3489.
Birch G.G., Cowell N.D. and Eyton D. (1970). A quantitative inves-

tigation of Shallenberger's sweetness hypothesis. J. Food Tech-
nol. 5: 277-280.
Bolte J.P., Schönhage F., Förster E., Knolle J. and Meyer zum
Büschenfelde K.H. (1973). Zur diagnostischen Bedeutung der Tre-
halose-Belastung bei Malassimilationssyndromen. Dtsch. med.
Wschr. 98: 1358-1362.
Brunet C., Lamare S. and Legoy M-D. (1998). Studies of specific cy-
clodextrin production starting from pure maltooligosaccharides
using Thermoanaerobacter sp. cyclodextrin glycosyltransferase.
Biocatalysis and Biotransformation 16: 317-327.
Candy D.J., Becker A. and Wegener G. (1997). Coordination and inte-

gration of metabolism in insect flight. Comp. Biochem. Physiol.
117B (4): 497-512.
Cerda J.J., Preiser H. and Crane R.K. (1972). Brush border enzymes
and malabsorption: elevated disaccharides in chronic pancreatic
insufficiency with diabetes mellitus. Gastroenterol. 62: 841.
Clarke F., Solkot R. and Corley R.C. (1939). Metabolism of certain

rare sugars. J. Biol. Chem. 131: 135-138.
Colaço C.A.L.S. and Roser B. (1995). Trehalose - a multifunctional

additive for food preservation. In "Food Packaging and Preser-
vation", M. Mathlouthi (Ed.), Blackie Professional, London, p.
123-140.
Colaço C., Sen S., Thangavelu M., Pinder S. and Roser B. (1992).
Extraordinary stability of enzymes dried in trehalose: simpli-
fied molecular biology. Bio/Technol. 10: 1007-1010.
Colaço C.A.L.S., Smith C.J.S., Sen S., Roser D.H., Newman Y., Ring
S. and Roser B.J. (1994). Chapter 14. Chemistry of protein sta-
bilization by trehalose. In: "Formulation and Delivery of Pro-
teins and Peptide", American Chemical Society, p. 222-240.
Courtois J.E. and Demelier J.F. (1966). Répartition de la tréhalase

chez l'homme et quelques mammifères. Bull. Soc. Chim. Biol. 48
(2): 277-286.
Crowe J.H., Crowe L.M., Carpenter J.F., Rudolph A.S., Aurell Wis-
trom C., Spargo B.J. and Anchordoguy T.J. (1988). Interactions
of sugars with membranes. Biochim. Biophys. Acta 947: 367-384.
Dahlqvist A. (1974). Enzyme deficiency and malabsorption of carbo-

hydrates. In: "Sugars in Nutrition", Sipple H.L and McNutt K.W.
(eds.), Academic Press, New York, p. 187-214.
Demelier J.-F., Labat J. and Courtois J.-E. (1975). Influence de

l'injection parentérale de tréhalose sur des mammifères à
équipements tréhalasiques distincts. C. R. Acad. Sc. (Paris)
Serie D, 280 (5): 669-672.
Dominick H.C. and Anspach K. (1975). Der Trehalose-Belastungstest

bei Malabsorptions-Syndrom. Mschr. Kinderheilk. 123: 432-433.
Elbein A.D. (1974). The metabolism of α,α-Trehalose. Adv. Carbo-

hydr. Chem. Biochem. 30: 227-256.
Eze L.C. (1989). Plasma trehalase activity and diabetes mellitus.

Biochem. Genetics 27 (9/10): 487-495.
FAO (1992a). Compendium of food additive specifications. FAO Paper

52/1, p. 357 (Carbohydrase from Aspergillus niger).
FAO (1992b). Compendium of food additive specifications. FAO Paper

52/1, p. 363 (Carbohydrase from Bacillus licheniformis).
FAO (1993). Compendium of food additive specifications. FAO Paper

52, p. 11 (Alpha-amylase from Bacillus subtilis).
Fiehring C., Koslowski H. Jung G. and Wedekind U. (1976). Unter-

suchungen über den Wert der Trehalaseaktivitätsbestimmung im
Kindesalter bezüglich der Resorptionsfähigkeit des Darmes.
Dt. Gesundh.-Wesen 31 (23): 1091-1095.
Funke G., Hutson R.A., Bernard K.A., Pfyffer G.E., Wauters G. and
Collins M.D. (1996). Isolation of Arthrobacter spp. from clini-
cal specimens and description of Arthrobacter cumminsii sp.
nov. and Arthrobacter woluwensis sp. nov. J. Clin. Microbiol.
34 (10): 2356-2363.
Galand G. (1989). Brush border membrane sucrase-isomaltase, mal-

tase-glucoamylase and trehalase in mammals. Comp. Biochem.
Physiol. 94B (1): 1-11.
Ghasempour H.R., Gaff D.F., Williams R.P.W. and Gianello R.D.

(1998). Contents of sugars in leaves of drying desiccation tol-
erant flowering plants, particularly grasses. Plant Growth
Regulation 24: 185-191.
Goddijn O. and Smeekens S. (1998). Sensing trehalose biosynthesis

in plants. Plant J. 14 (2): 143-146.
Green J.L. and Angell C.A. (1989). Phase relations and vitrifica-
tion in saccharide-water solutions and the trehalose anomaly.
J. Phys. Chem. 93: 2880-2882.
Gudmand-Høyer E., Fenger H.J., Skovbjerg H., Kernhansen P. and Rør-

baek Madsen P. (1988). Trehalase deficiency in Greenland.
Scand. J. Gastroenterol. 23: 775-778.
Gudmand-Høyer E. and Skovbjerg H. (1996). Disaccharide digestion and

maldigestion. Scand. J. Gastroenterol. 31 (Suppl. 216): 111-
121.
Gupta S.K., Chong S.K.F. and Fitzgerald J.F. (1999). Disaccharidase
activities in children: normal values and comparison based on
symptoms and histologic changes. J. Pediatr. Gastroenterol.
Nutr. 28: 246-251.
Heine W., Mohr Ch. and Münch Ch. (1996). Zuckerverbindungen als
Substrate für Wasserstoff-Atemgasteste: Korrelationen zur Do-
sis, orozökalen Transitzeit, Mikroflora und zu Nebenwirkungen.
Pädiatr. Grenzgeb. 34: 481-490.
Hietanen E. (1973). Interspecific variation in the levels of intes-

tinal alkaline phosphatase, adenosine triphosphatase and disac-
charidases. Comp. Biochem. Physiol. 46A: 359-369.
Hore P. and Messer M. (1968). Studies on disaccharidase activities

of the small intestine of the domestic cat and other carnivo-
rous mammals. Comp. Biochem. Physiol. 24: 717-725.
Iglesias H.A., Chirife J. and Buera M.P. (1997). Adsorption iso-

therm of amorphous trehalose. J. Sci. Fd. Agric. 75: 183-186.
Isichei U.P. and Gorecki T. (1993). Serum trehalase activities in

controlled and uncontrolled diabetes and the impact of oral
glucose, high carbohydrate and glycosuria on serum levels. Afr.
J. Med. med. Sci. 22: 5-11.
Jönsson K.Å., Bodemar G., Tagesson C. and Walan A. (1985). Varia-

tion of disaccharidase activities in duodenal biopsy specimens.
Scand. J. Gastroentrol 21: 51-54.
Kato M. (1999). Trehalose production with a new enzymatic system
from Sulfolobus solfataricus KM1. J. Mol. Catalysis B: Enzy-
matic 6: 223-233.
Kitching J. (1995). Trehalose (crystal) bacterial mutation assay.

Unpublished report of Huntingdon Research Centre Ltd., Hunting-
don (UK) for Hayashibara Biochemical Laboratories Inc., Oka-
yama-ken, Japan. p. 1-34.
Koch E.M. and Koch F.C. (1925). The presence of trehalose in yeast.
Science 61 (1587): 570-572.
Kurimoto M., Nishimoto T., Nakada T., Chaen H., Fukuda S. and
Tsujisaka Y. (1997a). Synthesis by an α-glucosidase of
glycosyl-trehalose with an isomaltosyl residue. Biosci.
Biotech. Biochem. 61 (4): 699-703.
Kurimoto M., Tabuchi A., Mandai T., Shibuya T., Chaen H., Fukuda
S., Sugimoto T. and Tsujisaka Y. (1997b). Synthesis of
glycosyl-trehalose by cyclomaltodextrin glucanotransferase
through the transglycosylation reaction. Biosci. Biotech.
Biochem. 61 (7): 1146-1149.
Kvasnikov E.I., Pisarchuk E.N., Ivanova L.L., Gushcha K.P. and

Andrienko V.I. (1986). Taxonomic importance of methods of
reproduction and types of separation of dividing cells in
species of the genus Arthrobacter. Mikrobiologiya 55 (4): 662-
668.
Labat-Robert J. (1982). Trehalases. In: "Developments in food

carbohydrate, Volume 3: Disaccharidases". Lee C.K. and
Lindley M.G. (eds.), Applied Science Publishers Ltd., London,
p. 84-106.
Lammert A.M., Schmidt S.J. and Day G.A. (1998). Water activity and
solubility of trehalose. Food Chem. 61 (1/2): 139-144.
Lee C.-K. and Birch G.G. (1975). Structural functions of taste in

the sugar series: binding characteristics of disaccharides. J.
Sci. Fd. Agric. 26: 1513-1521.
Leibowitz J. (1943). A new source of trehalose. Nature 152: 414.
Leibowitz J. (1944). Isolation of trehalose from desert manna.

Biochem. J. 38: 205-206.
Leslie S.B., Israeli E., Lighthart B., Crowe J.H. and Crowe L.M.
(1995). Trehalose and sucrose protect both membranes and
proteins in intact bacteria during drying. Appl. Environm.
Microbiol. 61 (10): 3592-3597.
Lina B.A.R. (1999). Sub-chronic (13-wk) oral toxicity study with

isomylase in rats. Unpublished report No. V99.646 of TNO Nu-
trition and Food Research Institute, Zeist, The Netherlands
for Hayashibara Co., Ltd., Okayama-ken, Japan.
LutkiC A., Šir S., Varga l., Primer S. and Turk R. (1985). Disac-
charidases in jejunal mucosa of children. Acta Pharm. Jugosl.
35: 61-69.
Madzarovová-Nohejlova J. (1973). Trehalase deficiency in a family.

Gastroenterol. 65: 130-133.
Maestracci D. (1976). Enzymic solubilization of the human intesti-
nal brush border membrane enzymes. Biochim. Biophy. Acta 433:
469-481.
MAFF 1991. Ministry of Agriculture, Fisheries & Food, Food Safety

Directorate, News Release (FSD 27/91) of April 16, 1991.
Mališ F., Lojda Z., Fric P. and Jodl J. (1972). Disaccharidases in

celiac disease and mucoviscidosis. Digestion 5: 40-48.
Mandai T., Shibuya T., Sugimoto T. and Miyake T. (1999). Non-

reducing saccharide and its production and use. US Patent Nr.
5,919,668, Japan.
Maruhn D., Fuchs I., Mues G. and Bock K. D. (1976). Normal limits
of urinary excretion of eleven enzymes. Clin. Chem. 22 (10):
1567-1574.
Maruta K., Hattori K., Nakada T., Kubota M., Sugimoto T. and
Kurimoto M. (1996). Cloning and sequencing of trehalose
biosynthesis genes from Arthrobacter sp. Q36. Biochim. Biophys.
Acta 1289: 10-13.
Maruta K., Nakada T., Kubota M., Chaen H., Sugimoto T., Kurimoto
M., Tsujisaka Y. (1995). Formation of trehalose from
maltooligosaccharides by a novel enzymatic system. Biosci.
Biotechnol. Biochem. 59: 1829-1834.
Masiak D., Martyniuk S. and Mysków W. (1991). Investigations of

Arthrobacter globiformis pupulations in soil using a
serological method. Zentralbl. Mikrobiol. 146: 125-129.
Mc Rae L.A. (1995). Trehalose crystals. Acute oral toxicity to the
rat. Unpublished report of Huntingdon Research Centre Ltd.,
Huntingdon U.K. for Hayashibara Biochemical Laboratories Inc.,
Okayama-ken, Japan.
Mehl P.M. (1997). Solubility and glass transition in the system α-

D-trehalose/water. J. Therm. Anal. 49: 817-822.
Mendes E., Brojo Proença E., Ferreira I.M.P.L.V.O. and Ferreira

M.A. (1998). Quality evaluation of Portuguese honey. Carbohydr.
Polym. 37: 219-223.
Miller D.P., de Pablo J.J. and Corti H. (1997). Thermophysical

properties of trehalose and its concentrated aqueous solutions.
Pharmaceut. Res. 14 (5): 578-590.
Miyake S. (undated). Effects of trehalose in food applications.

Internal document of Hayashibara Co. Ltd., Okayama-ken, Japan.
Morimoto H., Noro H. and Ohtaki H. (1979 a). Acute toxicity test of
Pseudomonas amyloderamosa (wet cells). Unpublished report Nr.
12110175-1 of Japan Food Research Laboratories, Tokyo, Japan,
for Hayashibara Co., Ltd., Okayama-ken, Japan.
Morimoto H., Noro H. and Ohtaki H. (1979 b). Acute toxicity test of
Pseudomonas amyloderamosa (culture filtrate). Unpublished
report Nr. 12110175-2 of Japan Food Research Laboratories,
Tokyo, Japan, for Hayashibara Co., Ltd., Okayama-ken, Japan.
Morimoto H., Noro H. and Ohtaki H. (1979 c). Acute toxicity test of
isoamylase (of Pseudomonas amyloderamosa origin). Unpublished
report Nr. 12110175-3 of Japan Food Research Laboratories,
Tokyo, Japan, for Hayashibara Co., Ltd., Okayama-ken, Japan.
Müller J., Wiemken A. and Aeschbacher R. (1999). Trehalose

metabolism in sugar sensing and plant development. Plant Sci.
147: 37-47.
Murray I.A., Coupland K., Smith J.A., Ansell I.D. and Long R.G.
(2000). Intestinal trehalase activity in a UK population: es-
tablishing a normal range and the effect of disease. Br. J.
Nutr. 83: 241-245.
Nagaoka K., Taniguchi M. and Yamauchi H. (1997). Development of

Elisa for enzymes used in trehalose production and residual en-
zymes in trehalose. Unpublished report of Fujisaki Institute,
Hayashibara Biochemical Labs., Inc., Japan.
Nakada T., Maruta K., Mitsuzumi H., Kubota M., Chaen H., Sugimoto
T., Kurimoto M. and Tsujisaka Y. (1995). Purification and char-
acterization of a novel enzyme, maltooligosyl trehalose treha-
lohydrolase, from Arthrobacter sp. Q36. Biosci. Biotech. Bio-
chem. 59 (12): 2215-2218.
Nakada T., Maruta K., Tzusaki K., Kubota M., Chaen H., Sugimoto T.,
Kurimoto M. and Tsujisaka Y. (1995). Purification and proper-
ties of a novel enzyme, maltooligosyl trehalose synthase, from
Arthrobacter sp. Q36. Biosci. Biotech. Biochem. 59 (12): 2210-
2214.
Nicolaus B., Gambacorta A., Basso A.L., Riccio R., De Rosa M. and
Grant W.D. (1988). Trehalose in Archaebacteria. System. Appl.
Microbiol. 10: 215-217.
Niederst C. and Dauça M. (1985). Activités des hydrolases rénales
au cours du développement pré- et post-natal de la souris. Can.
J. Physiol. Pharmacol. 63: 731-734.
Niessen K.H., Schmidt K. and Brügmann G. (1975). Disaccharidasen

der Dünndarmschleimhaut bei Säuglingen und Kindern. Z. Gastro
enterol 6: 565-572.
O'Brien J. (1996). Stability of trehalose, sucrose and glucose to

nonenzymatic browning in model systems. J. Fd. Sci. 61 (4):
679-682.
Oku T. and Okazaki M. (1998). Transitory laxative threshold of tre-

halose and lactulose in healthy women. J. Nutr. Sci. Vitaminol.
44: 787-798.
Oku K., Sawatani I., Chaen H., Fukuda S. and Kurimoto M. (1998).
Trehalose content in foods. Nippon Shokuhin Kagaku Kogaku Kai-
shi 45 (6): 381-384.
Pointner H., Flegel U., Bayer P.M. and Schmidt P. (1974). Anwendung
des Trehalose-Toleranztests zur Untersuchung der Dünndarmre-
sorption bei Patienten unter chronischer Hämodialyse. Wiener
Klin. Wschr. 22: 691-693.
Portmann M.O. and Birch G. (1995). Sweet taste and solution proper-
ties of α,α-trehalose. J. Sci. Fd. Agric. 69: 275-281.
Qiu L.H. and Bedding R. (1999). Low temperature induced cryoprotec-

tant synthesis by the infective juveniles of Steinernema carpo-
capsae: biological significance and mechanisms involved. Cryo –
Lett. 20 (6): 393-404.
Riby J., Sunshine S. and Kretchmer N. (1990). Renal trehalase:
Function and development. Comp. Biochem. Physiol. 95A (1): 95-
99.
Rodeck U. and Dominick H. Chr. (1983). Trehalose-load-test in gas-

troenterology. Monatsschr. Kinderheilkd. 131: 19-22.
Roser B. (1991). Trehalose, a new approach to premium dried foods.

Trends Food Sci. Technol. 7: 166-169.
Ruppin H., Domschke W., Domschke S. and Classen M. (1974). Intesti-

nale Disaccharidasen bei juvenilem Diabetes mellitus. Klin.
Wschr. 52: 568-570.
Sasai-Takedatsu M., Taketani S., Nagata N., Furukawa T., Tokunaga

R., Kojima T. and Kobayashi Y. (1996). Human trehalase: charac-
terization, localization, and its increase in urine by renal
proximal tubular damage. Nephron 73: 179-185.
Sato S., Okamoto K., Minami R., Kohri H. and Yamamoto S. (1999).
Trehalose can be as a parenteral saccharide source in rabbits.
J. Nutr. 129: 158-164.
Schmid H., Biedermann K., Luetkemeier H., Weber K. and Wilson J.

(1998). Subchronic 13-week oral toxicity (feeding) study with
trehalose in mice. Unpublished report (RCC Project 639213) of
RCC Group, Ittingen, Switzerland for Hayashibara Co., Ltd.,
Okayama-ken, Japan.
Silljé H.H.W., Paalman J.W.G., ter Schure E.G., Olsthoorn S.Q.B.,

Verkleij A.J., Boonstra J. and Verrips C.T. (1999). Function of
trehalose and glycogen in cell cycle progression and cell vi-
ability in Saccharomyces cerevisiae. J. Bacteriol. 181 (2):
396-400.
Singer M.A. and Lindquist S. (1998a). Thermotolerance in Saccharo-

myces cerevisiae: the yin and yang of trehalose. Trend Bio-
techn. 16 (11): 460-468.
Singer M.A. and Lindquist S. (1998b). Multiple effects of trehalose

on protein folding in vitro and in vivo. Molecular Cell 1: 639-
648.
Sola-Penna M. and Meyer-Fernandes J.R. (1998). Stabilization

against thermal inactivation promoted by sugars on enzyme
structure and function: why is trehalose more effective than
other sugars? Arch. Biochem. Biophys. 360 (1): 10-14.
Steiner A. and Cori C.F. (1935). The preparation and determination

of trehalose in yeast. Science 82 (2131): 422-423.
Stewart L. C., Richtmyer N. K. and Hudson C. S. (1950). The prepa-

ration of trehalose from yeast. J. Nat. Inst. Health 72: 2059-
2061.
Streeter J.G. and Bhagwat A. (1999). Biosynthesis of trehalose from

maltooligosaccharides in Rhizobia. Can. J. Microbiol. 45: 716-
721.
Sugimoto T. (1995). Production of trehalose by enzymatic starch

saccharification and its applications. Shokuhin Kogyo (Food
Industry) 38(10): 34-39.
Sussich F., Urbani R., Princivalle F. and Cesàro A. (1998). Poly-
morphic amorphous and crystalline forms of trehalose. J. Am.
Chem. Soc. 120: 7893-7899.
Tabuchi A., Shibuya T., Sugimoto T. and Miyake T. (1999).

Crystalline maltosyl glucoside, and its production and use. US
Patent Nr. 5,912,330.
Thevelein J.M. (1984). Regulation of trehalose mobilization in

fungi. Microbiol. Rev. 48 (1): 42-59.
Thompson S.N. (1999). Blood sugar formation from dietary carbohy-

drate is facilitated by the pentose phosphate pathway in an in-
sect Manduca sexta Linnaeus. Biochim. Biophys. Acta – Gen. Sub.
1472 (3): 565-575.
Tonkova A. (1998). Bacterial cyclodextrin glucanotransferase. En-

zyme Microb. Techn. 22: 678-686.
Ushijima K., Fujisawa T. and Kretchmer N. (1995). Study on diges-

tion and absorption of trehalose in human intestine. Dig. Ab-
sorp. 18: 56-57.
Van Delft J.H.M. (1999). Bacterial reverse mutation test with

isoamylase. Unpublished report No. V99.526 of TNO Nutrition
and Food Research Institute, Zeist, The Netherlands for Haya-
shibara Co., Ltd., Okayama-ken, Japan.
Van Elburg R.M., Uil J.J., Kokke F.T.M., Mulder A.M., van de Broek
W.G.M., Mulder C.J.J. and Heymans H.S.A. (1995). Repeatabil-
ity of the sugar-absorption test, using lactulose and manni-
tol, for measuring intestinal permeability for sugars. J. Pe-
diatr. Gastroenterol. Nutr. 20: 184-188.
Van Handel E. (1969). Do trehalose and trehalase function in renal

glucose transport? Science 163 (187): 1075-1076.
Van Handel E. (1970). Serum trehalase: assay and normal values.

Clin. Chim. Acta 29: 349-353.
Vázquez C., Escobar H., Polanco I., Codoceo R. and Vitoria J.C.
(1975). Malabsorción de los hidratos de carbono en el niño.
An. Esp. Pediat. 8: 105-194.
Waalkens-Berendsen D.H. (1998). Oral embryotoxicity/teratogenicity

study with trehalose in rats. Unpublished report No. V98.551
of TNO Nutrition and Food Research Institute, Zeist, The
Netherlands for Hayashibara Co., Ltd., Okayama-ken, Japan.
Welsh J.D., Poley J.R., Bhatia M. and Stevenson D.E. (1978). Intes-
tinal disaccharidase activities in relation to age, race, and
mucosal damage. Gastroenterol. 75: 847-855.
WHO (1972). WHO Technical Report Series 488, 9-13 (Comments on sub-
stances on the Agenda: Enzymes).
WHO (1987). WHO Food Additives Series 20: 3-7 (Carbohydrase (α-
amylase) from Bacillus lichenformis).
WHO (1988). WHO Food Additives Series 22: 11-14 (Enzymes derived
from Aspergillus niger).
WHO (1991a). WHO Food Additives Series 28: 63-66 (Alpha-amylase

from Bacillus stearothermophilus).
WHO (1991b). WHO Food Additives Series 28: 67-70 (α-Amylase from
Bacillus subtilis).
WHO (1998). WHO Food Additives Series 40: 91-100 (Maltogenic amy-
lase).
WHO (1999). WHO Food Additives Series 42: 183-202 (γ-cyclodextrin).
Winegar R.A. (1997a). Evaluation of trehalose in the CHO chromosome

aberration assay. Unpublished Report No. G018-97 of SRI Inter-
national, Menlo Park (CA) for Hayashibara Biochemical Laborato-
ries, Inc., Okayama-ken, Japan.
Winegar R.A. (1997b). Evaluation of trehalose in the mouse micronu-

cleus assay. Unpublished Report No. G019-97 of SRI Interna-
tional, Menlo Park (CA) for Hayashibara Biochemical Laborato-
ries, Inc., Okayama-ken, Japan.
Wolterbeek A.P.M. and Waalkens-Berendsen D.H. (1999a). Oral two-

generation reproduction study with trehalose in rats. Unpub-
lished report No. V99.280 of TNO Nutrition and Food Research
Institute, Zeist, The Netherlands for Hayashibara Co., Ltd.,
Okayama-ken, Japan.
Wolterbeek A.P.M. and Waalkens-Berendsen D.H. (1999b). Oral embryo-

toxicity/teratogenicity study with trehalose in New Zealand
white rabbits. Unpublished report No. V98.797 of TNO Nutri-
tion and Food Research Institute, Zeist, The Netherlands for
Hayashibara Co., Ltd., Okayama-ken, Japan.
Wyatt G.R. and Kalf G.F. (1957). The chemistry of insect hemolymph.
II. Trehalose and other carbohydrates. J. Gen. Physiol. 40:
833-847
Yokobayashi K. (1988). Isoamylases (Pseudomonas amyloderamosa,

etc.). Handbook of Amylases and Related Enzymes, pp. 125-130.
Yoshida K. (1993). Serum trehalase activity in patients with

rheumatoid arthritis. Clin. Chim. Acta 215: 123-124.
Table 1: Applications of trehalose and maximum levels of use
[The data contained in this table are related to marketing know-how for which the
applicant claims confidentiality]
TREHALOSE_EU_PUBLIC_TABLE_1_120500.xls
Table 2 Estimated intake of trehalose from the combined proposed
uses in food1)
Age Group Eating occasion Intake per user (g)

mean 90th perc.
Children (age 2-12 years)
Per Day 5.2 10.9
Breakfast 3.6 5.8

Lunch2) 4.1 7.8
Dinner3) 4.8 9.2
Snack 3.7 7.5
Other occasions 5.8 11.9
Teenagers (age 13-19 years)
Per Day 7.5 15.2
Breakfast 5.1 10.2

Lunch2) 6.6 12.5
Dinner3) 7.1 17.3
Snack 5.2 9.9
Adults (age > 20 years)
Per Day 7.2 16.4
Breakfast 3.9 7.6

Lunch2) 6.1 15.0
Dinner3) 8.1 18.6
Snack 5.5 11.9
1
Based on USDA CSFII 1994-96 data, chewing gum intake not included
2
Lunch combines reported consumption at both lunch and brunch
3
Dinner combines reported consumption at both dinner and supper
Table 3: Acute toxicity of trehalose
Species Sex Route LD50 Reference

(g/kg bw)
Mouse Male and female oral >5 Atkinson & Thomas (1994 a)
Male and female intravenous >1 Atkinson & Thomas (1994 a)
Rat Male and female oral > 16 Mc Rae (1995)

Male and female oral > 5 Atkinson & Thomas (1994 b)
Male and female intravenous > 1 Atkinson & Thomas (1994 b)
Dog Male oral >5 Atkinson & Thomas (1994 c)

Male intravenous >1 Atkinson & Thomas (1994 c)
Table 4 : Results of assays for the genotoxicity of trehalose
End-point Test object Concentration Result Reference
Bacterial gene S. typhimurium TA 1535, 312-5000 Negative a Kitching, 1995

mutation TA 1537, TA 98, TA 100; µg/plate
E.coli WP2 uvr A trp
Chromosomal Chinese hamster ovary (CHO) 312-5000 Negative a Winegar, 1997

aberration b cells µg/plate
Micronucleus Swiss-Webster mouse bone 1250-5000 mg/kg bw Negative Winegar, 1997

formation c marrow (i.p., single dose)
a
With and without exogenous metabolic activation (MA)
b
Two assays were conducted: treatment times 3 h (with and without MA); 3 h (with MA) and 21 h (without MA).
Harvest times, 21 and 45 h
c
Harvest times, 24 and 48 h
TREHALOSE
Figure 1: Trehalose is a disaccharide consisting of 2 glucose moieties

linked by a glycosidic bond that joins their anomeric carbons.
TREHALOSE_ACNFP_FIG_1_150200.doc
Figure 2: Action of trehalose producing enzymes
4 1 4 1 4 1 4 1
Maltooligosaccharide
4 1 4 1 4 1
MTSase
(EC 5.4.99.15)
n
4 1 4 1
1
MTHase
(EC 3.2.1.141)
n
1
Glucose (n=0), maltose (n=1)
or maltooligosaccharide (n>2) TREHALOSE
TREHALOSE_EU_PUBLIC_FIG_2_120500.ppt
Figure 3: Flow chart of trehalose production
Starch
Slurry prepartion
Liquefaction Thermostable α-amylase
Isoamylase
Saccharification MTSase
MTHase
Decoloration CGTase
Deionization Glucoamylase
α-amylase
Concentration
First Crystallization
Centrifugation
Wet Crystals 1 Mother Liquid 1
Concentration
Second Crystallization
Centrifugation
Wet Crystals 2 Mother Liquid 2
Chromatographic Separation
Mixing / Drying / Cooling Glucose fraction Trehalose fraction
Concentration Concentration
TREHALOSE
TREHALOSE_EU_PUBLIC_FIG_ 3_120500.xls
Annex 1
Data requirements for the evaluation of trehalose pro-

duced by a novel process (according to Commission Rec-
ommendation 97/618/EC and ACNFP decision tree)
TREHALOSE_EU_PUBLIC_ANNEX1_COVERSHEET_MANU1_120500.doc
Company: Bioresco Product: Trehalose
You have identified the novel food as belonging to:
Class 6 - a food produced using a novel process.
You therefore need to supply information to satisfy the

following structured schemes:
I. Specification of the Novel Food

II. Effect of the production process applied to the NF
III. History of the organism used as the source of the NF
IX. Anticipated intake/extent of use of the NF
X. Information from previous human exposure to the NF or
its source
XI. Nutritional information on the NF
XII. Microbiological information on the NF
XIII. Toxicological information on the NF
Select OK to continue.
Scheme I - Specification of the NF
There is sufficient information from scheme I.
Detailed outcome:
Depending on the derivation and composition of the NF, is

appropriate analytical information available on potentially
toxic inherent constituents, external contaminants and
nutrients?
YES
Is the information representative of the NF when produced on a

commercial scale?
YES
Is there an appropriate specification (including species,

taxon etc. for living organisms) to ensure that the NF
marketed is the same as that evaluated?
YES
OUTCOME: All appropriate analytical information and the

appropriate specification of the NF needs to be forwarded with
your application.
Scheme II - Effect of the production process applied to the NF
The result of scheme II is:
There is sufficient information from scheme II
Detailed outcome:
Does the NF undergo a production process?
YES
Is there a history of use of the production process for the

food?
NO
Does the process result in a significant change in the

composition or structure of the NF compared to its traditional
counterpart?
NO
OUTCOME: Provide all relevant information on the production

process.
Scheme III - History of the organism used as the source of the

NF
The outcome of scheme III is:
Sufficient information from scheme III.
Detailed outcome:
Is the NF obtained from a biological source, i.e. a plant,

animal or microorganism?
NO
OUTCOME: No information needed under this scheme.

Scheme IX - Anticipated intake/extent of use of the NF
The result of scheme IX is:
There is sufficient information from scheme IX.
Detailed outcome:
Is there information on the anticipated used of the NF based

on its properties?
YES
OUTCOME: You will need to provide all relevant information.

Scheme X - Information from previous human exposure to the NF

or its source
The result of scheme X is:
There is sufficient information from scheme X.
Detailed outcome:
Is there information from previous direct, indirect, intended

or unintended human exposure to the NF or its source which is
relevant to the Community situation with respect to
production, preparation, population, lifestyles and intakes?
YES
Is there information to demonstrate that exposure to the NF is

unlikely to give rise to nutritional, microbiological,
toxicological and /or allergenicity problems?
YES
OUTCOME: You will need to provide all relevant information.

Schemes XI, XII and XIII discuss requirements for nutritional,
microbiological and toxicological information.
Scheme XI - Nutritional Information on the NF
The result of scheme XI is:
There is sufficient information from scheme XI.
Detailed outcome:
Is there information to show that the NF is nutritionally

equivalent to existing foods that it might replace in the
diet?
YES
OUTCOME: You will need to provide all relevant information to

show the NF is nutritionally equivalent to existing foods that
it might replace in the diet.
Scheme XII - Microbiological Information on the NF
The result of scheme XII is:
There is sufficient information from scheme XII.
Detailed outcome:
Is the presence of any microorganisms or their metabolites due

to the novelty of the product/process?
YES
Is there information to show that the NF is unlikely to

contain microorganisms and/or their metabolites of adverse
public health significance?
YES
OUTCOME: You will need to provide all relevant information on

these microorganisms and/or their metabolites.
Scheme XIII - Toxicological information on the NF
There is sufficient information from scheme XIII.
Detailed outcome:
XIII TOXICOLOGICAL INFORMATION ON THE NOVEL FOOD
Is there a traditional counterpart to the NF that can be used

as a baseline to facilitate the toxicological assessment?
YES
Compared to the traditional counterpart, does the NF contain

new toxicants or changed levels of existing toxicants?
NO
Is there information which suggests that the NF might pose an

allergenic risk to humans?
NO
OUTCOME: There is a traditional counterpart to the NF that

can be used as a baseline to facilitate the toxicological
assessment, and compared to that baseline the NF does not
contain new toxicants or changed levels of existing toxicants.
Any allergenic risk to humans from the NF should also be

investigated, and the information presented. Controlled
allergenicity trials may need to be undertaken where
necessary.
All studies and any cited literature references which support

the answers to the above questions to establish the
toxicological safety of the NF should be forwarded with your
application.
Annex 2
Draft specifications
TREHALOSE_EU_ANNEX2_120500.doc
Trehalose
SYNONYMS α,α-trehalose
DEFINITION A disaccharide which consists of two glu-

cose moieties linked by a 1,1-glucosidic
bond. It is obtained from liquefied starch
by a multi-step enzymatic process.
Chemical name α-D-glucopyranosyl-α-D-glucopyranoside
C.A.S. number 99-20-7 (anhydrous)

6138-23-4 (dihydrate)
Chemical formula C12 H22 O11 (anhydrous)

C12 H22 O11 • 2H2O (dihydrate)
Structural formula
Formula weight 342.30 (anhydrous)

378.33 (dihydrate)
Assay Not less than 98% on an anhydrous basis
DESCRIPTION Virtually odorless, white or almost white

crystals (dihydrate) with a sweet taste
FUNCTIONAL USES Bulk sweetener, texturizer, stabilizer, hu-

mectant, formulation aid
TREHALOSE_ACNFP_ANNEX_2_150200.doc
-2-
CHARACTERISTICS
IDENTIFICATION
Solubility Freely soluble in water, very slightly

soluble in ethanol
Specific rotation [α]20 : +199° (5% aqueous solution of dihydrate)

D
Melting point 97°C (dihydrate)

210.5°C (anhydrous)
PURITY
Loss on drying Not more than 1.5% (60°C, 5 h) (Crystal wa-

ter of dihydrate is not released under
these conditions)
Total ash Not more than 0.05%
Lead Not more than 1mg/kg

Prepare a sample solution as directed for
organic compounds in the Limit Test and de-
termine by atomic absorption spectroscopy
Microbiological Total (aerobic) plate counts: < 300/g

criteria Coliforms: Negative by test
Salmonella: Negative by test
Yeast and molds: < 100/g
-3-
METHOD OF ASSAY Principle: Trehalose is identified by liq-
uid chromatography and quantified by com-
parison to a reference standard containing
standard trehalose.
Preparation of sample solution: Weigh accu-

rately about 3 g of dry sample into a 100-
ml volumetric flask and add about 80 ml of
purified, deionized water. Bring sample to
complete dissolution and dilute to mark
with purified deionized water. Filter
through a 0.45 micron filter.
Preparation of standard solution: Dissolve

accurately weighed quantities of dry stan-
dard reference trehalose in water to obtain
a solution having known concentration of
about 30 mg of trehalose per ml.
Apparatus: Liquid chromatograph equipped

with a refractive index detector and an in-
tegrator recorder.
Conditions:
Column: Shodex Ionpack KS-801 (Showa Denko Co.)

-length : 300 mm
-diameter : 10 mm
-packing : Shodex Ionpack KS-801
-temperature : 50°C
Solvent: Water
Flow rate: 0.4 ml/min
Injection volume: 8 µl
Procedure: Separately inject equal volumes

of the sample solution and the standard so-
lution into the chromatograph. Record the
chromatograms and measure the response of
trehalose peak. Calculate the quantity, in
-4-
mg, of trehalose in 1 ml of the sample so-
lution by the following formula:
RU
125 x C x --
RS
in which C is the concentration, in mg per

ml, of trehalose in the standard prepara-
tion, RU is the peak response of trehalose
in the sample preparation, and RS is the
peak response of trehalose in the standard
preparation.
-5-
Annex 3
Analytical data of representative batches of trehalose
Annex 4
Critical Control Points and Control Standards of the

production process
Specifications of applied raw material, process chemi-

cals, enzymes and ion exchange resins
Annex 5
Current applications and use levels of trehalose in

Japanese foods
Annex 6
Projection of trehalose intake by the dietary survey

approach

Trehalose Produced by A Novel Enzymatic Process

Uploaded by

Trehalose Produced by A Novel Enzymatic Process

Uploaded by

TREHALOSE

produced by a novel enzymatic process

Dossier prepared and submitted on behalf of

Author: Albert Bär PhD

Date: May 18, 2000

10. Summary ................................................ 57

Annex 1 Data requirements for the evaluation of trehalose produced by

Annex 2 Draft specifications

Annex 3 Analytical data of representative batches of trehalose

Annex 4 Critical Control Points and Control Standards of the produc-

Annex 5 Current applications and use levels of trehalose in Japanese

Annex 6 Projection of trehalose intake by the dietary survey approach

Trehalose is a naturally occurring disaccharide which consists of

The Advisory Committee on Novel Foods and Processes (ACNFP) assessed

Hayashibara Co., Ltd., Okayama (Japan) has developed a novel enzy-

The present application for authorization of trehalose produced by

For evaluation of a class 6 product the following information is

I. Specification of the Novel Food (NF)

II. Effect of the production process applied to the NF

III. History of the organism used as the source of the NF

IX. Anticipated intake/extent of use of the NF

XI. Nutritional information on the NF

XII. Microbiological information on the NF

XIII. Toxicological information on the NF

The information presented in this dossier is structured according to

Specifications as submitted to JECFA for evaluation at its 55th meeting

Analyses of representative batches of trehalose demonstrate that

In the pivotal safety studies (genotoxicity tests, embryotoxic-

(g anhydrous trehalose/100ml solution at 20°C): 43.03

(g trehalose dihydrate/100ml solution at 20°C): 46.63

Melting point (°C): 210.5 (anhydrous)5; 97.0 (dihydrate)6

Heat of solution (kJ/mol): + 20.7 (trehalose dihydrate)

2.3. Other properties relevant to the use of trehalose as a

Trehalose in aqueous solution (at concentrations from about 10-34%

Trehalose (dihydrate) has a very low hygroscopicity, similar to

2.3.3. Chemical stability

Trehalose (dihydrate) is stable on storage at ambient temperature.

2.3.4. Thermal stability

Under the temperature conditions applied typically during the proc-

3.1. Novelty of the process

Trehalose may be obtained by extraction from yeast. Trehalose ob-

The subject of the present submission is a novel enzymatic process

Trehalose produced by this enzymatic process is chemically identi-

In a first step, starch is liquefied by treatment with a thermo-

The flow scheme of the trehalose production process is presented in

3.3.1. Starch liquefaction

[This chapter reports on details of the production process for

3.3.2. Saccharification, production of trehalose

[This chapter reports on details of the production process for

[This chapter reports on details of the production process for

3.3.4. Concentration and crystallization

[This chapter reports on details of the production process for

Food-grade starch is used as the starting material for the treha-

The chemicals used as processing aids in the manufacturing process

The thermophilic α-amylase (EC 3.2.1.1), which is used for starch

MTSase and MTHase have been purified and characterized. Their

Isoamylase (EC 3.2.1.68) is used in the trehalose production proc-

In anticipation of a direct food use of isoamylase which is unre-

The acute toxicity of P. amyloderamosa (wet cells), culture fil-

In a subchronic (90-day) toxicity study 4 groups of rats

On the basis of these results, the isoamylase preparation from

CGTase (EC 2.4.1.19) is used in the trehalose production to in-

3.5. Potential impurities resulting from the production process

1.5.1. Impurities from raw material, process chemicals and en-

The food-grade starch contains small amounts of protein and inor-

All applied chemicals are food-grade and/or have otherwise a high

3.5.2. By-products from enzymatic saccharification

The treatment of the enzymatic depolymerisation products of starch

3.5.3. Purification steps

The proteinaceous impurities from the raw-material (starch) and the

The ionic impurities are removed in the two-step demineralization

The glycosidic impurities and extractives of the ion-exchange res-

In the novel production process which is the subject of this appli-

5.1. Intended uses in food

Being about 40-45% as sweet as sucrose (Birch et al., 1970; Sugi-

Trehalose can be used to make fruit fillings and toppings, cream