Toxicological evaluation of some food
    additives including anticaking agents,
    antimicrobials, antioxidants, emulsifiers
    and thickening agents


    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


    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.



         This substance has been evaluated for acceptable daily intake by
    the Joint FAO/WHO Expert Committee on Food Additives (see Annex 1,
    Refs Nos 6, 8 and 20) in 1961, 1964 and 1969.

         Since the previous evaluation no additional data have become
    available. Therefore the previous monograph remains unchanged and is
    reproduced in its entirety below.



         Biochemical studies have shown that NDGA has an inhibiting effect
    on a number of enzyme systems and it has been suggested that in some
    cases this inhibition in enzyme systems may be due to a denaturing
    action of the antioxidant (Tappel & Marr, 1954). Specific inhibition
    of peroxidase, catalase and ethyl alcohol dehydrogenase occurs with a
    concentration of 2  10-4M of the antioxidant. Non-specific
    inhibition of ascorbic acid oxidase, D-amino-acid oxidase, the
    cyclophorase system and urease at a concentration of 2  10-3M has
    been described (Tappel & Marr, 1954). The serum esterase system was
    markedly inhibited by four fat antioxidants (octyl gallate, ascorbyl
    palmitate, butylhydroxytoluene, and NDGA, resp.) in concentrations of
    0.1-1.0 mM. In the liver, the inhibition induced by octyl gallate was
    significantly smaller than it was by the other three antioxidants. The
    esterase activity in the fatty tissues (perirenal) did not appear to
    be significantly affected by the four antioxidants. It was concluded
    that the antioxidants examined could affect fat metabolism adversely
    by inhibiting the ester-cleaving systems in the serum and liver
    (Placer et al., 1964).

         The metabolism of NDGA has recently been studied in the rat. No
    free NDGA was found in lymph or kidney extracts after long- or short-
    term feeding studies. The first trace of the o-quinone metabolite was
    found in kidney extracts from rats fed 2% NDGA for 28 days. After
    80 days on test, 260 g o-quinone/g kidney tissue was extracted.

         The formation of o-quinone in the rat intestine, after single
    administration of 250 mg NDGA directly into the small intestine, was
    determined at 0, 1, 2, 4, 6 and 7.5 hours after dosing. The content
    of o-quinone in the ilium was found to be 0, 0, traces, 6, 480 and
    2760 g respectively. The content of the caecum after 7.5 hours was
    1620 g.



                                    Analysis of contents of:
                                         Ilium                Caecum

    Time after dosing (hour)  0   1   2       4    6      7.5     7.5

    o-Quinone (ug)            0   0   Trace   6    480    2 760   1 620

         Twenty-four hour urine samples collected from rats fed 2% NDGA
    for 36 days contained no detectable free NDGA but did contain up to
    56 g of o-quinone (Grice et al., 1968).

         The effect of NDGA in rat kidney has been studied in greater
    detail in rats fed 2% NDGA and sacrificed at various times ranging
    from one to 15 months from the start of the experiment. It was
    postulated that the o-quinone metabolite of NDGA was taken up by
    lysosomes of the proximal tubules, where it may have affected the
    permeability of the lysosomal membrane, or cause an inhibition of
    lysosomal enzymes, with subsequent destruction of the cells involved
    (Goodman et al., 1969).


    Acute toxicity

    Animal         Route      (mg/kg bw)         Reference

    Rat            oral       2 000 - 5 500      Lehman et al. (1951)

    Mouse          oral       2 000 - 4 000      Lehman et al. (1951)

    Mouse          i.p.       550                Lehman et al. (1951)

    Guinea-pig     oral       830                Lehman et al. (1951)

    Short-term studies


         No deleterious effects on the rate of growth or food intake were
    noted in mice fed a control diet and diets containing NDGA, gum
    guaiac, phenol and catechol at concentrations of 0.25% and 0.5%
    respectively. The average number of months on the diets varied from 6-
    1/2 to 7-1/2 for NDGA to 8.4-12.6 for the other compounds. Necrosis of
    the liver and spleen were occasionally found in all the groups,
    including the controls (Cranston et al., 1947a and 1947b).


         NDGA was found to be among the more strongly reacting compounds
    among several antioxidants tested for their capacity to induce skin
    sensitivity in the guinea-pig (Griepentrog, 1961).


         NDGA was fed to dogs at dietary levels of 0.1% (three dogs), 0.5%
    (four dogs) and 1.0% (five dogs) for one year. Adult dogs were fed the
    same range of dietary concentrations, the number in the groups being
    two, three and two respectively, with two controls. The growth curves
    of the young dogs showed some impairment in weight gain at the 0.5%
    levels but not at the 0.1% and 1.0% levels. No significant
    pathological changes were found which were attributable to the
    treatment and all the dogs were in good physical condition at the time
    of sacrifice (Cranston et al., 1947a).

    Long-term studies


         Chronic toxicity experiments were conducted over a two-year
    period, in which NDGA was compared with phenol, catechol and gum
    guaiac in concentrations of 0%, and 0.5% in rats (10 males per group).
    All four compounds showed some tendency to decrease the rate of growth
    as compared to the controls, catechol having the greatest and phenol
    the least effect. Massive haemorrhage in the caecum was observed in
    five rats fed NDGA at months 12, 13, 19, 20 and 21; one of the rats
    had a supperative inflammation near the caecum and another had a
    thrombosed artery and ulcer of the caecum (Cranston et al., 1947a and

         Concentrations of 0%, 0.5% and 1.0% NDGA were used with another
    series of rats (18 females per group). NDGA had little or no effect on
    growth or food intake except in the highest concentration, where there
    was a temporary decrease in growth associated with a decreased food

    intake. Histological study of the liver, spleen and kidneys showed no
    significant pathological effect. Single or multiple cysts in the
    mesentory (in the angle of the junction between the small and large
    intestine) were found in 10/12 surviving rats on 1% NDGA and in 15/17
    surviving rats on 0.5% NDGA. Nodules, slightly larger than normal and
    suggestive of the beginning cyst formation, were observed in 1/12
    surviving rats on 0.1% NDGA and in 1/13 surviving control rats. No
    haemorrhage in the caecum occurred in any of the animals in this
    experiment (Cranston et al., 1947a and 1947b).

         Two-year toxicity tests were also carried out with groups of 10
    male rats at 0%, 0.1%, 0.25%, 0.5% and 1.0% NDGA. Inflammatory caecal
    lesions were noted in 4/20 of the animals fed NDGA at levels of 0.5%
    and 1.0%, and slight cystic enlargement of the lymph nodes near the
    caecum in half these animals. The changes were absent at levels of 0%
    (controls), 0.1 and 0.28% NDGA. The gastrointestinal tract apart from
    the caecum was unaffected as were the other structures studied (Lehman
    et al., 1951; Nelson, 1947).

         In a recent study, NDGA fed to weanling rats (10 males and 10
    females per group, Wistar-derived strain), at levels of 0.5% or 1.0%
    in the diet for 74 weeks, caused cystic retinoloendotheliosis of
    paracaecal lymph nodes and vacuolation of kidney tubular epithelium
    (Grice et al., 1968). The retardation in growth and caecal changes
    confirmed previously reported work (Lehman et al., 1951; Cranston et
    al., 1947a and 1947b) for rats fed NDGA at the 0.5% and 1.0% levels.
    The o-quinone derived from NDGA was isolated from kidney tissue and
    identified by thin-layer chromatography and ultra-violet spectroscopy.
    Rats fed 2% NDGA in the diet for shorter periods of time exhibited
    similar pathological changes (Grice et al., 1968).


         A significant new finding is the conversion of
    nordihydroguaiaretic acid to the corresponding orthoquinone which may
    be the cause of the formation of mesenteric cysts. The rat seems to
    have a special susceptibility to this action.


         Not possible with data provided.


    Cranston, E. M. et al. (1947a) Unpublished report, dated 8 March

    Cranston, E. M. et al. (1947b) Fed. Proc., 6, 318

    Goodman, T. et al. (1969) Laboratory Investigation (In press)

    Grice H. C., Becking, G. & Goodman, T. (1968) Fd. Cosmet. Toxicol., 
         6, 155

    Griepentrog, F. (1961) Arzneimittel-Forsch., 11, 920

    Lehman, A. J. et al. (1951) Advanc. Food Res., 3, 197

    Nelson, A. A. (1947) Unpublished summary report of pathology dated
         4 November

    Placer, Z., Veselkova, Z. & Petrasek, R. (1964) Nahrung, 8, 707

    Tappel, A. L. & Marr, A. G. (1954) J. Agr. Food Chem., 2, 554

    See Also:
       Toxicological Abbreviations
       Nordihydroguaiaretic acid (FAO Nutrition Meetings Report Series 38a)
       Nordihydroguaiaretic acid (FAO Nutrition Meetings Report Series 46a)