Toxicological evaluation of some food
additives including anticaking agents,
antimicrobials, antioxidants, emulsifiers
and thickening agents
WHO FOOD ADDITIVES SERIES NO. 5
The evaluations contained in this publication
were prepared by the Joint FAO/WHO Expert
Committee on Food Additives which met in Geneva,
25 June - 4 July 19731
World Health Organization
Geneva
1974
1 Seventeenth Report of the Joint FAO/WHO Expert Committee on
Food Additives, Wld Hlth Org. techn. Rep. Ser., 1974, No. 539;
FAO Nutrition Meetings Report Series, 1974, No. 53.
POLYOXYETHYLENE (20) SORBITAN MONOESTERS OF LAURIC,
OLEIC, PALMITIC AND STEARIC ACID AND TRIESTER OF
STEARIC ACID
Explanation
These compounds have been evaluated for acceptable daily intake
by the Joint FAO/WHO Expert Committee on Food Additives (see Annex 1,
Ref. No. 7) in 1963.
Since the previous evaluation, additional data have become
available and are summarized and discussed in the following monograph.
The previously published monograph has been expanded and is reproduced
in its entirety below.
BIOLOGICAL DATA
BIOCHEMICAL ASPECTS
The coefficients of digestibility of the fatty acid moieties of
polyoxyethylene (20) sorbitan monooleate, monostearate and tristearate
were found to be 100%, 98% and 84%, respectively, when fed at
concentrations of 10% in the diet (Oser & Oser, 1957a).
Studies with 14C-labelled polyoxyethylene sorbitan monostearate
in rats showed that 6-10% was excreted in the urine, 2-7% was
recovered in the expired air, and the remainder was excreted in the
faeces. Labelling with 14C was confined to the sorbitol part of the
molecule (Wick & Joseph, 1956).
The amounts of 14C-containing compounds found in the urine and
expired air were considerably smaller than in similar studies with
14C-labelled sorbitan monostearate (Wick & Joseph, 1953).
Studies in six human subjects with polyoxyethylene (20) sorbitan
monooleate showed that the distribution of the polyoxyethylene moiety
was 3.9-5.8% in the urine and 90.4-98.3% in the stools; total recovery
of the polyoxyethylene in these investigations was 96.2-102.2% (Culver
et al., 1951).
Polyoxyethylene (20) sorbitan monooleate ("Tween 80") fed to rats
for one week in amounts of 0.1% and 1% of the diet augments the
absorption of fats when these are present in preparations of 10% to
33% of the diet, but not when fats comprise less than 7% of the diet
(Sergiel et al., 1971).
Polysorbate (80) and polysorbate (20) [= polyoxyethylene (20)
sorbitan monooleate and polyoxyethylene (20) sorbitan monolaurate] in
concentrations of 0.01% or less increase the absorption of lipid
soluble drugs (4-amino-antipyrine = ampyrone; secobarbitone),
indicating that these surfactants affect the permeability of
biological membranes (Levy & Anello, 1969; Anello & Levy, 1969).
However, administration of 0.03% of polysorbate 80 to rats
together with drugs (4-aminoantipyrine, salicylic acid, salicylamide)
did not affect their absorption from the small intestine (Levy &
Perälä, 1970).
Metabolic studies were carried out in rats with polyoxyethylene
(20) sorbitan monolaurate with 14C-label in the lauric acid moiety or
the polyoxyethylene moiety. Doses administered were 1 g/kg orally and
0.5 g/kg i.v. With the label in the fatty acid, the approximate
proportions of radioactivity 24 hours after oral administration were:
liver, 1.2%;urine, 2.5%; faeces and gastrointestinal contents, 4%;
carcass, 12%; expired CO2, 80%; after i.v. injection, they were
liver, 0.7%; urine. 5%; faeces and gastrointestinal contents, 2.5%;
carcass, 22%; expired CO2, 68%. With the label in the polyoxyethylene
moiety, the distribution of radioactivity 24 hours after oral
administration was different: liver, nil; urine, 8%; faeces, 90%;
carcass, nil; expired CO2, nil. After i.v. administration the
distribution was: liver 0.15%; urine, 83%; faeces, 11%; carcass, 2%;
expired CO2, nil. The findings indicate the fatty acid ester bond is
hydrolyzed and the fatty acid is utilized metabolically; however, the
polyoxyethylene moiety is poorly absorbed and is not subject to
metabolism (Nelson et al., 1966).
Metabolic studies were carried out in rats after oral
administration of polyoxyethylene (20) sorbitan monolaurate ("Tween
20") prepared with 14C-labelling of the polyoxyethylene or the lauryl
moieties. The 14C-lauryl moiety was rapidly absorbed and oxidized,
more so in fasted than non-fasted rats. After 24 hours, some 75% of
the administered radioactivity was recovered in expired CO2, and 4%
was not absorbed. The 14C-polyoxyethylene moiety was poorly absorbed,
87% of radioactivity remaining unabsorbed, and about 8.5% was
recovered in the urine. After i.v. injection of the substance, the
14C-lauryl moiety was metabolized as after oral administration.
The 14C-polyoxyethylene moiety was not metabolized, since no
radioactivity was recovered as 14C-CO2: most appeared in the urine,
but some was present in the faeces indicating biliary excretion.
Polyoxyethylene (20) sorbitan monooleate ("Tween 80") with a
14C-sorbital label was also administered to rats. After oral
administration, 91% of the radioactivity was recovered in the faeces,
2.1% in the urine, 1.6% in the carcass and none in expired CO2,
liver, kidney, spleen, adrenals, brain, gonads or fat (Treon et al.,
1967).
Polyoxyethylene (20) sorbitan monooleate ("Tween 80") in
concentrations of 0.2 and 1 mM did not affect bile salt transport in
isolated preparations of rat intestine; 4 mM slightly inhibited
transport (Holt, 1964).
Polyoxyethylene (20) sorbitan monooleate (polysorbate 80) in a
concentration of 0.5% decreased the rates of absorption from rat small
intestine perfused in situ of various drugs (sulfamethoxypyridazine,
diphenhydramine, salicylic acid, p-hydroxybenzoic acid) (Aoki et al.,
1969).
The absorption of aspirin from the gut into the blood in rats was
facilitated by simultaneous administration of polyoxyethylene (20)
sorbitan monooleate in a dose of 50 mg/kg, but not by 10 mg/kg (Cid et
al., 1971).
TOXICOLOGICAL STUDIES
Acute toxicity
The acute oral toxicity in the mouse, rat and hamster was shown
to be extremely low (Eagle & Poling, 1956; Hopper et al., 1949).
Polyoxyethylene (20) sorbitan monolaurate has an LD50 of
3.75 g/kg bw when injected i.v. into mice (Hopper al., 1949).
The accidental administration of a dose of 19.2 g of
polyoxyethylene (20) sorbitan monooleate per kg bw on two successive
days to an infant was followed by no ill effects, apart from purgation
(Chusid & Diamond, 1955).
Rat
"Tween 20" = polyoryethylene (20) sorbitan monolaurate:
Oral LD50 > 39 000 mg/kg
i.v. LD50 1 450 mg/kg
(male, 1680; female, 860)
"Tween 40" = polyoxyethylene (20) sorbitan monopalmitate:
Oral LD50 > 38 400 mg/kg
i.v. LD50 1 580 mg/kg
(male, 2320; female; 1350)
"Tween 60" = polyoxyethylene (20) sorbitan monostearate:
Oral LD50 > 38 000 mg/kg
i.v. LD50 1 220 mg/kg
(male, 1750; female, 1180)
"Tween 65" = polyoxyethylene (20) sorbitan tristearate:
Oral LD50 > 40 000 mg/kg
"Tween 80" = polyoxyethylene (20) sorbitan monooleate:
Oral LD50 > 38 000 mg/kg
i.v. LD50 1 790 mg/kg
(male, 2180; female, 1720)
(see Brandner, 1973)
A range of polyethylene sorbitan esters of fatty acids ("Tween
20, 21, 40, 60, 61, 65, 80, 81, 85") failed to produce reactions on
patch testing of each in 50 subjects. On instillation in the rabbit
conjunctival sac, most failed to produce a reaction even when the eye
was not subsequently washed and did not produce reaction when the eye
was washed (Treon et al., 1967).
Short-term studies
Mouse
Groups of 10 to 12 mice, studied over three to four months were
found to tolerate 2.5%, 5% and 10% polyoxyethylene (20) sorbitan
monostearate without showing any ill effects, but at the 15% level
there was some gastrointestinal disturbance, with reduced food intake
and some growth retardation (Brush et al., 1957).
Rat
Groups of young rats fed 3% and 5% of polyoxyethylene (20)
sorbitan monolaurate for eight weeks showed a significantly slower
weight gain as compared with controls; this was attributed to mild
diarrhoea. However, there were no apparent gross abnormalities nor any
significant histopathological findings in the treated animals (Krantz,
1943a).
When groups of 30 rats were fed 2% of polyoxyethylene (20)
sorbitan monooleate in their diet for three generations, no evidence
of alteration in their fecundity and growth pattern was found, nor
were there any histological findings in the livers and kidneys
(Krantz, 1946).
No toxic symptoms were found when rats were fed for eight weeks
on 2% and 5% polyoxyethylene (20) sorbitan monostearate in their daily
diet (Krantz, 1943b).
A 15-week feeding study using 25% polyoxyethylene (20) sorbitan
monostearate in the daily diet of male rats resulted in retardation of
growth as compared with control animals on the same basic diet. The
rats receiving the diet containing polyoxyethylene (20) sorbitan
monostearate exhibited transient diarrhoea, but the haematological
findings were normal and no abnormalities were revealed on gross
pathological and histological examination of the important viscera
(Krantz, 1949).
Polyoxyethylene (20) sorbitan monostearate was fed for 14 weeks
to groups of 12 male and 12 female weanling rats as 5% and 15% of a
soybean meal diet. There were no clinical manifestations of toxicity
throughout the experimental period, and the histopathological
examination of the animals at the end of the experiments showed no
abnormalities. Neither were deleterious effects observed when 5%
polyoxyethylene (20) sorbitan monostearate in the soybean basal diet
was fed to older rats for a period of 14 weeks. On the other hand,
polyoxyethylene (20) sorbitan monostearate fed to weanling rats as 5%
of a basal purified casein diet caused diarrhoea and retardation of
growth (Chow et al., 1951; Chow et al., 1953).
Hamster
Weanling hamsters (12 per group) were fed 1% and 5% of
polyoxyethylene (20) sorbitan monostearate in the diet for one year.
For the first six weeks growth and food efficiency were the same as in
controls. During the 12-month feeding period, the mortality and body
weights of the treated animals were not significantly different from
those of the controls, but chronic diarrhoea was exhibited at the 5%
level. When the animals were sacrificed at the end of the experiment,
no significant differences were noted between the treated animals and
the controls in regard to organ weights and pathological findings,
except that in the kidneys of those at the 5% level, there was a
significantly greater frequency of hyaline casts and chronic
interstitial nephritis, believed to be associated with the water
imbalance brought about by the chronic diarrhoea (Brush et al., 1957).
In another experiment with hamsters (10 in each group), which had
been fed 5%, 10% and 15% levels of polyoxyethylene (20) sorbitan
monolaurate, fatalities and chronic diarrhoea were unusually frequent,
and when the remaining animals were sacrificed at 28 to 39 weeks
highly significant changes were found in a number of organs (Eagle &
Poling, 1956).
A high mortality was also found in hamsters (10 per group) fed
diets containing 10% and 15% of polyoxyethylene (20) sorbitan
monolaurate. With 5% in the diet, growth was significantly retarded
and diarrhoea occurred (Poling et al., 1956).
Similar results were obtained when groups of 36 hamsters were fed
diets containing 5% and 15% of polyoxyethylene (20) sorbitan
monolaurate for 68 days (Harris et al., 1951).
Dog
Beagle puppies were fed for one year on diets containing 5% and
10% of polyoxyethylene (20) sorbitan monostearate; no abnormalities
were observed (Brush et al., 1957).
Monkey
When six monkeys were fed polyoxyethylene (20) sorbitan
monolaurate or monooleate at the rate of 1 g a day for periods of up
to 17 months there were no significant changes in weight gain or
histopathology (Krantz, 1943a, 1947a).
Chick
Polyoxyethylene (20) sorbitan monostearate and monolaurate fed to
groups of 12 chicks at levels of 0.1%, 1% and 2% of the diet for seven
weeks did not cause any adverse effects (Ringrose & Waller, 1959).
Long-term studies
Rat
Long-term feeding studies have been carried out on groups of
15-30 rats with polyoxyethylene (20) sorbitan monolaurate (Oser &
Oser, 1957b), at levels of 0.5% and 2% in the diet and with
polyoxyethylene (20) sorbitan monopalmitate (Wick & Joseph, 1956),
monostearate (Wick & Joseph, 1953), tristearate (Culver et al., 1951),
and monooleate (Eagle & Poling, 1956) at a level of 2%.
As compared with control animals, no abnormalities were found
that could be attributed to the experimental diets. In addition to
growth measurements, mortality statistics and studies of blood
chemistry, gross and histological examinations were made of the
following organs: brain, spleen, pancreas, thyroid, parathyroid,
prostate, pituitary, salivary and adrenal glands, bladder, liver,
kidney, bone marrow, heart, lung, testis, lymph nodes and muscle (Oser
& Oser, 1957a; Wick & Joseph, 1956; Wick & Joseph, 1953; Culver et
al., 1951; Eagle & Poling, 1956; Krantz, 1943c, 1947b, 1947c, 1947d,
1947e).
Four groups of rats, each containing 12 males and 12 females,
were fed for the whole life-span with diets containing 2%, 5%, 10% and
25% of polyoxyethylene (20) sorbitan monostearate. This study revealed
no effect of the ester at the 2% and 5% levels. However, the compound
produced marked diarrhoea at the 10% and 25% levels, as well as
enlargement of the caecum (slight to moderate at the 25% level, less
at the 10% level), and a questionable fatty change of a very slight
degree in the livers of the rats fed the 25% level (Fitzhugh et al.,
1959).
In another experiment, the effects of feeding polyoxyethylene
(20) sorbitan monostearate, tristearate and monooleate to groups of
12 males and 20 females over the whole life-span at dosage levels of
5%, 10% and 20% of the diet were studied. Observations were also made
on three successive generations. This extensive study included tests
on gestation and fertility, mortality, blood and urine constituents,
and histopathology. No abnormalities were found at the 5% level. At
the 10% and 20% levels, many of the animals, particularly the males,
had diarrhoea. The 20% level had some adverse effect on postnatal
survival, lactation efficiency, and duration of breeding activity.
Growth rate in the males and caloric utilization efficiencies were
slightly diminished (Oser & Oser, 1957a; Oser & Oser, 1956a, 1956b,
1957b).
Two colours which produce tumours when given subcutaneously,
Brilliant Blue FCF and Solid Green FCF (? le Vert solide FCF?) were
given orally in a concentration of 1% together with a surface active
agent, polyoxyethylene (20) sorbitan monostearate ("Tween 60") or
polyoxyethylene (20) sorbitan monooleate ("Tween 80"), in a
concentration of 5% to rats for two years. The presence of the
surfactants did not affect the incidence of tumours. In another
experiment, it was shown that they did not affect the absorption of
the dyes (Truhaut, 1970).
OBSERVATIONS IN MAN
4.5-6 g of polyoxyethylene (20) sorbitan monooleate were taken
daily by 100 adults - 10 for three to four years, 17 for two to three
years, 19 for one to two years, 54 for less than one year. No
deleterious effects could be demonstrated (Krantz, 1951).
These results have been confirmed by other authors (Steigmann,
1953; Waldstein et al., 1954).
Comments:
Some of the early short-term studies with these polyoxyethylene
sorbitan esters in rats and hamsters showed deleterious effects.
Subsequent work suggests that these were largely due to diarrhoea
resulting from a large amount of unabsorbed polyglycol, possibly
aggravated in some experiments by the use of an unsuitable basal diet.
Since that time there has been considerable improvement in testing
procedures, and more extensive long-term studies have been carried
out. It seems reasonable therefore to base the evaluation of these
substances on the levels causing no adverse effects indicated by the
results of the more recent investigations.
The significance of the local tumours which were produced by
injection has been discussed at the meeting of the Scientific Group
(1966). No increase in tumour incidence has followed the oral intake
of polyoxyethylene sorbitan esters. Furthermore, large doses of the
oleate and stearate have been well tolerated by human subjects.
EVALUATION
Level causing no toxicological effect
Rat: 50 000 ppm (5%) in the diet equivalent to 2500 mg/kg bw.
Estimate of acceptable daily intake for man
0-25 mg/kg bw.*
REFERENCES
Anello, J. A. & Levy, G. (1969) J. Pharm. Sci., 58, 721
Aoki, M. et al. (1969) Chem. Pharm. Bull., 17, 1109
Brandner, J. D. (1973) Unpublished report submitted by ICI America
Inc.
Brush, M. K. et al. (1957) J. Nutr., 62, 601
Chow, B. F. et al. (1951) Fed. Proc., 10, 378
Chow, B. F. et al. (1953) J. Nutr., 49, 563
Chusid, E. & Diamond, J. (1955) J. Pediat., 46, 222
Cid, E., Dresse, A. & Jaminet, Fr. (1971) Pharm. Acta Helv., 46, 377
Culver, P. J. et al. (1951) J. Pharmacol. exp. Ther., 103, 377
Eagle, E. & Poling, C. E. (1956) Food Res., 21, 348
Fitzhugh, O. G. et al. (1959) Toxicol. appl. Pharmacol., 1, 315
Harris, R. S., Sherman, H. & Jeter, W. W. (1951) Arch. Biochem., 34,
259
Holt, P. R. (1964) Proc. Soc. exp. Biol. Med., 117, 230
Hopper, S. H., Hulpieu, H. R. & Cole, V. V. (1949) J. Amer. pharm.
Ass., Sci. Ed., 38, 428
Jaminet, Fr. (1971) Pharm. Acta Helv., 46, 377
* As total polyoxyethylene (20) sorbitan esters.
Krantz J. C. jr (1943a) Unpublished report No. WER-124-88 to the Atlas
Chemical Co.
Krantz J. C. jr (1943b) Unpublished report No. WER-149-76 to the Atlas
Chemical Co.
Krantz J. C. jr (1943c) Unpublished report No. WER-149-76/160/188/A to
the Atlas Chemical Co.
Krantz J. C. jr (1946) Unpublished reports Nos. WER-149-130/123/A/B to
the Atlas Chemical Co.
Krantz J. C. jr (1947a) Unpublished report No. WER-149-123 to the
Atlas Chemical Co.
Krantz J. C. jr (1947b) Unpublished reports Nos. WEE-149-164/193/235/
A/B to the Atlas Chemical Co.
Krantz, J. C. jr (1947e) Unpublished reports Nos. WER-149-165/195/
237/A/B to the Atlas Chemical Co.
Krantz, J. C. jr (1947d) Unpublished reports Nos. WER-149-175/198/
234/A/B to the Atlas Chemical Co.
Krantz, J. C. jr (1947e) Unpublished reports Nos. WER-149-133/161/
189/228 to the Atlas Chemical Co.
Krantz, J. C. jr (1949) Unpublished report No. WER-149-270/A/B to the
Atlas Chemical Co.
Krantz, J. C. jr et al. (1951) Bull. Sch. Med. Maryland, 36, 48
Levy, G. & Anello, J. A. (1969) J. Pharm. Sci., 58, 494
Levy, G. & Perälä, A. (1970) J. Pharm. Sci., 59, 874
Nelson, M. F. et al. (1966) J. Food Sci., 31, 253
Oser, B. L. & Oser, M. (1956a) J. Nutr., 60, 367
Oser, B. L. & Oser, M. (1956b) J. Nutr., 60, 489
Oser, B. L. & Oser, M. (1957a) J. Nutr., 61, 149
Oser, B. L. & Oser, M. (1957b) J. Nutr., 61, 235
Poling, C. E., Eagle, E. & Rice, E. E. (1956) Food Res., 21, 337
Ringrose, A. T. & Waller, E. F. (1959) Toxicol. appl. Pharmacol., 1,
548
Sergiel, J.-P., Paris, R. & Clement, J. (1971) Cah. Nut. Diet, 5, 33
Steigmann, F., Goldberg, E. M. & Schoolman, H. M. (1953) Amer. J. dig.
Dis., 20, 380
Treon, J. F. et al. (1967) Page 381 in Application of surface active
Substances, Vol. III of Chemistry, Physics and Application of
Surface Active Substances, Gordon & Breach, New York
Truhaut, R. (1970) Bull. Acad. Nat. Med., 154, 789
Waldstein, S. S., Schoolman, H. M. & Popper, H. (1954) Amer. J. dig.
Dis., 21, 181
Wick, A. N. & Joseph, L. (1953) Food Res., 18, 79
Wick, A. N. & Joseph, L. (1956) Food Res., 21, 250