N-methylcarbamoylated valine of hemoglobin in humans after exposure to N,N-Dimethylformamide: Evidence for the formation of methyl isocyanate?

Article abstract of DOI:10.1021/tx000230r

N,N-Dimethylformamide (DMF) is reported to cause testicular germ-cell tumors in exposed workers. The reports, however, are not in line with results obtained in animal and in vitro experiments, where DMF was shown not to be mutagenic and also not to be carcinogenic. Considerable interest raised on the formation of a reactive intermediate, presumably methyl isocyanate (MIC), during metabolism of DMF in humans over the last years. We report the formation of N-methylcarbamoylated valine of hemoglobin (Hb) in blood samples from workers exposed to DMF in the polyacrylic fiber industry. N-Methylcarbamoylated Hb was formed by the reaction of MIC with Hb. For this purpose, Hb adducts were monitored by means of a modified Edman degradation involving the release of the N-terminal valine adduct in form of 3-methyl-5-isopropylhydantoin (MIH). For internal standardization of the method, 3-ethyl-5-isopropylhydantoin (EIH) was used. Separation and analysis of MIH and EIH were carried out by gas chromatography and mass spectrometry with electron impact ionization (GC/EIMS). Hb adducts in form of MIH were quantified in blood samples from exposed persons in concentrations between 26.1 and 412.0 nmol of MIH/g of globin. The observed adducts were proven to be identical to those derived from the in situ reaction between Hb and MIC. Taken together with the fact that only N-methylcarbamoylated Hb can undergo ring closure to the corresponding hydantoin, the reaction is indirect evidence for the occurrence of MIC in vivo. The formation of MIC directly in the cell and its possible distribution through the human body may lead to critical effects after exposure to DMF. Adducts were determined not to be totally specific for exposure to DMF since an identical adduct was also found in blood samples from the general population. However, concentrations were lower by a factor of about 100. The sources for background adducts are currently unknown.

Full text of DOI:10.1021/tx000230r

Chem. Res. Toxicol. 2001, 14, 833-840
833
N-Meth ylca r ba m oyla ted Va lin e of Hem oglobin in
Hu m a n s a fter Exp osu r e to N,N-Dim eth ylfor m a m id e:
Evid en ce for th e F or m a tion of Meth yl Isocya n a te?
Heiko Udo K a¨ fferlein*,†,‡ and J u¨ rgen Angerer†
Institut und Poliklinik f u¨ r Arbeits-, Sozial-, und Umweltmedizin, Friedrich-Alexander Universit a¨ t
Erlangen-N u¨ rnberg, Schillerstrasse-25/ 29, D-91054 Erlangen, Germany
Received October 31, 2000
N,N-Dimethylformamide (DMF) is reported to cause testicular germ-cell tumors in exposed
workers. The reports, however, are not in line with results obtained in animal and in vitro
experiments, where DMF was shown not to be mutagenic and also not to be carcinogenic.
Considerable interest raised on the formation of a reactive intermediate, presumably methyl
isocyanate (MIC), during metabolism of DMF in humans over the last years. We report the
formation of N-methylcarbamoylated valine of hemoglobin (Hb) in blood samples from workers
exposed to DMF in the polyacrylic fiber industry. N-Methylcarbamoylated Hb was formed by
the reaction of MIC with Hb. For this purpose, Hb adducts were monitored by means of a
modified Edman degradation involving the release of the N-terminal valine adduct in form of
3
-methyl-5-isopropylhydantoin (MIH). For internal standardization of the method, 3-ethyl-5-
isopropylhydantoin (EIH) was used. Separation and analysis of MIH and EIH were carried
out by gas chromatography and mass spectrometry with electron impact ionization (GC/EI-
MS). Hb adducts in form of MIH were quantified in blood samples from exposed persons in
concentrations between 26.1 and 412.0 nmol of MIH/g of globin. The observed adducts were
proven to be identical to those derived from the in situ reaction between Hb and MIC. Taken
together with the fact that only N-methylcarbamoylated Hb can undergo ring closure to the
corresponding hydantoin, the reaction is indirect evidence for the occurrence of MIC in vivo.
The formation of MIC directly in the cell and its possible distribution through the human
body may lead to critical effects after exposure to DMF. Adducts were determined not to be
totally specific for exposure to DMF since an identical adduct was also found in blood samples
from the general population. However, concentrations were lower by a factor of about 100.
The sources for background adducts are currently unknown.
In tr od u ction
yielded negative results (11). Moreover, multiple expo-
sures to other carcinogens (e.g., aromatic amines and
chromium) could not be completely excluded at the
workplace (12). Therefore, whether DMF is carcinogenic
is a question that is not yet solved.
1
N,N-Dimethylformamide (DMF) is one of the most
important industrial solvents with worldwide production
estimated to be ∼250 000 tons/year. DMF is commonly
used in the manufacture of films, fibers, coatings, and
for polyurethane lacquers. Human exposure to DMF
occurs through inhalation and dermal absorption (1). The
main toxic effects reported were hepatotoxicity and
developmental toxicity (2, 3). Walrath et al. (4) reported
no increase in the incidence of cancer in humans in one
epidemiological study. In several other reports, however,
testicular germ-cell tumors (5, 6) were identified among
workers who had been exposed to DMF. The results were
discussed in detail (7-9) because no carcinogenic effects
of DMF were observed in animal studies (10) and also
because the majority of various in vitro test systems
However, Mr a´ z and colleagues reported a fundamental
difference in the metabolism of DMF between humans
and animals (13, 14). N-Acetyl-S-(N-methylcarbamoyl)-
cysteine (AMCC) was found to be a major metabolite in
humans but less in rodents. On the basis of the urinary
excretion of AMCC and its precursor N-acetyl-S-(N-
methylcarbamoyl)glutathione, the formation of a reactive
intermediate, presumably methyl isocyanate (MIC), was
suggested during the metabolism of DMF (15, 16). MIC
may be generated either by oxidation of the formyl carbon
or N-oxidation of N-methylformamide (NMF) (17) (Figure
1
). Recently, Midorikawa et al. (18) reported the forma-
*
To whom correspondence should be addressed. Phone: (919) 558-
tion of carbon-centered or nitrogen-centered radicals
during the degradation of DMF. Radicals were also
suggested by Gescher (17) to be highly reactive interme-
diates during the suspected formation of MIC. The
formation of MIC itself was not described yet, however.
Beside the study by Midorikawa et al., only few articles
could be traced in the literature, which investigate the
mechanistic basis of DMF toxicity (19-21).
1
359. Fax: (919) 558-1404. E-mail: hkaefferlein@ciit.org.
†
Institut und Poliklinik f u¨ r Arbeits-, Sozial-, und Umweltmedizin.
Current address: CIIT Centers for Health Research, 6 Davis Drive,
‡
P.O. Box 12137, Research Triangle Park, NC 27709-2137.
1
Abbreviations: AMCC, N-acetyl-S-(N-methylcarbamoyl)cysteine;
DMF, N,N-dimethylformamide; EIC, ethyl isocyanate; EIH, 3-ethyl-
-isopropylhydantoin; MIC, methyl isocyanate; MIH, 3-methyl-5-iso-
propylhydantoin; MITC, methylisothiocyanate; NEVal, N-ethylcar-
bamoyl-D/L-valine; NMF, N-methylformamide; NMVal, N-methylcar-
bamoyl-D/L-valine.
5
1
0.1021/tx000230r CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/22/2001

8
34 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
K a¨ fferlein and Angerer
exposure to DMF. Ring closure to MIH is not possible
without prior N-methylcarbamoylation of the N-terminal
valine by MIC (Figure 3). This hypothesis was tested in
our studies by analyzing blood samples from workers
exposed to DMF in the synthetic fiber industry. The
results were compared to those derived from (1) in situ
reaction of MIC with Hb and (2) from control persons who
were not exposed to DMF or MIC.
Exp er im en ta l P r oced u r es
F igu r e 1. Suggested formation of methyl isocyanate during
the metabolism of DMF in humans according to Gescher (17)
via oxidative demethylation to N-methylformamide subse-
quently followed by C- or N-oxidation.
Ca u tion : MIC and ethyl isocyanate (EIC) are severe eye, skin,
and mucous membrane irritants and sensitizers. Most deaths
are a result of lung tissue damage (acute toxicity). Both sub-
stances are dangerous, when exposed to heat or flame oxidizers.
Mutations were seen in S. typhimurium and mouse lymphocytes
with exposure to MIC (31).
Ch em ica ls. The chemicals used for standard synthesis and
for GC/MS were obtained by Merck KgaA (Darmstadt, Ger-
many) and Sigma/Aldrich GmbH (Deisenhofen, Germany) and
are regularly found in a laboratory. MIC was synthesized
according to Kaushik et al. (32).
1
F T-NMR Sp ectr oscop y. H NMR, 400 MHz, was performed
on a J EOL GX400 spectrometer. TMS was used as internal
reference and chemical shifts (δ) are given in parts per million
F igu r e 2. Regular Edman degradation: formation of N-
phenylthiocarbamoylated valine of globin and subsequent ring
closure to 3-phenyl-5-isopropylthiohydantoin by heating.
relative to TMS. Samples were prepared in D
2
O (δ 4.80 ppm).
1
1
13
2
D-COSY H NMR and 2D-HETCOR H/ C NMR was carried
1
3
out to verify coupling. C NMR, 100 MHz, was carried out on
a J EOL PS100 spectrometer. The solvent signals derived from
acetone-d
6
(δ 29.95 ppm) were used for internal standardization.
C NMR signals were assigned using the distortionless en-
hancement by polarization transfer (DEPT) method to distin-
1
3
1
13
2 3
guish between CH, CH , and CH signals. All H and C NMR
spectra were recorded at 23 °C, with a pulse delay time of 6.0
s. A 2.2 and 1.0 s delay was used for 2D-COSY and 2D-
HETCOR, respectively. The raw NMR data was processed with
the program Nuts (Acorn NMR Inc., Livermore, CA). The
identity of the standards was also investigated by FAB-MS. The
spectra were recorded on a Micromass ZabSpec spectrometer
F igu r e 3. Formation of N-methylcarbamoylated valine of
globin after reaction of methyl isocyanate and globin. Addition
of concentrated acids and heating leads to the formation of
3
-methyl-5-isopropylhydantoin.
+
using a primary beam of Cs . The matrix was 3-nitrobenzyl
alcohol.
GC/MS In str u m en ta tion . GC/MS analysis was performed
on a GC HP 5890 Series II fitted with a mass selective detector
HP 5972 and a split/splitless injector system HP 7673 operating
in the splitless mode. The inlet purge off time was set to 1.0
min. The injector temperature was set to 280 °C. The transfer
line, ion source, and quadrupole were maintained at 280, 200,
Because MIC has electrophilic properties, it should
react with nucleophilic sites of macromolecules, e.g., with
amino groups in hemoglobin (22). From the literature it
is well-known that the N-terminal valine of hemoglobin
(
Hb) can be split off in form of its corresponding thio-
-6
hydantoin by a regular Edman degradation. For this
purpose, phenyl isothiocyanate is used as derivatizing
reagent (Figure 2). The reaction was modified by T o¨ rn-
qvist et al. (23) using pentafluorophenyl isothiocyanate
to monitor cancer initiators (especially alkylating sub-
stances) through Hb adducts at the N-terminal valine
and 100 °C, respectively. The ion source pressure was ∼3 × 10
Torr. EI with an electron energy of 70 eV was chosen as the
ionization mode at an electron multiplier voltage of 2100 V.
Samples analyzed by GC/EI-MS were injected on a capillary
column with the stationary phase HP-35MS (35%-diphenyl-
6
0
5%-dimethylarylene siloxane copolymer), 60 m with an i.d. of
.25 mm and a phase film thickness of 0.25 µm. Helium was
(
N-alkyl Edman). In her studies, the resulting thio-
used as carrier gas at a constant flow of 1.0 mL/min. 1 µL of
hydantoins were finally analyzed by GC/MS. Nowadays
the N-alkyl Edman method is of widespread use to
determine the internal dose of alkylating substances in
blood samples from both, rodents and humans (24-28).
From a chemical point of view isocyanates are capable
to undergo identical reactions such as isothiocyanates.
Therefore, thiocarbamoylation of valine by phenyl isothio-
cyanate can be taken over from carbamoylation of valine
by isocyanates in vivo. As a consequence, exposure to
isocyanates can be assessed by the formation of hydan-
toins (29). Using this procedure, 3-methyl-5-isopropyl-
hydantoin (MIH) could be analyzed in blood samples from
victims exposed to MIC after the Bhopal incident in 1984
the sample was injected. The initial column temperature at 70
°
C was held for 1 min and then raised to 160 °C at 8 °C/min.
The temperature was held for 17 min. Finally, ramp rate 2
raised the temperature to 270 °C at 25 °C/min and kept it
constant for 10 min. Beside GC/MS-EI, the instrument was also
operated in PCI mode. Methane served as the ionizing gas
(pressure ) 1.0 Torr). Samples analyzed by GC/PCI-MS were
injected on a capillary column with the stationary phase HP
Innowax [100% poly(ethylene glycol), cross-linked], 60 m with
an i.d. of 0.32 mm and a phase film thickness of 0.25 µm. Helium
was used as carrier gas. The initial column temperature of 120
°
C was programmed to rise to 180 °C at 10 °C/min, held for 25
min and then increased to 240 °C at 20 °C/min. The final
temperature was held for 25 min.
(
30). The formation of MIH, however, can be also used
Syn th esis of 3-Meth yl-5-Isop r op ylh yd a n toin . MIH was
synthesized similar to procedures already described by Ram-
as an indirect evidence for the formation of MIC after

Hb Adducts after Exposure to DMF
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 835
F igu r e 4. 1H NMR spectra (D
O) of 3-methyl-5-isopropyl-
2
hydantoin. Due to hindered rotation of the C5-C6 bond, methyl
groups have different shifts. The integration values are normal-
ized to hydrogen 2 (relative value: 1.00).
achandran et al. (33) and Venkateswaran et al. (34). A total of
9
36 mg of d/L-valine (8 mmol) were dissolved in 20 mL of double-
distilled water, and pH 9 was adjusted by addition of a 1 M
Na solution. Then 456 mg of MIC (430 µL, 8 mmol) were
2
O
2
directly added. After stirring for 2 h at RT, 80 mL of concen-
trated HAc/HCl (1:1, v/v) was added, and the solution was
refluxed for 15 min. After evaporation to dryness using a
vacuum rotator at 80 °C, a viscous substance was obtained.
Recrystallation in ethyl acetate yielded 999.6 mg (80.1%) MIH.
1
3
H NMR (Figure 4): δ 0.86 (d, CH
3
3
, J 1-2 ) 6.83 Hz); 1.04 (d,
, J 1-2 ) 6.83 Hz); 2.23 (m, CH); 2.99 (s, NCH ) and 4.17 (d,
), 18.4 (CH ); 24.5
3
CH
3
F igu r e 5. 1H NMR spectra (D
isopropylhydantoin (EIH) and N-ethylcarbamoyl-D/L-valine
NEVal). For signal assignment please refer to text. Integration
2
O) of the mixture 3-ethyl-5-
3
13
NCH, J 3-2 ) 3.66 Hz). C NMR: δ 15.3 (CH
3
3
(CH); 30.2 (NCH
3
); 63.2 (NCH); 160.2 (NCO); and 177.8 (NCON).
(
+
FAB-MS: m/z 157 [M + 1] and 289 [M + Cs ].
values are all normalized to hydrogen 2 (relative value: 1.00
for both EIH and NEVal).
Syn th esis of 3-Eth yl-5-isop r op ylh yd a n toin . EIH was
synthesized according to the procedure described for MIH.
However, EIC was used for N-ethylcarbamoylation. Neverthe-
less, the outcome was different. Two products were obtained
that may be assigned to EIH and the ring-opened product
N-ethylcarbamoylvaline (NEVal). The ratio was found to be
about 1:1.
2
3
3
3.50 (m, NCH );
(d, EIH-NCH,
2
3.99 (d, NEVal-NCH,
J 3-2 ) 3.42 Hz).
J 3-2 ) 4.40 Hz) and 4.14
13C NMR Sign a l Assign m en t for EIH a n d NEVa l. On the
basis of chemical shift and the DEPT spectra, 13C NMR signals
between 12 and 20 ppm could be assigned to CH groups, while
signals for CH-, NCH -, and NCH-carbons, were found at ∼30,
3
1H NMR Sign a l Assign m en t for EIH a n d NEVa l. Since
chemical structures of MIH and EIH are similar and NMR
spectra were recorded under identical conditions, 1H NMR
assignments for EIH (with the exception of the ethyl moiety)
2
34, and 60 ppm, respectively. Signals for nonprotonated CO-
carbons were between 150 and 190 ppm (Figure 6). On the basis
of the 2D-HETCOR spectrum signals at δ 15.2 and 18.4, 30.3,
and 62.8 ppm were assigned to 1, 2, and 3, respectively. As a
consequence (and also based on the 2D-HETCOR), the second
set of signals at δ 17.3 and 17.8, 29.4, and 58.8 ppm were
assigned to 1*, 2*, and 3*. The signal at δ 159.8 ppm (labeled
4) was nearly identical (within 0.4 ppm) to the NCO-group in
MIH and, therefore, assigned to the NCO-group of EIH. As a
consequence (and based on the DEPT spectrum), δ 172.1 ppm
was assigned to 4* (COOH) in NEVal. The signal at δ 177.3
ppm was nearly identical to the NCON-signal in MIH (δ 177.8
ppm) and occurred at twice the relative intensity compared to
4 or 4*. Since carbons 4 (in EIH) and 4* (in NEVal) are less
similar in chemical environments than carbons 5 and 5*, δ 177.3
ppm was assigned to both (5/5*), the NCON-group in EIH and
NEVal. Signals for EIH (2, 3, and 4) are always slightly higher
than those for NEVal (2*, 3*, and 4*). On the basis of the
relative height and the DEPT spectra, δ 12.2 and 33.8 ppm could
be assigned to 7 and 6 of the ethyl moiety from EIH. The second
set of signals (δ 13.0 and 35.3 ppm) could be assigned to 7* and
6* of the ethyl group from NEVal. The assignments for 6/6*
and 7/7* would be also in accordance with the 2D-HETCOR.
1
could be based on comparison with H NMR chemical shifts for
MIH (Figures 4 and 5). Signals at δ 0.85 and 1.03 ppm (labeled
1
, Figure 5) were assigned to EIH since chemical shifts and
3
coupling constants ( J 1-2) were found to be essentially identical
to those of MIH (Figure 4). H correlations from hydrogen 1 to
1
hydrogens 2 (δ 2.22 ppm) and 3 (δ 4.14 ppm) could be
established by 2D-COSY. On the basis of relative integration,
δ 3.52 ppm was assigned to the methylene group of the ethyl
moiety (4) and connection to the methyl group 5 (δ 1.15 ppm)
was established by 2D-COSY.
An additional set of signals was observed in the 1H NMR
spectrum, reflecting the presence of a second molecule, presum-
ably the open form NEVal (Figure 5). Signals at δ 1.08 and 1.10
ppm were assigned to the methyl moieties of the isopropyl group
in NEVal (labeled 1*). Connectivities from 1* to 2* and 3* could
be established by 2D-COSY. On the basis of the relative
integration and 2D-COSY, δ 3.10 and 1.30 ppm were assigned
to 4* and 5*, respectively. Since 4, 4*, and 5, 5* in EIH and
NEVal are only based on relative integration to the respective
isopropyl protons, it is possible that these assignments are
1
3
interchangeable. H NMR: δ 0.85 (d, EIH-CH
3
,
J
1-2 ) 6.83
3
Hz); 1.03 (d, EIH-CH
1-2 ) 7.08 Hz); 1.10 (d, NEVal-CH
3
, J 1-2 ) 6.84 Hz); 1.08 (d, NEVal-CH
3
,
2
The signals can be assigned to either EIH or NEVal since signal
assignment is based only on relative integration (for H NMR) or
relative height (for C NMR).
3
3
J
3
,
J
1-2 ) 7.08 Hz); 1.15,
1
2
13
1
.30 (t, CH
2
CH
3
); 2.22 (m, EIH-CH); 2.38 (m, NEVal-CH); 3.10,

8
36 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
K a¨ fferlein and Angerer
was healthy, male, nonsmoker, nonexposed to DMF or MIC, and
did not take any drugs prior to blood sampling. Red blood cells
were separated from the plasma and cells were washed and
lysed. Then, the sample was divided in two parts each contain-
ing 2 mL of lysed blood. A total of 20 µL of MIC was added to
sample 1, while sample 2 served as a control. After both samples
were shaken rapidly, they were incubated at 37 °C for 2 h. Then
globin was isolated and stored at -28 °C until analysis was
carried out. A double experiment was carried out.
P op u la tion Sa m p les. MIH levels from 35 persons exposed
to DMF in the polyacrylic fiber industry were studied. Workers
were employed in different parts of the facility, e.g., working
directly at the spinning machines, preparing DMF/water solu-
tions, washing fibers, or crimping fibers (post-fiber treatment).
In the company there is a potential co-exposure to carbon
2 2
disulfide (CS ). However, CS is not known to form a carbam-
oylating intermediate in its metabolism. All study subjects were
male. Eighteen persons were smokers, one person took snuff,
and 16 persons were nonsmokers. Exposure assessment was
carried out for the workers by measuring NMF in their urine
samples using a standardized method of the Deutsche Fors-
chungsgemeinschaft DFG (38). The levels were determined to
be between 1.3 and 46.5 mg/L (mean ) 17.0 mg/L, median )
9
.8 mg/L). The results were compared to those derived from
samples of 42 healthy persons who were not known to have been
exposed to MIC or DMF. Twenty-two persons were male, 20
female. Smoking status was assessed by questionnaire (12
smokers, 30 nonsmokers). The persons did not use any drugs
with carbamoylating properties (e.g., disulfiram) and did not
participate in chemoprevention studies or clinical trials for drug
development. NMF in all of their urine samples was determined
to be lower limit of detection (<0.1 mg/L).
Meth od Ca libr a tion a n d Va lid a tion . Calibration stand-
ards were prepared in the range of 5 and 500 nmol of MIH/g of
globin (0.78 and 78.0 µg of MIH/g of globin). For this purpose,
3
.12 mg of MIH were dissolved in 100 mL of ethanol. From this
stock solution, six different calibration standards between 156
and 15.6 mg/L in ethanol were prepared. One milliliter of
standard 1 was added to 200 mg of globin, 1 mL of standard
F igu r e 6. 13C NMR spectra of the mixture 3-ethyl-5-isopro-
pylhydantoin (EIH) and N-ethylcarbamoyl-D/L-valine (NEVal).
For signal assignment please refer to text.
2
to 200 mg of globin, etc., and the globin samples were dried in
a vacuum desiccator overnight. The globin was derived from
nonsmoking persons not exposed to DMF or MIC. Blank samples
containing concentrated HAc/HCl (v/v, 3 mL) were also prepared
to check for possible MIH contamination during sample cleanup.
In every analytical series a complete set of calibration standards
and one blank sample were included. Using this kind of
calibration, losses during sample preparation were compensated.
To prepare a solution of the internal standard (IS), 3.4 mg of
the EIH/NEVal-mixture were dissolved in 100 mL ethanol. After
sample preparation, the quotient of the area counts under the
curve between MIH and EIH was plotted as a function of the
concentration of MIH. The calibration curve was used for the
calculation of MIH concentrations (nmol/g of globin) in each
blood sample. The method was validated. The linearity of the
calibration curve was determined up to a concentration of 1000
nmol/g of globin. Precision in the series (n ) 8) and from day to
day (n ) 8) was determined at two concentrations (25 and 100
nmol/g of globin), while (relative) recovery (n ) 5) was deter-
mined at 50 and 200 nmol/g of globin.
However, since the 13C signals for the ethyl moiety were only
assigned based on their relative height (and the corresponding
1
H signals on their relative integration), 6/6* and 7/7* may be
1
3
2
interchangeable. C NMR: δ 12.2, 13.0 (CH
CH ); 17.3 (NEVal-CH ); 17.8 (NEVal-CH ); 18.4 (EIH-CH
9.4 (EIH-CH); 30.3 (NEVal-CH); 33.8, 35.3 (CH
2
CH
3
); 15.2 (EIH-
);
2
); 58.8 (NEVal-
3
3
3
3
2
2
NCH); 62.8 (EIH-NCH); 159.8 (EIH-NCO); 172.1 (NEVal-
COOH); and 177.3 (EIH and NEVal-NCON). FAB-MS: m/z 171
+
[
+
EIH + 1]; 303 [EIH + Cs ]; 189 [NEVal + 1], and 321 [NEVal
+
Cs ].
P r ep a r a tion of Globin . A total of 2 × 10 mL of blood was
drawn from the vein of exposed and nonexposed persons using
a disposable syringe containing EDTA as anticoagulant. Red
blood cells were separated from the plasma by centrifugation
of the blood samples. Erythrocytes were washed with 0.9%
saline and then lysed. The cell fragments were removed by
centrifugation. After blood samples were lysed, globin was
isolated immediately to avoid artifactual formation of adducts
Sa m p le P r ep a r a tion a n d GC/MS An a lysis. For the analy-
sis of N-methylcarbamoylated valine of globin, 100 µL of the
internal standard were added to 200 mg globin in a 20-mL-
screwed-top vial. Then the globin was dissolved in 3 mL of a
mixture of HAc/HCl (1:1, v/v) and incubated for 1 h at 110 °C.
After the sample cooled, it was neutralized stepwise using 10
M NaOH solution. Saturated saline solution (1 mL) was added,
and the sample was shaken rapidly for 10 min. The pH was
adjusted to 3-5 using concentrated HCl, and then the sample
was extracted with 5 mL of ethyl acetate. The sample was
shaken for 5 min and centrifuged at 3000g for 10 min. The
extraction step was repeated. The organic phases were combined
in a new vial and evaporated to ∼1.5 mL. The solution was
(35). To prepare between 150 and 250 mg of globin, a 50 mM
hydrochloric acid solution in 2-propanol was added to 2 mL of
lysate. Then, ethyl acetate was added slowly to the supernatant,
and the solution was stored at 4 °C overnight. The precipitated
globin was separated by centrifugation, washed several times
with ethyl acetate and n-hexane, and dried in a vacuum
desiccator (36, 37). Globin was stored at -28 °C until analysis
was carried out.
In Situ Exp er im en ts. The theoretical formation of N-
methylcarbamoylated Hb at the N-terminal valine after expo-
sure to MIC was tested in in-situ experiments. For this purpose,
a blood sample was drawn from one volunteer. The individual

Hb Adducts after Exposure to DMF
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 837
only ∼2 levels above physiological pH, it is within the
bounds of probability that MIC is transported in its
“active” form throughout the body. On the other side,
MIC was not directly determined in body fluids yet.
Therefore, distribution in vivo may also occur via a
semistabile, reversible mechanism, e.g., methylcarbam-
oylated thiols or methylcarbamoyl phosphate.
In contrast to MIH, different results were obtained for
the synthesis of EIH. A considerable amount of ring-
opened NEVal was traced after EIC reacted with D/L-
valine for 2 h and reflux in concentrated acids for 15 min.
A ratio of ∼1:1 for EIH/NEVal was found according to
1
the H NMR spectrum. In analogy to the formation of
MIH the kinetics for the formation of EIH was studied
in dependence on the reaction times for step 1 and step
2
. First, the reaction time for step 1 (D/L-valine with EIC)
was increased to 6 and 12 h, while reaction time for step
(reflux in concentrated acids) was kept constant. Then,
F igu r e 7. (A) Fragmentation of 3-methyl-5-isopropylhydantoin
(
m/z 156) yields m/z 114 (McLafferty rearrangement) and m/z
2
5
7. The ratio was found to be 1:160:35, respectively. The high
abundance of m/z 114 in EI-MS is due to stabilization by electron
delocalization (mesomeric effect). (B) Fragmentation of 3-ethyl-
reaction time for step 1 was kept constant (2 h), while
reflux in concentrated acid was extended to 1 and 2 h.
Neither of the experiments changed the outcome as was
5
-isopropylhydantoin is nearly identical. m/z 128 is formed by
McLafferty rearrangement and m/z 100 by subsequent loss of
ethene and H-transfer. The ratio for m/z 170:128:100 was
determined to be 1:280:135, respectively.
1
determined by H NMR. Always a ratio ∼1:1 was found
1
for EIH/NEVal. The results by H NMR show that EIC
can react with the amino group of D/L-valine to form
NEVal. NEVal, however, is only converted to EIH in a
minor extend. Decreased orbital overlapping caused by
increasing repulsive interactions (sterical hindrance)
during hydantoin formation may be of major role for the
observed outcome.
Separation of EIH and NEVal using various organic
solvents and solvent mixtures for recrystallation was not
successful probably because of their structural similarity.
No attempts were made to separate the compounds using
column chromatography because the mixture is only used
as internal standard. The ratio of ∼1:1 was not assumed
to undergo any changes during the cleanup procedure,
which may influence the quantitative result. One can
argue that a ratio of ∼1:1 and difficulties to separate the
substances may also point to the possibility that for some
reason the ring-closed (or ring-opened) D- and L-isomers
were obtained. However, based on stereochemistry alone,
reaction of D/L-valine with a nonchiral molecule should
not produce two products of different chirality.
transferred to a 1.7-mL-crimp-top vial and evaporated to
dryness under a stream of nitrogen. The sample was dried in a
vacuum desiccator over NaOH overnight and resolved in 1 mL
of ethyl acetate again on the following day. After shaking for at
least 15 min and centrifugation for 10 min at 3500g the
supernatant was transferred to a new 1.7-mL-screwed-top vial
and concentrated to 500 µL under a stream of nitrogen. One
microliter was analyzed by GC/MS. Quantitative determination
of Hb adducts derived from workers employed in the polyacrylic
fiber industry and control subjects was carried out by EI-MS
and multiple ion detection of three characteristic ions for MIH
and EIH. The dwell time was maintained at 80 ms/ion trace.
Recorded ion traces for MIH were m/z 156, 114, and 57, while
m/z 170, 128, and 100 were recorded for EIH (Figure 7). The
R R
observed retention times were, t (MIH), 24.6 min, t (EIH), 26.7
min. Recorded ion trace for MIH during PCI-MS was m/z 157
+
[
M + 1] after proton transfer from CH
5
. The retention time
was determined to be t
R
) 28.6 min. PCI-MS was used to verify
the outcome by GC/EI-MS. However, the method was not used
for the quantification of adducts.
Relia bility of th e Meth od . Differences in yields for
the formation of hydantoins derived from the N-terminal
valine are well-known from the literature. Calibration
using the final hydantoins or the single amino acids (in
our case MIH and NMVal) can lead to different results
compared to radiolabeled Hb adduct standards. On the
other side, there is only a negligible difference between
the results derived from calibration using radiolabeled
Hb standards and those derived from calibration with
peptide standards (41, 42). Ring closure to the hydantoin
and the presence of at least one peptide bond (which is
cleaved during cleanup) is therefore of crucial relevance
during Hb adduct analysis. As a consequence, peptide
standards or radiolabeled Hb standards should be used
for calibration in Hb adduct analyses, which are based
on the formation of hydantoins. With such standards, the
N-terminal function of Hb is simulated during cleanup,
and different conversion yields to the corresponding
hydantoin are taken into account. To study the differ-
ences between dipeptide and hydantoin calibration, the
dipeptide standards N-methylcarbamoyl-D/L-valine-D/L-
leucine anilide and N-ethylcarbamoyl-D/L-valine-D/L-leu-
cine anilide were also prepared, and the two calibration
Resu lts a n d Discu ssion
Syn th esis of Sta n d a r d s. A simple and rapid synthe-
sis for MIH is presented. For this purpose, D/L-valine was
carbamoylated with MIC to NMVal (step 1). Instead of
isolation, NMVal was subsequently treated with concen-
trated acids to give MIH (step 2). Step 1 was carried out
in water over a period of 2 h at pH 9 since basic pH levels
were reported to be advantageous for the reaction of D/L-
valine and isocyanates (39, 40). The pH adjustment was
2 2
carried out using a freshly prepared 1 M Na O solution.
The kinetics of the formation of MIH was studied in
dependence on the reaction time for step 1 (15, 30, 60,
and 120 min). A strong increase in yield of MIH was
investigated with increasing time. Yields of MIH were
determined to be 23.8, 59.2, 76.1, and 80.1%, respectively.
On the other side, step 2 was always completed after 15
min under reflux in concentrated acids. As a consequence,
the rate-determining step is the formation of NMVal.
Overall, the experiment points to a remarkable stability
of MIC in aqueous solutions, which may be due to the
2 2
use of Na O as an alkalizing agent. However, as pH 9 is

8
38 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
K a¨ fferlein and Angerer
Ta ble 1. Relia bility Cr iter ia for th e Deter m in a tion of Hb
a
Ad d u cts a fter Exp osu r e to DMF
precision (n ) 8)
relative
concentration
in series
day to day
recovery (n ) 5)
2
5
00
00
5
0
7.2
12.4
89.6
92.7
1
2
6.8
7.1
a
All concentrations are given in nanomoles of MIH per gram
of globin, while precision and relative recovery is given as a
percentage. n ) number of investigated samples.
F igu r e 8. Chromatogram derived from a blood sample of an
exposed worker in the polyacrylic fiber industry containing 73.8
nmol of 3-methyl-5-isopropylhydantoin/gram globin (m/z 156 not
shown). The ratio of m/z 156:114:57 was determined to be 1:171:
procedures were directly compared.3 Surprisingly, no
differences were observed. One explanation may be that
ring-closure to MIH in the presence of concentrated acids
and heat is nearly 100% once NMVal is formed. Since
dipeptide standards are not commercially available, must
be synthesized via a multistep procedure and are not
advantageous compared to the use of MIH and EIH, there
is no need to use them in the particular case of ex-
posure to DMF or MIC. Therefore, future biological
monitoring can be carried out using MIH for calibration,
which makes the procedure easier from a practical point
of view.
Under the conditions described here for sample prepa-
ration and the operational conditions for GC/EI-MS, a
detection limit of 0.4 nmol of MIH/g of globin was
calculated from a signal-to-noise ratio of 3:1 investigating
m/z 114. The limit of detection for two ions (m/z 114, 57),
however, is only 5 nmol/g because of the high abundance
of m/z 114. The calibration curve was determined to be
linear up to 1000 nmol/g of globin. Precision and recovery
for the whole method were found to be in an acceptable
range (Table 1).
In Situ Exp er im en ts. In the globin samples derived
from the in situ experiments MIH could be clearly
identified in MIC treated blood. Using GC/EI-MS and GC/
PCI-MS MIH was identified based on its retention time
and peak ratios compared to a standard solution of MIH
in ethyl acetate (10 mg/L). The amount of MIH formed
in the in situ experiments was sufficient to obtain a total
ion count spectrum of MIH by MS. However, the results
showed that PCI was less sensitive than EI. Possibly,
negative chemical ionization (NCI-MS) may lead to better
results. Unfortunately, we were not able to carry out GC/
NCI-MS. However, the results derived from EI-MS and
PCI-MS directly confirm those obtained by Ramachan-
dran et al. (33), who demonstrated that MIC after
inhalation serve as a N-methylcarbamoylating substance
in vitro and in vivo in rats and rabbits. In our in situ
samples not treated with MIC, MIH was identified in low
amounts near the limit of detection using GC/EI-MS
operating in the single ion monitoring mode for m/z 114.
However, no peak could be detected in the chromatogram
using GC/PCI-MS, which may be due to its lower
sensitivity compared to EI-MS.
3
8.
Ta ble 2. Descr ip tive Sta tistica l An a lysis for Tw o
a
P op u la tion Sa m p les In vestiga ted in th e P r esen t Stu d y
investigated
subjects
range
mean median 5%
95%
exposed workers
26.1-412.0 120.3 109.7 29.8 214.3
21.3-465.0b 123.6b 121.1b 25.3 221.1
b
b
(n ) 35)
general population <LOD-16.6
1.8
1.4
0.4
2.8
(
n ) 42)
a
All concentrations are given in nanomoles of MIH per gram
of globin. n ) number of investigated persons, LOD ) limit of
detection, 5%, 95% ) 5 and 95 percentiles. b Concentrations
obtained by using dipeptide standards for calibration instead of
MIH.
workers. In five spot checks, the presence of MIH was
also verified by GC/PCI-MS. No statistical differences in
MIH levels could be observed between smokers and
nonsmokers (nonparametric Wilcoxon test).
In comparison to a previous publication (43), where we
reported a mean value of 450.8 nmol of MIH/g of globin
in 10 individuals exposed to DMF, Hb adduct levels were
found to be lower in the present study. The difference
may be mainly due to different exposure situations at
the workplace. The exposure of the workers to DMF was
verified by measurements of NMF in urine samples.
NMF analysis in the previous publication (43) revealed
concentrations between 8.8 and 84.6 mg/L (mean ) 43.9
mg/L). Although NMF is only an index of daily exposure
to DMF, the results indicated high exposures to DMF at
4
the workplace in the previous study. However, in the
study presented here, NMF levels were determined to
be between 1.3 and 46.5 mg/L (mean ) 17.0 mg/L).
Therefore, the exposure situation was lower, which is in
line with lower Hb adduct concentrations.
Assuming that MIC is responsible for both, formation
of MIH and AMCC, MIH concentrations should be lower
in rodents since Mr a´ z et al. (13) reported decreased
formation of AMCC in rodents compared to humans. This
suggestion is supported by the observation that metabo-
lism of DMF to AMCC and MIH is closely tied together.
Built-up of MIH levels paralleled the excretion course of
AMCC after a single ip dose of 1000 mg of DMF/kg of
body wt in rats (44). In addition, the levels of MIH and
cumulative excretion of AMCC between the first and
fourth day also correlated. Unfortunately, the results in
rats by Mraz et al. (44) cannot be directly compared to
those presented here in humans since dosage and route
of uptake greatly differ from the situation at the work-
place.
P op u la tion Sa m p les. Thirty-five globin samples from
persons exposed to DMF in the polyacrylic fiber industry
were analyzed. A GC/EI-MS chromatogram derived from
a globin sample of an exposed individual is shown in
Figure 8. MIH could be identified in all blood samples
based on its retention time and peak ratios. A mean of
1
20 nmol/g of globin (Table 2) was found in exposed
3
H.U.K., Sieber, W., Burkhart-Reichl, A. and J .A. (2000) unpub-
4
lished results.
The biological exposure index for NMF in urine is 15 mg/L.

Hb Adducts after Exposure to DMF
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 839
Ack n ow led gm en t. The authors wish to thank the
Deutsche Forschungsgemeinschaft (DFG) for their finan-
cial support of this work (AN 107/15-3 and AN 107/15-
4
). Many thanks are also due to the following persons,
companies, and institutes: Dr. A. Burkhart-Reichl from
Acordis Acrylic Fibers, Kelheim, Germany, for providing
blood samples from workers exposed to DMF; Dr. W.
Bauer and Dr. W. Donaubauer from the Institut f u¨ r
Organische Chemie, Friedrich-Alexander Universit a¨ t
1
13
Erlangen, Germany, for measuring the H, C, 2D-COSY,
2
D-HETCOR NMR, and FAB-MS spectra; Dr. S. J .
F igu r e 9. Chromatogram derived from a blood sample of a
person not exposed to DMF or MIC. Nevertheless a small
amount of MIH could be quantified (1.7 nmol/g globin). Only
m/z 114 can be observed at such low concentrations.
Sumner and Dr. T. R. Fennell from CIIT for their help
in assigning NMR shifts, and Dr. B. Kuyper for editorial
assistance.
To compare our results to those derived from persons
nonexposed to DMF or MIC, blood samples of 42 indi-
viduals from the general population were analyzed. In
contrast to exposed workers, MIH concentrations of the
controls were up to 100 times lower and were determined
to be in the range of 0.4 and 16.6 nmol of MIH/g of globin
Refer en ces
(1) Lauwerys, R. R., Kivits, A., Lhoir, M., Rigolet, P., Houbeau, D.,
Buchet, J . P., and Roels, H. A. (1980) Biological surveillance of
workers exposed to dimethylformamide and the influence of skin
protection on its percutaneous absorption. Int. Arch. Occup.
Environ. Health 45, 189-203.
(
Table 2). A chromatogram derived from a control indi-
(2) Kennedy, G. L., J r. (2001) Biological effects of acetamide, form-
amide, and their monomethyl and dimethyl derivatives: an
update. Crit. Rev. Toxicol. 31, 139-222.
vidual is presented in Figure 9. No statistical differences
in MIH concentrations between males and females were
determined. Although a slight trend to higher MIH levels
could be observed in smokers, the difference between
smokers and nonsmokers was also not significant.
In contrast to reagent blanks, blood samples of the
general population contained small amounts of MIH.
Therefore, MIH is not totally selective for the exposure
to DMF or MIC at the workplace. Since the general
population is not exposed to DMF or MIC, the question
remains as to why MIH was found in their blood samples.
Possible explanations are speculative at the moment and
should be investigated in future studies. The occurrence
of methyl isothiocyanate (MITC) in the diet may be
important. MITC was detected in wine samples (45, 46).
Moreover, thioglucosinolates, which can release MITC
during enzymatic degradation, are present in cabbage
and other cruciferous vegetables (47). MITC can yield
(
3) Fail, P. A., George, J . D., Grizzle, T. B., and Heindel, J . J . (1998)
Formamide and dimethylformamide: reproductive assessment by
continuous breeding in mice. Reprod. Toxicol. 12, 317-332.
4) Walrath, J ., Fayerweather, W. E., Gilby, B. G., and Pell, S. (1989)
A case-control study of cancer among Du Pont employees with
potential for exposure to dimethylformamide. J . Occup. Med. 31,
(
4
32-438.
(5) Ducatman, A. M., Conwill, D. E., and Crawl, C. (1986) Germ-cell
tumors of the testicle among aircraft repairman. J . Urol. 136,
8
34-836.
(
6) Levin, S. M., Baker, D. B., Landrigan, P. J ., Monaghan, S. V.,
Frumin, E., Braithwaite, M., and Towne, W. (1987) Testicular
cancer in leather tanners exposed to dimethylformamide. Lancet
3
33, 1153.
(7) Chen, J . L., and Kennedy, G. L. (1988) Dimethylformamide and
testicular cancer. Lancet 334, 55.
(
8) Calvert, G. M., Fajen, J . M., Hills, B. W., and Halperin, W. E.
(1990) Testicular cancer, dimethylformamide, and leather tan-
neries. Lancet 336, 1253-1254.
(9) Gollins, W. J . (1991) Dimethylformamide and testicular cancer.
Lancet 337, 306-307.
methylamine and CO
2
through the intermediacy of MIC
(
10) Malley, L. A., Slone, T. W., Van Pelt, C., Elliott, G. S., Ross, P.
E., Stadler, J . C., and Kennedy, G. L. (1994) Chronic toxicity/
oncogenicity of dimethylformamide in rats and mice following
inhalation exposure. Fund. Appl. Toxicol. 23, 268-279.
11) International Agency for Research on Cancer (1989) IARC
Monographs on the Evaluation of Carcinogenic Risk to Humans.
Some Organic Solvents, Resin Monomers and Related Compounds,
Pigments and Occupational Exposures in Paint Manufacture and
Painting, Vol. 47, Lyon.
12) Ducatman, A. M. (1989) Dimethylformamide, metal dyes, and
testicular cancer. Lancet 335, 911.
13) Mr a´ z, J ., Cross, H., Gescher, A., Threadgill, M. D., and Flek, J .
(1989) Differences between rodents and humans in the metabolic
toxification of N,N-dimethylformamide. Toxicol. Appl. Pharmacol.
(48). A completely different explanation, however, would
be carbamoylation of Hb subsequently followed by meth-
ylation. Carbamoylated Hb (CHb) is formed in humans
mainly via the reaction of Hb with cyanate, a product of
in vivo urea dissociation. Studies within the 90s (49, 50,
(
5
1) revealed high levels of CHb in healthy, nonexposed
persons (between 200 and 450 nmol/g of Hb). Methylation
of CHb, e.g., by S-adenosylmethionine (52), would yield
N-methylcarbamoylated Hb.
(
(
Overall, the results evidence the formation of Hb
adducts at the N-terminal valine of globin after exposure
to DMF. Hb adducts may serve as a new biomarker on
the molecular level to assess exposure to DMF. Therefore
the method may be used in future biological monitoring.
The study also provides an indirect evidence for the
formation of MIC as a reactive intermediate during
metabolism of DMF in humans since (1) the observed Hb
adducts were proven to be identical to those derived from
in situ reaction of human blood samples with MIC and
9
8, 507-516.
(
14) Mr a´ z, J ., Gescher, A., Cross, H., Shaw, A. J ., and Flek, J . (1991)
New findings in the metabolism of N,N-dimethylformamide-
consequences for evaluation of occupational risk. Sci. Total.
Environ. 101, 131-134.
15) Kestell, P., Threadgill, M. D., Gescher, A., Gledhill, A. P., Shaw,
A. J ., and Farmer, P. B. (1987) An investigation of the relationship
between the hepatotoxicity and the metabolism of N-alkylform-
amides. J . Pharmacol. Exp. Ther. 240, 265-270.
(
(
(
(
16) Mr a´ z, J ., and Nohov a´ , H. (1992) Absorption, metabolism and
elimination of N,N-dimethylformamide in humans. Int. Arch.
Occup. Environ. Health 64, 85-92.
(2) adducts were also identical to those which were
reported after the Bhopal incidence in 1984. Therefore,
the pathological outcome (toxic hepatitis, malformations,
etc.) after exposure to DMF may be mediated by MIC.
As Hb adducts are accepted biomarkers of mutagenic
relevance (53), the study also shed some new light on the
suspected carcinogenicity of DMF.
17) Gescher, A. (1993) Metabolism of N,N-dimethylformamide: key
to the understanding of its toxicity. Chem. Res. Toxicol. 6, 245-
2
51.
18) Midorikawa, K., Murata, M., Oikawa, S., Tada-Oikawa, S., and
Kawanishi, S. (2000) DNA damage by dimethylformamide: Role
of hydrogen peroxide generated during degradation. Chem. Res.
Toxicol. 13, 309-315.

8
40 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
K a¨ fferlein and Angerer
(
(
(
19) Cordeiro, R. F., and Savarese, T. M. (1986) Role of glutathione
depletion in the mechanism of action of N-methylformamide and
N,N-dimethylformamide in a cultured human colon carcinoma cell
line. Cancer Res. 46, 1297-1305.
20) Shaw, A. J ., Gescher, A., and Mr a´ z, J . (1988) Cytotoxicity and
metabolism of the hepatotoxin N-methylformamide and related
formamides in mouse hepatocytes. Toxicol. Appl. Pharmacol. 95,
(36) Mowrer, J ., T o¨ rnqvist, M., J ensen, S., and Ehrenberg, L. (1986)
Modified Edman degradation applied to hemoglobin for monitor-
ing occupational exposure to alkylating agents. Toxicol. Environ.
Chem. 11, 215-231.
(37) Bader, M., Lewalter, J ., and Angerer, J . (1995) Analysis of
N-alkylated amino acids in human Hb: evidence for elevated
N-methylvaline levels in smokers. Int. Arch. Occup. Environ.
Health 67, 237-242.
(38) DFG, Deutsche Forschungsgemeinschaft (1997) in Analyses of
Hazardous Substances in Biological Materials (Angerer, J ., and
Schaller, K. H., Ed.) Vol. 5, pp 97-108, VCH Weinheim.
(39) Rivett, D. E., and Wilshire, J . F. K. (1965) Reactions of
N,N-diphenylcarbamoyl chloride. Aust. J . Chem. 18, 1667-
1675.
(40) M o¨ ller, M., Henschler, D., and Sabbioni, G. (1998) Synthesis and
spectroscopic characterization of 4-chlorophenyl isocyanate ()1-
chloro-4-isocyanatobenzene) adducts with amino acids as potential
dosimeters for the biomonitoring of isocyanate exposure. Helv.
Chim. Acta 81, 1254-1263.
(41) T o¨ rnqvist, M., Magnusson, A. L., Farmer, P. B., Tang, Y. S.,
J effrey, A. M., Wazneh, L., Beulink, G. D., van der Waal, H., and
van Sittert, N. J . (1992) Ring test for low levels of N-(2-
hydroxyethyl)valine in human Hb. Anal. Biochem. 203, 357-
360.
(42) Pauwels, W., Farmer, P. B., Osterman-Golkar, S., Severi, M.,
Cordero, R., Bailey, E., and Veulemans, H. (1997) Ring test for
the determination of N-terminal valine adducts of styrene 7,8-
oxide with Hb by the modified Edman degradation technique. J .
Chromatogr. B 702, 77-83.
(43) Angerer, J ., G o¨ en, T., Kr a¨ mer, A., and K a¨ fferlein, H. U. (1998)
N-methylcarbamoyl adducts at the N-terminal valine of globin
in workers exposed to N,N-dimethylformamide. Arch. Toxicol. 72,
309-313.
1
62-170.
21) Mr a´ z, J ., J heeta, P., Gescher, A., Hyland, R., Thummel, K., and
Threadgill, M. D. (1993) Investigation of the mechanistic basis
of N,N-dimethylformamide toxicity. Metabolism of N,N-dimeth-
ylformamide and its deuterated isotopomers by cytochrome P450
2
E1. Chem. Res. Toxicol. 6, 197-207.
(
(
(
22) Ehrenberg, L., Granath, F., and T o¨ rnqvist, M. (1996) Macromol-
ecule adducts as biomarkers of exposure to environmental mu-
tagens in human populations. Environ. Health Perspect. 104
(Suppl. 3), 423-428.
23) T o¨ rnqvist, M., Mowrer, J ., J ensen, S., and Ehrenberg, L. (1986)
Monitoring of environmental cancer initiators through Hb adducts
by a modified Edman degradation method. Anal. Biochem. 154,
2
55-266.
24) Walker, V. E., Wu, K. Y., Upton, P. B., Ranasinghe, A., Scheller,
N., Cho, M. H., Vergnes, J . S., Skopek, T. R., and Swenberg, J .
A. (2000) Biomarkers of exposure and effect as indicators of
potential carcinogenic risk arising from in vivo metabolism of
ethylene to ethylene oxide. Carcinogenesis 21, 1661-1669.
25) Walker, V. E., MacNeela, J . P., Swenberg, J . A., Turner, M. J .,
J r., and Fennell, T. R., (1992) Molecular dosimetry of ethylene
oxide: formation and persistence of N-(2-hydroxyethyl)valine in
hemoglobin following repeated exposures of rats and mice. Cancer
Res. 52, 4320-4327.
(
(26) Tareke, E., Rydberg, P., Karlsson, P., Eriksson, S., and T o¨ rnqvist,
M. (2000) Acrylamide: a cooking carcinogen? Chem. Res. Toxicol.
1
3, 517-522.
(44) Mr a´ z, J ., G a´ lova, E., Nohov a´ , H., Bou sˇ kov a´ , and Hornychov a´ , M.
(1998) Formation of N-methylcarbamoyl adduct with N-terminal
valine of globin following the exposure to N,N-dimethylformamide
(
27) Perez, H. L., Segerb a¨ ck, D., and Osterman-Golkar, S. (1999)
Adducts of acrylonitrile with hemoglobin in nonsmokers and in
participants in a smoking cessation program. Chem. Res. Toxicol.
th
(DMF) and N-methylformamide (MF) in rats, 4 International
1
2, 869-873.
Symposium on Biological Monitoring in Occupational and Envi-
ronmental Health, 23-25 Sept. 1998, p 125, Seoul, Korea.
(45) Saito, I., Yamada, S., Oshima, H., Ikai, Y., Oka, H., and Hay-
akawa, J . (1994) Analysis of methyl isothiocyanate in wine by
gas chromatography with dual detection. J . AOAC Int. 77, 1296-
1299.
(
28) Tavares, R., Borba, H., Monteiro, M., Proenca, M. J ., Lynce, N.,
Rueff, J ., Bailey, E., Sweetman, G. M., Lawrence, R. M., and
Farmer, P. B. (1996) Monitoring of exposure to acrylonitrile by
determination of N-(2-cyanoethyl)valine at the N-terminal posi-
tion of Hb. Carcinogenesis 17, 2655-2660.
(
29) Sabbioni, G., Hartley, R., Henschler, D., Hollrigl-Rosta, A., K o¨ ber,
R., and Schneider, S. (2000) Isocyanate-specific hemoglobin
adduct in rats exposed to 4,4′-methylenediphenyl diisocyanate.
Chem. Res. Toxicol. 13, 82-89.
(46) Rostron, C. (1992) Methyl isothiocyanate in wine. Food Chem.
Toxicol. 30, 821-823.
(47) Fenwick, G. R., Heaney, R. K., and Mullin, W. J . (1983) Glucosin-
olates and their breakdown products in food and food plants. Crit.
Rev. Food Sci. Nutr. 18, 123-201.
(48) Valentine, W. M., Amarnath, V., Amarnath, K., and Graham, D.
G. (1995) Characterization of protein adducts produced by N-
methyldithiocarbamate and N-methyldithiocarbamate esters.
Chem. Res. Toxciol. 8, 254-261.
(49) Balion, C. M., Draisey, T. F., and Thibert, R. J . (1998) Carbam-
ylated hemoglobin and carbamylated plasma protein in hemo-
dialyzed patients. Kidney Int. 53, 488-495.
(50) Han, J . S., Kim, Y. S., Chin, H. J ., J eon, U. S., Ahn, C., Kim, S.,
Lee, J . S., J ang, I. J ., and Shin, S. G. (1997) Temporal changes
and reversibility of carbamylated hemoglobin in renal failure. Am.
J . Kidney Dis. 30, 36-40.
(51) Stim, J ., Shaykh, M., Anwar, F., Ansari, A., Arruda, J . A., and
Dunea, G. (1995) Factors determining hemoglobin carbamylation
in renal failure. Kidney Int. 48, 1605-1610.
(52) T o¨ rnqvist, M., and Kautiainen, A. (1993) Adducted proteins for
identification of endogenous electrophiles. Environ. Health Per-
spect. 99: 39-44.
(53) IPCS, International Program on Chemical Safety (1993) Bio-
markers and risk assessment, Environmental Health Criteria 155,
WHO, Geneva.
(
30) Sriramachari, S., Rao, G. J ., Sharma, V. K., J adhav, R. K., Saraf,
A. K., and Chandra, H. (1991) GC-NPD and GC-MS analysis of
preserved tissue of Bhopal gas disaster: evidence of methyl car-
bamoylation in post-mortem blood. Med. Sci. Law 31, 289-293.
31) Hazardous Substance Data Bank, HSDB (1998) Methyl isocyanate
and ethyl isocyanate. National Library of Medicine, (http://
toxnet.nlm.nih.gov/).
(
(
(
32) Kaushik, M. P., Sikder, A. K., and J aiswal, D. K. (1987) A
convenient laboratory synthesis of methyl isocyanate. Current Sci.
5
6, 1008.
33) Ramachandran, P. K., Gandhe, B. R., Venkateswaran, K. S.,
Kaushik, M. P., Vijayaraghavan, R., Agarwal, G. S., Gopalan, N.,
Suryanarayana, M. V., Shinde, S. K., and Sriramachari, S. (1988)
Gas chromatographic studies of the carbamoylation of Hb by
methyl isocyanate in rats and rabbits. J . Chromatogr. 426, 239-
2
47.
(
34) Venkateswaran, K. S., Raghuveeran, C. D., Gopalan, N., Agarwal,
G. S., Kaushik, M. P., and Vijayaraghavan, R. (1992) Monitoring
of haemoglobin-methyl isocyanate adduct by high-performance
liquid chromatography with diode array detector. Biochem. Int.
2
8, 745-750.
(
35) T o¨ rnqvist, M. (1990) Formation of reactive species that lead to
hemoglobin adducts during storage of blood samples. Carcino-
genesis 11, 51-54.
TX000230R