Tetrachloroethylene oxide: hydrolytic products and reactions with phosphate and lysine.

Article abstract of DOI:10.1021/tx020028j

Tetrachloroethylene, or perchloroethylene (PCE), has considerable industrial use and is of toxicological interest because of a variety of effects. Most of the existing literature presents PCE oxide as a critical intermediate in the oxidative metabolism of PCE to Cl(3)CCO(2)H, oxalic acid, and products covalently bound to proteins, including trichloroacetyl derivatives of lysine. PCE oxide was synthesized by photochemical oxidation of PCE and characterized. Decomposition at neutral pH (t(1/2) = 7.9 min at 0 degrees C, 5.8 min at 23 degrees C, 2.6 min at 37 degrees C) yielded only trace ( approximately 1%) Cl(3)CCO(2)H; the major products identified were CO (73% yield) and CO(2) (63% yield). In phosphate buffer (0.10 M) a major product was identified as oxalyl phosphate. Oxalyl chloride also reacted to form CO and CO(2) in aqueous solution and to form oxalyl phosphate in neutral phosphate buffer. Oxalyl phosphate decomposed to oxalic acid (t(1/2) = 53 min at 37 degrees C) but did not react with lysine. Reaction of PCE oxide with free lysine yielded the oxalic acid amide derivatives of lysine plus lysine dimers in which cross-linking of the amino groups involved oxalo linkage. The reaction of PCE oxide with albumin yielded mainly N(6)-oxalolysine and some (<5%) N(6)-trichloroacetyllysine. We propose a reaction pathway for PCE oxide based on our previous studies with trichloroethylene oxide, in which C-C bond scission is a major product of reaction in aqueous buffer and yields CO and CO(2). Oxalyl species are proposed as intermediates and prominent acylating species formed in the reactions of the epoxide. The formation of Cl(3)CCO(2)H in cytochrome P450 reactions is postulated to result from intramolecular migration within an enzyme intermediate.

Full text of DOI:10.1021/tx020028j

1096
Chem. Res. Toxicol. 2002, 15, 1096-1105
Tetr a ch lor oeth ylen e Oxid e: Hyd r olytic P r od u cts a n d
Rea ction s w ith P h osp h a te a n d Lysin e
Tadao Yoshioka, J oel A. Krauser, and F. Peter Guengerich*
Department of Biochemistry and Center in Molecular Toxicology, Vanderbilt University
School of Medicine, Nashville, Tennessee 37232-0146
Received April 8, 2002
Tetrachloroethylene, or perchloroethylene (PCE), has considerable industrial use and is of
toxicological interest because of a variety of effects. Most of the existing literature presents
PCE oxide as a critical intermediate in the oxidative metabolism of PCE to Cl3CCO2H, oxalic
acid, and products covalently bound to proteins, including trichloroacetyl derivatives of lysine.
PCE oxide was synthesized by photochemical oxidation of PCE and characterized. Decomposi-
tion at neutral pH (t1/2 ) 7.9 min at 0 °C, 5.8 min at 23 °C, 2.6 min at 37 °C) yielded only trace
(∼1%) Cl3CCO2H; the major products identified were CO (73% yield) and CO2 (63% yield). In
phosphate buffer (0.10 M) a major product was identified as oxalyl phosphate. Oxalyl chloride
also reacted to form CO and CO2 in aqueous solution and to form oxalyl phosphate in neutral
phosphate buffer. Oxalyl phosphate decomposed to oxalic acid (t1/2 ) 53 min at 37 °C) but did
not react with lysine. Reaction of PCE oxide with free lysine yielded the oxalic acid amide
derivatives of lysine plus lysine dimers in which cross-linking of the amino groups involved
oxalo linkage. The reaction of PCE oxide with albumin yielded mainly N6-oxalolysine and some
(<5%) N6-trichloroacetyllysine. We propose a reaction pathway for PCE oxide based on our
previous studies with trichloroethylene oxide, in which C-C bond scission is a major product
of reaction in aqueous buffer and yields CO and CO2. Oxalyl species are proposed as
intermediates and prominent acylating species formed in the reactions of the epoxide. The
formation of Cl3CCO2H in cytochrome P450 reactions is postulated to result from intramolecular
migration within an enzyme intermediate.
products by the action of cysteine conjugate â-lyase (6,
7). The reaction product is mutagenic (7). The other
pathway involves oxidation by P450 (8). Human P450s
1A2, 2B6, and 2C8 have been reported to activate PCE
to a product causing micronucleus formation in lympho-
blastoma cells (9). Evidence for both the GSH conjugation
and oxidative pathways has been found in rats (6).
Cl3CCO2H and oxalic acid are the two major oxidation
products identified in vivo or in liver microsomal incuba-
tions (6, 8, 10, 11). The pathway leading to these products
is usually presented with PCE oxide as an obligate
intermediate (Scheme 1) (1, 5, 6, 8, 10, 12-14). Evidence
In tr od u ction
Tetrachloroethylene, or perchloroethylene (PCE),1 is
produced to the extent of about 108 kg annually in the
United States alone (1, 2). This compound is used
extensively in the dry cleaning industry, is employed as
a solvent in degreasing applications, and is a common
contaminant in waste dump sites (1, 2).
A number of toxicities have been associated with PCE
administration in experimental animals, primarily in-
volving the liver as the target organ (2). Effects include
fatty liver degeneration, necrosis, liver enlargement,
abnormal liver function, decreased ATP levels, and
increased serum transaminases (2). The effects in hu-
mans are less well-defined beyond general narcosis, but
some of the same liver effects seen in animal models have
been documented following high exposure (1). PCE has
been shown to cause hepatocellular carcinomas in mice
but not rats (3, 4). The mechanism of tumorigenesis is
unclear; PCE is not very mutagenic in standard test
systems, even in the presence of activating enzymes (5).
Metabolic activation is generally considered to be
involved in all PCE toxicity, except general central
nervous system depression (2). Two pathways are opera-
tive. One involves enzymatic conjugation with GSH to
form a conjugate that can be processed to reactive
Sch em e 1. Liter a tu r e Sch em e for F or m a tion a n d
Rea ction s of P CE Oxid e (2)
has been presented for the formation of N6-trichloro-
acetylLys in proteins following microsomal oxidation of
PCE, and the pathway involving PCE oxide (and its
hydrolytic opening to Cl3CCO2H) is the basis of assess-
ments of risk estimated from physiologically based phar-
macokinetics and other modeling (15, 16). However, in
the literature (a meeting proceedings account) PCE oxide
was reported to be hydrolyzed to Cl3CCO2H rather
* To whom correspondence should be addressed. Telephone: (615)
322-2261. Fax: (615) 322-3141. E-mail: guengerich@toxicology.mc.
vanderbilt.edu.
1 Abbreviations: PCE, perchloroethylene (tetrachloroethylene); TCE,
trichloroethylene; MOPS, 3-(N-morpholino)propanesulfonate; ESI, elec-
trospray ionization; CBZ, carbobenzyloxy; Orn, ornithine.
10.1021/tx020028j CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/23/2002

Reactions of Tetrachloroethylene Oxide
Chem. Res. Toxicol., Vol. 15, No. 8, 2002 1097
Sch em e 2. Str u ctu r es of Lys Ad d u cts
D2O or CDCl3. 13C shifts are reported relative to CH3OH or 1,4-
dioxane (23), and 31P shifts are reported relative to H3PO4. Mass
spectrometry was carried out with a Finnagan-MAT TSQ7000
series mass spectrometer (Finnigan, Sunnyvale, CA). For LC/
MS analysis, a Waters 2690 separation module was used
(Alliance, Waters, Milford, MA).
inefficiently (3-4% yield) and did not form oxalic acid at
neutral pH (14).
Studies in this laboratory with the related trichloro-
ethylene (TCE) oxide have shown that this epoxide is not
the source of the major enzymatic oxidation product,
chloral hydrate (17). Chloral hydrate results from a
chloride ion migration within a P450 enzyme intermedi-
ate, which yields TCE oxide as a minor product (17). The
hydrolysis of TCE oxide is dominated by C-C bond
scission, yielding CO and HCO2H (17, 18). This scission
appears to occur following opening of the epoxide to an
Syn th esis of P CE Oxid e (24). PCE (180 mL, distilled over
P2O5) was added to a photochemical reaction vessel and warmed
in an 80 °C water bath. O2 was bubbled through the sample for
10 min and the system was illuminated (Hanovia medium-
pressure 450 W mercury arc lamp filtered by a sleeve of Pyrex)
to start the reaction under a constant flow of O2 (∼300 mL
min-1). After 7 h the lamp was turned off, and the reaction
mixture was cooled in an ice bath and then washed with a cold
1 N NaOH solution to remove trichloroacetyl chloride. The
residual PCE mixture was dried over Na2SO4, and PCE oxide
was obtained by distillation (yield 5% based on PCE; bp 35-38
°C at 65 mmHg).
Quantitative analysis was done using proton-decoupled 13C
NMR with an inverse gated decoupling method and delay time
of 20 s between successive transmitter pulses. The spectrum
showed three singlets at 90.6, 105.2, and 120.4 ppm, corre-
sponding to PCE oxide (61%, w/w), hexachloroethane (2%, w/w),
and residual PCE (37%, w/w), respectively (concentrations were
estimated based on the integration of the spectrum measured
in the addition of a known amount of CCl4). The signal at 90.6
ppm is in agreement with the assignment of Campbell and Vogl
(24).
2
acyl halide, as judged by 18O and H labeling studies (18).
Isotopic labeling studies also indicated that Lys does not
react directly with TCE oxide but rather with an acyl
halide product (glyoxal chloride) (18). The hydrolysis of
the very unstable epoxide of vinylidene chloride seems
to follow a course similar to that of TCE oxide, with
chloride and hydride ion migration in an enzyme inter-
mediate (19), although labeling studies have not been
done to characterize the mechanism of epoxide hydrolysis
in detail.
In light of the findings on the mechanisms of hydrolysis
and reaction of TCE oxide (17, 18, 20-22), the reported
low yields of Cl3CCO2H and oxalic acid (14), and the
general industrial and toxicological significance of PCE,
we reinvestigated the mechanism of hydrolysis of PCE
oxide, the role of the epoxide in PCE oxidation, and
reactions of the epoxide with free and protein-bound Lys
(Scheme 2). Our results indicate that, as in the case of
TCE oxide, only very limited Cl- migration occurs and
the hydrolysis is characterized by facile C-C bond
scission to CO and CO2. The enzymatic formation of
trichloroacetyl chloride and oxalic acid is probably ex-
plained by an enzyme intermediate that forms trichlo-
roacetyl chloride and oxalyl chloride.
Syn th esis of Oxa lyl P h osp h a te (25). Oxalyl chloride (1.0
mL, 12 mmol) was added dropwise to a stirred solution of 2.0
M K2HPO4 (10 mL, 20 mmol) in an ice bath. After 20 min, the
reaction mixture was diluted with H2O (400 mL) and applied
to a Dowex 1X8 column (Cl- form, 2.2 × 12 cm) at a flow rate
of 1.2 mL min-1 (4 °C). Elution was performed as described
previously (26): 100 mL of 50 mM MgCl2, followed by 100 mL
of H2O and then a 500 mL linear gradient of 0.1 to 1.0 M KCl.
The flow rate was 1.5 mL min-1, and 9 mL fractions were
collected. Fractions were analyzed for oxalic acid and oxalyl
phosphate using HPLC (vide infra), and the fractions containing
oxalyl phosphate were pooled and stored at -80 °C. To deter-
mine the concentration of oxalyl phosphate, 0.2 mL of 1 N HCl
was added to 2.0 mL of each pooled fraction and the samples
were heated at 37 °C for 30 min. By quantifying oxalic acid by
HPLC (vide infra) and inorganic phosphate (27) formed from
oxalyl phosphate, the concentration of oxalyl phosphate in the
stock solution was determined to be 21 mM. 31P and 13C NMR
spectra of oxalyl phosphate were measured after adjusting the
pH to 8 and adding D2O [final 10% (v/v)], using a Bruker DRX
Exp er im en ta l P r oced u r es
Ca u tion ! PCE oxide and the related acyl chlorides are
reactive acylating agents and should be handled carefully. Gloves
and other protective clothing should be used appropriately.
Sp ectr oscop y. Unless otherwise noted, 1H, 13C, and 31P
NMR spectra were recorded on a Bruker DPX 300 AVANCE
spectrometer operating at 300.1, 75.5, and 121.4 MHz, respec-
tively (Bruker, Billerica, MA). Samples were prepared in either

1098 Chem. Res. Toxicol., Vol. 15, No. 8, 2002
Yoshioka et al.
500 NMR spectrometer at 202.5 and 125.7 MHz, respectively.
methyl ester‚HCl (vide supra) in 6 mL of pyridine, and stirred
at room temperature for 2.5 h. The reaction mixture was
concentrated in vacuo; the residue was dissolved in CH2Cl2 and
washed with 1 N HCl and then H2O and dried over Na2SO4.
Evaporation of the organic solvent gave a solid (91%), which
was purified by recrystallization from a CH3OH/H2O mixture.
A 6 mL solution of HBr in CH3CO2H (30%, w/v) was added to
the compound (700 mg) and the reaction proceeded at room
temperature for 40 min, followed by addition of (C2H5)2O to give
a precipitate (97% yield). The compound (560 mg) was then
hydrolyzed with 1 N NaOH in 5 mL aqueous CH3OH to give
the title compound, which was neutralized by addition of 1 N
HCl, followed by purification by using AG1X8 ion-exchange resin
(acetate form; 1.0 × 15 cm) with elution with H2O. After
lyophilization, the solid (76% yield, overall yield 67%) was
obtained as a fine powder (recrystallized from aqueous C2H5-
OH): mp 124 °C (decomp.); 1H NMR (D2O) δ 1.37 (m, 2H, γ-H),
1.67 (m, 2H, δ-H), 1.81 (m, 1H, â -H1), 1.89 (m, 1H, â-H2), 2.97
(t, 2H, J ) 7.5 Hz, ꢀ-H), 4.18 (dd, 1H, J ) 5.2 and 7.8 Hz, R-H);
13C NMR (D2O) δ 22.7 (γ), 27.0 (δ), 31.5 (â), 39.8 (ꢀ), 55.9 (R),
161.0 (CONH), 178.5 (CO2H); MS, m/z 347 (MH+).
(6) Bis(N6-L-lysyl)oxa lyl Am id e. To 3.0 mmol of L-Lys in
3.0 mL of 1 N NaOH (3.0 mmol), 190 µL (0.5 eq) of diethyl
oxalate was added. The reaction was stirred for 15 min in an
ice-water bath and then at room temperature. After 1.5 h at
room temperature, 2.5 mL of 1 N HCl was added; the solid was
collected and dried to give the product (40% yield): mp 280 °C
(decomp.); 1H NMR (D2O-DCl) δ 1.50 (m, 2H, γ-H), 1.61 (m, 2H,
â-H), 1.99 (m, 2H, δ-H), 3.31 (t, 2H, J ) 6.7 Hz, ꢀ-H), 4.13 (t,
1H, J ) 6.3 Hz, R-H); 13C NMR (D2O-DCl) 22.1 ppm (γ), 28.2
ppm (â), 29.8 ppm (δ), 39.5 ppm (ꢀ), 53.2 ppm (R), 161.4 ppm
(CONH), 172.3 ppm (CO2H); MS, m/z 347 (MH+).
31P NMR showed a peak at -0.50 ppm (using 85% H3PO4 as an
external standard) and 13C NMR showed peaks at 161.0 (J CP
10.3 Hz) and 164.4 ppm (J CP ) 7.3 Hz) (using dioxane as an
external standard).
)
Syn th esis of Lys Der iva tives. (1) N6-Tr ich lor oa cetyl-L-
Lys. To 5.0 mmol of L-Lys in 2.5 mL of 2 N NaOH (5.0 mmol),
1.0 mL (1.5 equiv) of ethyl trichloroacetate was added in one
portion and the reaction was stirred for 2 h at room temperature.
The solid that was formed was collected, washed with CH3CN,
and then dried to give the product (23% yield, fine needles from
aqueous C2H5OH): mp 204 °C (decomp.); 1H NMR (D2O) δ 1.51
(m, 2H, γ-H), 1.72 (m, 2H, â-H), 1.91 (m, 2H, δ-H), 3.41 (t, 2H,
J ) 6.9 Hz, ꢀ-H), 3.76 (t, 1H, J ) 6.3 Hz, R-H); 13C NMR (D2O)
22.4 ppm (γ), 28.2 ppm (δ), 30.8 ppm (â), 41.4 ppm (ꢀ), 55.3 ppm
(R), 92.3 ppm (COCCl3), 165.1 ppm (COCl3), 175.4 ppm (CO2H);
MS, m/z 291/293/295/297 (MH+).
(2) N2-Tr ich lor oa cetyl-L-Lys. N6-Carbobenzyloxy (CBZ)-L-
Lys benzyl ester‚HCl was prepared (28) by heating N6-CBZ-L-
Lys (9.81 g, 35.0 mmol) and p-toluenesulfonic acid (6.7 g, 35
mmol) in a mixture of 100 mL of toluene and 66 mL of benzyl
alcohol for 20 h, using a Dean-Stark trap. The yield was 65%.
To 1.5 mmol of N6-CBZ-L-Lys benzyl ester in 8 mL of CH2Cl2
and 0.8 mL of (C2H5)3N, trichloroacetyl chloride (0.22 mL, 1.3
equiv) was added; the reaction was stirred at room temperature
for 50 min. The reaction mixture was diluted with CH2Cl2 and
washed sequentially with 1 N HCl, 1 N NaOH, and H2O and
dried over Na2SO4. Evaporation of the organic solvent gave a
pale yellow syrup (92%). The compound was then hydrolyzed
with 1 NaOH in aqueous CH3OH (200 mL, room temperature,
1 h) to the corresponding carboxylic acid derivative (80%), which
was treated with 4 mL of HBr in CH3CO2H (30%, w/v) for 1 h,
followed by the addition of (C2H5)2O to give N2-trichloroacetylLys‚
HBr (91%, overall yield 67%). Neutralization with NaHCO3
followed by purification using a Sep-Pak Vac 35 cm3 cartridge
(C18, 10 g, Waters) gave free base (fine powder from aqueous
C2H5OH): mp 191 °C, (decomp); 1H NMR (D2O) δ 1.42 (m, 2H,
γ-H), 1.66 (m, 2H, δ-H), 1.83 (m, 1H, â-H1), 1.96 (m, 1H, â-H2),
2.98 (t, 2H, J ) 7.6 Hz, ꢀ-H), 4.20 (dd, 1H, J ) 5.1 and 8.1 Hz,
R-H); 13C NMR (D2O) δ 22.7 (γ), 27.0 (δ), 31.2 (â), 39.9 (ꢀ), 57.2
(R), 92.2 (CCl3), 164.2 (COCCl3), 178.0 (CO2H); MS m/z 291/293/
295/297 (MH+).
(7) (N2,N6-L-Lysyl)oxa lyl Am id e. N2-Ethoxycarbonylcarbo-
nyl-N6-CBZ-L-Lys methyl ester was prepared in the following
procedure: ethyl chlorooxoacetate (0.27 mL, 1.2 equiv) was
added to N6-CBZ-L-Lys methyl ester‚HCl (2.0 mmol) in
a
mixture of 8 mL of CH2Cl2 and 1.0 mL (C2H5)3N at 0 °C, and
the reaction was stirred for 1.5 h, yielding a pale yellow syrup
after concentration in vacuo (0.86 g, 91% yield). N2-CBZ-L-Lys
methyl ester (2.3 mmol) (28, 30) was added to 1.6 mmol of N2-
ethoxycarbonylcarbonyl-N6-CBZ-L-Lys methyl ester and the
reaction stirred at room temperature for 2 days. The reaction
mixture was evaporated and the residue was dissolved in CHCl3
and washed with 0.5 N HCl and dried over Na2SO4. Concentra-
tion in vacuo and purification using silica gel flash chromatog-
raphy give the corresponding oxalyl amide (57%). To 500 mg of
the compound, 4 mL of HBr in CH3CO2H (30%, w/v) was added
and the reaction kept at room temperature for 1 h followed by
addition of (C2H5)2O. After removal of the (C2H5)2O by decanta-
tion, the precipitate was extracted with H2O and NaOH was
added (to pH ∼10) and the reaction was allowed to stand at
room temperature to give sodium salt of the title compound.
Neutralization followed by purification by using a Sep-Pak Vac
35 cm3 cartridge (C18, 10 g, Waters) gave the product in 66%
yield (overall yield 38%) (fine powder from aqueous C2H5OH):
(3) N6-Oxa lyl-L-Lys. The compound was synthesized accord-
ing to a literature procedure (29) (yield 45%, fine needles from
aqueous 2-propanol): mp 182 °C (decomp.); 1H NMR (D2O) δ
1.42 (m, 2H, γ-H), 1.58 (m, 2H, â-H), 1.90 (m, 2H, δ-H), 3.24 (t,
2H, J ) 6.7 Hz, ꢀ-H), 3.92 (t, 1H, J ) 6.5 Hz, R-H); 13C NMR
(D2O) 22.2 ppm (γ), 28.4 ppm (â), 30.2 ppm (δ), 39.6 ppm (ꢀ),
54.1 ppm (R), 163.8 ppm (COCO2H), 165.5 ppm (COCO2H), 173.8
ppm (CO2H); MS m/z 219 (MH+).
(4) N2-Oxa lyl-L-Lys. Ethyl chlorooxoacetate (0.27 mL, 1.2
equiv) was added to 2.0 mmol of N6-CBZ-L-Lys benzyl ester HCl
(vide supra) dissolved in 8 mL of CH2Cl2 [and 1 mL of (C2H5)3N]
and stirred at 0 °C for 1.5 h. The reaction mixture was diluted
with CH2Cl2 and washed with 1 N HCl, 1 N NaOH, and then
H2O and dried with Na2SO4. Evaporation of the organic solvent
gave a pale yellow syrup (91%). The compound was then
hydrolyzed with 0.7 g of NaOH in a mixture of 5 mL of (CH3)2-
NCHO and 10 mL (1 h, room temperature) to give the corre-
sponding dicarboxylic acid derivative (82%), which was then
treated with 6 mL of HBr in CH3CO2H (30%, w/v) for 1.5 h
followed by addition of (C2H5)2O to give the HBr salt of the title
compound in quantitative yield (overall yield 75%). Neutraliza-
tion with NaHCO3 yielded the free base (prisms from aqueous
1
mp 242 °C (decomp.); H NMR (D2O) δ 1.43 (m, 4H, γ-H), 1.68
(m, 4H, δ-H), 1.89 (m, 4H, â-H), 3.01 (t, 1H, J ) 7.4 Hz, ꢀ-H),
3.33 (dt, 1H, J ) 2.4, 6.8 Hz, ꢀ′-H), 3.74 (t, 1H, J ) 5.7 Hz, R′-
H), 4.22 (dd, 1H, J ) 5.0 and 8.0 Hz, R-H); 13C NMR (D2O):
22.3 and 22.7 (γ), 26.7 and 28.4 (δ), 30.7 and 31.5 (â), 39.8 and
39.9 (ꢀ), 55.3 and 55.8 (R), 161.1 and 161.5 (CONH), 175.4 and
178.6 (CO2H); MS, m/z 347 (MH+).
Syn t h esis of Or n it h in e (Or n ) Der iva t ives. (1) N5-
Tr ich lor oa cetyl-L-Or n . To 5.0 mmol of l-Orn‚HCl in 5.0 mL
of 2 N NaOH (10 mmol), 1.0 mL (1.5 equiv) of ethyl trichloro-
acetate was added in one portion and stirred for 1.5 h at room
temperature. After the reaction mixture was neutralized by
adding 3 N HCl, the reaction mixture was concentrated in vacuo
to give a white solid. After cooling, the solid was collected and
then dried to give the product (36% yield, fine needles from
aqueous C2H5OH): mp 202 °C (decomp.); 1H NMR (D2O) δ 1.67
(m, 2H, γ-H), 1.88 (m, 2H, â-H), 3.37 (t, 2H, J ) 6.8 Hz, δ-H),
1
C2H5OH) mp 225 °C (decomp.); H NMR (D2O) δ 1.43 (m, 2H,
γ-H), 1.67 (m, 2H, δ-H), 1.85 (m, 1H, â -H1), 1.96 (m, 1H, â-H2),
2.97 (t, 2H, J ) 7.4 Hz, ꢀ-H), 4.36 (dd, 1H, J ) 4.9 and 8.8 Hz,
R-H); 13C NMR (D2O) δ 22.7 (γ), 26.8 (δ), 30.8 (â), 39.8 (ꢀ), 53.6
(R), 165.0 (COCO2H), 165.9 (COCO2H), 176.4 (CO2H); MS m/z
219 (MH+).
(5) Bis(N2-L-lysyl)oxa lyl Am id e. Oxalyl chloride (0.13 mL,
0.5 eq) was added in portions to 3.0 mmol of N-CBZ-L-Lys

Reactions of Tetrachloroethylene Oxide
Chem. Res. Toxicol., Vol. 15, No. 8, 2002 1099
3.73 (t, 1H, J ) 5.9 Hz, R-H); 13C NMR (D2O) δ 24.5 (γ), 28.4
(â), 41.2 (δ), 55.0 (R), 92.2 (COCCl3), 165.2 (COCl3), 175.0
(CO2H); MS, m/z 277/279/281/283 (MH+).
residual PCE and PCE oxide. Twenty microliters of concentrated
H2SO4 was added to the aqueous layer, and Cl3CCO2H was
extracted with (C2H5)2O (3 × 1 mL). The combined layers were
evaporated under a stream of N2 to dryness, and the residue
was dissolved in 100 µL of 100 mM potassium MOPS buffer (pH
7.4) and analyzed by HPLC (vide infra). The extraction efficien-
cies of Cl3CCO2H from the MOPS and phosphate buffers were
90 and 84%, respectively.
R ea ct ion s of E lect r op h iles w it h Lys. The indicated
amounts of PCE, PCE oxide, oxalyl chloride, or trichloroacetyl
chloride (solutions in dry CH3CN) were reacted with Lys (final
concentration of 10 mM; unless otherwise indicated) in 100 mM
potassium MOPS or sodium phosphate buffer (pH 7.4) or H2O
in an ice bath for 90 min. In control experiments, PCE oxide
was reacted in H2O for 90 min in an ice bath, followed by the
addition of Lys to the reaction mixture, and allowed to stand
for more 90 min. Adducts with Lys were analyzed using HPLC
and LC/MS (vide infra).
(2) N5-Oxa lyl-L-Or n . The compound was synthesized ac-
cording to a literature procedure (29) [yield 45%, crystals from
1
aqueous (CH3)2CO]: mp 158 °C (decomp.); H NMR (D2O-DCl)
δ 1.74 (m, 2H, γ-H), 1.98 (m, 2H, â-H), 3.36 (t, 2H, J ) 6.8 Hz,
δ-H), 4.16 (t, 1H, J ) 6.2 Hz, R-H); 13C NMR (D2O-DCl) δ 24.3
(γ), 27.6 (â), 39.4 (δ), 53.0 (R), 159.8 (COCO2H), 162.3 (COCO2H),
172.1 (CO2H); MS m/z 205 (MH+).
(3) Qu a n tita tion of P CE Oxid e. PCE oxide solutions were
diluted into dry CH3CN, and the concentration was estimated
(at room temperature) using a colorimetric assay with 4-(4-
nitrobenzyl)pyridine reagent (31)(50 mM in ethylene glycol-
acetone (43:7, v/v)), using ꢀ524 ) 880 M-1 cm-1, which was
calculated with the use of a sample in which the concentration
of PCE oxide was quantified using decoupled 13C NMR (vide
supra).
Reactions with bovine serum albumin were done as described
previously (20) and, following digestion with proteinase K, the
amino acids were derivatized with phenylisothiocyanate. In the
analysis of the albumin Lys adducts, the Orn derivatives (also
phenylthiocarbamates) were utilized to generate internal stan-
dard curves (20).
HP LC a n d MS. Determination of Cl3CCO2H, oxalic acid, and
oxalyl phosphate was performed by using an octadecylsilane
column (Beckman Ultrasphere ODS, 4.6 × 250 mm, Beckman,
San Ramon, CA) with the solvent 10 mM tetra-n-butylammo-
nium hydroxide (adjusted pH to 7.0 with H3PO4): CH3CN (4:1,
v/v) at a flow rate of 1.0 mL min-1 and detection at 210 nm
(18). Determination of adducts with Lys was performed by using
the same column with 10 mM NH4CH3CO2 (pH 6.5) at a flow
rate of 0.7 mL min-1 and detection at 210 nm.
Mea su r em en t of t1/2 of P CE Oxid e. (1) In 3-(N-Mor p h oli-
n o)p r op a n esu lfon a te (MOP S) Bu ffer . To each test tube (on
ice) containing 100 µL of 100 mM potassium MOPS (pH 7.4),
10 µL of 0.1 M PCE oxide in CH3CN was added (final concentra-
tion of the oxide was ∼10 mM). At appropriate times, 500 µL of
the 4-(4-nitrobenzyl)pyridine reagent was added and the samples
were kept at room temperature for 5 min. At that time, 500 µL
of (C2H5)3N-acetone (1:1, v/v) was added and the absorbance was
measured at 535 nm.
(2) In CH3CN. A solution of 100 mM PCE oxide in CH3CN
was kept in a Teflon-sealed vial at room temperature (23 ( 1
°C). At appropriate times, 10 µL of the solution was dissolved
in 1.0 mL of 100 mM potassium MOPS buffer (pH 7.4). The
mixtures were kept in an ice bath for 15 min and then the
amount of Cl3CCO2H was analyzed by HPLC (vide infra).
Mea su r em en t of P oten tia l P r od u cts of P CE Oxid e. (1)
CO. Determination of CO was done with an electrochemical
sensor (Draeger Pac III, Draeger, Pittsburgh, PA) (18). To each
Teflon-sealed vial containing 990 µL of 100 mM sodium phos-
phate or potassium MOPS (pH 7.4) buffer, 10 µL of 80 mM PCE
oxide solution (in CH3CN) was added (final concentration 0.80
mM), and the mixture was stirred for 90 min in an ice bath. At
appropriate times, the vial was placed in a circuit containing
the detector and a peristaltic pump and CO was determined.
(2) CO2. Determination of CO2 was done with an electro-
chemical sensor (pHoenix CO2 electrode attached to an Orion
model 920A meter). A solution of 100 mM sodium phosphate or
potassium MOPS (pH 7.4) buffer (30 mL), containing oxalyl
chloride (2 mM) or PCE oxide (10 mM) was kept in an ice bath
for 10 min (or 90 min in the case of PCE oxide). After the
reaction, a 1 M citrate buffer (pH 4.5) and 1 N HCl were added
to adjust the pH to ∼5.1, just before measurements were made.
Calibration curves were prepared according to the manufactur-
er’s instructions by using NaHCO3 in both MOPS and phosphate
buffers containing 1% (v/v) CH3CN (at 0 °C).
For HPLC/MS (electropray) (ESI) analysis, the HPLC condi-
tions were same as those for the determination of the adducts
(NH4CH3CO2), and 200 µL effluent min-1 was introduced into
the mass spectrometer (18, 20).
Direct inlet ESI MS (positive ion mole) was used to establish
molecular masses of chemicals synthesized in this work.
Micr osom a l In cu ba tion s. Liver microsomes were prepared
from phenobarbital-treated rats (33) and incubated (5 µM P450)
with 1.0 mM PCE and an NADPH-generating system (34) in
100 mM potassium MOPS buffer (pH 7.4) at 37 °C. Reactions
were terminated by the addition of 50 µL of concentrated H2-
SO4. Samples were extracted three times with 1.0 mL of ethyl
acetate in the analysis of oxalic acid or with (C2H5)2O for Cl3-
CCO2H, and the combined organic phase was concentrated to
dryness under an N2 stream. The residue was dissolved in a
solution of 20% CH3CN (v/v) [for oxalic acid, or 25% CH3CN (v/
v) for Cl3CCO2H] in 10 mM tetra-n-butylammonium hydroxide
(buffer to pH 6.5) with H3PO4, and 30 µL aliquots were injected
into a 4.6 × 250 mm Beckman octadecylsilane HPLC column
equilibrated with the injection buffer. A210 was monitored for
the estimation of Cl3CCO2H and oxalic acid, utilizing external
standard curves. Recoveries for Cl3CCO2H and oxalic acid were
81 and 17%, respectively.
(3) HCO2H. HCO2H was analyzed by HPLC as described
previously (18).
(4) HCHO. HCHO was estimated colorimetrically using the
Nash procedure (32).
(5) Oxa lic Acid . To 990 µL of MOPS or phosphate buffer (0
°C), 10 µL of 100 mM PCE oxide solution in CH3CN was added
(final concentration of PCE oxide 1.0 mM), and the mixture was
stirred for 90 min. At appropriate times, 30 µL of the reaction
mixture was directly injected onto HPLC (vide infra) for
measurement of oxalic acid. The determination of oxalic acid
by HPLC was also done with a reaction mixture consisting of
960 µL of the each buffer and 40 µL of 250 mM PCE oxide in
CH3CN (final concentration of oxide 10 mM) in an ice bath for
90 min.
(6) Cl3CCO2H. To 6 mL of MOPS or phosphate buffer kept
in an ice bath, 60 µL of 100 mM PCE oxide in CH3CN (final
concentration PCE oxide 1.0 mM) was added. At appropriate
times, aliquots of the reaction mixture (1.0 mL) were transferred
to test tubes and washed with (C2H5)2O (3 × 1 mL) to remove
Resu lts
Rea ctivity of P CE Oxid e. In CH3CN solution, PCE
oxide rearranged quantitatively to trichloroacetyl chlo-
ride with first-order kinetics. Because trichloroacetyl
chloride was hydrolyzed quantitatively to the correspond-
ing acid when buffer was added, the amount of trichlo-
roacetyl chloride was determined as Cl3CCO2H. The rate
constant was 0.0183 ( 0.0009 h-1 (t1/2 ) 39 h at 23 ( 1
°C). At -20 °C, rearrangement to trichloroacetyl chloride
was not observed over the course of 100 h.
The decomposition of PCE oxide in 100 mM potassium
MOPS (pH 7.4) at 0 °C was analyzed by measuring the
residual amount of the oxide with 4-(4-nitrobenzyl)-

1100 Chem. Res. Toxicol., Vol. 15, No. 8, 2002
Yoshioka et al.
Ta ble 1. Decom p osition P r od u cts of P CE Oxid ea
products formed
(nmol/µmol of PCE oxide)
oxalic
acid
oxalyl
CO CO2 phosphate
buffer
CCl3CO2H
12
potassium MOPS
(0.10 M)
sodium phosphate
(0.10 M)
<3
<3
730 630
60 50
<2
10
180
a
Incubations were at 0 °C for 90 min except for measurement
of oxalyl phosphate (120 min). Initial concentrations of PCE oxide
were 0.80 mM for the measurement of CO and oxalyl phosphate,
10 mM for oxalic acid, and 1.0 mM for the other measurements.
Ta ble 2. Decom p osition P r od u cts of Oxa lyl Ch lor id ea
products formed
(nmol/µmol of oxalyl chloride)
F igu r e 1. Time dependence of formation of oxalyl phosphate
from PCE oxide in 100 mM sodium phosphate buffer (pH 7.4)
at 0 °C. The initial concentration of PCE oxide was 1.0 mM.
The curve is a fit to a single exponential with a rate constant of
0.056 min-1 ((0.004 min-1).
oxalic
acid
oxalyl
phosphate
buffer
CO
CO2
910
potassium MOPS (0.10 M) 970
<3
<2
a
Incubations were for 5 min at 0 °C. Initial concentrations of
PCE oxide were 1.0 mM for measurement of CO, CO2, and oxalyl
phosphate and 10 mM for measurement of oxalic acid.
pyridine reagent. The reagent did not give the charac-
teristic purple pigment (ꢀ535) with trichloroacetyl chloride,
Cl3CCO2H, oxalyl chloride, or oxalic acid, which are
potential products formed from PCE oxide. The kinetics
were pseudo-first-order, and the rate constant of the
decomposition of PCE oxide was 0.088 ( 0.006 min-1 (t1/2
) 7.9 min). At 23 °C, the rate constant was 0.12 ( 0.02
min-1 and at 37 °C the rate constant was 0.26 ( 0.16
min-1. The rate constants were the same in both MOPS
and phosphate buffers (0.10 M each).
Decom p osition P r od u cts of P CE Oxid e. Decompo-
sition products of PCE oxide were measured in both 100
mM potassium MOPS and sodium phosphate buffers (pH
7.4) at 0 °C. PCE itself was stable and yielded none of
the decomposition products under the same conditions
(results not shown). Cl3CCO2H was formed from PCE
oxide in both the buffers, although in very low amounts
(Table 1). Although oxalyl chloride has been considered
to be one of the active intermediates formed from PCE
oxide, oxalic acid was not detected in either of the buffers.
CO and CO2 were the major products (73 and 63 mol %)
in MOPS buffer. However, in the phosphate buffer, the
amounts of CO and CO2 formed were much less. In the
phosphate buffer, formation of oxalyl phosphate (18 mol
%) was observed.
F igu r e 2. Hydrolysis of oxalyl phosphate in 100 mM sodium
phosphate buffer (pH 7.4) at 37 °C. The relative peak area refers
to the integral of the HPLC peak. The curve is a fit to a single
exponential with a rate constant of 0.013 min-1 ((0.004 min-1).
experiments on the formation of and hydrolysis of oxalyl
phosphate were done. The reaction of oxalyl chloride with
sodium phosphate buffer to give oxalyl phosphate at 0
°C was rapid (within 5 min) and the amount of oxalyl
phosphate formed was a linear function of the concentra-
tion of phosphate from 50 to 250 mM at pH 7.4 (results
not shown). In the reaction of PCE oxide in 100 mM
sodium phosphate buffer (pH 7.4) at 0 °C, oxalyl phos-
phate was formed more slowly and the yield was 18 mol
%, based on PCE oxide (Figure 1). The observed first-
order rate constant was 0.056 ( 0.004 min-1, which is
slightly less than the rate of decomposition of PCE oxide
(note no effect of 0.10 M phosphate on rate of PCE oxide
decomposition, vide supra).
Oxalyl phosphate was not stable in the phosphate
buffer (especially at higher temperature) and decomposed
slowly with first-order kinetics, with concomitant forma-
tion of oxalic acid. The rate constant of decomposition
(hydrolysis to oxalic acid and inorganic phosphate) in 100
mM sodium phosphate at 37 °C was 0.013 ( 0.004 min-1
(t1/2 ) 53 min) (Figure 2). At 0 °C, hydrolysis of the
phosphate was not observed during the course of 120 min.
No CO formation was observed in the decomposition of
oxalyl phosphate (results not shown).
Neither HCO2H nor HCHO was detected as a product
(limit of detection 0.1% yield).
Decom p osition P r od u cts of Oxa lyl Ch lor id e. Al-
though oxalic acid was not detected as a decomposition
product of PCE oxide (vide supra), oxalyl chloride has
been considered as reactive product resulting from the
decomposition of PCE oxide. The decomposition products
of oxalyl chloride, therefore, were investigated under the
same conditions used for the reactions of PCE oxide. CO
and CO2 were the major products formed in MOPS buffer
(Table 2) and oxalic acid was not detected. The formation
of oxalyl phosphate was also observed in the phosphate
buffer.
F or m a tion a n d Rea ction s of Oxa lyl P h osp h a te.
Because oxalyl phosphate was formed from PCE oxide
(Table 1) and oxalyl chloride in phosphate buffer, kinetic

Reactions of Tetrachloroethylene Oxide
Chem. Res. Toxicol., Vol. 15, No. 8, 2002 1101
F igu r e 3. Formation of Cl3CCO2H and oxalic acid in NADPH-
fortified rat liver microsomes. The rate of formation of Cl3CCO2H
(O) and oxalic acid (b) were 0.32 and 0.06 nmol min-1 (mg
protein)-1, respectively.
F igu r e 4. HPLC of products of reaction of PCE oxide and Lys.
(A) Reference standards. (B) The reaction contained 10 mM of
each reactant (in H2O). The identities of the peaks were
established by tR values and subsequently by HPLC/MS (Figure
5). N2-Oxalyl ) N2-oxalylLys; N6-oxalyl ) N6-oxalylLys; N2,N2-
bis ) bis (N2-lysyl)oxalyl amide; N2,N6-bis ) (N2,N6-lysyl)oxalyl
amide; N6,N6-bis ) bis (N6-lysyl)oxalyl amide.
Micr osom a l Oxid a tion of P CE. P450-catalyzed oxi-
dation of PCE has been reported to yield Cl3CCO2H and
oxalic acid, with Cl3CCO2H predominating (6, 8). We
carried out oxidation of PCE with liver microsomes
prepared from phenobarbital-treated rats and observed
the same pattern (Figure 3), with respective rates of 0.32
nmol of Cl3CCO2H and 0.06 nmol of oxalic acid formed
min-1 (mg microsomal protein)-1
.
P r od u cts of Rea ction s w ith F r ee Lys. Reaction of
PCE oxide with Lys was investigated to gain insight into
the reactive electrophile(s) formed from the oxide. Reac-
tions of trichloroacetyl chloride, oxalyl chloride, and
oxalyl phosphate with Lys were also performed in H2O
or 100 mM potassium MOPS or 100 mM sodium phos-
phate buffer (pH 7.4) at 0 °C. Trichloroacetyl chloride
(final 10 mM) reacted with Lys (final 10 mM) to yield
trace amounts of the N-trichloroacetyl derivatives in all
three solutions. Oxalyl chloride (10 mM) reacted with Lys
(10 mM) to give an adduct that corresponded to the major
adduct formed from PCE oxide and Lys (vide infra),
although in low yield. Oxalyl phosphate (2.1 mM) did not
react with Lys (2 mM) in H2O. PCE (10 mM) was also
unreactive toward Lys (10 mM) in H2O, as expected. PCE
oxide (10 mM) gave several adducts with Lys (10 mM)
in the three solutions. The reaction product of the oxide
with Lys in H2O (90 min) was analyzed using LC/MS,
with five products at tR 3.40, 6.28, 9.15, 12.3, and 19.0
min (Figure 4). The small peak at tR ) 3.82 min was
unreacted Lys. By comparison with the authentic com-
pounds, the compounds with tR ) 6.28, 9.15, and 12.3
min were assigned to bis(N2-lysyl)oxalyl amide, (N2,N6-
lysyl)oxalylamide, and bis(N6-lysyl)oxalyl amide, respec-
tively. Bis(N2-lysyl)oxalyl amide was the major adduct
and was also formed from the reaction of oxalyl chloride
with Lys in H2O, as described above. These oxalyl amides
showed MH+ ions at m/z 347 (Figures 5 and 6, panels
B-D). The peak at tR 3.40 min (m/z 219, Figure 6A) was
shown to be N2-oxalylLys but not N6-oxalylLys by com-
parison of tR values. The identity of the compound eluting
at tR 19.0 min (m/z 218, Figure 6E, cf. Figure 4B) remains
unknown.2 Control experiments indicated no formation
F igu r e 5. HPLC/MS analysis of products of reaction of PCE
oxide with Lys. The reaction proceeded in H2O at 0 °C for 90
min. N2-Oxalyl ) N2-oxalylLys; N2,N2-bis ) bis (N2-lysyl)oxalyl
amide; N2,N6-bis ) (N2,N6-lysyl)oxalyl amide; N6,N6-bis ) bis
(N6-lysyl)oxalyl amide.
of these Lys adducts when the oxide hydrolysis preceded
the addition of Lys under the same conditions.
The amounts of the bis Lys oxalyl amides increased
with increasing concentrations of PCE oxide (Figure 7).
P r od u cts of Rea ction of P CE Oxid e w ith Albu -
m in . Bovine serum albumin was reacted with varying
amounts of PCE oxide and the Lys adducts were sepa-
rated and quantified by HPLC/MS after derivatization,
with the use of appropriate Orn derivatives as internal
standards. In this system, the formation of bis adducts
should be rare, and only the N6-adducts will be formed.
The major product was the N6-oxalylLys derivative, and
the N6-trichloroacetylLys adduct accounted for <5% of
the total (Figure 8).
2
+
The molecular ion would correspond to an NH4 adduct of an
internally cyclized oxalyl Lys adduct, but attempts to synthesize such
a standard were unsuccessful.

1102 Chem. Res. Toxicol., Vol. 15, No. 8, 2002
Yoshioka et al.
F igu r e 6. Mass spectra of products of reaction of PCE oxide with Lys. Chromatograms are shown in Figure 5. N2-Oxalyl ) N2-
oxalylLys; N2,N2-bis ) bis (N2-lysyl)oxalyl amide; N2,N6-bis ) (N2,N6-lysyl)oxalyl amide; N6,N6-bis ) bis (N6-lysyl)oxalyl amide.
F igu r e 7. Formation of Lys adducts from PCE oxide as a
F igu r e 8. Formation of Lys adducts in albumin treated with
PCE oxide. Bovine serum albumin (1.0 mg mL-1) was incubated
with varying concentrations of PCE oxide for 90 min at 0 °C
and the Lys adducts were quantified by HPLC/MS following
derivatization with phenylisothiocyanate.
function of PCE oxide concentration. Lys (1.0 mM) was incu-
bated with varying concentrations of PCE oxide in H2O (0 °C,
90 min) and the products were analyzed by HPLC/UV as
described.
Discu ssion
oxalyl equivalents with nucleophiles. The limited forma-
tion of oxalic acid and Cl3CCO2H is not consistent with
the products of microsomal oxidation of PCE, suggesting
that the epoxide is not an obligatory intermediate in the
P450 reaction (Scheme 3).
The prominence of C-C bond scission was not noted
in previous hydrolytic studies with PCE oxide, except in
one report by Henschler et al. (14), in which the reported
yields of CO and CO2 were ∼30%. The yields of CO and
CO2 in our study were 73 and 63%, respectively (Table
1). Henschler et al. (14) reported the formation of CO2 in
an amount equivalent to CO, although the method of
In light of the attention given to PCE oxide as an
oxidation product of PCE (1, 2, 5, 6, 8, 10, 12-14, 35),
we prepared this epoxide and characterized some of its
properties and its behavior in several reactions. At 37
°C, the t1/2 in aqueous buffer (pH 7.4) was 2.6 min,
compared to the t1/2 of 11.6 min reported by Kline et al.
(36).3 Analysis of the hydrolysis products indicated the
prominence of C-C bond scission and the reaction of
3
The discrepancy may be due in part to the fact that the rate was
determined in a 33% acetone solution in the work of Kline et al. (36).

Reactions of Tetrachloroethylene Oxide
Chem. Res. Toxicol., Vol. 15, No. 8, 2002 1103
Sch em e 3. P ostu la ted Oxid a tion of P CE by P 450a
phate underwent first-order hydrolysis to oxalic acid,
with t1/2 ) 59 min at 37 °C (Figure 3). Thus, under the
conditions used in Table 1, no oxalic acid would be
expected in the time frame of the experiment. The
patterns of PCE oxide products obtained in the reactions
done in the absence and presence of phosphate differ
(Table 1), but as indicated, the kinetics of PCE oxide
decomposition were identical. Thus, the epoxide must be
hydrolyzed, and then a product reacts rapidly with
phosphate (Scheme 4), rather than have a nucleophilic
attck of phosphate on PCE oxide itself. The phosphate
appears to be a poor leaving group, as judged by the
stability and the lack of reactivity of oxalyl phosphate
with Lys. The formation of such a phospho-adduct may
be a general issue in reactions that generate potential
acylating agents and possibly alkylating agents, in that
phosphate ions may block other reactions. Some of the
microsomal studies reported with PCE had been done in
0.1 M phosphate buffers (6, 35). Whether such reactions
(of reactive products of PCE oxide) occur with DNA
phosphates in unknown.
a
In the Fe-O- intermediate, intermediate, the broken arrow
indicates that epoxide formation competes with acid chloride
formation.
Sch em e 4. P ostu la ted Mech a n ism s of
Decom p osition a n d Rea ction s of P CE Oxid e
The reaction of PCE oxide with free Lys yielded
primarily oxalyl amides, consistent with other reactions
such as the formation of oxalyl phosphate. Thus, reaction
of the oxalyl species with a nucleophile appears to trap
it, in a reaction that competes with C-C bond scission
(Scheme 4). The mono-oxalyl amides were formed at both
the N2 and N6 atoms, with N2 predominant due to its
lower pKa (37). Cross-linked products containing two Lys
groups bound with an oxalo species were formed. The
Lys-N2 adducts are, in a sense, artifacts in the sense that
they are only found in the models and would not be
formed in protein reactions. We detected only trace
amounts of trichloroacetyl derivatives in the reactions
of PCE oxide with free Lys (Figures 4 and 5). Reaction
of PCE oxide with albumin yielded predominately oxalo
Lys derivatives as opposed to trichloroacetyl Lys deriva-
tives (Figure 8). This pattern is similar to that observed
with TCE oxide (20), in that the amount of (di/tri)-
chloroacetyl Lys is low compared to another amide.
However, the major product in the TCE oxide reacton
was N6-formylLys. Here with PCE oxide, an adduct
formed from the 2-carbon entity is dominant.
analysis was unspecified. We estimated CO2 formation
from PCE oxide using a selective electrode.4 The results
support the view that C-C bond scission yields CO and
CO2 (Scheme 4). No HCHO or HCO2H was detected, and
the formation of CH3OH seems highly unlikely. Oxalyl
chloride also decomposes to CO and CO2 (Table 2). A
mechanism for C-C bond cleavage is presented in
Scheme 4, based on the more extensive studies with TCE
oxide (18). Alternate possibilities can be considered, such
as a scission to generate free phosgene which could go
on to hydrolyze. However, no evidence for Lys adducts
generated from phosgene was apparent (Figures 4 and
5).
In previous work with TCE oxide, we found that the
yields of several hydrolysis products were lower in the
presence of phosphate buffer (17). The same pattern was
observed in this work with PCE oxide (Table 1), where
much less CO and CO2 were recovered. The results
suggested that oxalyl phosphate might be formed, and
HPLC analysis of the PCE oxide hydrolysis reaction
yielded a peak subsequently identified as oxalyl phos-
phate by co-chromatography with authentic material and
its demonstrated conversion to oxalic acid. Oxalyl phos-
The course of products formed from PCE oxide can be
compared with products of microsomal oxidation of PCE.
The literature indicates the production of Cl3CCO2H >
oxalic acid (6, 8), reflecting the in vivo products, and the
same pattern was observed here (Figure 3). This pattern
contrasts with the products of the hydrolysis of PCE oxide
(Table 1). Further, Dekant and his associates have shown
that the major protein recovered from a microsomal PCE
oxidation system is N6-trichloroacetylLys and apparently
not an oxalyl or other derivative (38). Why is there a
difference in products between P450 oxidation of PCE
and PCE oxide hydrolysis? We propose that the apparent
discrepancy can be explained with proposals we have
provided for TCE (Scheme 3) (17) and vinylidene chloride
(19). The P450 intermediate is a tetrahedral FeO-PCE
entity that can collapse via Cl- migration to yield
trichloroacetyl chloride or via formation of PCE oxide,
with the former being favored. Also, the enzyme (P450)
4
A number of other methods were considered and all were unsat-
isfactory. Gas production was not appropriate because of the known
production of CO. Ba2+ precipitation proved unreliable. Acid-base
titrations of HCO3- were not quantitative because of the formation of
HCl.

1104 Chem. Res. Toxicol., Vol. 15, No. 8, 2002
Yoshioka et al.
trophilic reactivity of the metabolically formed epoxides. Arch.
Toxicol. 39, 7-12.
(14) Henschler, D., and Hoos, R. (1982) Metabolic activation and
deactivation mechanisms of di-, tri-, and tetrachloroethylenes.
Adv. Exp. Med. Biol. 136A, 659-666.
(15) Ruden, C. (2002) The use of mechanistic data and the handling
of scientific uncertainty in carcinogen risk assessments. Regul.
Toxicol. Pharmacol. 35, 80-94.
(16) Loizou, G. D. (2001) The application of physiologically based
pharmacokinetic modelling in the analysis of occupational expo-
sure to perchloroethylene. Toxicol. Lett. 123, 59-69.
(17) Miller, R. E., and Guengerich, F. P. (1982) Oxidation of trichlo-
roethylene by liver microsomal cytochrome P-450: evidence for
chlorine migration in a transition state not involving trichloro-
ethylene oxide. Biochemistry 21, 1090-1097.
(18) Cai, H., and Guengerich, F. P. (1999) Mechanism of aqueous
decomposition of trichloroethylene oxide. J . Am. Chem. Soc. 121,
11656-11663.
(19) Liebler, D. C., and Guengerich, F. P. (1983) Olefin oxidation by
cytochrome P-450: evidence for group migration in catalytic
intermediates formed with vinylidene chloride and trans-1-
phenyl-1-butene. Biochemistry 22, 5482-5489.
(20) Cai, H., and Guengerich, F. P. (2000) Acylation of protein lysines
by trichloroethylene oxide. Chem. Res. Toxicol. 13, 327-335.
(21) Cai, H., and Guengerich, F. P. (2001) Reaction of trichloroethylene
oxide with proteins and DNA: instability of adducts and modula-
tion of functions. Chem. Res. Toxicol. 14, 54-61.
(22) Cai, H., and Guengerich, F. P. (2001) Reaction of trichloroethylene
and trichloroethylene oxide with cytochrome P450 enzymes:
inactivation and site of modification. Chem. Res. Toxicol. 14, 451-
458.
may favor formation of oxalic acid from the intermediate.
In conclusion, the general features of PCE oxide
reactions resemble those demonstrated with other halo-
genated epoxides. Hydrolysis is dominated by C-C bond
scission (Table 1) (17-19). PCE oxide is more stable than
the unsymmetrical epoxides, as previously shown by
others (5, 13, 14). Some Cl- migration does occur, even
in the absence of Lewis acids (39). Reaction of PCE oxide
with Lys generates primarily amides arising from acyl
halides (or their equivalents). As documented with vi-
nylidene chloride oxide (40) and TCE oxide (18), the
electrophile reacting with the nucleophile (Lys) is not the
epoxide itself (which would not yield a stable adduct) but
a reaction hydrolysis product.
The analysis of protein adducts with PCE oxide has
been limited to Lys adducts in this study, which are
stable. However, in previous work with TCE oxide, we
found that the majority of protein adducts were unstable
and were released (nonenzymatically) from proteins with
a t1/2 of ∼1 h (21, 22), and similar behavior might be
expected with the PCE oxide adducts. This aspect of the
work is being pursued at this time. Finally, the relevance
of any protein adducts to biological functions and toxicity
remains to be considered.
Ack n ow led gm en t. This work was supported by U.S.
Public Health Service Grants R01 ES10546 and P30
ES00267. We thank M. Voehler for his assistance with
some of the NMR experiments and M. L. Manier for
assistance with the MS work.
(23) Gottlieb, H. E., Kotlyar, V., and Nudelman, A. (1997) NMR
chemical shifts of common laboratory solvents as trace impurities.
J . Org. Chem. 62, 7512-7515.
(24) Campbell, R. W., and Vogl, O. (1977) A practical synthesis of
tetrachloroethylene oxide. J . Macromol. Sci.-Chem. A11, 515-
534.
(25) Kofron, J . L., and Reed, G. H. (1990) Synthesis of oxalyl phosphate
and processing of the acyl phosphate by phosphoenolpyruvate-
dependent enzymes. Arch. Biochem. Biophys. 280, 40-44.
(26) Kim, J ., and Duraway-Mariano, D. (1996) Phosphoenolpyruvate
mutase catalysis of phosphoryl transfer in phosphoenolpyruvate;
kinetics and mechanism of phosphorus-carbon bond formation.
Biochemistry 35, 4628-4635.
(27) Ames, B. (1966) Assay of inorganic phosphate, total phosphate
and phosphatases. Methods Enzymol. 8, 115-118.
(28) Coughlin, D. J ., and Wood, R. (1994) Preparation of glycyltyrosyl-
(N-diethylenetriaminepentaacetic acid)lysine hydrochloride (gyk-
dtpa), U. S. Patent Appl. 92-890519, cited as Statutory Invent.
Regist.
(29) Nunn, P. B., Partridge, L. Z., and Euerby, M. R. (1989) Synthesis
and characterization of ω-N-oxalyl derivatives of some L- and
D-diamino acids. J . Chem. Res. M, 2401-2415.
(30) Balajthy, Z. (1991) An improved preparation of N-benzyloxycar-
bonyl-L-lysine methyl ester hydrochloride. Org. Prep. Proced. Int.
23, 375-376.
(31) Epstein, J ., Rosenthal, R. W., and Ess, R. J . (1955) Use of p-(4-
nitrobenzyl)pyridine as analytical reagent for ethylenimines and
alkylating agents. Anal. Chem. 27, 1435-1439.
(32) Nash, T. (1953) The colorimetric estimation of formaldehyde by
means of the Hantzsch reaction. Biochem. J . 55, 416-421.
(33) Guengerich, F. P., Dannan, G. A., Wright, S. T., Martin, M. V.,
and Kaminsky, L. S. (1982) Purification and characterization of
liver microsomal cytochromes P-450: electrophoretic, spectral,
catalytic, and immunochemical properties and inducibility of eight
isozymes isolated from rats treated with phenobarbital or â-naph-
thoflavone. Biochemistry 21, 6019-6030.
(34) Guengerich, F. P. (2001) Analysis and characterization of enzymes
and nucleic acids. In Principles and Methods of Toxicology (Hayes,
A. W.) 4th ed., pp 1625-1687, Taylor & Francis, New York.
(35) Greim, H., Bonse, G., Radwan, Z., Reichert, D., and Henschler,
D. (1975) Mutagenicity in vitro and potential carcinogenicity of
chlorinated ethylenes as a function of metabolic oxirane forma-
tion. Biochem. Pharmacol. 24, 2013-2017.
Refer en ces
(1) U. S. Departament of Health and Human Services (1993) Toxi-
cological profile for tetrachloroethylene, Public Health Service,
Agency for Toxic Substances and Disease Registry, Atlanta.
(2) Reichert, D. (1983) Biological actions and interactions of tetra-
chloroethylene. Mutat. Res. 123, 411-429.
(3) National Cancer Inst. (1977) Bioassay of tetrachloroethylene for
possible carcinogenicity. National Toxicology Program Technical
Report, 77-813.
(4) National Cancer Inst. (1986) Carcinogenesis bioassay of tetra-
chloroethylene. National Toxicology Program Technical Report,
232.
(5) Henschler, D. (1977) Metabolism and mutagenicity of halogenated
olefinssa comparison of structure and activity. Environ. Health
Perspect. 21, 61-64.
(6) Birner, G., Rutkowska, A., and Dekant, W. (1996) N-Acetyl-S-
(1,2,2-trichlorovinyl)-L-cysteine and 2,2,2-trichloroethanol. Two
novel metabolites of tetrachloroethene in humans after oc-
cupational exposure. Drug Metab. Dispos. 24, 41-48.
(7) Volkel, W., Friedewald, M., Lederer, E., Pahler, A., Parker, J .,
and Dekant, W. (1998) Biotransformation of perchloroethylene:
Dose-dependent excretion of trichloroacetic acid, dichloroacetic
acid, and N-acetyl-S-(trichlorovinyl)-L-cysteine in rats and hu-
mans after inhalation. Toxicol. Appl. Pharmacol. 153, 20-27.
(8) Costa, A. K., and Ivanetich, K. M. (1980) Tetrachloroethylene
metabolism by the hepatic microsomal cytochrome P-450 system.
Biochem. Pharmacol. 29, 2863-2869.
(9) White, I. N. H., Razvi, N., Gibbs, A. H., Davies, A. M., Manno,
M., Zaccaro, C., De Matteis, F., Pahler, A., and Dekant, W. (2001)
Neoantigen formation and clastogenic action of HCFC-123 per-
chloroethylene in human MCL-5 cells. Toxicol. Lett. 123, 129-138.
(10) Daniel, J . W. (1963) The metabolism of 36Cl-labeled trichloro-
ethylene and tetrachloroethylene in the rat. Biochem. Pharmacol.
12, 795-802.
(11) Ikeda, M., Ohtsuji, H., Imamura, T., and Komoike, Y. (1972)
Urinary excretion of total trichloro-compounds, trichloroethanol,
and trichloroacetic acid as a measure of exposure to trichloro-
ethylene and tetrachloroethylene. Brit. J . Industr. Med. 29, 328-
333.
(36) Kline, S. A., Solomon, J . J ., and Van Duuren, B. L. (1978)
Synthesis and reactions of chloroalkene epoxides. J . Org. Chem.
43, 3596-3600.
(37) Mahler, H. R., and Cordes, E. H. (1971) Biological Chemistry, 2nd
ed., Harper & Row, New York.
(12) Powell, J . F. (1945) Trichloroethylene: absorption, elimination,
and metabolism. Brit. J . Industr. Med. 2, 142-145.
(13) Henschler, D., and Bonse, G. (1977) Metabolic activation of
chlorinated ethylenes: dependence of mutagenic effect on elec-
(38) Birner, G., Richling, C., Henschler, D., Anders, M. W., and
Dekant, W. (1994) Metabolism of tetrachloroethene in rats:

Reactions of Tetrachloroethylene Oxide
Chem. Res. Toxicol., Vol. 15, No. 8, 2002 1105
identification of Nꢀ-(dichloroacetyl)-L-lysine and Ne-(trichloro-
acetyl)-L-lysine as protein adducts. Chem. Res. Toxicol. 7, 724-
732.
(40) Liebler, D. C., Meredith, M. J ., and Guengerich, F. P. (1985)
Formation of glutathione conjugates by reactive metabolites of
vinylidene chloride in microsomes and isolated hepatocytes.
Cancer Res. 45, 186-193.
(39) McDonald, R. N. (1969) Rearrangements of R-halo epoxides and
related R-substituted epoxides. In Mechanisms of Molecular
Migrations (Thyagarajan, B. S.) pp 67-107, Wiley, New York.
TX020028J