N-alkylformamides are metabolized to N-alkylcarbamoylating species

Mar 1, 1990 - Mechanisms of DrugToxicity Research Group,Pharmaceutical Sciences Institute, Department of Pharmaceutical Sciences, Aston University, ...
0 downloads 0 Views 1MB Size
Chem. Res. Tonicol. 1990, 3, 357-362

357

N-Alkylformamides Are Metabolized to N-Alkylcarbamoylating Species by Hepatic Microsomes from Rodents and Humans Hilary Cross, Renuka Dayal, Ruth Hyland, and Andreas Gescher* Mechanisms of Drug Toxicity Research Group, Pharmaceutical Sciences Institute, Department of Pharmaceutical Sciences, Aston University, Aston Triangle, Birmingham B47ET, U.K. Received March 1, 1990

Hepatotoxic formamides such as N-methylformamide (NMF) and Nfl-dimethylformamide

(DMF) are metabolized in vivo to N-acetyl-S-(N-methylcarbamoy1)cysteinevia oxidation a t the formyl carbon, which yields a reactive intermediate. The hypothesis was tested that this biotransformation route can be studied in vitro with hepatic fractions. NMF was incubated with microsomes or cytosol obtained from BALB/c mice, and metabolically generated N-methylcarbamoylating species were analyzed after derivatization with ethanol in base to furnish ethyl N-methylcarbamate. Generation of metabolite was catalyzed by microsomes, but not by cytosol. Detection of the N-methylcarbamoylating species was dependent on the presence in the incubation mixture of NMF, viable microsomes, NADPH, and a thiol-containing agent such as glutathione. Metabolite formation was inhibited by SKF 525-A (3 mM) and abolished when the incubation atmosphere consisted of an air/carbon monoxide mixture (1:l)instead of air. Metabolism was not induced by pretreatment of mice with phenobarbital or 8-naphthoflavone. N-Ethylformamide and the DMF metabolite N-(hydroxymethy1)-N-methylformamide, but not DMF, were metabolized by microsomes to the N-alkylcarbamoylatingmetabolite at a measurable rate. NMF metabolism was also observed with liver microsomes from Sprague-Dawley rats or from humans. In the case of rat microsomes the rate of metabolism was half of that measured with murine microsomes. The results suggest that (i) the metabolic toxification of NMF can be studied in hepatic microsomes and (ii) the oxidation of the formyl moiety in N-alkylformamides is catalyzed by cytochrome P-450.

Introductlon N-Alkylformamides such as NMF' (CH3NHCHO)and DMF [(CH,),NCHO] are polar solvents frequently used in the chemical laboratory and in industrial processes. NMF possesses antineoplastic properties in mouse tumor models (1-3),and its potential application as a cancer chemotherapeutic agent was evaluated in a number of phase 1and phase 2 clinical trials (4-8). The outcome of these trials did not support a role for NMF in the therapy of human malignancies, not least due to the hepatotoxicity, occasionally very severe, which was observed as one of the dose-limiting toxicities in patients treated with NMF. Manifestations of liver damage have also been reported recently in workers occupationally exposed to DMF under conditions of poor industrial hygiene (9). The hepatotoxicity of NMF can be reproduced in rodents (10-13),and the mechanism by which N-alkylformamidescause liver damage in mice has been shown to be intrinsically linked to their metabolism (13).NMF is extensively metabolized in vivo in rodents and patients (3,12-17)and in suspensions of mouse hepatocytes (18). The metabolic pathway of NMF, which appears to be associated with the generation of the hepatotoxic lesion, leads via oxidation at the formyl moiety to a reactive intermediate, possibly methyl isocyanate (CH3NCO)(13),which reads with GSH to yield the glutathione conjugate SMG' (16)(seeFigure 1). SMG is further metabolized to the mercapturate AMCC' as a major urinary excretion product of NMF (13,15).AMCC Abbreviations: AMCC, N-acetyl-S-(N-methylcarbamoy1)cyeteine; DMF, NJV-dimethylformamide; GSH, glutathione; HMMF, N-(hydroxymethy1)-N-methylformamide;NEF, N-ethylformamide; NMF, N-methylformamide;SEG, S-(N-ethylcarbamoy1)glutathione:SMG. S(N-methylcarbamoy1)glutathione.

is also a urinary metabolite of DMF in humans and rodents (19,20).Whereas the parent formamides are devoid of direct cytotoxic properties, N-methylcarbamic acid thioesters, such as SMG and AMCC, are cytotoxic and are not innocuous detoxification products (21). In view of the pivotal role that metabolism plays in determining the toxicity of N-alkylformamides, we wished to test the hypothesis that the metabolic oxidation of N-alkylformamides, which leads ultimately to a urinary thioester, is catalyzed by hepatic microsomes. To that end we obtained liver microsomes from mice, rats, and humans and studied their ability to generate N-alkylcmbamoylatingmetabolites using NMF, NEF (CH3CH2NHCHO),and DMF and its major in vivo metabolite HMMF [ (CH20H)CH3NCHO] (22)as substrates.

Experimental Procedures Sources of Microsomes and Chemicals. Mice of the BALB/c or CBA/CA strains (18-23 g) and Sprague-Dawley rata (90-130 g), all of male sex, were purchased from Bantin and Kingman Ltd. (Hull, U.K.). Excess samples of healthy human liver tissue after graft reduction of donor liver were obtained from the Liver Transplant Unit at the Queen Elizabeth Hospital, Birmingham, U.K. These tissues originated from four organ donors, three male and one female, aged between 23 and 35 years. In two cases the postmitochondrial supernatant of human liver samples was stored at -70 O C for 2 and 8 weeks, respectively, before use. Cofactors (NADPH, NADP, glucose 6-phosphate, glucose-6phosphate dehydrogenase), GSH, N-acetylcysteine, and 8-naphthoilavone were obtained from Sigma Chemical Co. Ltd. (Poole, U.K.). Formamides except HMMF were purchased from Aldrich Chemical Co. (Poole, U.K.) and purified by distillation, so that they were >99% pure (GLC). HMMF was synthesized as described before (23) and contained inevitably up to 20% NMF.

0893-228~/90/2703-0357$02.50/0 0 1990 American Chemical Society

358 Chem. Res. Toricol., Vol. 3, No. 4, 1990 0 I

I

-

CH3

DMF

CH3

HMMF

NMF

0

II

0

,/'\,/"

~-4%

-L-

I CH3

described for mice were obtained with microsomes from BALB/c animals, unless stated otherwise. In some experiments mice were pretreated (i.p.) with either phenobarbital (0.2 mL, 80 mg/kg, dissolved in saline) or 8-naphthoflavone (0.3 mL, 50 mg/kg, dissolved in corn oil) once daily on 4 consecutive days prior to the preparation of microsomes. Sample Preparation a n d Analytical Procedures. Incubations were terminated by addition of ethanol (3 mL, 4 "C) to a sample (1.5 mL) of the incubate. Metabolically generated N-alkylcarbamoylatingactivity, most likely N-alkylcarbamic acid thioester, for example, SMG in the presence of GSH with NMF as substrate (see Discussion), was derivatized and analyzed by GLC as described by Mriz (24) with the modifications suggested by Shaw et al. (18) and Mrlz et al. (20). Derivatization of SMG or SEG with ethanol in alkali yields ethyl N-methylcarbamate (CH3NHCOOC2H5)or ethyl N-ethylcarbamate (CzH5NHCOOC2H6),respectively, which were quantified by GLC using a Pye Unicam Series 204 chromatograph,under the conditions described before (18, 20). Calibration curves were established by using authentic SMG and SEG. Generation of metabolite was linear with concentration of microsomal protein in the mixture. Rates of metabolite production given under Results were calculatedfrom amounta measured within 20- or 4omin incubation, but only when linearity between metabolite formation and incubation time had been established. N-Demethylationof aminopyrine was measured by colorimetricquantification of formaldehyde according to Nash (25). Cytochrome P-450 levels were measured spectroscopically as described by Gibson and Skett (26). The protein content of the microsomal suspension was determined according to the method of Lowry et al. (27).

0

I

CH3

Cross et al.

O=C=)SH-CH

4

/I

GSH

C GS'

H

\N/

I

CH3

NMF

,i" 0

COOH HN -CH-CH2I I COCH,

'I

S /c\N/H I CHI

AMCC

Figure 1. Metabolism of DMF and NMF. The structure in brackets is a postulated intermediate. SMG and SEG were prepared by Drs. D. H. Han and P. G. Pearson, University of Washington, Seattle, as described recently (21). SKF 525-A (proadifen) was a gift from Smith Kline and French Ltd. (Welwyn, U.K.). Preparation of Microsome@and Microsomal Incubations. Liver homogenate (25%) was prepared in phosphate-buffered (50 mM, pH 7.4) KC1 solution (0.154M). Microsomes were obtained in the usual way by differential centrifugation of homogenate first at (9 X 10a)gfor 20 min and then at 1@g for 1h in a Beckman L8-6OM ultracentrifuge. The microsomal pellet was suspended in phosphate buffer (50 mM, pH 7.4), recentrifuged at 105gfor 1h, and resuspended. In control incubates microsomes were used after inactivation by heating in a boiling water bath for 10 min. Incubations were performed, in duplicate, by using glass vials in a shaking water bath at 37 "C. Mixtures contained microsomes pooled from three or four mice, equivalent to 0.25 g of liver, NADPH (10 mM), or a NADPH-generating system [glucose 6phosphate (20 mM), NADP (10 mM), glucose-6-phosphate dehydrogenase (4 units)], MgClz (50 mM), GSH (10 mM), and formamides (1-10 mM) in a final volume of 2 mL. In control experiments preformed NADPH or NADPH generated in situ supported the reaction rate equally. In some cases SKF 525-A (0.5-3mM) waa added. Incubations were commenced by addition of Substrate and carried on for up to 60 min thereafter. Mixtures were incubatedunder (i) air in open vials, (ii) nitzagen in stoppered vials connected to a Nz bottle, or (iii) mixtures of carbon monoxide/air or nitrogen/air (1:l each) using a system of connected flasks which allowed evacuation and replacement of gas. Results

B

A I

c-1

D

Results Metabolism of NMF and NEF by Mouse Liver Microsomes. Figure 2 shows representative gas chromatograms of extracts of incubates with mouse liver microsomes. The detection of the N-methylcarbamoylating species was dependent on the presence of NMF, NADPH, and GSH in the incubation mixture. Generation of metabolite was abolished when viable microsomes were replaced by hepatic cytosol or heat-inactivated microsomes. Microsomal formation of metabolite was related in a linear fashion to both NMF concentration and incubation time (Figure 3A). When NMF was replaced by NEF as substrate, an N-ethylcarbamoylating species was detected by GLC. The microsomal biotransformation of NEF (Figure 3B) proceeded at a rate which was considerably faster than that seen with NMF: within 30 min NEF at 1 or 5 mM generated, respectively, 7 and 3 times as much N-alkylE

I

F

!

i2 "

I,

I

II

I 1 Pmin

G

1 1 Zmm

RETENTION TIME (min)

Figure 2. Gas-liquid chromatograms of derivatized samples of a mixture of SMG and NMF (A) and of incubation mixtures of NMF (10 mM) (B-F)or HMMF (10 mM) (G)at the start of the incubation (B) and after 60 min (C-G), with viable (B, C, E X ) or hmbinactivated microsomes (D) in the presence (B-D, F,G) or absence of GSH (E) and in the presence (B-E, G) or absence (F)of NADPH. Chart speed was reduced 1.2-1.5 min after each sample injection on the column. Peak I coeluted with NMF, peak I1 with quinoline, which was used as internal standard. Arrow indicates position of ethyl N-methylcarbamate peak. Details of the GLC method have been described before (18, 20).

Chem. Res. Toxicol., Vol. 3, No. 4, 1990 359

Microsomal Metabolism of N - Alkylformamides 40 I

A

I

1mM 5mM A 10mM

L

0

a-

I

I

10

20

I 30

I 40

-4

I 50

0 I-

T

o*8t+T GsH

I

1

I 60

TIME OF INCUBATION (min)

B

+-cetylcysteine

B Control

T

1mM 5mM A 10mM 0

/

PB

T

SKF 0.5mM

I C

-1

0

10

20

30

40

50

PB

I

60

TIME OF INCUBATION (min)

Figure 3. Time course of microsomal metabolism of NMF (A) and NEF (B)to N-alkylcarbamoylating species. Values are the mean SD of 3 experiments.

*

carbamoylating metabolite as did NMF at the same concentrations. The N-methylcarbamoylating metabolite was not detected in mouse liver microsomes in the presence of NMF when GSH was omitted from the incubation mixture. However the metabolite was found when GSH was replaced by N-acetylcysteine in the incubate. With Nacetylcysteine as trapping reagent the concentration of N-methylcarbamoylating metabolite measured was comparable to that seen in the presence of GSH (Figure 4A). In order to find out which enzymes may be involved with the catalysis of NMF biotransformation, we investigated the effect of the following conditions on the production of N-methylcarbamoylatingspecies: (i) replacement of air in the incubation atmosphere with nitrogen or a carbon monoxide/air mixture, (ii) inclusion of SKF 525-A in the incubate, and (iii) pretreatment of mice with phenobarbital. Figure 4B illustrates some of the results obtained in these experiments. Exposure to carbon monoxide abolished the metabolism of NMF. NMF was also not metabolized under an atmosphere of nitrogen (result not shown). Coincubation with SKF 525-A (3 mM) inhibited NMF metabolism by 67%, and pretreatment of mice with phenobarbital (80 mg/kg) for 4 days did not increase the rate of microsomal metabolite formation (Figure 4B). Treatment with phenobarbital augmented cytochrome P-450 levels by 67% from 1.6 f 0.3 nmol/mg of microsomal protein in control mice to 2.7 f 0.4 nmol/mg of mi-

Figure 4. Rate of metabolism of NMF (10 mM) (A, B)or aminopyrine (5 mM) (C) in mouse liver microsomes: (A)in the presence of GSH or N-acetylcysteine; (B, C) on incubation either under an atmosphere of mixtures of carbon monoxide/air or nitrogen/air (1:l each) or with SKF 525-A (SKF), or using microsomes from mice pretreated with phenobarbital (PB). Values are the mean f SD of 4-6 experiments.

crosomal protein (mean f SD, n = 4). Thus metabolism of NMF to the N-methylcarbamoylating species was decreased from 0.42 nmol/(nmol of cyt P-450.min) in control microsomes to 0.28 nmol/(nmol of cyt P-450-min) in microsomes obtained from phenobarbital-pretreated mice. Likewise, pretreatment of mice with the enzyme inducer P-naphthoflavoneor heating microsomal mixtures (45 "C) for 10 min prior to addition of substrate, which affects flavin monooxygenase activity, failed to alter the rate at which NMF was metabolized (results not shown). For comparison, Figure 3C shows the effect of carbon monoxide, SKF 525-A, and phenobarbital pretreatment on the metabolism of aminopyrine to formaldehyde in mouse microsomes. Comparison between NMF-Metabolizing Ability of Rodent and Human Microsomes. Different rodent species and mouse strains possess different susceptibilities toward the hepatotoxic potential of NMF (II,12,28).In order to study whether this difference is related to a difference in the microsomal metabolism of NMF, the formation of the N-methylcarbamoylating metabolite was

Cross et al.

360 Chem. Res. Toxicol., Vol. 3, No. 4, 1990

40

T I

t

A BALB/c MOUSE CEAICAMOUSE

~

0

10

20 30 40 50 TIME OF INCUBATION (min)

60

Figure 5. Time course of metabolism of NMF (10 mM) by microsomes obtained from BALB/c or CBA/CA mice or Sprague-Dawley rats. Values are the mean f SD of 4 experiments.

compared in microsomes from mice of the BALB/c and CBA/CA strains and Sprague-Dawley rats. Figure 5 shows that the NMF-metabolizing capacity of microsomes from the two mouse strains was similar,whereas rat microsomes catalyzed the generation of N-methylcarbamoylating activity a t half the rate measured in mice. Human microsomes also produced the N-methylcarbamoylating metabolite from NMF (10 mM) in the presence of NADPH and GSH. Formation of the metabolite by liver microsomes from four human adults was linear with time for 60 min and varied between 0.17 and 0.68 nmol/(mg of microsomal protein-min) [mean f SD 0.34 f 0.24 nmol/(mg of microsomal protein-min)]. These values compare with 0.60 f 0.08 and 0.32 f 0.11 nmol of metabolite/(mg of microsomal protein-min) for microsomes from BALB/c mice and Sprague-Dawley rats,respectively. In two human liver samples levels of cytochrome P-450 were also measured, and in these microsomes metabolism of NMF to the N-methylcarbamoylating species occurred at a rate of 0.39 and 0.46 nmol/(nmol of cyt P-4Wmin). These values are similar to that found in mice [0.42 nmol/(nmol of cyt P-450-min)l. Microsomal Metabolism of DMF and HMMF. Metabolic conversion to N-methylcarbamoylating species was not detectable on incubation of DMF with murine or human hepatic microsomes. Nevertheleas, a peak coeluting with authentic NMF was found on GLC analysis of incubates with DMF, but quantification was not performed. NMF is likely to originate from the primary DMF metabolite HMMF, which is relatively stable in aqueous solution but decomposes on heating to NMF and formaldehyde (30). Incubation of mouse or human liver microsomes with HMMF afforded N-methylcarbamoylating activity, which eluted as ethyl N-methylcarbamate on the GLC column (Figure 2G). At 5 and 10 mM HMMF 1.5 f 0.2 and 1.4 f 0.1 nmol of N-methylcarbamoylating metabolite was measured, respectively, per milligram of microsomal protein after 20-min incubation with mouse liver microsomes. This amount did not change significantly when the incubation period was prolonged to 60 min. In human microsomes metabolite formation from HMMF (10 mM) was linear with time and amounted to 46 f 34 pmol/(mg of microsomal proteimmin) (mean f SD, n = 4).

Dlscusslon Almost 20 years ago NMF and DMF were claimed to be metabolized by rat liver microsomes to formaldehyde, which was measured colorimetrically (31). However, in confirmatory experiments levels of formaldehyde generated

by microsomes on incubation with NMF or DMF were indistinguishable from formaldehyde levels in control microsomes (3). Therefore, the results described above demonstrate for the first time that (i) N-alkylformamides undergo metabolism in vitro in a liver homogenate fraction and (ii) hepatic enzymes which catalyze the metabolism of formamides are located in microsomes but not in cytosol. In view of what is known about the metabolism of N-alkylformamides in vivo and in suspensions of mouse hepatocytes (see Introduction) we propose that the carbamoylating metabolite found in microsomal incubates in the presence of GSH was mainly or exclusively S-(N-alkylcarbamoyl)glutathione,SMG in the case of NMF and SEG in mixtures with NEF. Metabolite formation was only detected when a thiol-containing agent such as GSH was present in the incubation mixture. The dependence of metabolite detection on the presence of GSH in the microsomal incubate might indicate a dual role for GSH (i) as nucleophilic recipient of the N-alkylcarbamoyl moiety and (ii) as cofactor in the enzyme-catalyzed oxidation of the formamide. However, the finding that N-acetylcysteine was as efficient in generating and/or trapping the N-methylcarbamoylating metabolite as was GSH suggests that GSH is not specifically required as cofactor for Nformyl oxidation. It is possible to explain the necessity of a thiol compound to be present for metabolite detection by the fact that the metabolic progenitor of SMG is too short-lived and reactive to survive unless trapped. This interpretation is consistent with two recent findings: (i) Radioactivity derived from 14C-labeledNMF was covalently bound to microsomal protein in incubations with murine liver microsomes in the absence of GSH but not in ita presence (29); (ii) depletion of hepatocytic GSH by pretreatment of mice with buthionine sulfoximine (1600 mg/kg i.p.) or by incubating cells with diethyl maleate (0.02%) exacerbated NMF-induced cytotoxicity in suspensions of hepatocytes and reduced generation of the N-methylcarbamoylating metabolite from NMF (5 mM) by 59 and 90%, respectively.2 The cofactor requirements of the microsomal production of the N-methylcarbamoylating metabolite from NMF and the lack of effect of pretreatment of mice with phenobarbital or P-naphthoflavone are compatible with the suggestion that the metabolism of NMF is catalyzed by cytochrome P-450, but not by phenobarbital- or P-naphthoflavone-inducible isozymes. Work in our laboratory is currently directed toward the characterization of the enzymes involved. The degree of hepatotoxicity of NMF in vivo is dependent on rodent species (11,12)and strain of mouse (281, and the BALB/c mouse appears to be the animal most susceptible to the hepatotoxic potential of NMF. Metabolism of NMF was observed with microsomes obtained from mice of the BALB/c and CBA/CA strains, from rats and humans. The strain difference in toxicity was not reflected by a difference between BALB/c and CBA/CA mice in the rate of microsomal NMF metabolism. However, the lower susceptibility of rats toward the hepatotoxicity of NMF compared to that of mice (11, 12) may be related to the difference observed here in the respective abilities of hepatic microsomes to catalyze the formation of the N-methylcarbamoylating metabolite. The capacity of liver microsomes from different human donors to metabolize NMF to SMG varied by a factor of 4, which might partially explain the individual variability in incidence and severity of hepatic damage observed in the clinical trials

* A. J. Shaw and A. Gescher, unpublished result.

Microsomal Metabolism of N-Alkylformamides (4-8). Interestingly, in these trials hepatotoxicity did not seem to be clearly dose-related. DMF, like NMF, undergoes metabolism in vivo to AMCC (19),even though its major urinary metabolite is HMMF (22)(see Figure 1). The generation of the mercapturate is only a minor biotransformation route of DMF in rodents, but a prominent one in humans (20).However, we could not detect N-methylcarbamoylating activity in suspensions of either human or rodent microsomes in the presence of DMF. This finding could be interpreted to indicate that DMF has to be converted to another species, presumably HMMF and/or NMF, before metabolism to SMG and AMCC can occur and that microsomal DMF methyl oxidation, which was observed but not quantified, operates at a very slow rate. It is also conceivable that formation of SMG from metabolically generated HMMF (or NMF) in these incubations is inhibited by DMF or one of its metabolites. HMMF, which decomposes to NMF very slowly (30),afforded SMG on incubation with murine or human microsomes, but at a much slower rate than that observed with NMF at the same concentration. In mouse microsomes the amount of SMG generated from HMMF was only one-tenth of that produced by NMF. Whereas in human microsomes metabolite production was linear with time, it did not increase beyond 20-min incubation time in mouse microsomes, and it was not linear with HMMF concentration. Some of these observations can perhaps be explained by the necessity of decomposition of HMMF to NMF for metabolic attack to occur at the formyl carbon. Alternatively, the affinity of HMMF for the metabolizing enzyme might be considerably lower than that of NMF. The metabolic transformation of N-alkylformamides to S-(N-alkylcarbamoy1)glutathioneis linked to the mechanism by which these agents cause hepatotoxicity (13). We have argued that the ultimate hepatotoxic species via which NMF causes liver damage may well be methyl isocyanate (13),but unequivocal proof for this contention has still to be presented. The results described here demonstrate that microsomes are a suitable experimental system in which this important novel metabolic route of formamides can be observed; thus the results render further investigation of mechanistic and enzymic details of the toxification reaction in vitro feasible.

Acknowledgment. This work was supported by grants from the Health and Safety Executive and the Medical Research Council of Great Britain. We thank Dr. A. Strain and colleagues at the Liver Unit, Queen Elizabeth Hospital, Birmingham, U.K., and Dr. J. K. Chipman, School of Biochemistry, University of Birmingham, for access to human liver samples; Dr. A. J. Shaw, University of Nottingham, Nottingham, U.K., for preparatory experiments; Drs. D. H. Han and P. G. Pearson, University of Washington, Seattle, for providing SMG and SEG; Dr. T. A. Baillie, University of Washington, for helpful suggestions and discussion; and NATO for a grant enabling collaboration. Registry No. DMF, 68-12-2;GSH, 70-18-8;HMMF, 2054632-1;NEF, 627-45-2;NMF, 123-39-7;NADPH, 53-57-6;cytochrome P-450,9035-51-2.

References (1)Clarke, D. A., Philips, F. S., Sternberg, S. S., Barclay, R. K., and Stock, C. C. (1953)Effecta of N-methylformamide and related compounds in sarcoma 180. R o c . SOC.Exp. Biol. Med. 84, 203-207. (2)Furst, A., Cutting, W. C., and Gross, H. (1955)Retardation of growth of Ehrlich ascites tumor by formamides and related com-

Chem. Res. Toxicol., Vol. 3, No. 4, 1990 361 pounds. Cancer Res. 15,294-299. (3)Gescher, A., Gibson, N. W., Hickman, J. A., Langdon, S. P., Roas, D., and Atassi, G. (1982)N-Methylformamide: Antitumor activity and metabolism in mice. Br. J. Cancer 45,843-850. (4)Laird Myers, W.O., Karnofsky, D. A., and Burchenal, Y. H. (1956)The hepatotoxicity of N-methylformamidein man. Cancer 9,945-954. (5)McVie, J. G., ten Bokkel Huinink, W. W., Simonetti, G., and Dubbelman, R. (1984) Phase I trial of N-methylformamide. Cancer Treat. Rep. 68,607410. (6)Ettinger, D. S.,Spremulli, E. N., Matook, G. M., Diamond, I., and Calabresi, P. (1985)Phase I study of N-methylformamide in patients with advanced cancer. Cancer Treat. Rep. 69,489-493. (7)Eisenhauer, E. A., Weinerman, B. H., Kerr, I., and Quirt, I. (1986) Toxicity of oral N-methylformamide in three phase I1 triala: A report from the National Cancer Institute of Canada Trials Group. Cancer Treat. Rep. 70,811-813. (8)Rowinsky, E. K., Grochow, L. B., Hantel, A., Ettinger, D. S., Vito, B. L., and Donehower, R. C. (1989)Assessment of N-methylformamide (NMF) administered orally on a three times weakly schedule: a phase I study. Invest. New Drugs 7,317-325. (9)Redlich, C. A., Beckett, W. S., Sparer, J., Barwick, K. W., Riely, C. A., Miller, H., Sigal, S. L., Shalat, S. L., and Cullen, M. R. (1988)Liver disease associated with occupational exposure to the solvent dimethylformamide. Ann. Intern. Med. 108,680-686. (10)Whitby, H., Gescher, A., and Levy, L. (1984)An investigation of the mechanism of hepatotoxicity of the antitumor agent Nmethylformamide in mice. Biochem. Pharmacol. 33, 295-302. 11) Scailteur, V., and Lauwerys, R. (1984)In vivo metabolism of dimethylformamide and relationship to toxicity in the male rat. Arch. Toxicol. 56, 87-91. 12)Tulip, K., and Timbrell, J. A. (1988)Comparativehepatotoxicity and metabolism of N-methylformamide in rats and mice. Arch. Toxicol. 62,167-176. 13)Kestell, P.,Threadgill, M. D., Gescher, A., Gledhill,A. P., Shaw, A. J., and Fanner, P. B. (1987)An investigation of the relationship between hepatotoxicity and the metabolism of N-alkylformamides. J. Pharmacol. Exp. Ther. 240,265-270. 14) Kestell, P., Gescher, A., and Slack, J. A. (1985)The fate of N-methylformamide in mice-Routes of elimination and characterization of metabolites. Drug Metab. Dispos. 13,587-592. (15)Kestell, P., Gledhill, A. P., Threadgill, M. D., and Gescher, A. (1986)S-(N-Methylcarbamoy1)-N-acetylcysteine:A urinary metabolite of the hepatotoxic experimental antitumor agent Nmethylformamide (NSC3051) in mouse rat and man. Biochem. Pharmacol. 35,2283-2286. (16)Threadgill, M. D., Axworthy, D. B., Baillie, T. A,, Farmer, P. B., Farrow, K. C., Gescher, A., Kestell, P., Pearson, P. G., and Shaw, A. J. (1987)Metabolism of N-methylformamidein mice: Primary kinetic deuterium isotope effect and identification of S-(Nmethylcarbamoy1)glutathione as a metabolite. J. Pharmacol. Exp. Ther. 242,312-319. (17)Tulip, K., Timbrell, J. A., Nicholson, J. K., Wilson, I., and Troke, J. (1986)A proton magnetic resonance study of the metabolism of N-methylformamide in the rat. Drug Metab. Dispos. 14,746-749. (18)Shaw, A. J., Gescher, A., and Mrb, J. (1988)Cytotoxicity and metabolism of the hepatotoxin N-methylformamide and related formamides in mouse hepatocytes. Toxicol. Appl. Pharmucol. 95, 162-170. (19)Mrb, J., and TureEek, F. (1987)Identification of N-acetyl-S(N-methylcarbamoyl)cysteine,a human metabolite of N,N-dimethylformamide and N-methylformamide. J. Chromutogr. 414, 399-404. (20)Mrb, J., Cross, H., Gescher, A., Threadgill, M. D., and Flek, J. (1989)Differences between humans and rodents in the metabolic toxification of N,N-dimethylformamide. Toricol. Appl. Pharmacol. 98,507-516. (21)Han, D.-H., Pearson, P. G., Baillie, T. A., Dayal, R., Tsang, L. H., and Gescher, A. (1990)Chemical synthesis and cytotoxic properties of N-alkylcarbamic acid thioesters, metabolites of hepatotoxic formamides. Chem. Res. Toxicol. 3, 118-124. (22)Kestell, P.,Gill, M. H., Threadgill, M. D., Gescher, A,, Howarth, 0. W., and Curzon, E. H. (1986)Identification by proton NMR of N-(hydroxymethy1)-N-methylformamideas the major urinary metabolite of N,N-dimethylformamide in mice. Life Sci. 38, 719-724. (23)Gate, E.N., Threadgill, M. D., Stevens, M. F. G., Chubb, D., Vickers, M. M., Langdon, S. P., Hickman, J. A., and Gescher, A. (1986)Structural studies on bioactive compounds. 4. A struc-

362 Chem. Res. Toxicol., Vol. 3, No. 4, 1990 ture-antitumor activity study on analogues of N-methylformamide. J. Med. Chem. 29, 1046-1052. (24)Mr&, J. (1988)Gan chromatographic method for the analysis of N-acetyl-S-(N-methylcarbamoyl)cysteine, a metabolite of N,Ndimethylformamide and N-methylformamide in human urine. J. Chromatogr. 431,361-368. (25)Nash, T.(1953)The colorimetric estimation of formaldehyde by means of the Hantzsch reaction. Biochem. J. 55,416-421. (26)Gibson, G. G., and Skett, P. (1986)Introduction to Drug Metabolism, pp 246-248, Chapman and Hall, London. (27)Lowry, 0.H.,Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951)Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193,265-275.

Cross et al. (28)Pearson, P. G.,Gescher, A., and Harpur, E. S. (1987)Hepatotoxicity of N-methylformamide in mice. I. Relationship to glutathione status. Biochem. Pharmacol. 36,381-384. (29)Pearson, P.G., Gescher, A., Harpur, E. S., and Threadgill, M. D. (1987) Hepatotoxicity of N-methylformamide in mice. 11. Covalent binding of metabolites of '%-labelled N-methylformamide to hepatic proteins. Biochem. Pharmacol. 36, 385-390. (30)Brindley, C.,Gescher, A., and Ross, D. (1983)Studies of the metabolism of dimethylformamide in mice. Chem.-Biol. Interact. 45,387-392. (31) Barnes, J. R., and Ranta, K. E. (1972) Metabolism of dimethylformamide and dimethylacetamide. Toxicol. Appl. Pharmacol. 23,271-276.