Bioactivation of S-(2,2-dihalo-1,1-difluoroethyl)-L-cysteines and S- (trihalovinyl)-L-cysteines by cysteine S-conjugate β-lyase: Indications for formation of both thionoacylating species and thiiranes as reactive intermediates

Article abstract of DOI:10.1021/tx960049b

The covalent binding of reactive intermediates, formed by β-elimination of cysteine S-conjugates of halogenated alkenes, to nucleophiles was studied using ¹⁹F-NMR and GC-MS analysis. β-Elimination reactions were performed using rat renal cytosol and a β-lyase model system, consisting of pyridoxal and copper(II) ion. S-(1,1,2,2-Tetrafluoroethyl)-L-cysteine (TFE-Cys) was mainly converted to products derived from difluorothionoacetyl fluoride, namely, difluorothionoacetic acid, difluoroacetic acid, and N- difluorothionoacetylated TFE-Cys. In the presence of o-phenylenediamine (OPD), as a bifunctional nucleophilic trapping agent, the major product formed was 2-(difluoromethyl)benzimidazole. This product results from initial reaction of difluorothionoacetyl fluoride with one of the amino groups of OPD, followed by a condensation reaction between the thionoacyl group and the adjacent amino group of OPD. In incubations with S-(2-chloro-1,1,2- trifluorofluoroethyl)-L-cysteine (CTFE-Cys) and S-(2,2-dichloro-1,1- difluorofluoroethyl)-L-cysteine (DCDFE-Cys), formation of thionoacylated cysteine S-conjugates was also observed by GC-MS analysis, indicating formation of the corresponding thionoacyl fluorides. However, according to ¹⁹F-NMR analysis, chlorofluorothionoacyl fluoride-derived products accounted for only 10% of the CTFE-Cys converted. In the presence of OPD, next to the corresponding 2-(dihalomethyl)benzimidazoles, 2- mercaptoquinoxaline was identified as the main product in incubations with CTFE-Cys. When chlorofluorothionoacylating species were generated from the unsaturated S-(2-chloro-1,2-difluorovinyl)-L-cysteine (CDFV-Cys), 2- (chlorofluoromethyl)benzimidazole and 2-mercaptoquinoxaline were also found as OPD adducts. However, with CDFV-Cys the ratio of 2- (chlorofluoromethyl)benzimidazole to 2-mercaptoquinoxaline was 12-fold higher than in the case of CTFE-Cys. These results suggest an important second mechanism of formation of 2-mercaptoquinoxaline with CTFE-Cys. The formation of 2-mercaptoquinoxaline could also be explained by reaction of OPD with 2,3,3-trifluorothiirane as a second reactive intermediate for CTFE-Cys. Comparable results were obtained when comparing OPD adducts from DCDFE-Cys and TCV-Cys. Both DCDFE-Cys and TCV-Cys form dichlorothionoacylating species. However, DCDFE-Cys forms 21-fold more 2-mercaptoquinoxaline than TCV-Cys, which may be explained by its capacity to form 3-chloro-2,2-difluorothiirane next to dichlorothionoacyl fluoride. In order to explain the apparent differences in the preference of thiols to form different reactive intermediates, free enthalpies of formation (Δ(f)G) of thiolate anions and their possible rearrangement products, thionoacyl fluorides and thiiranes, derived from TFE-Cys, CTFE-Cys, and DCDFE-Cys, were calculated by ab initio calculations. For TFE-thiolate, formation of difluorothionoacetyl fluoride is energetically favored over formation of the thiirane. In contrast, the thiirane pathway is favored over the thionoacyl fluoride pathway for CTFE- and DCDFE-thiolates. The results of these quantum chemical calculations appear to be consistent with the experimental data.

Full text of DOI:10.1021/tx960049b

1092
Chem. Res. Toxicol. 1996, 9, 1092-1102
Bioa ctiva tion of S-(2,2-Dih a lo-1,1-d iflu or oeth yl)-L-
cystein es a n d S-(Tr ih a lovin yl)-L-cystein es by Cystein e
S-Con ju ga te â-Lya se: In d ica tion s for F or m a tion of both
Th ion oa cyla tin g Sp ecies a n d Th iir a n es a s Rea ctive
In ter m ed ia tes
J an N. M. Commandeur,* Laurence J . King,† Luc Koymans, and
Nico P. E. Vermeulen
Leiden/ Amsterdam Center for Drug Research, Department of Pharmacochemistry, Division of
Molecular Toxicology, Free University, De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands
Received March 22, 1996X
The covalent binding of reactive intermediates, formed by â-elimination of cysteine
S-conjugates of halogenated alkenes, to nucleophiles was studied using 19F-NMR and GC-MS
analysis. â-Elimination reactions were performed using rat renal cytosol and a â-lyase model
system, consisting of pyridoxal and copper(II) ion. S-(1,1,2,2-Tetrafluoroethyl)-L-cysteine (TFE-
Cys) was mainly converted to products derived from difluorothionoacetyl fluoride, namely,
difluorothionoacetic acid, difluoroacetic acid, and N-difluorothionoacetylated TFE-Cys. In the
presence of o-phenylenediamine (OPD), as a bifunctional nucleophilic trapping agent, the major
product formed was 2-(difluoromethyl)benzimidazole. This product results from initial reaction
of difluorothionoacetyl fluoride with one of the amino groups of OPD, followed by a condensation
reaction between the thionoacyl group and the adjacent amino group of OPD. In incubations
with S-(2-chloro-1,1,2-trifluorofluoroethyl)-L-cysteine (CTFE-Cys) and S-(2,2-dichloro-1,1-
difluorofluoroethyl)-L-cysteine (DCDFE-Cys), formation of thionoacylated cysteine S-conjugates
was also observed by GC-MS analysis, indicating formation of the corresponding thionoacyl
fluorides. However, according to 19F-NMR analysis, chlorofluorothionoacyl fluoride-derived
products accounted for only 10% of the CTFE-Cys converted. In the presence of OPD, next to
the corresponding 2-(dihalomethyl)benzimidazoles, 2-mercaptoquinoxaline was identified as
the main product in incubations with CTFE-Cys. When chlorofluorothionoacylating species
were generated from the unsaturated S-(2-chloro-1,2-difluorovinyl)-L-cysteine (CDFV-Cys),
2-(chlorofluoromethyl)benzimidazole and 2-mercaptoquinoxaline were also found as OPD
adducts. However, with CDFV-Cys the ratio of 2-(chlorofluoromethyl)benzimidazole to
2-mercaptoquinoxaline was 12-fold higher than in the case of CTFE-Cys. These results suggest
an important second mechanism of formation of 2-mercaptoquinoxaline with CTFE-Cys. The
formation of 2-mercaptoquinoxaline could also be explained by reaction of OPD with 2,3,3-
trifluorothiirane as a second reactive intermediate for CTFE-Cys. Comparable results were
obtained when comparing OPD adducts from DCDFE-Cys and TCV-Cys. Both DCDFE-Cys
and TCV-Cys form dichlorothionoacylating species. However, DCDFE-Cys forms 21-fold more
2-mercaptoquinoxaline than TCV-Cys, which may be explained by its capacity to form 3-chloro-
2,2-difluorothiirane next to dichlorothionoacyl fluoride. In order to explain the apparent
differences in the preference of thiols to form different reactive intermediates, free enthalpies
of formation (∆fG) of thiolate anions and their possible rearrangement products, thionoacyl
fluorides and thiiranes, derived from TFE-Cys, CTFE-Cys, and DCDFE-Cys, were calculated
by ab initio calculations. For TFE-thiolate, formation of difluorothionoacetyl fluoride is
energetically favored over formation of the thiirane. In contrast, the thiirane pathway is favored
over the thionoacyl fluoride pathway for CTFE- and DCDFE-thiolates. The results of these
quantum chemical calculations appear to be consistent with the experimental data.
FE), hexafluoropropene, and 1,1,2-trichloro-3,3,3-tri-
fluoropropene, as reviewed recently (1-3). Formation of
the cysteine S-conjugates results from initial enzymic
conjugation of the cysteinyl thiol group of glutathione to
the parent haloalkenes, followed by subsequent hydroly-
sis by γ-glutamyltranspeptidases and dipeptidases present
In tr od u ction
Cysteine S-conjugates are intermediates in the bioac-
tivation mechanisms responsible for the nephrotoxicity
of a number of halogenated alkenes, such as hexachloro-
1,3-butadiene, tetrafluoroethylene (TFE),1 chlorotrifluo-
roethylene (CTFE), 1,1-dichloro-2,2-difluoroethylene (DCD-
1
Abbreviations: CDFV-Cys, S-(2-chloro-1,2-difluorovinyl)-L-cys-
teine; CTFE-Cys, S-(2-chloro-1,1,2-trifluoroethyl)-L-cysteine; S-(2,2-
dibromo-1,1-difluoroethyl)-L-cysteine; DCDFE-Cys, S-(2,2-dichloro-1,1-
difluoroethyl)-L-cysteine; MOP, 3-mercapto-2-oxopropionic acid; OPD,
o-phenylenediamine; TCV-Cys, S-(1,2,2-trichlorovinyl)-L-cysteine; TFE-
Cys, S-(1,1,2,2-tetrafluorethyl)-L-cysteine; TFE-PMS, N-(difluorothiono-
acetyl)-S-(1,1,2,2-tetrafluoroethyl)-L-cysteine.
* To whom correspondence should be addressed. Tel: 020-4447595;
fax: 020-4447610; e-mail: command@chem.vu.nl.
† Present address: School of Biological Sciences, University of
Surrey, Guildford, Surrey GU2 5XH, U.K.
X Abstract published in Advance ACS Abstracts, August 15, 1996.
S0893-228x(96)00049-5 CCC: $12.00 © 1996 American Chemical Society

Bioactivation of Cysteine Conjugates of Haloalkenes
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1093
F igu r e 2. Anticipated products of reactions between o-phen-
ylenediamine (OPD) and dihalothionoacylating reactive inter-
mediates (dihalothionoacyl halides and dihalothioketenes) and
3-halo-2,2-difluorothiirane. X1, X2, and X3 represent fluorine and/
or chlorine atoms. Anticipated products from R-thiolactones are
not shown since these reactive intermediates are only formed
from bromine-containing cysteine S-conjugates (18).
F igu r e 1. Proposed reactive species formed upon the â-elim-
ination reactions from S-(2,2-dihalo-1,1-difluoroethyl)-L-cysteine
(A) and S-(trihalovinyl)-L-cysteine S-conjugates (B).
(19). However, these authors did not find experimental
proof for the formation of thiiranes from CTFE-Cys and
DCDFE-Cys.
in several tissues. The toxicity of cysteine S-conjugates
of haloalkenes is directed to the proximal tubule of the
kidney, due to the presence of active uptake mechanisms
and the high activity of the bioactivating enzyme, cys-
teine S-conjugate â-lyase (â-lyase), in this part of the
nephron (1, 3). Cleavage of cysteine S-conjugates of
haloalkenes by â-lyase results in the formation of reactive
intermediates which may bind covalently to essential
biomacromolecules, such as proteins, lipids, DNA, and
RNA (4-8), thus initiating processes leading to cell death,
mutagenicity, or carcinogenicity. Reactive intermediates
which have been proposed for S-(2,2-dihalo-1,1-difluoro-
ethyl)-L-cysteine and S-(trihalovinyl)-L-cysteine (9-17)
are shown in Figure 1.
The detection of difluorothionoacylated products from
S-(1,1,2,2-tetrafluoroethyl)-L-cysteine (TFE-Cys, Figure
1A: X1,X2 ) F, F) is consistent with thionoacylation of
target nucleophiles by difluorothionoacetyl fluoride formed
by R-elimination of a fluoride anion from the initial
thiolate (11, 13, 14, 16). Similar evidence has been
obtained for the formation of the analogous reactive
intermediate, chlorofluorothionoacetyl fluoride, from S-(2-
chloro-1,1,2-trifluoroethyl)-L-cysteine (CTFE-Cys, Figure
1A: X1,X2 ) F, Cl) (10, 12, 14-16). However, the
relatively low yields of chlorofluorothionoacetylated nu-
cleophiles and the accompanying high yield of fluoride
anion observed in 19F-NMR studies may indicate insta-
bility of the adduct formed or the formation of another
reactive intermediate. Based on the formation of gly-
oxylate, Finkelstein et al. (18) proposed R-thiolactones
(Figure 1A) as reactive intermediates formed from bro-
mine-containing cysteine S-conjugates. However, be-
cause glyoxylate formation was not observed in the case
of TFE-Cys, CTFE-Cys, and S-(2,2-dichloro-1,1-difluoro-
ethyl)-L-cysteine (DCDFE-Cys, Figure 1A: X1,X2 ) F, Cl),
this kind of reactive intermediate does not appear to be
formed from non-bromine-containing cysteine S-conju-
gates. Formation of 2,2,3-trifluorothiirane, as another
alternative reactive intermediate, has originally been
proposed by Dohn et al. (9). Very recently, the formation
of trihaloalkenes next to glyoxylate in incubations with
bromine-containing cysteine S-conjugates was claimed to
be the first experimental proof for formation of thiiranes
In the present study, the nature of covalent binding of
â-elimination products of three S-(2,2-dihalo-1,1-difluo-
roethyl)-L-cysteine conjugates was investigated using GC-
MS and 19F-NMR. Because it is anticipated that the
initial products of the reaction of nucleophiles with
thionoacyl halides and thiiranes still contain an electro-
philic center, o-phenylenediamine (OPD) was chosen as
a trapping agent. With OPD, the introduced electrophilic
center can rapidly react intramolecularly with the adja-
cent free amino group (Figure 2). Comparable bifunc-
tional nucleophilic trapping agents are DNA bases, such
as adenosine, which have been used to trap bifunctional
electrophiles, such as 2-bromoacetaldehyde, forming
ethenoadenosine (20). However, the advantage of OPD
over adenosine is that the products formed can easily be
analyzed by GC-MS.
It was anticipated that analysis of OPD adducts might
discriminate between dihalothionoacylating reactive in-
termediates (dihalothionoacyl halides and dihalothio-
ketenes) and 3-halo-2,2-difluorothiiranes (Figure 2). The
possible contribution of thiiranes was investigated by
comparing OPD adducts formed from S-(2,2-dihalo-1,1-
difluoroethyl)-L-cysteine with those formed from S-(tri-
halovinyl) S-conjugates, which can be bioactivated to
similar thionoacylating reactive species (17) but cannot
form thiirane species (Figure 1B).
The results in the present study support the formation
of 3-halo-2,2-difluorothiiranes next to thionoacyl fluorides
from S-(2,2-dihalo-1,1-difluoroethyl)-L-cysteine conju-
gates. A theoretical study, using ab initio calculations
of the free enthalpies of formation (∆fG) of reactants and
products, was also performed to compare the thermody-
namics of the proposed rearrangement reactions.
Ma ter ia ls a n d Meth od s
Ch em ica ls a n d Syn th esis. Pyridoxal hydrochloride, o-
phenylenediamine (OPD), copper(II) chloride hydrate, and 1,2-
difluoro-1,1,2,2-tetrachloroethane were obtained from Aldrich
(Brussels, Belgium). 1,2-Difluoro-1,2-dichloroethane was pre-
pared by heating 1,2-difluoro-1,1,2,2-tetrachloroethane in pres-

1094 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Commandeur et al.
ence of zinc dust in ethanol, by the same procedure as described
for the synthesis of 1,1-dichloro-2,2-difluoroethylene (21). Di-
fluoroacetic acid (DFA) was obtained from Acros Chimica (Geel,
Belgium), and chlorofluoroacetic acid was prepared according
to Young and Tarrant (22). Difluoroacetic acid anhydride was
prepared from difluoroacetic acid according to Sawicki (23). TFE-
Cys, CTFE-Cys, S-(2,2-dichloro-1,1-difluoroethyl)-L-cysteine (DCD-
FE-Cys), S-(2,2-dibromo-1,1-difluoroethyl)-L-cysteine (DBDFE-
Cys), S-(1,2-dichlorovinyl)-L-cysteine (1,2-DCV-Cys), S-(2,2-
dichlorovinyl)-L-cysteine (2,2-DCV-Cys), and S-(1,2,2-trichloro-
vinyl)-L-cysteine (TCV-Cys) were synthesized as described
previously (12, 17, 24). Deuterium-labeled CTFE-Cys, [ethyl-
2-D]-S-(2-chloro-1,1,2-trifluoroethyl)-L-cysteine, was synthesized
according to the same method as described for CTFE-Cys, after
exchanging all acidic protons of the reagents for deuterium
atoms prior to the reaction. Exchange of protons was achieved
by repeatedly dissolving the reagents in deuterium oxide and
subsequent freeze-drying. 1H-NMR, 19F-NMR, and GC-MS
proved the exchange of the proton of the 2-chloro-1,1,2-trifluo-
roethyl moiety by a deuterium atom.
144 (21), 88 (30), 84 (9), 43 (100).
After completion of the dehydrohalogenation reaction, the
reaction mixture was acidified to pH 2, filtered to remove silver
salts, and extracted by ethyl acetate. The ethyl acetate fraction
was evaporated to dryness, and the residue was treated with
50 mL of 6 N hydrochloric acid at 80 °C for 16 h, in order to
N-deacetylate N-acetyl-S-(2-chloro-1,2-difluorovinyl)-L-cysteine.
After evaporation of solvent by rotavap, the residue was purified
by preparative thin layer chromatography using silica gel as
stationary phase and 1-propanol/water (70/30 v/v) as eluent. The
fraction at Rf 0.6, which was UV-active and ninhydrin-reactive,
was collected. 1H-NMR analysis revealed that the final product
was free from N-acetyl-S-(2-chloro-1,2-difluorovinyl)-L-cysteine
and N-acetyl-S-(1,2-dichloro-1,2-difluoroethyl)-L-cysteine: δ (D2O)
3.30-3.50 (2H, m), 4.20-4.35 (1H, m). GC-MS analysis of the
product, after subsequent acetylation by acetic anhydride and
methylation by ethereal diazomethane, revealed the two isomers
of the methyl esters of N-acetyl-S-(2-chloro-1,2-difluorovinyl)-
L-cysteine as the sole products. This indicates that the final
product consists of S-(2-chloro-cis-1,2-difluorovinyl)-L-cysteine
and S-(2-chloro-trans-1,2-difluorovinyl)-L-cysteine. Because both
isomers were anticipated to yield identical products upon the
â-elimination reactions, no attempts were undertaken to sepa-
rate both isomers.
2-Hydr oxyqu in oxalin e an d 2-Meth oxyqu in oxalin e. 2-Hy-
droxyquinoxaline was obtained from Aldrich-Chemie (Steinhein,
West Germany). In our GC-MS conditions as described below,
2-hydroxyquinoxaline demonstrated a very poor chromatogra-
phy yielding a broad peak ranging from 8.7 to 9.1 min and the
following electron-impact mass spectrum: m/ z (intensity %) 146
(100, M•+), 118 (68), 91 (39), 64 (28), 63 (32).
Methylation of 2-hydroxyquinoxaline by diazomethane yielded
a sharp peak with a retention time of 8.53 min and the electron-
impact mass spectrum: m/z (intensity %) 160 (100, M•+), 132
(84), 131 (98), 104 (48), 90 (20), 77 (42), 63 (40), 51 (37). A minor
product with retention time of 6.95 min also showed a molecular
ion at m/ z 160. We propose that this product might be derived
from N-methylation of the tautomeric form of 2-hydroxyquino-
line: m/ z (rel intensity %, assignment) 160 (100, M•+), 159 (47,
M•+ - H), 131 (56), 130 (32), 103 (34), 90 (57), 76 (14), 75 (13),
64 (16), 63 19).
Syn th esis of N-Acetyl-S-(2-ch lor o-1,2-d iflu or ovin yl)-L-
cystein e. The strategy to synthesize S-(2-chloro-1,2-difluorovi-
nyl)-L-cysteine was the preparation of N-acetyl-S-(2-chloro-1,2-
difluorovinyl)-L-cysteine, followed by deacetylation under acidic
conditions, as used previously (12).
In order to synthesize N-acetyl-S-(2-chloro-1,2-difluorovinyl)-
L-cysteine, the method as described by Gandolfi et al. (25) was
used. In short, 5 g (30 mmol) of N-acetyl-L-cysteine was
dissolved in 250 mL of liquid ammonia, and 1.4 g (60 mmol) of
metallic sodium was added in small pieces. Subsequently, 20
g (150 mmol) of 1,2-dichloro-1,2-difluoroethylene was added
slowly to the reaction mixture which was kept at -50 °C. After
4 h, the cooling was removed, and the ammonia and excess 1,2-
dichloro-1,2-difluoroethylene were evaporated overnight by a
slow nitrogen stream while stirring. The residue was dissolved
in 100 mL of 2 N hydrochloric acid and extracted two times with
100 mL of ethyl acetate. The ethyl acetate fractions were
combined, treated with active charcoal, and filtered. The solvent
was removed by rotaevaporation under vacuum, yielding ap-
proximately 6 g of a thick yellow oil. However, according to
1
analysis by H-NMR and GC-MS, the sole product formed was
N-acetyl-S-(1,2-dichloro-1,2-difluoroethyl)-L-cysteine, indicating
that under these reaction conditions an addition reaction took
place instead of the expected substitution reaction. Similar
results were obtained when the reaction was performed in
methanol instead of liquid ammonia (data not shown).
2-Mer ca p toqu in oxa lin e a n d 2-Meth ylth ioqu in oxa lin e.
2-Mercaptoquinoxaline was prepared from 2-hydroxyquinoxa-
line by the procedure described by Scheeren et al. (26). Anhy-
drous NaHCO3 (4 mmol) was added slowly with stirring to a
mixture of 2-hydroxyquinoxaline (1 mmol) and tetraphosphorus
decasulfide (0.6 mmol) in diethyl ether, and refluxed for 5 h.
The organic phase was evaporated to dryness, the residue was
treated with 5 mL of 0.2 N HCl at 40 °C for 2 h and cooled, and
the suspension was extracted with ethyl acetate (3 × 5 mL).
The combined ethyl acetate fractions were evaporated to dryness
to obtain the product. 1H-NMR of the product: (CDCl3, relative
to 3-(trimethylsilyl)propionic acid): δ (assignment, intensity,
multiplicity) 7.30-7.50 (Hb, 2H, m), 7.55-7.85 (Hc, 2H, m), 8.65
(Hc, s). GC-MS analysis: retention time 8.25 min; electron-
impact mass spectrum: m/ z (intensity %, assignment) 162 (100,
Characterization of N-acetyl-S-(1,2-dichloro-1,2-difluoro-
ethyl)-L-cysteine: 1H-NMR (D2O): δ (intensity, multiplicity) 2.00
(3H, s), 3.20-3.65 (2H, m), 4.45-4.55 (1H, m), 6.62 (1H, d of d,
3
2J FH ) 48 Hz, J FH ) 6 Hz). Electron-impact GC-MS of methyl
ester: m/ z (rel intensity %, multiplicity, assignment): 309 (0.1,
2Cl, M•+), 250 (3.5, 2Cl, M•+ - COOCH3 and M•+ - NH2COCH3),
208 (7.6, 2Cl), 176 (18), 144 (8), 134 (14), 117 (19), 88 (57), 43
(100).
We prepared N-acetyl-S-(2-chloro-1,2-difluorovinyl)-L-cysteine
by stirring a 50 mL solution of 50 mM N-acetyl-S-(1,2-dichloro-
1,2-difluoroethyl)-L-cysteine in cold 0.1 N NaOH in the presence
of 100 mM silver nitrate. The progress of the reaction was
followed by analyzing 100 µL fractions by GC-MS. The 100 µL
fractions were added to 2 mL of 2 N hydrochloric acid and
extracted with 3 mL of ethyl acetate. The ethyl acetate layer
was separated, treated with ethereal diazomethane, and ana-
lyzed by electron-impact GC-MS. Using this procedure, two
products, with retention times of 11.40 and 11.50 min, in a ratio
of approximately 1:2 were obtained. We attribute these two
products to the two isomers which were anticipated, having a
cis- and trans-(difluorovinyl) moiety, respectively. The electron-
impact mass spectra of both isomers were identical and consis-
tent with that of N-acetyl-S-(2-chloro-1,2-difluorovinyl)-L-
cysteine.
M
•+), 134 (10), 129 (24, M•+ - SH), 118 (75, M•+ - CS), 108 (8),
102 (21), 90 (10).
Methylation of a solution of 2-mercaptoquinoxaline in ethyl
acetate by ethereal diazomethane yielded 2-methylthioquin-
oxaline: 1H-NMR (CDCl3, relative to 3-(trimethylsilyl)propionic
acid): δ (assignment, intensity, multiplicity) 2.62 (Hd, 3H, s),
7.50-7.70 (Hb, 2H, m), 7.82-8.02 (Ha, 2H, m), 8.55 (Hc, 1H, s).
GC-MS analysis: retention time 8.45 min; electron-impact mass
spectrum: m/ z (rel. intensity %, assignment) 176 (100, M•+),
175 (32, M•+ - H), 161 (13, M•+ - CH3), 143, (27, M•+ - SH),
134 (17), 131 (25, -CHS), 103 (25), 102 (27), 90 (13).
2-(Diflu or om eth yl)ben zim id a zole was prepared from di-
fluoroacetic anhydride (3 mmol) and OPD (2 mmol) in dry
acetonitrile, according to the method described by Sawicki (23).
After removal of the solvent, the residue was dissolved in 2 N
HCl and extracted with ethyl acetate to remove diacylated OPD;
the product was recovered by ethyl acetate extraction of the
EI-GC-MS of methyl ester: m/ z (rel intensity %, multiplicity,
assignment): 273 (0.05, 1Cl, M•+), 214 (4, 1Cl, M•+ - COOCH3
and M•+ - NH2COCH3), 211 (6, 1Cl, M•+ - C2F2), 172 (10, 1Cl),

Bioactivation of Cysteine Conjugates of Haloalkenes
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1095
aqueous phase adjusted to pH 9 with NaOH. 1H-NMR of the
product (CDCl3, relative to 3-(trimethylsilyl)proprionic acid): δ
(assignment, intensity, multplicity) 6.92 (Hb, 1H, t, 2J FH ) 53.4
Hz), 7.32-7.42 (Ha, 4H, m), 11.5 (Hc, 1H, broad peak). 19F-NMR
of the product (H2O, relative to trifluoroacetic acid): δ -41.93
(d, 2J FH ) 51.01 Hz). GC-MS analysis: retention time 6.88 min;
electron-impact mass spectrum: m/ z (rel. intensity %, assign-
ment) 168 (100, M•+), 148 (30, M•+ - HF), 129 (19, M•+ - HF -
F), 121 (18), 118 (14, M•+ - CF2), 103 (18), 90 (37).
Methylation of the ethyl acetate solution of 2-(difluoromethyl)-
benzimidazole by ethereal diazomethane yielded N-methyl-2-
(difluoromethyl)benzimidazole. GC-MS analysis: retention time
6.57 min; electron-impact mass spectrum: m/ z (rel. intensity
%, assignment) 182 (100, M•+), 181 (23, M•+ - H), 167 (5, M•+
- CH3), 163 (10, M•+ - F), 161 (5, M•+ - H - HF), 147 (3, M•+
- HF - CH3), 131 (23, M•+ - CHF2), 90 (8).
2-(Dich lor om et h yl)b en zim id a zole was prepared from
dichloroacetyl chloride (3 mmol) and OPD (2 mmol) in dry
acetonitrile, according to the method described by Sawicki (23).
Because the major product in this reaction mixture was bis-
dichloroacetylated OPD, no 1H-NMR spectrum of pure 2-(dichlo-
romethyl)benzimidazole could be obtained. GC-MS analysis,
however, showed formation of a product (amounting to 10% of
the reaction products) with a mass spectrum consistent with
that of 2-(dichloromethyl)benzimidazole. GC-MS analysis: re-
tention time 9.06 min; electron-impact mass spectrum: m/ z (rel.
intensity %, multiplicity, assignment) 200 (18, 2Cl, M•+), 165
(88, 1Cl, M•+ - Cl), 135 (19), 129 (20), 105 (100), 77 (40).
Methylation of reaction mixture by ethereal diazomethane
converted this product in a compound with a mass spectrum
consistent with that of N-methyl-2-(dichloromethyl)benzimida-
zole. GC-MS analysis: retention time 9.15 min; electron-impact
mass spectrum: m/ z (rel. intensity %, multiplicity, assignment)
214 (19, 2Cl, M•+), 179 (100, 1Cl, M•+ - Cl), 143 (10), 129 (7),
102 (8).
F igu r e 3. 19F-NMR spectra of 16 h incubations of S-(1,1,2,2-
tetrafluoroethyl)-L-cysteine (TFE-Cys) (15 mM) in a â-lyase-
mimetic model consisting of pyridoxal (0.25 mM) and Cu(II) (100
µM) in 50 mM potassium borate buffer, pH 8.6. (A) In the
absence of Cu(II). (B) In the presence of Cu(II). TFE-PMS,
N-(difluorothionoacetyl)-S-(1,1,2,2,-tetrafluoroethyl)-L-cys-
teine; DFA, difluoroacetic acid; DFTA, difluorothio(no)acetic
acid.
Ca lcu la tion of F r ee En th a lp ies of F or m a tion of Rea c-
tive In ter m ed ia tes. Free energies of formation of the thiol
compounds and possible rearrangement products were calcu-
lated by ab initio energy calculations using the quantum
chemical program package GAMESS (27) on the Cyber 205 of
the Academic Computer Centre of Amsterdam (SARA). The
geometries of the parent compound and reaction products were
fully optimized, implying variation of bond distances, bond
angles, and torsion angles, using the STO-3G minimal basis set
(28). After geometry optimization, a SV 6-31G SCF-LCAO-MO
energy calculation was performed; d-orbitals were included in
calculating sulfur-containing compounds (29).
In cu ba tion of Cystein e Con ju ga te in th e P r esen ce of
â-Lya se Mim etic Mod el. Cysteine S-conjugates (15 mM) were
incubated at 37 °C in 50 mM sodium borate buffer (pH 8.6) in
the presence of pyridoxal (0.25 mM) and metal ion (100 µM). In
reaction mixtures containing the trapping agent OPD (4 mM),
pyridoxal was increased to 2 mM. To identify fluorine-contain-
ing reaction products, cysteine S-conjugates were incubated at
an initial concentration of 15 mM; after 16 h, trifluoroacetic acid
(TFA) was added as internal standard and the incubation
mixtures were analyzed with 19F-NMR.
Free enthalpies of hydration of halide ions were obtained from
Friedman and Krishnan (30).
Resu lts
In cu ba tion s of S-(2,2-Dih a lo-1,1-d iflu or oeth yl)-L-
cyst ein e Con ju ga t es. When the fluoro-containing
S-(2,2-dihalo-1,1-difluoroethyl)-L-cysteine conjugates (15
mM) were incubated for 16 h in the presence of 0.5 mM
pyridoxal and 100 µM Cu(II), the reaction mixtures with
CTFE-Cys and DCDFE-Cys developed a dark brown
polymeric material. The reaction mixture of TFE-Cys,
however, remained a clear pale yellow solution. After
16 h, the parent cysteine S-conjugates were not detected
by 19F-NMR anymore, indicating their complete degrada-
tion in the â-lyase model system (Figures 3B, 4B, and
5B). In the presence of 4 mM OPD, however, significant
amounts of cysteine S-conjugates were still present, and
might be due to the binding of OPD to pyridoxal. In the
absence of Cu(II), the unchanged cysteine S-conjugates
were still observed after 16 h (Figures 3A, 4A, and 5A)
and there was no evidence for the formation of any other
fluorinated products.
For GC-MS analysis of products, incubations were acidified
to pH 2 and subsequently extracted by ethyl acetate. Ethyl
acetate fractions were injected on GC-MS directly and after
methylation by ethereal diazomethane.
In cu ba tion of Cystein e S-Con ju ga tes in th e P r esen ce
of Ra t Ren a l Cytosol. Rat renal cytosol was prepared from
kidneys of male Wistar rats (180-200 g) as described previously
(12). Cysteine S-conjugates (4 mM) were incubated in 50 mM
sodium borate buffer (pH 8.6) at 37 °C in the presence of 4
mg/mL cytosolic protein and 0.5 mM R-keto-γ-methiolbutyric
acid. After 30 min of incubation, reaction mixtures were
acidified to pH 2 and extracted two times with ethyl acetate.
Combined ethyl acetate fractions were treated with ethereal
diazomethane, concentrated, and analyzed by GC-MS.
In str u m en ta l Con d ition s. All GC-MS analyses were car-
ried out on a Hewlett Packard 5890/MSD system. A CP Sil-SE
30 capillary coloumn (25 m, 0.22 mm i.d.) obtained from
Chrompack Ned. BV. was used. The operation temperatures
were 280 °C (split injector) and 280 °C (ion source). Electron-
impact ionization (electron energy of 70 eV) was used. The
carrier gas was helium, at a flow rate of about 3 mL/min, and
the column head pressure was 80 kPa. The column temperature
was programmed from 60 °C (2.5 min) to 280 °C at 20 °C/min.
19F-NMR spectra of incubations were measured directly in 5
mm NMR tubes using a Bruker MSL 400 system operating at
376.43 Hz. Chemical shifts are with reference to trifluoroacetic
acid as internal standard.
The results of 19F-NMR and GC-MS analysis of the
reaction products of the S-(2,2-dihalo-1,1-difluoroethyl)-
L-cysteine conjugates are discussed separately.
(1) TF E-Cys. Upon 19F-NMR analysis of incubations
of TFE-Cys with the â-lyase-mimetic system, several
fluorine-containing products were observed (Figure 3B).
These products are similar to those described previously
for â-cleavage of TFE-Cys by rat renal cytosol (11). In
addition to fluoride anion, difluoroacetic acid (DFA), and
difluorothio(no)acetic acids (DFTA), formation of the self-

1096 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Commandeur et al.
F igu r e 4. 19F-NMR spectra of 16 h incubations of S-(2-chloro-
1,1,2-trifluoroethyl)-L-cysteine (CTFE-Cys, 15 mM) in a â-lyase-
mimetic model consisting of pyridoxal (0.25 mM) and Cu(II) (100
µM) in 50 mM potassium borate buffer, pH 8.6. (A) Incubation
in the absence of Cu(II); (B) incubation in the presence of Cu(II);
(C) details of the 19F-NMR spectrum (B); (D) details of the 19F-
NMR spectrum of the incubation of 4 mM CTFE-Cys with rat
renal cytosol (5 mg/mL). CFA, chlorofluoroacetic acid.
F igu r e 6. Detail of 19F-NMR spectra of 16 h incubations of
S-(1,1,2,2-tetrafluoroethyl)-L-cysteine (15 mM) in the â-lyase-
mimetic model in the presence (top) or absence (bottom) of 4
mM o-phenylenediamine (OPD). The â-lyase-mimetic model
consisted of 2 mM pyridoxal and 0.1 mM Cu(II) in 50 mM
potassium borate buffer, pH 8.6.
(Figure 7A) identical to that of 2-(difluoromethyl)benz-
imidazole. Methylation by diazomethane yielded a peak
at 6.57 min with a mass spectrum identical to that of
synthetical N-methyl-2-(difluoromethyl)benzimidazole (Fig-
ure 7D).
GC-MS analysis of cytosolic incubations with TFE-Cys
showed formation of the same products as in the â-lyase
model system. However, in addition, dimethylated
S-(1,1,2,2-tetrafluoroethyl)-3-mercapto-2-oxopropionic acid
(TFE-MOP) was also observed,2 pointing to transamina-
tion or oxidative deamination of TFE-Cys in cytosol.
Addition of (aminooxy)acetic acid (1 mM) to the cytosolic
incubation blocked â-elimination reaction; however, for-
mation of TFE-MOP was increased 4-fold, suggesting
that amino acid oxidases are the major source of this
deamination product in rat renal cytosol (data not
shown).
(2) CTF E-Cys. At the end of incubation of CTFE-Cys
with the â-lyase-mimetic system, fluoride anion was the
major product in the 19F-NMR spectrum (Figure 4B). In
addition, the spectrum revealed relatively small amounts
of chlorofluoroacetic acid (doublet at δ -61.31 ppm (2J FH
) 52.2 Hz)) and doublets at δ -56.38 ppm (J ) 50.3 Hz)
and -57.48 ppm (J ) 50.8 Hz) (Figure 4C). Also
observed were compound(s) with chemical shifts similar
to but significantly different from CTFE-Cys at δ -9.25
to -12.75 ppm and at δ -72.9 ppm. When CTFE-Cys
was incubated at a concentration of 4 mM with rat renal
cytosol (5 mg/mL) for 16 h, similar products were
observed (Figure 4D).
F igu r e 5. 19F-NMR spectra of 16 h incubations of S-(2,2-
dichloro-1,1-difluoroethyl)-L-cysteine (DCDFE-Cys, 15 mM) in
a â-lyase-mimetic model consisting of pyridoxal (0.25 mM) and
Cu(II) (100 µM) in 50 mM potassium borate buffer, pH 8.6. (A)
Incubation in the absence of Cu(II). (B) Incubation in presence
of Cu(II).
adduct N-(difluorothionoacetyl)-S-(1,1,2,2-tetrafluoroeth-
yl)-L-cysteine (TFE-PMS; also called pseudomercapturic
acid (11)) was observed. Based on the ratio of integrals
at the end of the incubation, the concentration of the
fluoride anion was estimated to be approximately 2.4-
2.8 times higher than the sum of the difluoro(thiono)-
acetyl-containing compounds. GC-MS analysis of meth-
ylated extracts showed formation of TFE-PMS (retention
time 7.82 min) as a major product and, in addition, a
small amount of N-(difluoroacetyl)-S-(1,1,2,2-tetrafluo-
roethyl)-L-cysteine (TFE-PMO (11)) (retention time 6.7
min).
When nonenzymic incubations were carried out in the
presence of 4 mM OPD after 16 h of incubations, a new
fluorine-containing product with δ -41.92 ppm (doublet,
J ) 51.0 Hz) was detected upon analysis by 19F-NMR
(Figure 6, top panel). The chemical shift and coupling
constant were identical to those of synthetical 2-(difluo-
romethyl)benzimidazole. The identity of this product was
further supported by GC-MS analysis of an ethyl acetate
extract revealing formation of a product with a retention
time (6.92 min) and electron-impact mass spectrum
GC-MS analysis of methylated extracts revealed the
formation of two products with slightly different retention
2
For identification, references of 3-mercapto-2-oxopropionic acid
S-conjugates were prepared by incubating cysteine S-conjugates (4 mM)
for 60 min in the presence of 1 mg/mL amino acid oxidase (Sigma, St.
Louis, MO) and 40 U/mL catalase (Boehringer Mannheim GmbH,
Mannhein, Germany). Incubations were performed in 50 mM potas-
sium phoshate (pH 7.4) at 37 °C. Due to keto-enol tautomerisation of
the 3-mercapto-2-oxopropionic acid S-conjugate, methylation by diazo-
methane also occurs at the enol oxygen. Because the double bond
formed can be cis- and trans-substituted, two peaks are observed in
GC-MS-analysis.

Bioactivation of Cysteine Conjugates of Haloalkenes
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1097
F igu r e 7. Electron-impact mass spectra of the 2-(dihalomethyl)benzimidazoles formed in incubations of S-(1,1,2,2-tetrafluoroethyl)-
L-cysteine (TFE-Cys, 20 mM) (A), S-(2-chloro-1,1,2-trifluoroethyl)-L-cysteine (CTFE-Cys, 20 mM) (B), or S-(2,2-dichloro-1,1-
difluoroethyl)-L-cysteine (DCDFE-Cys, 20 mM) (C) in a â-lyase-mimetic model consisting of pyridoxal (2 mM), Cu(II) (100 µM), and
OPD (4 mM) in 50 mM potassium borate buffer, pH 8.6. Treatment of the 2-(dihalomethyl)benzimidazoles with ethereal diazomethane
converted these compounds into the corresponding N-methyl-2-(dihalomethyl)benzimidazoles (D, E, and F).
times, 9.85 vs 9.91 min, but with almost identitical
electron-impact mass spectra (Figure 8A). These prod-
ucts were attributed to the pseudomercapturic acid
resulting from N-chlorofluorothionoacetylation of CTFE-
Cys; the fragmentation pattern was analogous to that of
dimethylated TFE-PMS (11), except for substitution of
fluorine by chlorine atoms in several fragments. Due to
the introduction of a racemic N-(chlorofluorothionoacetyl)
group into L-cysteine, two diastereomeric products were
detected by capillary GC. The relative intensity of these
products on GC-MS approximately correlated with the
relative intensity of the doublets at δ -56.38 ppm and
at δ -57.48 ppm observed by 19F-NMR, suggesting that
these two doublets represent the N-(chlorofluorothiono-
acetyl) groups of the two diastereomeric pseudomercap-
turic acids. Based on the ratio of integrals of the 19F-
NMR spectrum, at the end of incubation, the concentra-
tion of the fluoride anion was more than 10 times higher
than the total of chlorofluorothionoacetyl-containing
products.
time and mass spectrum identical to that of synthetical
2-mercaptoquinoxaline (Figure 9A). Methylation of this
extract showed formation of 2-methylthioquinoxaline
(Figure 9B), which has a retention time of 8.49 min.
When incubations were performed in presence of [ethyl-
2-D]-S-(2-chloro-1,1,2-trifluoroethyl)-L-cysteine, the mo-
lecular ion of 2-methylthioquinoxaline increased one
mass unit, proving incorporation of a two-carbon frag-
ment derived from the 2-chloro-1,1,2-trifluoroethyl moiety
of CTFE-Cys.
The retention time and mass spectrum of the minor
peak at 8.70 min in the nonmethylated extract were
identical to those of 2-hydroxyquinoxaline. Methylation
of this product resulted in a retention time shift to 8.50
min as was observed with the reference compound.
Although 2-methoxyquinoxaline and 2-methylthioqui-
noxaline show almost identical retention times, these two
products can easily be distinguished by selected ion
chromatography at m/ z 160, which is absent in 2-meth-
ylthioquinoxaline, and at m/ z 176, which is absent in
2-methoxyquinoxaline.
19F-NMR analysis of the reaction mixture of CTFE-Cys
in presence of OPD did not result in any additional
fluorine-containing products (data not shown) in com-
parison to that shown in Figure 4C. However, using GC-
MS analysis of ethyl acetate extracts showed peaks at
8.01 and 8.25 min as well as a broad peak at 8.70 min,
which were not present in incubations without OPD. The
EI-mass spectrum of the peak at 8.25 min had a retention
The EI-mass spectrum of the peak at 8.01 min revealed
a molecular ion of m/z 184, having a chlorine isotope
pattern (Figure 7B). This mass spectrum is consistent
with that of 2-(chlorofluoromethyl)benzimidazole. Al-
though the identity of this adduct was not proved
unequivocally by synthesis of the reference compound,
its fragmentation pattern and retention time are consis-

1098 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Commandeur et al.
ylation of the presumed 2-(chlorofluoromethyl)benzimi-
dazole yielded a new compound with retention time 7.91
min and a mass spectrum consistent with that of N-meth-
yl-2-(chlorofluoromethyl)benzimidazole (Figure 7E).
As in the case of TFE-Cys, in incubations of CTFE-
Cys with rat renal cytosol the same products could be
identified as in the â-lyase model system. However, the
yield of products derived from chlorofluorothionoacyl
fluoride appeared to be relatively higher in the cytosolic
incubations. This may be due to decomposition of the
chlorofluorothionoacyl moieties during overnight incuba-
tion with the â-lyase-mimetic system (16).
In cytosolic incubations of CTFE-Cys, two peaks at-
tributed to cis- and trans-isomers of dimethylated S-(2-
chloro-1,1,2-trifluoroethyl)-3-mercapto-2-oxopropionic acid
(CTFE-MOP) were also observed by GC-MS-analysis,2 at
6.66 and 6.98 min, pointing to a transamination or
oxidative deamination reaction in cytosol.
(3) DCDF E-Cys. As in incubations with CTFE-Cys,
fluoride anion was the major product observed in incuba-
tions of the â-lyase model with DCDFE-Cys (Figure 5B).
The very weak signal observed at approximately -5 ppm
also points to a DCDFE-Cys-derived product. Analysis
of methylated ethyl acetate extracts of acidified reaction
mixtures by GC-MS also pointed to the formation of small
amounts of the pseudomercapturic acid, resulting from
N-dichlorothionoacetylation of DCDFE-Cys (Figure 8B),
with a retention time of 11.66 min.
When DCDFE-Cys incubations were carried out in
presence of OPD, different OPD-dependent peaks were
observed upon GC-MS analysis of ethyl acetate extracts.
GC-MS analysis of ethyl acetate extracts of reaction
mixtures revealed the presence of 2-mercaptoquinoxaline
(retention time 8.25 min; mass spectrum, Figure 9A) as
well as 2-(dichloromethyl)benzimidazole (retention time
9.09 min; mass spectrum, Figure 7C). Treatment of the
ethyl acetate extracts by ethereal diazomethane con-
verted these products to 2-(methylthio)quinoxaline (re-
tention time 8.48 min; mass spectrum, Figure 9B) and
N-methyl-2-(dichloromethyl)benzimidazole (retention time
9.15 min; mass spectrum, Figure 7F). The presence of
m/ z 160 in the mass spectrum at 8.48 min also points to
the formation of hydroxyquinoxaline in these incubations.
In cytosolic incubations of DCDFE-Cys, two peaks
attributed to cis- and trans-isomers of dimethylated
S-(2,2-dichloro-1,1-difluoroethyl)-3-mercapto-2-oxopropi-
onic acid (DCDFE-MOP) were also observed by GC-MS-
analysis,2 at 7.89 and 8.15 min, again pointing to a
transamination or oxidative deamination reaction in
cytosol.
F igu r e 8. Electron-impact mass spectra of dimethylated
pseudomercapturic acids (PMS) formed in incubations of S-(2-
chloro-1,1,2-trifluoroethyl)-L-cysteine (CTFE-Cys) (A) and S-(2,2-
dichloro-1,1-difluoroethyl)-L-cysteine (DCDFE-Cys) (B) in
a
â-lyase-mimetic model consisting of pyridoxal (0.25 mM) and
Cu(II) (100 µM) in 50 mM potassium borate buffer, pH 8.6.
Fragmentations, as indicated by the dotted lines, are consistent
with those obtained from the dimethylated pseudomercapturic
acid formed from S-(tetrafluoroethyl)-L-cysteine (11).
Com p a r ison of P r od u cts F or m ed fr om S-(2,2-
Dih a lo-1,1-d iflu or oeth yl)-L-cystein e Con ju ga tes a n d
S-(Tr ih a lovin yl)-L-cystein e Con ju ga tes. It has been
reported that adducts of chlorofluorothionoacyl fluoride
are labile at basic pH (16), so that part of the products
after overnight incubations with a â-lyase model already
may have been degraded; therefore, it was decided to
compare the reaction products of S-(trihalovinyl)cysteine
S-conjugates and S-(2,2-dihalo-1,1-difluoroethyl)-L-cys-
teine conjugates using enzymic incubations of only 30
min. To avoid decomposition of initially formed products
as much as possible, these incubations were extracted
immediately and, after methylation, analyzed by GC-MS.
(1) S-(2-Ch lor o-1,2-diflu or ovin yl)-L-cystein e (CDFV-
Cys). GC-MS analysis of extracts of incubations of
CDFV-Cys, in the presence of rat renal cytosol and OPD,
revealed the formation of similar OPD adducts as ob-
F igu r e 9. Electron-impact mass spectra of 2-mercaptoquinoxa-
line (A) formed in incubations of S-(2-chloro-1,1,2-trifluoroethyl)-
L-cysteine (CTFE-Cys, 20 mM) or S-(2,2-dichloro-1,1-difluoro-
ethyl)-L-cysteine (DCDFE-Cys, 20 mM) in a â-lyase-mimetic
model consisting of pyridoxal (2 mM), Cu(II) (100 µM), and OPD
(4 mM) in 50 mM potassium borate buffer, pH 8.6. Treatment
of 2-mercaptoquinoxaline with ethereal diazomethane converted
this compound into 2-methylthioquinoxaline (B).
tent with those of 2-(difluoromethyl)benzimidazole and
2-(dichloromethyl)benzimidazole (Figure 7A-C). Meth-

Bioactivation of Cysteine Conjugates of Haloalkenes
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1099
Ta ble 1. P ea k Ar ea s of Ion Ch r om a togr a m s for Qu in oxa lin e a n d Ben zim id a zole Com p ou n d s
OPD adduct analyzed tret m/ z
2-methylthioquinoxaline 8.49
substrate
ratioa
56.8
CTFE-Cys
176
N-methyl-2-(chlorofluoromethyl)benzimidazole
7.90
198/200b
1CDFV-Cys
2-methylthioquinoxaline
N-methyl-2-(chlorofluoromethyl)benzimidazole
8.50
7.90
176
4.6
198/200b
DCDFE-Cys
TCV-Cys
2-methylthioquinoxaline
N-methyl-2-(dichloromethyl)benzimidazole
8.49
9.15
176
101.1
4.7
214/216c
2-methylthioquinoxaline
N-methyl-2-(dichloromethyl)benzimidazole
8.49
9.15
176
214/216c
a
Ratio of peak area of 2-methylthioquinoxaline relative to that of benzimidazole-products; peak areas were obtained by integration of
b
the ion chromatograms acquired by selected ion monitoring at the indicated m/ z values. The ratio of the integrals of m/ z 198 to 200
was 3 to 1 due to the chloro isotope pattern. c The ratio of the integrals of m/ z 198 to 200 was 1.5 to 1 due to the dichloro isotope pattern.
served in incubations of CTFE-Cys. In methylated
extracts a peak was observed at 7.90 min with a retention
time and mass spectrum identical to the N-methyl-2-
(chlorofluoromethyl)benzimidazole formed in the incuba-
tion of CTFE-Cys. This indeed indicates the formation
of the anticipated chlorofluorothionoacylating species as
shown in Figure 1B (X1, X2, X3 ) Cl, F, F). At 8.50 min,
2-methylthioquinoxaline and minor amounts of 2-meth-
oxyquinoxaline could also be identified. The relative
amounts of 2-methylthioquinoxaline and N-methyl-2-
(chlorofluoromethyl)benzimidazole were determined by
integrating the corresponding peaks in ion chromato-
grams at m/ z 176 and 198, respectively. From Table 1
it is clear that the ratio of 2-methylthioquinoxaline to
N-methyl-2-(chlorofluoromethyl)benzimidazole was 12.3-
fold lower when compared to incubations with CTFE-Cys
as substrate.
In cytosolic incubations of CDFV-Cys, a cluster of
peaks with mass spectra consistent with dimethylated
S-(2,2-dichloro-1,1-difluoroethyl)-3-mercapto-2-oxopropi-
onic acid (DCDFE-MOP) were also observed by GC-MS
analysis, between 6.68 and 6.80 min, pointing to tran-
samination or oxidative deamination of CDFV-Cys in
cytosol. The cluster can be explained by the fact that
both CDFV-Cys isomers form cis- and trans-isomers of
dimethylated MOP S-conjugates.
(2) S-(Tr ich lor ovin yl)-L-cystein e (TCV-Cys). GC-
MS analysis of extracts of incubations of TCV-Cys, in the
presence of rat renal cytosol and OPD, revealed the
formation of similar OPD adducts as observed in incuba-
tions of DCDFE-Cys. In methylated extracts a peak was
observed at 9.15 min with retention time and mass
spectrum identical to that of synthetical N-methyl-2-
(dichloromethyl)benzimidazole. This product is indica-
tive for the formation of the anticipated dichlorothiono-
acylating species (Figure 1B: X1, X2, X3 ) Cl, Cl, Cl).
Formation of dichlorothionoacylating species from TCV-
Cys has also been demonstrated previously (17). As in
the case of CDFV-Cys, also in these incubations 2-meth-
ylthioquinoxaline and minor amounts of 2-methoxyqui-
noxaline could be identified (Table 1). The relative
amounts of 2-methylthioquinoxaline and N-methyl-2-
(dichloromethyl)benzimidazole were determined by in-
tegrating the corresponding peaks in ion chromatograms
at m/ z 176 and 214, respectively. In incubations with
TCV-Cys as substrate, the ratio of 2-methylthioquinoxa-
line to N-methyl-2-(dichloromethyl)benzimidazole was
21.5-fold lower when compared to that in incubations
with DCDFE-Cys as substrate (Table 1).
2-oxopropionic acid (DCDFE-MOP) were observed by GC-
MS analysis, at 9.15 and 9.25 min,2 pointing to transam-
ination or oxidative deamination of TCV-Cys in cytosol.
F r ee En th a lp ies of Rea r r a n gem en t P a th w a ys of
F lu or in a ted Eth a n eth iol Com p ou n d s. Free enthal-
pies of formation (∆fG) of presumed reactive intermedi-
ates derived from cysteine S-conjugates of TFE, CTFE,
and DCDFE were calculated by ab initio calculation. The
differences between the calculated ∆fG values of possible
rearrangement products, thiiranes, thionoacyl fluorides,
and the corresponding halide anions, are summarized in
Table 2. For TFE-thiolate, rearrangement to thionoacyl
fluoride is energetically favored by 14 kcal/mol over
formation of the corresponding thiirane. In contrast, for
both CTFE-thiolate and DCDFE-thiolate, formation of
thiirane is energetically favored by 66 and 70 kcal/mol,
respectively, over formation of the thionoacyl fluorides.
Because of the relatively high free enthalpies of hydra-
tion (∆Ghydr) of fluoride and chloride anions released,
-104 and -76 kcal/mol, respectively (30), hydration may
play an important role in directing the rearrangement
reaction in aqueous solution, depending upon which type
of halide anion is released by the different rearrangement
reactions. In the case of TFE-thiolate, both rearrange-
ments yield a fluoride anion, but formation of thionoacyl
fluoride is still energetically favored over formation of
thiirane by 14 kcal/mol. However, with both CTFE- and
DCDFE-thiolate anions, the nature of released halide
anion differs between the thiirane pathway, releasing a
chloride anion, and the thionoacyl fluoride pathway,
releasing a fluoride anion. Because the free enthalpy of
hydration of the fluoride anion is more exothermic than
that of the chloride anion, the difference between the
thionoacyl fluoride pathway and the thiirane pathway
is reduced by 28 kcal/mol (Table 2). However, for both
CTFE- and DCDFE-thiolate anions, the thiirane pathway
remains energetically favored by 38 and 42 kcal/mol,
respectively, over the thionoacyl fluoride pathway. How-
ever, the contribution of differences in hydration enthal-
pies of the thiirane and thionoacyl fluoride products
cannot be included because these are not known.
Discu ssion
In the present study, the nature of the reactive thiol
compounds resulting from â-elimination reactions of
various cysteine S-conjugates was studied. OPD was
used as a trapping agent in order to discriminate between
thionoacyl fluorides and thiiranes. The o-amino groups
of OPD were anticipated to favor ring closure to stable
low-molecular-weight products which can easily be de-
tected by GC-MS. Thiiranes were anticipated to form
In cytosolic incubations of TCV-Cys, two peaks with
mass spectra consistent with cis- and trans-isomers of
dimethylated S-(2,2-dichloro-1,1-difluoroethyl)-3-mercapto-

1100 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Commandeur et al.
Ta ble 2. Ca lcu la ted F r ee En th a lp ies of F or m a tion of th e Decom p osition P r od u cts of F lu or in a ted Eth a n eth iola te
Com p ou n d s
free enthalpies of formation
decomposition products (reactive intermediate + halide ion)
Σ(∆fG)a
Σ[(∆fG) + (∆GhydrX-)]b
TFE-thiolate:
difluorothionoacyl fluoride + F-
2,2,3-trifluorothiirane + F-
-871.327 a.u.
-871.305 a.u.
-871.493 a.u.
-871.471 a.u.
∆ -0.022 a.u.
(-13.8 kcal/mol)
∆ -0.022 a.u.
(-13.8 kcal/mol)
CTFE-thiolate:
chlorofluorothionoacyl fluoride + F-
2,2,3-trifluorothiirane + Cl-
-1231.372 a.u
-1231.477 a.u.
-1231.538 a.u.
-1231.598 a.u.
∆ -0.105 a.u.
(+65.9 kcal/mol)
∆ +0.060 a.u.
(+37.6 kcal/mol)
DCDFE-thiolate:
difluorothionoacyl fluoride + F-
3-chloro-2,2-difluorothiirane + Cl-
-1591.420 a.u.
-1591.532 a.u.
-1591.586 a.u.
-1591.653 a.u.
∆ +0.112 a.u.
(+70.3 kcal/mol)
∆ +0.067 a.u.
(+42.0 kcal/mol)
a
Calculated free enthalpies of formation (∆fG): 2,2,3-trifluorothiirane, -771.955 a.u.; 3-chloro-2,2-difluorothiirane, -1132.010 a.u.;
difluorothionoacyl fluoride, -771.977 a.u.; chlorofluorothionoacyl fluoride, -1132.022; dichlorothionoacyl fluoride, -1492.070 a.u.; F-,
-99.35 a.u.; Cl-, -459.522 a.u. Free enthalpies of hydration of halide anions (∆GhydrX-): F-, -104 kcal/mol (-0.166 a.u.); Cl-, -76
b
kcal/mol (0.121 a.u.) (28).
quinoxaline compounds exclusively, whereas thionoacyl
fluorides were anticipated to condense to benzimidazole
compounds (Figure 2).
and products. Our ab initio calculations revealed that
the TFE-thiolate anion rearrangement via the thionoacyl
fluoride pathway would be energetically favored over the
thiirane pathway by 14 kcal/mol (Table 2). Interestingly,
in the case of CTFE- and DCDFE-thiolate anions, it was
found that formation of the corresponding thiiranes
would be thermodynamically favored over that of thion-
oacyl fluoride even when differences in the hydration
energies of the halide anions released were considered
(Table 2). When known free energies of hydration of
halide anions were included, the thiirane route was still
thermodynamically favored over the thionoacyl fluoride
route for CTFE- and DCDFE-thiolate anions by 38 and
42 kcal/mol, respectively (Table 2). Because hydration
enthalpies of the fluorinated thiiranes and thionoacyl
fluorides are not known, their contribution to the ther-
modynamic control cannot be calculated. However, as
these compounds are uncharged and lipophilic, these
contributions are expected to be low.
Finkelstein et al. detected glyoxylic acid in incubations
of bromine-containing cysteine conjugates, which was
attributed to hydrolysis of R-thionolactones as reactive
intermediates (18). In incubations of TFE-Cys, CTFE-
Cys, and DCDFE-Cys, they were not able to detect
glyoxylate, indicating that R-thionolactones may not be
formed from non-bromine-containing cysteine conjugates.
The results in the present study are supportive for their
observations. Because R-thionolactones are formed via
the thionoacyl fluoride pathway (Figure 1), this route is
not energetically favored for CTFE-Cys and DCDFE-Cys.
For TFE-Cys, the thio(no)acetic acid formed by hydrolysis
of difluorothionoacyl fluoride appears to be relatively
stable, because it is detected by 19F-NMR (Figure 3).
In order to find experimental proof for the formation
of thiirane products, incubations were performed in the
presence of o-phenylenediamine (OPD). In incubations
with CTFE-Cys the formation of 2-(chlorofluoromethyl)-
benzimidazole was indicative for formation of chlorofluo-
rothionoacetyl fluoride. However, in addition, the major
OPD adduct in CTFE-Cys incubations was identified as
2-mercaptoquinoxaline. 2-Mercaptoquinoxaline can be
explained by a cyclization reaction of the initial chloro-
fluorothionoacyl adduct of OPD, followed by dehydroha-
The results of the present investigation demonstrate
that with all S-(2,2-dihalo-1,1-difluoroethyl)cysteine S-
conjugates studied dihalothionoacyl fluorides are formed.
This was demonstrated by the formation of self-adducts
to the parent cysteine S-conjugates (Figure 6). In the
case of TFE-Cys, the so-called pseudomercapturic acid
(11) resulting from covalent binding of difluorothiono-
acetyl fluoride to TFE-Cys itself was the major product.
Similar results have been obtained by Hayden et al. (13)
using a nonenzymic model consisting of pyridoxal 5′-
phosphate and copper(II). In incubations in the presence
of OPD, a high concentration of 2-(difluoromethyl)-
benzimidazole was observed, pointing to a intramolecular
condensation reaction of the initially difluorothiono-
acylated OPD (Figure 7).
In incubations with CTFE-Cys and DCDFE-Cys, the
corresponding pseudomercapturic acids were also ob-
served, but only in relatively low amounts according to
19F NMR analysis (Figures 4B and 5B). In CTFE-Cys
incubations, the ratio of the integrals of the fluoride anion
and the products derived from chlorofluorothionoacetyl
fluoride was approximately 10 to 1. It was shown by
Fisher et al. (16) that a lysine adduct of chlorofluo-
rothionoacyl fluoride slowly degrades to non-fluorine-
containing products and fluoride anion. At pH 7.4, 50%
of the adduct was degraded after 4 days of incubation.
When our model system was incubated at pH 7.4, the
fluoride anion was still the major product after 16 h of
incubation (data not shown). Because instability of
chlorofluorothionoacyl amides apparently is not ex-
tremely high, we investigated whether alternative reac-
tive intermediates might be formed from CTFE-Cys next
to chlorofluorothionoacetyl fluoride.
As proposed previously by Dohn et al. (9), a possible
alternative reactive intermediate derived from CTFE-Cys
is 2,2,3-trifluorothiirane (Figure 1). In the present study,
the preference for rearrangement of thiolate anions to
either thiiranes or thionoacyl fluorides was investigated
theoretically as well by quantum chemical ab initio
calculations of free enthalpies of formation of reactants

Bioactivation of Cysteine Conjugates of Haloalkenes
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1101
logenation reaction (Figure 2) or by a reaction of OPD
with 2,2,3-trifluorothiirane (Figure 2, route a). In order
to distinguish between these two possibilities, additional
incubations were performed with S-(2-chloro-1,2-difluo-
rovinyl)-L-cysteine (CDFV-Cys). It was anticipated that
CDFV-Cys can form a chlorofluorothionoacylating species
(chlorofluorothioketene and/or chlorofluorothionoacyl fluo-
ride, Figure 1B) but that it cannot form a thiirane
compound. As expected, 2-(chlorofluoromethyl)benzimi-
dazole was indeed identified in incubations with CDFV-
Cys. The presence of 2-mercaptoquinoxaline in incuba-
tions of CDFV-Cys demonstrates that the initial chloro-
fluorothionoacyl adduct of OPD indeed reacts intramo-
lecularly, forming a 6-membered ring. This indeed
supports the observation of Fisher et al. (16) that
chlorofluorothionoacetyl moieties are reactive centers
prone to hydrolysis or alkylation. Next to 2-mercapto-
quinoxaline also small amounts of 2-hydroxyquinoxaline
could be identified. This product may result from hy-
drolysis of 2-mercaptoquinoxaline or 2-fluoroquinoxaline
(Figure 2) or can be formed by reacting to glyoxylic acid
which can be anticipated as a hydrolysis product formed
from both chlorofluoroacetic acid and 2,2,3-trifluorothi-
irane. Dichloroacetic acid and difluoroacetic acid did not
react with OPD under the incubation conditions used,
according to GC-MS and 19F-NMR analysis (data not
shown), ruling out the contribution of dihaloacetic acids
to product formation. Dihalothioacetic acids are also not
likely to contribute to formation of 2-mercaptoquinoxaline
because reaction of thioacetic acids with amino groups
is known to result in release of the sulfur atom as
hydrogen sulfide (31).
As mentioned above for both CTFE-Cys and CDFV-
Cys, formation of 2-(chlorofluoromethyl)benzimidazole
and 2-mercaptoquinoxaline was observed. If a chloro-
fluorothionoacyl adduct of OPD would be the only inter-
mediate leading to these products, a fixed ratio of
2-(chlorofluoromethyl)benzimidazole and 2-mercaptoqui-
noxaline would be expected in incubations with CTFE-
Cys and CDFV-Cys. The fact that for CTFE-Cys the
relative amount of 2-mercaptoquinoxaline was 12-fold
higher than in the case of CDFV-Cys strongly suggests
a second route leading to this product. Based on our
quantum chemical calculations, and the fact that forma-
tion of 2-mercaptoquinoxaline via thiirane formation can
be rationalized (Figure 2), we suggest formation of 2,3,3-
trifluorothiirane as a second reactive intermediate for
CTFE-Cys.
ultimately resulting in a diconjugated product which
rapidly eliminates all its fluorine atoms on exposure to
water (32). Thus all the fluorine atoms are lost as
fluoride anion, but the sulfur atom is retained. The
formation of 2-mercaptoquinoxaline in the reaction of
OPD with the presumed 2,2,3-trifluorothiirane can be
regarded as an analogous reaction. 2,2-Difluorothiirane
is unstable and rapidly extrudes its sulfur atom at 0 °C
(33). Extrusion of sulfur from several thiiranes is
catalyzed effectively by thiolate anions attacking the
thiirane sulfur atom. In dimethyl sulfoxide, this process
resulted in formation of polysulfides (n ) 2-8) and 1,1-
difluoroethylene; electron-withdrawing substituents on
the thiirane ring promoted this reaction strongly (34).
However, the fact that in our experiments almost all
fluorine atoms of CTFE-Cys and DCDFE-Cys were
recovered as the fluoride anion indicates that formation
of fluorinated ethylenes can only be a minor route. Very
recently, glyoxylate and small amounts of trihaloalkenes
have been identified as products of bromine-containing
cysteine S-conjugates, indicative for thiirane formation
(19). However, these products could not be identified in
incubations with DCDFE-Cys, CTFE-Cys, and TFE-Cys.
In conclusion, the present study demonstrated that
dihalothionoacylating species are formed in the case of
S-(2,2-dihalo-1,1-difluoroethyl)-L-cysteine compounds (TFE-
Cys, CTFE-Cys, DCDFE-Cys) and S-(trihalovinyl)-L-
cysteine compounds (TCV-Cys and CDFV-Cys). For
CTFE-Cys and DCDFE-Cys, thiiranes are proposed to be
important reactive intermediates next to the dihalothi-
onoacyl fluorides. Ab initio calculations showed that
rearrangement to thiiranes is thermodynamically favored
over formation of thionoacyl fluorides for the thiolate
anions formed from CTFE-Cys and DCDFE-Cys. Our
results show that, after an initial alkylation reaction step,
the products formed from both dihalothionoacylating
species and thiiranes still contain an electrophilic center
capable of a second alkylation reaction. These electro-
philes therefore might lead to intra- and intermolecular
cross-links of biomacromolecules.
Refer en ces
(1) Commandeur, J . N. M., Stijntjes, G. J ., and Vermeulen, N. P. E.
(1995) Enzymes and transport systems involved in the formation
and disposition of glutathione S-conjugates. Pharmacol. Rev. 47,
271-330.
(2) Anders, M. W., Dekant, W. And Vamvakas, S. (1992) Glutathione-
dependent toxicity. Xenobiotica 22, 1135-1145.
(3) Commandeur, J . N. M., and Vermeulen, N. P. E. (1990) Molecular
and biochemical mechanisms of chemically-induced nephrotox-
icity: A review. Chem. Res. Toxicol. 3, 171-194.
(4) Anderson, P. M., and Schultze, M. O. (1965) Interactions of S-(1,2-
dichlorovinyl)-L-cysteine with proteins. Arch. Biochem. Biophys.
109, 615-621.
(5) Bhattacharya, R. K., and Schultze, M. O. (1973) Modification of
polynucleotides by a fragment produced by enzymic cleavage of
S-(1,2-dichlorovinyl)-L-cysteine. Biochem. Biophys. Res. Commun.
53, 172-181.
(6) Schrenk, D., and Dekant, W. (1989) Covalent binding of hexachlo-
robutadiene metabolites to renal and hepatic mitochondrial DNA.
Carcinogenesis 10, 1139-1141.
Comparable results were obtained when comparing
products derived from DCDFE-Cys and TCV-Cys. Both
compounds are anticipated to form dichlorothionoacylat-
ing reactive species (Figure 2: X1, X2 ) Cl, Cl). Indeed,
both 2-mercaptoquinoxaline and 2-(dichloromethyl)ben-
zimidazole were identified as adducts to OPD. In the
case of DCDFE-Cys, however, the relative amount of
2-mercaptoquinoxaline was 21-fold higher than in the
case of TCV-Cys, which only can form dichlorothionoa-
cylating agents. For the same reasons as discussed for
CTFE-Cys, this is indicative for formation of 3-chloro-
2,2-difluorothiirane as a second reactive intermediate
derived from DCDFE-Cys.
(7) Hayden, P. J ., and Stevens, J . L. (1990) Cysteine conjugate
toxicity, metabolism and binding to macromolecules in isolated
rat kidney mitochondria. Mol. Pharmacol. 37, 468-476.
As yet, only limited data are available about the
chemical properties of fluorine-containing thiiranes, and
their reactivity has only been studied in organic solvents.
Tetrafluorothiirane, a relatively stable compound, reacts
rapidly with nitrogen compounds and is attacked at its
carbon atoms by secondary amines, such as morpholine,
(8) Hayden, P. J ., Welsh, C. J ., Yang, Y., Schaeffer, W. H., Ward, A.
J . I., and Stevens, J . L. (1992) Formation of mitochondrial
phospholipid adducts by nephrotoxic cysteine conjugate metabo-
lites. Chem. Res. Toxicol. 5, 231-237.
(9) Dohn, D. R., Leininger, J . R., Lash, L. H., Quebbeman, A. J ., and
Anders, M. W. (1985) Nephrotoxicity of S-(2-chloro-1,1,2-trifluo-
roethyl)glutathione and S-(2-chloro-1,1,2-trifluoroethyl)-L-cys-

1102 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Commandeur et al.
teine, the glutathione and cysteine conjugates of chlorotrifluoro-
ethylene. J . Pharmacol. Exp. Ther. 235, 851-857.
(10) Dekant, W., Lash, L. H., and Anders, M. W. (1987) Bioactivation
mechanism of the cytoxic and nephrotoxic S-conjugate S-(2-chloro-
1,1,2-trifluoroethyl)-L-cysteine.Proc. Natl. Acad. Sci. U.S.A. 84,
7443-7447.
(11) Commandeur, J . N. M., de Kanter, F. J . J ., and Vermeulen, N. P.
E. (1989) Bioactivation of the cysteine-S-conjugate and mercap-
turic acid of tetrafluoroethylene to acylating reactive intermedi-
ates in the rat. Mol. Pharmacol. 36, 654-663.
(12) Commandeur, J . N. M., Brakenhoff, J . P. G., de Kanter, F. J . J .,
and Vermeulen, N.P. E. (1988) Nephrotoxicity of mercapturic
acids of three structurally related 2,2-difluoroethylenes in the rat.
Biochem. Pharmacol. 37, 4495-4505.
(13) Hayden, P. J ., Yang, Y., Ward, J . I. A., Dulik, D. M., McCann, D.
J ., and Stevens, J . L. (1991) Formation of difluorothionoacetyl-
protein adducts by S-(1,1,2,2,-tetrafluoroethyl)-L-cysteine metabo-
lites: Nucleophilic catalysis of stable lysyl adduct formation by
histidine and tyrosine. Biochemistry 30, 5935-5943.
(14) Harris, J . W., Dekant, W., and Anders, M. W. (1992) In vivo
detection and characterization of protein adducts resulting from
bioactivation of haloethene cysteine S-conjugates by 19F NMR:
Chlorotrifluoroethene and tetrafluoroethene. Chem. Res. Toxicol.
5, 34-41.
(15) Hayden, P. J ., Ichimura, T., McCann, D. J ., Pohl, L. R., and
Stevens, J . L. (1991) Detection of cysteine conjugate metabolite
adducts with specific mitochondrial proteins using antibodies
raised against halothane metabolite adducts. J . Biol. Chem. 266,
18415-18418.
(16) Fisher, M. B., Hayden, P. J ., Bruschi, S. A., Dulik, D. M., Yang,
Y., Ward, J . I. A., and Stevens, J . L. (1993) Formation, charac-
terization, and immunoreactivity of lysine thioamide adducts from
fluorinated nephrotoxic cysteine S-conjugates in vitro and in vivo.
Chem. Res. Toxicol. 6, 223-230.
(17) Dekant, W., Berthold, K., Vamvakas, S., Henschler, D., and
Anders, M. W. (1988) Thioacylating intermediates as metabolites
of S-(1,2-dichlorovinyl)-L-cysteine and S-(1,2,2-trichlorovinyl)-L-
cysteine formed by cysteine conjugate â-lyase. Chem. Res. Toxicol.
1, 175-178.
(21) Commandeur, J . N. M., Oostendorp, R. A. J ., Schoofs, P. R., Xu,
B., and Vermeulen, N. P. E. (1987) Nephrotoxicity and hepato-
toxicity of 1,1-dichloro-2,2-difluoroethylene in the rat. Indications
for differential mechanisms of bioactivation. Biochem. Pharmacol.
36, 4229-4237.
(22) Young, J . A., and Tarrant, P. (1949) The preparation of some
derivatives of chlorofluoroacetic acid. J . Am. Chem. Soc. 71, 2432-
2438.
(23) Sawicki, E. (1956) Polyfluoroacyl derivatives of carcinogenic and
allied amines. J . Org. Chem. 21, 376.
(24) Commandeur, J . N. M., and Vermeulen, N. P. E. (1990) Identi-
fication of N-acetyl-S-(2,2-dichlorovinyl)-L-cysteine and N-acetyl-
S-(1,2-dichlorovinyl)-L-cysteine as regioisometic mercapturic acids
of trichloroethylene in the rat. Chem. Res. Toxicol. 3, 212-218.
(25) Gandolfi, A. J ., Nagle, R. B., Soltis, J . J ., and Plescia, F. H. (1981)
Nephrotoxicity of halogenated vinyl cysteine compounds. Res.
Commun. Chem. Pathol. Pharmacol. 33, 249-261.
(26) Scheeren, J . W., Ooms, P. H. J ., and Nivard, R. J . F. (1973) A
general procedure for the conversion of a carbonyl group into a
thione group with tetraphosphorus decasulphide. Synthesis, 149-
151.
(27) Guest, M. R., and Kendrick, J . (1986) GAMESS User Manual,
An Introductory Guide, CCP/86/1, Daresbury Laboratory, imple-
mented on the Cyber 205 by M. F. Guest and J . H. van Lenthe.
(28) Collins, J . B., Van Schleyer, P. R., Binkley, J . S., and Pople, J . A.
(1976) Self-consistent molecular orbital methods. XVII. Geom-
etries and binding energies of second molecules. A comparison of
three basis sets. J . Chem. Phys. 64, 5142-5151.
(29) Frisch, M. J ., Pople, J . A., and Binckley, J . S. (1984) Self-consistent
molecular orbital methods. 25. Supplementary functions for
Gaussian basis sets. J . Chem. Phys. 80, 3265-3269.
(30) Friedman, H. L., and Krishnan, C. V. (1973) Thermodynamics of
ion hydration. In Water: A Comprehensive Treatise (Frank, F.,
Ed.) Vol. 3, pp 1-118, Plenum Press, New York.
(31) Duus, F. (1979) Thiocarbonyl compounds. In Comprehensive
Organic Chemistry (J ones, D. N., Ed.) Vol. 3, pp. 373-487,
Pergamon Press, Oxford.
(18) Finkelstein, M. B., Dekant, W., Kende, A. S., and Anders, M. W.
(1995) R-Thiolactones as novel intermediates in the cysteine
conjugate â-lyase-catalyzed bioactivation of bromine-containing
cysteine S-conjugates. J . Am. Chem. Soc. 117, 9590-9591.
(19) Finkelstein, M. B., Dekant, W., and Anders, M. W. (1996) Cysteine
conjugate â-lyase-catalyzed bioactivation of bromine-containing
cysteine S-conjugates: stoichiometry and formation of 2,2-dif-
luoro-3-halothiiranes. Chem. Res. Toxicol. 9, 227-231.
(20) Guengerich, F. P. (1994) Mechanisms of formation of DNA adducts
from ethylene dihalides, vinyl halides, and arylamines. Drug
Metab. Rev. 26, 47-66.
(32) Brasen, W. R., Cripps, H. N., Bottomley, C. G., Farlow, M. W.,
and Krespan, C. G. (1965) Tetrafluorothiirane. J . Org. Chem. 30,
4188-4193.
(33) Middleton, W. J ., Howard, E. G., and Sharkey, W. H. (1965)
Fluorothiocarbonyl compounds. I. Preparation of thio ketones,
thioacyl halides, and thio esters. J . Org. Chem. 30, 1375-1384.
(34) Huisgen, R. (1989) From thiiranes to thiocarbonyl S-sulfides;
recent results. Phosphorus, Sulfur Silicon Relat. Elem. 43, 63-94.
TX960049B