First draft prepared by
    Dr T. Kuiper-Goodman and Dr P.S. Nawrot
    Bureau of Chemical Safety
    Health and Welfare Canada
    Ottawa, Ontario, Canada


         The common potato,  Solanum tuberosum, contains toxic
    steroidal glycoalkaloids derived biosynthetically from cholesterol
    (Sharma & Salunkhe, 1989). In older literature (before 1954) these
    have been referred to only as 'solanine' or as total glycoalkaloids
    (TGA). The potato glycoalkaloids have not been evaluated previously
    by the Joint FAO/WHO Expert Committee.

         Potatoes that have been exposed to light in the field or during
    storage may become green, due to an accumulation of chlorophyll.
    This greening may affect only the surface (peel) or it may extend
    into the flesh of the potato. Exposure to light is only one of the
    stress factors affecting potatoes. Other pre- or post-harvest stress
    factors are mechanical damage, improper storage conditions, either
    as a tuber or after partial food processing, and sprouting (Sharma &
    Salunkhe, 1989).

         As a result of any of the above stress factors, there can be a
    rapid increase in the concentration of TGA, notably, alpha-solanine
    and alpha-chaconine, which gives the potatoes a bitter taste. These
    natural toxicants (stress metabolites) have insecticidal and
    fungicidal properties; each of the two major glycoalkaloids is
    normally present in all tubers in small amounts (< 5 mg/100 g of
    tuber fresh weight) (Table 1). The glycoalkaloids are formed in the
    parenchyma cells of the periderm and cortex of tubers, and in areas
    of high metabolic activity such as the eye regions. The
    glycoalkaloids are unevenly distributed throughout the potato, with
    a large part concentrated under the skin (Table 1). Some cultivars
    are more prone to develop elevated levels of TGA than others.
    Growing conditions may also affect the level of glycoalkaloids. None
    of cooking, baking, frying nor microwaving destroys the
    glycoalkaloids (Bushway & Ponnampalam, 1981).

        Table 1. Normal Levels of TGA in various tuber tissues

                                                 mg/100g FW

         whole tuber                             7.5 (4.3-9.7)
         flesh                                   1.2-5
         skin 2-3% of tuber                      30-60
         peel 10-15% of tuber                    15-30
         bitter tuber                            25-80
         peel from bitter tuber                  150-220

         1 Wood & Young, 1974
         In commercially available potato tubers destined for human
    consumption, as much as 95% of the TGA fraction consists of
    alpha-solanine and alpha-chaconine (Fig. 1) There is usually
    slightly more alpha-chaconine than alpha-solanine. These compounds
    are derivatives of the aglycone solanidine, each containing three
    sugar moieties. Solanidine itself may also be present in potato
    tubers. The remainder of the TGA fraction may consist of other
    glycoalkaloids or their aglycones (Sharma & Salunkhe, 1989). Other
    aglycones include demissidine, tomatidenol and
    5-solanidan-3alpha-ol. Alpha- And -solamarine are examples of
    glycoalkaloids derived from tomatidenol found in potatoes. Through
    plant breeding using wild potatoes other glycoalkaloids, such as
    commersonine and demissine, both derived from demissidine, and
    various leptines, derived from leptinidine, may be introduced
    (Sharma & Salunkhe, 1989).

         The most extensive review on  Solanum and solanine is by
    Jadhav  et al. (1981); other reviews are by Maga (1980), Dalvi &
    Bowie (1983), Morris & Lee, (1984), Morgan & Coxon (1987) and Sharma
    & Salunkhe (1989). Most of the toxicity data deals with
    alpha-chaconine and alpha-solanine. A Nordic view and assessment of
    the health risks from glycoalkaloids in potatoes was recently
    compiled (Slanina, 1990a,b).

    FIGURE 1


    2.1  Biochemical aspects

    2.1.1  Absorption, distribution, and excretion  Mice

         Groups of 4 female Swiss-Webster mice, weighing 20 g, were
    orally administered alpha-chaconine once at a dose of 10 mg/kg bw.
    Animals were sacrificed by exsanguination at 3, 6, 14, 72 and 120 h
    after dosing. Blood was obtained (0.1 ml/time point) by multiple
    incisions into the tail vein. Absorption was slow, with peak values
    in blood (0.82 g/ml) obtained after 14 h. Decrease in blood levels
    was slow, with 0.31 g/ml present at 120 h. The peak level in the
    liver (2.97 g/g) was reached 6 h after dosing. A second, but lower,
    peak of radioactivity was seen in the liver at 120 h, suggesting
    enterohepatic recycling. Cell fractionation studies showed that
    within liver cells, there was no preferential location for
    alpha-chaconine, and there was no binding of alkaloid to isolated
    RNA or DNA fractions (Sharma  et al., 1983).  Rats

         Male Fischer rats (180-250 g) were orally administered
    alpha-solanine, tritiated at the carbon atoms adjacent to the
    nitrogen atom and the double bond (Fig. 1), at a dose of 5 mg/kg bw.
    Blood samples were taken from the abdominal aorta at 1, 3, 6, 12,
    24, 48, 72, and 96 h (2 animals per time point). Radioactivity in
    the gastrointestinal tract started to disappear from 3 to 6 h after
    dosing. During the first 24 h interval the total urinary and faecal
    excretion was 78% of the dose, with most in the faeces. Maximum
    concentrations of radio-activity occurred near 12 h for all tissues,
    with the largest concentration in the kidneys, spleen, liver, and
    lungs, and the lowest concentration in the blood. By 24 h only 10%
    of the administered dose remained, and 12% was unaccounted for
    (presumed to be associated with other organs and the carcass). At
    that time the amount of tritium in the liver represented 1.54% of
    the dose, and for blood this was 0.375% (based on a total blood
    volume of 64.1 ml/kg bw). By 4 days 84% of the dose had been
    excreted by the faecal route, and urinary excretion accounted for
    10% of the dose (Nishie  et al., 1971).

         Male Sprague-Dawley rats (200-300 g) were orally administered
    alpha-chaconine, tritiated at the carbon atoms adjacent to the
    nitrogen atom and the double bond (Fig. 1), at a dose of 5 mg/kg bw.
    Blood samples were taken from the abdominal aorta at 1, 3, 6, 12,
    24, 48, 72, and 96 h (2 animals per time point). alpha-Chaconine was
    poorly absorbed since faecal elimination accounted for 60% of the
    dose within 12 h, and 80% of the dose within 24 h. Urinary excretion

    of tritium was 5% of the dose 3 to 6 h after dosing, and reached a
    plateau of 10% of the dose between 12 and 24 h. Maximum
    concentrations of radioactivity occurred between 6 to 12 h for all
    tissues, with the largest concentration in the liver. Intermediate
    concentrations were seen in the kidneys, spleen, and lungs, and the
    lowest concentrations were seen in the blood, brain and abdominal
    fat. At 24 h after dosing the amount of tritium associated with the
    liver represented 1.29% of the dose, and that with the blood was
    0.17% of the dose (based on a total blood volume of 64.1 ml/kg bw)
    (Norred  et al., 1976).  Hamsters

         Golden hamsters (130-150 g) were orally administered randomly
    tritiated alpha-chaconine at a dose of 10 mg/kg bw. At specified
    times (3, 12, 24, 72, and 168 h) hamsters (3 animals per time point)
    were exsanguinated by cardiac puncture. (The reviewers noted an
    error in reporting and have adjusted the results of the original
    report by changing ng to g). At 3 h after dosing, the highest
    concentration of alpha-chaconine was seen in the intestines,
    including intestinal contents (125 g/g), and this represented 63%
    of the administered dose. By 24 h these values were 75 g/g or 44%
    of the administered dose, and by 168 h they had declined to 1.73
    g/g or 0.92% of the administered dose. Peak blood (1.74 g/ml) and
    peak tissue levels (liver = 27.2 g/g) of alpha-chaconine for most
    tissues were seen by 12 h, and for the heart and kidneys by 24 h. By
    168 h after dosing, blood levels had declined to 0.29 g/ml. The
    ratio of liver concentration to blood concentration at 72 h was
    greater than at 24 h, indicating the possibility of enterohepatic
    recycling. Only small amounts of radioactivity were recovered from
    the faeces in the elimination phase (non-detectable at 3 h, 0.15% at
    24 h, to 0.24% of the administered dose by 168 h). In the urine
    these values increased from non-detectable at 3 h to 0.25% at 12 h,
    and to 21% by 168 h. These results suggest that most of the
    alpha-chaconine was absorbed, but that absorption from the
    gastrointestinal tract was slow. Much of the radioactivity appeared
    in various tissues in bound form (Alozie  et al., 1979a).  Humans

         Tritiated solanidine (dose not given, but expressed as
    radioactivity) was administered to 3 human volunteers (2 males, 1
    female) by iv injection. Blood and urine samples were collected at
    various times up to 150 h. Ninety per cent of tritium had
    disappeared from the blood within 20 minutes of injection. Presuming
    that radioactivity represented solanidine or its metabolites, three
    phases of elimination were identified in plasma with half-lives of 2
    to 3.7 min, 2 to 5 h, and 72 to 104 h, respectively. Within minutes
    of injection, the concentration of tritium in erythrocytes exceeded
    that in plasma. Erythrocytes were found to be a mobile reserve of

    solanidine, thereby delaying transfer of solanidine from vascular to
    extravascular compartments. Low rates of excretion were seen in
    urine and faeces, and together accounted for about 5% of the
    administered dose during the first 24 h. Thus a fraction in excess
    of 90% of the dose was sequestered somewhere in the body 24 h after
    dosing. After this time, the rate of elimination from the body was
    low, about 1-2% per day, corresponding to an overall half-life of 34
    to 68 days. The authors calculated that if absorption of solanidine
    were 1 mg/day, then with a fractional rate of excretion of 0.02, the
    body burden would be 50 mg. The authors suggested that mobilization
    from various storage loci could occur during times of 'metabolic
    stress', including pregnancy (Claringbold  et al., 1982).

         Mean levels of 1.56  1.17 (7 males) and 1.20  0.93 (27
    females) ng/ml solanidine were found, using radioimmunoassay, in
    human plasma samples obtained by a hospital clinic in the UK,
    collected in the morning before lunch (Matthew  et al., 1983).

         Thirty healthy males, aged 18-44 years, and 27 healthy females,
    aged 16-62 years, participated in a study in the UK designed to
    measure levels of serum solanidine in persons eating their usual
    diet (during the winter). Intake of the type of potato product
    (i.e., French fried, boiled or baked, and whether the skin was
    included) was recorded daily for one month, with arbitrary units,
    corresponding to approximate levels of TGA in those products,
    assigned to each product; the weight of product ingested was not
    measured. Serum samples were collected before the midday meal, and
    were analyzed by radioimmunoassay (detection limit 0.5 ng/ml). In
    males the mean level of solanidine was 10.8  5.4 ng/ml (range
    2.1-22.5 ng/ml), whereas in females the respective values were 7.9 
    4.3 (range 1.6-18.5). For both genders there was a significant
    correlation between serum solanidine levels and the alkaloid intake
    (expressed in units as indicated above) during the month (R = 0.878
    and R = 0.703, respectively). In two male subjects serum solanidine
    levels dropped to 0.5 ng/ml 2 to 3 weeks after they had been on a
    potato avoidance diet, indicating a relatively long serum half-life
    for solanidine. It was suggested by the authors that solanidine may
    be bound to blood constituents such as free sterols (Harvey  et al.,

         Eighteen healthy males, aged 20-45 years, and 15 healthy
    females, aged 19-63 years, from the London area in the UK,
    participated in a study designed to measure levels of total serum
    alkaloids (alpha-solanine + alpha-chaconine + solanidine) and
    solanidine in persons eating their usual diet (during the summer).
    For comparison, 5 males, aged 31-41 years, and 5 females, aged 31-67
    years, from the Uppsala area in Sweden also participated in this
    study. In Sweden, 2 of the males and 1 female consumed 200-300 g
    potatoes of 2 varieties high in TGA, including the skin, for 1 week
    (mean 24 mg TGA/100 g), giving an intake of approximately

    60 mg/person or 1 mg/kg bw/day. Blood samples were collected before
    the midday meal, and were analyzed by radio-immunoassay (detection
    limits for total alkaloids and solanidine were 0.4 and 0.5 ng/ml
    serum, respectively). The mean levels of serum solanidine were,
    respectively, 3.5 and 4.0 ng/ml in the UK and Swedish subjects
    eating their usual diets, whereas in those three Swedes consuming
    potatoes with a higher TGA content the mean serum solanidine level
    was 31 ng/ml (range 27.8-35.5). The respective serum total alkaloid
    levels were 12.0, 16.9 and 50 ng/ml. The mean serum total alkaloid
    concentration was about 2.7 times the solanidine concentration,
    which, according to the authors, suggests considerable metabolism in
    man of the glycoalkaloids alpha-chaconine and alpha-solanine (they
    represent the major proportion of alkaloids in potatoes) through
    hydrolysis of the sugar residues. It was suggested that hydrolysis
    could take place in the acid medium of the stomach, or at the site
    of absorption, or the ratio could reflect the preferential
    absorption of the more lipophilic solanidine. Alternatively,
    alpha-solanine and alpha-chaconine might be absorbed unchanged and
    metabolized within the body (Harvey  et al., 1985b).

         Blood serum levels of alpha-solanine, alpha-chaconine, and
    solanidine resulting from a single meal of mashed potatoes
    (equivalent to 1 mg TGA/kg bw/day) were monitored in 8 healthy
    subjects (HPLC, detection limit 1 ng/ml). Peak concentrations were
    achieved after 4-8 h; these were 3-11 ng/ml for alpha-solanine and
    6-21 ng/ml for alpha-chaconine. The 1:2 ratio was maintained for the
    duration of the experiment. After longer time intervals the level of
    solanidine was < 4 ng/ml. The serum half-lives for alpha-solanine
    and alpha-chaconine were 11 and 19 h, respectively (unpublished data
    by K.E. Hellens, cited by Slanina, 1990b).

    2.1.2  Biotransformation  Rats

         Male Fischer rats (180-250 g) were orally administered 5 mg/kg
    bw solanine, tritiated at the carbon atoms adjacent to the nitrogen
    atom and the double bond (Fig. 1). Approximately 65% of the
    radioactivity in the faeces was identified as solanidine. In urine
    72% of radioactivity was present as basic compounds of which 6% was
    identified as solanidine. Two other compounds, present at 80% and
    13%, possessed intermediate polarity with respect to solanine and
    solanidine (Nishie  et al., 1971).

         Male Sprague-Dawley rats (200-300 g) were orally administered
    alpha-chaconine, tritiated at the carbon atoms adjacent to the
    nitrogen atom and the double bond (Fig. 1) at a dose of 5 mg/kg bw.
    Urine and faecal samples were collected 24 h later. The major
    constituent in both faeces and urine was presumed to be solanidine
    because it showed the same Rf. Similarly, 25% of the radioactivity

    in the faeces was attributed to unchanged alpha-chaconine. In
    addition, 2 minor compounds, possessing intermediate polarity
    between solanidine and alpha-chaconine and representing 1-5% of
    total activity, were found in faecal and urine extracts. The authors
    concluded that the absorption and metabolism of alpha-chaconine was
    similar to alpha-solanine (Norred  et al., 1976).  Hamsters

         Golden hamsters (130-150 g) were orally administered randomly
    tritiated alpha-chaconine at a dose of 10 mg/kg bw. At specified
    times (3, 12, 24, 72, and 168 h), hamsters were exsanguinated by
    cardiac puncture (3 animals per time point) (see above Alozie
     et al., 1979a). Thin-layer chromatographic separation was
    performed on the chloroform soluble fractions from urine and faeces
    collected at various time intervals after dosing. In urine, over
    half of the eliminated radioactivity during the initial 24 h was due
    to unaltered alpha-chaconine. A major urinary metabolite was
    solanidine, which was the major peak by 72 h. In addition, 4 other
    unidentified metabolites were present at various concentrations. Two
    of these were the major peaks by 168 h after dosing. In faeces, much
    of the eliminated radioactivity was due to alpha-chaconine, and a
    major metabolite was solanidine. There were 2 additional
    unidentified metabolites present in about the same concentration as
    solanidine (Alozie  et al., 1979b).

    2.1.3  Effects on enzymes and other biochemical parameters

         Groups of male Sprague-Dawley rats (5/group) were fasted
    overnight and then given alpha-solanine by gavage at 0 and 250 mg/kg
    bw or i.p. at 0 and 20 mg/kg bw. In the orally dosed animals serum
    glutamic oxalacetic transaminase (SGOT) and serum glutamic pyruvic
    transaminase (SGPT) were increased, and cholinesterase activity was
    decreased, but the differences were not statistically significant.
    With the i.p.-dosed animals statistically significant increases of
    29 and 63% in SGOT and SGPT, respectively, and a 27% decrease in
    cholinesterase activity were observed. In addition, a significant
    inhibition of liver benzphetamine N-demethylase activity and a
    decrease in liver cytochrome P-450 were observed after i.p. dosing,
    whereas after oral dosing these differences were statistically
    insignificant (Dalvi, 1985).

    2.2  Toxicological studies

    2.2.1  Acute toxicity studies

         Oral LD50 values for solanine in rodents are considerably
    higher than LD50 values determined after intraperitoneal
    administration (Table 2), probably because these species do not
    absorb much of the solanine. Post-mortem examination failed to

    reveal the cause of death in rats that had been dosed orally
    (stomach tube). The oral LD50 values in rodents were 300 to > 500
    times the toxic dose of about 2 mg/kg bw and a lethal dose of 3 to
    6 mg/kg bw estimated for humans (see Section 2.3.1).

    Table 2. LD50 values in mg/kg bw

                             alpha-solanine       alpha-chaconine            solanidine

                        i.p.            p.o.           i.p.                  i.p.

    Mice                 32.31         >10002          19.25                 >5002
                         42.02                         27.56

    Rat                 753             5903

    Rabbit              <201,2                          506

    Rhesus              <407

    1 Patil et al. (1972)
    2 Nishie et al. (1971)
    3 Gull et al. (1970)
    4 Chaube & Swinyard (1976)
    5 Sharma et al. (1979)
    6 Nishie et al. (1975)
    7 Swinyard & Chaube (1973)
         Two rhesus monkeys died 48 h after an i.p. injection of
    40 mg/kg bw of total glycoalkaloids; one other died 2 h after having
    been dosed i.p. twice (24 h apart) with 20 mg solanine/kg bw
    (Swinyard & Chaube, 1973).

    2.2.2  Short term studies  Rabbits

         One group of 4 rabbits, each weighing about 950 g (strain not
    given), was fed normal potatoes (TGA content 7.5 mg/100 g) and 4
    more rabbits were fed greened potatoes (TGA content 20.4 mg/100 g)

    for a period of 20 days. After 4 to 6 days the latter group,
    consuming 49-53 mg TGA/kg bw/day, became dull and inactive; after 10
    days, diarrhoea, hair loss and weight loss occurred, which was
    followed by watering eyes, body rigidity and dullness. Protein
    digestibility (amount of protein from potatoes ingested less amount
    of protein excreted expressed as a percentage of protein from
    ingested potatoes) decreased by 45% from day 1. This was accompanied
    by a significant decrease in body weight, resulting in an average
    body weight of about 650 g for treated and 1150 g for 'control'
    rabbits at 25 days after cessation of feeding the experimental
    diets. One of the 4 treated rabbits died within 10 to 20 days. The
    control animals which consumed 20 to 23 mg TGA/kg bw/day were
    unaffected (Azim  et al., 1983).

         The same authors similarly fed 2 groups of 5 rabbits each
    (weight and strain not given) a control potato diet (7.5 mg
    TGA/100 g) and a high TGA potato diet (29.75 mg TGA/100 g) for 45
    days. Daily intake of TGA was 16.8 to 17.9 mg/kg bw in the control
    diet, and 73.9 to 75.0 mg/kg bw in the high TGA diet. Blood samples
    were collected from the ear vein every 15 days. RBC counts and
    haemoglobin concentrations were determined. A significant decrease
    in RBC counts was seen throughout feeding the high TGA diet, and
    this was 27.5% by 45 days. This compared to a decrease of 12.5% seen
    by 45 days in the control diet. Decreases in haemoglobin
    concentrations paralleled the findings with RBC counts. The authors
    suggested that these results indicate that the rabbits developed
    haemolytic anaemia. This could be explained by the metabolite
    solanidine increasing the permeability and fragility of RBC
    membranes (Azim  et al., 1984).  Monkeys

         Four time-mated rhesus monkeys were fed  ad libitum a diet of
    diced potatoes of the B5141-6 variety (since withdrawn from the
    market), containing on average 26 mg solanine per 100 g tuber for 25
    consecutive days during days 0-42 following mating. It subsequently
    became apparent that the monkeys were not pregnant. The monkeys
    ingested the equivalent of 0.77 to 1.02 mg/kg bw/day alpha-solanine
    or 3.08 to 4.07 mg/kg bw/day TGA. No adverse effects were observed
    (Swinyard & Chaube, 1973).

    2.2.3  Long-term/carcinogenicity studies

         No studies available.

    2.2.4  Reproduction studies  Rats

         Groups of Holzman rats, approximately 4 months of age, were
    mated, one male to 3 females. Increase in weight was taken as an
    indication of pregnancy, and afterwards the females were
    individually caged and given a basal diet of (I) ground lab chow;
    (II) ground lab chow to which was added 10% ground frozen potato
    sprouts; (III) 30 mg/kg diet solanine (commercial); (IV) 40 mg/kg
    diet solanine (commercial); and (V) 30 mg/kg diet solanine that was
    isolated from the frozen sprouts. The time on the test diets was
    variable, since increase in weight is not a sensitive indicator of
    pregnancy, and some dams dropped their litters within a few days on
    the test diet. They were then kept on the test diet until they had a
    second litter. No food consumption records were kept, but the rats
    readily ate the diets. With diets II to V, many of the pups died
    within 3 days of birth, evidently from starvation as indicated by an
    absence of milk in their stomachs. The percentage of pups
    successfully weaned was 82.6, 50.6, 31.0, 31.1, and 19.5 for diets I
    to V, respectively. All of the pups in 18/33 litters born of the
    rats eating the test diets died before reaching weaning age, whereas
    only one of 11 control litters was lost. The authors concluded that
    the toxicity of the potato sprout diet was due to 'solanine'. It was
    speculated by the authors that solanine may exert an anti-hormonal
    effect and prevent lactation in some sensitive dams. The reviewers
    feel that further studies to examine these effects and other aspects
    of reproduction are necessary (Kline  et al., 1961).

    2.2.5  Special studies on embryotoxicity/teratogenicity  Rats

         Potential teratogenicity of alpha-solanine and alpha-chaconine
    was investigated in four different experiments with Wistar rats
    weighing 175-200 g. In the first experiment, three groups of rats
    (9-10/group) were gavaged with alpha-solanine at dose levels of 0.3,
    1.0 or 3.0 mg/kg bw/day, from days 6 to 15 of gestation. In the
    second experiment, a group of 9 rats was given alpha-solanine by
    gavage at a dose level of 6 mg/kg bw/day, from days 7 to 10 of
    gestation. In the third experiment, groups of rats (3-4/group)
    received alpha-solanine at dose levels of 2, 10 or 25 mg/kg bw/day
    from days 8 to 11 of gestation. In the fourth experiment, a group of
    4 rats received alpha-chaconine by gavage at a dose level of
    1.5 mg/kg bw/day, from days 6 to 15 of gestation. In the first three
    experiments, concurrent control groups composed of 2 to 10
    rats/group were included. On day 22 of gestation, all females were
    sacrificed and the following parameters were investigated: corpora
    lutea, resorption sites, litter size, litter weight, and gross,
    visceral and skeletal fetal anomalies. The only adverse effect was
    observed in the first experiment. One fetus with craniorachischisis

    and exopthalmos (1/117), from the group receiving 3 mg/kg bw/day,
    and another with twisted pelvic limbs and absent tail (1/108), from
    the 0.3 mg/kg bw/day group were observed. No maternal toxicity was
    reported. The authors concluded that the observed effects were not
    treatment-related (Ruddick  et al., 1974).

         A group of 14 Wistar rats, weighing 175-200 g, was fed a diet
    containing about 73% of cooked and freeze-dried, visibly blighted
    parts of potato tubers, from days 1 to 22 of gestation. The intake
    of blighted potatoes was approximately 70 g/kg bw/day. The control
    group of 13 females was fed a diet containing the same amount of
    freeze-dried potatoes inoculated with heat-killed  Phytophthora
     infestans. The content of glycoalkaloids in the diets was not
    determined. All dams were sacrificed on day 22 of gestation and the
    standard parameters (corpora lutea, resorption sites, litter size,
    litter weight, and gross, visceral and skeletal anomalies) were
    investigated. There was no evidence of maternal toxicity, fetal
    toxicity nor teratogenicity (Ruddick  et al., 1974).  Hamsters

         Groups of hamsters (12-15/group), weighing about 100 g, were
    fed from days 5 to 10 of gestation diets of commercial hamster
    ration containing 50% freeze-dried, unblighted potato concentrate
    (group 1); 50%  Phytophthora infestans infected freeze-dried,
    blighted potato concentrate (group 2); or 50%  Alternaria solani
    infected freeze-dried, blighted potato concentrate (group 3). A
    group of 13 hamsters was fed commercial hamster ration only,
    throughout gestation (group 4). The content of glycoalkaloids in the
    diets was not determined. Food and water were provided  ad libitum.
    On day 15 of gestation, the dams were sacrificed, and fetuses were
    examined for gross, visceral and skeletal anomalies. Feed
    consumption, maternal body weight gain, litter size, number of
    resorptions and fetal weight were not affected by the treatment. The
    most frequent gross anomaly was haemorrhagic necrosis of the central
    nervous system, but the frequency of this effect was not
    treatment-related (group 1 - 1/114; group 2 - 3/153; group 3 -
    0/135; group 4 - 11/99) (Sharma  et al., 1978).

         Groups of Syrian hamsters (body weights and age not given) were
    gavaged on day 8 of gestation with alpha-chaconine, isolated from
    Arran Pilot potato sprouts, at levels of 165 mg/kg bw/day (23/group)
    or 180 mg/kg bw/day (14/group) and alpha-solanine, isolated from
    Arran Pilot potato sprouts, at a level of 200 mg/kg bw/day
    (37/group). Females of the vehicle control group (37/group) received
    vehicle material alone (2% ethanol at pH 5-6 in 1% carboxymethyl
    cellulose or in water). Animals were individually caged in a room
    maintained at 20-26 C and received food and water  ad libitum.
    Maternal toxicity was monitored by daily weighing and clinical
    observation. On day 15 of gestation, the dams were sacrificed and
    necropsied. The uteri were exposed and the number of resorptions and

    live fetuses were determined. The  corpora lutea were counted and
    all fetuses were examined for gross anomalies. Maternal mortality
    was observed in all treated dose groups; 4 and 6 dams died in the
    165 and 180 mg/kg bw/day alpha-chaconine dose groups, respectively,
    and 3 dams died in the 200 mg/kg bw/day alpha-solanine dose group.
    The days of gestation on which the dams had died were not indicated.
    No maternal mortality was observed in the vehicle control group. A
    high incidence of neural tube defects such as interparietal
    encephalocoele (11-13%) and exencephaly (5-12%) was observed in
    fetuses, of which the mothers were exposed to either alpha-chaconine
    or alpha-solanine. In the vehicle control group only 1 fetus of 393
    examined exhibited exencephaly. The authors concluded that the
    observed teratogenic effects were treatment-related. They also
    indicated that a number of fetuses exhibited CNS malformations
    without apparent toxicity or weight loss in the dam, making it
    unlikely that the malformations were secondary to maternal toxicity.
    Occasional short tail and minor digital anomalies were noted in
    fetuses from all experimental groups, but those effects were not
    treatment-related (Renwick  et al., 1984).  Rabbits

         Groups of New Zealand rabbits (2-6/group), weighing on average
    4 kg, were fed, throughout gestation, diets containing 50%
    freeze-dried, unblighted, potato concentrate, 50%  Phytophthora
     infestans infected freeze-dried, blighted, potato concentrate, or
    50%  Alternaria solani infected freeze-dried, blighted, potato
    concentrate. The content of glycoalkaloids in the diets was not
    determined. Prior to parturition (day not defined), all dams were
    sacrificed and fetuses removed and examined for gross, visceral and
    skeletal anomalies. During fetal examination particular attention
    was paid to any malformations of brain and spinal cord. Among 21
    fetuses examined in the  Phytophthora infestans blighted potato
    group, three fetuses (from three litters) exhibited incomplete
    closure of the caudal vertebral column, and two other fetuses were
    very small and had shortened appendages. Among 28 fetuses examined
    in the  Alternaria solani blighted potato group, two fetuses
    exhibited incomplete closure of the caudal vertebral column, one
    fetus had a very small brain (nearly half the normal size) and the
    cranial cavity was filled with fluid, and two other fetuses were
    abnormally small in size. All six abnormal fetuses were from
    different litters. None of the nine fetuses (two litters) from the
    unblighted potato group were affected. The authors concluded that
    feeding pregnant rabbits potatoes blighted with either of the fungi,
    at high concentrations in the diet, can produce a low incidence of
    the caudal vertebral column malformation. This result must be
    considered with caution since the small number of control litters
    examined does not permit an adequate estimate of the spontaneous
    incidence of this malformation in rabbits (Sharma  et al., 1978).  Miniature swine

         Groups of female miniature swine (2/group), weighing about
    39 kg, were fed laboratory diets containing 50% freeze-dried,
    unblighted, potato concentrate (group 1); 50%  Phytophthora
     infestans infected freeze-dried, blighted, potato concentrate
    (group 2); or 50%  Alternaria solani infected freeze-dried,
    blighted, potato concentrate (group 3), during the first half of
    gestation (about the first 57 days of gestation). The content of
    glycoalkaloids in the diets was not determined. At the end of
    gestation (the day was not indicated), all dams were sacrificed and
    necropsied. All fetuses were removed and examined for gross and
    visceral malformations. Depressed weight gain was observed in sows
    of group 3. One fetus of 15 examined from group 2 exhibited
    anencephaly with extensive internal hydrocephaly. Other fetuses from
    this and other groups were not affected. The authors concluded that
    feeding potatoes blighted with  Phytophthora infestans may be a
    causative factor in the production of anencephaly in miniature
    swine. However, the small sample size makes definite conclusions
    difficult (Sharma  et al., 1978).  Marmosets

         A group of 6 female marmosets  (Callithrix jacchus), five
    years of age, weighing about 375 g, previously producing normal
    offspring, was fed a diet containing freeze-dried concentrate of
    blighted potatoes (Kerr's Pink variety), at a level of 4.7 g/kg
    bw/day (equivalent to 0.9 mg/kg bw/day of glyco-alkaloids), for 50
    days, during either days 0-50 or 20-70 of gestation. A control group
    of 6 pregnant marmosets received a standard unsupplemented diet.
    Marmosets were sacrificed between days 80-120 of gestation and
    fetuses were examined for developmental anomalies. Four of 11
    fetuses in the blighted potato groups exhibited gross abnormalities,
    described as cranial osseous defects. Histological examination
    revealed replacement of bone by a collagenous membrane in the
    occipital area. Brain examination of affected fetuses revealed
    enlargement of the lateral ventricle. Eleven fetuses of the control
    group showed no gross abnormalities in any system. The investigators
    indicated that, during the 2 years of existence of the marmoset
    colony, similar defects had occurred once spontaneously in twin
    fetuses (spontaneously aborted) among 104 live births. The authors
    concluded that the occurrence of cranial dysplasia in 4 of 11
    fetuses, in the experimental group, is suggestive of teratogenicity
    of blighted potatoes in marmosets (Poswillo  et al., 1972, 1973).

         Three experiments were conducted to investigate possible
    teratogenic effects of different varieties of unblemished or
    blemished potatoes in 5 year-old, pregnant marmosets  (Callithrix
     jacchus), previously producing normal offspring. In the first
    experiment, a group of 6 female marmosets were fed freeze-dried

    concentrate of unblemished 'domestic potatoes' (Cornish White and
    King Edward varieties), at a level of about 4.7 g/kg bw/day,
    equivalent to 0.56 mg/kg bw/day of glycoalkaloids. In the second
    experiment a group of 7 female marmosets were fed freeze-dried
    concentrate of blemished 'industry rejected potatoes' (King Edward,
    Cornish White varieties), at a level of about 4.7 g/kg bw/day,
    equivalent to 0.78 mg/kg bw/day of glycoalkaloids. In the third
    experiment a group of 5 female marmosets were fed freeze-dried
    concentrate of King Edward variety potatoes, infected with  Erwinia
     carotovera (a bacterial pathogen responsible for 'blackleg'), at a
    level of about 4.7 g/kg bw/day, equivalent to 0.07 mg/kg bw/day of
    glycoalkaloids. Feeding trials were commenced 10 days postpartum of
    the previous litter, to cover the period of the expected postpartum
    oestrous and were carried throughout an undefined period of
    gestation. Female marmosets fed both the 'domestic potatoes' and
    'industry rejected potatoes' diets were allowed to proceed with
    pregnancy to term, and the offspring was grossly examined at birth
    and at regular intervals up to 6 months of age. Females fed
    'infected potatoes' (with  Erwinia carotovora) diet were sacrificed
    between days 90 to 110 of gestation, and fetuses were examined
    grossly and radiographically for abnormalities. Behavioural
    anomalies such as continuous clinging to parents or siblings, and
    prolonged weaning time, were observed in three sets of twins, born
    to dams of the second experiment. No anatomical abnormalities were
    observed in any experimental groups. The authors concluded that the
    significance of the behavioural abnormalities observed in this study
    cannot be determined at this stage but further observation of growth
    and development to sexual maturity may throw more light on this
    phenomenon (Poswillo  et al., 1973).  Chicken embryos

         Fertile chicken eggs (White Leghorn) were injected either with
    pure solanine, mixed glycoalkaloids or an ethanol extract (obtained
    from potatoes infected with  Phytophthora infestans) into the yolk
    sac, at levels ranging from 0.13 to 0.26 mg/egg, between 0 and 26 h
    of incubation. A high incidence of embryo mortality (20-27%) and
    increased incidence of abnormalities (16-25%) such as cranioschisis,
    celosoma, cardiac septal defects, rumplessness (absence of tail) and
    trunklessness (absence of trunk below the wing bud) were observed in
    treated embryos. The most frequent defect was rumplessness and
    trunklessness. In controls injected with chick Ringer or HCl
    solvent, the percentage of abnormal embryos was 9-10% and the
    mortality was 1-8% (Jelinek  et al., 1976; Mun  et al., 1975).

    2.2.6  Special studies on cholinesterase inhibition

         The inhibitory effect of alpha-solanine and solanidine, as well
    as an extract from potatoes, were studied using a 1:100 dilution of
    sera from 21 human individuals. These persons had been previously

    phenotyped as 'usual' (95% of population in Great Britain),
    'intermediate' (3-4% of the population), and 'atypical' (uncommon),
    using the acetylcholinesterase inhibitor dibucaine. At a
    concentration of 2.88 M and 3.14 M, respectively, alpha-solanine
    and solanidine were about equally effective causing 86.2  1.2 % and
    80.0  1.4 % inhibition in the 'usual' phenotype, and parallel
    effects to dibucaine in the other two phenotypes. The results with
    the potato extract were similar. The authors noted that it is not
    clear to what extent the toxic effects of solanine can be attributed
    to the inhibition of serum cholinesterase, but if it plays a role
    then individuals with the 'atypical' phenotype, would presumably be
    less susceptible (Harris & Whittaker, 1962).

         Male and female New Zealand rabbits (2 of each sex) were given
    a single i.p. dose of 20 or 30 mg solanine/kg body weight. These
    doses resulted in severe depression, with difficult breathing and
    prostration, and were lethal in 3 of the rabbits within 24 h. One
    rabbit survived. Blood samples were obtained at 15 to 225 min
    post-dosing; plasma and erythrocyte acetylcholinesterase activities
    were measured and compared to control samples taken from the same
    rabbit before dosing. Solanine was a weak to moderate inhibitor of
    both specific and non-specific cholinesterase. Maximum inhibition of
    plasma cholinesterase was seen at 80 min after injection, with the
    activity decreasing to about 45% of the control value; inhibition of
    erythrocyte cholinesterase was somewhat lower, and was maximally
    reduced to 68.6% at 85 min after injection.

         The same authors injected i.v. 5 doses of 6 mg/kg body weight,
    10 min apart, in one anaesthetized male dog (15 kg bw). Quick
    inhibition of serum cholinesterase was followed by rapid recovery.
    Erythrocyte cholinesterase was not inhibited (Patil  et al., 1972).

         Male Sprague-Dawley rats (3 per group) were injected i.p. with
    0, 10, 30 or 60 mg/kg bw of alpha-chaconine and sacrificed 3 h after
    dosing. All rats administered alpha-chaconine showed initial signs
    of depression, as well as other signs of poisoning by an
    anticholinesterase agent, such as respiratory depression. Following
    electrophoresis in acrylamide slabs, homogenates of brain (diluted
    1:6) showed 3 zones of acetylcholinesterase isoenzyme activity with
    a dose-related decrease in peak heights. Overall
    acetylcholinesterase activity, using a colorimetric method, was
    reduced to 79, 55, and 18% of the control value for the 3 respective
    dose groups. Heart acetylcholinesterase activity was reduced to
    about 40% of the control value in all treatment groups; plasma
    cholinesterase activity in controls was about 30% of that seen in
    brain homogenates, and was reduced to 50% in the 10 mg/kg bw dosage
    group, with no further reduction in rats given 30 mg/kg bw. The
    authors concluded that alpha-chaconine is a fairly potent inhibitor
    of cholinesterases (Alozie  et al., 1978).

         In an  in vitro assay the anticholinesterase activity of
    several glycoalkaloids was compared, using highly purified acetyl
    cholinesterase isolated from human and bovine erythrocytes (Sigma).
    Alpha-solanine and alpha-chaconine were equally effective, 100 M
    caused about 80% inhibition of both human and bovine enzymes.
    Tomatine was less effective, causing 40% and 50% inhibition of
    bovine and human enzymes, respectively. Solasine, solamargine and
    the aglycones solanidine, tomatidine and solasidine were
    ineffective. Over a range of Ph 5 to pH 8, pH of the medium was not
    very important. These results show that the nature of the aglycone
    moiety is important (Roddick, 1989).

    2.2.7  Special studies on genotoxicity

         Pure alpha-solanine (Sigma) at 0.01 to 0.05 mg/plate and
    extracts from potatoes were negative in the Ames test both with
    strains TA98 and TA100, and in the presence or absence of activation
    by S9 fraction from PCB-induced rat liver (Ness  et al., 1984).

         Alpha-solanine (25 and 250 M) tested negative in a
    DNA-cell-binding assay using Ehrlich ascites cells and  Escherichia
     coli cells mixed with 32P-labelled nucleic acids (Kubinski  et
     al., 1981).

    2.2.8  Special studies on mitotic index

         Cultured human fibroblasts were treated up to 40 h with 0, 4.1,
    8.3, 16.6, 33.3, and 66.6 g alpha-solanine/ml. At the highest dose
    there was an inhibition of growth, whereas at lower dose levels
    there was a stimulation of growth, as evidenced by an increase in
    the mitotic index from 1.7% in controls to 2.4% at the 4.1 g/ml
    dose level, which according to the authors was similar to the
    sex-hormonal type of effect exerted by estrogens on target tissues.
    Using pulse labelling with tritiated thymidine, it was shown that at
    5 g alpha-solanine/ml the mean cell cycle time decreased from 42.5
     1.87 h in controls to 28.5  0.29 h in treated cells. This was
    however accompanied by a 4 h increase in the period of DNA synthesis
    (S phase) in treated cells, and a decrease to virtually zero for the
    G1 phase. The authors concluded that if alpha-solanine reached the
    fetus, the observed types of effects could be hazardous to it, and
    could lead to malformations (Kirk & Mittwoch, 1975).

    2.2.9  Special studies on calcium transport

         Alpha-solanine (100 M at pH 7.4) caused a 90% inhibition of
    active calcium transport in rat duodenum when added to everted
    intestinal sacs (8 replicates)  in vitro. A Dixon plot revealed
    that the inhibition by alpha-solanine was non-competitive, and the
    inhibition constant was 25 M. The inhibition of active calcium
    transport was accompanied by a 40% decrease in oxygen consumption
    (Michalska  et al., 1985).

         When alpha-solanine was given to 12 male and female Wistar
    albino rats (5-6 weeks old) in their drinking water (5 mM, pH 6.4)
    for 12 days, calcium transport in duodenal sacs was reduced to about
    one-third of the control value, but oxygen consumption was not
    significantly reduced (Michalska  et al., 1985).

    2.3  Observations in humans

    2.3.1  Gastrointestinal and neurotoxic effects

         There have been many reported cases of human poisonings
    (sometimes fatal) due to the ingestion of greened or otherwise
    damaged potatoes. The symptoms of low grade solanine poisoning are
    acute gastrointestinal upset with diarrhoea, vomiting and severe
    abdominal pain. In more severe cases, neurological symptoms,
    including drowsiness and apathy, confusion, weakness, and vision
    disturbances, followed by unconsciousness and, in some cases, death
    have also been reported. The vital signs include fever, rapid and
    weak pulse, low blood pressure and rapid respiration. Onset of
    symptoms has ranged from minutes to 2 days after ingestion of toxic
    potatoes, with longer incubation periods generally associated with
    the more severe cases.

         As is usual with case histories of this nature, the available
    data are not complete. Over the years, various analytical methods or
    assays have been used to determine the concentration of 'solanine'
    in cases of suspected poisonings. With most of the older data, the
    estimate for solanine included the other glycoalkaloids, such as
    alpha-chaconine. McMillan and Thompson (1979) showed that
    gravimetric methods gave higher values than colorimetric methods. A
    few case reports, for which the reviewers have estimated the dose
    ingested, are given below. Additional reports were compiled by
    Morris and Lee (1984), who indicated that more than 2000 cases with
    about 30 deaths have been reported in the literature. Not all of
    these reports were available to the reviewers.

         Fifty-six German soldiers suffered typical 'solanine' poisoning
    after eating 1 to 1.5 kg cooked peeled potatoes containing 24 mg
    TGA/100 g (whole uncooked tubers contained 38 mg TGA/100 g). In a
    few cases jaundice and partial paralysis were also observed. If one
    assumes a body weight of 70 kg, the intake of 'solanine' was 3.4 to
    5.1 mg/kg bw (Pfuhl, 1899).

         In 18 separate households in Scotland, 61 persons suffered
    typical 'solanine' poisoning soon to several hours after eating
    potatoes. Persons not eating potatoes were not ill. One 5-year old
    died. The potatoes in that household contained 41 mg 'solanine'/100
    g. Assuming the child ate 200 g potatoes and had a bw of 18 kg, the
    lethal dose was estimated at 4.5 mg/kg bw. Assuming adults ate 500 g
    potatoes and had a bw of 60 kg, their intake of 'solanine' would
    have been 3.4 mg/kg bw (Harris & Cockburn, 1918).

         A small outbreak of solanine poisoning affected a family of
    four adults on three consecutive Sunday evenings in Great Britain,
    about 8 h after they had eaten 1 to 3 baked potatoes in their
    jackets (weight of potatoes not given). A 5th person who only ate
    the flesh of the potatoes was not affected. The severity of symptoms
    was related to the number of potatoes ingested, and consisted of
    abdominal pain, diarrhoea, and general malaise. Patients recovered
    within 24 h. The level of solanine was 50 mg/100 g tuber, as
    determined chemically and by cholinesterase inhibition. Assuming a
    weight of 150 g per potato, and body weights of 60 and 70 kg, the
    dose was estimated at 1.25 to 3.2 mg 'solanine'/kg bw (Wilson,

         Seventy-eight junior schoolboys in Great Britain became ill
    from solanine poisoning 7 to 9 h after eating two small boiled
    peeled potatoes each (weight of potatoes not given) as part of their
    lunch, and 17 were admitted to hospital. Symptoms included vomiting,
    diarrhoea, and general abdominal pain. Most of the boys developed a
    fever, suffered from headache, dizziness, mental confusion,
    hallucinations and their vision was affected. Three boys were
    comatose and stuporose on admission, with peripheral circulatory
    collapse. All were discharged 6-11 days following admission, and 4-5
    weeks later there were no sequelae. Tests for the presence of
    biocides, such as nicotine, organophosphorus or organochloride
    pesticides were negative. Six days after eating the meal, plasma
    pseudocholinesterase levels in 10 out of 17 schoolboys was subnormal
    (about 25% below the normal range for this age group). Red blood
    cell cholinesterase levels were normal. The source of toxic potatoes
    was traced to a bag of old potatoes that had been condemned for
    consumption because of their appearance, but that had inadvertently
    been cooked (peeling of the potatoes had been done by an automatic
    peeling machine). Insufficient potatoes were left over after the
    meal for direct chemical analysis. Solanine levels in the boiled
    peeled potatoes were therefore estimated from the  in vitro
    reduction in pseudocholinesterase activity in human plasma, using
    acetylcholine as a substrate, and were equivalent to 25-30 mg/100 g
    tuber of alpha-solanine. Assuming an intake of 200 g potatoes and a
    bw of 40 kg (age = 11-14 years), the reviewers estimate that the
    intake of 'solanine' by the schoolboys would therefore have been
    approximately 1.4-1.6 mg/kg bw. Because of the small margin of
    safety between normal potatoes and toxic potatoes, the authors
    speculated that in toxic potatoes other toxic steroids besides
    glycoalkaloids may be synthesized, such as sapogenins and saponins,
    which might enhance the toxicity of solanine alkaloids by promoting
    gastro-intestinal absorption or other means (McMillan & Thompson,

         In a recent (1983) poisoning associated with a school lunch
    programme, 61 of 109 school children and staff in Alberta, Canada,
    became ill, most within 5 minutes, after eating baked potato (weight

    of potato not given) containing 49.4 mg 'solanine' per 100 g
    (analytical method not indicated). Test results showed that there
    was no evidence that the illness occurred due to the presence of
    viruses, bacteria, moulds, pesticides or other chemicals in the food
    items or their containers. The potatoes had a slight tinge of green
    and had a bitter or unusual taste (noted by 44% of those affected),
    causing a burning sensation in the throat of 18% of those affected.
    The predominant symptoms in order of frequency were nausea (69%),
    abdominal cramps (43%), headache (33%), vomiting (11%), fever and
    diarrhoea (8%). The children recovered in about 3 h. The reviewers
    estimate that, assuming the children ingested 200 g, and had a bw of
    40 kg, the dose was about 2.5 mg 'solanine'/kg bw (Anon, 1984).

         Based on the available human data (Table 3), an intake of
    3-6 mg TGA/kg bw is considered a potentially lethal dose for humans,
    and >1 to 3 mg TGA/kg bw is considered a toxic dose for humans.
    Children may be more sensitive than adults. Other factors may be
    present in suspect potatoes and modulate the toxicity of the
    steroidal glycoalkaloids.

         No signs of acute toxicity were noted in 3 Swedish adult
    volunteers who ingested for 1 week a diet estimated to give an
    intake of 1 mg/kg bw TGA (Harvey  et al., 1985b).

        Table 3. Summary of published reports of solanine poisoning in humans

    Affected             Potato type      Quantity    Concentration      Estimated        Outcome          Reference
                                          Consumed    of TGA             Toxic Dose
                                                      mg/kg bw           mg/kg bw

    56                   peeled,          1-1.5 kg    24                 3.4-5.1          recovered        Pfuhl,
    (soldiers)           cooked                       (38)                                                 1899

    60 adults            potatoes         500 g ?     41                 3.4              recovered        Harris &
    1 child                               200 g ?                        4.5              1 fatal          Cockburn,
                                                                         (lethal)         (5 yr-old)       1918

    7 (family)           greened          ?           ?                  ?                2 fatal          Hansen, 1925

    50-60                shoots,          ?           27                 ?                1 fatal          Willimot,
    (Cyprus)             leaves                       49                                                   1933

    Prisoners            experimental     ?           ?                  2.8              recovered        Report cited
                                                                                                           by Ruhl,

    Child                potato           ?           ?                  ?                1 fatal          Report cited
                         berries                                                                           by Ruhl,

    4 (family            baked            1-3         50                 1.2-3.2          dose-related,    Wilson, 1959
    adults)              potatoes         potatoes                                        recovered
                         with skin        150-450 g


    Table 3 (continued)

    Affected             Potato type      Quantity    Concentration      Estimated        Outcome          Reference
                                          Consumed    of TGA             Toxic Dose
                                                      mg/kg bw           mg/kg bw

    78                   old potatoes     2 small     25-30              1-4-1.6          3 comatose,      McMillan &
    (schoolboys)                          potatoes                                        all              Thompson,
                                          200 g                                           recovered;       1979
                                                                                          more affected

    61                   baked potato     200 g       49                 2.5              recovered        Anon. 1984

         ? Data not available.

    2.3.2  Teratogenic effects

         In 1972, Renwick showed that areas with an increased incidence
    of neural tube defects (NTD) (anencephaly and spina bifida) were
    associated with areas where potato consumption was higher, and where
    potato blight was more common. The worldwide incidence of NTD varies
    from < 1 to about 7 per 1000 total births. He postulated that this
    disease was due to toxic factors in potatoes, such as
    alpha-chaconine and alpha-solanine. These antifungal compounds offer
    resistance to potato blight and increase in amount in blighted
    potatoes, infected with the fungus  Phytophthora infestans. Several
    studies have been conducted to prove or disprove this theory
    (Renwick, 1972). The same author more recently suggested that a long
    half-life of potato glycoalkaloids could lead to their retention in
    the body, and possible release early during pregnancy (Renwick,

         In a prospective study with women who had previously borne a
    child with NTD, 27 women did not handle or eat potatoes or potato
    containing foods after deciding on a future pregnancy, and
    throughout gestation; another 61 women, attending the same clinic,
    did not avoid potatoes. The allocation to the two groups was
    non-random, but voluntary. The groups did not differ significantly
    with respect to age distribution, social class, parity or history of
    outcome of previous pregnancies. The incidence of NTD was 8.7% in
    the group of women avoiding potatoes, and 3.6% in the group eating
    potatoes (p=0.58). This study failed to support the Renwick
    hypothesis, but the authors pointed out that the size of the groups
    was small (Nevin & Merrett, 1975).

         Although there is a geographical similarity between neural tube
    defect occurrence and potato blight in Canada, no annual or seasonal
    associations were demonstrated. The author concluded that
    socioeconomic factors were probably more important as a risk factor
    for NTD, but suggested that better exposure assessment to factors
    present in potatoes, at the level of the individual, would be
    necessary to resolve this question. Such prospective studies should
    also assess the significance of other risk factors (Elwood, 1976).

         An epidemiological study (prospective study) was conducted in
    Great Britain, whereby human serum specimens from 380 patients, who
    were being screened for NTD by measuring their serum
    alpha-fetoprotein at 15-22 weeks of gestation (most at 16 weeks),
    were also analyzed for potato glycoalkaloids, using a sensitive
    radioimmunoassay. The samples were analyzed blind, regardless of the
    outcome of pregnancy, which resulted in 210 NTD cases and 170 normal
    offspring. In most of the 9 centres studied, serum TGA and serum
    solanidine levels were higher (p <0.05 in 2 centres) in the women
    with a normal fetus than in those with a fetus affected by NTD.
    Although closure of the neural tube normally takes place at 4-5

    weeks of gestation, the authors felt that measurements at the later
    date might reflect glycoalkaloid exposure earlier during gestation.
    The results of this study are therefore the opposite of what one
    would expect if the ingestions of potatoes contributed to the
    etiology of NTD. Instead the authors suggested that avoidance of
    potatoes might contribute to a vitamin deficiency thereby increasing
    rather than decreasing the incidence of NTD (Harvey  et al., 1986).


         Numerous studies performed on a variety of experimental animal
    species to elucidate the toxicological properties of glycoalkaloids,
    including teratogenicity, have been evaluated.

         Cranial abnormalities have been observed in some teratogenicity
    studies with laboratory animals, particularly with the hamster at
    levels of 165-200 mg glycoalkaloids/kg bw/day. However, the
    suggested association of the consumption of blighted potatoes during
    pregnancy with increased incidences of spina bifida and anencephaly
    has not been substantiated.

         In a limited study in humans, the daily consumption of potato
    tubers containing approximately 24 mg glycolkaloids/100 g did not
    result in any signs of acute toxicity. However, human poisonings
    have been associated with the consumption of poor-quality potato
    tubers with elevated levels of glycoalkaloids. The signs of
    low-grade glycoalkaloid poisoning are acute gastrointestinal upset
    with diarrhoea, vomiting, and severe abdominal pain. In more severe
    cases, neurological symptoms, including drowsiness, apathy,
    confusion, weakness, and vision disturbances followed by
    unconsciousness, have also been reported.


         The Committee considered that, despite the long history of
    human consumption of plants containing glycoalkaloids, the available
    epidemiological and experimental data from human and laboratory
    animal studies did not permit the determination of a safe level of
    intake. The Committee recognized that the development of empirical
    data to support such a level would require considerable effort.
    Nevertheless, it felt that the large body of experience with the
    consumption of potatoes, frequently on a daily basis, indicated that
    normal glycoalkaloid levels (20-100 mg/kg) found in properly grown
    and handled tubers were not of concern. To support the continued
    safe use of potato tubers, those developing new cultivars, and
    others growing, harvesting, storing, processing, and consuming
    potatoes, should be aware of the possibility of inadvertently
    increasing the content of glfycoalkaloids to potentially toxic


    ALOZIE, S.O., SHARMA, R.P. & SALUNKHE, D.K. (1978). Inhibition of
    rat cholinesterase isoenzymes  in vitro and  in vivo by the potato
    alkaloid, alpha-chaconine.  J. Food Biochem., 2: 259-276.

    ALOZIE, S.O., SHARMA, R.P. & SALUNKHE, D.K. (1979a). Physiological
    disposition, subcellular distribution and tissue binding of
    alpha-chaconine (3H).  J. Food Safety, 1: 257-273.

    ALOZIE, S.O., SHARMA, R.P. & SALUNKHE, D.K. (1979b). Excretion of
    alpha-chaconine-3H, a steroidal glycoalkaloid from
     Solanum-tuberosum L. and its metabolites in hamsters.  Pharmacol.
     Res. Commun., 11: 483-490.

    ANON. (1979). Solanine poisoning [editorial].  Br. Med. J., 2:

    ANON. (1984). Solanine food poisoning associated with a school lunch
    program - Alberta.  Canada Diseases Weekly Report, Health and
     Welfare Canada, 10-18: 71.

    AZIM, A., SHAIKH, H.A. & AHMAD, R. (1983). Toxic effects of high
    glycoalkaloid feeding on the protein digestibility and growth of
    rabbits.  J. Pharm. Univ. Karachi., 2: 15-24.

    AZIM, A., SHAIKH, H.A. & AHMAD, R. (1984). Toxic effects of high
    glycoalkaloid feeding on the red blood cell counts and haemoglobin
    concentration of rabbit blood.  J. Pharm. Univ. Karachi, 3: 43-49.

    BMER, A. & MATTIS, H. (1924). [Solanine content of potatoes] Der
    Solaningehalt der Kartoffeln.  Z. Nahr. Genussm., 47: 97-127.

    BUSHWAY, R.J. & PONNAMPALAM, R. (1981). alpha-chaconine and
    alpha-solanine content of potato products and their stability during
    several modes of cooking.  J. Agric. Food Chem., 29: 814-817.

    CHAUBE, S. & SWINYARD, C.A. (1976). Teratological and toxicological
    studies of alkaloidal and phenolic compounds from  Solanum tuberosum
     L. Toxicol. Appl. Pharmacol., 36: 227-237.

    CLARINGBOLD, W.D.B., FEW, J.D. & RENWICK, J.H. (1982). Kinetics and
    retention of solanidine in man.  Xenobiotica, 12: 293-302.

    DALVI, R.R. & BOWIE, W.C. (1983). Toxicology of solanine: an
    overview.  Vet. Hum. Toxicol., 25: 13-15.

    DALVI, R.R. (1985). Comparative assessment of the effect of solanine
    administered orally and intraperitoneally on hepatic dysfunction in
    male rats.  Jpn. J. Vet. Sci., 47: 657-659.

    ELWOOD, J.M. (1976). Anencephalus, spina bifida and potato blight in
    Canada.  Can. J. Public Health, 67: 122-126.

    GULL, S.D., ISENBERG, F.M. & BRYAN, H.H. (1970). Alkaloid toxicology
    of  Solanum-tuberosum. Hort. Science, 5: 316.

    HANSEN, A.A. (1925). Two fatal cases of potato poisoning.  Science,
    61: 340-341.

    HARRIS, F.W. & COCKBURN, T. (1918). Alleged poisoning by potatoes.
     Am. J. Pharm., 90: 722-726.

    HARRIS, H. & WHITTAKER, M. (1962). Differential inhibition of the
    serum cholinesterase phenotypes by solanine and solanidine.  Ann.
     Hum. Genet., 26: 71-76.

    HARVEY, M.H., McMILLAN, M., MORGAN, M.R.A. & CHAN, H.W.-S. (1985a).
    Solanidine is present in sera of healthy individuals and in amounts
    dependent on their dietary potato consumption.  Hum. Toxicol., 4:

    HARVEY, M.H., MORRIS, B.A., McMILLAN, M. & MARKS, V. (1985b).
    Measurement of potato steroidal alkaloids in human serum and saliva
    by radioimmunoassay.  Hum. Toxicol., 4: 503-512.

    HARVEY, M.H., MORRIS, B.A., McMILLAN, M. & MARKS, V. (1986). Potato
    steroidal alkaloids and neural tube defects: serum concentrations
    fail to demonstrate a causal relation.  Hum. Toxicol., 5: 249-253.

    JADHAV, S.J., SHARMA, R.P. & SALUNKHE, D.K. (1981). Naturally
    occurring toxic alkaloids in foods.  Crit. Rev. Toxicol., 9:

    JELINEK, R., KYZLINK, V. & BLATTNY, C., Jr. (1976). An evaluation of
    the embryotoxic effects of blighted potatoes on chicken embryos.
     Teratology, 14: 335-342.

    KIRK, D. & MITTWOCH, U. (1975). Changes in the mitotic cycle induced
    by alpha-solanine.  Humangenetik, 26: 105-111.

    KLINE, B.E., VON ELBE, H., DAHLE, N.A. & KUPCHAN, S.M. (1961). Toxic
    effects of potato sprouts and of solanine fed to pregnant rats.
     Proc. Soc. Exp. Biol. Med., 107: 807-809.

    KUBINSKI, H., GUTZKE, G.E. & KUBINSKI, Z.O. (1981). DNA-cell-binding
    (DCB) assay for suspected carcinogens and mutagens.  Mutat. Res.,
    89: 95-136.

    MAGA, J.A. (1980). Potato glycoalkaloids.  Crit. Rev. Food Sci.
     Nutr., 12: 371-405.

    D.T. (1983). Determination of solanidine in human plasma by
    radioimmunoassay.  Food Chem. Toxicol., 21: 637-640.

    McMILLAN, M. & THOMPSON, J.C. (1979). An outbreak of suspected
    solanine poisoning in schoolboys: examination of criteria of
    solanine poisoning.  Q. J. Med., 48: 227-243.

    effect of alpha-solanine on the active calcium transport in rat
    intestine.  Gen. Pharmacol., 16: 69-70.

    MORGAN, M.R.A. & COXON, D.T. (1987). Tolerances: glycoalkaloids in
    potatoes. Ch. 7. In: Watson, D.H. (ed.).  Ellis Horwood series in
     food science and technology: Natural toxicants in food: progress
     and prospects, Ellis Horwood, Chichester, England, pp. 221-230.

    MORRIS, S.C. & LEE, T.H. (1984). The toxicity and teratogenicity of
     Solanaceae glycoalkaloids particularly those of the potato
     (Solanum tuberosum): a review.  Food Technol. Aust., 36: 118-124.

    MUN, A.M., BARDEN, E.S., WILSON, J.M. & HOGAN, J.M. (1975).
    Teratogenic effects in early chick embryos of solanine and
    glycoalkaloids from potatoes infected with late-blight,
     Phytophthora infestans. Teratology, 11: 73-77.

    NESS, E., JONER, P.E. & DAHLE, H.K. (1984). Alpha-solanine tested
    for mutagenicity with the Ames test.  Acta Vet. Scand., 25:

    NEVIN, N.C. & MERRETT, J.D. (1975) Potato avoidance during pregnancy
    in women with a previous infant with either anencephaly and/or spina
    bifida.  Br. J. Prev. Soc. Med., 29: 111-115.

    NISHIE, K., GUMBMANN, M.R. & KEYL, A.C. (1971). Pharmacology of
    solanine.  Toxicol. Appl. Pharmacol., 19: 81-92.

    NISHIE, K., NORRED, W.P. & SWAIN, A.P. (1975). Pharmacology and
    toxicology of chaconine and tomatine.  Res. Commun. Chem. Pathol.
     Pharmacol., 12: 657-668.

    NORRED, W.P., NISHIE, K. & OSMAN, S.F. (1976). Excretion,
    distribution and metabolic fate of 3H-alpha-chaconine.  Res.
     Commun. Chem. Pathol. Pharmacol., 13: 161-171.

    PATIL, B.C., SHARMA, R.P., SALUNKHE, D.K. & SALUNKHE, K. (1972).
    Evaluation of solanine toxicity.  Food Cosmet. Toxicol., 10:

    PFUHL, E. (1899). [Regarding an outbreak of illness due to poisoning
    by solanine in potatoes] ber eine Massenerkrankung durch Vergiftung
    mit stark solaninhaltigen Kartoffeln.  Deutsch. Med. Wochenschr.,
    25: 753-754.

    POSWILLO, D.E., SOPHER, D. & MITCHELL, S.J. (1972) Experimental
    induction of fetal malformation with "blighted" potato: a
    preliminary report.  Nature, 239: 462-464.

    R.F. & PRICE, K.R. (1973). Investigations into the teratogenic
    potential of imperfect potatoes.  Teratology, 8: 339-347.

    RENWICK, J.H. (1972). Hypothesis: anencephaly and spina bifida are
    usually preventable by avoidance of a specific but unidentified
    substance present in certain potato tubers.  Br. J. Prev. Soc. Med.,
    26: 67-88.

    RENWICK, J.H. (1982). Food and malformation.  Practitioner, 226:

    McLEAN, A.C.S. (1984). Neural-tube defects produced in Syrian
    hamsters by potato glycoalkaloids.  Teratology, 30: 371-381.

    RODDICK, J.G. (1989). The acetylcholinesterase-inhibitory activity
    of steroidal glycoalkaloids and their aglycones.  Phytochemistry,
    28: 2631-2634.

    RUDDICK, J.A., HARWIG, J. & SCOTT, P.M. (1974). Nonteratogenicity in
    rats of blighted potatoes and compounds contained in them.
     Teratology, 9: 165-168.

    RHL, R. (1951). [Contribution on the pathology and toxicology of
    solanine] Beitrag zur Pathologie und Toxikologie des Solanins.
     Arch. Pharm., 284: 67-74.

    SHARMA, R.P., WILLHITE, C.C., WU, M.T. & SALUNKHE, D.K. (1978).
    Teratogenic potential of blighted potato concentrate in rabbits,
    hamsters, and miniature swine.  Teratology, 18: 55-61.

    SHARMA, R.P., WILLHITE, C.C., SHUPE, J.L. & SALUNKHE, D.K. (1979).
    Acute toxicity and histopathological effects of certain
    glycoalkaloids and extracts of  Alternaria solani or  Phytophthora
     infestans in mice.  Toxicol. Lett., 3: 349-355.

    SHARMA, R.P., TAYLOR, M.J. & BOURCIER, D.R. (1983). Subcellular
    distribution of alpha-chaconine in mouse hepatocytes.  Drug Chem.
     Toxicol., 6: 219-234.

    SHARMA, R.P. & SALUNKHE, D.K. (1989). Solanum glycoalkaloids. In:
    Cheeke, P. R. (ed.).  Toxicants of Plant Origin, Vol. 1 Alkaloids,
    CRC Press, Boca Raton, Florida. pp. 179-236.

    SLANINA, P. (1990a). Assessment of health-risks related to
    glycoalkaloids ("solanine") in potatoes: a Nordic view. Report from
    the Nordic working group on food toxicology and risk assessment.
     Vr Fda, 43: 1-14.

    SLANINA, P. (1990b). Solanine (glycoalkaloids) in potatoes:
    toxicological evaluation.  Food Chem. Toxicol., 28: 759-761.

    SWINYARD, C.A. & CHAUBE, S. (1973). Are potatoes teratogenic for
    experimental animals?  Teratology, 8: 349-357.

    WILLIMOTT, S.G. (1933). An investigation of solanine poisoning.
     Analyst, 58: 431.

    WILSON, G.S. (1959). A small outbreak of solanine poisoning.
     Monthly Bulletin, Ministry of Health (London), 18: 207-210.

    WILSON, A.M., McGANN, D.F. & BUSHWAY, R.J. (1983). Effect of growth,
    location and length of storage on glycoalkaloid content of roadside
    stand potatoes as stored by consumers.  J. Food Prot., 46: 119-121.

    WOOD, F.A. & YOUNG, D.A. (1974). TGA in potatoes.  Agric. Can. Publ.
    1533: pp. 1-2.

    See Also:
       Toxicological Abbreviations