effects of the herbicide chlorthiamid on the olfactory mucosa

T o x i c o l o g y _ le t ters

E L S E V I E R Toxicology Letters 76 (1995) 203-208

Effects of the herbicide chlorthiamid on the olfactory mucosa

Catarina Eriksson*, Eva B. Brittebo

Department of Pharmacology and Toxicology, Uppsala Biomedical Center. SLU, Biomedicum, Box 573, S-751 23 Uppsala, Sweden

Received 20 June 1994; accepted 20 September 1994

Abstract

Chlorthiamid (2,6-dichlorothiobenzamide) and its major metabolite 2,6-dichlorobenzonitrile are olfactory toxicants with a high in vivo covalent binding in the olfactory mucosa of mice. This study showed that the cytochrome P450 (P450) inhibitors, metyrapone and sodium-diethyidithiocarbamate, abolished the chlorthiamid-induced toxicity (12 mg/kg; 0.06 mmol/kg) in C57B1/6 mice suggesting a P450-dependent toxicity. Incubation of ll4Cl-labelled chlorthiamid with rat olfactory microsomes showed a low NADPH-dependent oxidative covalent binding which was only 3-fold higher than that in liver microsomes. Thus the results do not support a major in situ metabolic activation of chlorthiamid and it is suggested that metabolic activation of the major chlorthiamid metabolite (2,6- dichlorobenzonitrile) is responsible for most of the covalent binding and toxicity of chlorthiamid at this site in vivo. Thiobenzamide (16 mg/kg; 0.12 mmol/kg), a dechlorinated chlorthiamid-analog, induced no marked morphological changes in the olfactory mucosa demonstrating that chlorines in the 2,6-position are important for the chlorthiamid- induced toxicity at this site.

Keywords: 2,6-Dichlorothiobenzamide; Thiobenzamide; Dichlobenil; Herbicide; Olfactory mucosa; Cytochrome P450

1. Introduction

Chlorthiamid (2,6-dichlorothiobenzamide; Fig. 1) and its major metabolite 2,6-dichlorobenzonitrile (dichlobenil; Fig. 1) are broad-spectrum herbicides [1,2] which are highly toxic to the olfactory nasal mucosa in mice and rats following single i.p. doses [3-5]. The chlorthiamid- and dichlobenil-induced lesions are similar; first there is

S S N II It 111 C - NH 2 C - NH 2 C

© I II III

* Corresponding author, Tel.: 46 18 174 506; Fax: 46 18 504 144.

Fig. 1. Chemical structures of (I) chlorthiamid, (II) thiobenzamide and (III) dichlobenil.

0378-4274/95/$09.50 © 1995 Elsevier Science Ireland Ltd. All rights reserved SSD1 0378-4274(94)03211 -O

204 C. Eriksson, E.B. Brittebo / ToxicoL Left. 76 (1995) 203-208

necrosis of the Bowman's glands in the lamina propia followed by detachment of the neuroepithelium. No morphological changes have been observed in other regions of the respiratory system or in the liver following administration of these herbicides in mice and rats.

The tissue-specific toxicity of dichlobenil is related to an oxidative, cytochrome P450 (P450)- dependent metabolic activation, depletion of GSH and covalent binding of metabolites in the olfactory mucosa [3-7]. Since previous studies have shown that chlorthiamid is irreversibly bound to the olfactory mucosa in vivo and decreases the metabolic activation of dichlobenil in mouse olfactory mucosa in vitro [3,4] it was considered of in- terest to examine the activation of chlorthiamid in its target tissue. In the present study the covalent binding of [14C]-labelled chlorthiamid in olfactory and liver microsomes and the effects of some P450 inhibitors on the toxicity of chlorthiamid in the olfactory mucosa have been studied. In addition, the effects of thiobenzamide (Fig. 1), a hepatotoxic dechlorinated analog of chlorthiamid, in the olfactory mucosa have also been examined.

2. Materials and methods

2.1. Chemicals Unlabelled chlorthiamid and dichlobenil were

obtained from Aldrich-Chemic (Steinheim, Ger- many). Thiobenzamide and sodium-dithionite were obtained from E. Merck (Darmstadt, Ger- many). 2,6-Dichlorothiobenz[14C]amide, with a specific activity of 0.41 mCi/mmol, was a gift from Prof. J. Bakke, Fargo, USA. 2,6-Dichlorobenzo- nitrile-[ring-14C], with a specific activity of 16.7 mCi/mmol, was obtained from Sigma Chem. Co. (St. Louis, MO). 2,6-Dichlorobenzonitrile-[14C - nitrile], with a specific activity of 9.6 mCi/mmol, was obtained from Amersham pie (Little Chalfont, Buckinghamshire, UK). The radioactive compounds were purified (> 98%) by thin layer chromatography on silica gel with ethylacetate/n- heptane (1:9) or acetone/n-heptane (1:1) as eluting agents. Metyrapone, glutathione (GSH) and sodium diethyldithiocarbamate (DEDTC) were obtained from Sigma Chem. Co.

2.2. Animals Female C57B1/6 mice (16-25 g) and female

Sprague-Dawley rats (150-190 g) were obtained from B&K Universal (Sollentuna, Sweden). The animals were housed in a temperature-controlled room at 22°C with a 12/12 h light/dark cycle and given a standard pelleted diet (Ewos AB, SSdert/il- je, Sweden) and tap water ad libitum.

2.3. Histopathology Saline or metyrapone (100 mg/kg; dissolved in

saline) were i.p. injected in groups of mice 30 rain before and 2 h after a single i.p. injection of chlorthiamid (12 mg/kg; 0.06 mmol/kg) dissolved in DMSO (1 /zl/g body wt.). Other groups of mice were given i.p. injections with DEDTC (350 mg/kg; dissolved in saline) or saline 20 rain before injection of chlorthiamid as above. Controls were given i.p. injections with DMSO or with DEDTC. In addition, groups of mice were given i.p. injections with thiobenzamide (8 or 16 mg/kg; 0.06 or 0.12 mmol/kg; dissolved in DMSO). At least 3 mice were used in each group.

The mice were anesthetized 24 h later with gas- eous CO 2 and exsanguinated. The entire nasal region and pieces of the livers were dissected. The nasal passages were gently flushed with 4% phosphate-buffered formaldehyde. Following fixa- tion, the nasal regions were decalcified in 5.5% EDTA in 4% formaldehyde for 2 weeks. The skulls were cut in 2 pieces which were embedded in meth- acrylate resin (Technovit, Hereaus Kulzer, Wehrheim, Germany). Transversal sections (2-3 #m) from the nasal regions (serial sections; > 10 glass slides/mouse) were cut and stained with hematoxylin/periodic acid-Schiff (PAS) or eosin or with toluidine blue. The livers were embedded in paraffin, sectioned (2 glass slides/mouse) and stained with hematoxylin/eosin.

2.4. Metabolic activation in vitro Microsomes of the rat olfactory mucosa and

liver were prepared as described previously [6]. Olfactory microsomes (0.08 mg protein) were added to incubation mixtures (0.5 ml) containing [14C]chlorthiamid (0.6 mM; 0.12 /~Ci) or [14C- nitrile]dichlobenil (0.6 mM; 0.11 /zCi) and an NADPH-generating system (5 mM glucose-6-

c Eriksson, E.B. Brittebo/Toxicol. Left. 76 (1995) 203-208 205

phosphate, 1.2 I U glucose-6-phosphate dehydro- genase and 1 m M N A D P +) in capped 10-ml vials. The effects o f sodium-dithionite (a reducer o f molecular oxygen and P450) and the nucleophilic G S H on the covalent binding of [14C]chlorthiamid in olfactory microsomes were also examined. In another experiment, olfactory and hepatic microsomes (0.2 mg) were added to incubation mixtures containing [14C]chlorthiamid as above or [14C-ring]dichlobenil (20 ~tM; 0.17/~Ci). After 20 min o f incubation at 37°C under air the reaction

was stopped by addit ion o f CHCl3:methanol (2:1) and the covalent binding o f radioactivity to protein was determined by liquid scintillation after repeated t reatment o f the protein precipitate with 1% sodium-dodecylsulphate and acetone as described previously [8].

3. Results

3.1. Histopathology 3.1.1. Chiorthiamid: The effects o f chlorthiamid

Fig. 2. Sections through the posterior olfactory nasal mucosa (dorsal meatus) of C57B1/6 mice killed 24 h following an i.p. injection of (A-C) chlorthiamid (12 mg/kg; 0.06 mmol/kg) or (D) thiobenzamide (16 mg/kg; 0.12 mmol/kg). (B) Metyrapone-treatment (100 mg/kg; i.p.) 30 min before and 2 h after the chlorthiamid-injection and (C) DEDTC-treatment (350 mg/kg; i.p.) 20 rain before the chlorthiamid injection. (A) In the mouse given chlorthiamid only the Bowman's glands are necrotic and there are signs of detachment of the neurovpithelium. (B and C) In the metyrapone- and DEDTC-treated mouse given chlorthiamid the Bowman's glands and the neurocpitbelium have a normal appearance. (D) In the thiobenzamide-treated mouse the Bowman's glands and the neuroepithelium have a normal appearance. The blood vessels and nerve bundles are intact in all mice. OE, olfactory epithelium; BG, Bowman's glands; N, nerves; V, blood vessels (toluidine blue-stained sections).

206 C Eriksson, E.8. Brittebo / Toxicol. Lett. 76 (1995) 203-208

(12 mg/kg; 0.06 mmol/kg) in the olfactory mucosa of mice 24 h following a single i.p. injection were similar to those previously reported [4]. In the dor- somedial olfactory region the neuroepithelium was slightly disorganized and showed signs of detachment, the Bowman's glands were necrotic or had disappeared completely and no red PAS-staining of glycoproteins in the contents of the glands could be observed (Fig. 2A). In other parts of the olfactory or respiratory mucosa, in glands situated around the maxillary sinuses or in the livers no morphological changes were observed.

3.1.2. Metyrapone + chlorthiamid: Treatment with metyrapone almost completely protected against the chlorthiamid-induced toxicity in the olfactory mucosa (Fig. 2B). No necrotic Bowman's glands were observed and PAS-positive red staining of glycoproteins occurred in the contents of the acini. No morphological changes were observed in the livers. Previous studies have shown that the olfactory mucosa of metyrapone-treated control mice has a normal appearance [5].

3.1.3. DEDTC + chlorthiamid: Treatment with DEDTC almost completely protected against the chlorthiamid-induced toxicity in the olfactory mucosa (Fig. 2C). No necrotic Bowman's glands were observed and the acini contained red PAS- stained material. No morphological changes were observed in the livers. In DEDTC-treated control mice the olfactory mucosa had a normal appearance and red PAS-positive material was present in the contents of the Bowman's glands throughout the olfactory region.

3.1.4. Thiobenzamide: In mice given thiobenzamide (0.06 and 0.12 mmol/kg; i.p.) the neuroepithelium and the Bowman's glands in the olfactory mucosa had a normal appearance (Fig. 2D). Following 0.12 mmol/kg the PAS-staining in- tensity of glycoproteins in the Bowman's glands was decreased in the dorsal meatus but in other parts of olfactory mucosa the PAS-staining inten- sity was similar to that of vehicle-treated controls. No morphological changes were observed in the livers.

3.1.5. Vehicle-treated control: In the DMSO- treated mice the olfactory mucosa had a normal appearance and red PAS-positive material was present in the contents of the Bowman's glands

Table i Covalent binding of [14C]chiorthiamid in rat microsomes

olfactory

Compound Concentration % of control a or amount/vial

Control 100 b,c Glutathione 3.0 mM 34 4- 4 b* Dithionite 4 mg 36 4. 12 b* -NADPH 37 4. 10 c*

aRat olfactory microsomes (0.08 b or 0.2 c mg) were incubated with 114Clchlorthiamid (0.6 mM) with or without an NADPH- generating system in a final volume of 0.5 ml at 37°C for 20 min. After termination, the protein precipitates were repeatedly extracted. The levels of covalent binding in the controls were 0.844.0.11 b and 0.724.0.23 c nmol/mg protein/20 min, respectively (mean 4. S.D.; n = 3-5). *P < 0.05; Student's t-test.

throughout the olfactory region. No morphological changes were observed in the livers.

3.2. Metabolic activation of chlorthiamid in vitro In rat olfactory and liver microsomes incubated

with [14C]chlorthiamid (0.6 raM) a low NADPH- dependent covalent binding of radioactivity was observed (Table 1). The NADPH-dependent binding of [14C]chlorthiamid in olfactory microsomes (0.46 ± 0.23 nmol/mg protein/20 rain) was > 3-fold higher than that in liver microsomes (0.12 ± 0.06 nmol/mg protein/20 rain; n = 4). Ad- dition of GSH or sodium-dithionite decreased the binding in olfactory microsomes to the same level as that observed in the absence of an NADPH- generating system. The level of [ 14C]chlorthiamid- binding in olfactory and liver microsomes in the absence of an NADPH-generating system was 37 and 65% of that in the presence of an NADPH- generating system. Due to paucity of radiolabelled compound it was not possible, however, to repeat the studies with a more efficient solvent extraction procedure or to perform studies for calculation of Vma x and K m.

In rat olfactory microsomes incubated with [14C-nitrile]dichlobenil (0.6 raM) the NADPH- dependent covalent binding of radioactivity (3.61 ± 0.09 nmoi/mg protein/20 min; n = 3) was more than 7 times higher than that observed at an

C Eriksson, E.B. Brittebo/Toxicol. Left. 76 (1995) 203-208 207

equimolar concentration of chlorthiamid. Olfac- tory microsomes incubated with [14C-ring]dichlobenil (20 ~M) as positive controls, showed a covalent binding of radioactivity in the same range as previously observed (data not shown) [6].

4. Discussion

Previous autoradiographic studies have shown that the levels of irreversible binding of the [14C]- labelled herbicides chlorthiamid and dichlobenil in the olfactory mucosa are considerably higher than that in the liver of mice given an i.v. injection [3]. Subsequent studies revealed that the covalent binding of a toxic dose ofdichlobenil is > 25 times higher in the olfactory mucosa than in the liver and that there is a considerably higher rate of activation in olfactory microsomes as compared to that in liver microsomes [6,7]. The present study showed a low, NADPH-dependent oxidative activation of [14C]-labelled chlorthiamid into a reac- tive species in olfactory microsomes but the level of binding was only 3 times higher than that in liver microsomes. The relatively low covalent binding of chlorthiamid in olfactory microsomes as compared to that in the liver therefore suggests that a major part of the covalent binding in vivo is not related to an in situ activation. Previous studies have shown that dichlobenil is a major chlorthiamid-metabolite in the urine [3,9]. Hence it cannot be excluded that a major part of the covalent binding and toxicity of chlorthiamid in vivo is related to a hepatic formation of dichlobenil and a subsequent activation of this stable chlorthiamid-metabolite in the olfactory mucosa. It is notable that the toxicity of chlorthiamid in the olfactory mucosa was abolished in mice pretreated with the P450 inhibitors metyrapone or DEDTC, also known to prevent the dichlobenil-induced toxicity at this site [5,10]. Furthermore the covalent binding of [14C]- labelled dichlobenil in olfactory microsomes was more than 7 times higher than that of [14C]- labelled chlorthiamid although previous studies have shown that the in vivo covalent binding and toxicity of these herbicides are in the same magnitude in the olfactory mucosa.

The results of this study also showed that thio-

benzamide, a dechlorinated chlorthiamid analog, did not induce any morphological changes in the olfactory mucosa when given in low doses (8-16 mg/kg), equimolar to toxic chlorthiamid doses. This is consistent with previous results showing that benzonitrile, a dechlorinated dichlobenil analog, did not induce any morphological changes in the olfactory mucosa [1 I]. Hence chlorines in the 2,6-position are important for the high chlorthiamid- and the dichlobenil-induced toxicity at this site. Previous studies have shown that single high doses (_> 150 mg/kg) of thiobenzamide will induce necrosis in the liver [12-15]. The thiobenzamide-induced hepatotoxicity is related to a sequential S-oxidation followed by covalent binding of metabolites [16-20]. Since S-oxidation of thiobenzamide has been demonstrated in the rabbit olfactory region in vitro [21] a covalent binding of thiobenzamide metabolites at this site cannot be excluded. Notably a decreased PAS- staining of glycoproteins in Bowman's glands was observed in thiobenzamide-treated mice (16 mg/kg). Absence of PAS-stainable glycoproteins in the Bowman's glands seems to be a sensitive mark- er for chemically-induced changes in this tissue structure [4,5] suggesting that thiobenzamide may possibly be cytotoxic at this site following high doses.

In conclusion the results of this study showed that pretreatment with P450 inhibitors abolished the chlorthiamid-induced necrosis in the olfactory mucosa. Since only a low covalent binding of chlorthiamid was found in olfactory microsomes the results do not support a major in situ activation of chlorthiamid. In addition, the studies in- dicate that chlorines at the 2,6-position are important for the chlorthiamid-induced toxicity in the olfactory mucosa.

Acknowledgements

We thank Margaretha Mattson for excellent technical assistance, supported by the Swedish Environmental Protection Agency. Drs. J. Bakke and V.J. Fell are thanked for the synthesis of radiolabelled chlorthiamid. Financial support was given by the Swedish Council for Forestry and Agricultural Research, Stiftelsen Oscar och Lili

208 C Eriksson, E.B. Brittebo / Toxicol. Lett. 76 (1995) 203-208

Lamms minne and the Medical Faculty, Univer- sity of Lund.

References

[I] Brown, V.K.H., Chambers, P.L., Ferrigan, L.W., Steven- son, D.E. and Williams, D.A. (1967) Toxicological studies with the herbicide 2,6-dichlorothiobenzamide (Chlorthiamid, 'Prefix'). Arch. Toxicol. 23, 42-51.

[2] Bcynon, K.I. and Wright, A.N. (1972) The fates of the herbicides chlorthiamid and dichlobenil in relation to residues in crops, soils and animals. In: F.A. Gunther, (Ed.), Residues Review, Springer-Verlag, Berlin, Vol. 43, pp. 23-53.

[3] Bakke, J.E., Larscn, G.L., Struble, C., Feil, V.J., Brandt, I. and Brittebo, E.B. (1988) Metabolism of 2,6-dichloro- bcnzonitrile, 2,6-dichlorothiobcnzamide in rodents and goats. Xenobiotica 9, 1063-1075.

[4] Brittebo, E.B., Eriksson, C., Feil, V., Bakke, J. and Brandt, I. (1991) Toxicity of 2,6-dichlorothiobcnzamide (chlorthiamid) and 2,6-dichlorobenzamide in the olfactory nasal mucosa of mice. Fundam. Appl. Toxicol. 17, 92-102.

[5] Brandt, I., Brittebo, E.B., Fell, V.J. and Bakke, J.E. (1990) Irreversible binding and toxicity of the herbicide dichlobcnil (2,6-dichlorobenzonitrile) in the olfactory mucosa of mice. Toxicol. Appl. Pharmacol. 103, 491-501.

[6] Eriksson, C. and Brittebo, E.B. (1991) Metabolic activation of the herbicide dichlobenil in the olfactory mucosa. Chem.-Biol. Interact. 79, 165-177.

[7] Brittebo, E.B., Eriksson, C. and Brandt, I. (1992) Effects of glutathione-modulating agents on the covalent binding and toxicity of dichlobenil in mice. Toxicol. Appl. Pharmacol. 114, 31-40.

[8] Baker, M.T. and Van Dyke, R.A. (1984) Metabolism- dependent binding of the chlorinated insecticide DDT and its metabolite, DDD, to microsomal protein and lipids. Biochem. Pharmacol. 33, 255-260.

[9] Griffiths, M.H., Moss, J.A., Rose, J.A. and Hathway, D.E. (1966) The comparative metabolism of 2,6-dichlorothiobcnzamide (Prefix) and 2,6-dichlorobenzonitrile in the dog and rat. Biochem. J. 98, 770-781.

[10] Eriksson, C. and Brittebo, E.B. (1995) Metabolic activation of the olfactory toxicant, dichtobenil, in rat

olfactory microsomes: comparative studies with p-nitro- phenol. Chem.-Biol. Interact. 94, 183-196.

II 1] Eriksson, C. Brandt, I. and Brittebo, E.B. (1992) Tissue- binding and toxicity of compounds structurally related to the herbicide dichlobenil in the mouse olfactory mucosa. Food Chem. Toxicol. 30, 871-877.

[12] Malvaldi, G. (1977) Liver changes following thiobenzamide poisoning. Experientia 33, 1200-1201.

[13] Hanzlik, R.P., Vyas, K.P. and Traiger, G.J. (1978) Substituent effects on the hepatotoxicity of thiobenzamide derivatives in the rat. Toxicol. Appl. Pharmacol. 46, 685-694.

[14] Chieli, E., Malvaidi, G. and Segnini, D. (1980) The hepatotoxicity of thiobenzamide-S-oxide. Toxicol. Lett. 7, 175-180.

[15] Grossman, S.J., Patrick, D.H., Kornbrust, D., Smith, P.F., Herold, E.G., Deluca, J.G. and Zacchei, A.G. (1991) Thiobenzamide-induced hepatotoxicity: effects of substituents and route of administration on the nature and extent of liver injury. Toxicol. Appl. Pharmacol. 111, 388-408.

[16] Neal, R.A. (1980) Microsomal enzymes and the toxicity of thiono-sulfur compounds. In: M.J. Coon, A.H. Con- hey, R.W. Estabrook, H.V. Gelboin, J.R. Gillette and P.J. O'Brien, (Eds.), Microsomes, Drug Oxidations, and Chemical Carcinogenesis, Vol. II, Academic Press, New York, pp. 791-797.

[17l Hanzlik, R.P. and Cashman, J.R. (1983) Microsomal metabolism of thiobcnzamide and thiobcnzamide S-oxide. Drug Metab. Dispos. 11,201-205.

[18l Tynes, R.E. and Hodgson, E. (1983) Oxidation of thiobcnzamide by the FAD-containing and cytochrome P- 450-dependent monooxygenases of liver and lung microsomes. Biochem. Pharmacol. 32, 3419-3428.

[19] Chieli, E. and Malvaldi, G. (1985) Role of the P450 dependent and FAD-containing monooxygenases in the bioactivation of thioacetamide, thiobenzamide and their suiphoxides. Biochem. Pharmacol. 34, 395-396.

[20] Chieli, E., Saviozzi, M., Puccini, P., Longo, V. and Gervasi, P.G. (1990) Possible role of the acetone- inducible cytochrome P-450IIEI in the metabolism and hepatotoxicity of thiobenzamide. Arch. Toxicol. 64, 122-127.

[21] Dahl, A.R., Bond, J.A., Petridou-Fischer, Sabourin, P.J. and Whaley, S.J. (1988) Effects of the respiratory tract on inhaled materials. Toxicol. Appl. Pharmacol. 93, 484-492.

effects of the herbicide chlorthiamid on the olfactory mucosa

Documents