INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
WORLD HEALTH ORGANIZATION
SAFETY EVALUATION OF CERTAIN
FOOD ADDITIVES
WHO FOOD ADDITIVES SERIES: 42
Prepared by the Fifty-first meeting of the Joint FAO/WHO
Expert Committee on Food Additives (JECFA)
World Health Organization, Geneva, 1999
IPCS - International Programme on Chemical Safety
SODIUM CARBOXYMETHYL CELLULOSE, ENZYMATICALLY HYDROLYSED
First draft prepared by
Dr J.B. Greig
Department of Health, Skipton House, London, United Kingdom
Explanation
Biological data
Biochemical aspects
Absorption, distribution, and excretion
Toxicological studies
Short-term studies of toxicity
Comments
Evaluation
References
1. EXPLANATION
Enzymatically hydrolysed sodium carboxymethyl cellulose is a
water-soluble dietary fibre with a relative molecular mass range of
800-10 000 Da, which is lower than that of sodium carboxymethyl
cellulose itself; for the same concentration, solutions of
enzymatically hydrolysed material have a lower viscosity than
carboxymethyl cellulose. It acts mainly as a stabilizer with
fat-extending properties. It can be used in formulating low-fat and
reduced-fat foods and soft drinks.
Enzymatically hydrolysed sodium carboxymethyl cellulose is
prepared from regular, food-grade carboxymethyl cellulose by partial
enzymic hydrolysis under mildly acidic conditions with a food-grade
cellulase enzyme from Trichoderma longibrachiatum. Carboxymethyl
cellulose itself was allocated an ADI 'not specified' at the
thirty-fifth meeting of the Committee (Annex 1, reference 88). The
cellulase enzyme was allocated an ADI 'not specified' at the
thirty-ninth meeting (Annex 1, reference 101). Enzymatically
hydrolysed sodium carboxymethyl cellulose has not been evaluated
previously by the Committee.
At the present meeting the Committee reviewed two studies in
which the properties of enzymatically hydrolysed sodium carboxymethyl
cellulose were compared with those of the parent material,
carboxymethyl cellulose.
2. BIOLOGICAL DATA
2.1 Biochemical aspects
2.1.1 Absorption, distribution, and excretion
The absorption and excretion of enzymatically hydrolysed sodium
carboxymethyl cellulose and carboxymethyl cellulose, each
radiolabelled with 14C in the carboxymethyl group, in rats were
compared over five days. The distribution of residual radioactivity in
the body was then determined. The radiolabelled materials were diluted
with 'Solka Floc' carboxymethyl cellulose or carboxymethyl cellulose
hydrolysate to provide materials with appropriate specific activity.
After a pilot study, two groups of four male and four female
conventional Wistar Wag/Rij rats were adapted for 14 days to diets
containing 5% of either material. The rats then received the
appropriate radiolabelled material at a dose of about 500 mg/kg bw,
equivalent to approximately 10 µCi per rat. Urine and faeces were
collected at intervals up to 120 h, and expired carbon dioxide was
collected at intervals up to 48 h. After 120 h, the rats were killed,
and 21 organs or tissues were collected, with the contents of the
gastrointestinal tract and the residual carcass; the blood was
separated into plasma and cells.
Little difference was seen between the two compounds in the
quantity of radiolabel in excreta: faecal excretion accounted for most
of the admini- stered dose (range of group means, 90-99%), with lesser
amounts in urine (1.3-2.0%) and expired air (0.6-0.9%). Peak
expiration of 14C-carbon dioxide occurred within the first 2 h after
dosing; the authors suggest that this was due to degradation of
low-molecular-mass compounds. Slightly more radiolabel was retained in
the carcasses of animals that received enzymatically hydrolysed
material than in those given carboxymethyl cellulose. A similar effect
was seen in individual organs, tissues, and body fluids, the largest
amounts being detected in fat, skin, muscle, liver, small intestine,
blood cells, and plasma. The tissue concentrations of both
enzymatically hydrolysed and parent carboxymethyl cellulose were
slightly higher in the liver and adrenals than in other tissues.
The relative molecular mass distribution of radiolabelled faecal
extracts was compared with that of the corresponding dosing solution
by gel permeation chromatography. The relative molecular masses of the
main peaks in the dosing solution differed from those in the faecal
extracts from animals treated with carboxymethyl cellulose but not for
animals treated with enzymatically hydrolysed material. The authors
concluded that the carboxymethyl cellulose was partially degraded
during passage through the gastrointestinal tract, resulting in the
formation of fragments similar in size to the preparation of
enzymatically hydrolysed sodium carboxymethyl cellulose. They were
unable to exclude the possibility that some labelled macromolecules
had been persorbed (van Ommen, 1993; Bär et al., 1995a).
2.2 Toxicological studies
2.2.1 Short-term studies of toxicity
Rats
Groups of 20 male and 20 female Crl:WI(WU)BR rats, about eight
weeks old, were fed either basal diet containing 10% added wheat
starch (controls) or diets containing 2.5, 5, or 10% enzymatically
hydrolysed sodium carboxymethyl cellulose or carboxymethyl cellulose
in place of the equivalent amount of wheat starch. The diets were
administered for 91-95 days to males and 98-102 days to females.
Ophthalmoscopic analyses were conducted in week 13. Haematological
examination was carried out on blood from 10 rats of each sex in each
group on day 85 (males) or 89 (females). Nineteen clinical chemical
parameters were analysed, including glucose levels on day 88 for males
and day 95 for females, while the other measurements were made at
necropsy.
One treated female rat died during the course of the experiment
from causes unassociated with administration of the diet. Rats fed 10%
of either material had diarrhoea but were otherwise healthy. At the
end of the experiment, the mean body weight of female rats fed the 10%
diets was significantly lower than that of the control group.
Dose-related increases in water intake were seen in weeks 4, 8, and
11. The food intake of animals at the higher doses tended to be
greater than that of rats at lower doses, although the conversion
efficiency of animals at 10% was only slightly lower. The mean food
intake of males and females given 10% carboxymethyl cellulose were 19
and 14 g per rat per day, equal to 6200 and 6800 mg/kg bw per day; the
corresponding figures for males and females given 10% enzymatically
hydrolysed material were 19 and 14 g per rat per day, equal to 5900
and 6600 mg/kg bw per day.
Ophthalmoscopic examination showed no treatment-related changes
in rats at the high doses. The only significant changes seen on
haematological examination were slight reductions in haemoglobin
concentration and haematocrit in females fed 10% enzymatically
hydrolysed sodium carboxymethyl cellulose. Clinical chemistry showed
some sporadic increases and a suggestion of a dose-related increase in
alkaline phosphatase and alanine aminotransferase activity with both
enzymatically hydrolysed and parent carboxymethyl cellulose, which was
more obvious in male rats. Urinary excretion of cations, particularly
sodium, and of citrate was altered in animals of each sex fed either
preparation and was associated with increased urine volume and pH and
with decreased urine density. Semiquantitative observations and
microscopic findings in the urine were unaffected in any manner
consistent with compound- or dose-related toxicity. The absolute and
relative weights of both the full and empty caecum, noted as caecal
enlargement at autopsy, were significantly increased in all treated
animals, with evidence of a dose-response relationship. The Committee
accepted that these changes are consistent with the incorporation in
the diet of an indigestible fibre which has a bulking effect (Bär,
1997; Newberne et al., 1988).
The observations in sections of the gastrointestinal tract taken
post mortem were consistent with the increased fluid intake of
animals at the high dose and were accompanied by histopathological
changes. The relative kidney weights were significantly higher in all
animals at the high dose, again with a clear dose-response
relationship. Statistically significant histopathological changes in
the kidney included papillomatous urothelial hyperplasia and pelvic
mineralization, generally graded only slight or very slight, in males
treated with the highest dose of enzymatically hydrolysed sodium
carboxymethyl cellulose and those given the lowest dose of
carboxymethyl cellulose. The incidence of simple and papillary
epithelial hyperplasia in the urinary bladder, generally only of
slight grade, was higher in males than in females and in animals
receiving the 10% dose of either material (Bär et al., 1995b; Til,
1992). These changes were attributed to an effect of the concentration
of sodium in the diet, which was about fourfold higher than in the
basal diet (Lalich et al., 1974; Bär, 1997).
3. COMMENTS
The Committee reviewed two studies in which the properties of
enzymatically hydrolysed sodium carboxymethyl cellulose were compared
with those of the parent material, carboxymethyl cellulose. The first
was a 90-day study of toxicity in which male and female rats received
diets containing 0, 2.5, 5, or 10% of either material. The second was
a comparative study of absorption, distribution, and excretion of
these two materials in rats.
Histopathological changes in the kidney and urinary bladder were
reported in the 90-day study in rats receiving either enzymatically
hydrolysed carboxymethyl cellulose or parent compound, which were
associated with changes in kidney weight, increased urine volume, and
increased excretion of some ionic species in the urine. Caecal
enlargement was also reported. For comparison, the Committee
considered the studies of the toxicity of carboxymethyl cellulose that
had been reviewed at its thirty-fifth meeting. It noted that, in early
studies that appeared to be comparable with regard to species, dietary
concentration of carboxymethyl cellulose, and duration of exposure, no
changes in organ weight or histopathological appearance were found
that were comparable to those reported from the 90-day study reviewed
at the present meeting.
Since the morphological changes in the kidney occurred in males
receiving the high dose of enzymatically hydrolysed carboxymethyl
cellulose and the low dose of carboxymethyl cellulose, the Committee
was not convinced that the response was of toxicological concern. The
observed epithelial hyperplasia of the urinary bladder was generally
graded slight or very slight and occurred in male rats fed a diet
containing 10% of either carboxymethyl cellulose. The Committee
concluded that this represents a response to a high intake of sodium
ions by rats of the strain used in the new study. Administration of
additional sodium (approximately four times that available in the diet
of the control group) could have led to the observed histopathological
changes and changes in urinary excretion. Also, the caecal
enlargement, diarrhoea, and other secondary changes were considered to
be the consequence of the accumulation of poorly absorbed,
water-soluble material in the caecum and colon and to be of no
toxicological significance. The NOEL for enzymatically hydrolysed
carboxymethyl cellulose was equal to 6000 mg/kg bw per day.
The metabolic study showed that passage of carboxymethyl
cellulose through the gastrointestinal tract results in partial
breakdown of the bonds between monomeric units, to result in a
material with a relative molecular mass equivalent to that of
enzymatically hydrolysed carboxymethyl cellulose. The amounts of
radiolabel excreted in the faeces, urine, and breath after
administration of radiolabelled versions of the two products were
quantitatively almost identical, providing additional evidence that
the end-products of digestion of carboxymethyl cellulose and
enzymatically hydrolysed carboxymethyl cellulose are similar in rats.
4. EVALUATION
The Committee concluded that these similarities are consistent
with no toxicologically significant difference between carboxymethyl
cellulose and enzymatically hydrolysed carboxymethyl cellulose.
Therefore, the Committee included the latter in the group ADI 'not
specified' with ethyl cellulose, ethyl hydroxyethyl cellulose,
hydroxypropyl cellulose, hydroxypropyl methyl cellulose, methyl
cellulose, methyl ethyl cellulose, and sodium carboxymethyl cellulose.
5. REFERENCES
Bär, A., van Ommen, B. & Timonen, M. (1995a) Metabolic disposition in
rats of regular and enzymatically depolymerized sodium
carboxymethylcellulose. Food Chem. Toxicol., 33, 901-907.
Bär, A., Til, H.P. & Timonen, M. (1995b) Subchronic oral toxicity
study with regular and enzymatically depolymerized sodium
carboxymethyl cellulose in rats. Food Chem. Toxicol., 33, 909-917.
Bär, A. (1997) Sodium carboxymethyl cellulose, enzymatically
hydrolyzed (CMC-ENZ). Unpublished dossier for the safety assessment of
CMC-ENZ from Bioresco Ltd, Binningen, Switzerland. Submitted to WHO by
Bioresco Ltd, Binningen, Switzerland.
Lalich, J.J., Paik, W.C.W. & Pradhan, B. (1974) Epithelial hyperplasia
in the renal papilla of rats: Induction in animals fed excess sodium
chloride. Arch. Pathol., 97, 29-32.
Newberne, P.M., Conner, M.W. & Estes, P. (1988) The influence of food
additives and related materials on lower bowel structure and function.
Toxicol. Pathol., 16, 184-197.
van Ommen, B. (1993) A comparative disposition study with
[14C]-CMC-hydrolysate and [14C]-CMC in rats. Unpublished report No. V
92.473 from TNO Nutrition and Food Research, Zeist, Netherlands.
Submitted to WHO by Bioresco Ltd, Binningen, Switzerland.
Til, H.P. (1992) Comparative sub-chronic (3-month) feeding study with
sodium carboxymethyl cellulose hydrolysate and sodium carboxymethyl
cellulose in rats. Unpublished report No. V 92.015 from TNO Nutrition
and Food Research, Zeist, Netherlands. Submitted to WHO by Bioresco
Ltd, Binningen, Switzerland.