DIQUAT                                JMPR 1972


    Diquat was evaluated at the Joint Meeting in 1970. Since the previous
    evaluation (FAO/WHO, 1971), some new experimental studies have been
    presented to fulfill the request made at that meeting for a further
    three-generation reproduction study, investigations on the mechanism
    of cataractogenesis in animals and clinical studies on factory workers
    exposed to diquat to assess cataractogenic risk.




    Levels of diquat in lung, muscle, kidney and liver have been measured
    at 1, 3, 5, 7 and 10 days after administration of 20 mg/kg
    intravenously to rats. Levels in lung and muscle were constantly
    lower, and in liver constantly higher, than those observed with
    paraquat. Up to five days, kidney levels of diquat and paraquat were
    comparable, after which diquat levels exceeded those of paraquat
    (Sharp et al., 1972).


    Special studies on cataractogenicity

    Cataractogenicity has been studied by Pirie and Rees (1970) and Pirie
    et al. (1970). These workers consider that such prolonged
    administration of diquat is required that cataract is unlikely to
    occur in man. Diquat cataracts differ from others in the rat. They
    develop through a posterior opacity, which may be rich in
    ribonucleoprotein, to a dense nuclear cataract, with cortical opacity,
    developing only at a late stage as the lens shrinks. In this respect,
    it resembles cataract due to irradiation, but is unlike it as no
    damage to cell nuclei of lens epithelium occurs. An important
    biochemical change is a fall in concentration of ascorbic acid in the
    lens and intraocular fluids. A major difference between cataracts
    following diquat administration and others (e.g. those developing from
    X-irradiation, galactose, diabetes, naphthalene feeding or senility)
    is that lens GSH remains high in diquat treated rats, even in the
    presence of severe cataracts. The concentration of diquat in the lens
    is below that of serum when diquat is administered intraperitoneally.
    Fresh extracts of bovine lens form free radicals of diquat in the dark
    by reduction with glutathione reductase and NADPH. This reaction is
    reversible. Sunlight catalyses the formation of free diquat radical in
    the presence of amino acids, aqueous humour or lens diffusate and also
    the oxidation of ascorbic acid in aqueous humour by diquat. At present
    it is possible only to speculate on the role of these biochemical
    reactions on the mechanism of cataract formation by diquat.

    Special studies on reproduction

    In a three-generation reproduction study, 3 groups of 12 male and 24
    female rats were fed on diet containing 0, 125 and 250 ppm diquat ion
    from the age of 35 days; thereafter, they and their progeny remained
    on these diets. After 100 days, two females were mated with one male
    to produce the F1a litter. The male was replaced if conception did
    not occur within three weeks. Second matings took place ten days after
    weaning. The F1b litter was used to provide the second generation
    parents. Subsequent generations were similarly produced. Animals
    receiving 500 ppm failed to maintain a normal rate of growth and their
    food intake was reduced. At the 125 ppm level, rats of the F2
    generation had body-weights significantly below that of controls at
    weaning and just prior to mating but not in the intervening period.
    Neither treatment level affected reproduction as shown by fertility,
    period between mating and litter production, mean litter size, number
    of stillborn and sex distribution. The mean body weight of the young
    at weaning was reduced at the 500 ppm level in F1a, F1b, F2a and F2b
    litters and at the 125 ppm level in F1b and F2a litters. Examination
    of all animals for changes in behaviour, congenital abnormalities and
    gross and microscopic pathology showed no differences between test and
    control groups, except for the occurrence of lens opacities. These
    were seen in animals receiving 500 ppm diquat and first appeared at
    91, 106 and 124 days in the parent, F1b and F2b generation adults.
    After about 280 days these generations showed an incidence of the
    lesion of 55%, 70% and 47%. The time of onset and incidence were thus
    no different in groups which had been exposed during intrauterine and
    neonatal periods. No opacities were found in the young or in adults of
    the 125 ppm groups (Fletcher et al., 1972).


    Observations have been made on 22 workers engaged in manufacturing
    and/or packing diquat dibromide for periods varying from 6 months to
    11´ years (average 3´ years). Nine are still employed by the company.
    Review of the medical records of the 22 workers over this period
    showed that there were 20 medical incidents which could be attributed
    to contact with diquat and that exposure was principally by splashing
    while drums were being filled. All were of a minor nature and due to
    contact effects on the eyes, nails and skin. No case of lens opacity
    has been observed in these or other workers exposed to diquat (Shaw,


    A three-generation reproduction study has confirmed that 500 ppm does
    not affect reproduction or produce teratogenic effects. This study has
    confirmed the cataractogenicity of this compound and has demonstrated
    that exposure of animals in utero and neonatally to diquat does not
    decrease the exposure time needed to develop cataracts. Human data are
    limited to cases of intermittent skin and inhalation exposure. Human
    subjects who have been exposed over several years to diquat and in
    some cases suffered injuries from direct eye, skin or nail contact,
    have not shown evidence of cataract formation. This parallels previous
    findings in animals. It seems unlikely that cataract formation will
    occur in humans exposed to low levels of diquat such as occur as
    pesticide residues in food. The acceptable daily intake was assessed
    from two-year studies in rats and dogs.


    Level causing no toxicological effect

         Rat: 10 ppm in the diet, equivalent to 0.5 mg/kg body-weight/day
              (corresponds to 0.36 mg diquat ion/kg body-weight/day)

         Dog: 68 ppm in diet, equivalent to 1.7 mg/kg body-weight/day
              (corresponds to 1.22 mg diquat ion/kg body-weight/day)


         0 - 0.005 mg/kg body-weight (equivalent to
         0 - 0.0036 mg diquat ion/kg body-weight.


    The uses and chemical characteristics of diquat together with its mode
    of action and toxicity have been reviewed by Wheeler (1971) and
    Baldwin (1969).


    Table 1 depicts residue data on various crops following desiccant use
    of diquat (Plant Protection Ltd., 1972).

    Residues of diquat and its degradation products in rapeseed and oil
    have been determined by Leahey and Allard (1971). Rape plants were
    desiccated with 14C-diquat at various rates of application and the
    oil and meal analysed. No detectable residues of diquat or its
    breakdown products were found in rapeseed oil when the seeds were
    harvested 7 days after desiccation, and residues of less than 0.02 ppm
    were found when seeds were harvested 14 days after desiccation. Values
    obtained from the analysis of the meal following varying application
    rates are given in Table 2. It was confirmed that a large part of the
    radioactive extract consisted of unchanged diquat.

        TABLE 1  Residues of diquat in various crops


                           Rate of           Interval                     Diquat found
    Crop                   application       harvest      Samples     range           mean
                           (g diquat/ha)     (days)       (no.)              (ppm)

    Wheat (grain)          600 - 1000        4 - 7        43          ND1 - 1.58    0.61
          (flour)          600 - 1000        4 - 7        8           ND  - 0.67      0.22

    Rice (with husk)       200 - 400         3 - 5        35          ND - 6.40       0.89
         (polished)        200 - 400         3 - 5        26          ND - 0.16       0.07

    Sorghum (grain)        400 - 600         4 - 10       25          ND - 5.90       0.81

    Rape seed              400 - 600         4 - 10       77          ND - 1.50       0.37

    Sunflower seed         600               4 -15        16          ND - 0.20       0.07

    Cotton seed            400 - 1000        10           14          ND - 0.98       0.37

    Potatoes               600 - 1000        4 - 10       36          ND - 0.25       0.03

    Peas                   300 - 1000        4 - 10       20          ND - 0.07       0.05

    Beans                  300 - 1000        4 - 10       15          ND - 0.57       0.10

    Poppy                  800               6 - 10       9           0.56 - 4.90     2.84

    sugar beet (root)      300 - 800         2 - 7        2           0.09 - 0.11     0.10

              (juice)      300 - 800         2 - 7        4           -               <0.01

    1  ND = not detected (usually <0.05 ppm)
    TABLE 2  Residues of diquat in rape seed meal


                                           Diquat ion      Radioactivity
             Rate of        Time between   equivalent      extracted
    Plant    application    spraying and   to the          with
    no.      (kg a.i./ha)   harvest        radioactivity   1.0 M HBr
                            (days)         detected        (%)

    1        0.3            7              0.37            83

    2        6.3            7              0.18            80

    3        0.6            7              0.96            85

    4        0.6            7              0.94            85

    5        1.1            7              0.48            67

    6        1.1            7              0.66            75

    7        0.3            14             0.52            86

    8        0.3            14             10.02           76

    9        0.6            14             1.49            75

    10       0.6            14             1.51            71

    11       1.1            14             3.22            85

    12       1.1            14             1.87            89


    General comments

    Diquat is a contact herbicide that kills or severely scorches all
    green herbage with which it comes into contact. It is not readily
    translocated. Furthermore, it is quickly rendered biologically
    inactive by sorption into clay minerals in the soil and is thus not
    mobile in soil or available for root uptake. These properties make it
    highly successful as a pre-planting or pre-crop-emergence total
    weedkiller or as a directed, inter-row spray between rows of emerged
    crops. Contamination may occasionally arise when spray is misdirected
    or drifts onto growing crops or when young seedlings emerge through

    dense swards of sprayed herbage containing diquat residues. In such
    cases severe contamination will kill or severely scorch the plants,
    and small residues (below 0.5 ppm) have been detected in the foliage
    of some crops (e.g., oats and maize) seven to eight weeks after
    application. However, the great majority of crops treated in this way
    show no detectable residues (<0.05 ppm) in edible parts when
    harvested from one to four months later (Plant Protection Ltd., 1972).

    In animals

    No residues of diquat were found in milk, urine or faeces of cows fed
    daily with 5 kg of ground sunflower seed, containing 1 mg of residual
    diquat, for periods of 185 and 257 days (Lembinski et al., 1971). No
    residues of diquat were found in the liver and kidneys of a calf born
    to a cow fed for 257 days with the ground seed or in the liver and
    kidneys of wethers fed daily for 141 days with 0.5 kg of ground seed
    containing 0.1 mg of diquat. The limits of detection were 0.01 ppm in
    milk and urine and 0.03 ppm in liver, kidneys and faeces.

    In plants

    Brian (1970) showed that the behaviour of diquat in plants was
    complex. Studies using tomato plants demonstrated that the activity
    was influenced by light before and after treatment, and that the
    apparent loss at intervals was not due to exudation from leaves or
    downward movement into the roots. When diquat was used as a silaging
    agent it was found that the protein content in the hay was increased
    and the glycidic content reduced (Jambrich, 1970). The moisture
    content of maize was not reduced significantly when diquat was used as
    a desiccant, but the application appeared to reduce the crop yield
    (Wilkins and Tetlow, 1970). The changes in plants occurring after
    application of diquat have been discussed by Dodge (1971).

    In soil

    The cationic exchange properties of diquat in soil clays, vermiculite
    and smectite has been investigated. Exchangeable diquat ion was
    replaced by potassium ion, and values are given for various samples. A
    direct relationship between the exchange and layer charge density was
    obtained (Dixon et al., 1970).

    In water

    Diquat was used at a rate of 4 lb/acre (surface) on two New York
    lakes; residues were less than 0.005 ppm in 4-8 days and a similar
    shoreline treatment showed no residue after 1 day (Sewell, 1970).


    Herbicides of the diquat type have been determined by a colorimetric
    method (Zhemchuzhin and Akimova, 1970) based on chloramine-T and
    chlorophenol red. A semi-quantitative thin-layer chromatographic
    method for the determination of diquat has been used for residues in
    bees (Mueller and Worseck, 1970). The polarographic response of diquat
    in five supporting electrolytes has been reported (Hance, 1970). The
    method of Calderbank and Yuen (1966) was used for the determination of
    diquat in sunflower seeds by Lembinski et al. (1971); the limit of
    detection was 0.03 ppm. The method was also used, with the
    modifications of Black et al. (1966), for residues in milk, urine,
    faeces, kidneys and liver after feeding animals with the sunflower
    seed. These procedures should be suitable for regulatory purposes.


    Since the evaluation of diquat in 1970 (FAO/WHO, 1971) further
    residues data have become available. These indicated a need to
    increase the recommended tolerance levels for residues of diquat in
    polished rice and potatoes; some new tolerances are also proposed.
    Tolerances on barley and wheat cover occasional required desiccant
    uses; the bulk of any cereals so treated should be used for animal
    feed or seed purposes only.



    The following tolerances are recommended to replace those listed in
    Annex 1 of FAO/WHO (1972).
                                                         diquat ion

         Barley, poppy seed, rice (in husk)              5
         Rapeseed, sorghum, wheat                        2
         Cottonseed                                      1
         Beans, sunflower seed                           0.5
         Rice (polished), potatoes, wheat flour          0.2
         Onions, maize, sugar beet, peas                 0.1
         Sesame, sunflower, rape, cotton seed oils       0.1
         Other vegetable crops                           0.05*

         Milk                     ) from the feeding     0.01*
         Meat and meat products   ) of treated forage    0.05*
         * at or about the limit of determination



    1.   Further study on the mechanism of cataractogenic activity.

    2.   Identification of the toxicologically active substance (i.e.,
         parent compound or metabolite).

    3.   If treated cereals are to be used for human consumption, further
         data would be required on residues occurring in barley, wheat,
         rye and oats and their products (flour, bread, beer, etc.)


    To FAO/WHO (1971), p. 549 and FAO/WHO (1972), p. 31: diquat
    (cation), column 4, after Peas, beans, sunflower seed for 0.1
    read 0.5.


    Baldwin, B.C. (1969) In Progress in photosynthesis research, (H.
    Metzner, Ed.), Tübingen III, 1737.

    Black, W.J.M., Calderbank, A., Douglas, G. and McKenna, R.H. (1966)
    Residues in herbage and silage and feeding experiments following the
    use of diquat as a desiccant. J. Sci. Fd. Agric., 17: 506-509.

    Brian, R.C. (1970) Effect of light environment on the activity and
    behaviour of diquat and paraquat in plants. Pestic. Sci., 1: 38-41.

    Calderbank, A. and Yuen, S.H. (1966) An improved method for
    determining residues of diquat. Analyst, 91: 625-629.

    Dixon, J.B., Moore, D.E., Agnihotri, N.P. and Lewis, D.E. (1970)
    Exchange of diquat in soil clays, vermiculite and smectite. Soil Sci.
    Soc. Amer. Proc., 34: 805-808.

    Dodge, A.D. (1971) Mode of action of the bipyridylium herbicides,
    paraquat and diquat. Endeavour, 30: 130-135.

    FAO/WHO. (1971) 1970 Evaluations of some pesticide residues in food.
    FAO/AGP/1970/M/12/1; WHO/Food Add./71.42.

    FAO/WHO. (1972) Pesticide residues in food. FAO Agricultural Studies
    No. 88; WHO Technical Report Series No. 502.

    Fletcher, K., Griffiths and Kinch, D.A. (1972) Diquat dibromide:
    three-generation reproduction study in rats. Report Imperial Chemical
    Industries Industrial Hygiene Research Laboratories. (unpublished).

    Hance, R.J. (1970) Polarography of herbicides - a preliminary survey.
    Pestic. Sci., 1: 112-113.

    Jambrich, J. (1970) Desiccants as silaging agents. Agrochemia, 10:

    Leahey, J.P. and Allard, J. (1971) Bipyridylium herbicides: residues
    in rape seed and oil following desiccation with diquat. Report TMJ
    674A Plant Protection Ltd. (unpublished)

    Lembinski, F., Ponikiewska, T., Trzebry, W. and Krzywinska, F. (1971)
    Ground seed of sunflower desiccated with "Reglone" as fodder for
    ruminants. Pamietnik Pulawski Prace IUNG, 49.

    Mueller, B. and Worzeck, M. (1970) Method for the semi-quantitative
    determination by thin-layer chromatography of diquat (Reglone) and
    paraquat (Gramoxone). Monatsh. Veterinaermed, 25: 560-561.

    Pirie, A. and Rees, J.R. (1970) Diquat cataract in the eye. Exptl. Eye
    Res., 9: 198-203.

    Pirie, A., Rees, J.R. and Holmberg, J.N. (1970) Diquat cataract:
    formation of the free radical and its reaction with constituents of
    the eye. Exptl. Eye Res., 9: 204-218.

    Plant Protection Ltd. (1972) Additional data to support the
    establishment of permanent tolerances for diquat in human food crops.

    Sewell, W.D. (1970) Diquat residues in two New York lakes. Proc.
    Northeast Weed Contr. Conf., 24: 281-282.

    Sharp, C.W., Allotengke, A. and Posner, H.S. (1972) Correlation of
    paraquat toxicity with tissue concentrations and weight loss of the
    rat. Toxic. appl. Pharmacol., 22: 241-251.

    Shaw, G.H. (1971) Report by Works Medical Officer, I.C.I.

    Wheeler, J.H. (1971) Diquat and paraquat. Proc. Ann. Calif. Weed
    Conf., 23: 154-157.

    Wilkins, R.J. and Tetlow, R.M. (1970) Effect of diquat and paraquat
    used as desiccants on the moisture content of maize for silage. Weed
    Res., 10: 288-292.

    Zhemchuzhin, S.G. and Akimova, N.N. (1970) Determination of herbicides
    of diquat type. Otkrytiya, Izobret, Prom. Obraztsy, Tovarnye Znaki,
    47(31): 133-135.

    See Also:
       Toxicological Abbreviations
       Diquat (HSG 52, 1991)
       Diquat (PIM 580F, French)
       Diquat (AGP:1970/M/12/1)
       Diquat (Pesticide residues in food: 1976 evaluations)
       Diquat (Pesticide residues in food: 1977 evaluations)
       Diquat (Pesticide residues in food: 1978 evaluations)
       Diquat (Pesticide residues in food: 1993 evaluations Part II Toxicology)