and 2,6-dichlorobenzonitrile in the dog and rat

Biochem. J. (1966) 98, 770

The Comparative Metabolism of 2,6-Dichlorothiobenzamide (Prefix)and 2,6-Dichlorobenzonitrile in the Dog and Rat

By M. H. GRIFFITHS, J. A. MOSS, J. A. ROSE AND D. E. HATHWAYTunstall Laboratory, Shell Research Ltd., Sittingbourne, Kent

(Received 22 July 1965)

1. A single oral dose of either [14C]Prefix or 2,6-dichlorobenzo[14C]nitrile to ratsis almost entirely eliminated in 4 days: 84.8-100.5% of 14C from [14C]Prefix isexcreted, 67-3-79.7% in the urine, and 85.8-97.2% of 14C from 2,6-dichlorobenzo-[14C]nitrile is excreted, 72.3-80.7% in the urine. Only 0-37 ± 0.03% of the dose of[14C]Prefix and 0-25 +0.03% of the dose of 2,6-dichlorobenzo[14C]nitrile are

present in the carcass plus viscera after removal of the gut. Rats do not show sex

differences in the pattern of elimination of the respective metabolites of the twoherbicides. The rates ofelimination of 14C from the two compounds in the 24hr. and48hr. urines are not significantly different (P > 0.05) from one another. 2. Afteroral administration to dogs, 85.9-106.1% of 14C from [14C]Prefix is excreted,66.6-80-9% in the urine, and 86.8-92.5% of 14C from 2,6-dichlorobenzo[14C]nitrileis excreted, 60-0-70-1% in the urine. Dogs do not show sex differences in thepattern of eliminating the metabolites of either Prefix or 2,6-dichlorobenzonitrile.3. Dogs and rats do not show species differences in the patterns of elimination ofthetwo herbicides. 4. Prefix and 2,6-dichlorobenzonitrile are completely metabolized;unchanged Prefix and 2,6-dichlorobenzonitrile are absent from the urine andfaeces, and from the carcasses when elimination is complete. In the hydrolysedurine of rats dosed with either [14C]Prefix or 2,6-dichlorobenzo[14C]nitrile,2,6-dichloro-3-hydroxybenzonitrile accounts for approx. 42% of the 14C, a further10-11% is accounted for by 2,6-dichlorobenzamide, 2,6-dichlorobenzoic acid,2,6-dichloro-3- and -4-hydroxybenzoic acid and 2,6-dichloro-4-hydroxybenzonitrilecollectively, and 25-30% by six polar constituents, of which two are sulphur-containing amino acids. 5. In the unhydrolysed urines of rats dosed with either[14C]Prefix or 2,6-dichlorobenzo[14C]nitrile, there are present free 2,6-dichloro-3-and -4-hydroxybenzonitrile, their glucuronide conjugates, ester glucuronides ofthe principal aromatic acids that are present in the hydrolysed urines, and twosulphur-containing metabolites analogous to mercapturic acids or premercapturicacids. 6. Prefix is thus extensively transformed into 2,6-dichlorobenzonitrile:R.CS.NH2 -+R.CN+H2S, where R = C6H3C12. However, the competitivereaction: R0CS.NH2+H20 -R.CO.NH2+H2S takes place to a very limitedextent.

2,6-Dichlorothiobenzamide (Prefix) (Yates, 1961)and 2,6-dichlorobenzonitrile (Belgian Patent, 1957)are promising herbicides with interesting biologicalproperties (Barnsley, 1960; Koopman & Daams,1960). Prefix is useful as a selective herbicide intransplanted rice. Agricultural application makesan understanding of the mammalian metabolismof the two herbicides desirable.

MATERIALS

2,6-Dichlorobenzonitrile. This was prepared by dehydra-tion in boiling acetic anhydride of anti-2,6-dichloro-benzaldoxime (Koopman, 1961), which was itself prepared

by reaction of 2,6-dichlorobenzaldehyde (Fierz-David &Blangey, 1952; Reich, 1917) with hydroxylamine mono-sulphonate (Koopman, 1961). 2,6-Dichlorobenzonitrilehad m.p. 1450 (Koopman, 1961, gives 144-1450).

2,6-Dichlorothiobenzamide (Prefix). This was prepared bythe addition of H2S to a solution of 2,6-dichlorobenzonitrilein dimethylformamide at 600, according to the method ofYates (1961). Prefix had m.p. 1520 (Yates, 1961, givesm.p. 1520).Purity of the herbicide,. The two herbicides, which had

been crystallized from benzene and ethanol to constantmelting point, migrated as single substances on thin-layerplates of Kieselgel G run with benzene; phenoxyethanol-AgNO3 spray (Mitchell, 1958) was used for locating thesubstance on the thin-layer plate. Infrared measurement

770

COMPARATIVE METABOLISM OF TWO HERBICIDES

was used as a further criterion of purity. The purity of eachcompound exceeded 99.9%.The unlabelled herbicides were used for dosing rats, in

which the cysteine-cystine pools had been labelled with35S (see the Methods section). Unlabelled herbicides werealso used as reference compounds to establish the presenceor absence of unchanged 14C-labelled herbicides in the urineand faeces of treated animals (see the Results section).

2,6-Dichlorobenzoic acid. This compound, m.p. 146-147°,was prepared from 2,6-dichlorobenzaldehyde (Fierz-David& Blangey, 1952; Reich, 1917) by oxidation with KMnO4 inacetone solution (Davies, 1921).

2,6-Dichlorobenzamide. This compound, m.p. 2020, wasprepared by treatment of 2,6-dichlorobenzonitrile withalkaline H202 (Reich, Salzmann & Kawa, 1917).

2,6-Dichloro-3-hydroxybenzonitrile. This was preparedfrom 2,6-dichloro-3-hydroxybenzaldehyde (Hodgson &Beard, 1926) as starting material. When the correspondingaldoxime (5.6g.) (Hodgson & Beard, 1926) was refluxed in20ml. of acetic anhydride for 2hr. and poured into water,2,6-dichloro-3-acetoxybenzonitrile was obtained as an oil,which crystallized (5.1 g., m.p. 1060). Hydrolysis, byrefluxing for 1 hr. with 3N-KOH under N2, and acidificationyielded 2,6-dichloro-3-hydroxybenzonitrile (4-1 g., m.p.196-198°) (Yates, 1962).

2,6-Dichloro-3-hydroxybenzoic acid. This was preparedby oxidation of 2,6 - dichloro - 3 - hydroxybenzaldehyde(Hodgson & Beard, 1926) withfreshlyprepared Ag20 accord-ing to the method of Pearl (1950). 2,6-Dichloro-3-hydroxy-benzoic acid crystallized from water, forming prisms of themonohydrate, m.p. 1260 (Mazzara & Bertozzi, 1900, givem.p. 123-124' for the monohydrate). Methyl 2,6-dichloro-3-methoxybenzoate. This compound, m.p. 570 (Mazzara &Bertozzi, 1900), was used as a reference compound in thegas-liquid chromatography (see the Methods section andTable 7). [The methyl ester and the two 0-methyl ethers,which were also used in gas-liquid chromatography (seeTable 7), were oils that failed to crystallize, and are nottherefore described in the Materials section.]

2,6-Dichloro-4-hydroxybenzonitrile. This was synthesizedfrom p-nitrosophenol as starting material. Reaction of anethanolic solution of p-nitrosophenol with HCI yielded2,6-dichloro-4-ethoxyaniline (Bargellini, 1929), which wasconverted into 2,6-dichloro-4-hydroxybenzonitrile bysuccessive diazotization, Sandmeyer reaction and de-ethylation. This reaction sequence gave a better yield offinal product from the same intermediate, 2,6-dichloro-4-ethoxyaniline, than the alternative sequence, de-ethylation,diazotization and Sandmeyer reaction, used by Strating &Backer (1943). 2,6-Dichloro-4-hydroxybenzonitrile crystal-lized from toluene, forming needles, m.p. 229-230° (Strating& Backer, 1943, give m.p. 228-230°).

2,6-Dichloro-3-hydroxythiobenzamide. This compound,m.p. 197-198°, was supplied by Dr J. Yates.

Bovine liver fl-glucuronidase. This enzyme (1g., 560000units) (Fishman, Springer & Brunetti, 1948) was obtainedfrom the Sigma Chemical Co., St Louis, Mo., U.S.A., andwas found to be free from hippuricase and sulphuric esterhydrolase activities.

Reagents and solvent8. All reagents and solvents were ofA.R. grade.

Radioactive chemical8. 2,6-Dichlorobenzo[14C]nitrile wassynthesized by reaction of 2,6-dichloroiodobenzene (Korner& Contardi, 1913) with Cu14CN and purified by column

chromatography on silicic acid. 2,6-Dichlorobenzo[14C]-nitrile (sp. activity 91 l,c/mg.) was eluted with methylenedichloride.

2,6-Dichlorothiobenz[14C]amide ([14C]Prefix) was pre-pared by addition of H2S to 2,6-dichlorobenzo[14C]nitrile(sp. activity 3.2,uc/mg.) and purified by column chromato-graphy on silicic acid. [14C]Prefix (sp. activity 3-2,uc/mg.)was eluted with methylene dichloride.The two 14C-labelled herbicides behaved identically with

authentic specimens of the unlabelled compounds on thin-layer plates of Kieselgel G run with benzene; radioauto-graphy did not reveal radioactive impurities. The radio-chemical purity of both compounds exceeded 99.9%.

L-[35S]Cystine (sp. activity 424lic/mg.) was obtainedfrom The Radiochemical Centre, Amersham, Bucks.

METHODS

Experiments with animals. Adult rats (approx. 2 monthsold, 150-200g. body wt.) were used (Porton strain main-tained as a specific pathogen-free colony in this Laboratory).For the metabolism experiments the animals were kept inall-glass metabolism cages under the conditions describedby Wright, Akintonwa, Crowne & Hathway (1965). In thefirst experiment a solution of [14C]Prefix (1-8mg., 3.2,uc/mg.) in 2ml. of aq. 5% (v/v) ethanol was administered tomale and female rats by stomach tube, and in the secondexperiment 0-80mg. (males) or 1-3mg. (females) of 2,6-dichlorobenzo[14C]nitrile (9.1 tc/mg.) in 2ml. of arachis oilwas administered to rats. Unrestricted food and water weresupplied, the urine and faeces were collected daily andstored at -29° and the expired gases were monitored for14CO2. In these experiments the rats were killed after 4 daysand the alimentary canal was removed. The carcass plusviscera, the alimentary canal and the skin plus hair werestored separately at -29°.

In another experiment 1X5 ml. ofan aq. 2% (w/v) NaHCO3solution containing 13,uc of L-[35S]cystinelml. was admini-stered intraperitoneally to four rats of each sex on 5consecutive days. On the fourth and fifth days two rats ofeach sex were each dosed by stomach tube with 1 ml. of asolution of Prefix (5mg.) in arachis oil and two rats of eachsex were each dosed by stomach tube with 1 ml. of a solutionof 2,6-dichlorobenzonitrile (5 mg.) in arachis oil. Un-restricted food and water were supplied and the urine wascollected on the fourth and fifth days and stored at -29°.This experiment is based on work of Gutmann & Wood(1950) and Marsden & Young (1958).Adult hounds (7-15kg. body wt.) were used (beagles

maintained for 2-3 years as a closed colony in this Labora-tory). For the metabolism experiments the animals werekept in stainless-steel metabolism cages under the conditionsdescribed by Wright et al. (1965). In one experiment eachdog swallowed a gelatin capsule containing a solution of7-90mg. of [14C]Prefix (2.8,uc/mg.) in ethanol (1ml.), andin another experiment each dog swallowed a gelatin capsulecontaining a solution of 0-92mg. of 2,6-dichlorobenzo[14C]-nitrile (8.52jc/mg.) in 1 ml. ofarachis oil. Unrestricted foodand water were supplied and the urine and faeces werecollected separately at daily intervals.

Fractionation of urines from treated animals. Solventextracts of the pooled urines from 12 rats were made bycontinuous extraction atpH 15-2-0 in a Schacherl apparatuswith ether and ethyl acetate successively. Evaporations

771Vol. 98

M. H. GRIFFITHS, J. A. MOSS, J. A. ROSE AND D. E. HATHWAYwere carried out in N2 under reduced pressure at below350

Hydrolysis of urinesfrom treated animals andfractionationof the hydrolysed urines. Preliminary investigations showedthat hydrolysis of the 14C-labelled metabolites in the urinesof treated animals gave the same chromatographic andelectrophoretic patterns of ether-soluble 14C-labelledconstituents irrespective of whether the hydrolysing agentwas fi-glucuronidase or 6N-HCI (3 hr. at 1100). The alter-native hydrolytic procedures did, however, yield somedifferent 14C-labelled constituents in the residual aqueousphases after ether extraction. Thus the 14C-labelledsulphur-containing metabolites (see below) were trans-formed by acid hydrolysis into 14C-Jabelled sulphur-containing amino acids, whereas they were unaffected byenzymic hydrolysis. For the purpose of identifying14C-labelled constituents in urine hydrolysates, the pooledurines of 12 treated animals were hydrolysed with acid. Thehydrolysed urines were exhaustively extracted by contin-uous liquid-liquid extraction (Schacherl apparatus) withether and ethyl acetate successively, and the sum of theseparate amounts of 14C in the solvent extracts and residualaqueous phase accounted quantitatively for the originallabel in the unhydrolysed urine. Further, when theindividual compounds in the solvent extracts and residualaqueous phase were separated on paper chromatograms andelectrophoretograms (for details see below), which in turnwere scanned and summed, an approximate balance ofradioactivity was obtained. This showed that acid hydro-lysis did not cause destruction of hydrolytic products withconsequent loss of radioactivity as 14C, 14C02 or 14C0labelled precipitates. This work was reproducible; otherbatches of pooled urines from 12 rats dosed with the 14C-labelled herbicides, when hydrolysed with acid in this way,also gave a consistent balance of 14C and identical propor-tions of the same 14C-labelled constituents.

Samples (5mg.) of each of the 14C-labelled constituentsthat were subsequently identified in the ether phase(2,6-dichlorobenzamide, 2,6-dichloro-3- and -4-hydroxy-benzonitrile, 2,6-dichlorobenzoic acid and 2,6-dichloro-3-hydroxybenzoic acid) and 2,6-dichloro-3-hydroxythio-benzamide were heated separately with 6N-HCI (5 ml.) for3hr. at 1100. Paper chromatograms of the ether extractsof the separate reaction mixtures were run in butan-l-olsaturated with 2N-NH3, and these were identical in everyrespect with the chromatograms of the initial authenticcompounds. In each case, unchanged starting material wasrecovered quantitatively from the reaction mixtures. Theseexperiments demonstrate the great stability of the hinderedamido, cyano and thioamido groups in these molecules andthe lack of reactivity of the chloro substituents in 2,6-positions. Since these constituents were stable to acid andsince, with the exception of 2,6-dichloro-3-hydroxythio-benzamide, they were formed by similar acid hydrolysis ofthe urines of animals dosed with the two 14C-labelledherbicides, it follows that the reactions by which they wereproduced from the (conjugated) metabolites in urine wereprobably straightforward hydrolytic reactions.

Chromatography on a silicic acid column. Mallinckrodt(100 mesh) silicic acid was washed and reactivated asdescribed by Wright et al. (1965). Ether-soluble 14C0labelled metabolites were successively eluted with solventsof increasing polarity; a typical fractionation is set out inTable 7.

Paper chromatography. Unconjugated 14C-labelled meta-bolites of [14C]Prefix and 2,6-dichlorobenzo[14C]nitrilewere separated by descending chromatography onWhatmanno. 1 filter-paper sheets at 29°, and Table 8 showsa typical separation of the ether-soluble 14C-labelledmetabolites. Paper chromatography in the same solventsystem was used for separating the more polar unconjugated14C-labelled metabolites and the 14C-labelled conjugatesin the unhydrolysed urines. All 14C- or 35S-labelledconstituents were detected and identified on paper byradioactive scanning with a Nuclear-Chicago Actigraph II,equipped with a gas-flow cell utilizing a helium-n-butane(98-5:1-5, v/v) mixture. Phenoxyethanol-AgNO3 (Mitchell,1958) was used as location reagent, and the fluorescence ofsome unconjugated metabolites in ultraviolet light servedas another means of location.

Thin-layer chromatography. This was carried out onKieselgel G (E. Merck A.-G., Darmstadt, Germany) asdescribed by Wright et al. (1965). The 14C-labelled meta-bolites were detected and identified on thin plates by radio-active scanning with a Desaga radiochromatogram scanner.Thin-layer chromatography was useful for separating14C-labelled aromatic acids and [14C]phenols; acetone-n-hexane (1:1, v/v) was used as solvent system. The purityof the two 14C-labelled herbicides was checked by thin-layerchromatography, benzene being the preferred solvent fordevelopment.

Gas-liquid chromatography. Chromatographs, eachequipped with an electron-capture detector, were used forcomparing retention times of 14C-labelled metabolites withthose of reference compounds. For the methyl esters,columns of working length 4ft. were filled with Diatopore Sthat had been coated with 3-8% of SE30 silicone oil asstationary phase. The running conditions were: tempera-ture, 1200; inlet pressure, 201b./in.2; flow rate for N2, notknown. For the O-methoxy and neutral constituents,columns of working length 2ft. were filled with untreatedCelite (100-120mesh) that had been coated with 5% ofphenyl diethanolamine succinate as stationary phase. Therunning conditions were: temperature, 1880; flow rate forN2, 50ml./min.

Electrolyte solutions. Formic acid-acetic acid buffer,pH2-0, was prepared by diluting a mixture of 98-100%formic acid (78ml.) and 99.6% acetic acid (148ml.) withwater to 2-51. Acetate-acetic acid buffer, pH5.2, was madeby mixing 0 2N-acetic acid (200ml.) and 0 2M-sodiumacetate (800ml.). Veronal-HCl buffer, pH9 0, was preparedby mixing 936 ml. of 0-1 M-veronal and 64ml. of 0-1 N-HCI.

Low-voltage electrophoresis. After electrophoresis onstrips of Whatman no. 1 filter paper in veronal-HCl buffer,pH9-0, at 12v/cm., [14C]phenols and 14C-labelled aromaticacids had migrated towards the anode; under theseconditions, 14C-labelled phenolic acids were separated from14C-labelled aromatic acids. Electrophoresis in acetatebuffer, pH 5-2, at 12 v/cm. caused 14C-labelled aromatic acidsto migrate towards the anode. After electrophoresis thepaper strips were infrared-heated in an air stream for 10min.to remove most of the liquid and were subsequently driedat 800 for 15 min. Zones of radioactivity were scanned withthe Nuclear-Chicago Actigraph 4ir system.

High-voltage electrophoresis. Electrophoresis on strips ofWhatman no. 1 filter paper in formic acid-acetic acidbuffer, pH2-0, in 1 hr. at 120v/cm. (Hathway, Mallinson &Akintonwa, 1965) caused movement of a 14C_ or 35S-labelled

772 1966


polar constituent towards the cathode with one-half of thespeed of glycine, whereas low-voltage electrophoresis inveronal-HCl buffer, pH9-0, caused the same 14C_ or35S-labelled polar constituent to migrate towards the anode.The dried electrophoretograms were scanned for radio-active zones.

Measurement of radioactivity. A Tri-Carb spectrometer(Packard Instrument Co. Inc., La Grange, Ill., U.S.A.) wasused for the measurement of 14C by liquid-scintillationcounting. Measurements of radioactivity in urine, faeces,gut and contents, carcass plus viscera and skin plus hair andof the 14CO2 in the expired gases were made by the methodsdescribed by Wright et al. (1965).Enzymic hydrolysi8 of the glucuronides of 2,6-dichloro-3-

and -4-hydroxybenzonitrile. Solvent extracts of the urine oftreated animals contained the glucuronides (RF values onpaper0-15 and 0 45 in butan- 1 -ol-2N-NH3) of2,6-dichloro-3-and -4-hydroxybenzonitrile (see the Methods and Resultssections). Separate elution of individual zones from suitablechromatograms provided material for enzymic hydrolysis.After evaporation of the solvent 5mg. of the residue wasdissolved in 0-15M-sodium acetate buffer, pH5-0 (lOml.),and incubated with fl-glucuronidase (500 units) at 370 for3hr., and the mixture extracted with ether. 2,6-Dichloro-3-or -4-hydroxybenzonitrile was detected on thin-layerplates of the ether extract run in acetone-n-hexane (1:1,v/v) with phenoxyethanol-AgNO3 reagent (Mitchell, 1958).Free glucuronic acid was detected with naphtharesorcinol(Tollens, 1908, 1910), and it was identified on paperchromatograms of the concentrated aqueous phase, runin butan-1-ol-acetic acid-water (4:1:1, by vol.), withnaphtharesorcinol (Elliott, Parke & Williams, 1959) andp-anisidine hydrochloride (Hough, Jones & Wadman, 1950).

RESULTSExcretion and retention of radioactivity in rats

dosed respectively with [14C]Preftx and 2,6-dichloro-benzo[14C]nitrile. After oral treatment of six

adult male and six adult female rats with either[14C]Prefix or 2,6-dichlorobenzo[14C]nitrile, themajor eliminative route was via the kidneys and theremAining radioactivity was almost entirely elimi-nated by the faecal route. With the two 14C-labelledherbicides, some 14C was present in the gut of someof the animals at the end of 96hr., especially whenthe animals were constipated on the third andfourth days (Tables 1 and 2). A single dose ofeither [14C]Prefix or 2,6-dichlorobenzo[14C]nitrile isalmost entirely eliminated from rats in 4 days:84-8-100-5% of 14C from [14C]Prefix and 85-8-97-2% of 14C from 2,6-dichlorobenzo[14C]nitrile areexcreted, these values including the contents of thegut at 96hr. Rats do not show a sex difference in thepattern of elimination of the respective metabolicproducts of either Prefix or 2,6-dichlorobenzo-nitrile.Very small amounts of 14C were found in the

carcass and viscera remaining after removal of thegastrointestinal tract from animals dosed with[14C]Prefix or 2,6-dichlorobenzo[14C]nitrile. Thevalues for 14C in the carcass of animals dosed with[14C]Prefix take into consideration the skin and hair(Table 1), whereas the skin plus hair of animalsdosed with 2,6-dichlorobenzo[14C]nitrile was as-sayed separately for 14C.

Excretion of 14C in the urines of rats dosed witheither [14C]Prefix or 2,6-dichlorobenzo[14C]nitrilewas very rapid, and the similar rates of eliminationfor the respective metabolic products of Prefix and2,6-dichlorobenzonitrile in rat urine do not precludethe possibility that 2,6-dichlorobenzonitrile mightbe an intermediate in the metabolism of Prefix(Tables 3 and 4). Radioactivity was slowly

Table 1. Excretion and retention of radioactivity in rats after oral administrationb of [14C]Prefix

Male and female rats (150-200g. body wt.) were treated by stomach tube with 1.8mg. of [14C]Prefix (3.2 1c/mg.)in 2 ml. of aq. 5% ethanol.

Recovery of 14C during 4-day collection (% of dose)

Urine68-079.774.574-879-767-371-8 171-471-874-772-076-0 J

Carcass plusExpired Gut and remaininggases contents viscera

<0-05

<0-05

0

2-5*0

0-50-10-40

0-41-60-70-18-6*

0-40-50-50-50-30-40-30-20-40-30-40-3

* Constipated animals (see Table 3).

Ratno.

123456789101112

Sex

6'CTIs

Y

Faeces20-418-316-017-115-617-116-118-018-120-917-99-2

Total88-8101-091-092-995-785-288-290-091-996-690-494-1

773Vol. 98

M. H. GRIFFITHS, J. A. MOSS, J. A. ROSE AND D. E. HATHWAY

Table 2. Excretion and retention of radioactivity in rate after oral admini8tration of2,6-dichlorobenzo[14C]nitrile

Rats (150-200g. body wt.) were treated by stomach tube with 0-80mg. (males) or 1-3mg. (females) of 2,6-dichlorobenzo[14C]nitrile (9.1,uc/mg.) in 2ml. of arachis oil.

Recovery of 14C during 4-day coHection (% of dose)

Expired Gut andgases contents

<0-02

<0-03

0-10-90-10-14-61-24-00-60-20-10-26.1*

Skin Carcass plusand remaininghair viscera0-1 0-10-3 0-30-1 0-10-2 0-20-2 0-50-2 0-20-8 0-30-6 0-20-7 0-30-5 0-30-5 0-20-5 0-3

* Constipated animal (see Table 4).

Table 3. Rate8 of faecal and urinary elimination of radioactivity in rat8 after oral admini8trationof [14C]PreftX

Male and female rats were treated with [14C]Prefix as described in Table 1.

Elimination of 14C (% of dose)

0-24hr.

Rat no. Sex123 d456 d7 Y

8 S9 y10 y11 V12 S

Faeces Urine15-3 66-111-8 76-39-5 72-4

13-8 72-14-2 75-110-4 65-2

No faeces 70-00-01 67-29-6 67-59-2 68-6

No faeces 67-36-0 71-2

24-48hr.

Faeces4-75-34-32-65-91-5

15-412-33-64-59-83-2

48-72hr.

Urine1-32-31-51-83-20-50-92-72-54-63-33-6

Faeces0-31-22-00-34-943-80-55-11-72-4

7-7No faeces

Urine0-40-60-40-60-81-10-60-60-81-10-90-7

72-96 hr.

Faeces Urine0-1 0-3

No faeces 0-50-3 0-30-4 0-40-51 0-61-3 0-50-1 0-40-6 0-83-2 1-04-7 0-50-4 0-4

No faeces 0-5

eliminated in the faeces, and some irregularity infaecal elimination may be due to coprophagy.Alimentary absorption is virtually complete and thetwo herbicides are rapidly metabolized. A bigproportion of the products of metabolism of eachherbicide enters the circulating blood and is rapidlycleared in the kidneys; the low rate ofelimination ofmetabolites in the faeces is probably due to entero-hepatic circulation.

Elimination of radioactivity in dogs dosed witheither [14C]Preftx or 2,6-dichlorobenzo[14C]nitrile.A single oral dose of either [14C]Prefix or 2,6-dichlorobenzo[14C]nitrile in dogs is almost entirelyeliminated in 4 days: 85-9-106-1% of 14C from[14C]Prefix and 86-8-92-5% of 14C from 2,6-dichlorobenzo[14C]nitrile are excreted. The majoreliminative route was via the kidneys and theremainder of the original label was almost entirely

Ratno.123456789101112

Sex Faeces

CT 19-0d 19-2CT 18-5&3 13-4&T 11-66' 19-1? 13-6? 15-1? 17-7? 16-3V 19-7? 11-5

Urine74-373-272-772-376-074-676-980-774-878-677-377-1

Total93-693-991-586-292-995-395-697-293-795-897-995-5

1966774


Table 4. Rates offaecal and urinary elimination of radioactivity in rate after oral administrationof 2,6-dichlorobenzo[14C]nitrile

Male and female rats were treated with 2,6-dichlorobenzo[14C]nitrile as described in Table 3.


0-24hr..at.no.SexFaeces Urine

Rat no. Sex Faeces Ur-me

24-48hr.

Faeces Urine

48-72hr.

Faeces Urine

72-96hr.

Faeces Urine1 CT 10-82 CT 1.53 & 164 CT 9-55 CT 2-96 CT 5.97 9 2.08 ? No faeces9 9? 4010 9 2-311 9 4.412 9 11.4

69*163'663'966-764-967-665*771-068-571*570*070-1

3.714*114-93.35.0

12-41-57-15.73.76-40.1

3-6 2*87.7 2-47-3 1-62*6 0*47-8 3.13.8 048-6 2-86-7 3.94*1 5-25-6 9-86*0 6-35-2 Nofaeces

1-2 1.71-3 1-31-1 041-2 0-21.7 0-61.0 0*41-7 7.31-6 4.11.5 2-81.0 0*50.9 2*61.1 Nofaeces

Table 5. Rates of faecal and urinary elimination of radioactivity in male and female dogsafter oral administration of either [140]Prefix or 2,6-dichlorobenzo[14C]nitrile

Each dog swallowed a gelatin capsule containing a solution of either 7-90mg. of [14C]Prefix (2.8,uc/mg.) inethanol (1 ml.) or 0-92 mg. of 2,6-dichlorobenzo[14C]nitrile (8.52,uc/mg.) in arachis oil (1 ml.).


BodyDog wt.no. (kg.)

0-24hr.

Sex Faeces Urine

1 14*8 6' No faeces2 14*5 6' 6-63 10-4 9 No faeces4 10.2 9 0-6

24-48hr._s

Faeces Urine

L14C]Prefix33-5 No faeces33 0 No faeces55-1 No faeces51-3 3.8

48-72hr.

Faeces Urine

18*5 24-416'2 9*220-8 24-518-2 No faeces

72-96hr.e U

Faeces lUrine

11.9 1-614*6 2-24.1 0-76-4 11-6

5 11-5 16-86 12-3 15*17 12-7 16-68 7-2 11-2

50-744-859.744*1

2,6-Dichlorobenzo[14C]nitrile4.3 10.7

10-2 11.21-8 6*65.3 13.7

1.5 2.3 0 4 0.81*5 2-9 0-4 1.13-8 3-2 0-2 0-62-6 6-2 1.4 2-3

eliminated by the faecal route (Table 5). Dogs donot show a sex difference in this pattem of elimina-tion of the metabolites of either Prefix or 2,6-dichlorobenzonitrile. Dogs and rats do not showspecies differences in the patterns of elimination ofthe metabolic products of either Prefix or 2,6-dichlorobenzonitrile. Excretion of 14C in the urinesof treated dogs was slower than in rats. Theelimination of 14C in the urines of dogs treated with[14C]Prefix was irregular and the rate of eliminationof 14C was different for the two sexes. This experi-ment was complicated by irregular defaecation ofthe animals, which were ill-adjusted to the diet,and the number of observations on the output of

14C in the urine was too small for a conclusion on thedifference in rate to be made.

Separation of fractions of chemically similarsubstances from the hydrolysed urines of rats dosedrespectively with [14C]Prefix and 2,6-dichlorobenzo-[14C]nitrile. Rats treated with Prefix or 2,6-dichlorobenzonitrile do not eliminate either herbi-cide unchanged. 2,6-Dichlorobenzonitrile may be a

metabilite of Prefix with a high turnover, since2,6-dichlorobenzonitrile is itself completely meta-bolized in rats, and to test this supposition themetabolic products of the two herbicides wereinvestigated.By successive extraction at pH 1-5-2-0 with ether

0-30-60*41.81-62-20.91-40-70.50*40-7

2-74-10.91-6

Vol. 98 775

M. H. GRIFFITHS, J. A. MOSS, J. A. ROSE AND D. E. HATHWAYand ethyl acetate, 80% of the radioactivity wasrecovered from the hydrolysed urines of rats dosedrespectively with [14C]Prefix and 2,6-dichloro-benzo[14C]nitrile. In both cases the ethyl acetateextract contained one-half of the amount of 14Cfound in the ether extract. Scanning revealed sixradioactive zones on paper chromatograms of theether-soluble material that had been run in butan-l-ol saturated with 2N-ammonia, and the fact thatthey reacted with phenoxyethanol-silver nitrate(Mitchell, 1958) was regarded as evidence that themain 14C-labelled metabolite in each of the zonescontained chlorine. These ether-soluble 14C-labelledconstituents were fractionated by colunm chromato-graphy into aromatic acids, neutral constituents andphenols (Table 6). The ethyl acetate extracts ofhydrolysed urines contained the more polar 14C_labelled constituents, which also reacted with thephenoxyethanol-silver nitrate reagent.A solution of each phenolic fraction in sodium

hydroxide was purified by solvent extraction.Electrophoresis in veronal-hydrochloric acid buffer,pH 9 0, caused phenolic 14C-labelled constituents tomigrate towards the anode, whereas in acetatebuffer, pH 5-2, they did not move. Major 14C-labelled constituents that were separated by thin-layer chromatography were identified as 2,6-dichloro-3- and -4-hydroxybenzonitrile by theirfluorescences in the ultraviolet and chromatographicproperties (Table 7).A solution of each neutral fraction was purified

by extraction with 2 N-sodium hydroxide, andelectrophoresis of these fractions in buffers, pH 2@0-9-0, did not cause movement. A single 14C-labelledconstituent was identified as 2,6-dichlorobenzamideby its chromatographic properties (Table 7).The acidic fractions were purified eitherby solvent

extraction of their solution in aqueous sodiumhydrogen carbonate or by thin-layer chromato-graphy (Table 7). After electrophoresis in acetate

Table 6. Separation of ether-soluble 14C-labelled constituents of the hydrolysed urinesinto fractions of chemically similar substances

Proportions of the ether-soluble material from the hydrolysed urines of rats (150-200g. body wt.) that had eachbeen dosed with 1-8mg. of [14C]Prefix (3.2,uc/mg.) and 0-8mg. of 2,6-dichlorobenzo[14C]nitrile (9.1 nc/mg.) werechromatographed on a silicic acid column (10 mm. diam. x 10 cm. long). Recovery was quantitative.

Fraction no.(10 ml.) Eluent

1-3 20% Acetone in n-hexane4-6 50% Acetone in n-hexane7-9 Acetone

Recovery of radioactiveconstituents

Phenolic fraction (80% of 14C)Neutral fraction (1% of 14C)Aromatic acids (19% of 14C)

Table 7. Fluorescences, RF values on paper and thin plates, and retention times on gas-liquid chromatographsfor the ether-soluble constituents in hydrolysed urines of rats dosed with either Preftx or 2,6-dichlorobenzonitrile

The labelled constituents were identified by comparing their chromatographic and fluorescence propertieswith those of the authentic unlabelled compounds.

RF

Constituents in hydrolysed urines2,6-Dichlorobenzamide2,6-Dichloro-3-hydroxybenzonitrile2,6-Dichloro-4-hydroxybenzonitrile2,6-Dichlorobenzoic acid2,6-Dichloro-3-hydroxybenzoic acidUnknown aromatic acid(2,6-Dichloro-4-hydroxybenzoic acid)

Paper inbutan-l-olsatd. with2N-NH30-920660-720.450-060-16

Thin-layerplates in

acetone-n-hexane(1:1, VfV)

0-709070-020-020-02

Fluorescencein u.v.

Dull violetBright violetDull violetDull violetDull violet

Retentiontimesfor the

methylatedderivatives*

(min.)15-04 0 L(a)3-51410°(b)

* After their separation, the acid, the two phenols and the phenolic acid respectively were exhaustively methylated withexcess of diazomethane in ether solution, and n-hexane solutions of the methylated derivatives were used for gas-liquidchromatography: (a) Stationary phase: 5% (w/w) phenyl diethanolamine succinate on Celite; flow rate, 50ml./min.;temperature, 1880. (b) Stationary phase: 3-8 (w/w) SE 30 silicone oil on Diatopore S; chromatograph not equipped withflow-meter; temperature, 1200.

776 1966


buffer, pH 5-2, acidic 14C-labelled constituentsmigrated towards the anode. Major 14C-labelledcomponents in the acidic fractions that were

separated by paper chromatography were identifiedas 2,6-dichlorobenzoic acid and 2,6-dichloro-3-hydroxybenzoic acid by their chromatographicproperties (Table 7). An unidentified 14C-labelledcomponent was possibly 2,6-dichloro-4-hydroxy-benzoic acid. After electrophoresis in veronal-hydrochloric acid buffer, pH9 0, 2,6-dichloro-3-hydroxybenz[14C]oic acid migrated twice as far as

2,6-dichlorobenz[14C]oic acid.Polar 14C-labelled constituents, which account

for the radioactivity in the residual aqueous phaseafter ether extraction, could be quantitativelyextracted with butan- 1-ol, but the resulting solutionwas contaminated with large amounts of the com-

ponents of normal urine, including electrolytes andwater. A solution, containing major proportions ofthese 14C-labelled constituents and smaller amountsof the contaminating substances, especially electro-lytes and water, was prepared by exhaustiveextraction with ethyl acetate; four main radioactivezones and two minor ones were present on paper

chromatograms. At least one 14C-labelled com-

ponent was adsorbed on to a column of cation-exchange resin in the H+ form and eluted with basicbuffers. Electrophoresis (120v/cm.) in acetic acid-formic acid buffer, pH2-0, caused a single radio-active peak to migrate towards the cathode withone-half of the speed of glycine. In acetate buffer,pH 5 2, there was no movement in an electric field,and electrophoresis in veronal-hydrochloric acidbuffer, pH9 0, caused a single radioactive peak tomove towards the anode. The presence of at leastone deacetylated [14C]mercapturic acid in thehydrolysed urines would account for these observa-tions, and this supposition was tested by labellingthe cysteine-cystine pools of rats with 35S andtreating these animals with the unlabelled herbi-cides (see the Methods section; Gutmann & Wood,1950; Marsden & Young, 1958). This experimentprovides clear evidence for the presence in thehydrolysed urines of two 35S-labelled constituents,RFO37 and 0 47 on paper. These 35S-labelledconstituents were absent from paper chromato-grams of the hydrolysed urine of the controlanimals that had been given [35S]cystine. Neitherof these 35S-labelled constituents could have beenconfused with 35SO42-, which ran entirely differentlyin the butan-1-ol-2N-ammonia solvent systememployed, and neither of them was due to organic[35S]sulphate esters which would have beenhydrolysed by the acid hydrolysis. These 35S-labelled hydrolysis products were ninhydrin-positive, and they migrated in an electric field(120v/cm.) in acetic acid-formic acid buffer,pH 2-0, towards the cathode and in veronal-

hydrochloric acid buffer, pH 9 0, towards the anode.The foregoing chromatographic and electrophoreticproperties of the 35S-labelled constituents corre-sponded to those of two 14C-labelled constituentspresent in the hydrolysed urines of animals dosedwith the 14C-labelled herbicides. The conclusion isdrawn that the hydrolysed urines of animals dosedwith either [14C]Prefix or 2,6-dichlorobenzo[14C]-nitrile and the hydrolysed urines of animals treatedwith [35S]cystine and the unlabelled herbicidescontained two sulphur-containing constituentsderived from metabolite(s) of the two herbicides,and that the physical and chemical properties ofthese constituents are consistent with the possessionof an associated amino acid residue. From theconspectus of knowledge on the mammalianmetabolism of foreign organic compounds, it isknown that there are two common types of sulphur-containing metabolites, mercapturic acids andsulphate esters (see Young & Maw, 1958). For thehydrolytic products, we are not concerned withsulphate esters and the sulphur-containing com-pounds in the hydrolysed urines may have beenproduced from mercapturic acid(s) or premercap-turic acid(s) by hydrolysis. The nature of theamino acid residue in each sulphur-containingconstituent has not been investigated, and thestructures of the two sulphur-containing con-stituents have not been elucidated; this wouldrequire the isolation ofpure compounds followed bycareful degradative studies. However, our know-ledge ofthe chemistry ofthe two herbicides indicatesthat the chloro substituents would not be expectedto undergo nucleophilic replacement, and that thereactivities of the cyano group of 2,6-dichloro-benzonitrile and of the thioamido group of Prefixare very hindered indeed. Thus the addition ofhydrogen sulphide to 2,6-dichlorobenzonitrile takesplace under forcing conditions, whereas the additionof mercaptans does not take place under anyconditions (Dr J. Yates, personal communication).The two sulphur-containing compounds in hydro-lysed urines of treated animals may therefore besubstituted 2,6-dichlorobenzonitriles, and this wastested by dethiolation with Raney nickel catalyst,when each of the sulphur-containing constituentsgenerated a compound with the chromatographicproperties of 2,6-dichlorobenzonitrile. The dataseem to be consistent with the partial-type formula(VII) (Scheme 1) suggested for these compounds.In the work-up of the hydrolysed urine from

animals dosed with the 14C-labelled herbicides, thepreviously mentioned 14C that remained in theaqueous phase after exhaustive extraction withether and ethyl acetate, and that could be com-pletely extracted with butan- l-ol, was shown to bealmost entirely due to further amounts of the twosulphur-containing constituents, RFO037 and 0-47

Vol. 98 777


Table 8. Proportions of 14C-labelled contituentM in the hydroly8ed urines of rat8 dosedrespectively with [14C]Prefix and 2,6-dichlorobenzo[14C]nitrite

The labelled constituents were identified by comparing their chromatographic and fluorescence properties withthose of the authentic unlabelled compounds.

Proportions of 14C-labelled metabolites(% of total 14C in the urines)

Constituents in hydrolysed Rats dosedurine with ... [14C]Prefix

2,6-Dichlorobenzamide < 1-02,6-Dichlorobenzoic acid 2-32,6-Dichloro-3-hydroxybenzoic acid 2-82,6-Dichloro-3-hydroxybenzonitrile 42'02,6-Dichloro-4-hydroxybenzonitrile 2-8Unknown aromatic acid (R).016) 2-1Ethyl acetate-soluble metabolites: (I) 3 0

(II) 4.9(III) 8*1(IV) 9.4(V) < 1.0(VI) < 1*0

Recovery ... ... > 77.4

2,6-Dichlorobenzo-[14C]nitrile

<1*0<1*05.0

41-74-01-33-06-89.06-52-21*5

> 81*0

CS * NH2

Cl(Cl

(I)

CN

Cl ICl

(II)

CN

Cl Cl

OH

(III)

CN

CI Cl

OH

(V)

C02H

Cl )Cl

OH

(IV)

CO2H

Cl g<Cl

OH

(VI)

CNCl Cl

S -1CH2] n * CH (NH2) *C(2H

(VII)

CO*NH2Cl Ci

(VIII)

CO2H

(IX)

Scheme 1. Transformations of Prefix and 2,6-dichlorobenzonitrile in the rat.

778 1966


on paper. Greater proportions of these compoundsthan those shown in Table 8 were therefore presentin the hydrolysed urines, from which they wereincompletely extracted with ethyl acetate. Thetwo sulphur-containing constituents are accordinglyimportant to the transformations of Prefix and2,6-dichlorobenzonitrile in the rat.The structures ofthe 14C-labelledconstituentsthat

have been investigated in the hydrolysed urine arerelated to those of the two herbicides in such a waythat they may have been formed direct from the14C-labelled metabolites by hydrolysis.

Relative proportions of the various constituents inthe hydrolysed urine of rats dosed with Prefix and2,6-dichlorobenzonitrile. Quantitative aspects of thecomparative metabolism of Prefix and 2,6-dichloro-benzonitrile in rats are summarized in Table 8.Pooled urines of 12 rats dosed with [14C]Prefix andpooled urines of 12 rats dosed with 2,6-dichloro-benzo[14C]nitrile were separately hydrolysed in theway that has been described, and radioactivitycorresponding to the different 14C-labelled con-stituents present in the various fractions wasmeasured. These results were reproducible, theexperiment being repeated with two further setsof 24 rats with identical results. Metabolism of thetwo herbicides yields the same constituents in thehydrolysed urine in strikingly similar proportions.For Prefix (I) (Scheme 1), m-hydroxylation accountsfor 44.8% of the constituents (III and IV), and, onthe assumption that the unknown aromatic acid is2,6-dichloro-4-hydroxybenzoic acid (VI), p-hyd-roxylation accounts for 4.9% of the constituents(V and VI). For 2,6-dichlorobenzonitrile (II), thecorresponding values are 46-7 and 5-3%. Meta-bolism of the two herbicides also yields approxi-mately similar proportions of polar constituents, ofwhich two of the ethyl acetate-soluble constituentsare sulphur-containing compounds of partial-typeformula (VII).

Since Prefix and 2,6-dichlorobenzonitrile haveessentially similar biological properties and herbi-cidal activities (Barnsley, 1960), and since 2,6-dichlorobenzonitrile is the main metabolite ofPrefix in plants and unsterilized soil (Barnsley,1960; Milborrow, 1965), the possibility arises thatPrefix might be transformed into 2,6-dichloro-benzonitrile in animals, and the fact that themetabolism of the two herbicides has given thesame constituents in the hydrolysed urine instrikingly similar proportions strongly suggestsan interrelationship of the metabolic pathwaysof the two herbicides. Prefix therefore appearsto be extensively transformed into 2,6-dichloro-benzonitrile:

R*CS.NH2 R*CN+H2Swhere R = C12C6H3

There is slight evidence that the competitivereactions:

R.CS.NH2+H20 -÷R.CO0NH2+H2SR.CO*NH2+H20 -R*RCO2H+NH3

take place to a very limited extent, since thehydrolytic product, 2,6-dichlorobenzoic acid (IX)(Scheme 1), is formed in vivo in greater quantityfrom Prefix than from 2,6-dichlorobenzonitrile;but the same quantity of 2,6-dichlorobenzamidewas measured in parallel experiments with the twoherbicides.The results are not consistent with direct

hydroxylation of Prefix and subsequent hydrolysisofthioamido products. The stability of 2,6-dichloro-3-hydroxythiobenzamide to boiling 6 N-hydrochloricacid, and its absence from the hydrolysed urines ofanimals dosed with Prefix, is strong evidence thathydroxylated thiobenzamides were not formedin vivo.Scheme 1 shows the transformations of Prefix and

2,6-dichlorobenzonitrile in rats.14C-labelled metabolites in the unhydrolysed urines

of rats dosed respectively with [14C]Preftx and 2,6-dichlorobenzo[14C]nitrile. Free 2,6-dichloro-3- and-4-hydroxybenzonitrile were extracted in smallquantity from the urine of rats dosed with eitherPrefix or 2,6-dichlorobenzonitrile. These phenolsmay have been excreted as their sulphate esters,which would probably have been hydrolysed underthe prevailing acidic conditions of our etherextraction (see Young & Maw, 1958). In otherexperiments, urine at pH 6-0 was extracted withether, and the material in the residual aqueous phasewas submitted to paper electrophoresis (13v/cm.) in0-025M-phthalate buffer, pH5-9, for various times.The rapidly migrating radioactive area from + 14 to+ 20cm. (Fig. 1) of strongly anionic substances wastentatively ascribed to strongly acidic sulphateesters; in another context similar assignments havealso been made (Williams, Meikle & Redemann,1964). The relative proportion of strongly anionicsubstances, when urines were extracted at pH 6*0,was almost the same as that of the free phenols,when the urines were extracted under more acidconditions.

Glucuronides of 2,6-dichloro-3- and -4-hydroxy-benzonitrile had R, 0415 and 0 45 on paper in thebutan-l-ol-2N-ammonia solvent system. Suchcompounds were also present in the urine of treatedanimals, and separation by paper chromatographyfollowed by hydrolysis of individual componentswith ,B-glucuronidase generated 2,6-dichloro-3- and-4-hydroxybenzonitrile and free glucuronic acid(for details see the Methods section).The aromatic acids that had previously been

found in the hydrolysed urines of rats dosed withthe two herbicides appear to be present as ester

779Vol. 98


140-

1~20-

~&100

80--0

40-

~ 0

+ IS 1 0 0

Distance (cm.)

Fig. 1. Electrophoretic separation of 14C-labelled meta-bolites in the residual aqueous phase after ether extractionof urine at pH6-0. Metabolites were separated from theresidual aqueous phase by electrophoresis on Whatmanno. 1 filter papers in 0 025M-phthalate buffer, pH5-9, at13 v/cm. for 2-5hr. ( ) or 1 0hr. (----).

glucuronides in the original urines. Hydrolysis with,B-glucuronidase caused the disappearance of radio-active spots from the paper chromatograms ofether-extracted urines and the simultaneous genera-tion of other radioactive spots due to the 14C_labelled aromatic acids that had been found in thehydrolysed urines.The urine of rats in which the cysteine-cystine

pools had been labelled with 35S and where theanimals were dosed with unlabelled herbicidescontained two 35S-labelled metabolites that wereabsent from the urine of control rats that had beengiven [35S]cystine. One of these substances could beextracted by ether (Rp020 on paper in butan-l-ol-2N-ammonia) and the other by ethyl acetate (R, 0 27on paper in butan-1-ol-2 N-ammonia). Thesemetabolites did not migrate in an electric field(120v/cm.) towards the cathode in acetic acid-formic acid buffer, pH 2 0, and they did not give acolour reaction with ninhydrin. The chromato-graphic and electrophoretic properties of the fore-going 35S-labelled metabolites corresponded tothose of two 14C-labelled metabolites present in theurines ofrats dosed with the 14C-labelled herbicides.The conclusion is drawn that the urines ofrats dosedwith [14C]Prefix or 2,6-dichlorobenzo[14C]nitrilecontained two sulphur-containing metabolites.Hydrolysis of these metabolites with 6N-hydro-chloric acid gave the two 14C-labelled sulphur-containing amino acid constituents, which migratedin an electric field towards the cathode in buffers ofpH 2 0 and towards the anode in buffers of pH9-0.The generation ofa single sulphur-containing amino

acid from each of the sulphur-containing meta-bolites, in the way that has been described, mayhave been brought about by elimination of an acylgroup from the amino nitrogen atom ofeach sulphur-containing acylamino acid metabolite. The condi-tions selected for hydrolysis of the urine were thosethat would have been sufficiently drastic to effectN-deacetylation of N-acetylcysteyl-containingmetabolites (Thomson, Barnsley & Young, 1963).The structures of the two sulphur-containingmetabolites have not been elucidated; at first sight,they may be analogous to 2,6-dichlorobenzonitrilemercapturic acids or premercapturic acids.

Quantitative assessments were made on theunhydrolysed urines of rats dosed with [14C]Prefixand 2,6-dichlorobenzo[14C]nitrile to give informa-tion for whole urines of the sort reported in Table 8for hydrolysed urines. Not much would be gained bythe inclusion of detailed information, since referencewould have to be made to unknown metabolitesand partially elucidated metabolites. Valuable con-firmation was obtained of the amounts of 14C_labelled constituents present in the various solventextracts of hydrolysed urines by the preparation ofsolvent extracts of hydrolysates of the correspond-ing solvent extracts of whole urine.

DISCUSSION

Treatment of animals with Prefix or 2,6-dichloro-benzonitrile has given rise to the expected metabolicproducts.

3- and 4-Hydroxylation of 2,6-dichlorobenzo-nitrile has now been demonstrated in vivo;hydroxylation is predominantly meta, owing tothe inductive effect of the 2,6-dichloro atoms. The2,6-dichloro substituents contribute to the rapidrate of hydroxylation of 2,6-dichlorobenzonitrilein vivo, since the slowness with which the metabolicproducts of benzonitrile and nitrobenzene areeliminated from the body has been attributed todifficulty associated with the hydroxylation ofaromatic compounds that have meta-directinggroups (Bray, Hybs & Thorpe, 1951; Smith, 1950;Smith & Williams, 1950). By analogy with chloro-benzene (Smith, Spencer & Williams, 1950) andnaphthalene (Corner & Young, 1955), the possibilitythat hydroxylation of 2,6-dichlorobenzonitrile invivo may involve intermediate formation of 2,6-dichloro-3,4-dihydro-3,4-dihydroxybenzonitrile isattractive, since this diol might itself be formedfrom an epoxide precursor, which would also beexpected to react with the thiol group of cysteine,glutathione or tissue proteins (Parke & Williams,1958) to yield the two sulphur-containing acylaminoacid metabolites that have been discussed.

In general, arylnitriles are metabolized to only a

780 1966

Vol. 98 COMPARATIVE METABOLISM OF TWO HERBICIDES 781

small extent by reactions involving the cyanogroup:

Ar.CN -*Ar.CO.NH2 -*Ar.CO2H

The reaction is a minor one, and its extent dependson the nature of the substituents in the benzenenucleus. With benzonitrile and o-tolunitrile, oxid-ation of the aromatic nucleus is the major pathwayof metabolism, but at least 10% of benzonitrileand 25% of o-tolunitrile are metabolized byhydrolysis of the cyano group to carboxyl (Brayet al. 1951). Less than 2% of the sterically hindered2,6-dichlorobenzonitrile is metabolized in vivoto 2,6-dichlorobenzamide plus 2,6-dichlorobenzoicacid. These compounds cannot be produced invitro by drastic acid hydrolysis, and it is noteworthythat the amide is stable to prolonged refluxing with6 N-hydrochloric acid. The hydroxylation products,2,6-dichloro-3- and -4-hydroxybenzonitrile, are alsopartially hydrolysed in vivo to 2,6-dichloro-3- and-4-hydroxybenzoic acid.In agreement with the known behaviour of other

arylnitriles, no evidence was found for the formationof hydrogen cyanide from 2,6-dichlorobenzonitrilein vivo. Thus when 2,6-dichlorobenzonitrile wasadministered orally, LD50 values (expressed asmg./kg.) were more than 3000 (rats), 866 (mice) andmore than 1250 (chickens). The toxicologicalproperties and symptomatology of acute poisoningare incompatible with release of hydrogen cyanide.

The authors thank their colleagues Mr P. A. Harthoornfor synthesizing the radioactive herbicides, Mr A. Richard-son for help with the gas-liquid chromatography and Dr J.Yates for providing some intermediates and referencecompounds. They are also grateful to Dr J. G. Wit, Instituutvoor Veterinaire Farmacologie der Rijksuniversiteit teUtrecht, for a gift of 2,6-dichloro-4-hydroxybenzonitrile.

REFERENCES

Bargellini, G. (1929). Gazz. chim. ital. 59, 16.Barnsley, G. E. (1960). Proc. 5th Conf. Briti8h Weed

Control Council, Brighton, pp. 597-616.Belgian Patent (1957). Belg. Patent 572662 (to N. V.

Philips Gloeilampenfabrieken).Bray, H. G., Hybs, Z. & Thorpe, W. V. (1951). Biochem. J.

48, 192.

Corner, E. D. S. & Young, L. (1955). Biochem. J. 61, 132.Davies, W. (1921). J. chem. Soc. 119, 873.Elliott, T. H., Parke, D. V. & Williams, R. T. (1959).

Biochem. J. 72, 193.Fierz-David, H. E. & Blangey, L. (1952). Grundlegende

Operationen der Farbenchemie, 8th ed., p. 154. Vienna:Springer-Verlag.

Fishman, W. H., Springer, B. & Brunetti, R. (1948).J. biol. Chem. 173, 449.

Gutmann, H. R. & Wood, J. L. (1950). Cancer Res. 10, 8.Hathway, D. E., Mallinson, A. & Akintonwa, D. A. A. (1965).

Biochem. J. 94, 676.Hodgson, H. H. & Beard, H. G. (1926). J. chem. Soc. p. 147.Hough, L., Jones, J. K. N. & Wadman, H. (1950). J. chem.

Soc. p. 1702.Koopman, H. (1961). Rec. Trav. chim. Pays-Bas, 80,1075.Koopman, H. & Daams, J. (1960). Nature, Lond., 186, 89.Korner, W. & Contardi, A. (1913). R.C. Accad. Lincei, 22,

828.Marsden, C. M. & Young, L. (1958). Biochem. J. 69, 257.Mazzara, G. & Bertozzi, V. (1900). Gazz. chim. ital. 30, 84.Milborrow, B. V. (1965). Weed Res. 5, 332.Mitchell, L. C. (1958). J. Ass. off. agric. Chem. Wash. 41,

781.Parke, D. V. & Williams, R. T. (1958). Rep. Progr. Chem.

55, 376.Pearl, I. A. (1950). Org. Synth. 30, 101.Reich, S. (1917). Bull. Soc. chim. Fr. 21, 223.Reich, S., Salzmann, R. & Kawa, D. (1917). Bull. Soc. chim.

Fr. 21, 217.Smith, J. N. (1950). Symp. biochem. Soc. 5, 15.Smith, J. N., Spencer, B. & Williams, R. T. (1950). Biochem.

J. 47, 284.Smith, J. N. & Williams, R. T. (1950). Biochem. J. 46,243.Strating, J. & Backer, H. J. (1943). Rec. Trav. chim. Pays-

Bas, 62, 57Thomson, A. E. R., Barnsley, E. A. & Young, L. (1963).

Biochem. J. 86, 145.Tollens, C. (1908). Hoppe-Seyl. Z. 56, 115.Tollens, C. (1910). Hoppe-Seyl. Z. 67,138.Williams, E., Meikle, R. W. & Redemann, C. T. (1964).

J. agric. Fd Chem. 12, 457.Wright, A. S., Akintonwa, D. A. A., Crowne, R. S. &Hathway, D. E. (1965). Biochem. J. 97, 303.

Yates, J. (1961). Brit. Patent 987 253 (to Shell ResearchLtd.).

Yates, J. (1962). Brit. Patent 902 586 (to Shell ResearchLtd.).

Young, L. & Maw, G. A. (1958). The Metabolism ofSulphurCompounds, p. 34 et seq. and p. 107 et seq. London:Methuen's Monographs on Biochemical Subjects.

and 2,6-dichlorobenzonitrile in the dog and rat

Documents