BHA was evaluated for acceptable daily intake for man (ADI) by
    the Joint FAO/WHO Expert Committee on Food Additives (JECFA) in
    1961, 1965, 1973, 1976, 1980, 1982, 1983 and 1986 (Annex 1,
    references 6, 11, 32, 41, 53, 59, 62 and 73). Toxicological
    monographs were issued in 1961, 1973, 1976, 1980, 1983 and 1986
    (Annex 1, references 6, 33, 42, 54, 63 and 74). At the 30th meeting
    of the JECFA (Annex 1, reference 73), a temporary ADI of
    0-0.3 mg/kg bw was established pending adequate studies in pigs and
    monkeys to explore the potential for BHA to cause esophageal
    hyperplasia in these species. A multigeneration reproduction study
    was also required. Since the previous evaluation, new data have
    become available and are summarized and discussed in the following
    monograph. In addition, the monograph reviews special studies of
    the effect of BHA on the stomach of experimental animals.



    Covalent binding to macromolecules

         Metabolites binding irreversibly to proteins were found in an
     in vitro study involving liver microsomes + NADPH, liver
    microsomes + cumene hydroperoxide, sheep seminal vesicle microsomes
    + arachidonic acid, or horseradish peroxidase + hydrogen peroxide
    (Newberne et al., 1986).

         In another  in vitro study (Cummings et al., 1985),
    3- tert-butylhydroxyanisole (3-BHA) metabolism was found to yield
    material which covalently bonded to protein, although the amount of
    these products was reduced directly in proportion to the amount of
    glutathione available.

         In a third study (Hirose et al., 1987a), however, no
    covalent binding to macromolecules was found in a 9000 g
    supernatant fraction derived from forestomach epithelium that had
    been incubated with 3-BHA in the absence or presence of NADPH.
    Looking at male F344 rats that had been given an intragastric
    injection of radioactively labelled 3-BHA, no metabolites were
    detected by thin layer chromatography in the forestomach or
    glandular stomach epithelium six hours after treatment. Assessing
    the macromolecular binding of 2- and 3-BHA, these same authors
    reported no binding to forestomach, glandular stomach, liver or
    kidney DNA or RNA, while binding to protein was similar at the
    four sites. They concluded that the activity of BHA is not related
    to any covalent binding property of BHA.

         In contrast to this, deStafney et al. (1986) reported that
    3-BHA bound  in vitro to liver microsomal proteins.  In vivo,
    3-BHA bound to forestomach microsomal proteins 14-fold more than to
    glandular stomach microsomal proteins and 12-fold more than liver
    microsomal proteins. With HPLC studies, these workers showed that
    there were peaks in the forestomach preparation that were not
    present in other preparations. The authors speculated that the
    effect of BHA was due to its ability to deplete thiols (accounting
    for the noticeable threshold of its effects), and the binding to
    tissue proteins or reactive BHA metabolites, possibly oxygen
    radicals resulting from the redox cycling of quinones and


         The elimination of BHA was studied in male F344 rats (Hirose
    et al., 1987b). Within 48 hours of ingestion of BHA, 87-96% of
    the BHA was found to be eliminated via urine, feces or respiration.
    The isomers 2-BHA and 3-BHA labelled at either the methoxy or
     tert-butyl site were used. For 3-BHA and 2-BHA, respectively,

    63.7% and 69.0% of the  tert-butyl label was found in the urine 
    and 28.8% and 18.1% in the feces (expired air was not examined for 
    these isotopes). For 3-BHA and 2-BHA, respectively, 49.8% and 46.5% 
    of the methoxy label was found in the urine, 28.3% and 29.6% in the 
    feces, while 8.3% and 13.7% was found in expired air.

         In a second study in which Sprague-Dawley rats were used
    (Ansari & Hendrix, 1985), 41% of the methoxy labelled 3-BHA was
    found in the urine while 53% was found in feces 48 hours after
    administration by gavage.

         In beagle dogs, within 48 hours of an J.p. injection of
    methoxy labelled 3-BHA, 50-80% of the label was found in the urine
    and 15-30% in feces (Takizawa et al., 1985).

    Toxicological Studies

    Special studies on the effect of BHA on the stomach


         Groups of 10 male NMRI mice received daily doses of 0 or
    1000 mg BHA/kg bw for 28 days via intubation of an arachis oil
    solution. At the end of this period,  macroscopic lesions
    were seen in the forestomach of the mice that resembled lesions
    in rats (Altmann et al., 1986).

         Male B6C3F1 mice were exposed to 0.5 or 1% BHA in their
    powdered diet for 104 weeks and groups of 10 animals were examined
    at 8 week intervals. Hyperplasia of the forestomach was first seen
    in the group exposed to the highest level of BHA after 64 weeks
    exposure at which time 30% of the animals showed this effect.
    Papillomas were observed in mice at both concentrations at week 80
    and were found in 10-40% of the animals thereafter. Carcinomas were
    detected beginning with week 88, but not in numbers significantly
    different from concurrent controls (a total of one carcinoma among
    all the low level groups and two among the high). The carcinomas
    seen were all well-differentiated. The authors state that the
    number of carcinomas is significant if compared to their historical
    control data (0 tumours in 244 mice) (Masui et al., 1986b; Ito
    et al., 1986b; Ito & Hirose, 1987).


         Male Fischer 344 (F344) rats were divided into groups of 5 and
    were fed diets (corn oil, pellets, or powder) containing 0, 0.1,
    0.25, 0.5, 1 or 2% BHA. After 9 or 27 days, they were sacrificed
    and their forestomachs examined histologically for proliferative
    changes in the squamous epithelium (Clayson et al., 1986). Groups
    of 5-15 rats on the same diet regimen were injected with
    radioactively labelled thymidine just prior to sacrifice and their

    forestomach squamous epithelia examined for incorporation of the label
    (the labelling index). Thymidine is taken up by replicating cellular
    DNA, the presence of which is indicative of induced cell proliferation.
    No effects on the labelling index were seen in rats fed 0.25% BHA or
    less for 9 days. Histopathologically, hyperplasia was observed only
    at 0.5% BHA and above, with the size of the forestomach area
    involved being proportional to the dose. After 9 days of exposure
    to 2% BHA in pellets, a localized four-fold thickening of the
    mucosa was found along the lesser curvature of the forestomach.
    Papillae and irregularly spaced rete ridges, acanthosis and
    hyperkeratosis were observed. Many mitotic figures were seen in the
    otherwise normal-appearing basal layers. There was also acute
    inflammatory cell infiltration in the underlying layers. After
    27 days, the now six-fold thickening had spread widely, being
    visible most markedly adjacent to the forestomach-fundic stomach
    junction. After 9 and 27 days with a pellet diet containing
    2% BHA, the labelling index had been increased approximately
    eight-fold in the pre-fundic region. With BHA given via corn oil,
    the labelling index after 27 days was only increased four-fold;
    however, the thickening had increased over twelve-fold in the
    prefundic region. In the mid-stomach region, the effect of
    corn oil vs. pellet feeding of BHA on the labelling index was
    similar (two-fold), while thickening was less pronounced with corn
    oil (four-fold vs. nine-fold). There were no significant differences
    between the effects of BHA given by pellet and given by powdered

         Male F344 rats in groups of 5 were exposed to 2% BHA in a
    powdered diet for 4 weeks in another study (Hirose et al.,
    1987c). Severe hyperplasia was observed mainly in the prefundic
    region of the stomach near the esophageal orifice. The forestomach
    showed whitish-colored thickening of the epithelium around the
    orifice of the esophagus and the limiting ridge of the forestomach,
    while focal patchiness was observed in the central region.
    Significant body weight gain reduction was also observed in these
    animals, although the liver weight was significantly increased
    compared to controls.

         In a third study (Clayson et al., 1986), groups of 5 male
    F344 rats were fed 0, 0.1, 0.25, 0.5 or 2% BHA (powder) for 13
    weeks. Body weight gain was seen to be significantly reduced in
    animals fed 2% BHA. Proliferative lesions also only developed in
    the forestomach epithelium of the 2% group. In these animals, there
    was thickening in the squamous epithelium and downward proliferation
    of the basal cells. Papillae and rete pegs as well as hyperkeratosis
    were also observed. The muscular layer of the forestomach, however,
    was normal. The labelling index was increased in animals exposed to
    no less than 0.5% BHA, being 2.5x normal after 9 days and 5.3x after
    91 days in the animals exposed to 2% BHA. One week after animals were
    removed to normal for all exposure groups, but the mucosal lesions
    reverted more slowly, still being visible even 9 weeks after cessation
    of exposure.

         As a follow up to this, Clayson et al., (1986) examined the
    forestomachs of rats that had been fed 2% BHA for 3 months and then
    basal diet for 12 months, and BHA for 6 months followed by basal
    diet for 9 months. The forestomachs of both of these groups were
    histologically nearly normal, with the 6 month treatment group
    showing a few downward projections from the normal-appearing
    epithelium. Two rats that had been exposed to 2% BHA for 12 months
    and then basal diet for 3 months, however, had squamous cell
    carcinoma of the forestomach, while others had papillary growths
    showing a high rate of proliferation as shown by thymidine

         In another study (Masui et al., 1986a; Ito & Hirose, 1987),
    10 male F344 rats were given a pellet diet containing 2% BHA for 24
    weeks. Another 20 received the same diet for 24 weeks and then were
    given food without BHA for 72 weeks to assess the effect of
    removing animals from BHA exposure. The forestomach of rats given
    BHA for 24 weeks showed epithelial thickening, especially at the
    limiting ridge. Only slight thickening, however, was seen near the
    limiting ridge in animals exposed to BHA and then removed from
    exposure for 72 weeks. Rats exposed to BHA for 24 weeks developed
    hyperplasia and papillomas in their forestomachs. These changes
    included upward proliferation with thickening of the stratified
    squamous epithelium often with interstitial proliferation, and
    downward proliferation of basal cells forming elongated ridges.
    Acute inflammatory reactions in the lamina propria or submucosa
    were also seen. In animals whose exposure was discontinued, upward
    proliferation of hyperplasia and papillomas had completely
    disappeared. Downward proliferation of basal cells persisted,
    however, in all rats, and papillomas were observed in 3 of the rats
    examined. In this group, however, no inflammation was observed, no
    dysplastic changes were seen in basal cells and no cancer was

         In a similar study (Ito et al., 1986a; Tamano et al.,
    1986), other workers reported that after 104 weeks exposure to 0,
    0.125, 0.25, 0.5, 1 or 2% BHA given by powdered diet to groups of
    50 male F344 rats, body weight gains over the period of observation
    were depressed in a dose-related manner, being statistically
    significant in rats receiving at least 0.5% BHA, even though there
    was no significant difference in food intake among any of the
    groups and non outward clinical symptoms. No significant pathology
    compared to controls was seen in animals fed BHA at any site other
    than the forestomach epithelium and significant forestomach damage
    was only seen in animals exposed to more than 0.5% BHA. This damage
    took the form of extensive raised lesions and was expressed in a
    dose-related manner with the most severely affected region being at
    the limiting ridge. At the end of the 104 week exposure period,
    100% of the animals exposed to 2% BHA had hyperplasia visible in
    the forestomach, 100% had papillomas, and 22% also displayed
    squamous cell carcinoma. These carcinomas were well differentiated,

    displaying keratinisation. No metastases were discovered, however.
    Animals exposed to 1% BHA showed 88% hyperplasia, 20% papilloma,
    but no carcinoma was seen. For rats exposed to 0.5%, 0.25% and 0.125%
    BHA, 32%, 14% and 2%, respectively, had hyperplasia visible, but no
    more severe lesions were reported.

         This work was confirmed by a study in which male F344 rats
    were exposed to 1 or 2% BHA in a powdered diet for 104 weeks with
    animals sacrificed in groups of 10 every 8 weeks during the course
    of the study (Masui et al., 1986b; Ito et al., 1986b).
    Hyperplasia of the forestomach was observed in a few animals of
    both treatment groups at the first observation time, 8 weeks after
    the start of the exposure. It was present in almost all animals
    examined from the high level exposure group from 16 weeks exposure
    onward and from 40 weeks exposure onward for the low level exposure
    group. In the high level group, papillomas were first seen at week
    8 and were present in virtually all animals examined in this group
    week 32. No papillomas were seen in the low level group for the
    first 48 weeks, but they were seen in 80-90% of all animals
    examined after week 56. The first animal with squamous cell
    carcinoma, seen only in the animals exposed to 2% BHA, was observed
    at week 48 and a second animal was observed at week 80, with the
    incidence slowly increasing thereafter so that by 104 weeks, 14% of
    the animals examined displayed this pathology. No changes of any
    kind were seen in the glandular stomach of any animal of any group.

         A 104 week exposure of male and female F344 rats to 0.5 or 2%
    BHA in pellets was also reported (Ito et al., 1982; Ito et al.,
    1983; Ito et al., 1985; Ito & Hirose, 1987). There were
    approximately 50 rats per exposure group. The mean body weights of
    both sexes, were reduced in those animals given 2% BHA with the
    difference being significant after 16 weeks on the diet. At the end
    of the exposure period, 100% of the males and 98% of the females
    exposed to 2% BHA displayed hyperplasia in the forestomach. By
    comparison, only 26% of the males and 20% of the females exposed to
    0.5% BHA had hyperplasia. For papillomas, 100% of the males and 96%
    of the females exposed to 2% BHA displayed this type of lesion,
    while 2% of the animals of both sexes exposed to 0.5% BHA. had this
    observation. Squamous cell carcinoma was seen only in animals
    exposed to 2% BHA and was observed in 35% of the males and 30% at
    that dose level. Grossly, most tumours were seen in the area of the
    limiting ridge of the forestomach. Tumours were greyish-white with
    nodules. Histologically, squamous cell carcinomas were sometimes
    well-differentiated and sometimes not. Those that were well-
    differentiated were keratinized with nuclear atypia and many
    mitotic figures present. Infiltration into the muscular layer and
    fatty tissue was seen. Three cases of metastases were observed. No
    significant incidence of tumours in any other organ were seen in
    any group. These diagnoses were also independently confirmed (Moch,

         Groups of 5-10 male and female Wistar rats were fed 2% BHA in
    a powdered diet for 1, 2 or 4 weeks (Altmann et al., 1985;
    Altmann et al., 1986). After 1 week, epithelial damage with mild
    hyperplasia and hyperkeratosis, and blood-filled cystic swellings
    were observed in the forestomach and livers were markedly enlarged.
    Progression of the hyperplasia and hyperkeratosis was seen after 2
    weeks. Body weight of those fed BHA was also significantly
    depressed. At 4 weeks, severe diffuse hyperplasia, acanthosis and
    hyperkeratosis in the forestomach mucosa, with the most pronounced
    lesions in the area of the limiting ridge, were seen. The effects
    seen, including weight differences, almost completely regressed,
    however, after an additional 4 weeks without BHA. In a second
    experiment, groups of 3 male rats received intubations of 1 g
    BHA/kg bw/day in arachis oil for 1, 2, 4, 8, 16 or 32 days. After
    only one day, increased mitotic activity, and, after 2 days, mild
    hyperplasia were seen in the forestomach. Inflammatory responses
    and superficial effects were not prominent, however, and the
    authors believe that the hyperplasia of the squamous epithelium did
    not result from the initial damage and subsequent regenerative
    activity. After the fourth intubation, the entire wall of the
    forestomach was thickened, wrinkled and swollen. After 8 days, the
    hyperplastic changes regressed. With the arachis oil gavage, the
    lesions appeared remote from the limiting ridge. Four weeks
    following the cessation of BHA intubation, the forestomach showed
    almost complete regression with only mild hyperplasia remaining. At
    no time during any of these experiments were changes seen in the
    glandular stomach or esophagus and there were no differences
    between the sexes.

         In a 90 day feeding study (Altmann et al., 1986), groups of
    10 male and 10 female Wistar rats received 0, 0.125, 0.5 or 2% BHA
    in crystalline form. The highest concentration led to marked
    hyperkeratosis and hyperplasia with epithelial dysplasia in some
    basal areas of the forestomach. Less pronounced lesions were seen
    at 0.5% and only mild lesions were seen in the 0.125% group. In a
    second experiment, 0, 0.025, 0.125 or 2% BHA was administered to
    groups of 20 male or 20 female rats dissolved in arachis oil.
    Again, pronounced hyperplasia was seen in the forestomach in the
    high dose group. None of the lower dose levels of BHA had any
    effect. In only one of the rats was the esophagus affected. In a
    study of reversibility of BHA damage, groups of 5 male and 5 female
    Wistar rats fed 2% BHA for 6, 12 or 15 months showed almost
    complete reversal of severe lesions of the forestomach after
    withdrawal from BHA for 7 months.

         In another study (Takahashi et al., 1986), male Wistar rats
    (10 per group) were fed 1 or 2% BHA (powder) for 32 weeks. A weight
    gain retardation in those animals receiving BHA was reported as
    were thickenings in the squamous cell mucosa of the forestomach as
    well as the presence of squamous cell papillomas. Neoplasms in
    animals receiving 2% BHA occupied most of the forestomach. These
    lesions villous nodules that were greyish-white in color. The
    surface epithelium showed hyperkeratosis with necrosis at the

    superficial layer and long processes of squamous cell epithelium.
    The incidence of papillomas was 100% in this group. Four cases
    (20% incidence) ofdownward growth into the submucosa were also
    seen. Animals on 1% BHA had single or multiple polypoid tumours
    in the forestomach with a 40% incidence of papillomas. No lesions
    in the glandular stomach or duodenum were seen in any exposure group.

         The development of forestomach lesions was found to be
    significantly quickened when Wistar rats in groups of 10 were
    subjected to a 2/3 partial hepatectomy and then fed 2% BHA in a
    powdered diet (Abraham et al., 1986). Under these conditions
    carcinomas were seen after only 3 months. No body weight gain
    depression was reported in this study, however. While only slight
    hyperplasia was seen in the forestomachs of animals after receiving
    BHA alone, those that were fed BHA following partial hepatectomy
    showed grossly visible tumours of the forestomach. In these latter
    animals, the forestomach mucosa was thickened and white in color
    with confluent nodular masses. All ten of the animals exposed to 2%
    BHA following partial hepatectomy displayed hyperplasia and had
    papillomas with marked hyperkeratosis. Half of the animals showed
    carcinoma as well. The carcinomas were well differentiated with
    dysmorphic changes, nuclear atypia and mitotic activity.
    Infiltration of the carcinoma in the muscle layer and fatty tissue
    was seen. Granulocyte, lymphocyte and macrophage infiltration of
    the submucosa was also reported. All other organs were normal in

         A group of 30 male Sprague-Dawley rats was given 1% BHA for 3
    months (Newberne et al., 1986). At the end of this exposure
    period, 66% of the animals had hyperplasia of the forestomach, 26%
    also showed papillomas, and 6% also had carcinomas. The labelling
    index showing cells of the forestomach that were undergoing active
    DNA synthesis in those animals exposed to BHA was over 11 times
    higher than in control animals. In a separate experiment, animals
    administered BHA by gavage were found to be even more severely
    affected than those fed BHA in the diet. In the former group 12/18
    animals were found to have carcinomas while only 2/20 in the latter
    had similar problems.


         A group of Syrian golden hamsters received a diet of 2% BHA
    for 28 days (Altmann et al., 1986). At the end of this period,
    macroscopic lesions in the forestomach of the hamsters were seen
    but were reported to be different from those in rats and mice. No
    hyperkeratosis was seen in the hamsters macroscopically, but the
    mucosae were less elastic and sometimes thickened and wrinkled.
    Macroscopically, mild hyperplasia and hyperkeratosis were observed.
    These were more pronounced in the females.

         In another study (Hirose et al., 1986d; Ito et al.,
    1986b), groups of approximately 30 male Syrian golden hamsters were
    fed 1% BHA in a powdered diet for 1 or 3 days or 1, 2, 3, 4 or 16
    weeks after which they were injected with radioactive thymidine (to
    enable a labelling index to be ascertained), sacrificed and
    examined. Animals exposed to BHA showed a body weight decrease
    compared to controls, but a liver weight increase. Focal thickening
    of the forestomach epithelium with or without ulceration and
    occasional covering with a dense keratin-like greyish-white
    substance was seen in hamsters exposed for at least 1 week. There
    were no abnormalities in any other organs examined. Severity of
    hyperplasia gradually increased with time of exposure. Papillomas
    were seen beginning with week 4. Neutrophil infiltration was also
    observed. An increase in the labelling index paralleled the
    severity of the lesions observed.

         In a similar study (Hirose et al., 1986b), fifteen hamsters
    were given a powdered diet containing 1% BHA for 20 weeks with 3
    being injected with radioactive thymidine just prior to sacrifice
    for labelling index determination. Again, body weight depression
    was seen in animals fed BHA. Thickening of the forestomach
    epithelium with white, keratin-like substance was also noted. All
    of the animals displayed severe hyperplasia with 60% additionally
    showing papillomatous lesions. The labelling index in the
    forestomach was almost three times that in control animals. No
    changes were seen in any other organ examined.

         These observations were extended to 104 weeks exposure to 1 or
    2% BHA (powder) in the next study (Masui et al., 1986b; Ito &
    Hirose, 1987). Groups of 10 male hamsters were observed at 8 week
    intervals. In these animals, hyperplasia of the forestomach was
    observed in all animals in both exposure groups starting at 8 weeks
    of exposure. Papillomas were likewise seen in almost all animals in
    the high level exposure group beginning at week 8, and in the low
    level group from week 16. Hyperkeratosis and downward growth of the
    tumour into the submucosa was often observed. Squamous cell
    carcinomas were observed in both treated groups beginning in week
    64. They were well differentiated and an invasion of the liver was
    seen in one case. After 104 weeks of exposure, hyperplasia was seen
    in 100%, 96% and 17% of the 2%, 1% and control animals, respectively.
    Likewise, 95%, 98% and 0% of these groups showed papillomas, and 10%,
    7% and 0%, respectively, carcinomas.

    Guinea pigs

         Guinea pigs fed a diet containing 1% BHA for 20 months did not
    exhibit any gross changes in the stomach (Ito & Hirose, 1987).


         Groups of 29 male and 30 female beagle dogs (also lacking a
    forestomach), were fed 0, 1.0 or 1.3% BHA for 180 days (Ikeda
    et al., 1986; Moch, 1986). Food consumption and body weight gain
    were reduced in the animals receiving the highest level of BHA and
    there was a liver weight increase in animals receiving BHA at
    either dose. Ultrastructural examination of the livers of animals
    on BHA showed proliferation of smooth endoplasmic reticulum and
    hepatocytic cytoplasmic myelinoid bodies. Light and electron
    examination of the stomach and lower esophagus showed no
    proliferative or hyperplastic lesions and no change in the number
    of cells.

         BHA at 0.25, 0.5 or 1.0% was fed to groups of 3-4 male or
    female beagle dogs for six months in a second study (Tobe et al.,
    1986). A dose-related retardation of growth was reported. Liver
    weights were increased but no histopathological changes were seen
    in that organ. There were no alterations in the stomach mucosa and
    there was no induced change in the mitotic index of the squamous
    epithelium of the distal esophagus.

         In a third report, adult beagle dogs fed a diet containing
    0-100 mg BHA/kg bw/day for one year exhibited no histopathological
    changes in any tissue examined (Ito & Hirose, 1987).


         Three different experiments were conducted involving pregnant
    young adult specific pathogen free (SPF) Danish Landrace gilts
    (Olsen, 1983; Wurtzen & Olsen, 1986; Moch, 1987). Combining the
    data of the three studies, groups of 9-13 pigs were fed pellets
    containing 0, 0.5%, 1.9% or 3.7% BHA (0, 50, 200, or 400 mg BHA/kg
    bw/day, respectively) for the first 110 days of pregnancy.
    Epithelial changes in the esophageal part of the stomach were
    similar for test and control groups. No papillomas and no changes
    in the oesophageal part of the stomach were similar for test and
    control groups. No papillomas and no changes in the glandular part
    of the stomach were reported. Linear yellow-brown, rough epithelium
    was seen in the entire length of the esophagus in a few pigs in the
    middle and high dose groups. In the first of the three experiments,
    one of the three animals in the middle dose group and two of three
    in the high dose group had esophageal lesions seen at gross
    necropsy. All three of these animals had histopathological
    examination of the esophageal lesions seen at gross necropsy and
    were diagnosed as having esophageal hyperplasia. In addition, one
    other, animal in the middle dose group had esophageal hyperplasia
    noted microscopically, although no lesion was seen at time of gross
    necropsy. In the second experiment, one of the three animals in the
    high dose groups had an esophageal lesion noted at gross necropsy.
    In the second experiment, one of the three animals in the high dose
    groups had an esophageal lesion noted at gross necropsy. No

    histological examination of this animal was performed, however. In
    the third experiment, one of the seven animals in the middle dose
    group and one of four animals in the high dose group had a lesion
    of the esophagus seen at gross necropsy. None of these animals had
    histological examination of the reported lesions. The authors
    concluded on the basis of the one middle dose level pig and the two
    high dose level pigs of the first study that BHA might have an
    effect on esophageal epithelium.


         Groups of 8 female cynomogus monkeys were given 0, 125 or
    500 mg BHA/kg bw by gavage in corn oil 5 times/week for 20 days,
    after which the dose was halved and continued for a total of 85 days
    (Iverson et al., 1986; Ito & Hirose, 1987). Although there were
    transitory dose-related changes in blood clinical values during the
    first part of the exposure, no value differed from the normal range
    nor were any abnormal fibroscopic observations made during the course
    of the exposure. Histopathological examination at the end of the
    exposure period showed no treatment related effects, although the
    mitotic index in the basal cell layer of the squamous epithelium of
    the distal esophagus was elevated by a factor of 1.9. Liver weight was
    also increased in animals exposed to BHA in a dose-related manner.

    Special studies on the effect of the molecular structure

         In Syrian golden hamsters (Hirose et al., 1986d), the
    observation was made that, in parallel with its relative power as
    an antioxidant, 3- tert-butylhydroxyanisole (3-BHA) is
    significantly better able to cause biological damage than
    2- tert-butylhydroxyanisole (2-BHA). Using radioactive labelling
    in F344 rats, however, 2-BHA has been shown to be incorporated into
    cells of the forestomach after 1 week of treatment at approximately
    twice the level of 3-BHA (Hirose et al., 1987b). Food grade BHA
    is primarily composed of 3-BHA.

         Three papers have reported investigations attempting to
    discover which part(s) of the chemical structure of BHA is critical
    to its actions. By comparing the ability of 13 structurally related
    phenolic compounds to cause lesions in the forestomach of Syrian
    golden hamsters after 20 weeks exposure, one report (Hirose et
    al., 1986b) concluded that the phenolic compounds must have both
    at least one hydroxy and exactly one  tert-butyl substituent to
    have strong activity in inducing forestomach tumours. Other
    structural variations produce their own spectra of biological
    consequences (Ito & Hirose, 1987). In the other two papers on this
    subject (Altmann et al., 1985; Altmann et al., 1986) in which
    Wistar rats were fed one of 12 chemicals related to BHA or BHA
    itself, the conclusion is made that the critical part of the BHA
    structure is the methoxy group.

    Special studies on potentiation or inhibition of carcinogenicity

         The ability of BHA to interact with known carcinogens has been
    studied extensively (Lindenschmidt et al., 1987; Hirose et al.,
    1986c; Ito et al., 1986b; Tsuda et al., 1987; Fukushima et
    al., 1987; Hirose et al., 1986a; Tsuda et al., 1984; Newberne
    et al., 1986; Takahashi et al., 1986; Moore et al., 1986; Ito
    et al., 1985; Williams et al., 1986; Masui et al., 1986c; and
    Chung et al., 1986). Several strains of rats and mice of both
    sexes have been fed BHA at various concentrations for various durations
    either before, after, or before, during and after treatment by various
    routes with various concentrations of several carcinogens administered
    for various lengths of time. The net effect of BHA plus the carcinogen
    has been examined in thirteen organs. What has been found is that BHA
    potentiates, inhibits, or has no effect on the ability of the
    carcinogen to act, depending on the exact chemical being tested and
    the circumstances under which the test is being conducted.


         Studies have been conducted that provide additional
    information on the proliferative changes observed in the
    forestomach of rats fed BHA. No new studies on the effect of BHA on
    the stomach and esophagus of species that do not have a
    forestomach, e.g., pigs or monkeys, were conducted. The data show
    that continuous exposure of the rat forestomach to 2% BHA in the
    diet for 6-12 months is necessary to produce squamous cell
    carcinoma. The data also show that the induction of mild
    hyperplasia can occur at levels of 0.125% of BHA in the diet but
    not at a level of 0.1% BHA. After reevaluating the data in pigs, it
    was concluded that the evidence that BHA produces hyperplasia in
    the esophagus of the pig is questionable. Moreover, these
    presumptive effects in pigs were reported to occur at levels of BHA
    significantly higher than those that produced the confirmed effects
    in the rat forestomach. Considering the absence of any significant
    adverse effects in two dog studies, it was concluded that further
    investigations in animals without forestomachs are not required.
    The human relevance of the rat studies, while inherently
    questionable because the target tissue in the rat has no human
    counterpart, cannot readily be ignored. Based on the dose
    dependence and reversibility of the lesions produced in the rat,
    discussed previously in the 1986 monograph (Annex 1, reference 74)
    and since confirmed by more recent studies, it was concluded that
    an ADI can be established.


    Level causing no toxicological effect

         Rat: 0.1% in the diet, equivalent to 50.0 mg/kg bw/day.

    Estimate of acceptable daily intake for man

         0-0.5 mg/kg bw.

    Further work or information


         Submission of the results of an ongoing reproduction study.


    Abraham, R., Benitz, K.F., Patil, G. & Lyon, R. (1986). Rapid
    induction of forestomach tumours in partially hepatectomized Wistar
    rats given butylated hydroxyanisole.  Experimental and Molecular 
     Pathology, 44, 14-20.

    Altmann, H.-J., Grunow, W., Wester, P.W. & Mohr, U. (1985).
    Induction of forestomach lesions by butylhydroxyanisole and
    structurally related substances.  Archives of Toxicology,
     Supplement 8, 114-116.

    Altmann, H.-J., Grunow, W., Mohr, U., Richter-Reichhelm, H.B. &
    Wester, P.W. (1986). Effects of BHA and related phenols on the
    forestomach of rats.  Food Chemistry and Toxicology, 24, 1183,

    Ansari, G.A.S. & Hendrix, P.Y. (1985). Tissue distribution and
    pharmacokinetics of 3- t-[ methyl-14C]butyl-4-hydroxy-anisole in
    rats.  Drug Metabolism and Deposition, 13, 535-541.

    Chung, F.-L., Wang, M., Carmella, S.G. & Hecht, S.S. (1986).
    Effects of butylated hydroxyanisole on the tumorigenicity and
    metabolism of N-nitrosdimethylamine and  N-nitrosopyrrolidine
    in A/J mice.  Cancer Research, 46, 165-168.

    Clayson, D.B., Iverson, F., Nera, E., Lok, E., Rogers, C. &
    Rodrigues, C. (1986). Histopathological and radioautographical
    studies on the forestomach of F344 rats treated with butylated
    hydroxyanisole and related chemicals.  Food Chemistry and
     Toxicology, 24, 1171-1182.

    Cummings, S.W., Ansari, G.A.S., Guengerich, F.P., Crouch, L.S. &
    Prough, R.A. (1985). Metabolism of 3- tert-butyl-4-hydroxyanisole
    by microsomal fractions and isolated rat hepatocytes.  Cancer 
     Research, 45, 5617-5624.

    De Long, M.J., Prochaska, H.J. & Talalay, P. (1985). Tissue-
    specific induction patterns of cancer-protective enzymes in mice by
     tert-butyl-4-hydroxyanisole and related substituted phenols.
     Cancer Research, 45, 546-551.

    deStafney, C.M., Prabhu, D.G., Sparnins, V.L & Wattenberg, L.W.
    (1986). Studies related to the mechanism of 3-BHA-induced neoplasia
    of the rat forestomach.  Food and Chemical Toxicology, 24,

    Fukushima, S., Sakata, T., Tagawa, Y., Shibata, M.-A., Hirose, M.
    & Ito, N. (1987). Different modifying response of butylated
    hydroxyanisole, butylated hydroxytoluene, & other antioxidants in
     N,N-dibutylnitrosamine esophagus and forestomach carcinogenesis
    of rats.  Cancer Research, 47, 2113-2116.

    Hirose, M., Hagiwara, A., Masui, T., Inoue, K. & Ito, N. (1986a).
    Combined effects of butylated hydroxyanisole and other antioxidants
    in induction of forestomach lesions in rats.  Cancer Letters, 30,

    Hirose, M., Inoue, K., Asamoto, M., Tagawa, Y. & Ito, N. (1986b).
    Comparison of the effects of 13 phenolic compounds in induction of
    proliferative lesions of the forestomach and increase in the
    labelling indices of the glandular stomach and urinary bladder
    epithelium of Syrian golden hamsters.  Carcinogenesis, 7,

    Hirose, M., Masuda, A., Inoue, T., Fukushima, S. & Ito, N. (1986c).
    Modification by antioxidants and  p,p'-diaminodiphenylmethane of
    17,12-dimethylbenz[a]anthracene-induced carcinogenesis of the
    mammary gland and ear duct in CD rats.  Carcinogenesis, 7,

    Hirose, M., Masuda, A., Kurata, Y., Ikawa, E., Mera, Y. & Ito, N.
    (1986d). Histologic and autoradiographic studies on the forestomach
    of hamsters treated  with  2- tert-butylated  hydroxyanisole,
    3- tert-butylated hydroxyanisole, crude butylated hydroxyanisole,
    or butylated hydroxytoluene.  Journal of the National Cancer
     Institute, 76, 143-149.

    Hirose, M., Asamoto, M., Hagiwara, A., Ito, N., Kaneko, H., Saito,
    K., Takamatsu, Y., Yoshitake, A. & Miyamoto, J. (1987a). Metabolism
    of 2-and 3- tert-butyl-4-hydroxyanisole (2- and 3-BHA) in the rat
    (II): Metabolism in forestomach and covalent binding to tissue
    macromolecules.  Toxicology, 45, 13-24.

    Hirose, M., Hagiwara, A., Inoue, K., Sakata, T., Ito, N., Kaneko,
    H., Yoshitake, A. & Miyamoto, J. (1987b). Metabolism of 2- and
    3- tert-butyl-4-hydroxyanisole (2- and 3-BHA) in the rat (I):
    Excretion of BHA in urine, feces and expired air and distribution
    of BHA in the main organs.  Toxicology, 43, 139-147.

    Hirose, M., Masuda, A., Imaida, K., Kagawa, M., Tsuda, H. & Ito, N.
    (1987c). Induction of forestomach lesions in rats by oral
    administrations of naturally occurring antioxidants for 4 weeks.
     Japanese Journal of Cancer Research, 78, 317-321.

    Ikeda, G.J., Stewart, J.E., Sapienza, P.P., Peggins III, J.O.,
    Michel, T.C., Olivito, V., Alam, H.Z. & O'Donnel Jr., M.W. (1986).
    Effect of subchronic dietary administration of butylated
    hydroxyanisole on canine stomach and hepatic tissues.  Food 
     Chemistry and Toxicology, 24, 1201-1221.

    Ito, N., Hagiwara, A., Shibata, M., Ogiso, T. & Fukushima, S.
    (1982). Induction of squamous cell carcinoma in the forestomach of
    F344 rats treated with butylated hydroxianisole.  Japanese Journal 
     of Cancer Research, 73, 332-334.

    Ito, N., Fukushima, S., Hagiwara, A., Shibata, M. & Ogiso, T.
    (1983). Carcinogenicity of butylated hydroxyanisole in F344 rats.
     Journal of the National Cancer Institute, 70, 343-352.

    Ito, N., Fukushima, S. & Tsuda, H. (1985). Carcinogenicity and
    modification of the carcinogenic response by BHA, BHT, and other
    antioxidants.  CRC Critical Reviews of Toxicology. 15, 109-150.

    Ito, N., Fukushima, S., Tamano, S., Hirose, M. & Hagiwara, A.
    (1986a). Dose response in butylated hydroxyanisole induction of
    forestomach carcinogenesis in F344 rats.  Journal of the National 
     Cancer Institute, 77, 1261-1265.

    Ito, N., Hirose, M., Fukushima, S., Tsuda, H., Tatematsu, M. &
    Asamoto, M. (1986b). Modifying effects of antioxidants on chemical
    carcinogenesis.  Toxicological Pathology, 14, 315-323.

    Ito, N. & Hirose, M. (1987). The role of antioxidants in chemical
    carcinogenesis.  Japanese Journal of Cancer Research, 78, 

    Iverson, F., Truelove, J., Nera, E., Lok, E., Clayson, D.B. & Wong,
    J. (1986). A 12-week study of BHA in the cynomolgus monkey.  Food 
     and Chemical Toxicology, 24, 1197-1200.

    Lindenschmidt, R.C., Tryka, A.F. & Witschi, H. (1987). Modification
    of gastrointestinal tumour development in rats by dietary butylated
    hydroxytoluene.  Fundamental and Applied Toxicology, 8, 474-481.

    Masui, T., Asamoto, M., Hirose, M., Fukushima, S. & Ito, N.
    (1986a). Disappearance of upward proliferation in rat forestomach
    papillomas induced by butylated hydroxyanisole.  Japanese Journal 
     of Cancer Research, 77, 854-857.

    Masui, T., Hirose, M., Imaida, K., Fukushima, S., Tamano, S. & Ito,
    N. (1986b). Sequential changes of the forestomach of F344 rats,
    Syrian golden hamsters, and B6C3F1 mice treated with butylated
    hydroxyanisole.  Japanese Journal of Cancer Research, 77,

    Masui, T., Tsuda, H., Inoue, K., Ogiso, T. & Ito, N. (1986c).
    Inhibitory effects of ethoxyquin, 4,4'-diaminodiphenylmethane and
    acetaminophen on rat hepatocarcinogenesis.  Japanese Journal of 
     Cancer Research, 77, 231-237.

    Moch, R.W. (1986). Pathology of BHA- and BHT-induced lesions.
     Food and Chemical Toxicology, 24, 1167-1169.

    Moch, R.W. (1987). Pathology Report, US Food and Drug
    Administration, "Butylated hydroxyanisole (BHA). Histopathology
    review of selected microslides from Danish landrace pigs", Office
    of Toxicological Sciences Center for Food Safety and Applied

    Moore, M.A., Thamavit, W., Tsuda, H. & Ito, N. (1986). The
    influence of subsequent dehydroepiandrosterone, diaminopropane,
    phenobarbital, butylated hydroxyanisole and butylated hydrotoluene
    treatment on the development of preneoplastic and neoplastic
    lesions in the rat initiated with di-hydroxy-di- N-propyl
     nitrosamine. Cancer Letters, 30, 153-160.

    Newberne, P.M., Charnley, G., Adams, K., Cantor, M., Roth, D. &
    Supharkarn V. (1986). Gastric and oesophageal carcinogenesis:
    Models for the identification of risk and protective factors.  Food 
     and Chemical Toxicology, 24, 1111-1119.

    Olsen, P. (1983). The carcinogenic effect of butylated hydroxyanisole
    on the stratified epithelium of the stomach in rat versus pig.
     Cancer Letters, 21, 115-116.

    Takahashi, M., Furukawa, F., Toyoda, K., Sato, H., Hasegawa, R. &
    Hayashi, Y. (1986). Effects of the four antioxidants on
    N-methyl- N'-nitrosoguanidine initiated gastric tumour
    development in rats.  Cancer Letters, 30, 161-168.

    Takizawa, Y., Matsuda, Y. & Yamasita, J. (1985). The absorption and
    excretion of butylated hydroxyanisole in beagle dogs.  Toxicology 
     Letters, 27, 27-34.

    Tamano, S., Hirose, M., Shibata, M.-A., Tagawa, Y., Fukushima, S.
    & Ito, N. (1986). Successful dose response of forestomach tumour in
    F344 rats induced by butylated hydroxyanisole (BHA).  Toxicology 
     Letters, 31, 207.

    Tobe, M., Furuya, T., Kawasaki, Y., Naito, K., Sekita, K.,
    Matsumoto, K., Ochiai, T., Usui, A., Kokubo, T., Kanno, J. &
    Hayashi, Y. (1986). Six-month toxicity study of butylated
    hydroxyanisole in beagle dogs. Food and  Chemical Toxicology, 24,

    Tsuda, H., Sakata, T., Shirai, T., Kurata, Y., Tamano, S. & Ito, N.
    (1984).  Modifications  of   N-methyl- N-nitrosourea  initiated
    carcinogenesis in the rat by subsequent treatment with antioxidants,
    phenobarbital and ethinyl estradiol.  Cancer Letters, 24, 19-27.

    Tsuda, H., Ogiso, T., Hasegawa, R., Imaida, K., Masui, T. & Ito, N.
    (1987). Inhibition of neoplastic development in rat liver, kidney,
    oesophagus and forestomach by 4,4'-diaminodiphenylmethane
    administration.  Carcinogenesis, 8, 719-722.

    Williams, G.M., Tanaka, T. & Maeura, Y. (1986). Dose-related
    inhibition of aflatoxin B1 induced hepatocarcinogenesis by the
    phenolic antioxidants butylated hydroxyanisole and butylated
    hydroxytoluene.  Carcinogenesis, 7, 1043-1050.

    Wurtzen, G. & Olsen, P. (1986). BHA study in pigs.  Food and 
     Chemical Toxicology, 24, 1229-1233.

    See Also:
       Toxicological Abbreviations
       Butylated hydroxyanisole (WHO Food Additives Series 5)
       Butylated hydroxyanisole (WHO Food Additives Series 10)
       Butylated hydroxyanisole (WHO Food Additives Series 21)