RED 2G
Explanation
This food colourant was evaluated by JECFA in 1977, and a
temporary ADI for man of 0-0.006 mg/kg bw was established (WHO, 1978).
Further work required included a multigeneration study, and studies on
bone marrow to elucidate the toxic effects on erythropoiesis. In 1979
none of these studies had been completed and JECFA extended the
previously established temporary ADI (WHO, 1980).
Since the previous evaluations, additional data have become
available and are included in this monograph.
BIOLOGICAL DATA
BIOCHEMICAL ASPECTS
Rats were injected intravenously with Red 2G. Bile was collected
for six hours and analysed. The recovery of the colour was on average
64% of the administered quantity (Priestly & O'Reilly, 1966). Biliary
excretion of i.v. administered Red 2G was also reported by Ryan &
Wright (1961).
In another experiment 250 mg/kg bw of Red 2G was administered by
gastric intubation to five male and five female rats. On average the
males excreted 61.8% of the dose in the urine and the females 71.5%,
42.2% of the dose was excreted in the urine as p-aminophenol in 48
hours, 9.2% as aniline in 24 hours and 3% as unreduced dye in 24
hours. The corresponding faecal excretion was 6.3, 1.0 and 0.1%. For
females urinary excretion amounted to 56.4% as p-aminophenol in 48
hours, 2% as aniline in 24 hours and 2.6% as unreduced dye in 24
hours. The corresponding faecal excretion was 8.6, 0.3 and 1.6%
(Walker, 1971).
The contents of a rat caecum were incubated at 37°C with a
solution of Red 2G in isotonic saline. At one-hour intervals a sample
of the incubate was filtered and Red 2G was estimated in the filtrate
by measuring the optical density. Two metabolites of Red 2G were
detected in the incubation mixture after separation by thin layer
chromatography on silica plates. One was 2-amino-8-acetamido-1-
naphtho-3,6-disulfonic acid, the other, aniline, was detected using
two different solvent systems. When Red 2G was incubated at 37°C with
liver homogenate the same two metabolites were detected (Jenkins et
al., 1966).
When a mixture of Red 2G and caecal contents were incubated
at 37°C darkening at the surface was observed. This was attributed
to oxidation of a metabolite of Red 2G, presumed to be a sulfur-
containing compound. Two groups of 12 rats were fed purified diet and
a purified diet containing 0.51% Red 2G respectively. Faeces were
collected and it was calculated that 48.2% of the sulfur derived from
Red 2G was excreted in the faeces (Unilever, 1974).
When rabbits were fed 0.5 g/kg bw of the colour the following
metabolites could be identified in urine over a period of 48 hours:
total p-aminophenol, 46%; p-aminophenylglucuronide, 37%; aniline, 0.6%
and o-aminophenol, 9%. The ratio of o-aminophenol to p-aminophenol was
the same for rabbits fed Red 2G and rabbits fed aniline previously
examined by Parke (1960) indicating that hydroxylation does not
necessarily precede fission of the azolinkage.
No binding of Red 2G to serum protein occurs (Jenkins et al.,
1966a).
TOXICOLOGICAL STUDIES
Special studies on aniline
Rat
Aniline dissolved in isotonic saline was administered either
intravenously to rats under anaesthesia or by stomach tube. Saline was
administered to control animals. Blood samples from the tail were
taken at 30-minute intervals at first and then at 60-minute intervals.
The no-effect dose of aniline was 20 mg/kg bw orally and 10 mg/kg bw
intravenously (Jenkins et al., 1972).
Groups of six male and six female rats were fed diets containing
0.098% aniline and molecular equivalent levels of p-aminophenol and
phenylhydroxylamine. Methaemoglobinaemia, Heinz bodies and splenic
enlargement were noted in rats which received either aniline or
phenylhydroxylamine. The no-effect single oral dose of aniline in rats
was 20 mg/kg bw (Jenkins et al., 1972).
Man
Aniline was administered orally to a volunteer for five days. The
dose was 10 mg on days 1 and 2 and 25 mg on days 3, 4 and 5. Urine
samples were tested for urobilinogen, glucose and protein and blood
tests included, haemoglobin, methaemoglobin, packed cell volume, serum
transaminases, alkaline phosphatase, thymol turbidity, serum proteins,
serum bilirubin and the staining for Heinz bodies. None of these tests
revealed an effect due to the ingestion of aniline, nor could Red 2G
be detected in the urine (Jenkins et al., 1972).
Single oral doses of 5 and 15 mg aniline had no effect in 20
human subjects; doses ranging from 25 to 65 mg significantly increased
the blood level of methaemoglobin to 2.46% but no Heinz bodies were
observed. Although human subjects are more sensitive to aniline
in vivo than rats, the methaemoglobin content of rat blood exposed
to phenylhydroxylamine in vitro exceeded that of human blood exposed
to phenylhydroxylamine in vitro. Glucose promoted the production of
methaemoglobin (Jenkins et al., 1972).
Special studies on haemotoxicity and haemopoiesis
Rats were fed 1-1.5 g/kg per day Red 2G for 75 days. The mean
peak Heinz body level was 80% falling to a maintained level of 30%.
Internal changes included a moderate though well controlled anaemia,
pronounced reticulocytosis, and splenomegaly (Rofe, 1957).
Methaemoglobin and Heinz bodies have been observed in acute human
aniline poisoning by a number of investigators (Freifield et al.,
1937; Hughes & Treon, 1954; Rodek & Westhaus, 1952). Methaemoglobin
formation occurred first followed by Heinz body appearance, the latter
being an irreversible process, Heinz bodies are formed most in vivo
but can be studied also in vitro.
It has been shown that phenylhydroxylamine is the N-hydroxylated
metabolite which produces most of the methaemoglobin observed in the
dog after injection of aniline. The reaction probably proceeds by
oxidation of phenylhydroxylamine to nitrosobenzene by O2 and Hb with
conversion of Hb to methaemoglobin (Kiese, 1959).
Since the rat is less susceptible to Heinz body producing agents
than the cat and possibly man, a procedure was devised which increased
the sensitivity of the rat to these agents by pretreating rats with
p-aminopropiophenone, 15 mg/kg bw subcutaneously (Unilever, 1974).
Methaemoglobinaemia is a reversible response to toxic injury and
depends on the integrity of the erythrocyte in vivo. Reduction of
methaemoglobin occurs through an NADH-dependent diaphorase system
which is deficient in subjects with hereditary mathaemoglobinaemia or
in young infants.
The effects of feeding aniline, para-aminophenol and
phenylhydroxylamine at a level of 0.1% in the diet of rats were
compared for 13 days. The results indicated that compared with a
control group, para-aminophenol had no effect on spleen weight,
aniline increased mean relative spleen weight by about 60% and
phenylhydroxylamine increased mean relative spleen weight by about
500%. Examination of the blood after feeding phenylhydroxylamine
for 11 days revealed a high incidence of Heinz bodies. Therefore,
the toxic effects of feeding aniline can be attributed to
phenylhydroxylamine which is a metabolite of aniline (Gellatly &
Burrough, 1966; Jenkins et al., 1966b).
Three groups of six four-week old female rats were fed in their
diet 0, 0.5% Red 2G or 0.093% aniline. All rats were fed ad libitum
for 19 days. The faeces of rats fed Red 2G were almost black; the
faeces of rats fed aniline were a normal colour. Both Red 2G and
aniline caused a similar increase in spleen weight, accelerated
erythropoiesis and haemosiderin content.
For rats fed purified diets containing 0. 1, 0.2 and 0.3% Red 2G
for two weeks there was a linear relationship between intake of Red 2G
and relative spleen weight. For rats fed purified diets containing
0.004, 0.006 and 0.012% phenylhydroxylamine for two weeks there was a
linear relationship between intake of phenylhydroxylamine and relative
spleen weight (Jenkins et al., 1967; Gellatly & Burrough, 1967).
For individual samples of rat blood and human blood the amount of
oxidation of haemoglobin to methaemoglobin was linearly related to the
logarithm of phenylhydroxylamine concentration. From the dose-response
curve it has been estimated that for rat blood the no-effect dose is
between 0.5 and 1 µg phenylhydroxylamine/ml blood. The response of
human blood to phenylhydroxylamine was more variable than the response
of rat blood. The no-effect concentration of phenylhydroxylamine
in vitro for human blood ranged from 0.46 to 4.1 µg/ml (blood). At
all levels of Red 2G fed to rats the proportion of Red 2G metabolized
to phenylhydroxylamine was constant (Unilever, 1974).
In vitro studies were conducted on samples of mouse, rat and
human blood with 100 and 200 mg% concentration of Red 2G and
acetylphenylhydrazine as a positive control. No Heinz bodies were seen
with Red 2G (BIBRA, 1965).
Red 2G was fed to male and female Colworth-Wistar rats at dietary
levels of 0 or 0.5% for two weeks. Some rats from each group were bled
by cardiac puncture (2 ml from males; 1 ml from females) to stimulate
erythropoiesis two weeks after administration of Red 2G had ceased. At
weekly intervals before, during and after feeding Red 2G, three to
five rats from each group were killed, liver, spleen and kidneys were
weighed and fixed for microscopic examination. Bone marrow samples
were taken for differential cell counts and blood samples were
examined for haemoglobin, methaemoglobin, red cell count, white cell
count, differential white cell count and PCV. In rats fed Red 2G,
methaemoglobinaemia, Heinz bodies, reticulocytosis, decreased
haemoglobin, PCV and red 3 cell count and increased spleen weight were
observed. In liver, spleen and bone marrow there was an increase in
erythropoiesis. Rats previously dosed with Red 2G responded to
bleeding in a similar manner to controls with increased erythropoiesis
observed in bone marrow. The results indicated that liver, spleen and
bone marrow respond to Red 2G with an increase in erythropoiesis but
there was no evidence that Red 2G had a toxic effect on erythropoiesis
(Jenkins et al., 1980).
Special studies on mutagenicity
In a series of microbial mutagenicity assays, food grade samples
of Red 2G were shown to possess weak mutagenic activity, inducing
repairable DNA damage and base-substitution mutations, only after
metabolic activation with rat-liver microsomes. The products of
reducing the azo group in Red 2G were non-mutagenic (Haveland-Smith et
al., 1979; Haveland-Smith & Combes, 1980, 1980a).
Special studies on reproduction
Two groups of 46 male and 46 female rats were fed 0.2% Red 2G in
their diet for 18 weeks and then mated for 10 days. The progeny were
weaned on the same diet and mated at 16 weeks and the F2 generation
was also weaned onto the same diet. No adverse effects were seen on
litter size, litter weight and weaning weight nor were there any
abnormalities at autopsy (Unilever, 1974).
Special studies on teratogenicity
Four groups of 20 female Colworth-Wistar rats were fed diets
containing 0, 40, 200 or 2000 ppm (0, 0.004, 0.02 or 0.2%) Red 2G on
days 0-19 of pregnancy (equal to 0, 3.5, 12.9 or 187.5 mg/kg bw per
day). Fifteen rats from each group were subjected to Caesarian section
on day 21 of gestation and the foetuses examined for visceral and
skeletal malformations; dams were examined for corpora lutea,
implantations and intrauterine deaths. Maternal blood, liver
and spleen, and foetel bone marrow and spleen were examined
histologically. Five rats from each group were allowed to litter
normally and to suckle the pups to weaning at 21 days, when the pups
were subjected to examination for malformations.
There was no evidence that Red 2G possessed foeto-toxic or
teratogenic potential at any of the dose levels studied. Treatment did
not affect normal parturition or postnatal development to weaning. The
only significant effect was the expected increase in the spleen weight
and erythropoietic activity in dams at the top dose level; no effects
on foetal spleen were observed (Cambridge et al., 1980).
Acute toxicity
LD50
Animal Route (g/kg bw) Reference
Mouse Oral 7.35 Unilever, 1974
i.p. 4.76 (4.13-5.85) Unilever, 1974
i.p. 3.0-4.0 BIBRA, 1965
Rat i.p. 6.35 (5.62-7.17) Unilever, 1965
Oral >5.0 Unilever, 1974
Guinea-pig Oral 4.81 (3.16-7.35) Unilever, 1974
i.p. 3.0 (1.83-4.91) Unilever, 1974
Rabbit Oral >5.0 Unilever, 1974
Chicken Oral >10.0 Unilever, 1974
No deaths occurred in six weanling rats following administration
of 5 g/kg bw (Unilever, 1974).
A dose of 5 g/kg bw was administered on each of two successive
days to a rabbit weighing 3.8 kg and a dose of 25 g/kg bw was
administered on each of two successive days to a rabbit weighing
4.3 kg. No signs of toxicity were observed and their red cells
contained no Heinz bodies (Unilever, 1974).
Histological studies in rats, rabbits and guinea-pigs doses as
given above showed extensive renal necrosis. In mice dosed orally with
Red 2G there was gross leptomeningeal vascular engorgement and focal
subarachnoid haemorrhage (Unilever, 1974).
No deaths occurred in three-week old and nine-week old chickens
following administration of 10 g/kg bw. There was no evidence of renal
necrosis (Unilever, 1974).
Short-term studies
Mouse
Five groups of 15 male and 15 female mice were given diets
containing 0.0, 0.01, 0.1, 1.0 and 2.0% Red 2G. Five mice of each sex
at each dose level were killed at 26, 55 and 96 days, when a full
autopsy and haematological investigation was carried out on each
animal. No adverse effect on growth or food consumption was evident in
any animal given Red 2G in the diet. Heinz bodies were seen at all
levels, the incidence being related to dose and duration of treatment.
The maximal effect was seen on day 26 and day 55 at a level of 0.1%
and below and on day 26 at 1.0 and 2.0%. Splenomegaly was seen at 2.0%
in both sexes and at 1.0% in females. Increased relative liver weights
were found in females at 2.0% throughout the tests and at 1.0% after
26 days treatment. The only pathological finding attributable to the
administration of Red 2G was increased haemosiderin in the Kupffer
cells of the liver at 2.0 and 1.0%. Haemosiderin was also present in
the spleen at 2.0% throughout the test; at 1.0, 0.1 and 0.01% the
incidence of haemosiderin increased with the duration and treatment
(BIBRA, 1965).
In another study Red 2G was fed for six weeks to five groups
of 10 mice at dietary levels of 0, 0.02, 0.1, 0.5 and 1.0%. The
toxic effects observed were development of Heinz bodies,
methaemoglobinaemia, splenic enlargement, accelerated splenic
macrophages. No toxic effects were observed in mice fed the diet
containing 0.02% (Unilever, 1974).
Rat
Six groups of five male and five female rats were given diets
containing 0.0, 0.05, 0.1, 0.5, 1.0 and 2.0% Red 2G for three weeks.
A further four groups of 10 male and 10 female rats were given diets
containing 0.0, 0.01, 0.05 and 0.1% Red 2G for two months. Retarded
growth associated with reduced food consumption was seen at 5.0 and
2.5% after nine days, with an initial retardation at 2.0%. No
effect on growth or food consumption was seen at lower levels of
administration. Macrocytosis, reticulocytosis and polychromasia were
evident at 5.0% with circulating normoblasts and a normoblastic
marrow. Heinz bodies were present in animals at 1.0% and above after
nine days and 1.0 and 0.5% after three weeks exposure. Signs
indicative of increased erythropoiesis could be seen in animals at all
levels down to 0.1%. Significant splenomegaly was evident at all
levels above 0.5% with scattered non-significant increases in spleen
weight at 1.0 and 0.05% after two months. Increased kidney weight was
also seen at 0.1% and above. On histological examination the only
change attributable to Red 2G was the increased haemosiderin seen in
the Kupffer cells of the liver, renal tubule cells and spleen in all
animals at 2.0% and in some at 1.0 and 0.5%. Blood samples were taken
from the tail of the rats. By nine days a definite haemolytic anaemia
with Heinz body formation was present at all levels from 0.5 to 2.0%
(BIBRA, 1965).
Three groups of 12 rats received Red 2G in the drinking-water at
levels of 0, 0.1 and 0.5% for 100 days. Heinz bodies were seen after
10 days in the red cells of rats fed 0.5% Red 2G in the drinking-
water, fewer were seen after 18 days and none or very few on later
occasions. A few Heinz bodies were occasionally seen in the red cells
of some of the rats fed 0.1% Red 2G in the drinking-water. The spleens
of rats fed 0.1% Red 2G were slightly larger than controls and the
spleen of rats fed 0.5% Red 2G were very much larger than controls.
Histological examination of livers of rats fed 0.5% Red 2G revealed an
increase in haemosiderin present in Kupffer cells and increased
erythropoietic activity. Histological examination of the spleens of
rats fed 0.5% Red 2G also revealed increased erythropoiesis and red
pulp engorgement. There was no effect of Red 2G on urine specific
gravity (Jenkins et al., 1966c; Gellatly et al., 1966).
Four groups of 24 male and 24 female rats were fed at dietary
levels of Red 2G which would ensure intakes of 100 x and 600 x the
assumed average daily dietary intake of Red 2G. This was achieved by
feeding sausage meat, containing 0, 30 and 180 ppm (0, 0.003 and
0.018%) Red 2G respectively in the diet at a level of 80%. The diets
containing Red 2G in sausage meat had no effect on growth, organ
function or organ weights. Blood tests also revealed no evidence of
toxicity. Histological examination revealed that, in the spleens of
rats fed sausage meat containing 180 ppm (0.018%) Red 2G, there was
increased erythropoiesis, increased splenic red pulp haemosiderin and
increased red pulp reticular impregnation with iron. No-effects on
spleen were seen in rats fed sausage meat containing 30 ppm (0.003%)
of Red 2G. Red 2G at 30 ppm (0.003%) and at 180 ppm (0.018%) in
sausage meat had no detectable histological effect on liver (Jenkins
et al., 1966d).
Long-term studies
Mouse
Five groups of 40 male and 40 female mice were fed diets
containing 0, 0.005, 0.025, 0.125 and 0.625% for 20 months. Splenic
enlargement and darkening were seen in mice fed dietary levels of
0.125 and 0.625% of Red 2G; in these animals there was accelerated
formation of red blood cells in the spleen and increased deposition of
iron in spleen and kidneys. There was no evidence of carcinogenicity
attributable to the feeding of Red 2G to mice. More than three-
quarters of the animals in each group survived for two years
(Unilever, 1974).
Rat
Five groups of 40 male and 40 female rats were fed at dietary
levels of 0.004, 0.016, 0.064 and 0.16% Red 2G. Rats fed diets
containing 0.064 and 0.16% of Red 2G showed splenic enlargement and
darkening attributable to increased storage of iron resulting from
haemolysis of red blood cells. Necrosis of splenic elastica was also
identified, this lesion being a sequel to prolonged splenic
enlargement. There was no evidence of carcinogenicity attributable to
the feeding of dietary levels of Red 2G up to 0.16% for two years.
Over half of each group of rats survived for two years (Unilever,
1974).
Two groups of 30 male and 30 female rats were fed a diet
containing 0 or 0.5% Red 2G. At the 0.5% level there was enlargement
and darkening of the spleen, attributable to accelerated splenic
erythropoiesis, increases splenic haemosiderin deposition and
extensive degeneration of splenic elastica. Biochemical studies of
blood and urine revealed no adverse effects on liver and kidneys which
could be attributed to feeding 0.5% Red 2G in the diet. More than half
the animals in the test group survived for two years (Unilever, 1974).
Comments
In the in vitro bacterial mutagenicity studies Red 2G only
induced mutations at very high concentrations of 10 mg/ml; no activity
was detectable at 1 mg/ml. Microsomal activation was required for
mutagenic activity and the reduction products of Red 2G were inactive.
Since Red 2G is substantially reduced by the gut microflora prior to
absorption (Walker, 1971) the concentrations of the dye reaching the
tissues (other than the intestinal mucosa) is likely to be very small
and considerably lower than that required to induce mutations in the
bacterial studies.
The teratology study and the study on the effects on
erythropoiesis were in response to the work required in the
previous evaluation by JECFA (WHO, 1980).
EVALUATION
Level causing no toxicological effect
Mouse: 0.025% in the diet, equal to 43-26 mg/kg/day for 80 weeks.
Rat: 0.016% in the diet, equivalent to 8 mg/kg/day for two years.
Estimate of acceptable daily intake for man
0-0.1 mg/kg bw.
FURTHER WORK OR INFORMATION
Desirable
Further studies on the mutagenicity of Red 2G and its major
subsidiary components in a range of test systems in vitro.
REFERENCES
BIBRA (1965) Acute and short-term feeding studies on Red 2G with
associated haematological investigations, British Industrial
Biological Research Association, Research Report 3/1965.
Unpublished report submitted to WHO
Cambridge, G. W. et al. (1980) Teratology study on Red 2G in the
Colworth Wistar Rat. Unpublished report from Unilever Research
Laboratories, submitted to WHO by Unilever Ltd.
Freifield, H., Schilowa, A. & Ludwinowsky, R. (1937) Heinz bodies and
methaemoglobin formation in poisoning by compounds containing
amino and nitro groups, Folia Haematol., 56, 333-342
Gellatly, J. B. M. & Burrough, R. (1966) Effects of diets containing
aniline and metabolites of aniline on spleen weights; pathology.
Unpublished report from Unilever Research Laboratories, submitted
to WHO by Unilever Ltd
Gellatly, J. B. M. & Burrough, R. (1967) Pathology of Red 2G and
phenylhydroxylamine in fat diets. Unpublished report from
Unilever Research Laboratories, submitted to WHO by Unilever Ltd.
Gellatly, J. B. M., Salmond, G. & Burrough, R. (1966) Sub-acute
toxicity of Red 2G; pathology. Unpublished report from Unilever
Research Laboratories, submitted to WHO by Unilever Ltd.
Haveland-Smith, R. B., Combes, R. D. & Bridges, B. A. (1979)
Methodology for the testing of food dyes for genotoxic activity:
Experiments with Red 2G (C.I. 18050), Mut. Res., 64, 241-248
Haveland-Smith, R. B. & Combes, R. D. (1980) Screening of food dyes
for genotoxic activity, Fd. Cosmet. Toxicol., 18, 215-221
Haveland-Smith, R. B. & Combes, R. D. (1980) Genotoxicity of the food
colours Red 2G and Brown FK in bacterial systems. Use of
structurally-related dyes and azo-reduction, Fd. Cosmet.
Toxicol., 18, 223-228
Hughes, J. P. & Treon, J. F. (1954) Erythrocytic inclusion bodies in
the blood of chemical workers, Arch. Hyg. occup. Med., 10,
192-202
Jenkins, E. P. et al. (1966) Metabolism of Red 2G by rat liver
homogenate. Unpublished report from Unilever Research
Laboratories, submitted to WHO by Unilever Ltd.
Jenkins, F. P. (1966a) Metabolism of Red 2G (protein binding
capacity). Unpublished report from Unilever Research
Laboratories, submitted to WHO by Unilever Ltd.
Jenkins, F. P. at al. (1966b) Effects of diets containing aniline
and metabolites of aniline on spleen weights. Unpublished
report from Unilever Research Laboratories, submitted to WHO by
Unilever Ltd.
Jenkins, F. P. et al. (1966c) Sub-acute toxicity of Red 2G.
Unpublished report from Unilever Research Laboratories, submitted
to WHO by Unilever Ltd.
Jenkins, F. P., Salmond, G. & Gellatly, J. B. M. (1966d) Sub-acute
toxicity of Red 2G in sausage meat. Unpublished report from
Unilever Research Laboratories, submitted to WHO by Unilever Ltd.
Jenkins, F. P. et al. (1967) Effect of Red 2G and phenylhydroxylamine
on spleen weight. Unpublished report from Unilever Research
Laboratories, submitted to WHO by Unilever Ltd.
Jenkins, F. P. et al. (1972) The no-effect dose of aniline in human
subjects and a comparison of aniline toxicity in man and rat,
Fd. Cosmet. Toxicol., 10, 671-679
Jenkins, F. P., Kennedy, S. & Williams, T. C. (1980) Erythropoiesis
in the Colworth Wistar rat fed Red 2G. Unpublished report from
Unilever Research Laboratories, submitted to WHO by Unilever Ltd.
Kiese, M. (1959) Significance of the oxidation of aniline to
nitrosobenzene for the formation of methemoglobin after
absorption of aniline into the organism, Arch. exp. Path.
Pharmak., 235, 360-364
Parke, D. V. (1960) Studies in detoxication 84. The metabolism of 14C
aniline in the rabbit and other animals, Biochem. J., 77,
493-503
Priestley, G. & O'Reilly, W. J. (1966) Protein binding and the
secretion of some azo dyes in rat bile, J. Pharm. Pharmacol.,
18, 41-45
Rodek, H. & Westhaus, H. (1952) Aniline poisoning in nurslings from
ink and labeling dyes, Arch. Kinderh., 145, 77-90
Rofe, P. (1957) Azo dyes and Heinz bodies, Brit. J. ind. Med., 14,
275-280
Ryan, A. J. & Wright, S. E. (1961) The excretion of some azo dyes
in rat bile, J. Pharm. Pharmacol., 13, 492-495
Unilever (1974) Unpublished review of the biological effects of food
colour Red 2G dated November 1974. Submitted to WHO by Unilever
Ltd.
Walker, R. (1971) Personal communication to E. M. den Tonkelaar,
National Institute of Public Health, Bilthoven, Netherlands
World Health Organization (1978) Twenty-first report of the Joint
FAO/WHO Expert Committee on Food Additives, Wld Hlth Org.
techn. Rep. Ser., No. 617, 94-102
World Health Organization (1980) Twenty-third report of the Joint
FAO/WHO Expert Committee on Food Additives, Wld Hlth Org.
techn. Rep. Ser., No. 648, 84-92