A new general pathway for synthesis of reference compounds of N-terminal valine-isocyanate adducts

Article abstract of DOI:10.1021/tx900278p

Adducts to Hb could be used as biomarkers to monitor exposure to isocyanates. Particularly useful is the measurement of carbamoylation of N-terminal valines in Hb, after detachment as hydantoins. The synthesis of references from the reactive isocyanates, especially diisocyanates, has been problematic due to side reactions and polymerization of the isocyanate starting material. A simpler, safer, and more general method for the synthesis of valine adducts of isocyanates has been developed using N-[(4-nitrophenyl)-carbamate] valine methylamide (NPCVMA) as the key precursor to adducts of various mono-and diisocyanates of interest. By reacting NPCVMA with a range of isocyanate-related amines, carbamoylated valines are formed without the use of the reactive isocyanates. The carbamoylated products synthesized here were cyclized with good yields of the formed hydantoins. The carbamoylated derivative from phenyl isocyanate also showed quantitative yield in a test with cyclization under the conditions used in blood. This new pathway for the preparation of N-carbamoylated model compounds overcomes the abovementioned problems in the synthesis and is a general and simplified approach, which could make such reference compounds of adducts to N-terminal valine from isocyanates accessible for biomonitoring purposes. The synthesized hydantoins corresponding to adducts from isocyanic acid, methyl isocyanate, phenyl isocyanate, and 2,6-toluene diisocyanate were characterized by LC-MS analysis. The background level of the hydantoin from isocyanic acid in human blood was analyzed with the LC-MS conditions developed.

Full text of DOI:10.1021/tx900278p

540
Chem. Res. Toxicol. 2010, 23, 540–546
A New General Pathway for Synthesis of Reference Compounds of
N-Terminal Valine-Isocyanate Adducts
Ronnie Davies,† Per Rydberg,*,† Emelie Westberg,† Hitesh V. Motwani,† Erik Johnstone,‡
and Margareta To¨rnqvist†
Department of Materials and EnVironmental Chemistry and Department of Organic Chemistry,
Stockholm UniVersity, SE-106 91 Stockholm, Sweden
ReceiVed August 11, 2009
Adducts to Hb could be used as biomarkers to monitor exposure to isocyanates. Particularly useful is
the measurement of carbamoylation of N-terminal valines in Hb, after detachment as hydantoins. The
synthesis of references from the reactive isocyanates, especially diisocyanates, has been problematic due
to side reactions and polymerization of the isocyanate starting material. A simpler, safer, and more general
method for the synthesis of valine adducts of isocyanates has been developed using N-[(4-nitrophenyl)-
carbamate]valine methylamide (NPCVMA) as the key precursor to adducts of various mono- and
diisocyanates of interest. By reacting NPCVMA with a range of isocyanate-related amines, carbamoylated
valines are formed without the use of the reactive isocyanates. The carbamoylated products synthesized
here were cyclized with good yields of the formed hydantoins. The carbamoylated derivative from phenyl
isocyanate also showed quantitative yield in a test with cyclization under the conditions used in blood.
This new pathway for the preparation of N-carbamoylated model compounds overcomes the above-
mentioned problems in the synthesis and is a general and simplified approach, which could make such
reference compounds of adducts to N-terminal valine from isocyanates accessible for biomonitoring
purposes. The synthesized hydantoins corresponding to adducts from isocyanic acid, methyl isocyanate,
phenyl isocyanate, and 2,6-toluene diisocyanate were characterized by LC-MS analysis. The background
level of the hydantoin from isocyanic acid in human blood was analyzed with the LC-MS conditions
developed.
Scheme 1. Common Commercial Isocyanates
Introduction
Organic mono-, di-, and polyisocyanates are common indus-
trial chemicals (Scheme 1). Monoisocyanates, such as phenyl
isocyanate (PIC) and methyl isocyanate (MIC), have been used
as synthetic intermediates in the chemical industry with MIC
in particular being used for the synthesis of various carbamate
pesticides. MIC is infamously known as the toxic agent in the
tragic chemical disaster in Bhopal, India, 1984, in which over
3000 persons died (1). Diisocyanates constitute the monomeric
subunits of polyurethanes (PURs). PURs have a wide range of
applications from flexible and rigid foams, elastomers, and
coatings to textiles (2, 3). PUR production constituted 12 million
metric tons worldwide in 2007 and is currently growing at the
rate of 5% each year due to the high demand of PUR products
(3). Diisocyanates used in the industry include toluene diiso-
cyanate (TDI; a mixture of the 2,4- and 2,6-isomers), 4,4′-
methylene diphenyl diisocyanate (MDI, both isomers and
homologues), and 1,6-hexamethylene diisocyanate (HDI)
(Scheme 1).
drolysis, and exposure to, for instance, TDI and MDI thus also
leads to internal exposure to the corresponding aromatic amines
(4, 5).
It was established after the Bhopal disaster that adducts to
N-terminal valine in hemoglobin (Hb) can be used for biomoni-
toring of isocyanate exposure (6, 7). Investigators were able to
measure MIC exposure through detachment of the carbamoy-
lated N-terminal valine of Hb by acid hydrolysis, to produce
3-methyl-5-isopropylhydantoin (MVH) (Scheme 2). Carbam-
oylation of other amino acids was, as well, detected in preserved
tissue samples (1, 7). Since then, carbamoylation of N termini
in Hb has also been used to identify and quantify internal
exposure to dimethylformamide (which metabolizes to MIC)
(8, 9) and TDI in humans (10) and to MDI, so far only in rats
(11).
The reactivity of mono- and diisocyanates, which makes them
useful for many purposes, stems from the highly electrophilic
carbon atom in the isocyanate group. In vivo, this reactivity
can lead to carbamoylation reactions with proteins, etc. Fur-
thermore, the corresponding amines are formed through hy-
* To whom correspondence should be addressed. Tel: +46-8-163769.
Fax: +46-8-163979. E-mail: per.rydberg@mk.su.se.
† Department of Materials and Environmental Chemistry.
‡ Department of Organic Chemistry.
Besides the occupational exposure to isocyanates, there is a
general endogenous exposure to isocyanic acid, which is
10.1021/tx900278p  2010 American Chemical Society
Published on Web 01/19/2010

N-Terminal Valine-Isocyanate Adducts
Chem. Res. Toxicol., Vol. 23, No. 3, 2010 541
Scheme 2. Carbamoylation and Hydrolysis of N-Terminal Valine
Scheme 3. Synthetically Produced Carbamoylation Using Amine and a Carbamate Group
produced in the spontaneous breakdown of urea (12). In patients
methylamide reacted with 4-nitrophenylchloroformate (see
Scheme 3). The approach to use such precursors is analogous
by the chemistry used in solid and liquid phase chemistry to
synthesize ureas from carbamates and amines (18-20). This
strategy could be especially useful for the synthesis of carbam-
oylvalines, as the “activated precursor” only reacts once. This
means that the risk for polymerization side reactions is
eliminated. Another great benefit will be that chemical libraries
can be build from a single activated carbamate precursor.
The reference substances of PIC synthesized in this work were
used to test the procedure to form hydantoin in human blood.
Conditions were established for LC-MS/MS analysis of the
hydantoins and were used for the analysis of carbamoylated
N-terminal valine in human Hb from the naturally occurring
isocyanic acid.
with kidney failure, urea levels have been monitored through
increased carbamoylation of N-terminal valine in Hb (13). The
measurement of carbamoylation of N termini in Hb was already
pioneered in 1973 by Manning et al. in the treatment of sickle
cell disease by cyanate (14).
For the measurement of hydantoins formed from N-carbam-
oylated terminal valine in Hb as a biomarker of internal exposure
to isocyanates, it is useful to have access to reference compounds
that mimic the adduct to N termini in Hb. The references should
be useful in studies of yield in the acid-catalyzed cyclization
reaction and of the extraction of the hydantoins, detached from
Hb or from isolated globin (15). An established pathway to
synthesize N-carbamoylated reference compounds and corre-
sponding internal standards for analysis of adducts from
isocyanates is to react a model compound of N-terminal valine
with the selected isocyanates (8, 10, 11, 16, 17). The formed
carbamoylated compound is then cyclized to produce the
hydantoin (Scheme 2). However, this route using isocyanates
as starting materials is problematic due to side reactions of the
mono- and diisocyanate with water, producing symmetrical urea
compounds. When using diisocyanates, polymerization side
reactions are a major problem, which lead to very low yields
of the desired products (our unpublished work). Furthermore,
the use of volatile and toxic isocyanates in the preparation of
reference compounds should be avoided. A strategy to overcome
such polymerization side reactions when preparing reference
standards from 2,4-TDI with single amino acids was introduced
by Sabbioni et al. in 2001 (10). Instead of using TDI, one of
the isocyanate group was masked by a nitro group. After the
carbamoylation reaction, this group was reduced to an amine.
This was an innovative solution, and improved yields could be
obtained as compared to the earlier methods. However, a major
drawback with this approach is that it includes several synthetic
steps with rather specific starting materials for the preparation
of the analytes of different isocyanates. In summary, there is a
need for a new simple and general synthetic route to prepare
N-carbamoylated reference compounds.
Materials and Methods
Caution: Chloroformates and aromatic amines are toxic and
were handled with protectiVe clothing in an efficient hood.
Chemicals. Ethyldi-isopropylamine was purchased from Aldrich
(Steinheim, Germany). Valine amide hydrochloride (VA-HCl) was
obtained from Novabiochem AG (La¨ufelflingen, Germany). Aniline
and ammoniumsulphate were obtained from Merck (Darmstadt,
Germany). NaOH was obtained from EKA Chemicals AB (Bohus,
Sweden). Methylamine (40%), L-valine methyl ester hydrochloride
(L-VME-HCl) (99%), and 4-nitrophenyl chloroformate (97%) were
obtained from Acros (Geel, Belgium). KMnO4 and 2,6-diamino-
toluene (97%) and isopropylhydantoin (VH) were obtained from
Aldrich. The hemolysate used for study of yield in cyclization
reaction was obtained from commercial human blood from Karo-
linska Hospital (Stockholm, Sweden). All other chemicals and
solvents used were of analytical grade.
1
Analytical Methods. All H NMR and 13C NMR (300 MHz)
were recorded on a Varian Mercury 400 spectrometer in CD3OD,
CDCl3, DMSO-d6, or (CD3)2CO solutions (50 mg/mL) at 25 °C
with TMS as the internal standard. 1H NMR and 13C NMR spectra
of compounds 2, 3, and 5-10 are also available as Supporting
Information.
Melting points were determined on a Bu¨chi 353 apparatus. TLC
was performed using silica gel 60 F-254 from Merck, and Rf values
were obtained by the application of 10 µg aliquots of the reaction
mixtures. Spots were visualized with UV (254 nm), ninhydrin, and
KMnO4.
The basic idea to attack the problem with low yields, etc.,
was to “invert” the chemistry for synthesis of N-carbamoylated
reference compounds. Instead of using electrophilically reactive
isocyanates, nucleophilic aliphatic or aromatic amines/diamines
were reacted with an electrophilically activated carbamate
precursor. The studied precursors were synthesized from valine
The LC-MS/MS system consisted of a Shimadzu Prominence
LC-system coupled to an API 3200 Q trap instrument (Applied
Biosystems). The mass spectrometer was operated using an

542 Chem. Res. Toxicol., Vol. 23, No. 3, 2010
DaVies et al.
electrospray ionization source in the positive ion mode (ESI+).
Acquisition and data processing were done with the Analyst
software version 1.5 (Applied Biosystems). The separation was
performed with an C18 column (5 µm, 1.0 mm × 150 mm, Ace).
The mobile phase system consisted of A, 0.1% formic acid (FA)
in water:acetonitrile (95:5, v/v), and B, 0.1% FA in water:ACN
(5:95,v/v). The gradient was 0% B for 1 min, followed by a linear
increase to 50% B in 1 min, and 50-100% B in 5 min followed
by isocratic 100% B in 3 min, with a flow rate of 75 µL/min. The
MS/MS was operated in product ion scan (PIS) mode and multiple
reaction monitoring (MRM) mode. The instrumental settings used
for the MRM were as follows: ion source temperature, 500 °C;
declustering potential, 30 V; collision cell exit potential, 25 V; ion
spray voltage, 5000 V; curtain gas (N2), 20; collision gas (N2), 5;
ion source gas (N2), 55; and turbo gas (N2), 70 (latter four are
arbitrary units from the Analyst software). Varied collision energy
(15-25 V), entrance potential (5-10 V), collision cell entrance
potential (7-15), and collision cell exit potential (1-5) were applied
for the different transitions used in MRM.
adjusted to 9 by dropwise addition of 1 M NaOH (0.1 mL).
The product and aniline were extracted with EtOAc (2 × 30
mL), and the organic phase was washed with 0.1 M NaOH (5
× 30 mL). Aniline was extracted from the organic phase by
vigorous shaking with 0.1 M HCl (5 × 30 mL). After the
organic solvent was evaporated, the product was crystallized
from EtOH/ACN (1:1, 10 mL) at 60 °C to yield 3 (42 mg, 0.17
mmol, 25%) as white crystals. Analysis: mp 275-276 °C, Rf,
1
0.3, EtOAc/DCM (1:1). H NMR (DMSO-d6, 25 °C): δ 0.81,
0.84 [2d, 6H, J ) 6.8 Hz, CH3(γ,γ′)], 1.90 [m, 1H, J ) 6.4,
5.9 Hz, CH(ꢀ)], 2.58 [d, 3H, J ) 4.4 Hz, CONHCH3], 4.01,
4.37 [dd, 1H, J ) 6.0 Hz, CH(R)], 6.31 [d, 1H, J ) 8.8 Hz,
CONHCH], 6.86 [t, 1H, J ) 7.2 Hz, para-arom], 7.19 [2t, 2H,
J ) 7.2 Hz, meta-arom], 7.33 [2t, 2H, J ) 7.2 Hz, ortho-arom],
7.97 [d, 1H, J ) 4.0 Hz, CONHCH3], 8.63 [s, 1H, arom-
NHCO]. 13C NMR (DMSO-d6, 25 °C): δ 18.24, 19.71 [CH3
(γ,γ′)], 25.85 [NH-CH3], 31.61 [CH(ꢀ)], 58.05 [CH(R)], 117.86,
121.51, 121.51, 129.15, 129.15, 140.9 [arom.], 155.36
[CO(NH)2], 172.47 [CONH2]. LC-MS (ESI+): m/z 250 {[M +
H]+, 10%}, 219 {[M - 31 (-CH3NH2)]+, 20%}, 191 {[M -
59 (-CONHCH3)]+, 20%}, 157 {[M - 92 (-C6H5NH)]+, 5%},
131 {[M - 119 (-C6H5NHCO)]+, 100%}.
Initially, a less optimized LC-MS system, without column
separation, was used for the characterization of synthezised
compounds and for studies of yield in the cyclization reaction.
Samples from the cyclization studies were also analyzed with the
optimized system as described above.
Methylcarbamoyl-valinmethylamide (MCVMA) (4). Dry
methylamine hydrochloride (MeNH3Cl) was prepared from 40%
aqueous methylamine (40% w/v, 50 mL) with 3 M HCl (300
mL) followed by evaporation. This salt (10 g) was alkalized
dropwise with 12 M KOH (15 mL) at 45 °C. The generated
MeNH2(g) was dried (MgSO4) and bubbled into a connected
flask containing NPCVMA (200 mg, 0.68 mmol) dissolved in
EtOH (10 mL), and nitrogen was used as the carrier gas. The
reaction was followed by TLC with EtOAc/DCM/acetic acid
(HOAc) (3:1:0.1), NPCVMA (Rf 0.5), the deprotonated p-
nitrophenolate (Rf 0.9), and pH paper for the generation of
MeNH2(g). After 6 h the reaction mixture was sealed and left
to stand overnight after which all NPCVMA was depleted. The
solvent was evaporated, and the residue was dissolved in 0.1%
HOAc (50 mL) and extracted with EtOAc (2 × 30 mL) to
remove the p-nitrophenol side product. The aqueous phase was
evaporated, and the product was crystallized from EtOH/ACN
1:1 (4 mL) to yield 4 (26 mg, 0.14 mmol, 21%) as white crystals.
Analysis: mp 265-269 °C. LC-MS (ESI+): m/z 188 {[M +
H]+, 10%}, 157 {[M - 31 (-CH3NH2)]+, 10%}, 131 {[M -
57 + H]+, 100%}.
Synthesis
Valine methyl amide (VMA) (1). L-VME-HCl (57 mmol,
9.5 g) was added to methylamine (40% w/v, 150 mL) while
stirring in an ice bath. The solution was left for 2 days at 22
°C. The solvent was evaporated, and dry tetrahydrofuran (THF)
(60 mL) was added to the solid residue. The white suspension
was ultrasonicated for 30 min to obtain dissolution. After the
solution was stored at +4 °C overnight, valine and methylamine
HCl were filtered off, and the eluate was evaporated to yield
the desired product 1 (4.4 g, 34 mmol, 60%) as an oil. Analysis:
1H NMR (CDCl3, 25 °C): δ 0.82, 0.99 [2d, 6H, J ) 6.8 Hz,
CH3(γ,γ′)], 1.4 [s, 2H, NH2], 2.13 [m, 1H J ) 3.7 Hz, CH(ꢀ)],
2.82, 2.83 [dd, 3H, J ) 5.2 Hz, NH-CH3], 3.24 [d, 1 H, J )
3.7 Hz, CH(R)].
N-[(4-Nitrophenyl)carbamate]valinmethylamide (NPCVMA)
(2). To a solution of 4-nitrophenyl chloroformate (0.57 g, 2.8
mmol) in dry THF (15 mL), VMA (0.74 g, 5.7 mmol) in THF
(15 mL) was dropped over a period of 10 min while stirring on
an ice bath under inert atmosphere (N2). The mixture was slowly
heated to ambient temperature while following the progress by
TLC [ethyl acetate/dichloromethane (EtOAc/DCM), 3:1]. After
2.5 h, the solvent was evaporated, and the solid product was
crystallized from ethanol (EtOH)/ACN (1:5, 75 mL) to yield 2
(0.59 g, 2.0 mmol, 71%) as white crystals. Analysis: mp
182-184 °C, Rf, TLC 0.4, EtOAc/DCM (3:1). 1H NMR (CDCl3,
25 °C): δ 1.0, 1.21 [2d, 6H, J ) 6.8 Hz, CH3(γ,γ′)], 2.13 [m,
1H J ) 6.8 Hz, CH(ꢀ)], 2.87 [d, 1H, J ) 4.8 Hz, CONHCH3],
3.94, 3.97 [dd, 1H, J ) 6.4 Hz, CH (R)], 5.79 [d 1H, J ) 8.8,
CONHCH], 5.89 [d, 1H, CONH CH3], 7.31 [d, 2H, J ) 6.8
Hz, arom], 8.24 [d, 2H J ) 6.8 Hz, arom.]. 13C NMR (CDCl3,
50 °C): δ 18.30, 19.37 [CH3(γ,γ′)], 26.50 [CONH-CH3], 31.75
[CH(ꢀ)], 61.15 [CH(R)], 115.9 122.1, 125.3, 126.4, 145.3, 153.6
[arom.], 156.05 [OCONH], 171.45 [CONHCH3].
Carbamoyl-valinmethylamide (CVMA) (5). Ammonium
chloride (10 g) was alkalized dropwise with 12 M KOH (15
mL) at 45 °C as described above to generate dry NH3(g), which
was bubbled into a solution of NPCVMA (200 mg, 0.68 mmol)
in EtOH (10 mL). The reaction conditions and purification
procedure were exactly as described above. The solid product
was crystallized from EtOH/ACN 1:1 (4 mL) to yield 5 (20
mg, 0.12 mmol, 17%) as white crystals. Analysis: mp 235-238
1
°C. H NMR (DMSO-d6, 25 °C): δ 0.77-0.84 [4d, 6H, J )
7.2 Hz, CH3(γ,γ′)], 1.87 [m, 1H J ) 6.4, Hz, CH(ꢀ)], 2.56 [d,
3H, J ) 4.4 Hz, CONHCH3], 3.91, 3.93, [dd, 1H, J ) 6.0 Hz,
CH(R)], 6.26, 6.33 [dd, 1H J ) 8.8 Hz, NHCONH2], 6.93, 8.11
[dd, 1H J ) 8.8 Hz, NHCONH2], 7.79, 7.84 [dd, 1H J ) 4.4
Hz, CONHCH3]. 13C NMR (DMSO-d6, 25 °C): δ 18.15, 19.70
[2 × CH3(γ,γ′)], 25.81 [NH-CH3], 31.54 [CH(ꢀ)], 58.41
[CH(R)], 158.01 [CO(NH)2], 172.87 [CONH]. LC-MS (ESI+):
m/z 174 [M + H]+.
Phenylcarbamoyl-valinmethylamide (PCVMA) (3).
NPCVMA (0.68 mmol, 0.20 g) and freshly distilled aniline (4
mL) were mixed together and stirred at 60 °C for 18 h. The
reaction was followed by TLC with EtOAc/DCM (1:1) for
visualizing the product with UV (Rf 0.3), NPCVMA (Rf 0.4),
aniline (Rf 0.8), and the byproduct p-nitrophenol (Rf 0.9). The
reaction was stopped when all NPCVMA was depleted. Water
(30 mL) was added to the reaction mixture, and the pH was
N-[(3-Amino-4-methylphenyl)carbamoyl]valinmethyl-
amide (ATCVMA) (6). NPCVMA (0.21 g, 0.71 mmol) in dry
THF (4 mL) was added to a solution of 2,6-diaminotoluene (85
mg, 0.70 mmol) (freshly purified in a silica gel column, mobile

N-Terminal Valine-Isocyanate Adducts
Chem. Res. Toxicol., Vol. 23, No. 3, 2010 543
Scheme 4. Reference Compounds Synthesized by the New Method from an Activated Carbamate Precursor
phase ACN, Rf 0.8) in dry THF (0.5 mL). The reaction was
performed at 60 °C for 4 days with magnetic stirring, and the
progress was followed by TLC with EtOAc/DCM (1:1, v/v, Rf
values were as follows: NPCVMA, 0.4; p-nitrophenol, 0.9; and
2,6-diaminotoluene, 0.3). The desired product precipitated as
gray/white crystals and was purified by evaporation of the
solvent and crystallization from ACN/EtOH 5:1 (100 mL) to
yield 6 (114 mg, 0.41 mmol, 58%) as white crystals. Analysis:
mmol, 65%) from compound 3. MVH (8) was crystallized from
CHCl3/pentane 1:4 (3 mL) as white needle crystals (mp
121-122 °C) to yield 18 mg (0.12 mmol, 52%) from compound
4, and VH (9) was crystallized from CHCl3/pentane 1:2 (6 mL)
to yield 10 mg (0.070 mmol, 62%) from compound 5 as white
crystals (mp 137-140 °C). ATVH (10) was purified on a silica
gel column eluted with EtOAc to yield 11 mg (0.044 mmol,
32%) from compound 6 as a white residue. Analysis: Compound
7: 1H NMR (CDCl3, 25 °C): δ 1.00, 1.08 [2d, 6H, J ) 6.9, 7.2,
CH3(γ,γ′)], 2.32 [m, 1H, J ) 7.2, 3.6 Hz, CH(ꢀ)], 4.05, 4.06
[dd, 1H, J ) 1.6, 1.2 Hz, CH(R)], 6.8 [s, 1H, NH], 7.37 [2t, 1
H, J ) 7.4 Hz, para-arom], 7.38 [2t, 2H, J ) 7.4 Hz, meta-
arom], 7.40 [d, 2H, J ) 7.4 Hz, ortho-arom]. 13C NMR (CDCl3,
25 °C): δ 16.03, 18.66 [2 × CH3(γ,γ′)], 30.63 [CH(ꢀ)], 62.19
[CH(R)], 126.22, 126.22, 128.29, 128.29, 129.13, 129.13, 131.45
1
mp 279-280 °C, Rf, TLC 0.5, EtOAc/DCM (1:1). H NMR
(DMSO-d6, 25 °C): δ 0.84, 0.86 [2 × d 6H, J ) 8.2 Hz,
CH3(γ,γ′)], 1.89 [m, 1H, CH(ꢀ)], 1.89 [s, 3H, arom-CH3], 2.60
[d, 3H, J ) 6.4, CONH-CH3], 4.03 [dd, 1H, J ) 3.6 Hz, CH(R)],
4.78 [ s, 2H, arom-NH2], 6.33 [d, 1H, J ) 10.8 Hz arom], 6.57
[d, 1H, CONHCH] 6.75 [t, 1H, J ) 10.8 Hz, arom.], 6.90 [d,
1H J ) 10.8, Hz, arom], 7.74 [s, 1H, arom-NH-CO], 7.95 [d,
1H, J ) 6.0, CONH CH3]. 13C NMR (DMSO-d6, 25 °C): δ
11.32 [arom-CH3], 17.97, 19.28 [2 × CH3(γ,γ′)], 25.40 [NH-
CH3], 31.28 [CH(ꢀ)], 57.85 [CH (R)], 109.53, 110.98, 112.29,
125.45, 138.04, 146.84 [arom.], 155.43 [CO(NH)2], 172.25
[CO]. LC-MS (ESI+): m/z 279 [M + H]+.
1
[arom.], 157.3 [NHCONH],172.5 [CO-CH]. Compound 8: H
NMR (CD3OD, 25 °C): δ 0.89, 1.05 [2d, 6H, J ) 6.8,
CH3(γ,γ′)], 2.17 [m, 1H, J ) 3.6,Hz, CH(ꢀ)], 2.95 [s, 1H, CH3,
NCH3], 3.99 [d, 1H, CH, J ) 3.6 Hz, CH(R)]. 13C NMR
(CD3OD, 25 °C): δ 14.89, 17.53 [2 × CH3(γ,γ′)], 22.97 [NCH3],
30.03 [CH(ꢀ)], 62.29 [CH(R)], 158.55 [CO-CH], 174.71 [NH-
Hydrolysis and Cyclization of Carbamoylated Com-
pounds To Form Hydantoins: 3-Phenyl-5-isopropylhydan-
toin (PVH) (7), MVH (8), 5-Isopropylhydantoin (VH) (9),
3-(3-Amino-2-methyl)phenyl-5-isopropylhydantoin (ATVH)
(10). The urea compounds were cyclized according to Mraz with
minor changes (16). The respective compounds 3 (40 mg, 0.17
mmol), 4 (40 mg, 0.23 mmol), 5 (20 mg, 0.12 mmol), and 6
(40 mg, 0.14 mmol) were separately dissolved in concentrated
HOAc and concentrated HCl (1:2, v/v, 3 mL) and heated at 70
°C for 18 h (see Scheme 4). The reactions were followed by
TLC with EtOAc/toluene (1:1) using UV for aromatic hydan-
toins and KMnO4 for aliphatic compounds. After the mixtures
were cooled, ammonium sulfate (1.5 g) and 10 M NaOH were
slowly added to obtain a pH of 7 (approximately 5 mL). The
solutions were extracted with EtOAc (4 mL), the organic phase
was evaporated, and the solid remains were crystallized as
follows: PVH (7) from acetone/heptane 1:4 (3 mL) as white
needle crystals (mp 128-130 °C) with a yield of 25 mg (0.11
1
CONH]. Compound 9: H NMR [(CD3)2CO, 25 °C]: δ 0.93,
1.05 [2d, 6H, J ) 6.8 Hz, CH3(γ,γ′)], 2.15 [m, 1H, J ) 3.6 Hz,
CH(ꢀ)], 4.03, 4.03 [dd, 1H, CH, J ) 3.6 Hz, CH(R)], 7.11 [s,
1H, CONH], 9.60 [s, 1H, (CO)2NH]. 13C NMR [(CD3)2CO, 25
°C]: δ 15.49, 17.98 [2 × CH3(γ,γ′)], 30.05 [CH3], 30.1 [CH(ꢀ)],
63.28 [CH(R)], 157.50 [NHCONH], 174.57 [CO-CH]. Com-
pound 10: 1H NMR (DMSO-d6, 25 °C): δ 1.03, 1.05, 1.08, 1.10
[2 × dd, 6H, J ) 6.8, 7.7 Hz, CH3(γ,γ′)], 1.96, 1.97 [2s, 3H,
arom-CH3], 2.33 [m, 1H, J ) 3.6 Hz, CH(ꢀ)], 4.09, 4.10, 4.11,
4.12 [2 × dd, 1H, J ) 3.6 Hz, CH(R)], 6.22 [d, 1H, J ) 13.6
Hz, CONH], 6.53, 6.61 [2 × dd, 1H, J ) 6.8 Hz, para-arom],
6.74 [2d, 1H, J ) 8.0 Hz, ortho-arom.], 7.10 [2 × dt, 1H, J )
4.0 Hz, meta-arom]. 13C NMR (DMSO-d6): δ 12.15, 12.16
[arom-CH3], 15.96, 16.45, 18.67, 18.93 [2 × CH3(γ,γ′)], 30.23,
30,64 [CH(ꢀ)], 62.33, 62,76 [CH(R)], 116.07,116.10, 118.17,
118.47, 120.52, 120.76, 127.06, 127.07, 130.77, 130.80 [arom],
145.96, 146.01 [arom-C-NH2], 157.24, 157.28 [NHCONH],

544 Chem. Res. Toxicol., Vol. 23, No. 3, 2010
DaVies et al.
Table 1. Fragmentation of the Studied Hydantoins Using LC-MS (ESI+) with Collision Energy 15-20 V
analyte
PVH (7)
MVH (8)
VH (9)
ATVH (10)
precursor
PIC
MIC
157 [M + H]+, 100%
isocyanic acid
143 [M + H]+, 20%
2,6-TDI
248 [M + H]+, 33%
LC-MS/MS 219 [M + H]+, 100%
(ESI+, PIS)
72 [M + H - 147, -adduct]+, 99% 72 [M + H - 85, -adduct]+, 57% 72 [M + H - 71, -adduct]+, 100% 149 [M + H - 99, -valine]+, 100%
120 [M + H - 99, -valine]+, 32% 129 [M + H - 28, -CO]+, 19% 115 [M + H - 28, -CO]+, 37%
191 [M + H - 28, -CO]+, 23%
72 [M + H - 176, -adduct]+, 25%
205 [M + H - 43, -propyl]+, 17%
176 [M + H - 43, -propyl]+, 10%
retention time 9.00
(min)
8.41
8.02
8.40 and 8.48 (double peak)
172.51, 172.61 [CO-CH]. Because of steric hindrance between
the methyl group in the toluene ring and the 2-carbonyl group
in the hydantoin system, compound 10 occurred in two
of VMA was a modification of the earlier protocol (23) with
exclusion of the time-consuming ion exchange step. The
modified synthesis allows the methylamine hydrochloride salt
and the hydrolyzed L-VME-HCl (valine) to fully precipitate at
+4 °C overnight to give sufficient yields (60%) of neutralized
VMA after the precipitate is filtered off.
1
conformers. This was observed as all signals measured by H
NMR and 13C NMR were in duplicates with only slightly
different chemical shifts.
The next step to improve the procedure for the synthesis of
the desired compounds was to synthesize a key precursor that
can be utilized for the synthesis of a wide range of valine adducts
of isocyanates by a reaction with amines instead of isocyanate
precursors. This was achieved by synthesis of an activated valyl-
carbamate, which was formed from carbamoylating VMA with
phenyl chloroformate derivatives (data not shown). Ultimately
chosen was 4-nitrophenyl chloroformate to form NPCVMA (2),
as the nitro group enhances the reactivity of the precursor. This
compound (2) was synthesized with 70% yield.
In the synthesis of PCVMA (3), freshly distilled aniline was
used in excess as the reactant and solvent, giving a faster rate
of reaction and minimizing side reactions. The formation of 7
and disappearance of 2 could be followed easily by TLC using
UV and by the formation of p-nitrophenol with its deep yellow
color. The desired product could easily be purified from side
products by extractions. A yield of 25% was obtained.
In the synthesis of MCVMA (4) from 2, various synthetic
routes were tested. The best result was obtained by reacting 2
with anhydrous methylamine dissolved in dry EtOH instead of
a 40% aqueous solution. By this route, the hydrolysis of 2 could
be avoided. The solution turned yellow within an hour showing
that p-nitrophenol was released from 2. Methylamine was
periodically added during the next 6 h to saturate the solution,
which finally was left to stand overnight, and 2 was depleted.
Although a few attempts were made to improve this reaction,
the yield did not exceed 30%.
Test of Acid-Catalyzed Conversion of PCVMA (3) to
PVH (7) in Blood Samples. A dilution series of PCVMA (3)
(n ) 5) in hemolysate was prepared by adding 100 µL of
90-1000 µM PCVMA in water to samples of hemolysate (0.5
mL) (n ) 5). The samples were then hydrolyzed and purified
according to Kwan et al. (21) with minor modifications.
Concentrated HCl:HOAc (1:1, v/v, 2 mL) was added to each
sample, which was heated for 18 h at 90 °C. After the tubes
were cooled, the pH was adjusted to 4 by 10 M NaOH (∼2
mL) (added dropwise at 0 °C). The samples were then extracted
with EtOAc (4 mL) by rocking the tubes for 10 min. The organic
phases were isolated after centrifugation (2000 rpm) and then
washed with 1 M NaHCO3 (2 mL) followed by H2O (2 mL).
The organic phases were transferred to new vessels, evaporated,
and redissolved in methanol (2 mL per sample). Calibration
samples (n ) 5) were prepared by addition of the corresponding
concentrations of the hydantoin PVH (7) to the washed extract
from blood samples hydrolyzed as described above. In addition,
a reproducibility test was done with the addition of 100 µL of
500 µM PCVMA (n ) 5) to samples of hemolysate, which were
treated as described above, to give a final concentration of 25
µM. The samples in these tests were analyzed by LC-MS/MS
by double injections for each sample.
As a further test of the method, the background level of
VH formed from isocyanic acid was analyzed in control Hb
samples. These analyses were done with the improved LC-
MS/MS conditions with separation and with more careful
sample preparation, with preisolation of globin (according
to Mraz et al., ref. 15). Globin was isolated by precipitation
from hemolysate (22) and worked up after hydrolysis for 1 h
at 100 °C according to Mraz et al. (15) with a few
modifications. Globin (50 mg) was dissolved in concentrated
HCl/HOAc (2:1, v/v, 2.5 mL) and heated for 1 h at 100 °C
in a Pyrex tube with screw cap (8 mL). Ammonium sulfate
(1.5 g) was then added. The mixture was neutralized by
addition of 10 M NaOH (pH 7.0-7.5) (approximately 3 mL).
The solution was extracted with EtOAc (4 mL), vortex-mixed,
and then centrifuged at 3000 rpm. The organic phase was
isolated and evaporated at 60 °C under a gentle stream of
nitrogen. The obtained residue was dissolved, and 1/5 was
transferred to a vial H2O:ACN (9:1, 100 µL) and analyzed
(5 µL, corresponding to 0.5 mg globin) by LC-MS/MS.
In the synthesis of the N-[(3-amino-2-methylphenyl)carbam-
oyl]valinmethylamide (ATCVMA) (6), 2,6-diaminotoluene was
reacted with 2 in THF at 60 °C for 4 days. This synthesis gave
a higher yield (57%) with little side reactions such as cross-
linking. It was found to be imperative to purify the 2,6-
diaminotoluene directly before use. In the synthesis of monoiso-
Results and Discussion
Synthesis. The use of VMA (1) as a model compound for
N-terminal valines in Hb is the more suitable choice than valine
or valine amide. The amide bond in VMA is expected to be
quite similar to a peptide bond during hydrolysis. The synthesis
Figure 1. Calibration line with different concentrations of PVH (upper
line). Cyclization to PVH with different concentrations of PCVMA
(dashed line).

N-Terminal Valine-Isocyanate Adducts
Chem. Res. Toxicol., Vol. 23, No. 3, 2010 545
A few control blood samples, without spiking with the
standard VH, were analyzed with improved conditions. The LC-
MS/MS conditions were optimized with regard to the separation
and the measured ions, and the blood was processed according
to Mraz et al. (15), using hydrolysis of isolated globin. Reference
standards (VH and PVH) were analyzed (see Figure 2). A
background level of VH was present as expected, and PVH was
not observed in the blood samples (see Figure 2). The observed
level of VH was in the range of earlier observed levels in human
Hb of about 200 nmol/g Hb (13). In the LC-MS analysis of the
adduct of VH detached from globin, using an injected volume
to 5 µL (corresponding to 0.5 mg of globin), a s/n ∼ 1000 was
obtained for the VH peak.
Future Applications of the New Synthesis Route. The aim
of this work was to develop a new and versatile synthesis
pathway for reference substances for adducts from isocyanates
to valine, useful for quantification of exposure to isocyanates.
Workup of human blood samples spiked with PCVMA gave
high yields of the formed hydantoin, PVH. The background of
VH formed from urea could be observed in a few samples of
human blood analyzed by LC-MS/MS. The developed synthetic
methods presented here could be used as a general route for
the synthesis of reference substances for accurate biomonitoring
of isocyanates and urea. Although not tested here, the more
important commercial diisocyanates, 4,4′-MDI, 2,4-TDI and
isophorone diisocyanate (IPDI), could also be synthesized from
their corresponding amines reacted with the activated precursor.
This approach to activate the nitrogen in valine methylamide
by 4-nitrophenyl chloroformate could relatively easily be
extended to include peptides. These “activated” peptide precur-
sors could then be reacted with selected amines for building
chemical libraries of N-carbamoylated reference peptide com-
pounds. Such reference compounds would be optimal for the
measurement of Hb adducts as the yields of detached hydantoin
might vary depending on the nature of the adduct. Pure
N-carbamoylated reference peptides, with and without isotope
substitution at the N-terminal amino acid, would possibly even
be a more ultimate choice for synthetic references and internal
standards.
Figure 2. LC-MS/MS (ESI+) MRM chromatograms of standard
hydantoins (VH and PVH) and analysis of hydrolyzed control blood
in which only VH was formed.
As the focus of this work was to develop a new synthetic
route, less effort was done to optimize the analytical procedure/
conditions. The high adduct levels of 102 nmol/g of VH (13) or
MVH (9) earlier observed in persons with renal failure or
dimethylformamide exposure, respectively, would be measurable
with the current procedure. This was also shown in our analysis,
with a high s/n ratio, of background VH in human blood. The
analysis of adducts from diisocyanates in exposed persons would
most probably need further development of the analysis. In
workers exposed to TDI, a corresponding adduct level of 70
pmol/g has been observed (10). This would probably be below
the LOQ of the method used here.
cyanate reference standards, a longer time in the reaction
chamber could increase the yields, as the use of UV in the
disappearance of NPCVMA could not be an adequate indicator.
Acid-catalyzed cyclization of the N-carbamoylated standards
(3-6) to form PVH (7), MVH (8), VH (9), and ATVH (10)
was performed in HOAc/HCl (1:2) and gave yields between
30 and 75% after recrystallization. The ring-closed hydantoins
1
were characterized with LC-MS, H NMR, or 13C NMR. Data
from LC-MS/MS fragmentation for the hydantoins are sum-
marized in Table 1. The hydantoins were also characterized by
GC-MS/MS (data not shown).
Analysis. In samples of hemolysate that were hydrolyzed and
extracted, the yield of PVH formed through cyclization of added
PCVMA was studied. The obtained levels corresponded to a
quantitative yield of PVH (see Figure 1). Calibration lines of
the PVH added to extracts from blood gave r2 values of 0.93 in
the analysis with the initially used conditions for LC-MS
analysis. In a reproducibility test of the method by Kwan et al.
(21) for hydrolysis and cyclization, a quantitative yield was
obtained, with a relative standard deviation of 17% (n ) 2 ×
5). Considering that this was a pilot study of the reproducibility
in the analysis of carbamoyl adducts as hydantoins, using a
preliminary analytical method, it was found acceptable with the
potential of improvements.
Acknowledgment. Economical support from the Sw. Council
for Working Life and Social Research (FAS), the Swedish
Research Council Formas, and the Swedish Cancer and Allergy
Foundation are acknowledged. Maria Anthanasiadou, Ph.D.,
Hans von Stedingk, M.Sc., and Carl Axel Wachtmeister, Prof.
Em., are gratefully acknowledged for excellent viewpoints.
Bjo¨rn Åkermark, Prof. Em., is acknowledged for support with
NMR analyses.
Supporting Information Available: 1H NMR and 13C NMR
spectra of compounds 2, 3, and 5-10. This material is available
free of charge via the Internet at http://pubs.acs.org.

546 Chem. Res. Toxicol., Vol. 23, No. 3, 2010
DaVies et al.
(12) Kraus, L. M., and Kraus, A. P., Jr. (2001) Carbamoylation of amino
acids and proteins in uremia. Kidney Int., Suppl. 78, 102–107.
(13) Wynckel, A., Randoux, C., Millart, H., Desroches, C., Gillery, P.,
Canivet, E., and Chanard, J. (2000) Kinetics of carbamylated haemo-
globin in acute renal failure. Nephrol., Dial., Transplant. 15, 1183–
1188.
(14) Manning, J. M., Lee, C. K., Cerami, A., and Gillette, P. N. (1973)
Gas chromatographic determination of the carbamylation of hemo-
globin S by cyanate. J. Lab. Clin. Med. 81, 941–945.
(15) Mraz, J., Duskova, S., Galova, E., Nohova, H., Krausova, P., Linhart,
I., and Simek, P. (2002) Improved gas chromatographic-mass spec-
trometric determination of the N-methylcarbamoyl adduct at the
N-terminal valine of globin, a metabolic product of the solvent N,N-
dimethylformamide. J.Chromatogr., B 778, 357–365.
(16) Mraz, J., and Bouskova, S. (1999) 2,4-Toluenediisocyanate and
hexamethylene-diisocyanate adducts with blood proteins: Assessment
of reactivity of amino acid residues in vitro. Chem.-Biol. Interact. 117,
173–186.
(17) Sabbioni, G., Lamb, J. H., Farmer, P. B., and Sepai, O. (1997)
Reactions of 4-methylphenyl isocyanate with amino acids. Biomarkers
2, 223–232.
(18) Izdebski, J., and Pawlak, D. (1989) A new convenient method for the
synthesis of symmetrical and unsymmetrical N,N′-disubstituted ureas.
Synthesis 423–425.
(19) Hutchins, S. M., and Chapman, K. T. (1994) A general method for
the solid phase synthesis of ureas. Tetrahedron Lett. 35, 4055–4058.
(20) Xiao, X., Ngu, K., Chao, C., and Patel, D. V. (1997) Selective solid
phase synthesis of ureas and hydantoins from common phenyl
carbamate intermediates. J. Org. Chem. 62, 6968–6973.
(21) Kwan, J. T. C., Carr, E. C., Bending, M. R., and Barron, J. L. (1990)
Determination of carbamylated hemoglobin by high-performance liquid
chromatography. Clin. Chem. 36, 607–610.
(22) Mowrer, J., To¨rnqvist, M., Jensen, S., and Ehrenberg, L. (1986)
Modified Edman degradation applied to hemoglobin for monitoring
occupational exposure to alkylating agents. Toxicol. EnViron. Chem.
11, 215–231.
References
(1) Sriramachari, S., and Chandra, H. (1997) The lessons of Bhopal [toxic]
MIC gas disaster scope for expanding global biomonitoring and
environmental specimen banking. Chemosphere. 34, 2237–2250.
(2) Krone, C. A., Ely, J. T. A., Klingner, T., and Rando, R. J. (2003)
Isocyanates in flexible polyurethane foams. Bull. EnViron. Contam.
Toxicol. 70, 328–335.
(3) Greza, A. (2008) Polyurethanes (PU), cross-linked plastics, 123-127.
Translated from Kunststoffe International 10/2008, pp 205-211.
(4) Ka¨a¨ria¨, K., Hirvonen, A., Norppa, H., Vainio, H., Rosenberg, C., and
Piirila¨, P. (2001) Exposure to 2,4- and 2,6-toluene diisocyanate (TDI)
during production of flexible foam: determination of airborne TDI and
urinary 2,4- and 2,6-toluenediamine (TDA). Analyst 126, 1025–1031.
(5) Sabbioni, G., Wesp, H., Lewalter, J., and Rumler, R. (2007) Deter-
mination of isocyanate biomarkers in construction site workers.
Biomarkers 12, 468–483.
(6) Ramachandran, P. K., Gandhe, B. R., Venkateswaran, K. S., Kaushik,
M. P., Vijayaraghavan, R., Agarwal, G. S., Gopalan, N., Surya-
narayana, M. V., Shinde, S. K., and Sriramachari, S. (1988) Gas
chromatographic studies of the carbamylation of haemoglobin by
methyl isocyanate in rats and rabbits. J. Chromatogr. 426, 239–247.
(7) Sriramachari, S., Rao, G. J., Sharma, V. K., Jadhav, 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 carbamylation in
post-mortem blood. Med. Sci. Law 31, 289–293.
(8) Mraz, J., Simek, P., Chvalova, D., Nohova, H., and Smigolova, P.
(2004) Studies on the methyl isocyanate adducts with globin. Chem.-
Biol. Interact. 148, 1–10.
(9) Ka¨fferlein, H. U., and Angerer, J. (2001) N-Methylcarbamoylated
valine of hemoglobin in humans after exposure to N,N-dimethylfor-
mamide: Evidence for the formation of methyl isocyanate. Chem. Res.
Toxicol. 14, 833–840.
(10) Sabbioni, G., Hartley, R., and Schneider, S. (2001) Synthesis of adducts
with amino acids as potential dosimeters for the biomonitoring of
humans exposed to toluenediisocyanate. Chem. Res. Toxicol. 14, 1573–
1583.
(11) Sabbioni, G., Hartley, R., Henschler, D., Ho¨llrigl-Rosta, A., Koeber,
R., and Schneider, S. (2000) Isocyanate-specific hemoglobin adduct
in rats exposed to 4,4′-methylenediphenyl diisocyanate. Chem. Res.
Toxicol. 132, 82–89.
(23) Helleberg, H., Westerholm, R., Ehrenberg, L., and Tornqvist, M. (1998)
Synthesis and characterization of N-alkylated L-valine methylamide
products from diol epoxide metabolites of fluoranthene and ben-
zo[a]pyrene. Acta Chem. Scand. 52, 797–805.
TX900278P