Cochineal, carmines and carminic acid were evaluated for
    acceptable daily intake for man by the Joint FAO/WHO Expert Committee
    on Food Additives in 1975 and 1977 (see Refs. 37 and 43 in the Annex).

         Since the previous evaluations, additional data have become
    available and the previous monographs have been expanded.

         Cochineal carmine is obtained from aqueous extracts of cochineal,
    which consists of the dried bodies of the female insect Dactylopius
    coccus costa. The colouring principle of carmines is the hydrated
    aluminium chelate of carminic acid, in which aluminium and carminic
    acid are thought to be present in the molar ratio 1:2 (Meloan et al.,

         In commercial products the colouring principle is present in
    association with ammonium, calcium, potassium or sodium cations,
    singly or in combination, and these cations may also be present in
    excess. Ammonium carmines exhibit solubility over a wide range of pH
    while calcium carmines are sparingly soluble at pH values below 7.
    Commercial products also contain proteinaceous material derived from
    the source insect, and may contain free carminate anion or small
    excesses of aluminium cations (Lloyd, 1980).



    No information on metabolism is available.


    Special studies on mutagenicity

         Carminic acid was negative in the Bacillus subtilis rec-assay
    for DNA-damaging ability (Kada et al., 1972).

         Carminic acid was not mutagenic for several strains of
    Salmonella typhimurium in the presence of liver microsomal
    preparations or enzymatic extracts of rat caecal microflora (Brown &
    Brown, 1976; Brown et al., 1977).

         Carminic acid did not produce reverse mutations in four strains
    (TA-1535, TA-1537, TA-98, TA-100) of Salmonella typhimurium when
    tested in presence and absence of liver microsomal (S9) fractions
    obtained from animals pretreated with phenobarbitone. There was no

    evidence of gene conversions when carminic acid was tested similarly
    in vitro with Saccharomyces cerevisiae D strain, nor of forward
    mutations in vitro or in vivo in host mediated assays using
    Schizosaccharomyces pombe (Barale et al., 1978). Similar results
    have been obtained in studies using Salmonella typhimurium TA-1538
    and Escherichia coli WP2 uvr A (Haveland-Smith & Combes, 1980).

    Special studies on skin sensitization

         Three subjects with lip lesions gave positive patch tests when
    tested with red lip salve containing calcium carmine, but negative
    reactions to colourless lip salve.

    Special studies on teratogenicity

         The embryotoxicity and teratogenicity of carmine have been
    studied in mice. Mice were killed on day 19 of gestation, after i.p.
    injection of lithium carmine or sodium carmine on day 8. Treated
    animals of both groups showed resorption rates (20%) higher than those
    of control groups (2%). The malformation rate was about 16% in the
    lithium carmine group and 2.5% after injection of sodium carmine. Only
    animals given sodium carmine showed an increase in the number of
    retarded foetuses (Schluter, 1970).

         Groups of mice were injected once with 2.5% lithium carmine at a
    dose of 150 mg/kg bw of carmine on days 6, 8, 10, 12 or 14 of
    pregnancy. A teratogenic effect was observed on the first three
    treatment days, with the maximum effect on day 8 (Schluter, 1971a,b).

         Four groups of 30 mated female rats were given daily 0, 200, 500
    or 1000 mg/kg bw of ammonium carmine by gastric intubation as aqueous
    solution during pregnancy days 0 to 20. A group of 17 similar animals
    received a solution of chlorides to provide an intake of sodium,
    potassium and ammonium ions equal to that resulting from the highest
    dose level of carmine. No adverse effects were noted on body weight,
    pregnancy rate, pre-implantation losses, the average number of live
    young litter weight or foetal weight. The group given the highest dose
    of carmine and the cations control had an increased number of
    implantation sites and of post-implantation losses. The latter was
    considered to be due to an inability to maintain the increased numbers
    of implantations rather than to an embryotoxic effect. No teratogenic
    effects were noted in the foetuses and the degree of ossification of
    those from the carmine treated rats tended to be more advanced than
    those from the control (Gaunt et al., 1976).

    Special studies over three generations

         Ammonia carmine was administered to Wistar rats over several
    generations at dietary concentrations designed to provide intakes of
    0, 50, 150 or 500 mg/kg bw per day. Animals of both sexes were used in
    groups of 36 for treatment with carmine and groups of 60 for the
    control. After a suitable period of treatment, the original animals
    (generation FoA) were mated to provide generation F1a and then remated
    to produce generation F1b. Generation F1a animals provided animals for
    generation F2, which in turn provided the final F3 generation.

         There were no effects on body weights, food and water intakes,
    fertility or organ weights in adults of generations FoA, F1a or F2
    which could be attributed to treatment. Post mortem examinations and
    organ weight measurements of pups of generations F1b, F2 and F3 did
    not reveal any differences between control and treated groups which
    could be related to treatment. Histopathological examination of pups
    of generation F3 revealed no treatment related effects. Survival,
    growth and development of pups in treated groups were similar to those
    of the control group apart from a slight delay in tooth eruption in
    the 150 and 500 mg/kg bw groups of generations F1b and F2. No delay in
    tooth eruption was seen in any of the treated groups of generation F3.

         In the teratological investigations, foetuses of all treated
    groups in the generation F3 were slightly more advanced in their
    degree of skeletal ossification compared to the control groups.
    Finally, post mortem examinations of the dams of generations FoA, F1a
    and F2 used in the teratology studies revealed no significant
    differences between control and treated animals except for slightly
    increased numbers of corpora lutea and post-implantation losses in the
    150 mg/kg bw group of generation F1a. These were considered to be
    unrelated to treatment (Grant et al., 1979).

    Acute toxicity

         No information available.

    Short-term studies


         Mice (number not stated) were given intraperitoneal injections of
    a 1 to 2% aqueous solution of the lithium salt of carminic acid for a
    period of 60 days. The only abnormality observed was proliferation of
    spleen tissue (Harada, 1931).


         Groups of 40 rats, equally divided by sex, received ammoniacal
    cochineal carmine in 0.4% aqueous agar by intubation at dosage levels
    of 0, 2.5, 5.0 and 10.0 g/kg bw five days per week for 13 weeks.

         Body weight was recorded bi-weekly. Blood counts were made three
    times. Gross and microscopic findings were not remarkable aside from a
    dose-related accumulation of colour in the tissues of the rats
    receiving the two higher dosage levels. No haematological effects were
    noted. At the two highest levels some decreased growth was apparent.
    Urine and faeces of the treated rats were coloured during the period
    of administration (Battelle, 1962).

         Groups of 50 weanling rats equally divided by sex were fed
    calcium carmine in the diet at levels of 0, 50, 250 and 500 mg/kg
    bw/day for 90 days. Blood counts, blood glucose, blood urea nitrogen
    and urinalyses were performed three times. No effects due to the
    carmine were reported in terms of growth, haematology and other
    clinical findings. Gross and microscopic pathology were not remarkable
    (FDRL, 1962).


         Five rabbits were given intravenous injections every five to
    seven days, of 3 to 10 ml of a 2 to.4% aqueous solution of the lithium
    salt of carminic acid. The treatment was continued for periods
    varying from 130 to 529 days. No tumours were observed, but great
    proliferation of the tissue of the spleen was noted (Harada, 1931).


         Cochineal and colouring principles derived from it were
    considered in the twenty-first report of the Expert Committee but at
    that time the toxicological data available were considered to be
    insufficient for an evaluation and the establishment of an ADI. The
    present Committee understood that the main substances used were
    ammonium carmine for alcoholic beverages and calcium carmine for
    foods. Their use was limited because of the small amount of cochineal
    produced. Recent reproduction studies with ammonium carmine had found
    no toxicologically significant effects. The Committee was informed
    that a long-term study had also been completed, but the data had not
    yet been submitted. The Committee allocated a temporary ADI of
    0-2.5 mg/kg bw for ammonium carmine, or equivalent amounts of the
    calcium, potassium, or sodium salts (the lithium salt is not
    acceptable for food-additive use). This was based on a no-effect
    level of 500 mg of ammonium carmine per kg of body weight in a
    multigeneration study. The Committee requested the submission of the
    results of the long-term study for evaluation at a future meeting.


    Level causing no toxicological effect

    Rat: 500 mg/kg bw per day.

    Estimate of temporary acceptable daily intake for man

    0-2.5 mg/kg bw.


    Required by 1982.

    Submission of the results of the long-term studies.


    Battelle Memorial Institute (1962) Unpublished report submitted to WHO

    Barale, R. et al. (1978) In: Galli, C. L., Paoletti, R. & Vettorazzi,
         G., eds, Proceedings of the International Symposium on Chemical
         Toxicology of Food, Elsevier, North Holland Biomedical Press,
         pp. 349-357

    Brown, J.P. & Brown, R. J. (1976) Mutation Res., 40, 203

    Brown, J. P., Roehm, G. W. & Brown, R. J. (1977) Environ, Mutagen
         Soc. 8th Ann. Meet., Abst., p. 33

    Food and Drug Research Laboratories (1962) Unpublished report
         submitted to WHO

    Gaunt, I. F., Clode, S. A. & Lloyd, A. G. (1976) Unpublished report
         from B.I.B.R.A., submitted to WHO - Studies of teratogenicity and
         embryotoxicity of carmine in the rat. Report 162/1/76, July 1976

    Grant, D., Conning, D. M. & Hawkins, R. I. (1979) Unpublished report
         from B.I.B.R.A., submitted to WHO - Multigeneration Toxicity
         studies in rats with carmine of cochineal. Report 230/1/79,
         December 1979

    Harada, M. (1931) cited by Hartwell, J. L.: Survey of compounds
         which have been tested for carcinogenic activity, 2nd ed., 1951

    Haveland-Smith, R. B. & Combes, R. D. (1980) Foods and Cosmetics
         Toxicology, 18, 215-221

    Lloyd, A. G. (1980) Food Chemistry, 5, 91-107 

    Meloan, S. N., Valentine, L. S. & Puchtler, H. (1971) Histochemie,
         27, 87

    Kada, T., Tutikawa, K. & Sadaie, Y. (1972) Mutation Res., 16, 165

    Sarkany, R. H., Meara, R. H. & Everall, J. (1961) Trans. St. John's
         Hosp. derm. Soc. (Lond.), 48, 39

    Schluter, G. (1970) Z. Anat. Entwickl.-Gesch., 131, 228

    Schluter, G. (1971a) Naunyn-Schmiedeberg's Archiv. Pharmak., 270,

    Schluter, G. (1971b) Naunyn-Schmiedeberg's Archiv. Pharmak., 270,

    See Also:
       Toxicological Abbreviations