Synthesis of Triazino Benzimidazol as Biopesticides
[1,2,4]Triazino [4,3 a] benzimidazol 4(10 H) ones(4) have been obtained by the reaction of 2 hydrazinobenzimidazole with ethyl pyruvate in neutral medium followed by hydrolysis and cyclization. Further, it has been found that these compounds exist in two tautomeric forms due to the labile hydrogen. The compounds display promising activity when screened for their antibacterial and antifungal activities.
Various 2 substituted benzimidazoles possessing antimicrobial activity have been reported by several group of workers. 2 Hydrazinobenzimidazoles are highly reactive compounds; some of the derivatives have been used as azodyes while some other derivatives show tuberculostatic and influenza virus inhibition activities. In pursuit of our search for new and better antimicrobial agents, we have now synthesized [1,2,4] triazinobenzimidazol 4(10 H) ones(4) by the reaction of 2 hydrazinobenzimidazoles with ethyl pyruvate. This procedure which involved three steps (Scheme 1) gave 4 in better yields than that reported by other workers from the reaction of ethyl pyruvate with other hydrazines.
2 Hydrazinobenzimidazoles on refluxing with ethyl pyruvate in ethanol for 6 7 hr gave ethyl an oxopropionate benzimidazol 2 ylhydrazones (2). Alkaline hydrolysis of 2 in 50% ethanol gave an oxopropionic acid benzimidazol 2 ylhydrazones (3) which underwent cyclization with gl. acetic acid affording 4 (Scheme 1).
In the PMR spectrum of 4d, the aromatic protons appeared as a multiplet in the region d. 7, 1 7.3, as in the case of hydrazone 3d. This indicates the cyclization to be linear giving 4 and not the angular isomer (4). If 4 had been formed, C9 H would have appeared slightly downfield due to deshielding effect of the carbonyl group.
The characterization data of all the new compounds are given in Table 1. The IR, PMR, 13C NMR and mass spectral data of only representative compounds have been given (see Experimental).
Methylation of 3 methyl [1, 2, 4] triazino [4, 3 a] benzimidazol 4(10H) one (4a) in alkaline medium with methyl iodide, gave two compounds (5a, 5b). One of them was found to be 3,10 dimethyl [1,2,4] triazino [4,3 a] benzimidazol 4 (10H) one (5a) and the other as 1,3 dimethyl [1,2,4] triazino [4,3 a] benzimidazol 4 (10 H) one(5b) on the basis of PMR data. The N methyl protons in 5b appeared at d 3.5 as compared to 5a where these appeared at 4.0, the data being in agreement with the respective environment. The formation of these products prompted us to make a new and important conclusion that [1, 2, 4] triazino [4, 3 a] benzimidazol 4(10H) ones exist in two tautomeric forms A and B. The former being the predominent form as indicated by the high yield of 5a.
Antimicrobial activity
Compounds 4a and 4c 4e of the series were evaluated for their antimicrobial activity following the method of Gould using streptomycin in antibacterial and mycostatin in antifungal activity as reference compounds.
All compounds were active against gram positive bacteria Staphylococcu saureus while none was active against gram negative bacteria. Compound 4e showed maximum zone of inhibition (15.0 mm) against Staphylococcus aureus. Its enhanced activity may be attributed to the presence of fluorine. Compounds 4d and 4e were active against all the fungi tested. All the compounds showed maximum activity against Aspergillus flavus. Compound 4a showed maximum zone of inhibition (10.0 mm) against the fungus Drechslera tetramera. The results are recorded in Table 2.
Experimental
IR spectra in KBr (nmax in cm-1) were recorded on a Perkin Elmer 577 spectrophotometer, PMR spectra in TFA on a Jeol FX 90Q spectrometer (89.55 MHz) using TMS as internal reference and 13CNMR spectra in DMSO d6, at 22.49 MHz (chemical shifts in both cases in d, ppm). The mass spectra were recorded on MS 30 and MS 50 Krantos mass spectrometers at an ionisation potential of 70 eV. Melting points were determined in open glass capillaries on a Gallenkamp melting point apparatus and are uncorrected. Elemental analyses were performed at CDRI, Lucknow.
Studies on Coumarin Derivatives
On the synthesis and evaluation of biological activities of coumarins, we report here the synthesis of 1 aroyl 1, 2 dihydro 3H naphtho [2, 1 b] pyran 3 ones (2) and the reactivity of their C 2 methylene groups toward some organic reagents aiming to synthesis new naphthopyrane derivatives which might have enhanced biological activities. Coumarin derivatives are known to exhibit bactericidal, bacterostatic, anticoagulant, and anticancerogenic, rodenticidic and antihelminthic activities.
Condensation of b (4 methoxy or 3,4 dimethyl benzoyl) acrylic acid with 2 naphthol in the presence of 75% sulphuric acid gave 1 (4 methoxy or 3,4 dimethylbenzoyl) 1, 2 dihydro 3H naphtho [2,1 b] pyran 3 one (2a or 2b) (Scheme 1) via ring closure of the intermediate b aroyl b (2 hydroxynaphthalen 1 yl)propionic acid (1), which gave positive acidity test and was formed exclusively and predominantly in the presence of 50% sulphuric acid.
Claisen Schmidt condensation of the naphthopyrones (2) with aromatic aldehydes such as benzaldehyde, 2 and 4 chloro benzaldehydes, and 3 and 4 nitrobenzaldehydes in the presence of piperidine afforded 1 aroyl 2 arylidene 1, 2 dihydro 3H naphtho [2,1 b] pyran 3 ones (3a f; Scheme 2) (Table 1).
Bromination of 2 in carbon tetrachloride at room temperature resulted in electrophilic substitution at both C 1 and C 2 to give 1,2 dibromo derivatives (4a and 4b; Table 1), while bromination of 2 in boiling acetic acid gave 1 aroyl 5 bromo 3H naphtho [2,1 b] pyran 3 ones (5a and 5b) (Scheme 2) which probably resulted from bromination at C 1 followed by dehydrobromination and subsequent bromination of the more reactive aromatic C 5.
Base catalyzed addition of 2 to 4 methoxybenzalacetophenone at 25° in the presence of sodium ethoxide yielded the expected Michael adducts 1 aroyl 2 (1 p methoxyphenyl 3 oxo 3 phenyl prop 1 yl) 1, 2 dihydro 3H naphtho [2,1 b] pyran 3 ones(6a and 6b) while in the presence of isobutylamine, 2 underwent cycloaddition to give 12 aroyl 8, 12 dihydro 8 isobutyl 11 (p methoxyphenyl) 9 phenyl 11H naphtho [1, 2 :5, 6] pyrano[2,3 b] pyridines (7a and 7b).
On the other hand at 120° the naphthopyrones (2) underwent base catalyzed cycloaddition with 4 methoxybenzalacetophenone and/or ethyl benzoylacrylate to give the corresponding cyclic Michael adducts depending upon the type of base used. In the presence of sodium ethoxide, the reaction yielded 1 aryl 2 benzoyl 3a, 11 c dihydro 3 (p methoxy phenyl/or carbethoxy) cyclopenta [d] naphtho [2,1 b] pyran 4(3H) ones (8a c) while in the presence of ammonium acetate it gave 12 aroyl 8, 12 dihydro 11 (p methoxyphenyl) 9 phenyl 11 H naphtho [1 , 2 : 5,6] pyrano [2,3 b] pyridines (7c and 7d) and in the presence of isobutylamine 4,7 dihydro 3 (2 hydroxynaphthyl) 1,7 diisobutyl 4 (p methoxyphenyl) 6 phenyl 2 (3,4 xylyl) 1H pyrrolo[2,3 b] pyridine (9) was obtained.
Experimental Procedure
Melting points recorded are uncorrected. IR spectra were recorded on a Unicam SP 1200 spectrophotometer using KBR wafer technique (nmax in cm-1), and PMR spectra on a Varian T 60 instrument using TMS as internal standard (chemical shifts in d scale).
b (p Anisoyl) b (2 hydroxynaphthalen 1 yl)propionic acid(1)
A mixture of 2 naphthol (7.2 g, 0.05 mol), b (p anisoyl) acrylic acid (10.3 g, 0.05 mol) and sulphuric acid (80 ml, 50%) was warmed on a water bath for 2 hr, cooled, and poured into 200 ml cold water. The solid separated was filtered, washed with water, dried and crystallised from benzene to give 1 as brown crystals, yield 80%.
1 Aroyl 1, 2 dihydro 3H naphtho [2, 1 b] pyran 3 ones (2)
A mixture of 2 naphthol (7.2 g, 0.05 mol), b (p anisoyl)acrylic acid/or b (3,4 dimethylbenzoyl) acrylic acid (0.05 mol) and sulphuric acid (80 ml, 75%) was warmed on a water bath for 2 hr, cooled and poured into 200 ml cold water. The solid separated was filtered, washed with water, dried and crystallised from benzene light petrol (80 100) to give 2a or 2b as orange crystals, yield 71 75%.
1 A royl 2 arylidene 1,2 dihydro 3H naphtho[2,1 b]pyran 3 ones(3)
A solution of 2 (0.01 mol), an aromatic aldehyde (benzaldehyde, 2 or 4 chlorobenzaldehyde, 3 or 4 nitrobenzaldehyde) (0.01 mol), piperidine (a few drops) and ethanol (50 ml) was refluxed for 10 hr. The reaction mixture was cooled and poured into dil. HCl (100 ml, 10%). The solid product was filtered, washed with water, dried and crystallised from an appropriate solvent (cf. Table 1 ).
1 Aroyl 1,2 dibromo 1,2 dihydro 3H naphtho [2,1 b]pyran 3 ones
(4a, b)
To a stirred solution of naphthopyrone (2a, b) (0.006 mol) in carbon tetrachloride (40 ml), bromine (0.006 mol) in CCl4 (10 ml) was added dropwise during 1 hr at 25°. The reaction mixture was left to stand for 48 hr at room temperature. The solid product which separated after evaporation of CCl4 was crystallised from benzene to give 4a or 4b as yellow crystals (cf. Table 1).
1 Aroyl 5 bromo 3H naphtho [2, 1 b] pyran 3 ones (5a, b)
A solution of naphthopyrone (2a, b) (0.006 mol), bromine (0.006 mol) and acetic acid (30 ml) was heated under reflux for 5 hr. The solid product which separated after concentration and cooling was filtered and crystallised from acetic acid to give 5a or 5b as reddish crystals (cf. Table 1).
Base catalyzed addition and cycloaddition of naphthopyrones (2a, b) to p methoxybenzalacetophenone/or ethyl benzoylacrylate. General procedures (A) At 25°: Formation of(6a, b) and (7a, b).
A mixture of naphthopyrone (2a, b) (0.015 mol), p methoxy benzalacetophenone (0.01 mol) and sodium ethoxide (0.02 mol) or isobutylamine (2 ml) in abs. ethanol (30 ml) was left to stand at 25° for 3 days and then poured into HCl (50 ml, 10%). The solid product was filtered, washed with water, dried and crystallised from a suitable solvent to give 6 or 7 (cf. Table 1).
Synthesis and Insect Growth Regulating Activity of Benzoylphenylureas
Disruption of metamorphosis with the use of insect growth regulators has been recognized as an attractive target for insecticidal action because the process is vital to the development of invertebrates. The renewed interest in cuticle biochemistry, particular ly in chitin synthesis stems from the discovery of bioactive benzoylphenylureas more than a decade ago. A thorough quantitative structure activity study was recently conducted for a large number of benzoylphenylureas (BPUs), with larvicidal activity and inhibition of cuticle formation in cultured integuments used as probing parameters. A number of insect growth regulators (IGRs) have been synthesized and evaluated in the laboratory and field conditions against Dipterous insects of medical and economic importance, particularly against mosquitoes.
The specific larvicidal spectra of certain compounds in this series were suggested to be due to innate differences in the metabolic mechanism besides differences in the physicochemical substituent effects on the ultimate activity. In this paper, we report the synthesis of a structurally modified class of BPU compounds by incorporating a methylene spacer between the benzamido and anilido segments of benzoylphenylurea.
These compounds were screened for their insect growth regulating activity against three different species of mosquito vectors.
Synthesis of the compounds with the desired structures was achieved through the new method. The yield of the product in most of the cases was around 80%. All the compounds were crystalline solids with high melting points (> 200°) and limited solubility in organic solvents. The IR spectra of the compounds showed two distinct carbonyl absorptions, one around 1650 and the other between 1675 and 1720 cm-1. The absorption at 1650 cm-1 may be due to the carbonyl function flanked between methylene and amino groups. The PMR spectra showed a doublet around d4.33 corresponding to the methylene group incorporated between benzamido and anilide moieties. The details of the spectral data are given in Table 1.
The biological activity of the compounds was determined against three vector species of mosquitoes. The preliminary screening results show that N [(4 methylphenylamino) carbonylmethyl] benzamide (BGA 3), N [(phenylamino) carbonyl methyl] 3,4 methylenedioxybenzamide (BGA 4) and N [(3 trifluoromethylphenylamino) carbonylmethyl] benzamide (BGA 7) are effective against Cx. quinquefasciatus. On testing at lower concentrations the respective El50 values of the above compounds were found to be 0.4714, 0.3821 and 0.1850 mg/litre, respectively.
The treatment showed various abnormalities in the larvae, pupae and emerging adults as frequently observed in the case of diflubenzuron (a benzoylphenylurea compound) treatment. The compound BGA 11, which differs from diflubenzuron (dimilinR) by a methylene spacer between the benzamido and anilide moieties, was found to be ineffective at the preliminary screening against all the three species tested. Hence, it could be stated that the biologi cal activity of the effective parent compound was lost during the incorporation of the methylene spacer. But some structural analogues of the same class of compounds are able to show the biological activity. So, further structural modifications may result in a derivative with optimal biological activity. The insect growth regulating activity of the compounds against Cx. quinquefasciatus in comparison with dimilin is given in Table 2.
The cost analysis of the compounds in comparison to the parent compound dimilin was carried out at the laboratory scale synthesis. The analysis shows that the most effective compound (BGA 7) of the class is ten times cheaper than dimilin. Moreover, this compound could be synthesized at a single stage reaction between hippuric acid and 3 aminobenzotrifluoride, without the involvement of an acid chloride intermediate. A suitable formulation of this compound may play a role in the control of the mosquito vector Cx. quinquefasciatus and thereby the reduction in the diseases transmitted by the same.
Melting points are uncorrected. Purity of the compounds was checked by TLC using pet. ether ethyl acetate (1:1) as irrigant. PMR spectra were recorded in a mixture of CDCl3 and DMSO d6 on a Hitachi R 600 spectrometer using TMS as internal standard and IR spectra in KBr on a Perkin Elmer 783 spectrophotometer.
Experimental Procedure
Synthesis
The synthesis of BGA type of compounds with substitution in the methylene spacer was described as an intermediate in the preparation of polypeptides and proteins involving a complex mechanism by condensing four different chemical components such as carboxylic acid, isonitrile, aldehyde and amine. We have modified the pathway for the synthesis of the compounds containing unsubstituted methylene spacer with starting materials as carboxylic acid, glycine and amine, by avoiding the involvement of highly toxic isonitriles. Synthesis of the compounds was carried out by reacting the substituted benzoyl chloride with glycine in alkali. Twelve compounds were synthesized by appropriate substituting the benzoyl and aminophenyl rings. The resultant benzoylglycines were purified by extracting the reaction product with hot carbon tetrachloride to remove the free benzoic acids and recrystallising from boiling water.
The second stage of reaction involved condensation of the benzoyl glycines with an equimolar quantity of an appropriate aniline at 135 ± 5° in an oil bath for 2 hr using a Dean Stark phase separator. The product obtained was extracted with hot chloroform followed by washing with 1N hydrochloric acid (3 x 25 ml), water, saturated sodium bicarbonate solution and finally with water. The solvent was removed by distillation and the product recrystallised from a hot mixture of diethyl ether and chloroform (1:1).
Insect growth regulating (IGR) activity against mosquito vectors
Third instar larvae of Culex quinquefasciatus, Aedes aegypti and Anopheles stephensi were obtained from the insectaries maintained at the Centre and used for the evaluation of IGR activity. The compound was dissolved in acetone to get 1% stock solution from which further dilutions were made. To achieve the required concentration, 1 ml of the stock solution of appropriate concentration was added to 249 ml of tap water in 500 ml beaker and stirred vigorously to ensure thorough mixing. Twenty five larvae were added to the test solution by means of a strainer or dropper. The larvae in each beaker were provided with food (yeast powder + dog biscuit). The larvae were exposed continuously till pupation and mortality of larvae and pupae was recorded and removed regularly. Pupae from the treated water were collected in small vials, kept in netted cages and observed for normal emergence, morphological aberrations, incomplete emergence and death.
Synthesis and Pesticidal Activities of Thiadiazolo s Triazine and Imidazol
The 1,3,4 thiadiazole ring is associated with diverse biological activities probably by virtue of incorporating a toxophoricts N = C-S- linkage, the importance of which has been well stressed in many pesticides. Likewise, symtriazine and imidazole derivatives are well known for their herbicidal, insecticidal, bactericidal and fungicidal activities.
These observations coupled with our interest in the synthesis of heterocyclic compounds and the fact that compactness and planarity of a molecule might augment its other biocidal activities as it does with herbicidal activity, the biolabile s triazine and imidazole nuclei were fused with 1, 3, 4 thiadiazole nuclei respectively to yield the titled compounds (Illa g) and (Va g) with the hope of achieving compounds of better biocidal action.
The required 2 amino 5 aryloxymethyl 1, 3, 4 thiadiazoles (la g) were prepared by the method of Maffii et al. The condensation of la g with furfural in methanol furnished the compounds Ila g which were converted into the titled compounds (Illa g) by 1,4 cycloaddition of phenyl isothiocyanate in xylene. On the other hand treatment of la g with chloroacetyl chloride in cold furnished the compounds IVa g which were converted into the titled compounds (Va g) by stirring with dry pyridine at room temperature (Scheme 1).
Fungicidal activity
All the seven compounds were screened for their fungicidal activity against Aspergillus Niger and Helminthosporium oryzae at 1000, 100 and 10 ppm concentrations. Carbendazim, a commercial fungicide was also tested for comparison. Compounds lIb, c, f and g were found to be promising fungicides while Va g were nearly inactive. The activity of the compound IIlf (86% at 1000 ppm) was quite comparable with carbendazim (92% at 1000 ppm), and hence further screening of this compound on a wider range of fungi as well as at more dilution is desirable. It is to be noted that all the four compounds (IIb, c, f and g) contain a polar chloro group in the phenyl ring on 1,3,4 thiadiazole nucleus indicating that the presence of halogen atom enhances the fungi toxicity of this series of compounds.
Bactericidal activity
The antibacterial activity of all the compounds (III and V) evaluated against the bacteria Staphylococcus aureus and Bacillus subtilis at 1000, 750, and 500 mg/litre concentrations. None of these compounds showed any notable activity except the compound (Illf) which was active against both the species (82% and 80% against S. aureus and B. subtilis, respectively at 500 mg/litre concentration). This shows that association of hydrocarbon residue (CH3) with a chloro group in the phenoxy moiety could enhance the activity.
Herbicidal activity
All the compounds were subjected to primary post emergent and pre emergent herbicide evaluation at the rates of 8.0, 4.0, 2.0, 1.0 and 0.5 kg/ ha. The test species were wheat, cocklebur, sunflower, velvet leaf and sicklepod. None of the compounds except IIlf was found to be promising herbicide. The activity of Illf was found to be 100% at 8.0 kg/ha and 80% at 0.5 kg/ha. Its activity is probably due to the presence of a 2,4 dichlorophenoxy moiety on 1,3,4 thiadiazole ring.
Although, various toxophoric groups and structures have been combined in the titled molecules with a hope of achieving compounds of better biocidal potency, the results are not very encouraging. This indicates that the activity of any compound may not necessarily be related to the numerical sum of all toxophores present in the molecule.
Synthesis and Fungicidal Activity of Thiazolidine and
2 Aryl Indoles
The synthesis and biological activities of novel heterocycles such as [1,3,4] oxadiazoloquinazolones, [1, 2, 4] triazolothiazoles, [1,3,4] thiadiazoloquinazolones 1,3,4 thiadiazolo s triazines, [1,2,4] triazoloquinazolones and [l,3,4] oxadiazolyl 2 azetidinones was reported. ln continuation of our this work we wish to report herein a convenient synthesis of another fused heterocyclic system [1,3,4] oxadiazino [5,6 b] indoles and a novel spiro heterocyclic spiro [3H indole 3,2 thiazolidine] 2,2 (1H) diones. The fungicidal activity of these compounds is also reported.
The key compounds isatine b (aryloxyacetylhydrazones) (I) and isatin b (aroylhydrazones) (II) were prepared by condensation of isatine with aryloxyacetylhydrazines and aroylhydrazines, res pectively. The cycloaddition of 2 mercaptoacetic acid to I in dioxane furnished the corresponding spiro [3H indole 3,2 thiazolidines] (III). The absence of > C = N peak in the IR spectra of III suggested the conversion of I into III. On the other hand, the synthon II on refluxing with aq. KOH furnished the desired compounds, 2 aryl [1,3,4] oxadiazino [5,6 b]indoles (III) (Scheme 1).
The absence of NH and > C = O peaks in the IR spectra of IV suggested the cyclization of II to IV.
Fungicidal activity
Compounds Ila IIg and IVa lVg were screened for their antifungal activity against Aspergillus flavus and Helminthosporium oryzae at 1000, 100 ppm and 10 ppm concentrations. The results have been compared with carbendazim, a commercial fungicide, tested under similar conditions.
Compound IIa showed better antifungal activity than IV. However, the activity of II decreased considerably upon dilution. Compounds lIb and IIe were found to show much better activity than any of the compounds of this series. However, compound lIe was more potent than lIb. The activity of II was almost comparable (89% of 1000 ppm) with that of the commercial fungicide carbendazim (97% at 1000 ppm). Further investigation of this compound on wider range of fungi as well as at more dilution is in progress.
Synthesis and Biological Activity of Benzothiazoles/Benzoxazoles/Benzimidazoles/imidazolidines
In view of the fact that a large number of derivatives of thia zoles, benzoxazoles, benzothiazoles, benzimidazoles and imidazoles have been found to exhibit a wide variety of pharmacological activity and our interest in thiazole nucleus, it was considered worthwhile to synthesize compounds bearing a thiazole nucleus linked to the benzothiazole, benzoxazole, benzimidazole and imidazole nuclei through an NH linkage and evaluate their antifungal and antibacterial activities. Only a limited number of molecules where an NH group is linked to two heterocyclic moieties has been described in literature.
The synthetic route is outlined in Scheme 1. Reaction of 2 amino 4 arylthiazoles (1a d) with carbon disulphide and methyl iodide in dimethyl formamide in the presence of strong sodium hydroxide solution gave the corresponding dimethyl N (4 aryl 2 thiazolyl) dithiocarbonimidodithioates (Ila d). These compounds (Ila d) on reaction with 2 aminothiophenol in refluxing dimethylformamide in the presence of one equivalent sodium hydroxide afforded Illa d in very good yield. The high nucleophilicity of the thiolate anion, generated in the reaction, led the formation of III in good yields and in a short period. In a similar way 2 heteroaryl amino derivatives of benzoxazoles (IVa d) were prepared from o aminophenol and II under nitrogen atmosphere to avoid darkening. In the preparation of V and VI the reaction conditions depended upon the diamine used. With o phenylenediamine a 1:1 mixture of reactants was heated under reflux until no more methylmercaptan was evolved. The nature of substituents on II affected the reaction period (10 16 hr generally). In case of ethylenediamine a threefold excess of it was used and the reaction started at room temperature which was completed at 100° after 10 12 hr. Structures of all the compounds were established by elemental analysis and spectral data.
Biological activity
All the compounds were tested for their antifungal and antibacterial activities. The antifungal activity was determined against Aspergillus niger and Aspergillus flavus at 10 and 25,mg/ml concentrations using the modified Czapeck Dox nutrient medium. The antifungal data revealed that all the compounds were mode rately toxic to the fungal species. Only compounds IIa, b and IIId were comparable with commercial fungicides (Carbendazim and Capta fol).
The antibacterial activity was evaluated by the cup plate agar diffusion technique at 10 and 25 mg/ml concentrations against E. coli and S. aureus. DMF was used as control in both the cases. All the compounds tested showed only marginal inhibition but better inhibition was observed at 25 mg/ml concentration in comparison to that at 10 mg/ml.
Antimicrobial Agents
The discovery of meconazole, clotrimazole, bifonazole, etc. in the treatment of topical and systemic fungal infections has created a great impetus in imidazole derivatives as antifungal agents. Furthermore, several benzofuran and benzoxazine derivatives also exhibit marked antibacterial, antimycotic and other pharmacological activities. In view of this, it was considered of interest to synthesize some new l [(benzofuran 2 yl)(3 oxo 1,4 benzoxazin 6 yl)methyl] 1 H imidazoles (5a f) with a view to evaluating their antifungal activity in vitro.
The key starting materials chloroacetylbenzoxazinones (2) were obtained by chloroacetylation of various benzoxazinones (1) under Friedel Crafts reaction conditions. Condensation of 2 with salicylaldehyde in acetone K2CO3 afforded the benzofuranoyl benzoxazinones (3) in good yields. The ketones 3 were reduced to the corresponding carbinols (4) with sodium borohydride in methanol. The alcohols 4 thus obtained were subjected to the imidazole transfer reaction using N,N thionylimidazole to obtain the desired imidazole derivatives 5 in fair yields. Two carbinols (4a and 4d) did not undergo the imidazole transfer reaction owing to their poor solubility in dichloromethane. However, these were converted to the respective imidazole derivatives via their chlorides. Thus, the carbinols (4a and 4d) were treated with thionyl chloride to give the corresponding chlorides (6a and 6d) which in turn were treated in situ with imidazole to give the respective imidazole derivatives (5a and 5d) (Scheme 1).
The structures of 3, 4 and 5 were confirmed by their elemental analyses (Table 1) and IR and PMR spectra (see experimental).
Antifungal activity
All the title imidazole derivatives (5a e) were tested for their antifungal activity in vitro against dermatophytes (Microsporum canis, Microsporum gypseum, Epidermophyton floccosum, Trichophyton rubrum and Trichophyton mentagrophytes), yeasts (Candida Albicans, Cryptococcus neoformans) and systemic fungi (Aspergillus fumigatus and Sporotrichun schenkii) using two fold agar dilution method according to the method described earlier.
Three compounds in this series (5b, 5c and 5e) were found to possess moderate antifungal activity in vitro and the activity was mainly restricted towards dermatophytes only. Thus, 5c was found to be the most potent compound of the series. It inhibited the growth of all the dermatophytes employed in the screening at 62 mg/ml. However, 5e was active against M. gypseum, T. mentagrophytes and E. floccosum only, with MIC values of 62 mg/ml. Compound 5b was less active as compared to 5c and 5e and the fungi, M. gypseum and E. floccosum only were found to be sensitive towards this compound at 62 mg/ml. All the other dermatophytes were inhibited by 5b and 5e at 125 250 mg/ml.
In addition to the above, compound 5c was also found to possess a low order of activity against systemic fungi by inhibiting their growth at 250 mg/ml. None of the compounds reported herein displayed any significant activity against yeasts at test concentration (250 mg/ml).
Synthesis and Antimicrobial Activities of Pyrazoles
Keeping in view the biological properties associated with pyrazoles, sulphonamides and pyrazoles with sulphamides moiety attached at different positions, it was considered of interest to synthesize some new 1 substituted 5 aryl 3 methyl 4 (N substituted p sulphamylbenzene azo)pyrazoles (11 26) and 1 substituted 3 (2 hydroxyaryl) 5 phenylpyrazoles (30 41) as potential antimicrobial agents.
3 (2 ,4 Dimethoxy 3 methylphenyl) 1 methyl propane l,3 dione (1) and 3 (2 ,4 dimethoxy 6 methylphenyl) 1 methylpropane l,3 dione (2) (prepared by the Claisen condensation of 2,4 dimethoxy 3 methylacetophenone and 2,4 dimethoxy 6 methylacetophenone respectively with ethyl acetate in the presence of pulverised sodium) on coupling with diazotised solution of different sulphonamides gave 3 (2 ,4 dimethoxy 3 methylphenyl) 1 methyl 2 (N substituted p sulphamyl benzeneazo)propane 1,3 diones (3 6) and 3 (2 ,4 dimethoxy 6 methylphenyl) 1 methyl 2 (N substituted p sulphamylbenzeneazo) propane 1,3 diones (7 10), respectively. These on cyclisation with 4 methylphenylhydrazine hydrochloride and 4 pyridylhydrazide gave 1 substituted 5 (2 , 4 dimethoxy 3 methylphenyl) 3 methyl 4 (N substituted p sulphamylbenzeneazo) pyrazoles (11 18) and 1 substituted 5 (2 4 dimethoxy 6 methylphenyl) 3 methyl 4 (N substitutedp sulphamylbenzeneazo)pyrazoles (19 26) (Table 1).
Similarly, 2 benzoyloxy 4 methoxyacetophenone, 2 benzoyloxy 4 methoxy 3 methylacetophenone and 2 benzoyloxy 4,6 dimethoxyacetophenone (prepared by Schotten Baumann reaction of the corresponding acetophenones with benzoyl chloride in the presence of pyridine) on Baker Venkataraman transformation in alkaline medium gave w benzoyl 2 hydroxy 4 methoxyacetophenone (27), w benzoyl 2 hydroxy 4 methoxy 3 methylacetophenone (28) and w benzoyl 2 hydroxy 4,6 dimethoxyacetophenone (29), respectively. These b diketones on cyclocondensation with hydrazine hydrate, phenylhydrazine, 4 methylphenylhydrazine hydrochloride and 2,4 dinitrophenylhydrazine, gave the corresponding pyrazoles (30 41) (Table 1).
The structures of all the compounds were established on the basis of elemental analysis and PMR spectral data.
Biological testing
All the title compounds prepared were screened in vitro against the bacteria Escherichia coli and Pseudomonas pyocyanea using cup plate agar diffusion method at 10, 25 and 50 mg/ml concentrations. Compounds 13, 14, 17, 18, 22, 26, 30, 32, 33, 34 and 39 showed moderate activity against both the bacterial species.
These compounds were also screened for their antifungal activity against, Aspergillus niger and Aspergillus flavous using Czapeck Dox nutrient medium at 10, 25 and 50 mg/ml concentrations. Compounds 11, 17, 18, 22, 39 and 41 inhibited the growth of A. niger to the extent of 72.2, 75.3, 80.1, 71.6, 59.1 and 82.3% respectively at 50 ug/ ml concentration. Compounds 11, 17, 18, 39, 40 and 41 inhibited the growth of A. flavous to the extent of 68.4, 68.9, 75.8, 49.5, 59.3 and 88.5% respectively at 50 mg/rnl concentration. Bavistin was used as the standard which showed 88% inhibition under similar conditions.
Experimental Procedure
Melting points were taken in sulphuric acid bath and are uncorrected. PMR spectra were recorded on a Perkin Elmer R 32 (90 MHz) spectrometer using TMS as internal standard (chemical shifts in d, ppm).
3 (2, 4 Dimethoxy 3 methylphenyl) 1 methylpropane 1, 3 dione (1)
Synthesis and Insecticidal Activity of New Substituted Hydroxyacetophenones
Acetophenones have been extensively used recently in the synthesis of different classes of compounds which behave as hypolipidemic, antibacterial and antiinflammatory agents and also as passive cutaneous anaphylaxis (PCA) inhibitors and anti juvenile hormones. Some of them show photochromism. The insecticidal and other biological activities of thiophosphorylated acetophenones prompted us to synthesise more compounds in this series with the hope of getting insecticides of low toxicity and less harmful effects. We wish to report in this paper the synthesis of 3 (N,N dialkylaminomethyl) 4 hydroxyacetophenones (4 9) and their corresponding thiophosphorylated compounds. The intra molecular H bonding arising due to lone pair of nitrogen and hydroxy group of these compounds has also been discussed.
Chloromethylation of 4 hydroxyacetophenone (1) by the method of Trave gave the corresponding chloromethylated derivative (2) which was converted into different N,N substituted aminomethyl derivatives (4 9; Table 1) by refluxing with different amines in dry benzene. Compound 2 was also hydrolysed with alkali to get the 3 hydroxymethyl derivative (3). The reaction of O,O diethylthio phosphoryl chloride with 3 9 in the presence of acetone and potassium carbonate at room temperature gave the corresponding thiophosphorylated derivatives (11 16; Table 1). The structural assignments of all these compounds were based on elemental analyses, mass IR and PMR spectral data.
In 4 hydroxyacetophenone (1) the carbonyl group is known to appear at a lower frequency! (1635). It was observed that when a chloromethyl or hydroxymethyl side chain was introduced at 3 position, the carbonyl band shifted to 1655 and 1663 respectively. Similarly in 3 (N, N dialkylaminomethyl) derivatives (4 10) the carbonyl bands appeared at a higher frequency compared to that in 1 (1640). In hydroxy frequency region the bands were observed for both free (3600 3660) and intramolecular bonded (3375 3450) hydroxyl groups in the case of 4, 7 and 9; however compounds 5,6 and 8 exhibited a band for free OH at 3450 3398.
From a study of the carbonyl band frequency of o hydroxyacetophenone and o methoxyacetophenone Hergert and Kurth arrived at a conclusion that the increased stability of the dipolar form of o hydroxy acetophenone and the consequent lowering in vCO band depend upon the greater ability of the hydroxyl group to donate electrons to the ring. Since the carbonyl bands of 2, 3 and 4 9 appeared at different frequencies, it can be expected that the carbonyl group must have different degrees of s orbital character of the C O s bond. This in turn gives different force constants of the carbonyl group and thus changes the frequency of nCO.
It is also known that the appearance of nCO at a lower frequency in 4 hydroxyacetophenone compared to acetophenone is attributed to the conjugation of C = O group with unsaturated chromophore11 and stabilization of the consequent dipolar form (1a) 10.
Therefore, it is the capacity of the O Z(Z = H, CH3 etc) group to donate electrons to the ring that finally determines the participation of the carbonyl group in conjugation and consequently the degree of s orbital character in the above compounds. Thus, it is clear from the above discussion that the capacity of the hydroxyl group in 2 9 to donate electron to the ring is affected.
The formation of intramolecular hydrogen bond in 2 hydroxybenzyl alcohol is well documented12. The formation of intramolecular hydrogen bond in o halophenols was investigated by Pauling and other workers and it was observed that a strong repulsion between the proton donor and acceptor exists. And from the examination of atomic population and bond order of this class of compounds it was observed that a major part of the electron density shifts from the phenolic proton to the proton accepting halogen. Dietrich has also suggested that in o halogenophenols the intramolecular interactions are mainly determined by H halogen (X) interaction in cis conformer and O X repulsion in cis and trans conformers and the O H X angle. Baker and Shulgin have shown that steric interactions become important in 2 dimethylaminophenol where the methyl groups force the nitrogen lone pair orbital towards the hydroxyl group. Rao have observed that in pyridine methanol system a slight increase in the electron density of the nitrogen atom occurs due to hydrogen bond formation.
The IR spectra of 2 9 indicated that the 4 OH group forms intramolecular hydrogen bonding with different hydrogen acceptors, i.e. chlorine in 2, oxygen in 3 and nitrogen in 4 9. It is expected that depending upon the steric factors. O-H X angle and basicity of the hydrogen acceptor, the degree of intramolecular hydrogen bond changes which in turn affects the acidity of the phenolic O H group and consequently its capacity to donate electron to the ring.
4 Hydroxy 3 (N,N diphenylaminomethyl)aceto phenone (9) exhibited two intense bands at 3600 and 3375 due to free and intramolecular hydrogen bonded OH respectively. From the intensity of the free hydroxyl bond it can be assumed that due to the two bulky phenyl rings, the less bulky free electron pair orbital points in such a way that it is not available as an acceptor in hydrogen bonding. Therefore, it is assumed that 9 exists as an equilibrium mixture of 9a, 9b. 9c and 9d. Similarly compounds 4 8 also exist as equilibrium mixtures of forms a d.
In both 6 and 9 the carbonyl band appears near the VC-O frequency of 4 hydroxyacetophenone (l), i.e. at a lower frequency (1652 and 1650 respectively) compared to that in 4, 5, 7 and 8, because the strong electron donating (towards the ring) capacity of O-H group is in these two cases are not affected much by small intramolecular hydrogen bonding.
The formation of six membered ring B through hydrogen bond may cause certain rigidity in the molecules (4 9). From the PMR spectra (Table 2) it can be observed that in compounds 4 8 the H 2 doublet appears at a higher field than the doublet of doublets (dd) for H 6, but the trend reverses in compounds 10 14 where the intramolecular hydrogen bonding is removed by methylation or thiophosphorylation of the 4 OH group. Therefore, the upfield shift of H 2 proton must arise from the steric effect of either the benzylic methylene protons or from the groups attached to nitrogen atoms. When Dreiding models for 4, 5, 6 and 9 were constructed, it was found that both the rings A and B remained almost on the same plane. The two alkyl groups and the two benzylic protons remained one each above and below the plane of the rings A and B. There was a probability that the b alkyI group might come over the H 2 proton and cause steric hindrance resulting in the upfield shift of the latter. Therefore we examined the model for a rigid system in 7 [N(R)=piperidino] and found that both rings A and B were almost planar and the b equiterial proton of a carbon atom to nitrogen could not exert any influence on the H 2 proton. Thus, the steric hindrance on H 2 and the resulting upfield shift of its PMR signal must arise from the two benzylic methylene protons with which it is making an angle of about 60°.
To confirm our above views we examined the PMR spectra of 6 acetyl 2 methoxy 4H 1,3,2 benzodioxaphosphorine 2 sulphide (17) and 2 methoxy 6 nitro 4H 1,3,2 benzodioxapho sphorin 2 sulphide (18) and found that there also the H 2 proton doublet appeared at a higher field compared to doublet of doublets for 6 H. On examining the model for 17 or 18 it was found that in these models the ring A and the two oxygen atoms of ring B remained on the same plane and the H 2 proton was making an angle of about 60° with the two benzylic methylene protons as in 4 9.
From the structural point of view, the compounds 2 9 have a sp3 carbon. atom between the aromatic ring and the proton acceptor X to minimise the delocalization between OH and X and does not make much impact on the dipole moment.
The 3 chloromethyl and 3 hydroxymethyl compounds (2 and 3) were found to be insoluble in CDCl3 and their PMR were recorded in acetone d6. It was observed that unlike in 4 8, the H 2 proton doublet in 2 and 3 appeared at a lower field compared to H 6 proton in the PMR spectrum. Since 2 and 3 also form hydrogen bonding with chlorine or oxygen atom of 3 substituted side chain, it was expected that H 2 proton would appear at a higher field compared to H 6 as in 4 8. To examine the solvent effect of acetone d6, we recorded the spectrum of 4 in this solvent and found that H 2 proton appeared at a relatively higher field compared to H 6 (see Table 2). Further, there was no difference in the pattern of the PMR spectrum from that recorded in CDCl3. Although it is premature at this stage to predict, without knowing in detail about the intramolecular hydrogen bonding in compounds 2 4 and the solvent effect of acetone d6 on them, we would like to suggest that the downfield shift of H 2 proton in 2 and 3 might arise from the electronegative chlorine or oxygen atom of the side chain.
Insecticidal activity
All the compounds (4 16) were screened for their insecticidal activity against Drosophila melanogaster (vinegar fly) by residue film evaporation method.at 1%, 0.5% and 0.2% concentrations. The mortality was observed after 24 hr. The screening results, recorded in Table 3, show that almost all the thiophosphorylated compounds (11 16) have significant insecticidal activity. Rest of the compounds has either poor or no activity.
Further, the screening results clearly indicate that the activity of compounds 11 16 lies only in the organophosphorus part of the molecules. Since, the 4 methoxy derivative (10) did not exhibit any appreciable activity, it was clear that the 4 OH group as such or the hydrogen bonding caused by it did not contribute to the insecticidal properties of the compounds. Similarly, the nitrogen part of the molecule was also proved ineffective because compound 15 showed the same activity as 11 14 and 16.
Syntheses of Sulfanilyl Derivatives
The present study forms part of our program on the chemistry of aromatic sulfonyl derivatives and their evaluation as candidate pesticides. A survey of the literature reveals that not much work has been reported on N4 (alkyl or aryl sulfonyl) sulfanilyl derivatives.
Reaction of o dichlorobenzene with excess chlorosulfonic acid in boiling chloroform, gave the 3,4 dichlorobenzenesuIfonyl chloride in 57% yield as compared with 81% reported earlier. Similarly m and p dichlorobenzene on reaction with excess of chlorosulfonic acid with or without the solvent afforded 2,4 and 2,5 dichloro benzene sulfonyl chlorides respectively in very good yields (80 90%). These dichlorobenzenesulfonyl chlorides, on condensation with excess aniline, gave the corresponding dichloro benzenesulfonanilides. Further reaction of these anilides with warm excess chlorosulfonic acid under the conditions employed for the successful chlorosulfonation of the analogous, dichlorobenzoic acid anilides caused N S bond cleavage to give the corresponding dichlorobenzenesulfonyl chlorides. However, when the reaction was carried out at lower temperature (10° to 10°), good yields (65 85%) of the dichlorobenzenesulfonyl sulfanilyl chlorides (1, 8. 19) (Scheme 1) were obtained (cf. ref. 5). With sulfonylanilides, N S bond cleavage was observed when chlorosulfonation was carried out at temperatures above 10°, although similar chlorosulfonation of carboxylic acid anilides occurred at 50 60° without appreciable
N S bond cleavage.
The difference in behaviour is presumably largely a reflection of the greater strength of the C N bond (184kcal/mol) as compared with the N S bond (111 kcal/mol). N4 (2,4 Dichlorobenzene sulfonyl)sulfanilyl chloride (1) was condensed with nucleophiles such as dimethylamine and hydrazine under standard conditions to give the derivatives (2,3). The hydrazide (3) was treated with acetone and benzaldehyde to afford the hydrazones (4, 5). The hydrazides (3, 11, 21) on refluxing with acetylacetone in ethanol for 3 hr furnished the 3, 5 dimethylpyrazole (6). 1 on reaction with sodium azide gave the azide (7), which on treatment with triethyl phosphite (Imol) in toluene at 0° furnished the phosphinimine (8). The 2, 5 dichlorosulfanilyl chloride (9) was similarly reacted with nucleophilic reagents to give the derivatives (10 16,18); the aziridine (17) was obtained by reaction of the azide (15) with norbornene(1 mol)in toluene for 6 hr (cf. ref. 8b). Analogous reactions of 3, 4 dichlorosulfanilyl chloride (19) gave the derivatives (20 25).
Attempts to prepare N4 acetamidobenzene sulfonylsulfanilyl derivatives (Table 1) via the chlorosulfonation of N4 acetylsulfanilyl anilide by treatment with excess chlorosulfonic acid failed. However, reaction of N4 acetylsulfanilyl chloride with the dimethylamine, morpholine and sodium azide gave the N4 acetylsulfanilyl derivatives which were deacetylated with hydrochloric .acid to give the corresponding sulfanilyl compounds. Thus N4 acetylsulfanilyl dimethylamide was deacetylated to N,N dimethylsulfanilylamide (50%), PMR(DMSO d6): d 7.20 6.56 (m, 4ArH), 5.90 (s, 2H, NH2, exchangeable with D2O), 2.05 [s, 6H, N(CH3)2]; N4 acetylsulfanilyl morpholidate to the morpholidate12 (58%), PMR (DMSO d6): d 7.40 6.60(m, 4ArH), 6.0 (s, 2H, NH2 exchangable with D2O), 3.6 2.8 (m, 8H, morpholino H) and N4 acetylsulfanilyl azide to the azide (61%). These were condensed with N4 acetylsulfanilyl chloride to afford the required products (26,27,30) (Table 1). The condensation could be carried out either in the presence of triethylamine in acetonitrile or sodium hydrogen carbonate in acetone; the latter system afforded a higher yield of a purer product. The azide (30) was obtained by the condensation of N4 acetylsulfanilyl chloride with sulfanilyl azide (1 mol) in the presence of sodium bicarbonate (1 mol) in aq. acetone for 3 hr. This, on reaction with triethylphosphite (1 mol) in toluene at 80° for 1 hr gave (31) and on treatment with norbornene (1 mol) in boiling THF for 6 hr furnished the aziridine(32). Acetone and benzaldehyde sulfanilyl hydrazones reacted with N4 acetylsulfanilyl chloride in pyridine to give low yields of the required products (28, 29).
Methanesulfonanilide on treatment with excess chloro sulfonic acid at 10 to 10 gave N4 (methanesulfonyl)sulfanilyl chloride (Table 1, 33) which was characterized as the derivatives (34 39). Reaction with aniline afforded N4 (methanesulfonyl) sulfanilylanilide, which like the corresponding N4 acetyl derivative, failed to form the sulfonyl derivatives with chlorosulfonic acid, under conditions successfully used for the dichlorobenzene and methyl sulfonanilides (Table 1). However, benzenesulfonanilide with excess chlorosulfonic acid at 10° to 0° gave the corresponding sulfonyl chloride (54%).
In the PMR spectra of the dichloro and acetamido benzene sulfonylsulfanilyl compounds (Table 1), the lowest field proton resonance is assigned to the SO2NH group due to the combined anisotropic effect of the two attached benzene rings, e.g. the acetone hydrazones (4,28). In the methane sulfonyl series (Table 1), the methylsulfamoyl proton appears at the lowest field and the SO2NH N proton resonance is merged with the aromatic proton resonances, e.g. compound (37), which is confirmed by the disappearance of part of the aromatic multiplet after treatment with deuterium oxide. The mass spectra of the compounds generally showed the molecular ions (M+), and fragment ions corresponding to M X, M SO2X, and M NHC6H4SO2X. In the acetamido series (Table 1) cleagage of the acetamido group was also observed.
Biological activity
Antibacterial screening was carried out by innoculation of agar plates as described by Steers et al. and the results were compared with penicillin as standard (100% control). The in vitro antifungal screening was performed using the standard glass slide spore germination test as described by Kirby and Frick.
In the preliminary in vitro antibacterial screening tests against Streptococcus faecalis, Clostridium perfringens, and Staphylococcus aureus at 50ppm, compounds 5, 10, 12, 13, 14, 15, 20 showed complete inhibition of the bacteria. The compounds were also screened against Botrytis cinerea in vitro at 50ppm, high activity was observed for compounds 4,12, 26, 36. Compounds were examined against Rhizoctonia solani and Pythium aphanodermatum in vivo by soil incorporation at 16 kg/ha, the best antifungal activity was shown by the azides (7, 24, 30).
Qsar of Fluridone
At least eight chemically different classes of herbicidal inhibitors have been demonstrated to affect phytoene desaturase as their essential mode of action. At the moment, qualitative and/or quantitative structure activity correlations (QSAR) are available only for five groups, namely phenoxybenzamides, phenylpyrida zinones phenylfuranones and phenoxynicotinamides. For a better understanding of the essential structural elements of a phytoene desaturase inhibitor and favorable substitution patterns more structure activity investigations with chemically different inhibitors are needed.
The phenylpyridinone, fluridone (1 methyl 3 phenyl 5 (3 (trifluoromethyl)phenyl (1H) pyridinone), belongs to the same type of bleaching herbicides as indicated above. In a preceding publication it has been demonstrated that this compound directly interacts with phytoene desaturase as a reversible noncompetitive inhibitor. In the present study our attempts are continued to analyze the structural requirements of typical phytoene desaturase inhibitors. Accordingly, a range of fluridone derivatives have been investigated for their inhibitory activity either in vivo or in vitro. Furthermore, quantitative structure activity relations have been calculated for fluridone analogs substituted at position 3 of the pyridinone ring.
MATERIALS AND METHODS
Aphanocapsa (= Synechococcus PCC 6714) was cultivated for 48 hr. Freshly harvested algal cells grown in the presence of bleaching compounds were directly used for carotenoid extraction or membrane preparation. Fusarium SG 4 mycelium grown for 5 days as described was freeze dried and stored at 15°C. Carotenoids were extracted with hot methanol (20 min, 65°C) and partitioned into 10% (v/v) diethyl ether/petrol. Total carotenoids were quantitated from this extract by their absorbance at 445 nm and related to dry weight of cells. I50 values for intact cells were determined by a Dixon type plot with 5 to 7 concentrations around the l50 value.
For in vitro carotenogenesis, Aphano capsa cells were resuspended in 0.1 M tris(hydroxymethyl) aminomethane (Tris) HCl buffer, pH 8.0, containing 5 mM dithiothreitol (DTT), and broken in a French press at 500 bar.
After centrifugation (12,000g, 15 min), the membrane pellet was resuspended in the same buffer. Fusarium SG 4 extracts, which form [14C]geranylgeranyl pyrophosphate from [14C]mevalonic acid, were obtained by suspending 0.1 g of powdered material in 0.8 ml of 0.4 M Tris HCl buffer, pH 8.0, containing 5 mM DTT, and centrifugation (10.000g, 10 min). The incubation mixture contained 0.2 ml of this supernatant, 0.1 ml of Aphanocapsa (thylakoid) mem branes equivalent to 0.15 mg of chlorophyll, ATP (5 mol), NAD+ (1 mmol), Mn2+ (3 mmol), and Mg2+ (2 mmol) made up to a total volume of 0.5 ml with water. The reaction was terminated after 2 hr by addition of 2.5 ml of methanol containing 6% KOH. After saponification for 20 min at 65°C, the carotenoids were partitioned into 10% (v/v) diethyl ether/petrol and separated by HPLC. The system used was a 25 cm Spherisorb ODS 1 5 mm column with iso cratic elution with acetonitrile/methanol/ propanol (85:10:5; v/v) at a flow rate of 1 ml/min. Radioactivity of the elution peaks was determined by a radioactive flow detector Ramona LS (Raytest, Strauben hardt, Germany). Inhibition ratios (IR) for incubations with a certain inhibitor concentration were calculated from the ratio of radioactivity accumulated in phytoene versus radioactivity in b carotene, related to the corresponding ratio of an untreated control (4 3). In a previous publication this value has been verified as a suitable in vitro parameter for QSAR. i50 and IR values are the means of three to five experiments. The standard deviation was in the range of 10%.
Details on the phenylpyridones employed including the procedure of their synthesis are given elsewhere. Quantitative structure activity analysis was carried out with R.A. Coburn s multiple linear regression program, QSAR PC, from Biosoft. Physicochemical parameters (Hansch Fujita hydrophobicity constants p and Hammett para constants sp) were provided by the data base of this program, except for the C6H4F p substituent which were estimated by comparison with the data for similar substituents. The Hammett para constants are derived from pka values of substituted benzoic acid.
RESULTS AND DISCUSSION
As it has been established, that most bleaching herbicides target the same enzyme reaction of questions about common structural elements in the bleaching molecules and about their interaction with the same or a similar inhibition site may be brought closer to an answer by modification of their structures and comparing their inhibitory properties. In case of phenylpyridinones two sets of derivatives could be investigated. For all compounds of Table 1, I50 values of bleaching activity have been determined from decreased contents of colored carotenoids in the cells. The most active compound was fluridone (no. 1). Replacement of the CF3 group by less lipophilic Cl (no. 2) moderately decreased its activity and replacement by COOH (no. 8) decreased it strongly. Inhibitory activity disappeared completely when one of the phenyl rings was missing and the aromatic nature of the pyrimidinone ring was lost simultaneously (nos. 10, 11). Replacement of one of the phenyl rings by phenoxy (no. 6) resulted in lower but substantial inhibition. Modification of the fluridone molecule from a N methyl to its N ethyl derivative (no. 5) was also accompanied by a negative effect on its bleaching activity. When the pyridinone ring of fluridone was completely saturated, the resulting piperidinone (no. 4) still showed reasonable activity with an I50 value of about 25 mm. The structure of this compound allows for a keto enol equilibrium. In this case, a charge separation between the heterocyclic nitrogen (positive) and the oxygen (negative) can be imagined, which might exert an additional effect on the inhibitory activity.
From the results of Table 1 it can be concluded that the aromatic nature of the pyridinone structure influences the level of activity but is not essential for the fluridone derivatives to interact as inhibitors of phytoene desaturase. Apparently, both phenyl rings at positions 3 and 5 are favorable. It has recently been reported for phenyltet rahydropyrimidines, which contain a non aromatic ring structure similar to a fluridone related piperidinone, that at least one of the two phenyl rings is essential for activity. As an N ethyl group instead of N methyl strongly affects the inhibitory activity in a negative way, the bulkiness of certain substituents and the overall shape of the substituted pyridinone moiety seems to be an important feature to influence herbicidal activity.
With respect to the substitution of the keto group at position 4 by other heteroatoms, a replacement by OCH3 increased the I50 value of the resulting pyridylium cation (no. 7) by a factor of about 100. A similar decrease of activity was observed for N (CH3)2 (no. 3). A thione group (no. 9) or even Cl (no. 12) led to very weak inhibition or to a compound which had lost all its inhibitory activity. These results are difficult to interpret. However, there seems to be a tendency of decreasing inhibitory activity depending on how strongly electrons are withdrawn from the heterocyclic nitrogen. Furthermore, electron density on the various substitutents at position 4 could exhibit an additional influence on activity.
All compounds in Tables 2 and 3 differ by a variation of the phenyl substituent at position 3 of the pyridinone ring. For this series of derivatives, I50 values with intact Aphanocapsa cells and the in vitro IR have been determined. These values have been compared to physicochemical parameters for the corres ponding substituents and lipophilicity p and their electronic properties sp were found to be important.
Only one compound, the C6H4F p analog, showed slightly better inhibitory activity than fluridone. All other substituents were less effective (Table 2). The dominant factor which determines their inhibitory properties is lipophilicity. A plot of pI50 versus it gave a good linear relationship with a correlation of 0.99 (Fig. 1A). However, the OC6H5 analog did not fit into this relationship. This might be due to the bulkiness of this group which has e.g., a B2 STERIMOL constant of 3.11 as compared to 1.71 for a phenyl group.
In a multiple regression analysis, lipophilicity p and the electronic parameter sp as independent variables gave a significant contribution to explain pl50. The resulting QSAR equation is given in Table 4A. The validity of the regression was confirmed by an F test and the significant contribution of both independent variables by a t test. Other parameters of the substituents did not improve this calculation.
In order to elucidate the relevance of the results from this QSAR analysis for the interaction of the pyridinone inhibitors with their target site, the phytoene desaturase, in vitro inhibition of the compounds of Table 2 was determined. For this purpose, the 10 compounds had to be divided into high and low activity groups. Then, the inhibition ratios have been measured for fixed concentrations of 0.3 and 10 mM, respectively (Table 3). Again, a dependency of inhibitory properties on lipophilicity could be demonstrated. The log IR from the first group gave a very good (r = 0.98) linear correlation with lipophilicity p (Fig. 1B). As shown for in vivo inhibition (pI50 values of Fig. 1A), the OC6H5 analog is an outlier and does not fit into this correlation.
Interference of Fluridone
Fluridone was one of the first developed bleaching herbicides and has been successfully used during the last 15 years for weed control in cotton. Its primary mode of action is inhibition of carotene interconversion at the level of phytoene. This colorless carotene is accumulated together with phytofluene in leaves after fluridone treatment, and formation of b carotene is impaired. As a consequence of the decreased levels of colored carotenoids, chlorophyll is destroyed and disruption of the chloroplast structure is observed in the light.
Accumulation of phytoene in the presence of fluridone indicates inhibition of the desaturation reaction. Direct interaction with phytoene desaturase at the enzyme level has been reported for several other bleaching herbicides and also for fluridone. It is a goal to find common features as well as differences of the known bleaching herbicides that interfere with phytoene desaturase. Helpful information may be obtained by comparing their binding sites, looking at their types of inhibition, and performing structure activity investigations with chemically modified compounds. Increasing our knowledge of inhibitor enzyme interaction may facilitate rational design of new potent phytoene desaturase inhibitors. In this context, the present publication concentrates on an enzyme kinetic study of fluridone inhibition of phytoene desaturase from Aphanocapsa. The results obtained allow the type of inhibition exhibited by this herbicide to be determined. Furthermore, it was demonstrated that Aphanocapsa membranes represent a good assay system for in vitro ki determinations of different substituted fluridone derivatives with a broad range of inhibitory activity.
MATERIALS AND METHODS
Aphanocapsa 6714 (= Synechocystis PCC 6714) and Fusarium SG 4 were cultivated for 48 hr and 5 days, respectively, as described. Carotenoids were extracted from harvested cells by suspending them in methanol containing 6% KOH and then heating them for 20 min at 60°C. This step and subsequent ones were all carried out in very dim light and under nitrogen, if possible. Carotenoids were partitioned into 10% (v/v) diethyl ether in petrol (bp 35 80°C). Total carotenoids were determined from this extract by absorbance at 445 nm and calculated with an average extinction coefficient of 2500. Subsequently, the carotenoid extracts were evaporated to dryness, resuspended in acetone, and sub jected to HPLC separation. The reversed phase HPLC system employed a Spherisorb ODS 1 5 mm column 25 mm in length and an isocratic solvent system of acetonitrile/methanol/2 propanol 85/10/5 (v/v/v) (7) with a flow rate of 1 ml/min. Carotenoids were identified with appropriate standards. Chlorophyll was extracted with hot methanol from cells in a manner similar to that used for carotenoids. Then, chlorophyll was quantitated spectrophotometri cally according to Mackinney.
In vitro carotenogenesis was carried out with membranes from Aphanocapsa prepared by French press treatment (500 bar) of cells suspended in 0.1 M Tris(hydroxy methyl)aminomethane (Tris) HCI buffer, pH 8.0, containing 5 mM dithiothreitol (DTT). Membranes were collected by centrifugation (12,000g, 15 min) and resuspended in the same buffer to a final chlorophyll concentration of 1.5 mg/ml. [14C] Geranylgeranyl pyrophosphate, the substrate for the Aphanocapsa membranes, was generated from [14C] mevalonic acid by a preparation from the fungal mutant Fusarium SG 4. Two hundred milligrams of powdered mycelium was mixed with 1.6 ml of 0.4 M Tris HCl buffer, pH 8.0, containing 5 mM DTT, and centrifuged for 10 min at 10.000g. Together with the Aphanocapsa membrane suspension (0.15 ml), 0.1 ml of the supernatant (3 mg protein/ml) was used in the incubation mixture which additionally contained K [2 l4C] mevalonic acid (0.5 mCi), ATP (5 mol), NAD+ (1 mmol), Mg2 + (2 mmol), and Mn2+ (3 mmol) in a total volume of 0.5 ml. Herbicides were applied in 5 ml methanol. The kinetics of Fig. 2 show that, in a first step, [14C]geranylgeranyl pyrophosphate was formed by preincubation of Fusarium SG 4 extracts with [14C]mevalonic acid and all cofactors present for 2 h. Varied amounts of the resulting crude geranylgeranyl pyrophosphate solution were used as substrate for the subsequent phytoene desaturase reaction with Aphanocapsa membranes. The reaction was terminated after 2 h by addition of 2.5 ml of methanol and 0.3 ml of 60% KOH solution. Heating, partitioning, and HPLC separation were done as described above for unlabeled carotenoids. The radioactivity in the phytoene and b carotene peaks was de termined on line with the radioactivity monitor Ramona LS (Raytest, Straubenhardt, Germany). As the Aphanocapsa membranes catalyze a reaction sequence from geranylgeranyl pyrophosphate to b carotene with phytoene as the only detectable intermediate, the sum of radioactivity in phytoene + b carotene accumulated over the total reaction period was used as the substrate concentration value and the radioactivity in b carotene as the product value. Binding of [14C] fluridone (6.68 mCi/mmol) was carried out by incubation of Aphanocapsa membranes in 0.1 M Tris HCl buffer, pH 8.0, with 0.1 nmol of radioactive herbicide in a total volume of 1 ml for 1 h. After centrifugation (10,000g, 10 min), residual radio activity was determined from the supernatant. Then the pellet was resuspended in 0.5 ml of the same buffer, and centrifugation was repeated.
The data in Tables 1 and 2 are means from three experiments; the standard deviation is ± 10%. In Figs. 1 to 3 data of typical experiments are shown. The carotenogenic activity of the membranes used for them may vary from one preparation to another. However, the overall picture and the biochemical information is always the same. The km values calculated from Fig. 2 and the ki values from Fig. 3 are very reproducible with standard deviations of less than 10% from three to five determinations. The herbicidal compounds used in this study are fluridone ( = EL 171; 1 methyl 3 phenyl 5 (3 (trifluoromethyl)phenyl) 4(1H) pyridinone), compound A (1 methyl 3 (3 (ethylacetyl)phenyl) 5 (3 (trifluoromethyl) phenyl) 4(1H) pyridinone), and compound B (1 methyl 3 phenyl 2,3 dehydropiperidi none.
RESULTS
Application of the herbicide fluridone to cultures of the cyanobacterium Aphanocapsa resulted in a decrease of all the major colored carotenoids present in this organism. Depending on the fluridone concentration of up to 1 mM, the acyclic glycoside myxoxanthophyll, b carotene and its hydroxylated and keto derivatives zeaxanthin and echinenone, all decreased more or less to the same extent (Table 1). With the highest herbicide concentration used, only 20% of the colored carotenoids of the control were retained. Simultaneous to the decrease of total colored carotenoids, phytoene, found at a very low level in the control, accumulated. At least at low concentrations of fluridone (0.1 mm), which only moderately affect colored carotenoids, the sum of colored carotenoids plus phytoene is similar to the control value. From the values of total colored carotenoids, a ki, value in the range of 0.2 mM was estimated. In addition to the decrease in carotenoids, lower levels of chlorophyll were observed in the presence of various concentrations of fluridone.
Direct interference of fluridone with phytoene desaturase was demonstrated in the in vitro experiment of Fig. 1. [l4C]Gera nylgeranyl pyrophosphate is converted by Aphanocapsa membranes and the labeled products of the carotenogenic pathway detected were phytoene and b carotene exclusively. The results are in parallel with the in vivo data. With increasing concentrations of fluridone less radioactivity was found in b carotene and more radioactivity was accumulated in phytoene. The nature of this inhibition of phytoene desaturase by fluridone was analyzed by enzyme kinetic studies and is presented in a Lineweaver Burk plot (Fig. 2). The substrate was varied and the radioactivity in the reaction product was determined in three sets of experiments: one as a herbicide free control, the two others with different concentra tions of fluridone. For all of them, straight lines were obtained in this double reciprocal plot with a common intersection at the abscissa which corresponds to noncompetitive nature for fluridone interaction. It is difficult to determine a km value for a heterogenous enzyme reaction. In our case, a suspension of membranes, in which the desaturation takes place and in which phytoene is generated, is employed. As the site of substrate conversion and product formation is exclusively in the Aphanocapsa membranes, the packed volume of the membranes per incubation was determined as 12 ml. Based on this value, a km value of 0.3mM was calculated for phyotene.
To find out whether fluridone binding is reversible, binding and replacement experiments were carried out with 14C labeled herbicide (Table 2). Phytoene desaturase is located in the thylakoid membranes. Therefore, an amount of fluridone with a radioactivity of about 1500 dpm was added to various amounts of thylakoid membranes that were quantitated by their chlorophyll content. Depending on the quantity of thylakoids, up to 30% of fluridone was bound. Reversibility of the binding was shown by washing the radioactive fluridone from the membranes. One wash released about 85% of the radioactive fluridone and a second wash released the rest of the bound fluridone.
For the noncompetitive and reversibly bound fluridone and two other pyridinone derivatives, ki values have been determined by a Dixon plot of inhibitor concentration versus inverse product formation at a constant substrate concentration (Fig. 3). The intersections of the straight lines with the abscissa gave a ki value of 0.08 mM for fluridone and 0.4 mM for compound A which carries an ethylacetyl group at position 3 instead of the unsubstituted phenyl ring. In the case of compound B in which both the 3 phenyl ring and the 3,4 double bond are lost, inhibitory activity is ex tremely low with a ki value of about 100 mM.
Absorption and Metabolism of Clomazone
Clomazone is a selective biopesticide used in soybean to control certain grass and broadleaved weeds. Clomazone is thought to inhibit an early step in the synthesis of terpenoids resulting in the absence of chlorophyll, carotenoids, and other terpenoids. Specifically, clomazone has been observed to inhibit either isopentenyl pyrophosphate isomerase or prenyl transferase, resulting in failure to produce plastidic terpenoids. However, clomazone inhibition of isopentenyl isomerase and prenyl transferase has not been observed in a different study. Past research has shown that both tolerant and susceptible plant species have the capacity to metabolize clomazone. Therefore, clomazone detoxi cation does not appear to account for soybean tolerance to the herbicide. Selectivity to clomazone has been speculated to involve either differential bioactivation of clomazone by sensitive species or differences of clomazone sensitivity at the site of action.
Although metabolism of clomazone has been studied in plants and plant cells, characterization of the metabolites produced is lacking. In seedlings of both soybean (tolerant) and velvetleaf (susceptible), we have demonstrated that clomazone metabolism occurs by oxidative cleavage followed by the benzyl moiety conjugating with glucose.
Cell cultures have been used to evaluate absorption and metabolism of herbicides in plant cells, and results have usually correlated well with observations made in intact plants. Advantages of cell cultures for this research were the uniform exposure of the cells to herbicide, absence of translocation, the relative high capacity of the cells to produce clomazone metabolites for iden tification purposes, and the possible absence of clomazone phytotoxicity in heterotrophic cells. The objectives of this study were to characterize the absorption and metabolism of clomazone by suspension cultured cells of soybean and velvetleaf and to characterize the major metabolites produced.
MATERIALS AND METHODS
Chemicals. Analytical clomazone (99.0% purity), [14C]CAR C1 (1.04 GBq/mmol; 98% purity), [14C]MET C (1.15 GBq/mmol; 98% purity), 5 OH clomazone, and 5 keto clomazone were provided by FMC Corp. 2 CBA and b glucosidase were purchased from Sigma Chemical Co.
Cell culture. Cell suspension cultures of soybean (cv. Corsoy 79) and velvetleaf were established and cultured in a modified Gamborg B5 medium as described previously (8). Medium consisted of Gamborg B5 major and minor salts supplemented with 0.3, 2.0, 0.6, and 100 mg liter-1 of kinetin, IAA (indole 3 acetic acid), picloram (4 amino 3 ,5 ,6 trichIoro 2 pyridinecarboxcylic acid), and myoinositol, respectively, and 25 g liter-1 sucrose. Suspension cells, established from callus produced from hypocotyl sections of each species, were subcultured every 14 days and grown on a rotary shaker at 25°C and 125 rpm.
Clomazone absorption and metabolism. Past research had indicated that clomazone metabolism may occur, in part, by oxidative cleavage. Therefore, we utilized two 14C labels of clomazone to follow each half of the clomazone molecule. MET C and
CAR C labeled the benzyl and hetero cyclic moieties of the clomazone molecule, respectively. Soyabean and velvetleaf cells (10 12 days after subculturing) were treated with either 1 mM MET C or 1 mM CAR C. Preliminary experiments indicated that ex posure to 1 (M clomazone for 14 days had no effect on the growth of either soybean or velvetleaf heterotrophic cells. Aliquots of 20 ml (containing 0.8 1.2 g fresh wt of cells) were taken at 1, 3, 6, 12, 24, and 48 hr after treatment. Cells were collected by vacuum filtration on glass fiber filter discs (Whatman GF/A, VWR Scientific, Chicago, IL) and immediately rinsed with 10 ml of ice cold incubation medium containing 1 mM unlabeled clomazone to remove unab sorbed [MC]clomazone. Cells were collected again by vacuum filtration. Cell fresh weights were determined and 4 ml 100% methanol was added. Cells were stored at 20°C prior to homogenization. Ten milliliters of 80% methanol was added and cells were homogenized using a Ten Broeck homogenizer.
Homogenized cells were prepared for chromatographic analysis as described previously. The homogenate was filtered through a glass fiber filter. The filtrate was concentrated by evaporation under a stream of N2 gas at room temperature and filtered through a 0.2 mm pore fluoropolymer membrane (Arco LC13, Gelman Sci ences, Inc., Ann Arbor, MI). Radiolabel remaining in the aqueous, suspension cell uptake medium (30 ml) was partitioned three times against 20 ml CH2Cl2. The CH2Cl2 phases were combined, a subsample was taken from both the aqueous and the CH2Cl2 phases, and the radioactivity present was determined by LSS. The remaining CH2Cl2 phase was evaporated to dryness in vacuo at 30°C and precipitates were dissolved in 1.5 ml methanol prior to chromatographic analysis.
Radiolabel remaining in the cellular debris was determined by oxidation and collection of I4CO2 Radioactivity collected was quantified by LSS. In all cases greater than 85% of the applied radioactivity was recovered.
Metabolite characterization. Cells were homogenized and [14C]extracted as described above. Methanol was removed in vacuo at 35°C and the aqueous concentrate (adjusted to 20 ml H2O) was partitioned three times with 15 ml CH2Cl2 (nonconjugate fraction). The three CH2CI2 phases were combined for I4C analysis. The aqueous phase (Polar I) was concentrated to 2 ml in vacuo at 50°C. Five milliliters of 0.1 M Na acetate buffer (pH 5.0) containing 50 U b glucosidase (E.G. 3.2.1.21) was added to cleave (b 1,6 sugar conjugates. After incubation for 12 hr at 37°C, another 50 U of b glucosidase was added. Following incubation at 37°C for another 12 hr, the aglycones were partitioned into CH2Cl2 (3 x 25 ml). The remaining aqueous phase (Polar II) was adjusted to pH 1.5 with 1 N HCl and partitioned immediately against CH2Cl2 (3 x 25 ml). All CH2Cl2 phases were individually evaporated in vacuo at 30°C to dryness. Samples were dissolved and stored in 0.5 ml of methanol prior to chromatographic analysis.
Chromatographic analysis. Separation of clomazone and clomazone metabolites was done using HPLC with a H2O/acetonitrile gradient and a C,8 reverse phase HPLC column. Gradient steps were: (a) 0 to 20% acetonitrile from 0 to 15 min, (b) 20 to 50% acetonitrile over the next 5 min, (c) 50% acetonitrile for the next 13 min, (d) 50 to 95% acetonitrile over the next 5 min, and (e) 95 to 0% acetonitrile over the next 5 min. Fractions were collected at 1 min intervals and radioactivity was determined by LSS. With this solvent system, clomazone and 2 CBA had retention times of 31.2 and 25.5 min, respectively.
GC/MS data on HPLC purified metabolites were obtained using a Varion 5890 gas chromatograph coupled to a Finnigan MAT ITD 800 ion trap detector mass spectrometer. The GC was equipped with a fused silica capillary column 15 m long by 0.25 mm i.d. containing a 0.25 mm bonded phase of Durabond 5 (J and W Scientific, Folsom, CA). The GC column was coupled directly to the ion trap manifold through the transfer line. The transfer line was maintained at 280ºC. The linear velocity of helium through the column was 26 cm sec-1. Splitless injections of 1 ml were made at an injection port temperature of 280°C. The GC oven was maintained at 50°C for 4 min and was increased at 6°C min-1 to a maximum of 300°C. The multiplier voltage was set at 1450 V. The ion trap detector was repetitively scanned from 50 to 450 amu in 1.0 sec.
Statistical analysis. All experiments were conducted twice with at least two replications of each treatment per experiment. Analysis of variance were performed on data expressed as nmol g-1 fresh wt. Means were compared with Fisher s least significant difference test.
Results
Clomazone uptake and metabolite retention. Cells of both soybean and velvetleaf retained more total radioactivity when treated with MET C than with CAR C (Figs. 1A 1D). Retention of I4C increased in both soybean and velvetleaf over time. Soybean treated with MET C initially had higher concentrations of radiolabeled compounds than did velvetleaf cells, but by 48 hr velvetleaf had higher concentrations than did soybean. Compared to velvetleaf, soybean cells treated with CAR C had slightly higher concentrations of I4C compounds at all times except 48 hr.
Clomazone metabolism. Soybean cells treated with MET C had clomazone concentrations of 0.7 to 1.0 nmol g-1 fresh wt throughout the experiment (Fig. 1A). Although the metabolite concentration increased over time, clomazone was metabolized more rapidly from 0 through 6 hr than from 6 through 48 hr.
Velvetleaf cells treated with MET C had clomazone concentrations from 0.1 to 0.3 nmol g-1 fresh wt (Fig. IB). This was lower than for soybean cells treated with MET C. Metabolite concentration in velvetleaf cells increased at a constant rate through 24 hr. A decrease in the rate of metabolite accumulation was observed between 24 and 48 hr. The metabolite concentration made up a higher percentage of total labeled compounds in velvetleaf cells than in soybean cells (93% versus 82% at 48 hr, respectively).
Soybean cells treated with CAR C had clomazone concen trations similar to those observed with soybean cells treated with MET C (0.7 to 1.0 nmol g-1 fresh wt) (Fig. 1C). However, the metabolite concentration was lower than that observed for MET C soybean cells. Thus, the percentage of labeled metabolites in CAR C soybean cells was lower than that observed for MET C treated soybean cells (55% versus 82% at 48 hr, respectively).
A similar difference between MET C and CAR C treated cells on uptake and metabolism was observed with velvetleaf (Figs. IB and ID). Clomazone concentrations were equal between CAR C and MET C velvetleaf cells; however, CAR C treated cells had lower metabolite concentrations than MET C treated cells. Thus, as was observed in soybean cells, velvetleaf cells treated with CAR C had a lower percentage of radiolabel as metabolites when compared to cells treated with MET C (85% versus 97% at 48 hr, respectively).
Extracellular metabolites. Both MET C and CAR C treated cells of soybean and velvetleaf had detectable levels of metabolites in the media (Fig. 2). However, greater amounts of water soluble clomazone metabolites were present in the media of CAR C treated cells than in MET C treated cells of both velvetleaf and soybean. Concentrations of metabolites in the media increased over time. No differences in the amounts of extracellular clomazone metabolites were observed between soybean and velvetleaf treated with MET C.
Metabolite characterization. Soybean and velvetleaf cells produced the same clomazone metabolites based on HPLC elu tion profiles (Fig. 3). Because velvetleaf metabolized clomazone more rapidly than soybean did (Fig. 1), higher concentrations of total metabolites were observed in velvetleaf than in soybean (Table 1). However, the differences in percentage distributions of the clomazone metabolites, if observed, were small between soybean and velvetleaf cells whether they were treated with CAR C or MET C (Table 1).
All of the metabolites were more polar than clomazone (Fig. 3). Velvetleaf produced higher concentrations of H2O soluble (Polar I) metabolites than soybean did (Table 2). Since more H2O soluble metabolites leaked out of the CAR C treated cells than the MET C treated cells (Fig. 2), lower amounts of H2O soluble metabolites were detected in the cells of each species with CAR C than MET C treatments.
A majority (>85%) of the radiolabel present in the nonconjugate fraction (initial CH2Cl2 phase) was clomazone for both soybean and velvetleaf cells. Confirmation of the identity of clomazone in this fraction was demonstrated by HPLC and mass spectral analysis. Mass spectra (El) of the isolated products from the different cells and labels were nearly identical to clomazone reference standard (Fig. 4). Characteristic molecular (M+) and base peak ions were observed at m/z 240 and m/z 125, respectively. Clomazone may also lose Cl to give (M Cl) m/z 204. The other minor metabolite had a retention time of 22 min and was present in both MET C and CAR C cells of soybean and velvetleaf.
Antidote Mode of Action
Studies have suggested that antidotes (safeners) for chloroacetamide herbicides protect crops by inducing glutathione conjugation of these herbicides. Antidote activity has been correlated with enhanced herbicide metabolism and with enhanced glutathione S transferase (GST)2 activity. Until recently the levels of parent herbicide and metabolites in tissues of plants grown in soil had not been assayed, raising the question of whether enhanced herbicide metabolism played a role in antidote action in vivo, and if so, whether enhanced metabolism was the primary mechanism of antidote action.
The first study in this series indicated that BAS 145 1383 (BAS) protected corn (Zea mays L.) from metazachlor injury by reducing the concentration of parent metazachlor in growing tissues, especially in the developing leaves. This decrease was attributed to (i) enhanced metabolism of metazachlor, (ii) decreased mobility of [14C] metazachlor and/or its metabolites, and (iii) slightly decreased absorption of metazachlor. The enhanced metabolism was not measured directly but was suggested as an explanation for the reduced amount of parent [14C] metazachlor as a percentage
of total radioactivity in BAS treated corn tissues .
The decreased mobility and absorption of radioactivity observed in BAS treated corn plants does not rule out the possibility that enhanced metabolism may be the primary or sole mechanism of antidote action. The reduced mobility of radioactivity in BAS treated plants could be due to the reduced levels of parent metazachlor, since the parent is more mobile than the metabolites, and especially the conjugates, of most herbicides. The reduced absorption of metazachlor could also be the consequence of reduced levels of parent metazachlor (i.e., the consequence of enhanced metazachlor metabolism) in BAS treated corn plants, since chloroacetamide herbicides inhibit cuticular wax development and since herbicide absorption can be increased by herbicide treatments that inhibit cuticle development (or conversely, herbicide absorption may be decreased by treatments such as BAS that decrease levels of chloroacetamide herbicides in plant tissues). Preservation of cuticle development by a herbicide antidote has been previously reported. Thus, we evaluated the possibility that the reduced mobility and absorption of radioactivity in corn plants treated with BAS and [14C] metazachlor were the consequence of enhanced metazachlor metabolism.
Corn cultivars DeKalb XL 25A and XL 55A were not effectively protected from EPTC by the antidote, dichlormid. Since dichlormid and BAS protect corn from metazachlor to the same extent and since both antidotes are dichloroacetamides, it seemed likely that these corn varieties would not be effectively protected from metazachlor injury by BAS. Dean el al. have further noted that the two major GST isozymes induced by the dichloroacetamide antidote, CGA 154281 [4 (di chloroacetyl) 3,4 dihydro 3 methyl 2H 1,4 benzoxazine], appeared to have activity on both chloroacetamide (metolachlor) and thiocarbamate (EPTC) herbicides. Furthermore, wheat seemed likely to be poorly protected from metazachlor by BAS because wheat has not been reported to be protected from chloroacetamide herbicides by dichlormid. If our hypotheses were correct, the reduced antidote responses could be verified using wheat and the two corn varieties, thereby allowing us to determine whether the decreased antidotal activity was also correlated with an altered effect on herbicide metabolism.
Experiments were conducted to further investigate the significance of metabolism, mobility, and absorption of metazachlor in the mode of action of BAS. The objectives of this study were to (i) directly assay the rate of metazachlor metabolism in the growing tissues of corn seedlings; (ii) evaluate metazachlor phytotoxicity and metabolism in plants with reduced responses to dichloroacetamide antidotes, including certain corn cultivars and wheat; (iii) evaluate the mobility of metazachlor metabolites; (iv) evaluate the effect of metazachlor and BAS treatments on subsequent metazachlor absorption in corn; and (v) evaluate the effect of BAS on glutathione (GSH) levels and GST activity.
MATERIALS AND METHODS
General
Experiments were conducted with corn cultivar Northrup King PX9144 unless otherwise indicated. Corn was grown in soil as previously described at 21°C with incorporated treatments of 0 or 5 parts per million by weight (ppmw) metazachlor and 0 or 1 ppmw BAS. Metazachlor at 5 ppmw inhibited corn growth and BAS at 1 ppmw prevented most of this inhibition. [14C]Metazachlor [pyrazol U 14C, sp act 11.6 mCi/mmol (1.55 MBq/mmol)] was used in all studies. All experiments were repeated.
[14C] Metazachlor metabolism studies
Four day old unemerged seedlings having shoots 3.0 to 4.5 cm long were removed from the soil and washed with water, and 2 cm root or apical shoot sections were excised. Fifteen root sections (0.30 0.35 g) or six shoot sections (0.40 0.45 g) were placed in 18 x 70 mm test tubes and pulse labeled for 30 sec by vacuum infiltration with 2.5 ml of aqueous 5 mM [14C]metazachlor. The shoots and solution were stirred gently with a vortex mixer during vacuum infiltration. After vacuum infiltration, the [14C] metazachlor solution was immediately removed by aspiration and the tissue was rinsed three times with 3 ml portions of water. The water rinses were also removed by aspiration and the moist tissue was incubated in the original test tubes at 25°C for the indicated periods. After the incubation, the sample was frozen on a dry ice acetone bath and stored at 20°C. Samples were ground with a polytron in 3.5 ml of 70% acetone and filtered through Whatman No. 3 filter paper. The polytron and filter were sequentially rinsed with 3.5 ml of 35% acetone and 3.5 ml of water. Extracts were partitioned three times with 10 ml of methylene chloride. The 14C in the aqueous phase of partitioned extracts was shown by thin layer chromatography to be metabolites of [14C] metazachlor and the 14C in the methylene chloride phase was confirmed to be parent [14C] metazachIor. Radioactivity in the two phases was quantified by liquid scintillation spectrometry. There were two replications per treatment.
Effect of BAS on Metazachlor Metabolism in Excised Tissues
Four studies were conducted:
Effect of metazachlor concentration. Plants were grown in soil treated with 5 ppmw metazachlor and 0 or 1 ppmw BAS. These treatments were chosen both here and in the time course (below) because they were the same as those used in the previous study. The effect of [14C] metazachlor concentration (0.5 50 mM) during the pulse label period on the rate of metazachlor metabolism was evaluated using a 30 min incubation.
Effect of soil treatments. Plants were grown in soil treated with 0 or 5 ppmw metazachlor with or without 1 ppmw BAS or 1 ppmw dichlormid. The 0 ppmw metazachlor treatment was included so that the effect of the antidote alone could be evaluated. Metazachlor metabolism in shoots and roots was evaluated using a 30 min incubation.
Time course. Plants were grown in soil treated with 5 ppmw metazachlor and 0 or 1 ppmw BAS. Atime course for metazachlor metabolism in shoots and roots was conducted for incubation periods ranging from 1 to 60 min.
Comparison of shoot tissues. Plants were grown in soil treated with 0 or 1 ppmw BAS. Metazachlor was omitted from the soil treatment in this experiment because metazachlor caused the leaves to adhere tightly to the coleoptile, thus making the dissection difficult. The shoot was cut at the coleoptile node, a 2 cm section of mesocotyl was excised, and the leaves were dissected from the coleoptile. Six shoots were dissected per replication. Metazachlor metabolism was evaluated separately for each tissue.
Effect of BAS on Metazachlor Phytotoxicity at Metabolism in Three Corn Cultivars and Wheat.
Growth studies were conducted with three corn cultivars, Northrup King PX9144, DeKalb XL 25A, and DeKalb XL 55A, and one wheat cultivar, Olaf. Fifteen corn seeds or 25 wheat seeds were planted in vermiculite and grown under conditions previously described. Treatments were applied in the initial watering solution. Metazachlor was applied at concentrations ranging from 0 to 300 mM for corn and from 0 to 10 mm. for wheat. BAS was applied at 0, 1, or 10 mm. Treatments which contained both metazachlor and BAS were applied as a single solution. Shoot height was evaluated 12 days after planting for corn cultivars Northrup King PX9144 and DeKalb XL 25A and 13 days after planting for corn cultivar DeKalb XL 55A and wheat. Untreated plants were at the two leaf growth stage at the time of harvest. There were two replications of each treatment.
Metabolism studies were conducted using separate plants grown as described above and treated with 0,1, or 10 mM BAS. Northrup King PX9144 seedlings were 4 days old and other seedlings were 5 days old at the time of evaluation; all seedlings were evaluated prior to emergence. Metabolism of [14C]metazachlor in apical shoot sections was evaluated as described above except that the incubation period was 15 min and 15 apical shoot sections of wheat (approximately 0.40 g) were used.
Mobility of Metazachlor Metabolites.
Corn seedlings were grown for 3.5 days at 21°C in vermiculite treated with 0 or 2 mM aqueous BAS. Metazachlor was omitted because of the previously described difficulties it causes in dissection. Seedlings with 2.5 to 4 cm long shoots were removed and washed. Five seedlings were placed in a 50 ml test tube for each replication. Seedlings were pulse labeled by submerging and aerating them in 25 ml of aqueous 5 mM [14C] metazachlor for 12 min. Seedlings were rinsed three times with 25 ml water, placed between layers of moist germination paper with the shoots protruding, and incubated at 21°C under an inverted 800 ml beaker lined with moist paper towels to maintain a high humidity. The germination paper was changed at 0.5, 1, 2, and 4 hr to minimize reabsorption of parent [l4C] metazachlor that may have diffused into the germination paper. The shoots were dissected into coleoptile, leaves, and mesocotyl. The tissues were weighed and frozen after 4 and 36 hr. Tissues were analyzed for parent metazachlor and metabolites as previously described. There were two replications of each treatment.
In a second experiment, the ability of metazachlor metabolites to diffuse out of corn shoot apical sections into water was evaluated. Corn shoot apical sections were excised, pulse labeled with [14C] metazachlor, and incubated for 30 min, as described for studies of metabolism in excised shoot tissues. This was followed by a 30 min incubation in 5 ml water to allow diffusion of metazachlor and metabolites out of the shoot. The levels of parent metazachlor and metabolite in corn tissue and water were analyzed as described for metabolism studies.
Effect of BAS and Metazachlor on [14C] Metazachlor Absorption.
Corn was grown in soil treated with 0 to 15 ppmw metazachlor and 0 or 1 ppmw BAS. Four day old unemerged seedlings with shoots 3 to 4.5 cm long were removed from the soil, washed, and placed between moist paper towels. Each seedling was placed in a 5 ml scintillation vial and treated with 10 1.5ml droplets containing 10 mM aqueous [14C] metazachlor (no organic solvent present). Five droplets were applied to the coleoptile and five were applied to the mesocotyl. Treated seedlings were incubated for 10 min under an inverted 800 ml beaker lined with moist paper towels to maintain high humidity. The shoot was then rinsed for 5 sec under a stream of acetone and 5 sec under a stream of chloroform to remove cuticular waxes and any associated 14C. The shoot was excised and absorbed radioactivity was determined by oxidation and liquid scintillation spectrometry. There were three seedlings per replication and three replications per treatment.
Effect of BAS on GSH Levels and GST Activity.
Corn was grown in soil treated with 0 or 1 ppmw BAS. GSH was assayed using equine GST and the substrate 1 chloro 2,4 dinitrobenzene as previously described (6). There were three replications for each treatment. GST was assayed as previously described with minor modifications. GST assays were conducted at 30°C and the reaction was started by adding [14C] metazachlor (5 mM final concentration). Enzymatic activity values were corrected for nonenzymatic and Time 0 controls. There were two replications for each time point.
RESULTS AND DISCUSSION
Effect of BAS on Metazachlor Metabolism in Excised Tissues
Effect of metazachlor concentration. The concentration of [14C]metazachlor used for pulse labeling did not significantly affect metazachlor metabolism in excised shoots when expressed as percentage of 14C absorbed, nor did metazachlor concentration affect the differences between antidoted and control treatments (Fig. 1). Results with roots were virtually identical (data not shown). The 5 mM metazachlor concentration was used in all subsequent studies.
Effect of soil treatments. BAS and chlormid increased the rate of metazachlor metabolism in corn shoots (Table 1). This correlates with the protection from metazachlor injury conferred by these two antidotes. Metazachlor soil treatment slightly increased the rate of [14C] metazachlor metabolism in corn shoots whether or not BAS was present in the soil. The effects of BAS and metazachlor treatments on metazachlor metabolism in the roots were similar to the effects in shoots (data not shown). The effect of dichlormid on roots was not evaluated. A similar stimulation of the metabolism of the chloroacetamide herbicide metolachlor by itself was previously reported.
Time course. Metazachlor was metabolized rapidly in unantidoted shoot and root tissues of corn but metabolism was even more rapid in tissues grown in BAS treated soil (Fig. 2). The half life of metazachlor was 58 and 14 min in roots (Fig. 2A) and 33 and 13 min in shoots (Fig. 2B) from unantidoted and antidoted plants, respectively. This is consistent with the previous dissection and growth studies which showed that BAS protected both roots and shoots and indirectly indicated that BAS enhanced the rate of metabolism of metazachlor in both tissues.
Comparison of shoot tissues. BAS soil treatment enhanced [14C] metazachlor metabolism in the coleoptile, developing leaves, and mesocotyl (Fig. 3). It appears that the effects of BAS on metabolism are not tissue specific. This suggests that metabolism of metazachlor in tissue adjacent to the developing leaves, i.e., the coleoptile and mesocotyl, could reduce the amount of metazachlor reaching the leaves and that the leaves themselves could further decrease the level of metazachlor present in BAS treated plants. This corresponds with measurements of metazachlor levels in corn plants grown in soil, in which BAS reduced the level of metazachlor in the mesocotyl, coleoptile, and leaves.
Effect of BAS on Metazachlor Phytotoxicity and Metabolism in Three Corn Cultivars and Wheat
All three corn cultivars showed substantial increases in the concentration of metazachlor required for 50% inhibition of growth (I50) as the BAS concentration increased from 0 to 10 mM (Table 2). These increases in tolerance were associated with increases in metazachlor metabolism rate. Wheat was much less tolerant of metazachlor than corn, as indicated by the lower I50 values for wheat. Wheat also was not protected from metazachlor by BAS, as indicated by the constant I50 values as BAS concentration increased. Metazachlor metabolism rates were correspondingly slower in wheat and these rates did not increase as BAS concentration increased (Table 2).
The corn cultivars, DeKalb XL 25A and XL 55A, were protected from metazachlor by BAS, contrary to our previously discussed expectation (see Introduction). It is possible that the previously reported apparent lack of dichloroacetamide antidote activity in these two corn cultivars is not repeatable, since no indication was given of the number of plants evaluated, the number of replications, the variability, or whether the experiment was repeated. The data reported here indicate that the level of antidotal activity is correlated with the rate of herbicide metabolism. When there is no antidotal activity, as in the case of wheat, there is likewise no effect of the antidote on herbicide metabolism rate (Table 2).
Mobility of Metazachlor Metabolites
[14C]Metazachlor absorbed by corn shoots should be almost entirely metabolized within 4 hr because it has a half life of only 33 min or less (Fig. 2). Therefore, if metabolites are immobile, as we have proposed, radiolabel should not move after 4 hr, even if a concentration gradient exists.
Most [14C] metazachlor was converted to metabolites in corn seedlings after incubation for 4 hr (Table 3). A strong concentration gradient (dpm/mg) of metabolites was present at 4 hr among the three tissues evaluated; the coleoptile had the highest concentration and the developing leaves had the lowest concentration. The concentration of radioactivity (dpm/mg) declined between 4 and 36 hr in all tissues as a result of growth but the concentration gradients remained. The total amount of radioactivity (dpm) remained constant in these tissues between 4 and 36 hr despite the concentration gradient. Similar results were obtained in both untreated and BAS treated plants (Table 3). Thus, it appears likely that the glutathione conjugate of metazachlor and its short term catabolic products are relatively immobile.
In a second experiment, the ability of metazachlor and its metabolites to diffuse out of corn shoot apical sections into water was also evaluated. Only 2.2% of the metabolites present in corn shoots diffused into water, versus 14.7% of the parent metazachlor. This observation also suggests that metazachlor metabolites are less mobile than the parent. The plasmalemma and/or tonoplast presumably act as barriers to movement of polar metabolites.
Effect of BAS and Metazachlor on [14C] Metazachlor Absorption
As the rate of unlabeled metazachlor applied to the soil increased, the amount of [14C] metazachlor absorbed increased (Fig. 4). BAS reduced the effect of soil applied metazachlor on [14C] metazachlor absorption, but when no metazachlor was applied to the soil (i.e., at 0 ppmw metazachlor in Fig. 4). BAS did not decrease absorption. This suggests that BAS reduced absorption indirectly; perhaps metazachlor alone inhibited synthesis of cuticular waxes, and BAS prevented this effect by enhancing the detoxification of metazachlor. These obser vations are consistent with previous reports that chloroacetamides inhibit cuticular wax development, that herbicide treated plants absorb more herbicide and that antidote treatment maintains cuticular wax integrity in the presence of the herbicide. These observations are consistent with the hypothesis that the apparent slight BAS induced decrease in metazachlor absorption previously observed in corn seedlings grown in the soil actually represented protection from a metazachlor induced increase in [14C] metazachlor absorption.
Effect of BAS on GSH Levels and GST Activity
Metazachlor is metabolized to the glutathione conjugate in corn. An antidote that induces metazachlor metabolism might do so either by increasing GSH levels or by increasing GST activity.
Untreated and BAS treated corn shoots contained 3.0 and 3.1 (mmol GSH/g fresh wt, respectively. Thus, the antidote had no significant effect on GSH levels.
GST activity was low in extracts of plants grown in untreated soil (Fig. 5), suggesting that a significant portion of metazachlor metabolism observed in untreated plants may be nonenzymatic. However, the relative importance of enzymatic and non enzymatic metabolism in unantidoted corn remains inconclusive because the concentration of the enzyme in the assay was diluted 70 fold compared to its concentration in vivo, and because some loss of activity could have occurred prior to the assay. Antidote treatment enhanced GST activity severalfold (Fig. 5). BAS induced metaza chlor metabolism is probably due to induction of GST activity.
Diclofop Resistance in Avena Fatua
The herbicide diclofop methyl, a member of the aryloxy phenoxypropionate class of herbicides, is commonly used in western Canada to control grass weeds, including wild oat (Avena fatua L.), in cereal grain crops. Selectivity between wild oat and wheat is based on differences in metabolism of diclofop methyl in the two species. In wheat, the herbicide is rapidly converted to an O glycoside following ring hydroxylation; this product is considered to be an irreversible detoxification product. In susceptible species, such as wild oat, the predominant metabolite is a glucose ester of diclofop (the parent acid), which can be hydrolyzed in vivo to regenerate the active form, diclofop.
The target site of diclofop and related herbicides is acetyl coenzyme A carboxylase (ACCase;3 EC 6.4.1.2), a key enzyme in acyl lipid biosynthesis. Many dicotyledonous plants are resistant to diclofop and other aryloxyphenoxypropionate herbicides, based on the low sensitivity of dicot ACCase to these herbicides. These species are also resistant to cyclo hexanedione herbicides, such as tralkoxydim and sethoxydim, which act on the same target enzyme. One grass species, Festuca rubra (red fescue), is resistant to both aryloxyphenoxypropionate and cyclo hexanedione herbicides; resistance is based on reduced sensitivity of F. rubra ACCase to these herbicides compared to that from susceptible species.
A diclofop resistant biotype of annual ryegrass (Lolium rigidum) has been identified in Australia, but the mechanism of resistance has not been determined. Several biotypes of wild oat have been identified in western Canada and elsewhere that are resistant to diclofop methyl. Recent characterization of some of the Canadian biotypes indicates that they are initially injured by the herbicide after application, but recover within 9 days after treatment.
The objective of this study was to examine the physiological and/or biochemical basis for resistance in two of these wild oat biotypes. This included examination of foliar absorption of diclofop methyl, translocation within the plants, metabolism of diclofop methyl, and comparison of the sensitivities of ACCase from the two resistant biotypes and a susceptible biotype to diclofop and tralkoxydim, a cyclohexanedione ACCase inhibitor.
MATERIALS AND METHODS
Plant material. Wild oat and wheat seeds were germinated in 9 cm petri dishes lined with Whatman No. 1 filter paper moistened with distilled water. After 2 to 3 days, seeds with emerged radicles were transferred to 195 ml Styrofoam cups containing coarse silica sand or vermiculite. The sand cultures were subirrigated with half strength Hoagland s solution. The plants were grown in a growth cabinet maintained at 22/18°C day/night temperatures, in a 16 hr photoperiod at 325 mE m-2 s-1. Plants were treated at the three leaf stage of development in all experiments, except where noted.
Diclofop methyl uptake and translocation. Solutions of [l4C] diclofop methyl ([U 14C] dichlorophenyl; sp act, 690.8 MBq g-1) were prepared in 10% aqueous ethanol so that 10 ml contained approximately 840 Bq. The herbicide solution (10 ml) was applied as 5 to 10 droplets to the second leaf using a Wiretrol capillary micropipette (Drumond Scientific Co.), at various time intervals after application; unabsorbed [14C] diclofop methyl was removed from the leaf surface by washing the treated area three times with 5 ml 10% aqueous ethanol. The I4C content of the leaf washes was determined by liquid scintillation spectrometry (LSS). The plants were then divided into the treated leaf, remainder of shoot, and root; the treated leaf was further subdivided into the treated area, the upper portion (i.e., toward the leaf tip), and the basal portion. The plant tissue was air dried at room temperature and combusted in a biological sample oxidizer and the 14C content of the different plant parts quantified by LSS.
Uptake of diclofop methyl was calculated as the total amount of radioactivity recovered in the plant, expressed as a percentage of the total applied. Translocation was calculated as 14C recovered in various plant parts, expressed a percentage of the total radioactivity taken up by the plants.
Metabolism of [14C] diclofop methyl. The metabolism of [l4C] diclofop methyl in the different wild oat lines was examined both by thin layer chromatography (TLC) and by high pressure liquid chromatography (HPLC). In the TLC experiments, plants were treated as described above and sampled 24 and 72 hr after treatment. The treated leaf was washed (as above) and the treated area homogenized and extracted twice with 5 ml 80% methanol. The ho mogenate was filtered through Whatman No. 1 filter paper and dried under nitrogen in a heating block at 37°C. The dried residue was redissolved in 400 ml 80% methanol, and a 50 mi.l aliquot removed and counted by LSS to determine total radioactivity content. The remainder of the sample was applied in a band to a plastic backed silica gel TLC plate (15 cm by 5 cm) and allowed to dry at air temperature. The plates were run in baths containing approximately 150 ml solvent (toluene: acetic acid: ethanol, 150:7:7) until the solvent front had advanced to the top of the plates. The plates were then cut into 15 I cm strips; each strip was placed in a scintillation vial, and 14C content determined by LSS. Counting efficiency was not significantly affected by the presence of the plastic strip in the scintillation vial.
The Rf values were calculated based on the location of 14C activity on the plates and were compared to standards of diclofopmethyl, diclofop, and ring hydroxylated diclofop. The Rf values of these compounds were determined by running standards using the same solvent system and locating the compounds on the plates by ultraviolet fluorescence. The Rf values of diclofopmethyl, diclofop, and aryl hydroxylated diclofop were 0.85, 0.35, and 0.15, respectively.
An attempt was made to identify polar metabolites (i.e., those with an Rf 60,000 fold resistant to topically applied abamectin. This resistance could not be suppressed by the synergists piperonyl butoxide or S, S, S tributyl phosphorotrithioate and did not confer cross resistance to lindane, dieldrin, crotoxyphos, dichlorvos, dimethoate, permethrin, or tetrachlorvinphos. In this paper we investigated the biochemical mechanisms and the genetic control of the >60,000 fold abamectin resistance found in the AVER strain of house fly.
MATERIALS AND METHODS
Insects. Three strains of house fly were used in this study: S + is a laboratory susceptible strain originally obtained from Dr. F. W. Plapp, Jr., of Texas A&M University, College Station; aabys is an insecticide susceptible strain from Dr. T. Hiroyoshi, University of Osaka, Japan, which has the recessive morphological markers ali curve, aristapedia, brown body, yellow eye, and snip wing on autosomes 1, 2, 3, 4, and 5, respectively; and the AVER strain that has high levels (>60,000 fold) of resistance to topically applied abamectin and moderate levels (35 fold) of resistance to formulated abamectin by residual exposure.
To examine the inheritance of abamectin resistance in the AVER strain we crossed AVER to the susceptible aabys strain en mass. Females of the appropriate strain were isolated every 8 hr to be certain they had not mated.
Chemicals. Abatmectin, abamectin 8,9 oxide, MK 243 (4 deoxy 4 epimethyI amino avermectin B1), and [5 3H]abamectin (sp act 10.8 Ci/mmol) were gifts from Merck Sharp and Dohme Research Laboratories (Rahway, NJ). Purity of [3H] abamectin was 99 4% by thin layer chromatography (TLC). All other compounds were purchased from commercial sources.
Insecticide bioassays. Insecticide bioassays were carried out by topical application and injection. Topical application was conducted using 3 to 5 day old female flies. Injection was carried out as follows: 0.25 ml of abamectin solution in acetone was injected into the metathoracic notum of 3 to 5 day old female flies using a 10 ml syringe (No. 701, Hamilton Co., Reno, NV). Each replicate consisted of 20 flies/dose and at least four doses. All tests were run at 25°C and were replicated six times. Mortality was assessed after 48 hr. Probit analysis was by the method of Finney as modified by Raymond.
Thin layer chromatography and radiography. TLC analysis was conducted on silica gel plates (60 F254, Merck). The solvent system used was ethyl acetate: toluene:2 propanol (10:3:1). The Rf value of abamectin in this solvent system was 0.61. Radioactive metabolites were detected by autoradiography, and unlabeled abamectin was detected by us visualization. Radioactive zones scraped off TLC plates were quantified by liquid scintillation counting (LSC). Liquiscint (National diagnostics. Palmetto, FL) was used as scintillation fluid and gave 47% counting efficiency.
Penetration studies. The time course of penetration was determined by applying 0.85 ng [3H] abamectin in 0.5 ml acetone to the thoracic notum of female flies (4 days old) of the AVER and S + strains and placing them, in groups of five, into scintillation vials. At this dose none of the flies showed symptoms of poisoning. At selected times the flies were transferred to another scintillation vial and rinsed twice in 5 ml of acetone for about 10 sec with gentle shaking. The body rinses were combined, evaporated, and analyzed by LSC. The holding vials were also analyzed by LSC. Percentages of unchanged [3H] abamectin contained in the body rinses and holding vials were identified using TLC. Each time point was replicated three times per experiment. Data are the average of four experiments.
Metabolism studies. In vivo metabolism of [3H] abamectin was conducted using an injection technique. A dose of 0.425 ng [3H]abamectin in 0.25 ml acetone was injected into the metathoracic notum of female flies (4 days old) of the AVER and S + strains using a 10 mI syringe (No. 701, Hamilton Co.). Groups of five treated flies were placed into scintillation vials. At this dose, none of the flies showed symptoms of poisoning. At selected times the flies were transferred to a new scintillation vial and homogenized in 10 ml of an ethyl acetate: water mixture (1:1) using a Biohomogenizer (Biospec Products, Inc., Bartlesville, OK). The homogenized solutions were then centrifuged at 1000 g for 5 min, and the 3H content in the ethyl acetate soluble and the water soluble fractions was measured by LSC. The ethyl acetate soluble fraction was reduced in vacuo and analyzed using TLC. The holding vials were also analyzed by LSC and TLC as described above. Unextractable materials were digested with 1 ml of Protosol for 12 hr and analyzed by LSC.
Receptor binding. Receptor binding was carried out by the methods of Schaeffer and Haines with slight modification. Adult house flies (4 days old) of the AVER and S + strains were frozen at 80°C for 1 hr. Immediately upon removal from the freezer, the container was shaken vigorously, and the frozen body parts were sifted through a 1.7 mm mesh sieve, which removed most of the heads, legs, and wings. Retained body parts were placed in a 1 liter beaker and gently swirled in 400 ml ice cold distilled water. Thoraces floated to the surface and were collected. The thoraces were first homogenized in 10 vol of ice cold 50 mM Hepes buffer (pH 7.4) using a Biohomogenizer, and then rehomogenized using a Wheaton Potter glass teflon homogenizer. The homogenate was filtered over two layers of muslin and the filtrate was centrifuged for 5 min at 1000g. The pellet was discarded, and the supernatant fraction was centrifuged for 20 min at 40,000g. The resulting pellet was resuspended in Hepes buffer to approximately 5 mg protein/ml. The membrane preparations (1.0 ml) were incubated with various concentrations (0.125 3.0 nM) of [3H]abamectin at 22°C for 45 min in the presence (nonspecific binding) or absence (total binding) of 1.25 p.M unlabeled abamectin in glass tubes (13 x 100 mm). For the equilibrium binding experiments, 1 mg protein/ml was used. The incubation was terminated by rapid filtration over Whatman GF/B filters (presoaked with 0.15% poly ethyleneimine and 0.5% Triton X 100). The filters were immediately rinsed with 15 ml (3 x 5 ml) of ice cold Hepes buffer containing 0.25% Triton X 100. The filters were placed into scintillation vials containing 10 ml of Liquiscint for 16 hr, and the radioactivity was determined by LSC. Specific binding was calculated by subtracting nonspecific from total binding. Protein concentration was determined according to the method of Bradford.
Results
Inheritance ofabamectin resistance. Reciprocal crosses between the aabys and AVER strains produced an F1 generation that gave nearly identical dose response lines (Table 1), indicating that abamectin resistance was not sex linked or due to cytoplasmic factors. Abamectin resistance was also highly recessive, having only a 3.5 to 5.4 fold resistance in the F1 compared to >60,000 fold for the AVER strain. While other recessive resistance mechanisms (ex. kdr) are known, the tremendous change in abamectin resistance between the F1 and parental AVER strain is quite remarkable. Males generally had lower LD50 values than females, probably due to the slightly smaller size of the males. The S+ strain was slightly less sensitive to abamectin compared to the aabys strain, although this difference is commonly noted between these strains and is most likely due to the larger size of the S + flies. Given the steep slopes of the dose response lines for aabys and the S+ strains, it is likely that these strains have little variability in their response to abamectin. Since the log dose response lines for the F1 progeny (aabys x AVER) also had steep slopes, this suggests that the AVER strain may be homozygous for the major mechanisms of resistance.
Toxicity tests. Toxicity tests by topical application and by injection were conducted against the S + and AVER strains. Results are shown in Table 2. The AVER strain showed only a 35 fold resistance to abamectin by injection. Since the AVER strain showed high levels (>60,000 fold) of resistance to topically applied abamectin, the dramatic change suggests that decreased cuticular penetration is an important mechanism of resistance in the AVER strain.
The AVER strain was found to be >4000 fold cross resistant to abamectin oxide with LD50 values of 0.02 and >100 mg/fly for the S+ (susceptible) and resistant AVER strains, respectively. MK 243 was reasonably toxic to S + house flies with an LD50 of 0.03 mg/fly. However, the AVER strain was very heterogeneous in its response to this compound, having a dose response line that was very flat: ranging from 5.7% kill at 0.03 mg/fly to 90% kill at 60 mg/fly. The approximate LDso was 0.4 mg/fly. Although the resistance ratio for this compound (»13) is much less than found toward abamectin oxide, the heterogeneity of response in the AVER strain suggests that high levels of resistance could be selected for by the use of abamectin oxide.
Penetration studies. The rate of penetration of radiolabeled abamectin is shown in Fig. 1. The penetration of radiolabeled abamectin was relatively slow, reaching only 53% after 8 hr in the S + strain. The rate constant for penetration of radiolabel was 2.4 fold slower in AVER than in S + with values of 1.37 and 3.28 x 10-2 hr-1, respectively. This suggests that decreased cuticular penetration is one of the mechanisms of resistance in the AVER strain. TLC and LSC analysis of the body rinses and holding vials at 8 hr after topical application showed that >98% of radiolabel recovered was [3H] abamectin in AVER and S+ strains. Nearly all of the applied radio label was recovered from the body rinses at all time points, while there was very little radiolabel remaining in the holding vials (i.e., less than 3% of total radiolabel even after 8 hr).
Metabolism studies. Because we lacked unlabeled standards of abamectin metabolites, no attempt was made to identify me tabolites. The metabolites of [3H] abamectin were, therefore, simplified to four groups: water soluble metabolites, abamectin in the ethyl acetate soluble fraction, unknown metabolites in the ethyl acetate soluble fraction, and unextractable materials. The results are summarized in Table 3. Almost 90% of the radiolabel was recovered from the body homogenates while less than 11% was recovered from the holding vials even 8 hr after treatment in both strains. In the body homogenates, the radiolabel from the water soluble fraction and the unknown metabolites in the ethyl acetate soluble fraction were constantly about two times higher in the S + strain than in the AVER strain. The radiolabel from the unextractable materials was also higher in the S + strain than in the AVER strain. These data suggest that increased metabolism is not a mechanism of abamectin resistance in the AVER strain. In the holding vials, unchanged abamectin was detected at slightly higher levels in the AVER strain than in the S+ strain. However, this is probably due to more abamectin leaking from the wound caused by the needle rather than as a result of true excretion because abamectin is lipophilic and unlikely to be excreted unchanged.
Receptor binding. Specific binding of [3H]abamectin at 2.5 nM to the membrane preparations of the AVER and S + strains increased linearly as a function of tissue protein concentration up to 1 mg protein/ml, and the specific binding was constantly about 1.8 times higher in the S + strain than in the AVER strain (Fig. 2). Nonspecific binding increased linearly, but was only 7% of total binding at 3 mg protein/ml. Specific [3H]abamectin binding to the membrane preparations was saturable with increasing concentrations of [3H]abamectin in both strains, and the saturated specific binding in the S + strain was about 1.5 times higher than that in the AVER strain (Fig. 3). The Scatchard analysis (Fig. 4) of these data yielded a straight line in both strains, indicating that both strains have a single class of [3HJabamectin binding sites. The equilibrium dissociation constant (x ± SD) was not significantly different between the AVER (KD = 0.74 ± 0.03 nM) and the S + (KD = 0.72 ± 0.02 nM). strains. However, the maximum number of binding sites (Bmaxi) was significantly different between the two strains, with the Bmax for the S+ strain (0.113 ± 0.008 pmol/mg protein) being 1.5 times higher than that of the AVER strain (0.077 ± 0.006 pmol/mg protein).
Chlorimuron Ethyl Metabolism in Corn
Chlorimuron ethyl, N (4 chloro 6 methoxypyrimidine 2 yl) N (2 ethoxycarbonyl benzenesulfonyl)urea, is a sulfonylurea herbicide used for weed control in soybean. Like other sulfonylurea herbicides, the primary mechanism of action of chlorimuron ethyl has been related to its inhibition of acetolactate synthase. Chlorimuron ethyl is metabolized at a much more rapid rate in tolerant soybean than in susceptible cock lebur or pigweed. Two products of chlorimuron ethyl metabolism in soybean have been identified: the free acid formed by hydrolysis of the ester and the homoglutathione conjugate formed by displacement of chlorine. Both products are inactive as inhibitors of acetolactate synthase and it has been concluded that the selectivity of chlorimuron ethyl between tolerant soybean and susceptible cocklebur and pigweed is related to metabolism.
Corn is susceptible to chlorimuron ethyl, and except for an indication that it may undergo hydroxylation, the metabolism of chlorimuron ethyl in corn has not been reported. Several reports have suggested that corn can be protected from sulfonylurea injury by herbicide safeners. However, these sulfonylurea herbicides are of types known to be metabolized in plants by oxidation followed by conjugation with glucose. More recently, chlorimuron ethyl injury to corn was partially alleviated by the use of the herbicide safener BAS 145 138 (l dichloroacetylhexahydro 3,3,8 a trimethylpyrrolo [1,2 a] pyrimidine 6 (2H one). Chlorimuron ethyl metabolism in soybean does not involve oxidation and BAS 145 138 is known to stimulate metabolism by conjugation with reduced glutathione (GSH).
The purpose of this study was to elucidate the routes of chlorimuron ethyl metabolism in corn. Since the primary route of metabolism of chlorimuron ethyl in soybean involves conjugation with homoglutathione, it seemed likely that metabolism in corn might involve conjugation with GSH. Soybean and several other legumi nous species contain homoglutathione rather than GSH and form homoglutathione conjugates by a process that is assumed to be analogous to GSH conjugation in corn. If chlorimuron ethyl was found to be metabolized primarily by conjugation with GSH in corn, it would then be possible to determine if BAS 145 138 protects corn from chlorimuron ethyl by stimulating the formation of a nontoxic GSH conjugate. The mechanisms of action of herbicide safeners that stimulate GSH conjugation have been well studied with herbicides that do not have a precisely known mechanism of action, such as the chloroacetamides and thiocarbamates. It is difficult to prove the mechanism of action of safeners using herbicides that do not have a known mechanism of action and whose metabolism has been studied primarily in tolerant species.
In this paper, the metabolism of chlorimuron ethyl was studied primarily in the roots of corn. The inhibition of acetolactate synthase in the roots of young seedlings is of particular importance because this causes a stunting of root growth and under mild stress can result in the death of the seedling. Metabolism studies were of short term, not exceeding 11 hr, because the primary objective of this study was to determine the rate and routes of initial metabolism. The objective of the second part of this study will be to determine the mode of action of BAS 145 138 as a safener for chlorimuron ethyl in corn.
MATERIALS AND METHODS
Chemicals
Chemicals were obtained from the following sources: chlorimuron ethyl (99.7%) and [phenyl 14C]chlorimuron ethyl (3.48 mCi/mmol, radiochemical purity 99%) (E.I. Du Pont de Nemours and Co., Wilmington, DE), analytical grade BAS 145 138 herbicide safener (1 dichloroacetylhexahydro 3, 3,8a trimethylpyrrolo[l,2 a] pyrimidin
6 (2H one) (98.4%) (BASF Corp., Research Triangle Park, NC), b glucosidase (EC 3.2.1.21) from almonds (Type II) and reduced glutathione (Sigma Chemical Co., St. Louis, MO), high performance liquid chromatography (HPLC) grade acetonitrile. HPLC grade methylene chloride from this source did not cause the oxidation of sulfide conjugates to the corresponding S oxides. The [phenyl-14C] chlorimuron ethyl used in these studies was usually diluted to a specific activity of 0.45 mCi/mmol with analytical grade chlorimuron ethyl.
2 Ethoxycarbonylbenzene sulfonamide was prepared by hydrolysis of chlorimuron ethyl at pH 3 for 24 hr at 40°C. It was purified by extraction into methylene chloride and by chromatography with HPLC System A [retention time (Rt) 10.5 min]. The GSH conjugate of chlorimuron ethyl (N (4 [S glutathionyl] 6 methoxypyrimidine 2 yl) N (2 ethoxycarbonylbenzenesulfonyl) urea was synthesized in 25% yield by the method of Brown and Neighbors (2). It was purified by HPLC System B (Rt 48 min).
Plant Treatment
Corn roots. Corn seed (Zea mays L., Northrup King Hybrid PX 9144) was placed between layers of paper toweling and moistened with 0 or 20 ppm BAS 145 138 in aqueous 0.2% acetone. The toweling was rolled into cylinders, placed vertically into beakers containing 0 or 20 ppm BAS 145 138 in 0.2% acetone, and placed in a growth chamber. After incubation for 4 days (21°C, 50% relative humidity, 14 hr photoperiod), the seedlings were removed and selected on the basis of uniformity of size and appearance (root length 5.5 ± 0.7 cm). Selected seedlings were placed horizontally on grated racks 2 cm above a water bath (22°C) and 12.3 ± 0.61 nmol of [14C]chlorimuron ethyl (0.45 mCi/mmoI) in 2.5 ml of 70% acetone was streaked onto the roots of each seedling. After 7 hr of incubation in the covered water bath, the seedlings were removed and dip rinsed in acetone (20 ml for 5 sec). The roots (60 ± 4 mg fresh wt/seedling) were excised into 1 cm sections, placed into tared centrifuge tubes containing ice water, and frozen at 20°C. Eighty seedlings were used in this experiment and the experiment was repeated three times. In each experiment, 40 seedlings were germinated in the presence of BAS 145 138 and 40 were germinated in the absence of BAS 145 138. Several additional experiments were conducted with seedlings treated with a lower dose (1.3 nmol/root) of a higher specific activity [14C] chlorimuron ethyl (3.48 mCi/mmol).
Plant Treatment
Corn coleoptiles. Corn seed was germinated for 5 days between layers of paper as previously described. After 5 days, nine coleoptiles were excised under water (1 g fresh wt), placed cut ends down into 2 ml of aqueous 5.3 mm [14C]chlorimuron ethyl, and incubated at 22°C. After 20 hr, the coleoptiles were removed and frozen in 1.8 ml of water. The frozen coleoptiles were extracted three times with 6 vol of 70% acetonitrile. The extract contained 47% of the 14C originally used in the treatment. The residual treating solution contained 50.6%.
Corn shoots. Corn seed was germinated between layers of paper as previously described. After 10 days, the resulting shoots were excised under water. The cut ends of the shoots were partially immersed (cut ends down) in 2 ml of 25.9 mM [14C] chlorimuron ethyl in aqueous 0.5% acetone (4 shoots/2.0 g fresh wt tissue/test tube). Three tubes of seedlings were incubated under fluorescent lights at 22°C. After 2.5 hr, the treating solution was removed by aspiration, the shoots were rinsed with 1.0 ml water, and 2.0 ml of water was added. The shoots were incubated for an additional 0.5, 3.0, and 5.5 hr. After incubation, the shoots were diced into 40 ml Teflon centrifuge tubes containing 6 ml of ice water and frozen.
Soybean roots. Soybean seeds [Glycine max (L.) Merr.] variety Evans were germinated between layers of moist paper toweling partially immersed in water as described for corn. After 5 days, 40 seedlings were treated by application of [14C] chlorimuron ethyl to the roots of each seedling (13.6 nmol/2.5 ml 70% acetone/seedling). The seedlings were incubated as previously described for corn. After 7 hr, the seedlings were rinsed with acetone and the roots (4.28 g) were excised and extracted. The extract from the roots contained 118 nmol of herbicide and metabolites/g fresh wt of roots.
Extraction of Tissue
Frozen tissue was thawed by adding acetonitrile to bring the concentration of acetonitrile in the extraction medium to 70%. Tissue was extracted three times with a 10:1 (v/w) ratio of 70% acetonitrile (4°C) using a Polytron homogenizer equipped with a PTA 10TS generator (Brinkman Instruments, Westbury, NY). Extracts were separated from cell debris by centrifugation at 3200g for 5 min. The cell debris was washed with acetone and the 14C present in the cell debris and in the 70% acetonitrile extracts was quantified by liquid scintillation spectrometry. The 70% acetonitrile extracts from each sample were combined, concentrated to ca. 5 ml under vacuum at 37°C, diluted with water to 40 ml, and partitioned three times against 20 ml volumes of methylene chloride. The methylene chloride and water soluble fractions were quantified by liquid scintillation spectrometry and concentrated under vacuum. The methylene chloride fractions were dissolved in 15% acetonitrile for analysis by HPLC as described later. The water soluble fractions were diluted to 15 ml with aqueous 5% acetonitrile, the 14C was quantified, and the solutions were applied to columns containing 3 g of a preparative 55 to 105 mm C18 liquid chromatography packing (Waters Division, Millipore Corp., Milford, MA). The columns were washed with 7 ml of water, eluted with 15 ml of aqueous 67% acetonitrile, and regenerated with 100% acetonitrile followed by water. Recovery of 14C was 99.2 ± 0.8% (six replicates), and 97.4 ± 1.2% of the recovered 14C was in the 67% acetonitrile eluate. The acetonitrile eluates were concentrated to dryness and dissolved in 15% acetonitrile for HPLC. In several cases, partitioning with methylene chloride was excluded and the 70% acetonitrile extracts were concentrated and fractionated directly on columns containing 3 g of C18 liquid chromatography packing.
Chromatography
Substrates and metabolites were analyzed or purified by HPLC on 3.9 mm x 30 cm columns of 10 mm C18 mBondapak (Waters Division, Millipore Corp.) eluted at 1.5 ml/min. Solvent System A consisted of a linear gradient from aqueous 25% acetonitrile/1% glacial acetic acid to aqueous 55% acetonitrile/1% glacial acetic acid in 30 min. Solvent System B consisted of a 30 min isocratic elution with aqueous 20% acetonitrile/1% glacial acetic acid, a 30 min linear gradient to aqueous 40% acetonitrile/1% glacial acetic acid, and a 10 min linear gradient to 99% acetonitrile/1% glacial acetic acid. Analyses were terminated by washing the HPLC columns with a 99% organic solvent. In addition, various isocratic HPLC systems were employed as described under Results. Column effluents were monitored for radioactivity with a Model LB 503 Berthold radioactivity monitor equipped with a 400 ml solid scintillator flow cell and also for uv at 254 nm with a Pharmacia dual path monitor UV 2. Thin layer chromatography (TLC) was performed on silica gel plates (250 mm x 5 cm x 20 cm Anasil HF silica gel plates, Analabs, Norwalk, CT) developed in butanol:glacial acetic acid: water (12:3:5, v/v/v). Radioactivity on these chromatograms was detected with a Bioscan System 200 imaging scanner and nonradioactive compounds were detected by uv at 254 nm.
Purification of Metabolites for Mass Spectrometry
Chlorimuron ethyl and metabolites of chlorimuron ethyl eluted from the open column of C18 were further purified by HPLC using solvent Systems A and B. HPLC System A was used in the initial separation of chlorimuron ethyl and two nonpolar metabolites from the polar metabolites. The individual metabolites resolved by HPLC Systems A and B were further purified as discussed in conjunction with the identification of these metabolites.
Mass Spectrometry
Metabolites were analyzed by mass spectrometry (MS) with a Varian MAT CH 5DF mass spectrometer. For fast atom bom bardment mass Spectrometry (FAB MS), samples (1.5 to 5 mg) dissolved in a matrix of 0.2 ml glycerol, 0.1 ml methanol, and 10 mg oxalic acid was applied to a copper probe tip. An lonTek Saddlefield gun was used to produce a xenon atom beam. Electron impact mass spectra were obtained using a solid sample probe with an El source. Precise mass measurements were made by high resolution peak matching.
Nuclear Magnetic Resonance (NMR) Spectrometry
Proton NMR Spectrometry was performed on a Brucker AM 400 M Hz NMR spectrometer. The samples (25 mg) were dissolved in 30 ml of methanol d4 and tet ramethylsilane was used as the internal standard.
b Glucosidase Hydrolysis
Metabolites suspected of being glucosides were treated with b glucosidase Type II from almonds using the procedures rec ommended by the Sigma Chemical Co. Approximately 5.6 nmol of metabolite in 400 ml water was added to 100 ml of b glucosidase (5.6 enzyme units, 1 mg protein) in 0.5 N, pH 5.0, sodium acetate buffer and incubated at 30°C. After 6.5 hr, the solutions were acidified with 1.5 ml of 1% glacial acetic acid and partitioned twice with 4 ml portions of methylene chloride. The 14C in the aqueous and methylene chloride soluble fractions was measured and the methylene chloride soluble fractions were concentrated and analyzed by HPLC and/or mass spectrometry. Little or no hydrolysis was observed in no enzyme controls.
Results
Uptake and Metabolism of Chlorimuron Ethyl
[l4C] Chlorimuron ethyl was readily absorbed by the roots of intact corn seedlings treated by surface application of the herbicide to the roots. Uptake was nearly complete in 1 hr. When chlorimuron ethyl was applied to the leaves in a similar manner, very little uptake was observed after 24 hr (data not shown). [14C]Chlorimuron ethyl was metabolized rapidly in the roots of corn seedlings treated with 21 nmol chlorimuron ethyl/g fresh wt of roots. Metabolism occurred at a linear rate of 2.3 nmol/g fresh wt/hr and the half life of the absorbed chlorimuron ethyl was ca. 4 hr (Fig. 1). [14C]Chlorimuron ethyl was also metabolized rapidly in excised leaves that were partially immersed in a dilute aqueous solution of [14C] chlorimuron ethyl. Following pulse treatment for 2.5 hr, the concentration of 14C in the leaves, expressed as [14C] chlorimuron ethyl, was 4.58 nmol/g fresh wt leaves. Metabolism of chlorimuron ethyl in the leaves was not linear, but the rate of metabolism, based on the half life of the herbicide (ca. 0.9 hr), was estimated to be ca. 2.5 nmol/g fresh wt/hr (Fig. 1). Because of lack of uptake [14C] chlorimuron ethyl metabolism was not monitored in intact corn leaves. The data in Fig. 1 are not a true comparison of the relative abilities of the roots and the leaves to metabolize chlorimuron ethyl since the methods of treatment and the levels of [14C] chlorimuron ethyl in the roots and shoots were different. However, it is clear that chlorimuron ethyl is metabolized at an appreciable rate in both organs.
[14C]Chlorimuron ethyl appeared to be metabolized by similar routes in both the roots and the leaves of corn. Chlorimuron ethyl and seven or eight metabolites were detected in both the roots and the shoots 7 hr following exposure to the herbicide (Fig. 2). Only the metabolites from the roots were actually isolated and identified. In the roots, the average recovery of I4C 7 hr following treatment with [l4C] chlorimuron ethyl was 96.2%, and the bound residue accounted for only 1.4% of the total residue. In the leaves, the average recovery of 14C 3 to 8 hr following treatment was 93.2%, and the bound residue accounted for an average of 1.0% of the total residue.
Detection of Metabolites
Three nonpolar, radioactive compounds were detected by HPLC of the methylene chloride soluble extracts of roots from corn treated with [14C] chlorimuron ethyl (Fig. 3). Six polar, radioactive metabolites were detected by HPLC of the corresponding water soluble extracts (Fig. 4). A radioactive peak observed during the HPLC of the water soluble fraction in HPLC System B (Rt 25 min, Fig. 4) was resolved into two major metabolites by HPLC with water:
Chlorimuron Ethyl
The methylene chloride soluble product with a retention time of 30 min in HPLC System A (Fig. 3) accounted for 50.5% of the radioactivity in the roots of corn seedlings 7 hr following treatment with [14C]chlorimuron ethyl. It was identified as chlorimuron ethyl by comparison to the standard by HPLC in System A and by FAB MS and El MS.
Metabolite I
Metabolite I accounted for 18.5% of the total 14C or 40.5% of the l4C labeled metabolites detected in the roots 7 hr following treatment with chlorimuron ethyl. It was purified from the methylene chloride fraction by successive HPLC with System A (Rt 20 min), with methanol: water: acetic acid (49.5:49.5:1) (Rt 29 min) and with water: acetonitrile (70:30) (Rt 23.2 min).
The positive ion FAB MS of metabolite I is shown in Fig. 5A. The intense ion cluster at m/z 431/433 [MH]+ is consistent with the presence of one chlorine and a molecular weight of 430. The ion at m/z 397 correspends to the reductive dechlorination of the MH+ ion. Reductive dechlorination during positive ion FAB MS has been previously reported and was also observed in this study during FAB MS of chlorimuron ethyl. The ion fragments of metabolite I at m/z 219, 202, and 175 appeared to be due to fragmentation of the hydroxylated pyrimidine moiety (Fig. 5A) while ions at m/z 213 and 197 appeared to be due to the substituted phenyl ring. The electron impact mass spectrum of metabolite I contained ion fragments derived from the hydroxylated pyrimidine ring (m/z 175, 70%) and the phenyl ring (m/z 210, 100%; m/z 184/185, 60%; m/z 120, 50%; and mtz 104, 100%), but a molecular ion was not observed. It was concluded that metabolite I was N (4 chloro 5 hydroxy 6 methoxypyrimidine 2 yl) N (2 ethoxy carbonylbenzenesulfonyl) urea.
Metabolite II
Metabolite II accounted for 4.1% of the total 14C or 9.0% of the 14C labeled metabolites in the roots 7 hr following treatment with chlorimuron ethyl. It was purified from the methylene chloride fraction by chromatography with HPLC System A (Rt 10 min) and by isocratic HPLC with water: acetonitrile: acetic acid (79:20:1) (Rt 14.3 min). Metabolite II cochromatographed with standard 2 ethoxycarbonylbenzene sulfonamide and the FAB MS of metabolite II and 2 ethoxycarbonylbenzene sulfonamide were nearly identical. The FAB MS of metabolite II is shown (Fig. 6). The spectra of metabolite II and the standard were characterized by a series of quasi molecular ions at m/z 230, 252, 268, 322, 344, and 360 that established the molecular weight of metabolite II to be 229. The intense ions at m/z 213 and 184 confirmed the identity of metabolite II to be 2 ethoxycarbonylbenzene sulfonamide.
Metabolite III
Metabolite III accounted for 1.5% of the total I4C or 3.3% of 14C labeled metabolites detected in the roots 7 hr following treatment with chlorimuron ethyl. It was isolated from the roots of corn plants that had been treated with [14C] chlorimuron ethyl and with [14C] chlorimuron ethyl plus BAS 145 138 safener. Based upon chromatographic evidence obtained with HPLC System B, it was concluded that metabolite III was produced by both treatments. Metabolite III was isolated by successive HPLC with System B (Rt 25 min) by isocratic HPLC with water: acetonitrile: acetic acid (83:16:1) and by a final step with HPLC System B. Metabolite III was contaminated with metabolite VI after the initial purification with HPLC System B, but these metabolites were resolved by HPLC with water: acetonitrile: acetic acid (83:16:1) (metabolite III, Rt 54 min and metabolite VI, Rt 64 min).
Metabolite III was hydrolyzed in a 91% yield by b glucosidase. The methylene chloride soluble, radioactive product from this hydrolysis was tentatively characterized as metabolite I by HPLC System A. Subsequently, metabolite I was converted to metabolite III in vivo in corn and an enzyme preparation from corn hydrolyzed metabolite III to metabolite I.
The FAB MS of metabolite III was characterized by a series of quasi molecular ions at m/z 593/595, 615/617, and 631/633 (Fig. 5B). These ions were consistent with the presence of one chlorine and a molecular weight of 592. The intense ion at m/z 431 (corresponds to the loss of the glucose moiety) and the similarity of this FAB MS to the FAB MS of metabolite I in the region from m/z 100 to 431 are evidence that metabolite III is the corresponding glucoside of metabolite I. The precise mass of the ion at m/z = 202.00200 agreed to within 0.0001 mass units to the calculated mass for the ion shown in Fig. 5B and confirmed its elemental composition. The ions at m/z 559 and 397 appear to be reductively clechlorinated forms of the ions at m/z 593 and 431, respectively. Metabolite III was concluded to be N (4 chloro 5 [O b D glucosyl] 6 methoxy pyrimidine 2 yl) N (2 ethoxycarbonylbenzenesulfonyl) urea.
Metabolite IV
Metabolite IV accounted for 3.4% of the total 14C or 7.4% of the l4C labeled metabolites detected in corn roots 7 hr following treatment with chlorimuron ethyl. It was purified from corn that had been treated with [14C] chlorimuron ethyl or with [14C) chlorimuron ethyl plus BAS 145 138. Based upon analysis with HPLC System B, it was concluded that metabolite IV was present in extracts from both treatments. Metabolite IV was purified by successive chromatography with HPLC System B (Rt 48 min), isocratic HPLC with methanol: water.glacial acetic acid (49.5:49.5:1) (Rt 19 21 min), and by HPLC System B (Rt 48 49 min).
When metabolite IV, purified as described above, was chromatographed in HPLC System B in the absence of 1% acetic acid, four radioactive peaks were eluted between Rt 24 and 41 min. When these peaks were combined, concentrated, and chromato graphed in HPLC System B in the presence of 1% acetic acid, a single peak at 48 min was observed. Rechromatography of this peak in HPLC System B in the absence of 1% acetic acid resulted in the elution of multiple peaks. A similar phenomena occurred when corn tissue was spiked with the synthetic 14C labeled GSH conjugate of chlorimuron ethyl, extracted, and analyzed by HPLC System B in the presence or absence of 1% acetic acid. Metabolite IV and the synthetic GSH conjugate chromatographed with retention times of 48 min in HPLC System B. The homoglutathione conjugate of chlorimuron ethyl, isolated from soybean root, also chromatographed in HPLC System B with a retention time of 48 min.
The FAB MS of the synthetic GSH conjugate and metabolite IV are compared (Figs. 7A and 7B). Metabolite IV produced a series of quasi molecular ions at m/z 686, 708. and 724; but the synthetic GSH conjugate, which apparently contained less salt, produced a single quasi molecular ion at mlz 686. The key fragmentation ions in the spectrum of the standard GSH conjugate (m/z 557, 431, 413, 225, 184, and 158) were also present in the spectrum of metabolite IV. The FAB MS of the homoglutathione conjugate, isolated from soybean, contained a series of quasi molecular ions at m/z = 738, 722, and 700 (14 mass units higher than metabolite IV) and fragmentation ions comparable to metabolite IV (Fig. 7C). ion fragments from the homoglutathione conjugate that contained the b alanine moiety occurred at 14 mass units higher than ion fragments from metabolite IV that contained the glycine moiety; therefore, it was possible to determine which ion fragments contained the C terminal b alanine or glycine moieties. It was concluded that metabolite IV was N (4 [S glutathionyl] 6 methoxypyrimidine 2 yl) N (2 ethoxycarbonyl benzenesulfonyl) urea.
Metabolite V
The FAB MS of metabolite IV contained two moderately intense ions (m/z 500,16%; and m/z 522, 6%) which were not in the spectrum of the standard GSH conjugate and which had no equivalent at the same mass or at 14 mass units higher in the spectrum of the homoglutathione conjugate isolated from soybean (Figs. 7A, 7B, and 7C). Subsequently, TLC of this preparation of metabolite IV showed that two radioactive metabolites were actually present, metabolite IV (the GSH conjugate, Rf 0.42, 80%) and a minor component, metabolite V (Rf 0.55,20%). The Rf of metabolite V was consistent with that of a cysteine conjugate. When metabolites IV and V were isolated from corn roots that had been incubated with chlorimuron ethyl for 11 hr instead of 7 hr, the ratio of metabolite IV to V changed from 4:1 to 2:1. It was concluded that the ions at m/z 500 and 522 (Fig. 7B) were the MH+ and MNa+ quasi molecular ions of metabolite V, N (4 [S cysteinyl] 6 methoxypyrimidine 2 yl) N (2 ethoxy carbonylbenzene sulfonyl)urea.
Metabolite VI
Metabolite VI accounted for 3.4% of the total 14C or 7.4% of the 14C labeled metabolites in the roots 7 hr following treatment with chlorimuron ethyl. Metabolite VI co chromatographed with metabolite III in HPLC System B, but after HPLC System B, it was resolved from metabolite III by isocratic HPLC with water.acetonitrile:acetic acid (83:16:1) (metabolite III, Rt 54 min and metabolite VI, Rt 63 min). Metabolite VI was rechromatographed on HPLC System B prior to analysis by FAB MS and NMR.
The FAB MS of this metabolite was characterized by the following series of quasi molecular ions: [MNa] +, 724, 6%; [MK] +, 740, 8%; [MNaK H] +, 762, 6%; [MK2 H], 778, 5%; [MNa2K H2] +, 784, 6%; and [MNaK2 H2], 800, 7%. These ions are consistent with a molecular weight of 701 and might be expected from either a hydroxylated GSH conjugate or the S oxide of the GSH conjugate of chlorimuron ethyl. Attempts to prepare the S oxide of the GSH conjugate of chlorimuron ethyl by oxidation of the synthetic GSH conjugate with hydrogen peroxide or with m chloroperbenzoic acid were not successful; however, these methods frequently fail with aryl GSH conjugates. A product with the same HPLC System B retention time as metabolite VI was prepared in low yield (4%) by the reaction of N (4 chloro 5 hy droxy6 methoxypyrimidine 2 yI) N (2 ethoxycarbonylbenzene sulfonyl) urea with GSH under the same conditions as used in the synthesis of the GSH conjugate of chlorimuron ethyl. This synthetic product, metabolite VI, metabolite IV, the homoglutathione conjugate of chlorimuron ethyl, S (2.4 dinitrophenyl) glutathione, and the GSH conjugate of propachlor have TLC Rf values of 0.38 to 0.42 in butanol:acetic acid :water (12:3:5). The cysteine conjugates of propachlor and chlorimuron ethyl have Rf values of 0.55 in this TLC system.
Proton NMR spectra (400 MH) were obtained from metabolite VI, synthetic metabolite IV (GSH conjugate), chlorimuron ethyl, primisulfuron, and three other substituted pyrimidinyl derivatives. Preliminary evidence indicated that metabolite VI was hydroxylated in the 5 position of the pyrimidine ring. The chemical shift of the H 5 pyrimidine proton of chlorimuron ethyl and metabolite IV was 6.29 ppm and in four other substituted pyrimidinyl standards it ranged from 6.06 to 6.33 ppm. In contrast, no protons with chemical shifts of 5.4 to 6.4 ppm were observed in the spectrum of metabolite VI. further suggesting that the 5 position of metabolite VI was hydroxylated. The chemical shifts of the protons on the phenyl rings of these metabolites and standards were from 7.5 to 8.0 ppm due to the sulfonyl and carboxylic ester substituents on the phenyl ring. Therefore, there were no interfering bands in this region of the spectra. That portion of the spectrum of metabolite VI due to the peptide side chain (1.9 to 4.6 ppm) was consistent with a GSH conjugate and was similar to the spectrum of metabolite IV, but this portion of the spectrum was not sufficiently resolved to be unambiguous. However, the methylene protons attributed to the cysteinyl residue of a GSH conjugate, usually observed between 2.8 and 3.1 ppm were present at 2.8 ppm in the spectra of both metabolites IV and VI. If metabolite VI was the S oxide of the GSH conjugate, a significant chemical shift of these methylene protons would be expected. Therefore, based on synthetic evidence, MW data from FAB MS, chromatographic comparisons with standards, and NMR spectrometry, it was concluded that metabolite VI was N (5 hydroxy 4 [S glutathionyl] 6 methoxypyrimidine 2 yl) N (2 ethoxycarbonylbenzenesulfony) urea.
Metabolite VII
Metabolite VII accounted for 1.4% of the total 14C or 3.4% of the MC IabeIed metabolites present in the roots 7 hr following treatment with chlorimuron ethyl. Based on HPLC evidence, metabolite VII appeared to be present in corn treated with either chlorimuron ethyl or chlorimuron ethyl plus BAS 145 138. It was purified from extracts pooled from both types of treatment. Purification was accomplished by two successive chromatographic steps with HPLC System B followed by HPLC with System B that had been modified by elimination of the glacial acetic acid. The retention time of metabolite VII in HPLC System B with 1% glacial acetic acid was 54 min and in the absence of 1% glacial acetic acid it was 52 min.
Metabolite VII appeared to be resistant to mild base hydrolysis. The retention time of metabolite VII in HPLC System B was not altered by treatment of metabolite VII with 0.5 N ammonium hydroxide at 60°C for 60 min. However, metabolite VII was hydrolyzed by b glucosidase from almonds. After hydrolysis for 6.5 hr, 90% of the radioactivity was soluble in methylene chloride. The retention time of the radioactive hydrolysis product in HPLC System A was 25.6 min and the retention times of metabolite I and chlorimuron ethyl in HPLC System A were 20.8 and 29 min, respectively. The negative ion FAB MS of metabolite VII was characterized by quasi molecular ions at 591/593 and 683/685 that confirmed the presence of one chlorine and a molecular weight of 592 while ions at m/z 244 and 270 were consistent with a metabolite hydroxylated in the phenyl ring (Fig. 5C). These data are consistent with a simple O b D glucoside formed by hydroxylation and conjugation of the phenyl ring of chlorimuron ethyl.
The radioactive hydrolysis product of metabolite VII, obtained by treatment with b glucosidase and extraction with methylene chloride, was purified by HPLC (1.9 u,g) and analyzed by El MS. A molecular ion was not observed in the El MS, but the presence of the following ion fragments is consistent with a metabolite hydroxylated in the phenyl ring:[HO C6H3(COOC2H5)(SO2NH2)+ at m/z 245 (10%), [HO C6H3(CO)(SO2NHCONH2)+ at m/z 243 (11%), [HO C6H3(CO) (SO2NCO)] + at m/z 226 (28%), [HO C6H3(COOH)(SO2NH2)] + at m/z 217 (10%), [HO C6H3(CO)(SO2NH2)+ at m/z 200 (43%), and [HO C6H3(CO)(SO2)]+ at m/z 184/185 (20%). The absence of hy droxylation in the pyrimidine ring was indicated by the following ions: [NH2 C4N2H (Cl) (OCH3)] + at m/z 159 (22%) and [NH2 C4N2H2 (Cl)] + at m/z 129 (20%). The interpretation of this El spectrum was based on a comparison with the El MS of chlorimuron ethyl, metabolite 1, 2 amino 4 chloro 6 methoxy pyrimidine, and 2 ethoxycarbonylbenzene sulfonamide (metabolite II). There was not sufficient metabolite to determine the position of hydroxylation on the phenyl ring by proton NMR. Metabolite VII was concluded to be N (4 chloro 6 methoxypyrimidine 2 yl) N (2 ethoxycarbonyI? [O b D glucosyl] ben zenesulfonyl) urea.
Superoxide Dismutase Inhibition by a Terthienyl
Over the past few years the secondary plant metabolite, a terthienyl (2,2 :5,2 terthiophene), has been a subject of consid erable interest because of its involvement in a wide variety of phototoxic actions. The phototoxic effect toward mosquito larvae and its potential application as a larvicide make its mechanism of action a topic of considerable importance. It has been reported that a terthienyl acts by Type II photodynamic action by sensitizing singlet oxygen. Recently, it has been shown that a terthienyl generates a superoxide anion radical in aqueous media. Mosquito larvae treated with a terthienyl in the presence of long wave ultraviolet light or sunlight show an accumulation of this compound in the anal gills and occurrence of gill membrane damage as a consequence of a terthienyl treatment as can be seen by the halide leakage technique. Cell/ membrane proteins are also shown to be the targets of toxicity of a terthienyl. The inhibition of glucose 6 phosphate dehydrogenase and malate dehydro genase and acetylcholine esterase are a few examples.
Another vital enzyme reported to be present in the organs rich in oxygen, viz. rete mirable and gas gland epithelium of marine fishes i.e., the superoxide dismutase, has never been located in the anal gills of the mosquito larvae. Once it was established that a terthienyl toxicity was oxyradical mediated, the significance of superoxide dismutase in the anal gills became apparent. Recently, a histo chemical method has been introduced by Laloraya et al. modifying the original negative staining method of Beauchamp and Fridovich. Using this technique, nitroblue tetrazolium, a free soluble yellow compound, upon reduction forms an intensely blue product (formazan) that is virtually insoluble in aqueous solutions and quickly precipitates, making it ideal for staining. In this negative staining technique, the enzyme superoxide dismutase dismutates the superoxide radical generated by riboflavin, thus making it unavailable for the electron acceptor nitroblue tetrazolium and thereby inhibiting the diformazan formation; whereas, in the absence of superoxide dismutase, the radical generated by riboflavin promotes diformazan formation. Thus, the achromatic zones represent areas showing superoxide dismutase activity, because of which this is called a negative staining method. In this study, we report our pioneer attempt to localize the enzyme superoxide dismutase in the anal gills of mosquito larvae and its fate after exposure to a terthienyl, successfully applying this technique.
MATERIALS AND METHODS
Reagents. Nitroblue tetrazolium, riboflavin, and Diethyldithio carbamic acid were obtained from Sigma Chemical Co. N,N,N,N, Tetramethylethylene diamine was from Loba Chemical Co., India, and a terthienyl was a gift from Dr. Arnason, University of Ottawa, Canada.
All of the four mosquito larval stages used for this study were reared in the laboratory and each stage was studied separately. Two different sets of experiments were performed. The first one involved superoxide dismutase activity and the second one determined superoxide dismutase activity after a terthienyl treatment.
Histochemical test for superoxide dismutase activity. Mosquito larvae were taken and cleaned with double distilled water, taking care not to injure the larvae. They were treated initially with 2.5 x 10-3 M nitroblue tetrazolium for 30 min. These were then transferred to an incubation medium containing 0.036 M potassium phosphate (pH 7.8), 0.028 M tetramethyl ethylene diamine, and 2.8 x 10-5 M riboflavin for 20 min. These larvae were then illuminated for 1 hr with a 15 W Phillips fluorescent lamp. For negative controls the incubation medium also contained diethyl dithiocarbamic acid, a known inhibitor of superoxide dismutase at a final concentration of 2 x 10-2 M, with all the abovementioned compounds.
Histochemical test for superoxide dismutase activity after a terthienyl treatment. Mosquito larvae were cleaned with double distilled water. They were then treated with a terthienyl at a concentration of 20 ppb and illuminated with an ultraviolet light lamp (wavelength, 366 nm) (Model UVL 21 Black Ray lamp, Ultraviolet Proc., Inc., San Gabriel, CA) for 1 hr. These larvae were then processed for the histochemical test outlined above.
All of the photochemical reactions were carried out in the absolute dark in aluminium foil covered petri dishes. After illumination, larvae were rinsed with 0.036 M potassium phosphate (pH 7.8) and subsequently with double distilled water. The anal gills were separated from the larvae and were mounted in glycerol jelly and viewed under bright field optics. Photographs were taken using ORWO NP 22 (ASA 125) black and white film. Exposure time was computed by a Nikon UFX II camera monitor unit and the film was processed using a fine grain developer and Amfix fixer (May & Baker, India).
Discussion
Discovery of the light dependent toxicity of a terthienyl against many numbers of organisms led to a number of studies whose main aim was to (i) determine whether the phototoxic reactions were oxygen dependent, (ii) establish whether the photodynamic reaction is Type I or Type II, and (iii) locate a site(s) affected in damaged cells. However, the main emphasis has been given to Type II photodynamic reactions, a consequence of the long lived triplet excited molecule and the high quantum yield of singlet oxygen formation. Recently, however, the recognition that Type I photodynamic reactions also operate in vitro has extended the range of possible mechanisms of toxicity in vivo. Unfortunately, no definitive experiments have yet compared the participation of Type I and Type II processes for a specific photo toxic effect in vivo. Tuveson et al. attempted such a comparison using several strains of Escherichia coli varying in repair capacities and catalase proficiency. Their study demonstrated that (i) the specific repair mechanisms tested did not play a role in protecting the organism from inactivation and (ii) the presence of H2O2 (which would have been generated from superoxide radical through a Type I photodynamic reaction) was of no detectable importance in cell lethality. In a separate study, an E. coli strain engineered to produce carotenoid pigments in the membrane exhibited virtually complete resistance to the photo sensitized damage induced by a terthienyl. This result strongly suggested that singlet oxygen generated by a terthienyl was lethal to the cells in vivo and that its target was the cell membrane.
Viruses are inactivated by a terthienyl in a light dependent manner, but only if they contain a membrane. Similarly, a terthienyl photosensitizes the conversion of supercoiled into open circular pBR322 DNA in vitro in a manner which appears independent of oxygen and which does not involve singlet oxygen, superoxide, or hydrogen peroxide when oxygen is present.
Parathion Metabolism by Soybean Lipoxygenase
Parathion (O,O diethyl p nitrophenyl phosphorothionate) is a widely used or ganophosphorus insecticide. It is generally accepted that the toxicity of parathion is due to the irreversible inhibition of acetyl cholinesterase in the nervous system. Parathion itself is a poor inhibitor of the acetyl cholinesterase but it is bioactivated by oxidative desulfuration to its oxygen analogue paraoxon, which is a potent inhibitor of acetytcholinesterase. Metabolism of parathion by the liver microsomal cytochrome P 450 dependent monooxygenase and serum A esterases has been well documented in various animal species.
Lipoxygenase (EC.1.13.11.12) is a group of ubiquitous enzymes found in several plants and animals. Lipoxygenase catalyzes the incorporation of molecular oxygen into polyunsaturated fatty acids to yield corresponding hydroperoxy fatty acids which subsequently break down either enzymatically or non enzymatically into hydroxy fatty acids, leukotrienes, lipoxins, and other products. Lipid peroxides are known to be produced in blood cells such as reticulocytes, leukocytes, and platelets and in tissues like lung and brain by lipoxygenases. Various lipid radicals are reactive compounds that support oxidation of different xenobiotics either directly or through mediation of enzymatic systems. Our laboratory has demonstrated the cooxidation of several hydrogen donors, epoxidation of benzo(a)pyrene 7,8 dihydrodiol and aldrin and sulfoxidation of thiobenzamide by lipoxygenase in the presence of linoleic acid. In addition to 13 hydroperoxylinoleic acid (13 HPOD) the hydroperoxides of other polyunsaturated fatty acids as well as inorganic peroxide such as hydrogen peroxide were shown to support xenobiotic oxidation. In this study, we examined the catalytic potential of lipoxygenase to mediate desulfuration and dearylation of parathion using a highly purified enzyme from soybean as a model.
MATERIALS AND METHODS
Chemicals. Soybean lipoxygenase type V (567,400 Sigma units/mg protein; MW 108 kDa), acetylcholinesterase (electric eel, sp act 355 units/mg protein), acetylthiocholine, 5,5, dithiobis 2 nitrobenzoic acid (DTNB), [14C]parathion (uniformly ring labeled, sp act 20.0 mCl/mmol), linoleic acid, nor dihydro guaiaretic acid (NDGA), phenidone, paraoxon, and p nitrophenol were the products of Sigma chemical company. Parathion (98.76%) was purchased from City Chemical Company. Petroleum ether, diethyl ether, and acetic acid were purchased from Fisher Scientific Company. Scintillation counting fluid was the product of ICN Biomedicals, Inc. 13 HPOD was biosynthe sised by incubating linoleic acid with soybean lipoxygenase and the fatty acid hydroperoxide was separated by thin layer chromatography.
Cholinesterase inhibition studies. Paraoxon production was estimated indirectly from the inhibition of acetylcholinesterse. Electric eel acetylcholinesterase was used as exogenous source of the enzyme to trap the oxon produced during lipoxygenase catalyzed metabolism of parathion. Parathion (25.0 mM), linoleic acid (1.0 mM), lipoxygenase (20.0 nM), and acetylcholinesterase (0.05 units) were incubated for 5.0 min at room temperature in 0.1 M phosphate buffer, pH 8.0. At the end of the incubation period, the reaction mixture was assayed directly for residual acetylcholinesterase activity by adding acetylthiocholine (1.5 mM) and DTNB (1.0 mM). The control incubation medium contained all the components except lipoxygenase. The results were compared with the standard curves of acetylcholinesterase inhibition by paraoxon to quantify the amount of paraoxon produced.
Radiometric method. Metabolism of [14C]parathion was studied in the presence of linoleic acid or 13 HPOD and lipoxygenase. The standard assay mixture, except otherwise stated, contained 50.0 mM Tris buffer (pH 8.0), 25.0 mM parathion (18.75 mM unlabeled parathion plus 6.25 mM [14C]parathion), 1.0 mM linoleic acid, and 20.0 nM enzyme in a final volume of 1.0 ml. The reaction mixture was incubated for 5 min at 37°C. The reaction was arrested with the addition of 250 ml of 10% trichloroacetic acid. The tubes were then frozen at 70°C until analyzed. In some experiments, linoleic acid was replaced with the desired concentration of 13 HPOD in the incubation mixture.
Quantitative analysis. The metabolites accumulated in the reaction mixture were extracted with 1.0 ml of methylene chloride. The organic phase was separated by centrifugation and concentrated under nitrogen. An aliquot of the concentrated extract was spotted on silica gel coated thin layer plate (Silica G pre coated plastic sheets from Brinkman Instruments Co). The plate was developed to a height of 15.0 cm using the solvent system containing petroleum ether, ethyl ether, glacial acetic acid (55:45:5 v/v). The Rf values observed in this solvent system were 0.8 for parathion, 0.6 for p nitrophenol, and 0.4 for paraoxon. The plate was removed from the developing chamber and dried and the locations of paraoxon and p nitrophenol were visualized under ultraviolet light. The spots were marked; the portions of the plate containing the metabolites were cut and placed in scintillation vial containing 10.0 ml of scintillation fluid and counted in a Beckman liquid scintillation counter. The data reported are corrected for the quenching and counting efficiency.
Dioxygenase activity. Dioxygenase activity of lipoxygenase was measured spectrophotometrically as an increase in the absorbance at 234 nm as described previously using assay medium containing 1.0 mM linoleic acid, 20.0 nM lipoxygenase enzyme, and the selected concentration of parathion in 50.0 mM Tris buffer (pH 8.0). The specific activity was calculated using an extinction coefficient of 25 mM-1 cm-1 for 13 HPOD. The lipoxygenase inhibitors NDGA and phenidone (at desired concentration) were preincubated with the enzyme for 2.0 min prior to the addition of parathion and linoleic acid.
Oxygen uptake studies. Oxygen uptake during metabolism of parathion was monitored with Clark type oxygen electrode and a Biological oxygen monitor (Yellow Spring Instruments Co., Inc) at 30°C. The reaction medium (final volume of 2.0 ml) contained 50.0 mM Tris buffer, pH 8.0, 1.0 mM linoleic acid, desired concentration of parathion, and 20.0 nM enzyme. NDGA and phenidone (at indicated concentration) were preincubated with the enzyme for 2.0 min prior to the addition of parathion and linoleic acid.
After several pilot runs, each experiment was repeated two or more times and the results of a typical experiment are presented in the tables and figures. In most cases, the experimental variation was about 10%.
RESULTS
The data in Table 1 show that lipoxy genase is capable of activating parathion to the more potent acetylcholinesterase inhibitor, paraoxon. The data also show that the amount of oxon produced was dependent on the concentration of parathion, linoleic acid, and the enzyme (Table 1). The maximal rate of metabolism was observed when incubation medium contained 25.0 mM parathion, 1.0 mM linoleic acid, and 20.0 nM soybean lipoxygenase. A decrease in the rate of paraoxon formation was evident when relatively higher concentrations of parathion, linoleic acid, and the enzyme were employed (Table 1).
[14C]Parathion was used in all other experiments. It was meta bolized to paraoxon and p nitrophenol in the presence of lipoxygenase and linoleic acid (Table 2). A measurable nonenzymatic conversion of parathion to paraoxon and p nitrophenol was noted in the presence of linoleic acid alone. However, a several fold increase in the formation of these metabolites was observed in the presence of the enzyme. The production of metabolites was in the ratio of approximately 1:2 for paraoxon and p nitro phenol, respectively.
The data on the effects of experimental conditions on parathion metabolism are presented in Figs. 1 and 2. As shown in Fig. 1, the rate of both paraoxon and p nitrophenol production increased with the incubation time up to 5 min and leveled off thereafter. The amount of products of desulfuration and dearylation reactions increased when the concentration of enzyme present in the incubation medium was increased up to 30 nM (Fig. 2A). The rate of metabolism of parathion was dependent upon the concentration of parathion (Fig. 2B). A lack of proportionality in the generation of metabolites was noted at parathion concentration >25 mM and this may be due to limited solubility of parathion or enzyme saturation. Both paraoxon and p nitrophenol production increased with an increase in linoleic acid concentration and maximum rate of production of these metabolites was noted when 1.0 mM linoleic acid was used (Fig. 2C). A marked decline in specific activity for paraoxon and p nitrophenol was observed when >1.0 mM linoleic acid was present in the reaction media. The data in Fig. 2D show the effect of pH of incubation medium on the parathion metabolism. P nitrophenol formation increased with increase in pH. A broad pH response curve was noted for paraoxon and maximum rate was observed at pH 8.0.
13 HPOD also supported the lipoxygenase catalyzed parathion metabolism in the absence of linoleic acid (Table 2). However, as compared to the linoleic acid supported reaction, significantly lower rates of enzymatic desulfuration and dearylation were observed when the reaction medium was supplemented with 13 HPOD (Table 2). The rate of parathion metabolism declined when higher (>22.5 mM) concentrations of 13 HPOD were used.
The data on the dioxygenase activity of lipoxygenase and oxygen uptake (Table 3) revealed a small change (4 5%) when up to 25.0 mM of parathion was present in the reaction medium. Higher parathion concentrations caused a marked inhibition of oxygen uptake and dioxygenase activity. A slightly greater than additive response in terms of inhibition of dioxygenase activity and oxygen consumption was observed when either NDGA or phenidone, the inhibitors of lipoxygenase, was present in the reaction medium in addition to parathion. As shown in Figs. 3A and 3B, both NDGA and phenidone, at indicated micromolar concentration, inhibited the desulfuration and dearylation reactions by 60 to 90%.
Depolarization Studies of Pyrethroids
Primary target sites of insecticidal pyrethroids are the nervous system of insects. Pyrethroids have various effects on excised nerve preparations. Nerve cord preparations discharge repetitively upon electrical stimulus after exposure to a certain class of pyrethroids. Other pyrethroids suppress the action potential. We have quantitatively analyzed the relationships between wholebody symptomatic activities against American cockroaches and house flies and nerve activities measured extracellularly from nerve cords excised from the cockroaches. Both repetitive and blocking neurophysiological effects are important in the toxicological effects on these insects. Depolarization has been observed intracellularly in single axons. We have measured this activity with crayfish axonal membranes used as model for the study of the mode of action of pyrethroids. Quantitative analysis has shown that the higher the neurophysiological activity, the higher the whole body symptomatic activities.
To examine more directly the origins of the symptoms of intoxication of American cockroaches and the nerve effects measured extracellularly, we measured the depolarizing potency of pyrethroids with cockroach giant axons using intracellular microelectrodes. We suggest that membrane depolarization is a major factor for block of nerve conduction leading to intoxication of insects.
MATERIALS AND METHODS
Compounds. Test compounds are listed in Table 1. Compounds 1 7 are the benzyl esters of the acid moiety of kadethrin (compound 17) and are called benzyl kadethrates in this paper. Compounds 1 6 and 8 17 are the same as those used previously (3, 4, 6, 11). (RS) a Cyano meta phenoxybenzyl kadethrate (compound 7) was prepared from the corresponding acid chloride and (RS) a cyano meta phenoxybenzyl alcohol in the presence of pyridine in dry benzene at room temperature. The structure was confirmed by PMR1 and elementary analyses. The acid moiety of compounds 1 7 and 17 had the (1R) cis configuration and that of compounds 8 16 had the (1R) trans configuration. The alcohol moiety of compounds 7, 10, and 14 was not stereochemically defined. The structures of the acid and alcohol moieties of the esters are shown in Tables 2 and 3. Pyrethroids and veratridine were stored as methanol solutions in a freezer at 20°C until use. Tetrodotoxin (TTX) was purchased from Sigma and stored as an aqueous solution in a refrigerator at 5°C.
Measurement of the membrane potential in cockroach giant axons. Central nerve cords excised from the adult male American cockroaches (Periplaneta americana) were used as the material. Unless otherwise noted, the nerve cord with the intact sheath was used. In some experiments, the sheath was partially broken between the 4th and 5th ganglia to give a desheathed nerve preparation. The preparations were immersed in a saline solution containing 210 mM NaCl, 2.9 mM KCI, 1.8 mM CaCl2, and 2.0 mM phosphate buffer (1.8 mM Na2HPO4 and 0.2 mM KH2PO4, pH 7.2). The membrane potential of the giant axon between the 4th and 5th ganglia was measured intracellularly. Glass capillary microelectrodes filled with 2 M potassium acetate with a resistance of 4 8 MW and 10 20 Mft for the desheathed and intact nerve preparations, respectively, were used through out the experiments. Nerve preparations with a resting potential more negative than 75 mV were selected for the experiments. Test solutions were prepared by addition of the stock solution of a test chemical into the saline solution. The final concentration of methanol in the solution was 1% or lower (v/v). Methanol at this concentration had no effect on the resting potential. Compounds were applied to the nerve preparation by external perfusion. Measurements were done at room temperature (23 ± 1°C).
RESULTS
Depolarization of the Resting Membrane
Kadethrin depolarized both the desheathed and intact nerve membranes (Figs. 1A and IB). With additional treatment with TTX (1.0 x 10-5 M), the membrane potential in the desheathed preparation recovered to a level more negative than that observed before treatment with kadethrin (Fig. 1A). TTX alone slightly hyperpolarized the axonal membrane in desheathed nerve cords that were not treated with kadethrin (Fig. 1C). In the intact nerve preparation (Fig. IB), however, the rate of recovery was much slower than that in the desheathed preparation, indicating that TTX did not readily penetrate the sheath of the cockroach nerve cord, probably because the toxin is highly hydrophilic. By increasing the duration and intensity of exposure to kadethrin, the membrane potential did not completely recover from the depola rization when the TTX was added (Fig. 2A). This toxin insensitive component of the depolarization was not observed when the compound was applied at a low concentration (Fig. 2B).
Depolarizing Activity in Axonal Membranes
To evaluate the depolarizing potency, we measured the maximum depolarization from the original resting potential within 60 min after the start of treatment with a compound. Usually, the membrane potential attained a steady state by this time even in the intact nerve preparations as shown in Fig. 2 for kadethrin with a desheathed nerve preparation. The dose response relationship for compound 3 with intact nerve preparations is shown in Fig. 3. From such a curve for each compound, the DC10 value (M), which is the concentration needed for membrane depolarization by 10 mV, was estimated graphically. Log (l/DC10) was used as an index of the depolarizing potency. The log (l/DC10) values are listed in Table 1.
Relationship between Neuroblocking and Excitatory Potencies and Depolarizing Potency
Previously, we measured extracellularly the neuroblocking and excitatory potencies of pyrethroids in terms of log(l/MBC) and log(1/MEC), respectively. The MBC and MEC are the minimum concentrations to block nerve conduction and to induce repetitive discharges in response to a single stimulus in the central nerve cord of American cockroaches. These values are plotted against the depolarizing potency in Fig. 4. The neuroblocking activity is almost linearly related to the depolarizing activity (Fig. 4B). There is, however, no clear cut relationship between the repetitive neu roexcitatory activity and the depolarizing activity (Fig. 4A).
Equation [1] was derived for the neuro blocking and depolarizing activities with the least squares method.
In this and the following equations, n is the number of data points, s is the standard deviation, r is the correlation coefficient, and F is the ratio of the variances of the observed and calculated values. The figures in parentheses are the 95% confidence intervals. Equation [1] was improved by the addition of the log P term, where P is the 1 octanol/water partition coefficient, to give Eq. [2].
Equation [2] indicates that the higher the depolarizing activity and the greater the hydrophobicity of compounds, the higher is the neuroblocking activity. The log (1/MBC) values calculated by Eq. [2] are listed in Table 1.
Correlation was not found between the neuroexcitatory and depolarizing activities even when the log P term was added.
Relationship between the Insecticidal Potency and the Depolarizing Potency
The insecticidal activity of several sets of pyrethroids against the American cockroaches have been represented in terms of log(1/MLD), where MLD is the minimum lethal dose (in moles) to kill the cockroaches within 24 hr after injection of compound together with piperonyl butoxide (50 mg/insect) and NIA 163882 (50 mg/ insect) that inhibit oxidative and hydrolytic metabolic mechanisms, respectively. The activity values are listed in Table 1. The activity was measured under conditions that minimized metabolic mechanisms; under these conditions, the insecticidal activity should simulate the activity at the target sites. Variations in the lethal activity were quantitatively analyzed with variations in the depolarizing activity, giving Eq. [3] as the best correlation.
