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.
CUPRIC SULFATE
Explanation
This compound has been evaluated for acceptable daily intake by
the Joint FAO/WHO Expert Committee on Food Additives (see Annex 1,
Ref. No. 22) in 1970.
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
Copper is an essential trace element and is a constituent of
plants and of animal and human tissues. The tissues containing the
largest concentrations are liver with 0.30-0.91 mg/100 g and brain
with 0.22-0.68 mg/100 g (Kehoe et al., 1940). The whole human body
contains 100-150 mg (Browning, 1969). At subcellular level a number of
enzymes, such as tyrosinase, contain Cu as part of their structure or
require it for proper functioning, e.g. catalase (Dawson & Mallette,
1945).
About 3.2 mg Cu is consumed daily in food (mainly in meat, eggs,
oils etc., oysters having the highest concentration with 27.4 mg/100
cals). Water provides 40-500 pg. The total daily intake in soft water
areas is calculated as: food 3200 pg, water 200 pg, beverages 300 pg,
air 2 pg. Excretion is calculated as: urine 60 pg, faeces 3640 pg,
sweat 2 pg/day (Schroeder et al., 1966).
Somewhat controversial evidence suggests that the metal is an
essential co-factor in haemoglobin synthesis and is involved in Fe
metabolism. Some animal diseases, especially severe anaemias, are
suspected to arise from nutritional copper deficiency. Copper
intoxication may cause acute haemolysis in sheep (Anon., 1966). In man
the average daily requirement for adults is estimated at 2 mg, and for
infants and children at 0.05 mg/kg bw (Fd. Std. Cttee, 1956; Browning,
1969). The copper content of various foods ranges from 20 to 400 ppm
(0.002% to 0.04%) (Underwood, 1962). The average daily dietary intake
for adults is estimated at 2 to 5 mg, of which up to 0.7 mg are
excreted in the urine (Browning, 1969). 0.8 mg are retained mainly in
the liver, kidney and intestine, while 1.40 mg are excreted in the
faeces. Increased intake appears to have little effect on urinary
output but faecal excretion may rise to 10 to 20 times the urinary
excretion. Absorption from the g.i. tracts is limited. Normal human
serum levels range from 68 to 90 mg/ml of which 95% is carried by the
alpha-globulin copper oxidase ceruloplasmin. The remainder is bound to
albumin or amino acids. In vitro studies on liver and kidney slices
using 64Cu-acetate demonstrated intra-cellular transport by histidine
and other amino acids (Neumann & Silverberg, 1966).
Rats fed 2.5 mg/day copper sulfate and sacrificed 1, 3, 6 and 24
hours later showed significant concentrations of Cu in kidneys, liver
and plasma (up to 2.7 pg/g in kidney and 1.1 pg/g in liver) (Decker et
al., 1972). The copper is attached to hepatic mitochondria and cell
nuclei, more being found in the nuclei at concentrations above
100 pg/g (Lal & Sourkes, 1971).
Copper and molybdenum levels become most critical when one or the
other is present in either deficient or toxic amounts. The level at
which molybdenum becomes toxic depends on the amount of copper in the
diet, and an excess of molybdenum can induce or intensify a deficiency
of copper. In addition, sulfate ion can act either to modify or
intensify the adverse effects of molybdenum. A similar but reverse
pattern occurs when molybdenum is deficient and copper is in excess
(Underwood, 1962; Gray & Daniel, 1964).
Continued intake of high levels of copper in experimental animals
leads to considerable accumulation in the liver. In the pig and the
rat this has resulted in lowered iron levels in haemoglobin and liver
and haemolytic jaundice in some stressed animals. Long-term
administration of even low concentrations of copper results in some
increased storage in the liver (O'Hara et al., 1960; Buntain, 1961;
Bunch et al., 1965; Harrison et al., 1954).
Effect on ascorbic acid availability was tested by giving guinea-
pigs copper sulfate or copper gluconate in drinking-water at levels
equivalent to 1600 ppm Cu (0.16% Cu) of the diet for 11 weeks. Animals
were sacrificed and examined grossly for scurvy and serum ascorbic
acid. No evidence of scurvy was found and serum levels of ascorbic
acid were not affected (Harrison et al., 1954).
TOXICOLOGICAL STUDIES
Acute toxicity
Substance Animal Route LD50 LD100 Reference
(mg/kg bw) (mg/kg bw)
Copper
chloride Rat Oral 140 Spector, 1956
Guinea-pig s.c. 100 Spector, 1956
Copper
nitrate Rat Oral 940 Spector, 1956
Copper
sulfate Mouse i.v. 50 Spector, 1956
Rat Oral 300 Spector, 1956
Guinea-pig i.v. 2 Spector, 1956
Rabbit i.v. 4-5 Spector, 1956
In animals ingestion of three ounces of 1% CuSO4 solution
produces intense g.i. tract inflammation (Browning, 1969). In mammals
injection or inhalation of copper and its compounds leads to
haemochromatosis, liver injury or lung injury (Browning, 1969).
Short-term studies
Rat
Young rats (100-150 g) were injected daily with CuCl2 solutions
at 0, 1, 2.5 and 4 mg/kg for 236 days. Controls showed no lesions.
Weight loss was evident in all treated groups and deaths occurred at
the two higher levels. Liver pathology showed necrotic cells in the
periphery of lobules with inflammation and regeneration, periportal
fibrosis, and nuclear hyperchromatism with large hyalinized cells.
Kidney lesions described were sloughing and degeneration of epithelial
cells of proximal convoluted tubules (Wolff, 1960).
Young (21-days old) albino rats were fed ad libitum for four
weeks on diets containing copper sulfate to give 0, 500, 1000, 2000
and 4000 ppm (0%, 0.05%, 0.1%, 0.2% and 0.4%) of added copper. The
daily food intake was less, the higher the copper content, the average
copper intakes being about 5, 8, 11 and 8 mg/rat/day respectively. All
the rats on the highest dose died in the first week; one out of eight
in the second highest dosage group died in the fourth week. It was
suggested that the deaths in the highest dosage group were due partly
to reduced food intake. The growth rate in the lowest dosage group was
slightly decreased, otherwise the rats appeared normal. There were
slight increases in the copper contents of blood and spleen and a
marked (14-fold) increase in copper content of the liver (Boyden et
al., 1938).
Copper sulfate at 0.135% and 0.406% (equivalent to 530 ppm and
1600 ppm copper, respectively) and copper gluconate at 1.14%
(equivalent to 1600 ppm Cu) were fed in the diet of rats for up to
44 weeks. A negative control group was also maintained. Each group
comprised around 25 male and 25 female rats.
Significant growth retardation, discernible at the twenty-sixth
week, occurred with the high level copper sulfate and the copper
gluconate. Mortality was increased in the high level copper sulfate
group and greatly increased (90% dead between four and eight months)
in the copper gluconate group. Four high level copper sulfate, copper
gluconate, and control rats were sacrificed between 30 to 35 weeks and
all survivors were sacrificed between 40 to 44 weeks. Haematology and
urine examinations were within normal limits except for high (83 mg%)
blood nonprotein nitrogen (NPN) in males ingesting the high level
copper sulfate and copper gluconate; the lower level copper sulfate
was just above the expected range of 60-70 mg% NPN). Serum levels of
ascorbic acid were not affected. Animals receiving copper gluconate
had hypertrophied uteri, ovaries and seminal vesicles. High level
copper sulfate and copper gluconate animals showed enlarged, distended
and hypertrophied stomachs, occasional ulcers, some blood, bloody
mucous in intestinal tract, and bronzed kidneys and livers.
Histopathology of the higher test level animals showed toxic
abnormalities in the liver and minor changes in the kidneys. Varying
degrees of testicular damage were noted in both high and low levels of
copper sulfate animals whereas control animals were normal. Liver,
kidney and spleen tissue-stored copper was elevated in all test
groups, liver being most pronounced. Liver-copper levels recorded per
100 g wet tissue at 40 weeks were: <2 mg (controls), 12-32 mg
(low copper sulfate), 38-46 mg (high copper sulfate) and at 30 weeks
56-75 mg (copper gluconate). Also noted was a marked depression in
tissue storage of iron in high level copper sulfate and copper
gluconate animals.
In conclusion, copper sulfate and copper gluconate at 1600 ppm
copper were toxic while copper sulfate at 530 ppm copper caused only
variable effects on testicular degeneration and tissue storage of
copper (Harrison et al., 1954).
Daily s.c. injection of 0.26 mg Cu for 80 days produced elevated
erythrocyte and plasma copper levels and raised caeruloplasmin levels
after a total dose of 3.64 mg Cu. The rises levelled out at 15.6 mg Cu
total though tissue levels continued to rise. Anaemia and diarrhoea
developed and mean survival was 67 days. Histology showed liver and
kidney damage and enlarged caeca. Survivors were mated and offsprings
were given 0.26 mg Cu daily for four weeks, then 0.65 mg/day for 8.5
months. Sixteen of the 37 offspring survived (Weedwanders et al.,
1968).
Rabbit
Copper acetate at 2 mg/g (2000 ppm (0.2%)) of diet fed to 21
rabbits through days 21 to 105 showed pigmentation in 17, cirrhosis in
9 and necrosis of the liver in 5; those with cirrhosis did not show
necrosis. Copper in the liver varied from 9.7-237 mg/100 g of wet
liver. A relationship was established in which a longer feeding period
resulted in a greater incidence of cirrhosis in the liver (Wolff,
1960).
Pig
Three-week-old pigs fed 250 (0.025%), 600 (0.06%) or 750 (0.075%)
ppm Cu in a fish meal diet showed depressed weight gain and feed
consumption while the same concentration of copper in soybean meal had
no effect. No gross pathological changes were seen in either group
(Clyde et al., 1969).
Sheep
Six out of 17 lambs fed from six to 12 weeks of age on a ration
containing 80 ppm (0.008%) copper developed spongy transformation of
the CNS white matter particularly in the region of the mid-brain, pons
and cerebellum with severe lesions in the superior cerebellar pedicles
(Doherty et al., 1969). Copper toxicity was found in three out of 170
housed lambs fed on a diet containing 20 ppm (0.002%) copper and 1 ppm
(0.0001%) molybdenum. The dead animals were well nourished but
jaundiced, with swollen, friable liver, metallic black kidneys and
myocardial haemorrhage. Some intravascular haemolysis was seen in one
lamb (Adamson et al., 1969). Sheep are highly susceptible to copper
poisoning and with over-dosage the liver may contain up to 1000 ppm
(0.1%) (Bull, 1949).
Turkey
Turkey poults can tolerate 676 ppm Cu (0.0676%) in the diet
without ill effect but growth was suppressed by over 810 ppm (0.081%).
These effects were counteracted by EDTA (Ouhra & Kratzer, 1968).
Long-term Studies
None available.
OBSERVATIONS IN MAN
Copper poisoning, with diarrhoea and vomiting, developed when
20 workmen drank tea containing 25 ppm (0.0025%) copper (Nicholas,
1968). Rashes were reported after drinking water containing 7.6 ppm
(0.00076%) copper (Paine, 1968) whilst jaundice and severe haemolytic
anaemia with elevations of serum SGOT, copper and caeruloplasmin were
seen in a child following repeated applications of copper sulfate to
extensive areas of severely burnt skin (Holtzman et al., 1966).
Mineral abnormalities occur in patients undergoing haemodialysis
when Cu levels may be raised (Mahler et al., 1971). With prolonged
i.v. infusions copper deficiency may occur (James & MacMahon, 1970).
Fatal oral human doses: Basic copper sulfate 200 mg/kg bw
Copper chloride 200 mg/kg bw
Copper carbonate 200 mg/kg bw
Copper hydroxide 200 mg/kg bw
Copper oxychloride 200 mg/kg bw
Large doses cause severe mucosal irritation and corrosion,
widespread capillary damage, hepatic and renal damage, CNS irritation
and depression. Sulphaemoglobinaemia and haemolytic anaemia have been
seen. The acetate and sulfate are very toxic especially the cupric
salts while cuprous chloride is the most toxic. Local skin corrosion
with eczema and eye inflammation occur. Copper sulfate has been used
in suicide attempts. Rapid transfer of absorbed Cu to red cells causes
haemolysis. Hepatic necrosis and renal tubular oedema with necrosis
are seen (Chuttani et al., 1965; Browning, 1969). Occupational copper
poisoning causes greenish hair and urine in coppersmiths and copper
colic. Inhalation of dust or vapour causes copper fume fever - brass
chills (Bur. Mines, 1953). Contact of food or soft acid water with
copper utensils may cause poisoning, but no haemochromatosis or liver
disease (Bur. Mines, 1953; Hueper, 1965; Browning, 1969). The
existence of chronic copper poisoning in man whether industrial or
non-industrial is debatable (Browning, 1969).
Newborn premature infants of about 1.2 kg bw were fed a milk diet
providing an average of 14 µg copper per kg per day (seven subjects)
or diet with a supplement providing an average of 173 µg copper per kg
per day (five subjects). The duration of the period of observation was
seven to 15 weeks. There were no differences in growth rate,
haemoglobin, serum protein or serum copper between the two groups
(Wilson & Lahey, 1960).
EVALUATION
There are no animal studies providing a no-effect level. However,
this does not preclude the evaluation of this essential trace element.
Reliance is placed on human epidemiological and nutritional data
related to background exposure to copper. The estimates quoted in the
tenth report of the Joint FAO/WHO Expert Committee are probably
conservative and more recent food analyses suggest that the daily
intake of 2 to 3 mg is likely to be exceeded by significant sections
of the population with no apparent deleterious effects. On this basis
there appears to be no reason to change the tentative assessment of
the maximum acceptable daily load of 0.5 mg/kg bw. This figure is
suggested on the understanding that the dietary levels of those
constituents which are known to affect copper metabolism, for example,
molybdenum and zinc, lie within acceptable limits.
REFERENCES
Adamson et al. (1969) Vet. Rec., 85, 368
Anon. (1966) Lancet, 1, 1082
Boyden, R., Potter, U. R. & Elvehjem, C. A. (1938) J. Nutr., 15, 397
Browning, E. (1969) Toxicity of Industrial Metals, II ed., London,
Butterworths
Bull, L. B. (1949) British Commonwealth Special Conference in
Agriculture in Australia 1949, p. 300, London, Her Majesty's
Stationery Office (1951)
Bunch, R. J. et al. (1965) J. An. Sci., 24, 995
Buntain, D. (1961) Vet. Rec., 73, 707
Bureau of Mines (1953) Information circular, 7666
Chuttani, H. K. et al. (1965) Amer. J. Med., 39, 849
Clyde, Parris & McDonald (1969) Can. J. Anim. Sci., 49, 215
Dawson, C. R. & Mallette, M. F. (1945) Advances in Protein
Chemistry, Vol. II, Academic Press
Decker, W. J. et al. (1972) Toxicol. appl. Pharmacol., 21, 331
Doherty et al. (1969) Res. Vet. Sci., 10, 303
Food Standards Committee (1956) Report on copper, London, Her
Majesty's Stationery Office
Gray, L. F. & Daniel, L. J. (1964) J. Nutr., 84, 31
Harrison, J. W. E., Levin, S. E. & Trabin, B. (1954) J. Amer. Pharm.
Ass., sci. Ed., 4, 722
Holtzman, W. A., Elliot, D. A. & Heller, R. H. (1966) New Engl. J.
Med., 275, 347
Hueper, W. C. (1965) UICC Symposium, Paris, Nov. 1965
James, B. E. & MacMahon, R. A. (1970) Med. J. Aust., 1161
Kehoe, R. A., Cholak, J. & Story, R. V. (1940) J. Nutr., 20, 85
Lal, S. & Sourkes, E. L. (1971) Toxicol. appl. Pharmacol., 18, 562
Mahler, D. J., Walsh, J. R. & Haynie, G. D. (1971) Amer. J. clin.
Path., 56, 17
Neumann, P. F. & Silverbeorg, M. (1966) Nature, 210, 416
Nicholas, P.O. (1968) Lancet, 11, 40
O'Hara, P. J., Newman, A. P. & Jackson, R. (1960) Aust. vet. J., 36,
225
Paine, C. H. (1968) Lancet, 11, 520
Schroeder, H. A. et al. (1966) J. Cron. Dis., 19, 1007
Spector, W. S. (1956) Handbook of Toxicology, Vol. 1, W. B. Saunders
Co.
Underwood, E. J. (1962) Trace elements in human and animal
nutrition, New York and London, Academic Press
Vohra, P. & Kratzer, F. H. (1968) Poult Sci., 47, 699
Weedwanders, R. E., Evans, A. W. & Nasdahl, W. W. (1968) I - Lancet,
88, 286
Wilson, J. F. & Lahey, M. E. (1960) Pediatrics, 25, 40
Wolff, S. (1960) Archives of Pathology, 69, 217