Diphenylthiourea, a common rubber chemical, is bioactivated to potent skin sensitizers

Article abstract of DOI:10.1021/tx100241z

Diphenylthiourea (DPTU) is a known skin sensitizer commonly used as a vulcanization accelerator in the production of synthetic rubber, for example, neoprene. The versatile usage of neoprene is due to the multifaceted properties of the material; for example, it is stretchable, waterproof, and chemical-and abrasion-resistant. The wide application of neoprene has resulted in numerous case reports of dermatitis patients allergic to DPTU. The mechanism by which DPTU works as a contact allergen has not been described; thus, the aim of the present study was to investigate if DPTU is a prohapten that can be activated by skin metabolism. The metabolic activation and covalent binding of ¹⁴C-labeled DPTU to proteins were tested using a skinlike cytochrome P450 (P450) cocktail containing the five most abundant P450s found in human skin (CYP1A1, 1B1, 2B6, 2E1, and 3A5) and human liver microsomes. The incubations were carried out in the presence or absence of the metabolite trapping agents glutathione, methoxylamine, and benzylamine. The metabolism mixtures were analyzed by LC-radiochromatography, LC-MS, and LC-MS/MS. DPTU was mainly metabolically activated to reactive sulfoxides resulting in desulfurated adducts in both enzymatic systems used. Also, phenylisothiocyanate and phenylisocyanate were found to be metabolites of DPTU. The sensitizing capacity of the substrate (DPTU) and three metabolites was tested in the murine local lymph node assay. Two out of three metabolites tested were strong skin sensitizers, whereas DPTU itself, as previously known, was negative using this mouse model. In conclusion, DPTU forms highly reactive metabolites upon bioactivation by enzymes present in the skin. These metabolites are able to induce skin sensitization and are probable causes for DPTU allergy. To increase the possibilities of diagnosing contact allergy to DPTU-containing items, we suggest that suitable metabolites of DPTU should be used for screening testing.

Full text of DOI:10.1021/tx100241z

Chem. Res. Toxicol. 2011, 24, 35–44
35
Diphenylthiourea, a Common Rubber Chemical, Is Bioactivated to
Potent Skin Sensitizers
Kristin Samuelsson,† Moa Andresen Bergstro¨m,‡ Charlotte A Jonsson,† Gunnar Westman,§
and Ann-Therese Karlberg*,†
Dermatochemistry and Skin Allergy, Department of Chemistry, UniVersity of Gothenburg, Gothenburg, Sweden,
Clinical Pharmacology & DMPK, AstraZeneca R&D Mo¨lndal, Mo¨lndal, Sweden, and Department of Chemical
and Biological Engineering/Organic Chemistry, Chalmers UniVersity of Technology, Gothenburg, Sweden
ReceiVed July 20, 2010
Diphenylthiourea (DPTU) is a known skin sensitizer commonly used as a vulcanization accelerator in
the production of synthetic rubber, for example, neoprene. The versatile usage of neoprene is due to the
multifaceted properties of the material; for example, it is stretchable, waterproof, and chemical- and
abrasion-resistant. The wide application of neoprene has resulted in numerous case reports of dermatitis
patients allergic to DPTU. The mechanism by which DPTU works as a contact allergen has not been
described; thus, the aim of the present study was to investigate if DPTU is a prohapten that can be
activated by skin metabolism. The metabolic activation and covalent binding of 14C-labeled DPTU to
proteins were tested using a skinlike cytochrome P450 (P450) cocktail containing the five most abundant
P450s found in human skin (CYP1A1, 1B1, 2B6, 2E1, and 3A5) and human liver microsomes. The
incubations were carried out in the presence or absence of the metabolite trapping agents glutathione,
methoxylamine, and benzylamine. The metabolism mixtures were analyzed by LC-radiochromatography,
LC-MS, and LC-MS/MS. DPTU was mainly metabolically activated to reactive sulfoxides resulting in
desulfurated adducts in both enzymatic systems used. Also, phenylisothiocyanate and phenylisocyanate
were found to be metabolites of DPTU. The sensitizing capacity of the substrate (DPTU) and three
metabolites was tested in the murine local lymph node assay. Two out of three metabolites tested were
strong skin sensitizers, whereas DPTU itself, as previously known, was negative using this mouse model.
In conclusion, DPTU forms highly reactive metabolites upon bioactivation by enzymes present in the
skin. These metabolites are able to induce skin sensitization and are probable causes for DPTU allergy.
To increase the possibilities of diagnosing contact allergy to DPTU-containing items, we suggest that
suitable metabolites of DPTU should be used for screening testing.
During the years, the allergenic activity of DPTU has been
studied in three different animal models: the murine local lymph
node assay (LLNA) (11), the sensitive local lymph node assay
(SLNA) (12), and the guinea pig maximization test (GPMT)
(13). DPTU was identified as a sensitizer in the GPMT and the
SLNA but was found to be nonsensitizing in the LLNA. This
disparity is most probably due to differences in the administra-
tion of the compound in the LLNA versus the SLNA and the
GPMT. In the two latter methods, the test compound is
administered intradermally in addition to topical application.
Intradermal injections seem to be important to obtain a positive
response to DPTU as the penetration of DPTU probably is low
when applied topically on intact skin (14).
Introduction
Diphenylthiourea (DPTU)1 (Scheme 1) is a skin sensitizer
included in the patch test series of shoe and rubber chemicals
for diagnosis of allergic contact dermatitis (ACD) (1). DPTU
is present in various rubber and plastic products due to its use
as a stabilizer in the manufacturing of polyvinyl chloride (PVC)
and as an accelerator in the production of neoprene. Neoprene
has become a widely used material since it is waterproof,
stretchable, and chemical- and abrasion-resistant. Its extensive
use is mirrored in case reports of ACD to DPTU from various
exposures, for example, from contact with wet suits, swim
goggles, running shoes, orthopedic devices, mouse pads, and
iPod holders (2-9). However, contact allergy caused by rubber
products is not only due to DPTU since other derivatives of
thioureas can be used in the synthesis of rubber (10).
The majority of contact allergens, known as haptens, are low
molecular weight chemicals able to react with nucleophilic
groups on proteins to form immunogenic hapten-protein
complexes. However, compounds inherently unreactive to
proteins may also induce ACD as a result of bioactivation to
reactive compounds in the skin. These compounds, prohaptens,
contain structural moieties susceptible to metabolic transforma-
tion into protein-reactive species able to cause contact allergy
(15, 16). For many years, the skin was considered to be an inert
passive structural barrier, but it is now obvious that the skin
plays a key role in the penetration and metabolism of topically
applied compounds (15, 17-19). Bioactivation generally in-
volves oxidative processes that are attributed to monooxyge-
* To whom correspondence should be addressed. Tel: + 46317869109.
Fax: +46317721394. E-mail: karlberg@chem.gu.se.
† University of Gothenburg.
‡ AstraZeneca R&D Mo¨lndal.
§ Chalmers University of Technology.
1 Abbreviations: ACD, allergic contact dermatitis; A:DBP, acetone:
dibutylphthalate; BA, benzylamine; DPTU, diphenylthiourea; DPU, diphe-
nylurea; ESI, electrospray interface; FMO, flavin-containing monooxyge-
nases; GPMT, guinea pig maximization test; LLNA, local lymph node assay;
MA, methoxylamine; MEST, mouse ear swelling test; PIC, phenylisocy-
anate; PITC, phenylisothiocyanate; PVC, polyvinyl chloride; RAD, radio-
activity detection; SI, stimulation index; SLNA, sensitive local lymph node
assay.
10.1021/tx100241z  2011 American Chemical Society
Published on Web 11/12/2010

36 Chem. Res. Toxicol., Vol. 24, No. 1, 2011
Samuelsson et al.
Scheme 1. Proposed Metabolic Pathways of DPTU in the Skinlike P450 Cocktail and Human Liver Microsomesa
a Structures within brackets are suggested intermediate compounds. Adducts A1-A8 were detected in coincubations with reactive metabolite trapping
agents glutathione (GSH), methoxylamine (MA), and benzylamine (BA).
nases, such as cytochrome P450 (P450) and flavin-containing
monooxygenases (FMO) (20, 21). The most common phase I
metabolizing enzymes present in the skin are P450s (15), and
the most abundant P450s found in human skin samples are
CYP1A1, 1B1, 2B6, 2E1, and 3A5 (22). A cocktail was
developed to reflect the metabolism in human skin (22) as there
are highly expressed cutaneous P450s that are not as pronounced
in the liver (23). The ratios between the individual P450
enzymes are the same in the cocktail as in the skin. As the ability
of a hapten or metabolite to form hapten-protein complexes is
pivotal for the development of contact allergy, bioactivation and
covalent binding to proteins can be assessed in vitro by
incubating radiolabeled compounds with microsomal prepara-
tions and measuring nonextractable radioactivity associated with
proteins. In addition to studying covalent binding, the structure
of the reactive metabolites can be probed using reactive
intermediate trapping experiments (24).
The thionocarbonyl group of thioureas, such as DPTU, is
known to be oxidized by FMOs and P450s to sulfenic (S-OH),
sulfinic (-SO2H), or sulfonic (-SO3H) acids, which can cause
alkylation of proteins (25, 26). Thioureas are metabolized in
vivo into desulfurated products. This is thought to be a result
of nucleophilic substitution or elimination reactions of the
sulfenic- and sulfinic acid metabolite intermediates. The oxidized
thiourea compounds are also believed to be responsible for the
toxicity of thioureas. Due to the instability of the intermediate
metabolites of thioureas, the formation of oxidation products
and the alkylating potential in metabolic mixtures have been
studied by indirect measurements in different assays. Studies
have shown that thioureas need to be metabolically activated
to become protein reactive (27, 28). However, thioureas can
also form reactive isothiocyanates by thermal decomposition
(29).
The aim of the present study was to investigate if DPTU can
act as a prohapten upon skin contact and to identify possible
reactive metabolites that could be formed in the skin. The
potential of DPTU to covalently bind to proteins was also
investigated. Furthermore, we were interested in studying the
formation of the putative reactive metabolite phenylisothiocy-
anate (PITC), since simultaneous reactions to PITC and DPTU
have been observed in patients (9). As isothiocyanates can be
metabolically converted into the corresponding isocyanate, we
were also interested in studying the formation of phenylisocy-
anate (PIC) from DPTU. The metabolism of 14C-labeleld DPTU
was examined in vitro using the skinlike P450 cocktail and
human liver microsomes. The covalent binding of DPTU and
metabolites to proteins was studied in the P450 cocktail and in
human liver microsome incubations and has to our knowledge
not been studied in this context before. To gain further insight
into the nature of the reactive metabolites formed, trapping
experiments using glutathione (GSH), methoxylamine (MA),
and benzylamine (BA) were performed. The sensitizing capaci-
ties of DPTU and three putative metabolites were investigated
using the LLNA in mice (30). In addition, elicitation studies
using the mouse ear swelling test (MEST) method were
performed (31).
Experimental Procedures
Chemicals and Biochemicals. 14C-DPTU was synthesized by
Isotope Chemistry at AstraZeneca R&D Mo¨lndal (Mo¨lndal, Swe-
den). A mixture of 14C6-DPTU and unlabeled DPTU with a specific
activity of 4.51 kBq/nmol was used for the microsomal incubations.
PITC (g99%), PIC (99%), and BA (g99%) were purchased from
Fluka (Buchs, Switzerland). Dibutyl phthalate (DPB, g99%) and
acetone (g99%) used in a 1:1 mixture as vehicle in the animal
experiments were purchased from Merck (Darmstadt, Germany)

Metabolism of Diphenylthiourea
Chem. Res. Toxicol., Vol. 24, No. 1, 2011 37
and Sigma-Aldrich (Schnelldorf, Germany), respectively. Human
liver microsomes and P450 enzymes were purchased from BD
Biosciences (San Jose´, CA) and Tebu-Bio (Offenbach, Germany),
respectively. OptiPhase Hisafe 3 from PerkinElmer (Waltham, MA)
was used as a scintillation cocktail.
(Wallac 1409, Wallac Oy, Kuopio, Finland). An activity of <500
dpm was considered satisfactory to ensure that the pellets had been
adequately washed. Each pellet was solubilized by heating and
shaking in aqueous sodium hydroxide (1 mL, 0.1 M) containing
sodium dodecyl sulfate (5%) at 40 °C overnight in a water bath.
The resulting solutions were analyzed for total radioactivity using
static liquid scintillation counting, and the protein content was
determined using the Markwell-modified Lowry assay with bovine
serum albumin as standard (32). The levels of nonextractable
radioactivity were expressed as pmol equivalents per milligram
protein (pmol equiv/mg protein) using the specific activity of the
test compound and the protein levels for each sample, according
to the following equation:
Synthesis. 1-Benzyl-3-phenylthiourea (A8 in Scheme 1). PITC
(10 mmol, 1.53 mL) was added to a solution of BA (10 mmol,
1.05 mL) in hexane (10 mL). The mixture was stirred at room
temperature for 1 h, and the product was collected by filtration
1
and washed with hexane yielding 86% of A8. H and 13C NMR
spectroscopy (400 and 100 MHz, respectively) were performed on
a JEOL Eclipse 400 instrument using deuterated DMSO as solvent.
Chemical shifts (δ) are reported using residual DMSO δ 2.5 and δ
1
1
50.0 as references for H and 13C shifts, respectively. H NMR: δ
8.0 (s, 1H), 7.18-7.42 (m, 10H), 6.3 (s, 1H), and 4.9 (d, 2H, J )
6 Hz). 13C NMR: δ 181.3, 139.7, 139.6, 129.2, 128.8, 128.0, 127.4,
124.9, 123.9, and 47.8. MS/MS 243 (M+ 45%), 119 (9%), 94 (31%),
and 91 (100%).
amount of binding (pmol equiv/mg protein) )
measured radioactive concentration (Bq/mL)
specific activity (Bq/pmol) × protein concentration (mg/mL)
1-Benzyl-3-phenylurea (A6 in Scheme 1). PIC (10 mmol, 1.09
mL) was added to BA (10 mmol, 1.53 mL) in hexane (10 mL)
during stirring. The mixture was stirred at room temperature for
1 h, and the product was collected by filtration and washed with
hexane. The yield was 91% of A6. 1H NMR: δ 4.3 8.5 (s, 1H), 7.3
LC/MS/RAD Analysis of Incubation-Mixture Supernatants.
The supernatants were analyzed using LC combined with radio-
activity detection (RAD) for quantification and MS for structural
elucidation of parent compound and metabolites. The LC system
(Agilent Series 1100, Palo Alto, CA) consisted of an autosampler,
a solvent degassing unit, and a gradient pump. The column oven
was set to 25 °C, and the sample tray temperature was set to 8 °C.
The mobile phases used were (A) H2O with 20 mM ammonium
acetate and (B) acetonitrile. Chromatographic separations were
performed on a Hypurity C18 LC column (4.6 mm i.d. × 150 mm;
particle size 5 µm) protected by an Optiguard CN column (1 mm
i.d.). A flow rate of 1 mL/min was used. All samples were diluted
with water to a concentration of 20% acetonitrile in water before
analysis, and the injection volume was 40 µL. Used was the
following LC gradient: The gradient program started with a 30 min
linear gradient from 10 to 30% B, followed by a 15 min gradient
from 30 to 60% B. An increase from 60 to 90% B during 25 min
was followed by 5 min of isocratic elution at 90% B, after which
the eluent composition returned to the initial conditions (10% B)
for 10 min. The total run time was 115 min.
RAD was performed with an online radioactivity detector
(Radiomatic Flo One model TR-525, Packard, Meriden, CT) with
a 100 µL flow cell and Ultima Flo AP scintillation fluid. The
counting results were analyzed using Laura (version 3, LabLogic).
For MS detection, ∼0.1 mL/min of the LC eluent was directed
to the mass spectrometer. The mass spectrometer used was a hybrid
quadrupole time-of-flight mass spectrometer (Micromass Q-Tof 2)
with an electrospray interface (ESI) and a LockSpray probe. Specific
mass spectrometric source conditions were as follows: cone voltage,
40 V; capillary voltage, 3.20 kV; MCP, 2000 V; TOF, 9.10 kV;
source temperature, 120 °C; and desolvation temperature, 320 °C.
MassLynx (version 4, Waters) was used for controlling the mass
spectrometer and for data evaluation. The mass range was 80-1000
Da, and MS spectra were acquired in the positive ionization mode.
MS/MS spectra were acquired using collision energies ranging
between 15 and 25 V. The MS/MS mass range was m/z 80-650.
A second LC-MS/MS instrument was used with the LC system
(Agilent Series 1200) consisting of an autosampler, a solvent
degassing unit, and a gradient pump. The column oven was set to
25 °C, and the sample tray temperature was set to 8 °C. The same
column and elution system was used as to the Agilent Series 1100.
For MS detection, about 0.4 mL/min of the LC eluent was directed
to the 6410B Triple Quad (Agilent) with an ESI interface. The data
were analyzed using MassHunter. This instrument was only used
in the product ion scan mode with the collision energies ranging
from 10 to 30 V. The fragmentor voltage was set to 50 V. The
mass range was from m/z 80 to parent ion plus 50 Da.
(m, 9H), 6.9 (t, 1H, 1.1), 6.6 (t, 1H, J ) 5.8), (d, 2H, J ) 5.8). 13
C
NMR: δ 155.8, 141.0, 140.9, 129.2, 128.8, 127.7, 127.3, 121.6,
118.2, and 43.3. MS/MS 227 (M+ 45%), 119 (9%), 94 (31%), and
91 (100%).
1,3-Diphenylurea (DPU; M3 in Scheme 1). DPU was obtained
in 80% yield by adding PIC (10 mmol, 1.07 mL) to a solution of
aniline (10 mmol, 0.91 mL) dissolved in hexane (5 mL) and stirred
at room temperature for 1 h. The product was collected by filtration
1
and washed with portions of hexane. H NMR: δ 8.6 (s, 2H), 7.4
(d, 4H, J ) 7.7 Hz), 7.3 (t, 4H, J ) 7.7 Hz), 6.9 (t, 2H, J ) 7.3
Hz). 13C NMR: δ 153.1, 140.3, 129.4, 122.4, and 118.7. MS/MS
213 (M+ 81%), 120 (2%), and 94 (100%).
Human Liver Microsomal and Skinlike P450 Cocktail
Incubations. 14C-DPTU (50 µM, 4.51 kBq/nmol) was incubated
with human liver microsomes (1 mg protein/mL) or the skinlike
P450 cocktail in the presence of NADPH (1 mM) and MgCl2 (5
mM) in potassium phosphate buffer (0.1 M, pH 7.4) in a total
volume of 500 µL. The P450 cocktail was prepared by mixing
human P450s expressed in bacterial along with cytochrome P450
reductase. The total P450 content was 56 pmol, a concentration
that was 2.5 times more than that used in the original protocol
according to Bergstrom et al. (22). The content of the skinlike P450
cocktail was 1A1 (9.0 pmol), 1B1 (5.0 pmol), 2B6 (0.088 pmol),
2E1 (28 pmol), and 3A5 (14 pmol). The total amount of protein in
the P450 incubations was 0.5 mg/mL. Incubations run in the absence
of NADPH or using heat-inactivated enzymes served as negative
controls. For heat inactivation, the enzymes were preincubated at
a temperature of >75 °C for 10 min. An additional control
experiment was performed without enzymes using unlabeled DPTU
to probe its decomposition in the conditions used. Trapping
experiments were carried out using either of the trapping agents,
GSH, MA, or BA, at a final concentration of 5 mM. The reactions
were initiated by the addition of NADPH after approximately 3
min of preincubation at 37 °C. The incubations were terminated
after 60 min, and the protein was precipitated by the addition of
ice-cold acetonitrile (0.5 mL). The samples were placed on ice for
10 min and were then centrifuged at 2600g at 4 °C for 15 min.
The supernatants were stored at -20 °C awaiting analysis. Also,
the protein pellets were retained for analysis of covalent binding
of radiolabeled compounds to proteins. All incubations were
performed in triplicate.
Covalent Binding to Proteins. The precipitated protein pellets
from the microsomal incubations were washed with methanol (2
× 1 mL), acetonitrile (2 × 1 mL), and acetonitrile:water (4:1) (2
× 1 mL). The pellets were disintegrated by vortex mixing, and the
washing solvent was removed to waste after centrifugation at 2600g
and 4 °C for 10 min. After the sixth wash, aliquots (200 µL) of the
washing liquid were analyzed using liquid scintillation counting
Experimental Sensitization Studies (LLNA). The LLNA (33)
was used to determine the sensitizing capacities of DPTU, DPU,
PITC, and PIC. All animal procedures were approved by the local
ethics committee. Female CBA/Ca mice, 8-12 weeks, in groups
of three animals, were exposed to a test compound dissolved in
A:DPB (1:1) in five different concentrations (Table 1). A sham-

38 Chem. Res. Toxicol., Vol. 24, No. 1, 2011
Samuelsson et al.
Table 1. Cell Proliferation Induced by DPTU, DPU, PITC,
and PIC Using the LLNAa
and the vehicle (A:DBP, 1:1), one on each ear. For each treatment,
the percent relative ear thickness was calculated from the sum of
the quotient of ear thickness before and 24 h after exposure divided
by the number of ears treated. For the vehicle-treated mice, the
mean of four ears was obtained, while the mean of two ears was
calculated for the sensitizers. Neither of the compounds increased
the ear thickness, indicating no irritation at the concentrations tested
(Table S1 in the Supporting Information).
Cross Reactivity Studies Using the MEST. The MEST experi-
ments were performed as described in literature (35) with some
modifications: change of mouse strain, vehicle, and omission of
adjuvant or other enhancement techniques to perform the experiment
using the same animal strain and vehicle as in the LLNA. Male
CBA/Ca mice 8-12 weeks old were used in groups of eight or
four animals (Table 2). Under isoflurane narcosis, induction was
carried out on days 0, 1, and 2 by topical applications on the shaven
backs (approximately 2 cm2) of the mice with 100 µL of the test
compounds in A:DBP (1:1) (Table 2). The ear thickness was
measured under isoflurane narcosis at day 7, after which the mice
were challenged topically on the dorsum of both ears with 25 µL
of test compound in the vehicle as shown in Table 2. Ear thickness
measurements were performed again after 24 and 48 h. The relative
ear thickness at these two time intervals after challenge was
calculated for each mouse using the following formula: relative
ear thickness (%) ) 100[(Ax/Bx) + (Ay/By)]/2 (A ) the ear thickness
after 24 h, B ) the ear thickness prior to exposure, X ) ear 1, and
Y ) ear 2). The mean relative ear thickness for each group of mice
is presented in Table 2.
EC3c
compound concentration
(% w/v)
DPM/lymph
node
SIb
%
mM
DPTU
0.10
1.0
2.5
5.0
-
-
710
981
994
824
1564
828
0.86
1.2
1.2
1.0
1.9
10
vehicled
DPU
0.10
1.0
-
-
1029
1096
915
1320
1779
966
1.1
1.1
0.95
1.4
1.2
2.5
5.0
10
vehicle
PITC
0.0010
0.010
0.10
1.0
2.0
vehicle
PIC
0.40
0.020
30
1.7
704
763
1758
7204
10533
1196
0.59
0.64
1.5
6.0
8.8
0.0010
0.10
1.0
5.0
10
1380
7077
7212
12224
15009
898
1.5
7.9
8.0
14
Statistical Analysis. The results from the MEST were statistically
evaluated using an exact permutation variant of Wilcoxons rank
sum test. The individual increase in relative ear thickness observed
of each mouse within the groups was used when groups 2-5 were
compared with control group 1.
17
vehicle
a The local lymph node experiments were performed as described in
the Experimental Procedures. b An increase in 3[H]-thymidine incorpo-
ration relative to vehicle-treated controls was derived for each
experimental group and recorded as a SI. c EC3 is the estimated
concentration required to induce an SI of 3 and is calculated using linear
interpolation. d Acetone:dibutylphthalate (A:DBP) (1:1).
Results
Identification of DPTU Metabolites. The metabolism of
14C6-DPTU was assessed, and metabolites were detected after
incubations with the skinlike P450 cocktail or human liver
microsomes. The metabolites and trapped adducts were sepa-
rated using reversed-phase HPLC and online RAD for quanti-
fication. MS/MS was used for structure elucidation. The total
conversion of DPTU after 60 min of incubation in human liver
microsomes and the P450 cocktail was 29 and 12%, respectively.
A loss of approximately 10% of parent was observed in the
control experiments without NADPH and is considered to be
due to decomposition of DPTU (data not shown). This was
confirmed in an additional control experiment run in the absence
of metabolizing enzymes.
treated control group was exposed to the vehicle alone. The highest
test concentrations used for DPTU and DPU were saturated
solutions at room temperature. The results of LLNA were obtained
as mean dpm/pooled lymph nodes of each animal group from which
the stimulation index (SI) was calculated, that is, the test group
dpm/control group dpm ratio. An EC3 value, that is, the estimated
concentration of the test material required to induce an SI of 3 (the
limit to be classified as a sensitizer), was calculated by linear
interpolation (34).
Irritation Study. Prior to the MEST (see below), a control
experiment was performed to determine what test concentrations
should be used to avoid irritation. Under isoflurane narcosis, the
ear thickness of na¨ıve male CBA/Ca mice, 8-12 weeks old, was
measured with a caliper before and 24 h after topical application
of vehicle, PITC, or PIC on the dorsum of the ears. Used
concentrations of the compounds are listed in Table S1 in the
Supporting Information. To reduce the number of animals, each
mouse was exposed to two different compounds, or one compound
Radiochromatograms from the human liver microsome in-
cubations with and without NADPH are shown in Figure 1. In
total, six metabolites were detected in the human liver mi-
crosome incubations without trapping agents. Four of the
metabolites were detected by MS and are denoted M1-M4 in
Table 2. Results from the MESTa
mean relative ear thicknessb ((SD)
no. of
animals
induction compound
(%, w/v)
challenge compound
(%, w/v)
observed reactionsc
at 24 and 48 h
group
24 h
48 h
1
2
8
8
8
4
4
PITC (2.0)
PITC (2.0)
PITC (2.0)
PITC (2.0)
PIC (0.50)
vehicled
104 (2.7)
101 (1.5)
113 (7.0)
112 (3.6)
121 (10.5)
104 (3.3)
101 (1.6)
113 (5.7)
111 (4.8)
140 (5.0)
-
-
+
+
+
DPTU (5.0)
PIC (0.50)
PITC (2.0)
PIC (0.50)
3
4e
5e
a The experiment was performed as described in the Experimental Procedures. b Relative ear thickness at 24 and 48 h after challenge as compared to
0 h was calculated for each mouse as follows: (%) ) 100(AX/BX + AY/BY)/2, where A ) the ear thickness after challenge (24 or 48 h), B ) the ear
thickness prior to challenge (0 h), X ) right ear, and Y ) left ear. c Groups 2-5 were compared to group 1 using an exact permutation variant of
Wilcoxons rank sum test. Groups 3-5 were all statistically distinguished from the control group 1, a positive (+) observed reaction. The p values were
<0.01 in all cases except for the reaction of group 4 at 48 h, which had a p value of <0.05. No increased ear thickness was observed in the DPTU
exposed mice as compared to the vehicle. d A:DBP (1:1). e Positive controls.

Metabolism of Diphenylthiourea
Chem. Res. Toxicol., Vol. 24, No. 1, 2011 39
Figure 1. LC radiochromatograms obtained from metabolism of 14C-
labeled DPTU in human liver microsomes. Panels A and B show
experiments with and without NADPH, respectively. The incubations
were run in triplicate and pooled before LC analysis. Each incubation
contained a substrate concentration of 50 µM at 0.5 mg protein/mL
and was incubated for 60 min at 37 °C. The metabolites (M1-M4)
are numbered in order of elution and were further identified with LC-
MS/MS. Peaks marked with * could not be detected by LC/MS. The
arrow denotes the metabolite M3, seen as a shoulder of a peak from a
NADPH-independent coeluting unknown compound.
order of elution time. No mass information, neither in positive
nor in negative ionization mode, could be obtained for the
remaining two DPTU metabolites, which are denoted *1 and
*2 (Figure 1A). In the P450 cocktail incubations, M1-M4 were
detected, however, the nonidentified peaks from the human liver
microsome incubations were not seen (Figure 2A). Amounts
of metabolites formed and masses of metabolites are presented
in Table 3. M1, a monooxygenated metabolite, was the major
metabolite formed in incubations with both the P450 cocktail
and the human liver microsomes. The minor metabolites
M2-M4 were formed via monooxygenation (M2), desulfuration
(M4), and combined oxygenation and desulfuration (M3).
Proposed metabolic pathways of DPTU are shown in Scheme
1. The compositions of DPTU and metabolites thereof are based
on LC-MS/MS data (Table 3), and all metabolites detected by
MS/MS have the same isotope pattern as DPTU. The m/z given
for the compounds is the unlabeled MH+. The MS/MS spectrum
of DPTU (MH+ at m/z 229) contains product ions at m/z 195,
136, and 94. Proposed structures of the DPTU product ions are
shown in the Supporting Information (Figure S1). M1 and M2
have pseudomolecular ions of m/z 245 corresponding to a net
mass gain of +16 Da and are thus monooxygenated metabolites.
Both M1 and M2 produced product ions at m/z 227, which
indicates that both compounds lose the mass of one water
molecule upon fragmentation of the parent at m/z 245. Hy-
droxylation on aromatic rings is a common metabolic pathway,
but loss of water is unlikely to occur for phenols when subjected
to collision-induced dissociation. Thus, on the basis of MS/MS
interpretation, M1 and M2 are postulated to be the sulfenic acid
Figure 2. LC radiochromatograms obtained from 14C-labeled DPTU
in the skinlike P450 cocktail. Panel A is without trapping agent, and
panels B-D show profiles after coincubation with GSH, MA, and BA,
respectively. The chromatograms are the results of three pooled reactions,
each performed with a substrate concentration of 50 µM at 0.25 mg protein/
mL for 60 min at 37 °C. The designated peaks of metabolites (M1-M4)
and adducts (A1-A8) are in order of elution and were further identified
with LC-MS/MS. Peaks marked with ‡ could not be detected by MS. The
arrow denotes the metabolite M3, seen as a shoulder of a peak from a
NADPH-independent coeluting unknown compound.
(S-OH) and hydroxylamine (N-OH) of DPTU, respectively.
S- and N-oxidations are catalyzed both by microsomal P450
and FMO (25). S-oxidation produces sulfenic (S-OH), sulfinic

40 Chem. Res. Toxicol., Vol. 24, No. 1, 2011
Samuelsson et al.
Table 3. Percentagesa, Masses, and MS/MS Data of DPTU
and Metabolites Formed in Human Liver Microsome and
P450 Incubations
Table 4. Relative Amounts of Adducts Formed in Reactive
Metabolite Trapping Experiments in Human Liver
Microsomesa
human liver
microsomes +
NADPH (%) NADPH (%)
adduct (retention time, min)
GSH (%)b
MA (%)
BA (%)
metabolites MH+
(tR, min) (m/z)
product ions
P450 +
A1 (28.9)
A2 (9.4)
41.0
ND
ND
ND
ND
ND
ND
ND
ND
11
41
ND
ND
ND
ND
ND
NQ
42
(m/z)
DPTU (50.3) 229 195, 136, 94
A3 (29.5)
A4 (67.0)
A5 (86.5)
A6 (54.1)c
A7 (58.7)
A8 (65.9)
M1 (29.1)
M2 (30.8)
M3b (58.1)
M4 (67.3)
*1 (49.2)
245 227, 195, 169, 152, 92
245 227, 152, 136, 110, 94
213 120, 94
47
1.5
1.1
20
15
15
64
20
7.7
9.6
17
1.5
ND
ND
ND
197 104, 94
12
*2 (81.2)
a ND, not detected. The limit of detection was set to 1000 cpm. NQ,
not quantified. b The relative amount (%) is calculated from the total
radioactivity of all metabolites and adducts detected in the incubations.
c A6 coeluted with DPTU and could therefore not be quantified.
a Areas in the radiochromatograms were used to calculate the percent
of formed metabolites. b The NADPH-independent compound from the
control experiment coeluting with M3 (Figures 1 and 2) was subtracted
from the incubations with NADPH.
(SO2H), and sulfonic acids (SO3H), summarized as sulfoxides
(SOxH) in Scheme 1. Thiourea sulfoxides are unstable and
difficult to isolate and identify. For this reason, only the sulfenic
acid of DPTU (M1) was identified in the metabolic mixture.
The MS/MS spectra of M1 and M2 are shown in the Supporting
Information (Figure S2). Using a synthetic standard, metabolite
M3 at m/z 213 was identified as diphenyl urea (DPU, Scheme
1). M3 produced the same MS/MS spectrum and had the same
retention time as the reference compound DPU (data not shown).
Metabolite M4 was the most nonpolar metabolite identified. This
compound was found in relatively low amounts and only in
incubates without trapping agents. The protonated molecular
ion had a m/z of 197 and is proposed to be diphenyl formamidine
(Scheme 1). Its MS/MS product ions at m/z 104 and 94
correspond to C7H6N and protonated aniline.
the same as those found for A4. The ion at m/z 209 is analogous
to the product of A4 at m/z 134 but contains BA as trapping
agent instead of MA. From A7, protonated BA is formed and
detected as m/z 108. Collision-induced dissociation of BA
products identified in this study did all produce the mass of 91,
which likely is the tropylium ion C7H7+ (38). Proposed structures
of adducts A2, A3, and A5 found in the incubates of MA and
NADPH are shown in Scheme 1. A2 is the result of a
dihydroxylated product of a DPTU-sulfoxide, while A3 results
from the corresponding monohydroxylated product. A5 was
identified as a DPTU dimer trapped by MA. The MS/MS spectra
of A2, A3, and A5 as well as proposed structures of product
ions are shown in Figure S3 in the Supporting Information.
Adducts A6 and A8 (m/z at 227 and 243, respectively) were
detected in incubations with BA (Figure 2). Their MS/MS
spectra and proposed fragment structures are shown in Figure
S4 in the Supporting Information. Retention times and MS/MS
spectra of A6 and A8 are identical to the synthetic standards,
and their structures are shown in Scheme 1. A6 and A8 are
formed from the reactive metabolites PIC and PITC (Scheme
1), respectively. PITC may also be formed as a thermal
decomposition product of DPTU (38), and small amounts of
A8 were also found in the controls (data not shown). Isothio-
cyanates may be metabolically converted into the corresponding,
and even more reactive, isocyanate PIC (39, 40), and thus, the
adduct A6 was formed as a result of further metabolism of PITC
to PIC. The product ions obtained from A6 and A8 are m/z
136, 119, 94, 91, and 119, 94, and 91, respectively. m/z 136
corresponds to PITC and m/z 119 to C6H5-NCNH2; the latter
was also present as a product ion of A4 and A7. The observed
fragment ion at m/z 94 is suggested to be anilinium ion, and
the ion at m/z 91 is (C7H7)+ as previously reported (38).
Protein Binding. As the formation of hapten-protein
complexes is a key event in contact allergy, the covalent binding
of radiolabeled products to proteins in 14C6-DPTU-containing
microsomal incubates was investigated. To determine the protein
binding propensity of DPTU before and after metabolic activa-
tion, the retained protein pellets from the skinlike P450 cocktail
and human liver microsome incubations were washed and
analyzed. Incubations run with NADPH were used to investigate
the protein reactivity of DPTU following metabolic activation.
Incubations run in the absence of NADPH, or heat-inactivated
microsomes were used to determine the inherent reactivity of
DPTU. Incubations run in the presence of the reactive metabolite
trappers GSH, MA, and BA were used to probe the nature of
the protein-reactive species. The covalent binding results are
presented in Figure 4. Protein binding was observed in both
human liver microsome and in skinlike P450 incubations and
in both the presence and the absence of NADPH. The human
Formation and Identification of Trapped Adducts. The
formation of reactive intermediate metabolites from DPTU in
incubations with the skinlike P450 cocktail and human liver
microsomes was investigated using GSH, MA, and BA as
reactive metabolite trapping agents. The incubates were analyzed
using LC-MS/RAD, and radiochromatograms from the P450
incubations are shown in Figure 2. The results from the P450
incubations were qualitatively similar to those of the human
liver microsome incubations except for the two compounds not
detected with MS. Peaks marked with ‡ could not be identified
by MS detection. In the trapping experiments, one GSH adduct
(A1, Figure 2B), four MA adducts (A2-A5, Figure 2C), and
three BA adducts (A6-A8, Figure 2D) were found. All adducts
had the same isotope pattern as DPTU and were subjected to
MS/MS analysis. The proposal of structures for adducts A1,
A4, and A7 (Scheme 1) is based on MS/MS data, and the fact
that thioureas are known to be bioactivated into reactive sulfenic,
sulfinic, and sulfonic acid metabolites. These acids will readily
react with amino and thiol compounds to form guanidines and
isothioureas, respectively (36, 37). The relative amounts of
adducts formed in the trapping experiments of the P450
incubations are shown in Table 4. The guanidine and isothiourea
adducts are the major adduct type formed in the various
incubations. The MS/MS spectra and proposed fragment
structures for the GSH adduct A1, the MA adduct A4, and the
BA adduct A7 are shown in Figure 3. Similar to DPTU and
M1, A4 and A7 produced a fragment at m/z 195, which is
suggested to be a nitriluim ion (Figure 3). The product ion of
A4 at m/z 210 is loss of the methoxy group from the MA moiety,
and m/z 119 is subsequent loss of one aniline group. The loss
of C6H5-N from A4 gave the product ion of m/z 134, whereas
+
m/z 93 corresponds to C6H5NH2 . Upon collision of adduct A7,
the product ions were observed as follows: 209, 195, 119, 108,
and 91. The m/z 195 and 119 product ions formed from A7 are

Metabolism of Diphenylthiourea
Chem. Res. Toxicol., Vol. 24, No. 1, 2011 41
Figure 3. MS/MS spectra of the three major adducts produced (A1, A4, and A7) from both human liver microsomes and skinlike P450 cocktail.
liver microsome incubations in the presence of NADPH revealed
that 22% of the total radioactivity added to the incubations was
protein bound. This is approximately two times more than that
of the NADPH-free or heat-deactivated incubations (Figure 4),
thus showing that DPTU has undergone metabolic activation.
For the incubations with the skinlike P450s, the difference in
binding between the NADPH supplemented and the NADPH-
free incubations is not as pronounced, 10 and 7%, respectively.
A decrease in metabolism-dependent covalent binding to
proteins was observed in all human liver microsome experiments
containing trapping agents when compared to NADPH without
trapper (Figure 4A). This was also observed in the skinlike P450
incubations but was not as pronounced when using MA as
trapping agent (Figure 4B). Interestingly, decreased levels of
binding were also observed in the NADPH-free experiments
using the P450 cocktail incubations with GSH and MA and for
all NADPH-free trapping experiments using the human liver
microsomes, which could be due to trapping of reactive
degradation products (Figure 4).
Sensitizing Capacity in Mice. The LLNA was used to
determine the sensitizing capacities of DPTU and the three
metabolites DPU (M3), PITC, and PIC. As various thiourea
sulfoxides previously have been reported to be unstable and
difficult to isolate (41, 42), the oxidized products of DPTU were
neither synthesized nor tested. The murine LLNA is the method
of choice according to the OECD Guidelines for predictive
testing of the sensitizing capacity of compounds (43). However,
it has been shown to give false negative results when testing
thioureas (12, 13). Despite in previous attempts to provide a
more sensitive LLNA, we tried an alternative vehicle and used
saturated solutions as the highest test concentrations for DPTU
and DPU. Saturated solutions are found to increase the degree
of penetration of a chemical into the epidermis, which is an
important factor in the sensitization process in the LLNA (44).

42 Chem. Res. Toxicol., Vol. 24, No. 1, 2011
Samuelsson et al.
ear thickness was detected after PIC challenge; that is, cross
reactivity was demonstrated (Table 2). As positive controls, mice
were sensitized and challenged with either PITC (group 4) or
PIC (group 5), showing that these are contact sensitizers able
to induce a challenge response. The results are presented as
mean relative ear thickness after 24 and 48 h in Table 2.
Discussion
We have, for the first time, shown that DPTU can act as a
prohapten. DPTU was bioactivated to reactive metabolites in
vitro by the major enzymes found in the skin as well as by
human liver microsomes. Four metabolites and eight trapped
adducts were identified from incubations with an enriched
skinlike P450 cocktail and human liver microsomes, respectively
(Scheme 1). Metabolites were also shown to undergo covalent
binding to proteins following NADPH-dependent bioactivation
of DPTU, an experiment previously not used for studying
prohaptens in contact allergy. The observed metabolism in the
skinlike P450 cocktail indicates that DPTU can be metabolized
to potent sensitizers in human skin. In addition to metabolic
activation, DPTU was also found to undergo nonenzymatic
activation to reactive species.
As contact allergens enter the body via the skin, the metabolic
activity in this tissue is of major importance for topically applied
prohaptens. To study skin metabolism, we used an earlier
developed skinlike P450 cocktail (22). Also, human liver
microsomes, commonly used to study in vitro metabolism, were
used as a positive control for metabolic activation of DPTU.
There are several hepatic P450s expressed in the skin, and four
of the five most common P450s present in the cocktail are not
highly expressed in the liver (2E1 is the exception) (23, 46).
Two of the metabolites (*1 and *2 in Figure 1) were found in
the human liver microsome incubates but not in the skinlike
P450 incubates. This variation is likely to be due to the
difference in enzyme composition between the two oxidative
systems used. Unfortunately, the structures of these metabolites
could not be revealed as they were not detected with MS. The
metabolites formed in the human liver microsome incubates
might be produced in human skin in vivo by other enzymes
except for the five P450s present in the cocktail.
The major metabolite formed in both the P450 cocktail and
the human liver microsome experiments was the sulfenic acid
of DPTU (M1 in Table 3). M1 can either be further oxidized to
reactive sulfinic and sulfonic acids or react directly with
nucleophiles, making it an interesting compound in mechanistic
investigations of contact allergy. Urea derivatives are often
formed by metabolism from the corresponding thioureas, and
this was also observed in the current study as DPU (M3) was
identified as one of the metabolites. DPU was found in low
amounts and coeluted with a degradation product. The MS
subtraction of DPU at m/z 213 was only found in incubates
with NADPH. Ureas are not considered to be toxic and DPU
was found to be a nonsensitizer in the LLNA. The diphenyl-
formamidine metabolite (M4) was only found in the incubations
without trapping agents, indicating that it is a reactive metabolite
and probably involved in the formation of the sulfamidine
adducts A1 and A4 and the guanidine adduct A7. These
identified sulfamidine and guanidine adducts are mainly thought
to be formed by substitution reactions of sulfoxides from
oxidized DPTU. They were the major adducts formed in both
oxidative systems and constituted more than 40% of the
metabolites and adducts detected in the GSH-, MA-, and BA-
containing incubations, respectively (Table 4). These adducts
are formed via oxidation of the thiono group, suggesting this
Figure 4. Covalent binding of 14C-DPTU in human liver microsomes
(A) and the skinlike P450 cocktail (B). Error bars represent the standard
deviation of the mean of the replicates. The first bar in panel A is the
result from heat-inactivated enzymes. The other bars in both A and B
show the results from active enzymes but with or without NADPH (+
or -) and coincubations with the three reactive metabolite trapping
agents, MA, BA, and GSH.
However, neither DPTU nor DPU provoked sensitization
(stimulation indices of more than three) at the concentrations
tested. PITC and PIC were, as expected, both sensitizers with
EC3 values of 0.40 (30 mM) and 0.020% (1.7 mM), respec-
tively. PITC, previously assessed as an intermediate sensitizer
(11) in the GPMT, was found to be a strong sensitizer according
to LLNA. The more electrophilic isocyanate PIC, previously
identified as a sensitizer in the MEST method (45), was found
to be an extreme sensitizer using the LLNA. The SI values and
the test concentrations used are presented in Table 1.
Cross-Reactivity Study in Mice (MEST). The MEST
method was chosen as an in vivo method to elucidate a possible
cross-reactivity between the DPTU and the metabolite PITC,
as well as between PITC and PIC. Because of the low ability
of thioureas to induce sensitization in mice after topical
application, the mice were induced with the strongly sensitizing
metabolite PITC and challenged with the prohapten DPTU, in
an attempt to increase the sensitivity of the method. In PITC-
sensitized mice, PITC itself was able to induce an ear swelling
challenge response (Table 2). However, no increased ear
thickness was obtained after DPTU challenge in the PITC-
sensitized mice; thus, no cross-reactivity was observed. This is
probably due to the low penetration of DPTU as discussed
above. The cross-reactivity between PITC and PIC was probed
by challenging PITC-sensitized mice with PIC. An increased

Metabolism of Diphenylthiourea
Chem. Res. Toxicol., Vol. 24, No. 1, 2011 43
Scheme 2. Suggested Regeneration Mechanism of DPTU
from M1 via GSH Conjugation
product and thus be present in, for example, neoprene and PVC
products, as these are produced at high temperatures. Thus, it
is likely that people using DPTU-containing items are simul-
taneously exposed to DPTU and PITC.
The NADPH-dependent formation of metabolites and adducts
together with LLNA experiments show that DPTU can be
bioactivated into reactive intermediates able to cause skin
sensitization. However, which route of activation of DPTU
(metabolism or thermal instability) that is the most important
in developing contact allergy depends on the situation of
exposure. In real life, neoprene products are used for extended
periods of time, in close contact with the skin, and often under
occlusion, which enhances the skin absorption and risk of
sensitization. Diagnostic patch testing using DPTU in dermatitis
patients can cause false negative results, as the dose of the
metabolites becomes very small as the skin penetration of DPTU
is limited. To increase the possibilities of diagnosing contact
allergy to DPTU-containing items, we suggest that suitable
metabolites of DPTU should be used for screening testing. This
is an objective for future studies.
initial transformation to be of significant importance in the
development of contact allergy to DPTU. The identification of
adducts A6 and A8 (formed via BA trapping of PIC and PITC,
respectively) constitutes indirect evidence that the isocyanate
and the isothiocyanate were formed metabolically from DPTU.
The PITC adduct (A8) was also found in the NADPH-free
controls but in lower amounts as compared to the NADPH-
supplemented incubation mixtures. This is probably due to
decomposition of DPTU to PITC under the conditions used and
shows that DPTU forms reactive species also nonenzymatically.
The formation of protein complexes with haptens is a key
event in contact allergy. Covalent binding of prohaptens in
metabolism experiments in vitro has to our knowledge not been
studied in the field of contact allergy. The NADPH-dependent
adduct formation and covalent binding to proteins after incuba-
tion with DPTU found in this study make the activation of
DPTU into sulfoxides, PITC, and PIC in vivo likely. Several
thiourea-containing compounds have previously been shown to
form reactive sulfoxide intermediates with alkylating potential
of the metabolites (37, 47). The radioactivity observed in the
pellet from the NADPH-free control experiments is thought to
correspond to binding of PITC, formed from degradation of
DPTU (38). Although previous studies (37, 47) support the
opinion that thioureas need to be oxidized into unstable
intermediary metabolites to achieve an alkylating potential, it
cannot be excluded that DPTU also is inherently reactive toward
nucleophiles. Lower levels of protein binding were found in
the trapping experiments, since the trapping agents compete with
the nucleophilic amino acids in the proteins. GSH was found
to be the most efficient inhibitor of covalent binding. The fact
that only one GSH adduct was identified supports that a
NADPH-dependent recycling mechanism of DPTU takes place
(Scheme 2) (25); however, no disulfide GSH adduct of DPTU
was detected in the incubates. Overall, when studying prohaptens
in contact allergy, in vitro covalent binding assays could be of
future interest as a good complement to reactive metabolite
trapping studies with prohaptens.
Acknowledgment. We thank Petri Karhunen for skillful
technical assistance. At Astra Zeneca R&D, we thank Claire
Milne for the opportunity of collaboration and Cecilia Ericsson,
Isotope Chemistry team, for providing the radiolabeled DPTU.
We also thank Lars Weidolf at Astra Zeneca R&D and Lars G.
Nilsson for valuable discussions and for reading the manuscript
and Martin Gillstedt for statistical calculations. In addition, we
thank Astra Zeneca R&D for instrumentation. The work was
performed within the Centre for Skin Research, University of
Gothenburg.
Supporting Information Available: Figures of MS/MS
spectra for DPTU, M1, and M2 and adducts A2, A3, A5, A6,
and A8 including a table of the relative ear thickness results
from the irritation study. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) De Groot, A. C., and Frosch, P. J. (2001) Patch test concentrations
and vehicles for testing contact allergens. In Textbook of Contact
Dermatitis (Rycroft, R. J. G., Menne´, T., Frosch, P. J., and Lepoittevin,
J. P., Eds.) Springer, Berlin-Heidelberg.
(2) Alcantara, M., Martinez-Escribano, J., Frias, J., and Garcia-Selles, F. J.
(2000) Allergic contact dermatitis due to diphenylthiourea in a
neoprene slimming suit. Contact Dermatitis 43, 224–225.
(3) Aplin, C. G., Bower, C., Finucane, K., and Sansom, J. E. (2001)
Contact allergy to IPPD and diphenylthiourea in an orthopaedic brace.
Contact Dermatitis 45, 301–302.
(4) Boehncke, W. H., Wessmann, D., Zollner, T. M., and Hensel, O. (1997)
Allergic contact dermatitis from diphenylthiourea in a wet suit. Contact
Dermatitis 36, 271.
(5) Garcia-Morales, I., Garcia Bravo, B., and Camacho Martinez, F. (2003)
Occupational contact dermatitis caused by a personal-computer mouse
mat. Contact Dermatitis 49, 172.
Our interest in studying PITC as a metabolite of DPTU
stemmed from a previous study showing simultaneous patch
test reactions to PITC and DPTU in patients with suspected
rubber allergy (9). The results from the current metabolism
experiments showed that PITC is formed following bioactivation
of DPTU. Isothiocyanates are known sensitizers (2), and
accordingly, PITC was also found to be a strong sensitizer in
our study. Furthermore, it is known that isothiocyanates (PITC)
can be enzymatically converted to isocyanates (48-50), as was
indicated in the present investigation where PIC was identified
as an adduct in the oxidative systems. Experimental studies in
mice showed cross-reactivity between PITC and PIC, cor-
roborating the in vitro results. Because of the thermal instability
of DPTU (38), PITC could also be formed as a degradation
(6) Meding, B., Baum, H., Bruze, M., Roupe, G., and Trulsson, L. (1990)
Allergic contact dermatitis from diphenylthiourea in Vulkan heat
retainers. Contact Dermatitis 22, 8–12.
(7) Sakata, S., Cahill, J., and Nixon, R. (2006) Allergic contact dermatitis
to thiourea in a neoprene knee brace. Australas. J. Dermatol. 47, 67–
69.
(8) Villarreal Balza de Vallejo, O. (1997) Contact dermatitis from
diphenylthiourea in a knee brace. Contact Dermatitis 36, 166–167.
(9) Fregert, S., Trulson, L., and Zimerson, E. (1982) Contact allergic
reactions to diphenylthiourea and phenylisothiocyanate in PVC
adhesive tape. Contact Dermatitis 8, 38–42.
(10) Kanerva, L., Estlander, T., and Jolanki, R. (1994) Occupational allergic
contact dermatitis caused by thiourea compounds. Contact Dermatitis
31, 242–248.
(11) Fregert, S., Dahlquist, I., and Trulsson, L. (1983) Sensitization capacity
of diphenylthiourea and phenylisothiocyanate. Contact Dermatitis 9,
87–88.

44 Chem. Res. Toxicol., Vol. 24, No. 1, 2011
Samuelsson et al.
(12) Ikarashi, Y., Tsuchiya, T., and Nakamura, A. (1993) A sensitive mouse
lymph node assay with two application phases for detection of contact
allergens. Arch. Toxicol. 67, 629–636.
(31) Gad, S. C., Dunn, B. J., Dobbs, D. W., Reilly, C., and Walsh, R. D.
(1986) Development and validation of an alternative dermal sensitiza-
tion test: the mouse ear swelling test (MEST). Toxicol. Appl.
Pharmacol. 84, 93–114.
(32) Markwell, M. A. K., Haas, S. M., Bieber, L. L., and Tolbert, N. E.
(1978) A modification of the Lowry procedure to simplify protein
determination in membrane and lipoprotein samples. Anal. Biochem.
87, 207–210.
(33) Kimber, I., Hilton, J., Dearman, R. J., Gerberick, G. F., Ryan, C. A.,
Basketter, D. A., Scholes, E. W., Ladics, G. S., Loveless, S. E., House,
R. V., et al. (1995) An international evaluation of the murine local
lymph node assay and comparison of modified procedures. Toxicology
103, 63–73.
(34) Basketter, D. A., Lea, L. J., Dickens, A., Briggs, D., Pate, I., Dearman,
R. J., and Kimber, I. (1999) A comparison of statistical approaches to
the derivation of EC3 values from local lymph node assay dose
responses. J. Appl. Toxicol. 19, 261–266.
(35) Gad, S. C., Dunn, B. J., Dobbs, D. W., Reilly, C., and Walsh, R. D.
(1986) Development and validation of an alternative dermal sensitiza-
tion test: the mouse ear swelling test (MEST). Toxicol. Appl.
Pharmacol. 84, 93–114.
(36) Maryanoff, C. A., Stanzione, R. C., and Plampin, J. N. (1986)
Reactions of oxidized thioureas with amine nucleophiles. Phosphorus
Sulfur 27, 221–232.
(37) Stevens, G. J., Hitchcock, K., Wang, Y. K., Coppola, G. M., Versace,
R. W., Chin, J. A., Shapiro, M., Suwanrumpha, S., and Mangold, B. L.
(1997) In vitro metabolism of N-(5-chloro-2-methylphenyl)-N′-(2-
methylpropyl)thiourea: Species comparison and identification of a
novel thiocarbamide-glutathione adduct. Chem. Res. Toxicol. 10, 733–
741.
(38) Lizarraga, E., Zabaleta, C., and Palop, J. A. (2008) Mechanism of
thermal decomposition of thiourea derivatives. J. Therm. Anal.
Calorim. 93, 887–898.
(39) Lee, M. S. (1992) Oxidative conversion by rat liver microsomes of
2-naphthyl isothiocyanate to 2-naphthyl isocyanate, a genotoxicant.
Chem. Res. Toxicol. 5, 791–796.
(40) Lee, M. S. (1996) Enzyme induction and comparative oxidative
desulfuration of isothiocyanates to isocyanates. Chem. Res. Toxicol.
9, 1072–1078.
(41) Miller, A. E., Bischoff, J. J., and Pae, K. (1988) Chemistry of
aminoiminomethanesulfinic and -sulfonic acids related to the toxicity
of thioureas. Chem. Res. Toxicol. 1, 169–174.
(42) Poulsen, L. L., Hyslop, R. M., and Ziegler, D. M. (1979) S-
Oxygenation of N-substituted thioureas catalyzed by the pig liver
microsomal FAD-containing monooxygenase. Arch. Biochem. Biophys.
198, 78–88.
(43) OECD (2002) Guideline for the testing of chemicals. Skin sensitisation:
Local lymph node assay. OECD Guideline 429.
(44) Zatz, J. L. (1993) Modification of Skin Permeation by Solvents. In
Skin Permeation (Satz, J. L., Ed.) pp127-148, Allured Publishing
Corporation, Wheaton.
(45) Karol, M. H., and Kramarik, J. A. (1996) Phenyl isocyanate is a potent
chemical sensitizer. Toxicol. Lett. 89, 139–146.
(46) Merk, H. F. (2009) Drug skin metabolites and allergic drug reactions.
Curr. Opin. Allergy Clin. Immunol. 9, 311–315.
(13) Ikarashi, Y., Ohno, K., Momma, J., Tsuchiya, T., and Nakamura, A.
(1994) Assessment of contact sensitivity of four thiourea rubber
accelerators: Comparison of two mouse lymph node assays with the
guinea pig maximization test. Food. Chem. Toxicol. 32, 1067–1072.
(14) Kimber, I., Hilton, J., Botham, P. A., Basketter, D. A., Scholes, E. W.,
Miller, K., Robbins, M. C., Harrison, P. T., Gray, T. J., and Waite,
S. J. (1991) The murine local lymph node assay: Results of an inter-
laboratory trial. Toxicol. Lett. 55, 203–213.
(15) Smith, C. K. and Hotchkiss, S. A. M. (2001) Enzymes and pathways
of xenobiotic metabolism in skin. Allergic Contact Dermatitis:
Chemical and Metabolic Mechanisms, pp 89-117, Taylor and Francis,
London.
(16) Smith, P. B., and Crespi, C. (2002) Thiourea toxicity in mouse C3H/
10T1/2 cells expressing human flavin-dependent monooxygenase 3.
Biochem. Pharmacol. 63, 1941–1948.
(17) Smith, C. K., Moore, C. A., Elahi, E. N., Smart, A. T., and Hotchkiss,
S. A. (2000) Human skin absorption and metabolism of the contact
allergens, cinnamic aldehyde, and cinnamic alcohol. Toxicol. Appl.
Pharmacol. 168, 189–199.
(18) Merk, H. F., Baron, J. M., Heise, R., Fritsche, E., Schroeder, P., Abel,
J., and Krutmann, J. (2006) Concepts in molecular dermatotoxicology.
Exp. Dermatol. 15, 692–704.
(19) Kuby, J. (1994) Immunology, 2nd ed., W. H. Freeman Company, New
York.
(20) Guengerich, F. P. (2001) Common and uncommon cytochrome P450
reactions related to metabolism and chemical toxicity. Chem. Res.
Toxicol. 14, 611–650.
(21) Guengerich, F. P., and Shimada, T. (1991) Oxidation of toxic and
carcinogenic chemicals by human cytochrome P-450 enzymes. Chem.
Res. Toxicol. 4, 391–407.
(22) Bergstrom, M. A., Ott, H., Carlsson, A., Neis, M., Zwadlo-Klarwasser,
G., Jonsson, C. A., Merk, H. F., Karlberg, A. T., and Baron, J. M.
(2007) A skin-like cytochrome P450 cocktail activates prohaptens to
contact allergenic metabolites. J. InVest. Dermatol. 127, 1145–1153.
(23) Du, L., Neis, M. M., Ladd, P. A., Lanza, D. L., Yost, G. S., and
Keeney, D. S. (2006) Effects of the differentiated keratinocyte
phenotype on expression levels of CYP1-4 family genes in human
skin cells. Toxicol. Appl. Pharmacol. 213, 135–144.
(24) Evans, D. C., Watt, A. P., Nicoll-Griffith, D. A., and Baillie, T. A.
(2004) Drug-protein adducts: an industry perspective on minimizing
the potential for drug bioactivation in drug discovery and development.
Chem. Res. Toxicol. 17, 3–16.
(25) Ziegler, D. M. (1980) Microsomal Flavin-Containing Monooxygenase:
Oxygenation of Nucleophilic Nitrogen and Sulfur Compounds. In
Enzymatic Basis of Detoxification (Jakkoby, W. B., Ed.) Academic
Press, New York.
(26) Ziegler, D. M. (1988) Flavin-containing monooxygenases: catalytic
mechanism and substrate specificities. Drug. Metab. ReV. 19, 1–32.
(27) Guo, W. X., and Ziegler, D. M. (1991) Estimation of flavin-containing
monooxygenase activities in crude tissue preparations by thiourea-
dependent oxidation of thiocholine. Anal. Biochem. 198, 143–148.
(28) Onderwater, R. C., Commandeur, J. N., Menge, W. M., and Ver-
meulen, N. P. (1999) Activation of microsomal glutathione S-
transferase and inhibition of cytochrome P450 1A1 activity as a model
system for detecting protein alkylation by thiourea-containing com-
pounds in rat liver microsomes. Chem. Res. Toxicol. 12, 396–402.
(29) Wang, S., Gao, Q., and Wang, J. (2005) Thermodynamic analysis of
decomposition of thiourea and thiourea oxides. J. Phys. Chem. B 109,
17281–17289.
(47) Decker, C. J., and Doerge, D. R. (1992) Covalent binding of 14C-
and 35S-labeled thiocarbamides in rat hepatic microsomes. Biochem.
Pharmacol. 43, 881–888.
(48) Lee, M. S. (1996) Enzyme induction and comparative oxidative
desulfuration of isothiocyanates to isocyanates. Chem. Res. Toxicol.
9, 1072–1078.
(49) Lee, M. S. (1992) Oxidative conversion by rat liver microsomes of
2-naphthyl isothiocyanate to 2-naphthyl isocyanate, a genotoxicant.
Chem. Res. Toxicol. 5, 791–796.
(50) Lee, M. S. (1994) Oxidative conversion of isothiocyanates to isocy-
anates by rat liver. EnViron. Health Perspect. 102 (Suppl. 6), 115–118.
(30) Kimber, I., Hilton, J., Dearman, R. J., Gerberick, G. F., Ryan, C. A.,
Basketter, D. A., Scholes, E. W., Ladics, G. S., Loveless, S. E., House,
R. V., et al. (1995) An international evaluation of the murine local
lymph node assay and comparison of modified procedures. Toxicology
103, 63–73.
TX100241Z