IPCS INCHEM Home

    ß-CYCLODEXTRIN

    First draft prepared by
    Professor R. Walker
    School of Biological Sciences
    University of Surrey, Guildford U.K.

    1.  EXPLANATION

         ß-Cyclodextrin is a cyclic heptamer composed of seven glucose
    units joined "head-to-tail" by alpha-1,4 links.  It is produced by
    the action of the enzyme, cyclodextrin glycosyl transferase (CGT),
    on hydrolyzed starch syrups.  CGT is obtained from  Bacillus
     macerans, B. circulans or related strains of  Bacillus.

         As a result of its cyclic structure, ß-cyclodextrin has the
    ability to form inclusion compounds with a range of molecules,
    generally of molecular mass of less than 250.  It may serve as a
    carrier and stabilizer of food flavours, food colours and some
    vitamins.  Intake of ß-cyclodextrin from use as a food additive has
    been estimated at 1-1.4 g/day.  Other applications in decaffeination
    of coffee/tea and in reducing the cholesterol content of eggs by
    complexation followed by separation of the complex would make a much
    lower contribution to intakes.

         ß-Cyclodextrin has not been reviewed previously by the Joint
    FAO/WHO Expert Committee on Food Additives.

    2.  BIOLOGICAL DATA

    2.1  Biochemical aspects

    2.1.1.  Absorption, distribution, and excretion

         ß-cyclodextrin is resistant to hydrolysis by acid (Szeftli &
    Budai, 1976), alpha- and ß-amylases and yeast.  It is thus not
    readily digested in the upper gastrointestinal tract by gastric or
    pancreatic enzymes (Jodal  et al. 1984).

         In a preliminary comparative metabolism study in rats, groups
    each of 2 animals were given 14C-labelled alpha-cyclodextrin,
    ß-cyclodextrin or gelatinised potato starch by gavage as
    approximately 2.5 ml of a 2.5% aqueous solution.  Urine, faeces and
    expired air were collected for 17-23 hours after which residual
    radioactivity was assayed in the gastro-intestinal tract (caecum and
    caecal contents separately), liver, kidneys, heart, lung, spleen,
    gonads and residual carcass.  Animals receiving ß-cyclodextrin
    metabolized the compound more slowly than starch, as indicated by
    the time course of elimination of 14CO2 in expired air, but by the
    end of collection the total amount of the dose excreted by this
    route (48.6% and 66.8% of the dose as 14CO2 after 17 and 23 hours
    respectively) was similar to that of the group given starch; 
    excretion of activity in the urine (3.6-5.1%) and faeces (0-5.4%),
    and the residual levels in organs and carcass were also similar in
    the groups given ß-cyclodextrin and starch  (Andersen  et al.
    1963).

         ß-Cyclodextrin, purity >77%, and glucose, uniformly labelled
    with 14C, were administered to rats by gavage as solutions in 20%
    aqueous dextran at dose levels of 13 mg/kg bw for glucose and 36 and
    313 mg/kg bw for ß-cyclodextrin.  In the case of glucose, blood
    levels of radioactivity peaked within 10-30 minutes of dosing
    whereas with the lower dose of ß-cyclodextrin, peak blood levels
    were observed between 4 and 11 hours following administration.  At
    the lower dose level of ß-cyclodextrin, the respired radioactivity,
    as a percentage of the dose, was similar to that after dosing with
    glucose whereas at the higher dose a smaller percentage was
    respired.  In the eighth hour after the high dose (313 mg/kg bw) no
    more than 3-50ppm ß-cyclodextrin was detectable in blood.  It was
    concluded that ß-cyclodextrin is not absorbed to a significant
    extent from the stomach or small intestine of rats but that
    hydrolysis to open chain dextrins/glucose occurs in the large
    intestine by a combination of the action of the gut microflora and
    endogenous amylases.  This process may be saturated, and unabsorbed
    material can be excreted in faeces at high dose levels.  The oxygen
    consumption of slices of rat small intestine was measured in the
    presence of glucose, maltose, starch and ß-cyclodextrin as
    substrates; the rate of oxygen consumption was increased by all

    substrates except ß-cyclodextrin (Szejtli  et al. 1980; Gerlóczy
     et al. 1981; 1985).

         The intestinal absorption, digestibility by the colonic
    microflora, and urinary excretion of ß-cyclodextrin were studied. 
    Using everted sacs of rat small intestine  in vitro and ligated gut
    loops  in vivo, absorption was shown to be slow, concentration-
    dependent, not saturable and not inhibited by phloretin; this
    indicates that a passive transport process is involved.  Rat caecal
    microflora were able to utilize ß-cyclodextrin under anaerobic
    conditions  in vitro, indicating that the compound may be
    hydrolyzed to glucose by bacterial enzymes.  It was concluded by the
    authors that ß-cyclodextrin may be utilized by the rat but only
    indirectly by the activity of the gut flora (Szabo  et al.
    1981a,b).

         Fasted male Sprague-Dawley rats were given a dose of 1 500 mg
    ß-cyclodextrin (about 3 235 mg/kg bw).  Only small amounts (0.6-4%
    of the dose) were excreted in the faeces in a 60 hour period post-
    dosing and negligible amounts (0-0.9% of the dose) remained in the
    gastrointestinal tract.  In a separate experiment using a similar
    dose, a large proportion was shown to be converted to glucose
    between 3 and 8 hours after dosing (Suzuki & Sato, 1985).

         In a 13 week short-term study in beagle dogs (see section
    2.2.2.3) animals received ß-cyclodextrin at dietary levels of 1.25,
    2.5, 5 or 10%, equal to mean daily doses of 570, 1 234, 2 479 or 4
    598 mg ß-cyclodextrin/kg bw.  Twenty-four-hour urine samples were
    analyzed for ß-cyclodextrin at weeks 7 and 13, faeces at week 13 and
    serum was collected 1, 3, and 6 hours after dosing during week 13.

         The urinary concentrations of ß-cyclodextrin at week 7 were
    332±76, 510±163, 1 222±491 and 4 218±1 833 mg/l in the respective
    dose groups; the corresponding values at week 13 were 631±707,
    467±335, 817±395 and 2 206±528 mg/l.  Although, because of the daily
    dietary dosing regime, it is not possible simply to express the
    amount excreted as a percentage of dose, approximately 0.6-1.7% of
    the mean 24-hour dose was excreted in 24-hour urine and the
    percentage was not dependent on dose.  In week 13, faecal
    concentrations of ß-cyclodextrin (based on dry matter) in the
    respective dose groups were 1.3, 1.1, 1.8 and 6.5%.  Serum levels of
    ß-cyclodextrin increased in a dose-dependent manner and with time
    after dosing up to 6 hours.  No ß-cyclodextrin was detected in the
    lowest dose group and only low levels (1 mg/l increasing to 5 mg/l
    in the 2.5% dose group;  serum levels increased from 4 to 8 mg/l in
    1 to 6 hours in the 5% dose group and from 8 mg/l to 46 mg/l in the
    10% group.  These results indicate that a small proportion of the
    dose may be absorbed from the gastrointestinal tract in the dog
    (Smith  et al. 1992).

    2.1.2.  Biotransformation

         Twenty-four out of thirty strains of  Bacteroides isolated
    from the human colon were able to degrade ß-cyclodextrin and utilize
    it as a sole carbon source.  Detailed examinations carried out on
    the dextrinases from two strains,  Bacteroides ovatus 3524 and
     Bacteroides distasonis C18-7 indicated that cyclodextrinase
    activity was predominantly cell bound and inducible in both
    organisms.  The products of hydrolysis of crude cyclodextrinase
    preparations from these two organisms were markedly different, the
    former producing only glucose whereas the latter produced a series
    of maltooligomers (Antenucci & Palmer, 1984).

          In vitro, ß-cyclodextrin was resistant to hydrolysis by
    purified alpha-amylase from  Aspergillus oryzae.  On incubation for
    24 h at an initial ß-cyclodextrin concentration of 15.8 mM, only 17%
    was degraded; the breakdown products were glucose (12%), maltose
    (2%) and maltotriose (3%) (Jodal  et al. 1983).

    2.2.  Toxicological Studies

    2.2.1.  Acute toxicity studies

         The results of acute toxicity studies with ß-cyclodextrin are
    summarized in Table 1.

    Table 1. Results of acute toxicity studies with ß-cyclodextrin.

                                                                 
    Species    Sex     Route      LD50        Reference
                               (mg/kg bw)
                                                                 
    Mouse       M      oral      >3 000       Sebestyén (1980)
    Mouse       F      oral      >3 000       Sebestyén (1980)
    Mouse       M      i.p.      372          Sebestyén (1980)
    Mouse       F      i.p.      331          Sebestyén (1980)
    Mouse       M      s.c.      419          Sebestyén (1980)
    Mouse       F      s.c.      412          Sebestyén (1980)
    Rats        M      oral      >5 000       Sebestyén (1980)
    Rats        F      oral      >5 000       Sebestyén (1980)
    Rats        M      i.p.      373          Sebestyén (1980)
    Rats        F      i.p.      356          Sebestyén (1980)
    Rats      M & F    i.v.      788          Frank  et al. (1976)
    Rats        M      s.c.      >1 000       Sebestyén (1980)
    Rats        F      s.c.      >1 000       Sebestyén (1980)
    Rats      M & F    dermal    >2 000       Sebestyén (1980)
    Rats      M & F    inh       >4.9*        Busch  et al. 1985
    Dog         M      oral      >5 000       Sebestyén (1980)
    Dog         F      oral      >5 000       Sebestyén (1980)
                                                                 
    * milligrams/litre of air for 4 hours

    2.2.2.  Short term toxicity studies

    2.2.2.1.  Mice

         In a repeat-dose study of very limited scope which was not
    reported in detail, mature male mice (number not stated) were given
    daily oral doses of 6 ml of a 1% solution of ß-cyclodextrin for 15
    days.  No effects were observed on body weight or relative liver
    weight.  ß-Cyclodextrin was detected in excreta (unclear whether
    urine and/or faeces) by paper chromatography but could not be
    detected in liver or gastrointestinal tract (limit of detection not
    stated)(Miyazaki  et al. 1979).

    2.2.2.2.  Rats

         In a 13 week oral toxicity study in Long-Evans rats, body
    weight 80-100 g at commencement, ß-cyclodextrin (purity not
    specified) was administered by gavage as a suspension in aqueous 1%
    methylcellulose to groups of 10 male and 10 female animals at daily
    dose levels of 200, 400 or 600 mg/kg bw; controls (15 animals of
    each sex) received an equal volume of 1% methylcellulose solution. 
    Following high mortality (4 males, 3 females) due to misdosing in
    the 400 mg/kg bw group, a repeat experiment was carried out at this
    dose level with a separate control group of 5 animals.  Body weight
    and food intake were recorded weekly and urinalysis, haematological
    and clinical biochemical were performed at termination.  At autopsy,
    heart, lung, liver, kidneys and spleen were weighed, and examined
    macroscopically and histologically together with gonads, stomach,
    intestine, pancreas, adrenals and brain.

         No abnormal clinical signs were observed and there were no
    deaths other than 1 male in the low-dose group and 1 female in the
    high-dose groups, attributed to misdosing.  There were no
    significant treatment-related changes in mean body weight, food
    consumption, or relative organ weights at termination.  No effects
    due to ß-cyclodextrin were observed in haematological parameters
    (Hb, haematocrit, MCH, RBC, total and differential leucocytes) or in
    the clinical biochemical indices ASAT, Alk-P-ase, BUN, bilirubin or
    creatinine;  dose-related but non-statistically significant changes
    were reported for ALAT in females (decreased) and Alk-P-ase
    (decreased in males, increased in females) but these were not
    considered to be of toxicological relevance.  Urinalysis for colour,
    pH, protein, glucose, urobilinogen, bilirubin, ketones and sediment
    gave similar results between treated animals and controls except for
    blood detected in the urine of some females of the low-dose group. 
    There were no significant, dose-related changes in the incidence of
    any lesions in any of the tissues examined histologically but there
    was a high background incidence of parasitic and bacterial
    infection.

         The power of this study was limited by the small number of
    animals used and the incidence of lesions due to infection but,
    within these limits, the NOEL was 600 mg/kg bw/day, the highest dose
    tested (Sebestyén, 1979; Tury, Aobos-Kovacs & Somogyvari, 1979).

         Groups of 17 (19 at the highest dose) male and female Sprague-
    Dawley rats, 4 weeks old at commencement of the study, were given
    ß-cyclodextrin by gavage in aqueous suspension at daily dose levels
    of 0, 100, 400 or 1 600 mg/kg bw  After 3 months administration an
    interim sacrifice was made as follows: Control, 5 males & 4 females;
    100 mg/kg bw/d group, 3 males & 5 females; 400 mg/kg bw/d group, 5
    males & 4 females; 1600 mg/kg bw/d group, 2 males & 4 females.  The
    remaining animals were maintained on the same dosing regime to 6
    months.  Food and water intake and body weights were determined
    weekly.  At termination, urinalysis (pH, protein, glucose, ketone
    bodies, blood, bilirubin, urobilinogen) was carried out on 5 animals
    of each sex in each group; haematological (RBC, WBC, haemoglobin and
    haematocrit) and serum biochemical analyses (protein,
    albumin/globulin, GOT, GPT, Alk-P-ase, BUN, bilirubin, total
    cholesterol and glucose) were carried out on all survivors.  At
    autopsy, the following organs were weighed: brain, pituitary,
    thyroid, thymus, heart, lung, liver, kidneys, adrenals, spleen,
    pancreas, testes or ovaries; these organs and stomach and intestinal
    tract were examined histologically (haematoxylin & eosin).

         A total of 18 animals died during the study as follows:
    100 mg/kg bw/d, 3 males; 400 mg/kg bw/d, 2 females; and 1600 mg/kg
    bw/d, 6 males and 7 females.  It was claimed that these deaths were
    due to misdosing, as no abnormalities were detected other than
    "pneumonia-like" lung pathology.  There was a small decrement of
    weight gain of both sexes in the top-dose group only, otherwise
    weight gain, food and water intake were similar to controls and
    there were no treatment-related effects on organ weights, urinalysis
    or haematological parameters.   Serum biochemical indices generally
    were within the normal range although some significant differences
    from controls were observed, notably a dose-related increase in
    alk-P-ase in males and a decrease in blood glucose in females of the
    top two dose groups.   Gross and histopathological examination did
    not reveal any treatment-related abnormalities.  If the deficit in
    weight gain at the top-dose level is considered an adverse effect in
    the absence of other, pathological, changes, the NOAEL for this
    study is 400 mg/kg bw/day by gavage (Makita  et al., 1975).

         Following a two-week pilot study, a 26 week oral toxicity study
    was conducted in Long Evans rats, body weight 80-100 g at
    commencement, in which ß-cyclodextrin (purity not specified) was
    administered by gavage as a suspension in aqueous 1% methylcellulose
    to groups of 15 male and 15 female animals at daily dose levels of
    200, 400 or 600 mg/kg bw; controls received an equal volume of 1%
    methylcellulose solution.  Additional groups of 6 animals of each
    sex were similarly dosed for 6 months followed by a 2-month recovery

    period prior to autopsy.  Body weight and food intake were recorded
    weekly and urinalysis, haematological and clinical biochemical
    examinations were performed at termination.  At autopsy, heart,
    lung, liver, kidneys, spleen and testes were weighed; macroscopical
    and histological examinations (haematoxylin, eosin, Oil red O, and
    PAS) were carried out on heart, lungs, liver, kidneys, spleen,
    gonads, stomach, intestine, pancreas, adrenals, lymph nodes and
    thymus.

         In the course of the study there were 5 deaths, not dose-
    related and attributed to intercurrent disease.  No adverse effects
    of treatment on clinical condition were seen in any dose group. 
    Food consumption was reduced in all treated male groups between
    weeks 18 and 24, body weight gain was reduced in males of the two
    highest dose groups between weeks 6 and 21 leading to a 10% deficit
    in weight gain in the top-dose group; females showed little change
    in food consumption or weight gain.  Statistically significant
    differences were observed in some haematological and clinical
    biochemical parameters but in most cases these were not
    systematically related to treatment and/or were within normal
    physiological ranges.  Dose-related increases in blood glucose were
    observed in all groups, significantly so in females of all dose
    groups and in males of the top-dose group.  Statistically
    significant increases, seen only at the top dose level, occurred in
    total bilirubin and Ca (males only), and in Cl, total protein and
    albumin/globulin ratio (females only) but were considered not to be
    of toxicological significance.  No significant treatment-related
    changes in organ weights were observed except for a dose-related
    reduction in spleen weight, significant in all female treatment
    groups and in males of the top-dose group.  However, these changes
    did not appear to associated with functional or histological
    alterations in the spleen and no treatment-related histological
    abnormalities were observed in heart, lung, liver, kidney, adrenals,
    gastrointestinal tract, gonads, lymph nodes, pancreas or thymus; no
    nerve tissue was examined histologically.

         In the 2 month withdrawal period following 6 months of
    treatment, the deficit in body weight in the high-dose males was
    largely recovered.  A significant increase in relative lymphocyte
    numbers was reported in all treated males but not in females. 
    Changes were noted in some organ weights, notably increased relative
    weights of lungs and kidneys in males of the top two dose groups, a
    decreased relative spleen weight in males in the top-dose group and
    decreased relative liver weights in females of the low and high (but
    not intermediate) dose groups.  These changes were small and
    unaccompanied by obvious functional changes, and were not considered
    to be of physiological significance.  Blood glucose levels remained
    higher than controls in all treated male groups and in the top-dose
    female group but were within the normal range as were the other
    clinical biochemical parameters.  The reductions in spleen weights
    seen in both sexes (up to 40% reduction in top dose males) were

    reported not to be accompanied by histopathological changes but
    individual histological details were not provided and a no-observed-
    effect-level cannot be established from this study (Gergely, 1982;
    Mészáros & Vetési, 1982).

         A 90-day study was conducted by dietary administration of
    ß-cyclodextrin to male and female Sprague-Dawley-derived OFA rats. 
    The test material, 99.7% pure on a dry weight basis, was
    administered to groups of 20 male and 20 female animals, 6-8 weeks
    old at commencement, by replacing starch in a semi-synthetic diet at
    levels of O, 1.25, 2.5, 5 or 10%; a "carbohydrate control" group
    received lactose at a dietary level of 10%.  Body weight and food
    intake were recorded weekly; water intake was monitored 3 times per
    week.  Ophthalmological examinations were conducted on all animals
    at the start and on 10 animals from the two control groups and the
    highest dose group at termination.  Urine volumes and concentrations
    of ß-cyclodextrin were recorded at 6 weeks and at the end of the
    study.  Blood and urine biochemistry, haematology and histopathology
    examinations (controls and high-dose group only) were carried out at
    termination.  The histopathological studies included histochemical
    examination (Perl's reaction) for ferric iron in liver, kidneys,
    spleen and lymph nodes.  Serum analyses included cholesterol,
    triglycerides, glucose, urea, creatinine, total bilirubin, total
    protein, albumin, GOT, GPT, Alk-P-ase, Na, K, Ca and Cl-;
    urinalyses included glucose, urea, uric acid, creatinine, total
    protein, Na, K and Ca.  Haematological examinations included
    haemoglobin, PCV, MCH, MCHC, MCV, RBC, total and differential
    leucocyte counts and prothrombin time.  At autopsy, relative organ
    weights were determined for brain, gonads, kidneys, spleen, thymus,
    caecum, heart and liver.

         One male from the lowest ß-cyclodextrin dose group died in the
    course of the study but there were no indications that the death was
    related to treatment.  There were inconsistent differences in food
    consumption between groups but no significant difference between the
    10% ß-cyclodextrin and the 10% lactose groups.  The dose achieved in
    the top-dose group was approximately 4.4 g/kg bw/day and 5.3 g/kg
    bw/day for males and females respectively.  No significant
    differences were recorded in water consumption between treatment
    groups in males but the lactose control group showed a reduced
    intake in weeks 4, 5 and 6.  With females there was a slight
    increase in water consumption for the 5 and 10% ß-cyclodextrin
    groups.  No significant differences in body weight were observed in
    any of the groups, except for the male lactose control group, and
    there were no dose-related adverse effects on haematology, serum
    biochemistry or urine composition.  A small fraction of the dose of
    ß-cyclodextrin was recovered in urine of animals of the top two dose
    groups and represented 0.1-0.3% of the highest dose administered. 
    The absolute and relative filled caecal weights of the rats, of both
    sexes, were increased following administration of diets containing

    ß-cyclodextrin or lactose (a common feature in rats receiving poorly
    absorbed and slowly digestible carbohydrates).  No treatment-related
    effects which were considered by the authors as indicative of a
    toxic response were found from the histopathological examination and
    it was concluded that ß-cyclodextrin appeared to lack toxicological
    activity at the doses tested (Olivier  et al. 1991).

         A review of the above study was carried out by an  ad hoc
    Scientific Advisory Group which essentially confirmed the authors'
    conclusions.  In noting that in the caeca of female rats of the high
    ß-cyclodextrin dose group there was a significant increase in
    sub-mucosal lymphoid follicles the reviewers considered that this
    was not of toxicological importance as it is commonly associated
    with caecal enlargement.  The increase in the presence of sinus
    macrophages in the lymph nodes seen in both males and females of the
    high-dose group was considered to be a physiological rather than a
    toxic response to the high levels of ß-cyclodextrin in the diet. 
    The review group did not comment on the observation that there was a
    low incidence of hepatic focal fibrosis in animals from the top-dose
    group (3/20 and 1/20 in males and females respectively) which was
    not seen in controls or in the group receiving lactose.  The group
    concluded unanimously that this subchronic study was well designed
    and properly conducted (Blumenthal  et al. 1990).

    2.2.2.3  Dogs

         The oral toxicity of ß-cyclodextrin (purity not specified) was
    examined in a 24-week study in beagles.  Groups of 3 males and 3
    females were given daily oral doses of 0, 100, 250 or 500 mg/kg bw,
    corrected for body weight changes every three weeks.  The material
    was administered in boluses made from egg yolk and dried breadcrumbs
    before the standard diet.  Food intake, clinical signs (behaviour,
    pulse rate, respiratory rate) were recorded initially and at the
    third, sixth, twelfth, eighteenth and twenty-fourth week of
    treatment; at the same time blood samples were collected for
    haematology (Hb, PCV, MCHC, WBC, differential leucocytes) and
    biochemical analyses (ASAT, ALAT, Alk-P-ase, BUN, glucose,
    bilirubin, total protein, inorganic P and Ca).  At autopsy, the
    following organs were weighed: liver, spleen, kidneys, gonads,
    heart, lung and brain; in addition to these organs, histological
    examinations were performed on adrenals, stomach, small and large
    intestine, pancreas, spinal cord and tissues showing gross lesions.

          All animals survived, but one dog from the top-dose group
    developed fever, lost appetite and showed catarrhal symptoms during
    the last week of the study, necessitating treatment with
    anti-distemper serum and antibiotics for four days; treatment with
    ß-cyclodextrin was continued during this period.  Diarrhoea was
    observed in some cases but did not appear to be due to
    administration of ß-cyclodextrin and the condition resolved
    spontaneously or after treatment with Tannocarbon and bolus

    astringents for 1-2 days.  No treatment-related changes in pulse or
    respiratory rates were detected.  There were no significant dose-
    dependent changes in blood biochemical parameters, although the
    values for bilirubin were very scattered, presumably due to
    haemolysis, at 0 and 3 weeks.  Haematological examination also did
    not reveal any treatment-related  changes; an increased eosinophil
    count in some cases was due to parasitic infestation, confirmed at
    autopsy.  There were no significant changes in organ weights
    although absolute and relative liver weights tended to be lower in
    all treated groups, and mean relative and absolute spleen weights
    were increased in all treated groups.  Histopathological changes
    observed were attributed to the method of sacrifice (magnesium
    sulfate injection) and were not dose-related.  The poor condition of
    some of the animals due to parasitic infestations and the agonal
    changes due to the method of sacrifice limited the sensitivity of
    the study but no obvious compound-related toxicity was detected
    (Haraszti, 1978; Tury  et al. 1978).

         In a 13-week study, groups of 2 male and 2 female beagles were
    given ß-cyclodextrin in the diet at levels of 0, 1.25, 2.5, 5 or 10%
    (the top-dose group received the test material at 5% of the diet for
    the first week and 10% thereafter).  The purity of the
    ß-cyclodextrin tested was stated to be > 99.0%.  The animals were
    acclimatised for four weeks prior to administration of the test
    compound during which time they received inoculations and
    anthelminthic treatment.  The condition of the animals and food
    intake were recorded daily and body weight was measured weekly. 
    Ophthalmoscopic examinations were performed at the beginning and end
    of the study; blood was collected for comprehensive haematology and
    clinical biochemical analysis prior to dosing and on weeks 6 and 13;
    total 24-hour urine samples were collected for analysis in weeks 7
    and 13.  Urine samples, and faeces collected during week 13, were
    analysed for ß-cyclodextrin as were blood samples collected 1, 3 and
    6 hours after feeding.  At termination organ weights were recorded
    for adrenals, brain, heart, kidneys, liver, lungs, pancreas,
    pituitary, spleen, thymus, thyroids and gonads.  An extensive range
    of tissues was fixed at autopsy but histopathological examination
    was limited to the alimentary tract (8 levels), kidneys, liver,
    pancreas, urinary bladder and all macroscopically abnormal tissues.

         The achieved group mean intakes of ß-cyclodextrin averaged over
    the 13 weeks of the study were 570, 1234, 2479 and 4598 mg/kg bw/day
    in the 1.25, 2.5, 5 and 10% dietary dose groups, respectively.  In
    the course of the study, liquid faeces were noted spasmodically in
    all animals, including controls, but the incidence was higher in the
    top-dose group (46.7%) than in controls (11.8%) or the 5% dose group
    (15.7%) and the higher frequency was considered to be treatment-
    related; there was also a significant decrement of weight gain in
    the top-dose group only, not associated with reduced food intake. 
    No other clinical symptoms attributable to treatment were observed
    and the ophthalmoscopic examinations were normal.  Haematological

    parameters were similar to controls except that red cell counts,
    haematocrit and haemoglobin levels were statistically significantly
    reduced in the top-dose group at weeks 6 and 13, and in the 5% dose
    group at week 6 only; however, the changes were small and similar
    trends were observed in the acclimatisation period so the
    toxicological significance was considered equivocal.  Reductions
    were observed in the group mean serum levels of cholesterol, HDL and
    ß-lipoprotein at weeks 6 and 13 for dogs in the 5% and particularly
    10% dose groups; in the top-dose group, total protein, albumin,
    calcium and phospholipid levels were slightly reduced at weeks 6 and
    13, and sodium levels were slightly reduced in week 13 only.  Other
    biochemical parameters were unaffected by treatment.  Urinalysis
    indicated that protein levels were elevated in the 5 and 10% dose
    groups at weeks 6 and 13 but there were no other notable findings
    due to treatment.  At autopsy, organ weights were generally
    unaffected by treatment; the group mean absolute thymus weight in
    the top-dose group was significantly lower than controls (p<0.5)
    but similar to the lowest-dose group and there was no dose-related
    trend with the mean group weight being non-significantly higher than
    controls in the 5% dose group.  It was concluded that treatment had
    no conclusive effect on organ weights.  No macroscopic abnormalities
    were observed and there were no observable treatment-related changes
    in the tissues examined histologically.  The authors concluded that
    the 2.5% dose level (equal to 1234 mg/kg bw/d) was a NOEL and that,
    in the absence of any treatment-related macroscopic or microscopic
    pathological findings, the other effects noted were indicative of
    only a mild toxic response.  However, the small number of animals
    and the restricted histological examination limited the power of
    this study.  ß-Cyclodextrin in dose-dependent concentrations was
    detected in urine, faeces and blood of these animals dosed orally
    (see  Biochemical Aspects above) (Smith  et al. 1992).

    2.2.3.  Long-term toxicity/carcinogenicity studies

         No information available.

    2.2.4.  Reproduction studies

         No information available.

    2.2.5.  Special studies on teratogenicity

    2.2.5.1.  Rats

         A teratogenicity study was conducted in CFY rats in which
    groups of 40 mated females (30 in control group) with positive
    vaginal smears were given ß-cyclodextrin by gavage in 1% methyl
    cellulose suspension at dose levels of 0, 200, 400 and 600 mg/kg
    bw/dy on days 7-11  post coitus.  The animals were killed on day 21
    and the number of implants, resorptions, live and dead fetuses,

    fetal weights and rates of congenital anomalies recorded (no details
    were given of the methodology of the teratological examinations).

         The conception rate was poor (about 30%) giving only 11-12
    pregnant animals in each group.  Five animals from the top and mid-
    dose groups and 1 from the low-dose group died during the study, the
    deaths being attributed to bronchopneumonia due to mis-dosing. 
    Maternal weight gain was reduced in a dose-related manner but there
    were no changes in mean number of implants (12.7 -14.1), resorptions
    (1.3-7.3%) fetal viability (93-99%) or fetal weight (3.5-3.9 g). 
    There were 5 congenital anomalies reported out of 628 fetuses, 2 in
    the low-dose group (hydronephrosis, cardiac anomaly), 1 in the mid-
    dose group (cardiac anomaly) and 2 in the top-dose group (both
    absence of right kidney);  no anomalies were seen in the control
    group.  No details were given of any examination for skeletal
    anomalies.  The authors state that the incidence of congenital
    malformations corresponded to the "spontaneous" incidence but no
    anomalies were seen in controls and no historical data were
    presented to support this conclusion (Jellinek  et al, undated).

         Following a pilot investigation at dose levels of 500 and 2 500
    mg/kg bw/day a teratology study was performed in Wistar rats given
    ß-cyclodextrin as a suspension in 1.25% aqueous methyl cellulose by
    gavage at doses of 0, 100, 500 and 2 500 mg/kg bw/day on days 7-16
    of pregnancy.  The group sizes were 26-28 sperm positive females, of
    which 22-25/group proved to be pregnant.  The dams were sacrificed
    on day 21 and the following parameters monitored: number of corpora
    lutea and implantations, preimplantation loss, embryonic and late
    fetal mortality, and number of viable fetuses.  About 50% of the
    fetuses from each litter were examined for soft tissue defects
    (Wilson's technique) and the remainder were cleared, stained
    (Alizarin Red or Alcian Blue/Alizarin red) and examined for skeletal
    anomalies.

         The doses used in this study had no effect on the clinical
    condition, food consumption or weight gain of the dams and there
    were no significant effects on intra-uterine mortality, viable
    fetuses or on the incidence or type of congenital malformation.  The
    anomalies recorded in this study were regarded as sporadic and
    unrelated to treatment, and had been seen previously among fetuses
    from over 600 control dams.  Under the conditions of this study,
    there was no evidence of fetotoxicity or teratogenicity (Druga,
    1985).

         In a further teratogenicity study in Sprague-Dawley rats,
    groups of 24-30 mated females were given ß-cyclodextrin as a
    suspension in 1% aqueous methyl cellulose by gavage at doses of
    0, 1 250, 2 500 and 5 000 mg/kg bw/day on days 7-16 of pregnancy. 
    The dams were sacrificed on day 21 and the following parameters
    monitored: number of corpora lutea and implantations, live and dead
    fetuses and fetal size.  About 50% of the fetuses from each litter

    were examined for visceral anomalies (Wilson's technique) and the
    remainder were cleared, stained (Alizarin Red) and examined for
    skeletal anomalies.

         A slight growth retardation was observed in the highest dose
    group, significant from day 8 to 21, which was associated with a
    reduced food intake; otherwise there was no effect of treatment on
    weight gain of the dams.  No mortality which could be ascribed to
    the test compound was observed but 5 animals in the top-dose group
    died as a result of misdosing and oesophageal perforation, reducing
    the effective number from 30 to 25.  There were no statistically
    significant differences between groups in any of the following:
    uterine weight (full and empty), placental weight, weight of
    fetuses, number of fetuses (all were alive), number of implantation
    sites, resorptions and corpora lutea or sex ratio.  There were no
    significant effects on the incidence or type of congenital
    malformation.  The anomalies recorded in this study were regarded as
    unrelated to treatment.  Under the conditions of this study, there
    was no evidence of fetotoxicity or teratogenicity at doses of up to
    5 000 mg/kg bw;  retardation of weight gain of the dam seen at this
    dose level was not apparent at a dose of 2 500 mg/kg bw/dy (Leroy
     et al. 1991).

    2.2.5.2.  Rabbits

         Following a pilot investigation at dose levels of 250, 500 and
    1 000 mg/kg bw/day a teratology study was performed in groups of
    12-14 thalidomide-sensitive New Zealand white rabbits.  The dams
    were artificially inseminated and given ß-cyclodextrin as a
    suspension in 1.25% aqueous methyl cellulose by gavage at doses of
    0, 150, 300 and 600 mg/kg bw/day on days 7-19 of gestation.  The
    dams were sacrificed on day 28 of gestation and the following
    parameters monitored: number of corpora lutea and implantations,
    preimplantation loss, embryonic and late fetal mortality, and number
    of viable fetuses.  All of the fetuses from each litter were
    examined for external and visceral abnormalities on the day of
    autopsy and the fetuses then fixed, stained (Alizarin Red) and
    examined for skeletal anomalies.

         At the doses used, ß-cyclodextrin had no effect on the clinical
    condition, food consumption or weight gain of the dams and there
    were no significant effects on intra-uterine mortality, fetal size
    or number of viable fetuses.  The incidence of minor congenital
    malformations was slightly but not significantly elevated in all
    treated animals and was not dose-dependent.  The anomalies recorded
    in this study were regarded as sporadic and unrelated to treatment,
    and  under the conditions of this study, there was no evidence of
    fetotoxicity or teratogenicity (Dóczy 1985).

    2.2.6.  Special studies on genotoxicity

         The results of genotoxicity studies with ß-cyclodextrin are
    summarized in Table 2.

    2.2.7  Special studies on nephrotoxicity

         The experimental use of ß-cyclodextrin in dialysis fluids has
    been associated with a characteristic nephrosis.  Groups of 4 rats
    of 100-125 g body weight were given single subcutaneous injections
    of ß-cyclodextrin at dose levels of 225, 450 or 900 mg/kg bw and
    sacrificed 12, 24, 48 or 96 hours later.  In a repeat-dose study,
    similar animals were given doses of 225, 450, 675 or 900 mg/kg daily
    for 1, 2, 3, 4 or 7 days and killed 24 hours after the last dose. 
    The kidneys were examined by light and electron microscopy.  A
    characteristic nephrosis was observed, manifested as a series of
    alterations in the vacuolar organelles of the proximal convoluted
    tubule.  The changes started with an increase in apical vacuoles and
    appearance of giant lysosomes followed by extensive vacuolation,
    cell disintegration and amorphous mineralization.  These lesions
    were evident following a minimal single dose of 675 mg/kg bw and a
    crude dose-response relationship was established.  The earliest
    manifestations, midcellular cytoplasmic vacuoles, were observable
    24 hours after injection.  Following repeated dosing, light
    microscopic lesions were found in one rat given 225 mg/kg bw/dy for
    4 days and daily injections of 450 mg/kg bw resulted in severe
    nephrosis but no deaths; all the animals given repeated doses of
    900 mg/kg bw/dy died within 4 days.  It was concluded by the authors
    that intracellular concentration of non-metabolisable ß-cyclodextrin
    by the lysosomal pathway represents a "perversion" of the
    physiologic function of the proximal tubule, leading to cell death
    (Frank  et al. 1976).

         In studies of the use of ß-cyclodextrin in peritoneal
    dialysates to accelerate removal of i.v. phenobarbital in adult
    rats, it was observed that a number of the animals given 1.5%
    ß-cyclodextrin solution in physiological phosphate i.p. at a level
    of 15% of body weight died overnight.  In a follow-up study, blood
    urea nitrogen (BUN) was determined after administration of
    ß-cyclodextrin orally as 10 ml of a 6% suspension or i.p. as a 0.75%
    solution in phosphate buffer.  Following fasting for 5 hours and
    gavage with single or three daily doses, BUN was within the normal
    range 24 hours after the last dose.  Intra-peritoneal administration
    of ß-cyclodextrin was accompanied by a significant increase (3-4
    fold) in BUN 24 and 72 hours after administration.  The BUN levels
    fell after 100 hours and returned to normal (Perrin  et al. 1978).


        Table 2.  Results of genotoxicity studies with ß-cyclodextrin.

                                                                                                                                 

    Test system                    Test Object                      Concentration            Results     Reference

                                                                                                                                 

    Host-mediated assay            E.coli WP2uvr A trp-,            Doses to rat of          -           Igali 1978
    in rat                         S.typhimurium TA1538             0, 100 or 
                                   1 000 mg/kg bw

    Chromosome aberration          Long Evans rat (?)bone           0, 200, 400 or           -*          Czeizel 1978
                                   marrow bw/d for 3 m.             600 mg/kg 

    Mouse micro-nucleus            Mouse bone-marrow                100 mg/kg bw             -           Weill, 1988
    test

    Sex-linked recessive           Drosophila melanogaster          1.6, 8 & 16 mM           -           Parádi 1987
    lethal  mutation

    Ames test (1)                  S. typhimurium TA98,             0, 0.1, 0.5, 1,          -           Weill, 1987
                                   TA100, TA1535, TA1537,           2 & 4 mg/plate
                                   TA1538

    HPRT mutation                  V79 Chinese hamster              10, 30, 100, 300         -           Marzin  et al. 1990
    (6-thio-guanine                cells                            & 1 000 µg/ml
    resistance) (1)

    In vitro chromosome            Human lymphocytes                100, 300 &               -           Marzin  et al. 1991
    aberration test (1)                                             1 000 µg/ml
                                                                                                                                 

    *The number of cells examined was small relative to normal guidelines
    (1) With and without rat liver S9 fraction
    

         Subcutaneous administration of ß-cyclodextrin at a daily dose
    level of 450 mg/kg bw in saline on 7 consecutive days resulted in
    polyuria and proteinuria, a doubling in relative kidney weight and a
    decrease in the activities of succinic dehydrogenase, alkaline
    phosphatase, glucose-6-phosphatase and ß-glucuronidase in proximal
    convoluted tubules (Hiasa  et al. 1981).

    2.2.8  Special studies on skin irritancy/sensitisation

    2.2.8.1  Guinea-pigs

         Evaluation of the cutaneous delayed hypersensitivity of
    ß-cyclodextrin was carried out in albino Dunkin-Hartley guinea-pigs
    using groups of 20 animals of both sexes.  Seven applications
    (induction phase) and challenge were carried out using 0.4 g
    ß-cyclodextrin moistened with 0.5 ml water.  Macroscopic
    examinations were carried out 6, 24 and 48 hours after removal of
    the occlusive patches.  Histopathological examinations were carried
    out on 6 animals showing doubtful reactions at 6 hours.  No delayed
    hypersensitivity reactions were provoked by this protocol (Mercier,
    1990).

    2.2.8.2  Rabbits

         A primary dermal irritation study was conducted in New Zealand
    white rabbits by application of 0.5 g of test compound moistened
    with 0.5 ml saline was applied to the shaved dorsal skin of 3
    animals under occlusion for 24 hours.  The mean primary irritation
    score was 0.50 (minimally irritating) based on a barely perceptible
    erythema after 24 hours, there was no eschar or oedema and the
    treatment sites were normal by 24 hours after removal (Reagan &
    Becci, 1985).

         A primary dermal irritation study in albino rabbits using an
    abraded skin protocol.  The index of primary cutaneous irritation
    which was obtained (0.01) classified ß-cyclodextrin as non-irritant
    (Leroy  et al. 1990).

    2.2.9  Special studies on eye irritancy

         In an ocular irritancy/corrosion test in albino rabbits,
    ß-cyclodextrin was classified as slightly irritant (Leroy  et al.
    1990).

    2.2.10  Special studies on tumour promotion

         Subcutaneous injection of ß-cyclodextrin at a dose of 450 mg/kg
    bw daily for 7 days during week 3 increased the number and size of
    renal tubular cell tumours in inbred Wistar rats treated with
     N-ethyl- N-hydroxyethylnitrosamine (EHEN) in the diet at a
    concentration of 1 000 mg/kg in the preceding 2 weeks.  After 32

    weeks, the incidence of tumours was 50% in animals treated with the
    EHEN alone and 100% in animals subsequently given ß-cyclodextrin. 
    In addition, ß-cyclodextrin promoted the development of renal
    tumours in rats treated with 500 mg EHEN/kg diet, which was a sub-
    threshold dose for renal tubular cell tumourigenesis (Hiasa  et al.
    1982).

    2.2.11  Special studies on cell-membranes

         The interactions between cyclodextrins and membrane
    phospholipids, liposomes and human erythrocytes were studied
     in vitro.  ß-Cyclodextrin did not alter the differential scanning
    calorimetric behaviour of phospholipids, and did not increase the
    permeability of dipalmitoyl-phosphatidylcholine liposomes.  No
    effects on active or passive transport of 42K or 86Rb into
    erythrocytes was observed at concentrations of up to 10-2mol/litre
    but at 1.7x10-2mol/litre, ß-cyclodextrin caused a significant
    increase in passive transport and 8-10% haemolysis (Szejtli  et al.
    1986).

         ß-Cyclodextrin induced haemolysis of human erythrocytes
     in vitro in isotonic solution with a threshold concentration of
    3 mM (3 400 mg/l).  Swelling of erythrocytes, associated with
    release of cholesterol from the membrane, was observed at lower
    concentrations, approximately 20% of the membrane cholesterol was
    released from the membrane at a concentration of 2 mM (2 300 mg/l)
    ß-cyclodextrin (Irie  et al. 1982).

    2.3  Observations in humans

    2.3.1  Absorption, distribution, metabolism and
    excretion

         The fate of ß-cyclodextrin the human gastrointestinal tract was
    studied in ileostomy subjects and in normal volunteers after
    administration of 10 g in a fasting state or after 3 doses of 10 g
    daily with meals.  In the ileostomy subjects, the recovery of
    ß-cyclodextrin in the ileal effluent was 97±10% and 91±5%
    respectively.  In normal subjects, in which utilization was
    estimated by the breath hydrogen technique and analysis of stools,
    breath hydrogen was low, and insignificant levels of ß-cyclodextrin
    were detectable in faeces.  It was concluded that ß-cyclodextrin is
    hardly hydrolyzed or absorbed in the human small intestine but is
    fermented by colonic microflora with minimal apparent hydrogen
    production (Flourié  et al. 1992).

    2.3.2  Human tolerance studies

         In three successive periods of one week, eighteen healthy
    males, aged 23±2 years, were given doses of 0, 24 or 48 g

    ß-cyclodextrin/day in addition to the normal diet.  The volunteers
    were randomly assigned to the treatment groups in what was stated to
    be a placebo-controlled, double blind protocol and the test compound
    was administered in a chocolate drink equally divided over three
    meals.  At the high dose level "to avoid too drastic influences on
    bowel function" the subjects were given 24 g ß-cyclodextrin on the
    first day, 36 g on the second day and 48 g on days 3-7.  Tolerance
    was evaluated by subjective assessment of abdominal complaints using
    a questionnaire.  At the end of each 7-day period, breath hydrogen
    concentration was measured.  One of the volunteers was withdrawn
    from the study on after three days and replaced with a substitute
    because of too many adverse events, resembling lactose intolerance
    (abdominal cramps, nausea, diarrhoea) which were not reported before
    the start of the study.  It is not clear what dose this volunteer
    received before withdrawal or whether the symptoms preceded
    treatment.

         There was a significant increase in complaints of flatulence
    (p<0.05) at the higher intake level; other scores of abdominal
    complaints, reported defaecation patterns and breath hydrogen were
    stated not to change significantly.  The authors concluded that the
    dose of 24 g ß-cyclodextrin/day was well tolerated on a short term
    basis (van Dokkum & van der Beek 1990).

    2.3.3  Sensitization/irritation

         In a repeated insult occlusive patch test in 58 subjects
    (1 male and 57 females aged 21 to 68 years) ß-cyclodextrin did not
    induce irritation or allergic contact dermatitis.  Three subjects
    showed scattered, transient, barely perceptible to mild, non-
    specific patch test responses during the induction or challenge
    phases of the study, none of which were irritant or allergic in
    nature (Alworth   et al. 1985).

    3.  COMMENTS

         Metabolic studies in animals and humans consistently indicate
    that ß-cyclodextrin is poorly hydrolyzed or absorbed in the upper
    gastrointestinal tract but is largely utilized following hydrolysis
    by the gut microflora in the lower gut.  A small proportion of
    ingested ß-cyclodextrin may be absorbed intact.

         A number of acute and short-term toxicity studies were reviewed
    which indicated low toxicity by the oral route, although most of
    these studies used smaller numbers of animals or more limited
    histological examination than would normally be appropriate for
    establishing an ADI.  In a well-conducted short-term toxicity study
    in rats, there were no effects of toxicological significance other
    than caecal enlargement and an increased number of macrophages in
    intestinal lymph nodes at doses of up to 10% ß-cyclodextrin in the
    diet; these effects are a common feature of poorly absorbed
    polysaccharides.

         In  in vitro stidies, ß-cyclodextrin sequestered cholesterol
    from erythrocyte membranes and caused haemolysis, but only at
    concentrations much higher than those seen in the blood of dogs
    given ß-cyclodextrin at a level of 10% in the diet.  No effects on
    mucosal cells of the gastrointestinal tract were seen in the high-
    oral-dose studies.  ß-Cyclodextrin was non-genotoxic in a range of
    tests, and it does not have a structure likely to be associated with
    such activity.  Given its poor bioavailability and lack of
    genotoxicity, the Committee concluded that a long-term
    carcinogenicity study was not required for the evaluation of this
    substance.

         When administered parenterally to rats, ß-cyclodextrin was
    nephrotoxic, but no renal toxicity was observed in any of the short-
    term toxicity studies using oral administration.  In dogs, the
    urinary excretion of unchanged ß-cyclodextrin was low, even when the
    compound was given at a dose level of 10% in the diet, indicating
    that it is unlikely that systemic levels following oral
    administration would be high enough to cause renal toxicity.

    4.  EVALUATION

         The Committee was informed that a 1-year oral toxicity study on
    ß-cyclodextrin in dogs was under way, and requested the results of
    this study to complete the evaluation of this compound.

         Despite its low toxicity, the Committee was concerned about the
    possible sequestering effects of ß-cyclodextrin on lipophilic
    nutrients and drugs.  In particular, further data on the effects of
    ß-cyclodextrin on the bioavailability of lipophilic nutrients are
    required.

         The Committee concluded that there were sufficient data to
    allocate a temporary ADI of 0-6 mg/kg bw for ß-cyclodextrin, based
    on a NOEL of 2.5% in the diet, equal to 1230 mg/kg bw/day in the
    study in dogs and a safety factor of 200.

         The results of the 1-year study in dogs and information on the
    effects of ß-cyclodextrin on the bioavailability of lipophilic
    nutrients are required by 1995.

         As this is a novel product with a wide range of potential
    applications, the Committee requested further information on the
    range of possible production methods that could be used in its
    manufacture.

    5.  REFERENCES

    ALWORTH, K., SHANAHAN, R.W. & DORMAN, N. (1985)  Clinical safety
    evaluation of beta cyclodextrin: Repeated Insult Patch Test. 
    Unpublished report of Essex Testing Clinic Inc. Submitted to WHO by
    Société Roquette Fréres, Lestrem, France.

    ANDERSEN, G.H., ROBBINS, F.M., DOMINGUES, F.J., MOORES, R.G. & LONG,
    C.L. (1963)  The utilization of Schardinger dextrins by the rat. 
     Toxicol. Appl. Pharmacol., 5, 257- 266.  Submitted to WHO by
    Société Roquette Fréres, Lestrem, France.

    ANTENUCCI, R.N. & PALMER, J.K. (1984)  Enzymatic degradation of
    alpha- and ß-cyclodextrins by  Bacteroides of the human colon.
     J. Agric. Food Chem. 32, 1316-1321.  Submitted to WHO by Société
    Roquette Fréres, Lestrem, France.

    BLUMENTHAL, H., FLAMM, W.G., FORBES, A. & MUNRO, I. (1990) 
    Unpublished report of a Science Advisory Group convened to review
    data on toxicity and food applications of ß-cyclodextrin. Submitted
    to WHO by Flamm Associates, Reston, VA, USA.

    BUSCH, B., BIESEMEIER, J.A. & BECCI, P.J. (1985)  Acute inhalation
    toxicity study of beta cyclodextrin in Sprague Dawley rats.
    Unpublished report No. 8456 of Food and Drug Research Laboratories
    Inc. submitted to WHO by Société Roquette Fréres, Lestrem, France

    CZEIZEL, E. (1978)  Study of the chromosomal mutagenic effect of
    beta-cyclodextrin in Long Evans rats. Unpublished report of the
    National Institute of Public Health, Budapest, Hungary.  Submitted
    to WHO by Société Roquette Fréres, Lestrem, France.

    D²CZY, K. (1985)  Report on the teratology study of
    beta-cyclodextrine in New Zealand white rabbits. Unpublished report
    of the Institute for Drug Research, Budapest, Hungary, submitted to
    WHO by Société Roquette Frères, Lestrem, France

    DRUGA, A. (1985)  Report on the teratology study of
    beta-cyclodextrin in LATI:HAN:Wistar rats.  Unpublished report of
    the Institute for Drug Research, Budapest, Hungary, submitted to WHO
    by Société Roquette Frères, Lestrem, France

    FLOURIÉ, B., MOLIS, C., ACHOUR, L., DUPAS, H., HATAT, C. & RAMBEAUD,
    J.C. (1992)  Digestibility of ß-cyclodextrin in the human intestine. 
    Unpublished report of INSERM U 290 submitted to WHO by Société
    Roquette Frères, Lestrem, France

    FRANK, D.W., GRAY, J.E. & WEAVER, R.N. (1976) Cyclodextrin nephrosis
    in the rat.  Am. J. Pathol., 83, 367-382.  Submitted to WHO by
    Société Roquette Fréres, Lestrem, France.

    GERGELY, V. (1982)  Six months chronic toxicity study of
    beta-cyclodextrin in Long-Evans rats.  Unpublished report. 
    Submitted to WHO by Société Roquette Fréres, Lestrem, France.

    GERL²CZY, A., F²NAGY, A. & SZEJTLI, J. (1981)  Absorption and
    metabolism of ß-cyclodextrin by rats.  Proceedings of the 1st
    International Symposium on Cyclodextrins, Budapest, Hungary, (Ed. J.
    Szejtli) Akadémiai Kiadó, Budapest; and Reidel Publishing,
    Dordrecht, The Netherlands.  Submitted to WHO by Société Roquette
    Fréres, Lestrem, France.

    GERL²CZY, A., F²NAGY, A., KERESZTES, P., PERLAKY, L. & SZEJTLI, J.
    (1985)  Absorption, distribution, excretion and metabolism of orally
    administered 14C-ß-cyclodextrin in rat.   Arzneimittel-Forsch./Drug
     Res. 35, 1042-1047.  Submitted to WHO by Société Roquette Fréres,
    Lestrem, France.

    HARASZTI, J. (1978)  Twenty-four week oral toxicity study of
    beta-cyclodextrin in beagle dogs. Unpublished report of Institute of
    Obstetrics, Department of Surgery, Budapest, Hungary.  Submitted to
    WHO by Société Roquette Fréres, Lestrem, France.

    HIASA, Y., OSHIMA, M., KITAHORI, Y., YUASA, T., FUJITA, T., IWATA,
    C., MIYASHIRO, A. & KONISHI, N. (1981) Histochemical studies of
    ß-cyclodextrin nephrosis in rats.  J. Nara. Med. Ass. 32, 316-326.

    HIASA, Y., OSHIMA, M., KITAHORI, Y., KONISHI, N., FUJITA, T. &
    YUASA, T. (1982) ß-cyclodextrin: Promoting effect on the
    development of renal tubular cell tumours in rats treated with
     N-ethyl- N-hydroxyethylnitrosamine.  J.Natl.Cancer Inst., 69,
    964-967.

    IGALI, S. (1978)  Study of the gene-mutation inducing effect of
    beta-cyclodextrin by host- mediated assay.  Unpublished report of
    the National Institute of Radiobiology, Budapest, Hungary. 
    Submitted to WHO by Société Roquette Fréres, Lestrem, France.

    IRIE, T., OTAGIRI, M., SUNADA, M., UEKAMA, K., OHTANI, Y., YAMADA,
    Y. & SUGIYAMA, Y. (1982) Cyclodextrin-induced hemolysis and shape
    changes of human erythrocytes  in vitro. J.Pharm.Dyn., 5, 741-744.

    JODAL, I., KANDRA, L., HARANGI, J., NÛNÛSI, P. & SZEJTLI, J. (1983) 
    Unpublished report submitted to WHO

    JODAL, I., KANDRA, L., HARANGI, J., NÛNÛSI, P. & SZEJTLI, J. (1984)
     Stärke, 36, 140.

    JELLINEK, H., KADAR, A. & CZEIZEL, E. (undated)  Unpublished report
    of the 2nd Department of Pathology, Semmelweiss University of
    Medicine, Budapest, Hungary submitted to WHO by Société Roquette
    Frères, Lestrem, France

    LEROY, P., OLIVIER, PH., WILS, D., COURBOIS, F. & MINET, P. (1990) 
    Local tolerance tests in the rabbit: index of primary skin
    irritation; index of eye irritation, beta-cyclodextrose (Kleptose)
    Unpublished report No. 90014 of the Biotoxicology Laboratory,
    Roquette Fréres submitted to WHO by Société Roquette Fréres,
    Lestrem, France.

    LEROY, P., OLIVIER, PH., JOURDAIN, R., MINET, P., WILS, D.,
    VERWAERDE, F., POTIER, M.N., BONIFACE, M., DEMARQUILLY, C., SIOU, G.
    & GUFFROY, M. (1991)  Teratogenicity study of orally administered
    ß-cyclodextrin in rats. Unpublished report No. 90021 of the
    Biotoxicology Laboratory, Roquette Fréres submitted to WHO by
    Société Roquette Fréres, Lestrem, France.

    MAKITA, T., OJIMA, N., HASHIMOTO, Y., IDE, H., TSUJI, M. & FUJISAKI,
    Y. (1975)  Chronic oral toxicity study of ß-Cyclodextrine (ß-CD) in
    rats.   Oyo Yakuri, 10, 449-458.  Submitted to WHO by Société
    Roquette Fréres, Lestrem, France.

    MARZIN, D., VO PHI, H., NESSLANY, F., LAGACHE, D. & DEHOUCK, M.P.
    (1990)  Mutagenicity test by investigating point mutation at the
    HPRT locus in V79 Chinese hamster cell culture (6-thioguanine
    resistance) carried out with the compound ß-cyclodextrine.  
    Unpublished report of the Institute Pasteur de Lille submitted to
    WHO by Société Roquette Frères, Lestrem, France

    MARZIN, D., VO PHI, H., NESSLANY, F., LAGACHE, D. & DEHOUCK, M.P.
    (1991) Test for chromosome aberrations by  in vitro human
    lymphocyte metaphase analysis.  Unpublished report of the Institute
    Pasteur de Lille submitted to WHO by Société Roquette Frères,
    Lestrem, France

    MERCIER, O. (1990)  Test to evaluate the sensitizing potential by
    topical applications in the guinea pig: the epicutaneous
    maximisation test. Unpublished report of Hazleton France No. 010360
    submitted to WHO by Société Roquette Frères, Lestrem, France

    MÉSZÛROS, J. & VETÉSI, F. (1982) A six month chronic toxicity study
    in Long-Evans rats; Expert opinion. Unpublished report submitted to
    WHO by Société Roquette Frères, Lestrem, France.

    MIYAZAKI, Y., KAWAHARA, K. & NOZOE, M. (1979) Influence of
    cyclodextrin on mice.  Ann. Rep. Coll. Agric., Saga Univ. 47,
    17-22.  Submitted to WHO by Société Roquette Fréres, Lestrem,
    France.

    OLIVIER, P., VERWAERDE, F. & HEDGES, A.R. (1991) Subchronic toxicity
    of orally administered Beta-Cyclodextrin in rats.  J. American Coll.
     Tox., 10, 407-419.  Submitted to WHO by Société Roquette Fréres,
    Lestrem, France.

    PARÛDI, E. (1987)  Study of the recessive lethal mutation-inducing
    effect of beta-cyclodextrin by Drosophila SLRL assay.  Unpublished
    report of 31st August 1987.  Submitted to WHO by Société Roquette
    Fréres, Lestrem, France.

    PERRIN, J.H., FIELD, F.P., HANSEN, D.A., MUFSON, R.A. & TOROSIAN, G.
    (1978).  ß-cyclodextrin as an aid to peritoneal dialysis. Renal
    toxicity of ß-cyclodextrin in the rat.  Chem. Path. Pharmacol.,
    373-376.  Submitted to WHO by Société Roquette Fréres, Lestrem,
    France.

    REAGAN, E.L. & BECCI, P.J. (1985)  Primary dermal irritation study
    of beta cyclodextrin in New Zealand White rabbits.  Unpublished
    report No. 8456A of Food and Drug Research Laboratories Inc.
    submitted to WHO by Société Roquette Frères, Lestrem, France.

    SEBESTYÉN, G. (1979) Three-months oral toxicity study of
    beta-cyclodextrin in Long-Evans rats. Report of Chinoin
    Pharmaceutical and Chemical Works.  Submitted to WHO by Société
    Roquette Fréres, Lestrem, France.

    SEBESTYÉN, G. (1980) The acute LD50 values of beta-cyclodextrin in
    CFY rats, CFLP mice and mongrel dogs.  Report of Chinoin
    Pharmaceutical and Chemical Works.  Submitted to WHO by Société
    Roquette Fréres, Lestrem, France.

    SMITH, T.G., COX, R.A., BUIST, D.P., CROOK, D., HADLEY, J.C. & C.
    GOPINATH (1992) Beta-cyclodextrin toxicity to dogs by repeated
    dietary administration for 13 weeks.  Preliminary study. 
    Unpublished report No. ROQ 2/911089 of Huntingdon Research Centre
    Ltd. Submitted to WHO by Société Roquette Frères, Lestrem, France

    SUZUKI, M. & SATO, A. (1985) Nutritional significance of
    cyclodextrins: Indigestibility and hypolipemic effect of
    alpha-cyclodextrin.  J. Nutr. Sci. Vitaminol., 31, 209-223. 
    Submitted to WHO by Société Roquette Fréres, Lestrem, France.

    SZABO, P., FERENCZY, T., SERF²Z², J., SZEJTLI, J. & LIPTÛK, A.
    (1981a)  Absorption and elimination of cyclodextrin derivatives by
    rabbits and rats.  Proceedings of the 1st International Symposium
    on Cyclodextrins, Budapest, Hungary, (Ed. J. Szejtli) Akadémiai
    Kiadó: Budapest and Reidel Publishing: Dordrecht, Holland, 115. 
    Submitted to WHO by Société Roquette Fréres, Lestrem, France.

    SZABO, P., FERENCZY, T., SERF²Z², J. AND SZEJTLI, J. (1981b)
    Investigation of the absorption and elimination conditions of
    ß-cyclodextrin and its derivatives. Drugs, Biochemistry and
    Metabolism, MATRAFÜRED, 167-171.

    SZEJTLI, J., GERL²CZY, A. & F²NAGY, A. (1980)  Intestinal absorption
    of 14C-labelled ß- cyclodextrin in rats.  Arzneimittel-Forsch./Drug
     Res., 30, 808-810.  Submitted to WHO by Société Roquette Fréres,
    Lestrem, France.

    SZEJTLI, J. & BUDAI, Z. (1976)  Acta Chimica Acad. Sci. Hung. 91,
    73-80

    SZEJTLI, J., CSERHÛTI, Y. & SZÖGYI, M. (1986) Interactions between
    Cyclodextrins and cell membrane phospholipids.  Carbohydrate
     Polymers, 6, 35-49.

    TURY, E., DOBOS-KOVÛCS, M & SOMOGYVÛRI, K. (1978)  The pathological
    and histo-pathological study of Beagle dogs subjected to
    beta-cyclodextrin treatment for twenty-four weeks. Unpublished
    report of the Departments of Surgery and Pathology, University of
    Veterinary Science, Budapest, Hungary.  Submitted to WHO by Société
    Roquette Fréres, Lestrem, France.

    TURY, E., DOBOS-KOVÛCS, M & SOMOGYVÛRI, K. (1979)  The pathological
    and histo-pathological study of Long Evans rats subjected to
    beta-cyclodextrin treatment for three months. Unpublished report of
    the Departments of Surgery and Pathology, University of Veterinary
    Science, Budapest, Hungary.  Submitted to WHO by Société Roquette
    Fréres, Lestrem, France.

    Van DOKKUM, W. & Van der BEEK, E.J. (1990)  Tolerance of
    ß-cyclodextrin in man.  Unpublished report No. V 90.419 of TNO, to
    AVEBE, Foxhol, The Netherlands, submitted to WHO by Dr R Scheuplein,
    USFDA, Washington, DC, USA.

    WEILL, N. (1987)  Unpublished report No. 709207 of Hazleton-IFT
    submitted to WHO by Société Roquette Frères, Lestrem, France

    WEILL, N. (1988)  Unpublished report No. 801201 of Hazleton-IFT
    submitted to WHO by Société Roquette Frères, Lestrem, France.


    See Also:
       Toxicological Abbreviations
       beta-Cyclodextrin (WHO Food Additives Series 35)
       beta-CYCLODEXTRIN (JECFA Evaluation)