ESTRAGOLE

    Explanation

         This flavouring agent was assessed for the first time by the
    Joint Expert Committee on Food Additives.

    BIOLOGICAL DATA

    BIOCHEMICAL ASPECTS

    Metabolism

         Approximately 20% of a dose of estragole was excreted in the
    urine of outbred CD-1 mice as a conjugate of 1'-hydroxy estragole
    (Drinkwater et al., 1976).

         Solheim & Scheline (1973) found that O-demethylation to
    p-hydroxyallylbenzene and oxidation of the allylic side chain to
    2-hydroxy-3-(p-methoxyphenyl) popionic acid or anisic acid (excreted
    largely as p-methoxyhippuric acid) are the main metabolic reactions.
    Other reactions yield 1-(p-methoxyphenyl) allyl alcohol,
    p-methoxyphenyl acetic acid and the latter's glycine conjugate. At
    least 60-65% (possibly 90%) of the doses administered were accounted
    for.

    TOXICOLOGICAL STUDIES

    Special studies on pharmacological aspects

         Estragole caused contractions of isolated frog muscles followed
    by relaxation. This effect was reduced in potassium-depolarized
    muscles or in the presence of quinidine or manganese. Estragole
    appears to cause these effects by producing a persistent
    depolarization of the muscle membrane (Zutshi et al., 1977).

         Estragole was found to be strongly active in prolongation of
    hexobarbital narcosis and zoxazolamine paralysis following
    intraperitoneal doses of 2.5, 10, 40, 160 and 640 mg/kg in mice. This
    implies that it inhibits hepatic microsomal enzyme functions (Fujii et
    al., 1970).

    Special studies on carcinogenicity

         Estragole and its 1-hydroxy metabolite caused significant
    increases in the incidences of hepatocellular carcinomas in male CD-1
    mice that received the compound by subcutaneous injection at 1-22 days
    of age. Doses given were 4.4 and 5.2 µmol (Drinkwater et al., 1976).

         1-hydroxyestragole did not demonstrate liver-mediated
    mutagenicity in a bacterial test system (Drinkwater et al., 1976).

         Estragole has no cytotoxic activity against HeLa cells (Stoichev
    et al., 1967). The ED50 (dose required for 50% cell growth inhibition)
    was reported at 5.23 × 104 Molar. This value implies that it has less
    potency as a tumour growth inhibitor than delta-9-tetrahydrocannabinol
    (Nichols et al., 1977).

    Special studies on dermal toxicity

         Estragon oil applied full strength to intect or abraded rabbit
    skin for 24 hours under occlusion was irritating (Shelanski, 1973). It
    was also irritating when applied undiluted to the backs of hairless
    mice (Urbach & Forbes, 1973). Tested at 4% in petrolatum, it produced
    no irritation after a 48-hour closed-patch test in 25 human subjects
    (Kligman, 1972).

    Acute toxicity

                                                                       

                                            LD50
    Species/route of       Mouse              Rat             Rabbit
    administration         g/kg              g/kg              g/kg

                                                                       

    Oral                   1.25              1.23
                       Jenner et al.     Opdyke et al.
                          (1964)            (1976)

    Dermal                                   1.82                5
                                         Jenner et al.     Opdyke et al.
                                            (1964)            (1976)

    Intraperitoneal        1.26              1.03
                      Coujelle et al.   Coujelle et al.
                          (1958)            (1958)
                                                                       

    Chronic/sub-chronic toxicity

         Rats given four daily oral doses of 605 mg/kg showed minor liver
    damage (Taylor et al., 1964).

    OBSERVATIONS IN MAN

         A maximation test (Kligman, 1966) was carried out on 25
    volunteers. The material was tested at a concentration of 4% in
    petrolatum and produced no sensitization reactions (Kligman, 1972).

    Comments

         A single sample of estragole has been tested in the CD-1 strain
    of mouse by the intraperitoneal and oral route which resulted in
    increased incidence of hepatocellularcarcinoma.

         Because no no-effect level has been demonstrated no ADI could be
    established. In view of the importance of the compound and the limited
    amount of data, the Committee would be pleased to see this work
    repeated in another strain of mouse.

    EVALUATION

    No ADI allocated.

    REFERENCES

    Coujelle, M. & Meynier D. (1958) Comptes rendus hebdomadaires
         des seances, Academie des Sciences, 246, 1465

    Drinkwater, N. R. et al. (1976) J. Nat. Cancer Inst., 57 (6), 1323-1331

    Fujii, K. et al. (1970) Toxicol. Appl. Pharmacol., 16 (2), 482-494

    Jenner, P. M. et al. (1964) Food Cosmet Toxicol., 2, 237-343

    Kligman, A. M. (1966) J. invest. Derm., 47, 393

    Kligman, A. M. (1972) Report to RIFM, 1 November

    Nichols, D. E., Mason, D. L. & Jackson, L. B. (1977) Life Sciences, 21
         (9), 1245-1248

    Opdyke, D. L. J. (1976) RIFM mongraphs on fragrance materials, Food
         Cosmet. Toxicol., 14, 603

    Shelanski, M. V. (1973) Report to RIFM, 30 January

    Solheim, E. & Scheline, R. R. (1973) Xenobiotica, 3 (8), 493-510

    Stoichev, S. et al. (1967) Bulg. Sci., 20 (12), 1341-1344

    Taylor, J. M., Jenner, P. M. & Jones, W. I. (1964) Toxicol. Appl.
         Pharmacol., 6, 378

    Urbach, F. & Forbes, P. D. (1973) Report to RIFM, 8 February

    Zutshi, S. K. & Bhagwat, A. W. (1977) Curr. Sci., 46 (7), 223-224
    

    See Also:
       Toxicological Abbreviations
       ESTRAGOLE (JECFA Evaluation)