First draft prepared by Dr D.L. Grant,
    Toxicological Evaluation Division
    Health and Welfare Canada.  


         Trichlorogalactosucrose (TGS) has been evaluated previously by
    the 33rd Joint FAO/WHO Expert Committee on Food Additives in 1989
    (see Annex 1, reference 83).  The Committee allocated a temporary
    ADI of 0-3.5 mg/kg b.w. and indicated that the following further
    studies or information were required:

         1.   Information on the absorption and metabolism of TGS in
              humans after prolonged oral dosing.

         2.   Results of studies to ensure that TGS produces no adverse
              effects in people with insulin-dependent and maturity-
              onset diabetes.

         3.   Results of further studies in rats on the elimination of
              TGS from pregnant animals and from the fetus, to exclude
              the possibility of bioaccumulation.

         4.   Results of a short-term rat study on 6-chlorofructose.

         Trichlorogalactose (1,6-dichloro-1,6-dideoxy-beta-D-fructo-
    furanosyl-4-chloro-4-deoxy-alpha-D-galactopyranoside) (TGS), is
    derived from sucrose by selective replacement of the three hydroxy
    groups at positions 4',1' and 6' by chlorine atoms, which greatly
    increases sweetness.  At room temperature, in water, TGS is 600-650
    times sweeter than sucrose at a concentration of 4-5%.  

         In acid solution, TGS hydrolyses slowly to its constituent
    monosaccharides, 4-chlorogalactose (4-CG) and 1,6-dichlorofructose
    (1,6-DCF).  This process is influenced by temperature and pH.  Under
    the extreme conditions (treatment of TGS with 0.11N aqueous
    hydrochloric acid at 68 C for 72 h) used to produce sufficient
    quantities of 4CG and 1,6-DCF for toxicological testing, an aqueous
    solution was obtained composed of:  1,6-dichlorofructose (47.5%); 4-
    chlorogalactose (49.1%); 6-chlorofructose (0.3%); 1-chlorofructose
    (0.2%) and TGS (1.2%).

         Since the previous evaluation, the results of additional
    studies and additional information have become available.  At the
    present meeting, the Committee evaluated new and existing data
    which, are summarized and discussed in the following addendum to the


    2.1  Special study on palatability

    2.1.1  Rat

         Groups of 10 female Sprague-Dawley, CD strain rats had free
    access to two  bottles containing either tap water or TGS solution
    at concentrations ranging from 20 2560 mg/ml for 32 days.  Rats
    showed a range of individual variability in their preference for TGS
    vs. water.  As a group, however, they displayed a distinct
    preference for TGS over water at concentrations up to 320 mg/100 ml. 
    At concentrations above 640 mg/100 ml, water was preferred over the
    TGS solution.  There was no mortality and no signs of clinical
    toxicity were observed (Amyes & Aughton, 1987).

         Groups of 20 young female Sprague-Dawley rats, Crl:  CDBR
    strain had access to two feeding jars containing either basal diet
    or a mixture of basal diet containing TGS.  The starting
    concentration of TGS was 50 ppm and was doubled every fourth day to
    a final concentration of 3200 ppm (32 days).  No statistically
    significant difference in the selection of basal diet or TGS
    containing diet was observed at 50 or 100 ppm.  A statistically
    significant preference for TGS diet was observed at 200 ppm.  At the
    400 ppm concentration, basal and TGS-containing diets were consumed
    in similar amounts.  At TGS concentration of 800 ppm or higher, a
    statistically significant preference was displayed for basal diet. 
    The total intake of food by treated rats was similar to that of
    their controls.  No death, nor apparent effect on the appearance,
    behavior, or body weight was reported (Amyes & Aughton, 1988).

    2.2  Short-term studies

    2.2.1  Rat

         Groups of Sprague-Dawley rats were given TGS (dissolved in
    water) by gavage at dose levels of 4000 mg/kg bw/day (15 rats/sex)
    for periods of 4, 9, or 13 weeks, respectively.  Corresponding
    control groups (receiving water) with the same number of animals as
    the treated groups were included.  A significant increase of caecal
    weights was observed in rats of both sexes at all dose levels. 
    Sporadic significant changes in some parameters of haematology and
    clinical chemistry and some differences in organ weights (in some
    groups at some time points) were observed.  The changes were not
    clearly dose-related and not considered as treatment-related.  Other
    investigated parameters were not adversely affected, these included: 
    clinical signs of toxicity, mortality, body weight gains, food and
    water consumption, urinalysis, gross and histopathology (Perry  et
     al., 1988).



    2.1  Biochemical aspects

    2.1.1  Absorption, distribution and excretion

         No information available.

    2.1.2  Biotransformation

         The metabolic fates of 1,6-DCF and its metabolites were
    evaluated and included experiments to determine the potential for
    1,6-DCF to metabolize to 6-chlorofructose (6-CF). 

         An anaesthetized rat with bile duct and ureter cannulas was
    injected intravenously with a single dose of 14C-6-chloro-6-
    deoxyfructose (14C-6CF; 1.1 Ci) at a level of 1.4 mg/kg b.w.  Bile
    and urine samples were collected up to 5 h and analyzed for
    radioactivity.  The radioactivity in the blood was also determined. 
    About 17% of the 14C(14% in urine and 3% in bile) was detected
    within 5 h.  No significant amounts or radioactivity were detected
    in blood.  Thin layer chromatography of urine revealed 5 radioactive
    components, none of which co-chromatographed with 1,6-dichloro-1,6-
    dideoxymannitol (DCM) or with 6-CF.

         Two anaesthetized rats with ureter cannulas were given 14C-
    1,6-dichloro-1,6,dideoxyfructose (14C-1,6-DCF; 4.4 Ci)
    intraduodenally at a dose level of 5 mg/kg bw.  Maximum levels of
    14C in blood occurred 15 min. after dosing.  The level decreased
    with time although radioactivity (about 12% of the dose) was still
    present at 360 min.  The distribution of radioactivity between RBC
    and plasma was constant between 15 and 360 min.  About 80% of the
    14C dosage was associated with the RBC.  About 50% of the 14C
    dosage was eliminated in the urine within 360 min. 

         Three rats with cannulated bile ducts received intravenously a
    single dose of 14C-1,6-dichloro-1,6-dideoxyfructose (14C-1,6-DCF)
    at a level of 100 mg/kg bw and urine samples were collected for the
    following 6 h.  The major urinary metabolite was DCM.  Other
    metabolites were widely distributed over various fractions and were
    not identified (Hughes  et al., 1988).

    2.2  Genotoxicity

    2.2.1  Results of mutagenicity assays with TGS-HP

    Test System    Test Object    Concetration   Result    Reference

    Dominant       Mouse          30-270 mg/kg   Negative  Bootman &
    lethal         gavage 5 day   bw/day                   Whalley, 1983
    assay          (in vivo)


        2.2.2  Results of mutagenicity assays with 1,6-DCF
    Test System         Test Object      Concentration         Results    Reference
                                         of 1,6-DCF
    Micronucleus        Mouse,           415-1600 mg/kg bw     Negative   Ivett, 1988a
    assay               gavage,
                        bone marrow
                        (in vivo)

    Micronucleus        Mouse,           1000-2500 mg/kg bw    Negative   Ivett, 1988b
    assay               gavage,
                        bone marrow
                        (in vivo)

    Sister chromatid    Mouse,           200-2000 mg/kg bw     Negative   Ivett, 1988
    exchanged           gavage,
                        bone marrow
                        (in vivo)

    Covalent DNA        Rat, gavage      21 mg/kg bw           Negative   Martin, 1989
    binding (liver,     (in vivo)
    kidney, small
    intestine, colon,
    stomach, bone


         No information available.

    2.2  Toxicological studies

    2.2.1  Special studies on neurotoxicity  Mouse

         Groups of male CD-1 mice (8 mice/group) were dosed daily (by
    gavage) with 6-chlorofructose (6-CF) (one of the TGS hydrolysis
    products obtained by treating TGS with aqueous hydrochloric acid
    [0.11N] at 68C for 72 h) at dose levels of 240 or 480 mg/kg b.w.
    day for 28 days.  Mice were weighed daily and examined for hind limb
    paralysis at the same time.  Negative control group animals were
    given water while positive controls received 6-chloroglucose (6-CG)
    at a level of 480 mg/kg b.w./day.  The results showed that 6-CF at
    both dose levels caused dose-related hind-limb paralysis in some
    treated animals (240 mg/kg bw/day: 2/8; 480 mg/kg b.w./day: 7/8)
    starting between 4 to 6 days post-dosing (Ford & Waites, 1982).

    2.2.3  Special studies on reproduction  Rat

         Groups of mature male CD rats (3-8 rats/group) were exposed
    orally to 6-CF at dose levels of 6, 12, 18, or 48 mg/kg b.w./day for
    14 days.  Treated male rats were mated with untreated females during
    the final 7 days of dosing.  Mated females were killed 10 days after
    mating and live embryos, resorptions and the corpora lutea were
    counted.  The results showed that males exposed to 6-CF at dose
    levels of 18 or 48 mg/kg b.w./day became infertile (conception rate: 
    control - 79%; treated groups - 0%).  Three weeks after cessation of
    treatment the conception rates had recovered to the level of control
    groups.  The lowest dose level of 6 mg/kg/b.w./day did not have any
    adverse effect on the fertility of male rats (Ford & Waites, 1978a;
    Ford  et al., 1981; Ford & Waites, 1982).

         Three groups of male Sprague Dawley rats of the CD strain (6
    rats/group) were given 6-chlorofructose dissolved in water by gavage
    twice daily at dose levels of 0, 3 or 9 mg/kg b.w./day for 10 weeks. 
    An additional group consisting of 6 males received 6-chlorofructose
    twice daily at a level of 27 mg/kg b.w./day for 4 weeks.  After 7
    days of treatment, males were paired on a one-to-one basis with
    untreated females.  Each morning the females were examined for
    copulation plugs and presence of spermatozoa in a vaginal smear. 
    Quantity and quality of spermatozoa were assessed.  Females which
    had not conceived during the first four days of pairing were

    sacrificed and their reproductive organs examined.  The pairing
    procedure was repeated at weekly intervals for 10 or for 8 weeks
    except for males of the 27 mg/kg b.w./day dose group which were
    paired weekly for 4 weeks.  Males were observed daily for signs of
    clinical toxicity and mating behavior.  Body weight gains were
    recorded weekly.  After completing the series of matings, the males
    were killed, necropsied and the following organs weighed:  testes,
    epididymis, prostate and seminal vesicle.  Testes and epididymis of
    the control and 9 mg/kg b.w./day dose groups were
    histopathologically examined.  On days 8-10  post coitum, females
    were sacrificed and the uterine horns checked for the presence of
    implantation sites.  For each male, mating performance, conception
    rate and fertility index were calculated.

         Males receiving 27 mg/kg b.w./day became infertile for the
    entire treatment period (4 weeks), but regained their fertility
    within the first 7 days of recovery period.  Females mated by males
    of the 9 mg/kg b.w./day dose group showed a reduction in the number
    of implantation sites, indicating a possible decrease in male
    fertility.  The fertility of males of the 3 mg/kg b.w./day dose
    group was not affected (Tesh,  et al., 1984).

    Acute toxicity TGS-HP
    Species     Sex        Route        LD50        Reference
                                     (mg/kg bw)

    Rat         female     oral         4450        Campbell et al.


         At its present meeting the Committee reviewed new and
    previously available data.  Although no new data on the absorption
    and metabolism of TGS in humans were received, the Committee
    concluded that there was no indication that these processes would
    change on prolonged oral dosing.  This conclusion was drawn from the
    comparative metabolic data for TGS in various species, including
    humans, and the lack of evidence of toxicity in extensive animal
    studies.  Nevertheless, the Committee recognized that the data did
    not address all possibilities, particularly the potential effects of
    adaptation of the gastrointestinal microflora.

         No specific studies on possible adverse effects of TGS in
    people with insulin-dependent and maturity-onset diabetes had been
    performed.  However, the Committee decided that this concern could
    be satisfactorily addressed through consideration of data which
    showed that TGS had no effect on the secretion of insulin in humans
    or rats, blood glucose levels or carbohydrate metabolism. 
    Furthermore, the Committee was aware of proposed studies invovling
    both types of diabetics.

         On re-assessment of the overall data on TGS, including the
    metabolic data in various species, including humans and pregnant and
    non-pregnant rabbits, and in the absence of any significant finding
    in the two-generation reproduction study in rats, the Committee
    concluded that the question on the accumulation of TGS in pregnant
    animals and fetuses was satisfactorily addressed, and that there was
    no evidence to suggest a difference in metabolism in pregnant and
    non-pregnant animals.

         The Committee reviewed additional studies relating to the
    possible toxicity of the potential breakdown product of TGS,6-
    chlorofructose.  In a short-term study (28 days) in which 6-
    chlorofructose was administered at 240 and 480 mg per kg of body
    weight per day, male mice showed hind-limb paralysis.  In addition,
    three special studies were conducted to assess reproductive function
    in rats.  Administration of 6-chlorofructose at 18-48 mg per kg of
    body weight per day for 7-14 days caused a loss of fertility in male
    rats.  In two of these studies, the no-effect-levels were 3 and 6 mg
    per kg of body weight per day.  The Committee noted, however, that
    6-chlorofructose is only a potential breakdown product of TGS. 
    While a hypothetical maximum exposure of 1.15 g per kg of body
    weight per day to humans would occur if TGS were subject to extreme
    conditions, such as 0.1 mol/l GC1 at 68 C for 72 hours, the
    Committee expected that exposure to 6-chlorofructose would be
    virtually nil under all foreseeable storage or physiological

         The Committee received a request to reconsider the title
    "trichlorogalactosucrose" it had adopted at its thirty-third meeting
    (Annex 1, reference 83) and to rename it "sucralose".  At that time,
    the Committee established guidelines for designating titles for
    specifications monographs and selected the title in accordance with
    those guidelines.  The Committee considered the information
    submitted at the present meeting and concluded that criteria 1
    (names established by international organizations) and 2 (names
    established by governmental legislation) were not applicable.  It
    did not believe that the information submitted in relation to
    criterion 3 (names established by common usage) was sufficiently
    compelling to change the name, and reaffirmed its view, reached in
    relation to criterion 4 (available scientific, common or trivial
    names), that "trichlorogalactosucrose" was appropriate.  In this
    regard, the Committee noted that the name "sucralose" had been
    designated as a synonym in the specifications monograph.  The
    existing tentative specifications were revised and the Committee
    agreed to deleted the "tentative" classification.


         Finally, the Committee concluded that since the 2-year study in
    rats, which included a period of exposure to TGS  in utero,
    represented a greater proportion of the lifetime of the species than
    did the 1-year study in dogs, the former study should be used for
    the purposes of setting an ADI.  A safety factor of 100 was
    therefore applied to the no-observed-effect level in the long-term
    study in rats (1500 mg per kg of body weight per day), and an ADI of
    0-15 mg per kg of body weight was allocated.

         Additional immunotoxicity studies to assess the significance of
    observed weight changes in the spleen and thymus and changes in
    lymphocyte counts in rats were considered to be desirable.  As yet,
    a causal relationship between these findings and high levels of
    exposure to TGS cannot be excluded.


    AMYES, S.J. & AUGHTON, P. (1987).  Acceptability of aqueous
    solutions of sucralose, saccharin and aspartame to female rats. 
    Unpublished report for Life Science Research Ltd., Suffolk, U.K. 
    Submitted to the World Health Organization by Tate & Lyle.

    AMYES, S.J. & AUGHTON, P. (1988).  An investigation of the
    acceptability of rats of diet containing sucralose.  Unpublished
    report for Life Science Research Ltd., Suffolk, U.K.  Submitted to
    the World Health Organization by Tate & Lyle.

    BOOTMAN, J & WHALLEY, H.E. (1983).  Dominant lethal study in the
    mouse after subacute treatment with an equimolar mixture of the
    hydrolysis products derived from TGS.  Unpublished report from the
    Life Science Research Ltd., Essex, U.K.  Submitted to the World
    Health Organization by Tate & Lyle.

    CAMPBELL, A.H., HARKINS, R.W. & LORD, G.H. (1980).  Acute oral
    toxicity studies in the rat.  Unpublished report from Johnson &
    Johnson Research Foundation, New Brunswick, New Jersey, U.S.A. 
    Submitted to the World Health Organization by Tate & Lyle.

    FORD, W.C.L. & WAITES, G.M.H. (1978a).  Chlorinated sugars:  A
    biochemical approach to the control of male fertility.   Int. J.
     Andrology, Suppl., 2, pp. 541-564.

    FORD, W.C.L. & WAITES, G.M.H. (1978b).  A reversible contraceptive
    action of some 6-chloro-6-deoxy sugars in the male rat.   J. Reprod.
     Fert., 52, 153-157.

    FORD, W.C.L., HARRISON, A. & WAITES, G.M.H. (1981).  Effects of 6-
    chloro-6-deoxysugars on glucose oxidation in rat spermatozoa.   J.
     Reprod. Fert., 63, 67-73.

    FORD, W.C.L. & WAITES, G.M.H. (1982).  Activities of various 6-
    chloro-6-deoxysugars in (S)a-chlorohydrin in producing
    spermatocoeles in rats and paralysis in mice and in inhibiting
    glucose metabolism in bull spermatozoa  in vitro.    J. Reprod.
     Fert., 65, 177-183.

    HUGHES, H.M., CURTIS, C.G. & POWELL, G.M. (1988).  1,6-Dichloro-1,6-
    dideoxyfructose:  metabolism in the rat (supplementary report). 
    Unpublished report from University College, Cardiff, U.K.  Submitted
    to the World Health Organization by Tate & Lyle.

    IVETT, J.L. (1988a).   1,6-Dichloro-1,6-dideoxy-D-fructose:   in
     vivo micronucleus assay in the mouse.  Unpublished report from the
    Hazleton Laboratories America, Inc., Maryland, U.S.A.  Submitted to
    the World Health Organization by Tate & Lyle.

    IVETT, J.L. (1988b).  1,6-Dichloro-1,6-dideoxy-D-fructose:  single
    exposure dose selection study for the  in vivo micronucleus assay
    in the mouse.  Unpublished report from the Hazleton Laboratories
    America, Inc., Maryland, U.S.A.  Submitted to the World Health
    Organization by Tate & Lyle.

    IVETT, J.L. (1988c).  1,6-Dichloro-1,6-dideoxy-D-fructose:   in vivo
    sister chromatid exchange assay in the mouse.  unpublished report
    from the Hazleton Laboratories America, Inc., Maryland, U.S.A. 
    Submitted to the World Health Organization by Tate & Lyle.

    MARTIN, C.M. (1988).  Study to evaluate the covalent DNA binding
    potential of 1,6-dichlorofructose in organs of treated rats. 
    Unpublished report from the Microtest Research Limited, York, U.K. 
    Submitted to the World Health Organization by Tate & Lyle.

    PERRY, C.J., HUDSON, P. & FINN, J.P. (1988).  Sucralose:  Special
    investigation in the rat of the effect of administration by oral
    gavage on body weight and selected organs.  Unpublished report from
    Inveresk Research International, Musselburgh, Scotland.  Submitted
    to the World Health Organization by Tate & Lyle.

    TESH, J.M., ROSS, F.W., WILLOUGHBY, C.R. & SECKER, R.C. (1984).  6-
    Chlorofructose:  Antifertility evaluation in the male rats. 
    Unpublished report from the life Science Research Ltd., Suffolk,
    U.K.  Submitted to the World Health Organization by Tate & Lyle.

    See Also:
       Toxicological Abbreviations