In vivo studies on the metabolism of the monoterpenes S-(+)- and R-(-)-carvone in humans using the metabolism of ingestion-correlated amounts (MICA) approach

Article abstract of DOI:10.1021/jf010157q

The major in vivo metabolites of S-(+)- and R-(-)-carvone in a metabolism of ingestion correlated amounts (MICA) experiment were newly identified as α,4-dimethyl-5-oxo-3-cyclohexene-1-acetic acid (dihydrocarvonic acid), α-methylene-4-methyl-5-oxo-3-cycloh

Full text of DOI:10.1021/jf010157q

J. Agric. Food Chem. 2001, 49, 4069−4075
4069
In Vivo Stu d ies on th e Meta bolism of th e Mon oter p en es S-(+)- a n d
R-(-)-Ca r von e in Hu m a n s Usin g th e Meta bolism of
In gestion -Cor r ela ted Am ou n ts (MICA) Ap p r oa ch
Wolfgang Engel†
Deutsche Forschungsanstalt fu¨r Lebensmittelchemie, Lichtenbergstrasse 4, D-85748 Garching, Germany
The major in vivo metabolites of S-(+)- and R-(-)-carvone in a metabolism of ingestion correlated
amounts (MICA) experiment were newly identified as R,4-dimethyl-5-oxo-3-cyclohexene-1-acetic acid
(dihydrocarvonic acid), R-methylene-4-methyl-5-oxo-3-cyclohexene-1-acetic acid (carvonic acid), and
5-(1,2-dihydroxy-1-methylethyl)-2-methyl-2-cyclohexen-1-one (uroterpenolone) on the basis of mass
spectral analysis in combination with syntheses and NMR experiments. Minor metabolites were
identified as reduction products of carvone, namely, the alcohols carveol and dihydrocarveol. The
previously identified major in vivo metabolite in rabbits, 10-hydroxycarvone, could not be detected,
indicating either concentration effects or interspecies differences. Metabolic pathways for carvone
in humans including oxidation of the double bond in the side chain and, to a minor extent 1,2- and
1,4 + 1,2-reduction of carvone, are discussed. No differences in metabolism between S-(+)- and
R-(-)-carvone were detected.
Keyw or d s: Metabolism of ingestion correlated amounts (MICA); urinary metabolites; R,4-dimethyl-
5-oxo-3-cyclohexene-1-acetic acid; R-methylene-4-methyl-5-oxo-3-cyclohexene-1-acetic acid; 5-(1,2-
dihydroxy-1-methylethyl)-2-methyl-2-cyclohexen-1-one; carvone; carveol; dihydrocarveol
INTRODUCTION
terpene concentration might have an effect on the
structure of the metabolism product itself as previously
shown for pine oil metabolites in humans. In one study
cis- and trans-verbenol were found (8), and in another
case myrtenol, borneol, and bornyl acetate (9) appeared
to be the major metabolites.
Only a few studies have been undertaken to clarify
the metabolism of terpenes in man (10). These were
either focused on terpenes for use as drugs (11) or on a
special group of individuals with unusually high expo-
sure to terpenes (8, 12). In addition, Zlatkis and Liebich
(13) identified terpenes (carvone, limonene, and terpi-
neol) in the urine of unexposed persons, indicating a
considerable intake of terpenes via food and cosmetics.
To our knowledge no data are available for terpene
metabolism in humans focusing on terpene intake
corresponding to a normal diet. The only data available
refer to medical treatment, occupational exposure, or
accidental poisoning with high amounts of terpenes. The
purpose of the present study was, therefore, to obtain
initial insight into the in vivo metabolism of low doses
of carvone in humans. To obtain data as close as possible
to a daily intake situation, the metabolism of ingestion-
correlated amounts (MICA) approach was developed.
Carvone was chosen for this experiment because it is a
constituent of most essential oils and is widely used as
a flavor compound, for example, for toothpaste and other
cosmetics.
Terpenes such as citral, carvone, limonene, terpineol,
or cineol play an important role in food. They are
responsible for many distinct aromas, in particular those
of spices, and their amount is sometimes correlated with
the quality of the spice. Their occurrence in foods ranges
from some micrograms to even grams per kilogram.
After ingestion, terpenes are well absorbed, together
with fat, which acts as the carrier. It is generally
accepted that metabolism of terpenes starts with hy-
droxylation or epoxidation by microsomal monooxidases.
Primary metabolites formed are further transformed
into more polar compounds, such as diols or carboxylic
acids, and excreted via urine as highly water soluble
adducts with sulfuric acid or glucuronic acid or as
mercapturic acids.
Despite the ubiquitous occurrence and importance of
terpenes in food, little is known about the metabolic fate
of terpenes in mammals, especially humans. Most of the
data available have been obtained from rats (1-4),
rabbits (5, 6), or different antipodean species (7 and
references cited therein) feeding on terpene-rich euca-
lyptus diets. However, it remains unclear whether the
data from these studies can be applied to humans as
there may be significant interspecies differences in
metabolism. Another shortcoming is that the doses
applied to the test animals were much higher (up to
1000 mg/kg of body weight) compared to amounts
present in a normal diet. Often the animals have been
administered toxic doses of the terpene (6) to obtain
large quantities of metabolites, for example, for NMR
experiments. However, there is some evidence that
MATERIALS AND METHODS
Ch em ica ls. S-(+)-Carvone, R-(-)-carvone, carveol, dihydro-
carveol, N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA),
sodium chlorite, sodium hypochlorite solution, 2,2,6,6-tetra-
methylpiperidine-1-oxyl radical (TEMPO), and m-chloroper-
benzoic acid (MCPBA, 70%, technical grade) were from Fluka
† Telephone 0049 (0)89 28914375; fax 0049 (0)89 28914183;
e-mail wolfy@dfa.leb.chemie.tu-muenchen.de
10.1021/jf010157q CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/25/2001

4070 J. Agric. Food Chem., Vol. 49, No. 8, 2001
Engel
(Buchs, Switzerland). â-Glucuronidase H2 from Helix pomatia
(100000 units of â-glucuronidase and 4500 units of sulfatase
per milliliter) was from Sigma (Steinheim, Germany), and
carragheen type 1 was from Serva (Heidelberg, Germany).
Glycerol was from Merck (Darmstadt, Germany).
Develop m en t of th e MICA Exp er im en t. The study was
approved by the “Freie Ethikkommission Mu¨nchen”. Six
volunteers [three males (one smoker and two nonsmokers) and
three females (two smokers and one nonsmoker)] received a
controlled diet starting with lunch 24 h prior to the experiment
to avoid uncontrolled intake of terpenes. Any intake of food,
sporting activity, and urine volumes were documented by the
volunteers. The amounts were adjusted to the eating habits
of each participant but were observed strictly throughout the
following days of the experiment.
Id en tifica tion of Meta bolites. GC-MS runs of each frac-
tion of test and control urine (TF, AF, PF, and NF; derivatized
and underivatized) were recorded. Metabolites were localized
by mass trace comparison. Each mass trace starting from m/z
43 to 250 was compared. Substances that were absent in
control were considered to be metabolites. Structures were
proposed from the obtained mass spectral data and compared
to those of authentic samples either commercially available
or synthesized as described.
R efer en ce Com p ou n d s. Authentic standards of R,4-di-
methyl-5-oxo-3-cyclohexene-1-acetic acid (dihydrocarvonic acid),
R-methylene-4-methyl-5-oxo-3-cyclohexene-1-acetic acid (car-
vonic acid), 5-(1,2-dihydroxy-1-methylethyl)-2-methyl-2-cyclo-
hexen-1-one (uroterpenolone), 5-[1-(hydroxymethyl)ethenyl]-
2-methyl-2-cyclohexen-1-one (10-hydroxycarvone), and 2-methyl-
5-(1-methyl-1-oxiranyl)-2-cyclohexen-1-one (8,9-epoxycarvone)
were prepared according to the procedures described below.
All syntheses started from S-(+)- or R-(-)-carvone, retaining
stereochemistry at C-4 of the menthane skeleton.
The diet of each of the following days consisted of the
following food. In the morning, each participant received
chocolate muesli (50-150 g) with milk (75-200 mL); for lunch,
a light meal was prepared consisting of potatoes with a little
butter, chicken breast, and a salad with yogurt dressing; every
volunteer had to drink 500 mL of milk at lunchtime; in the
evening each participant was free to eat any of the following
foods: bread, rolls, milk products, and meat products entirely
free of spices. For drinks, milk, water, coffee, and tea were
allowed. No fruit juices were consumed throughout the experi-
ment. After 24 h of “adjustment”, within the next 24 h the
diet was continued and the 24 h urine (control) was collected.
Then a single dose of carvone (0.5 mmol, ∼1 mg/kg of body
weight) was ingested as a solution in full-fat milk (500 mL) at
lunchtime, and again a 24 h urine sample (test) was collected.
All urine samples were stored immediately in a freezer at -20
°C. Total diet period was 72 h. To avoid intake of terpenes
from toothpaste, a terpene-free toothpaste was prepared.
Tooth p a ste Recip e. Sodium dodecyl sulfate (SDS; 10 g)
was dissolved in water (119.6 g; solution 1). Carragheen type
1 (9 g) and ethanol (18 g) were mixed, and glycerol (180 g)
was added in small portions under vigorous stirring with a
mechanical stirrer (solution 2). The SDS solution 1 (99 g) was
added in small portions to solution 2, resulting in solution 3.
Finely powdered CaCO3 (90 g) was placed in a mortar, and
solution 3 (204 g) was added in small portions. After each
addition, the resulting mixture was homogenized thoroughly.
The toothpaste contained about 30% CaCO3, 20% water, 40%
glycerol, 4% ethanol, 2% SDS, and 2% carragheen type 1.
Isola tion of Meta bolites fr om Ur in e. The frozen urine
samples were warmed to room temperature. The pH of the
samples was between 6.1 and 6.5 with a total volume ranging
from 0.7 to 4.2 L. The differences between test and control
urine volume of an individual participant were in the range
of (5%. An aliquot of 100 mL was withdrawn and adjusted to
pH 5.0 with concentrated HCl. A solution of glucuronidase and
sulfatase from Helix pomatia (100 µL) was added to both
samples. The samples were hydrolyzed for 72 h at 37 °C and
adjusted to pH 2.0 with concentrated HCl, and the metabolites
were extracted with peroxide free diethyl ether [2 × 200 mL;
total fraction (TF)]. The acids were separated from TF by
washing with saturated NaHCO3 solution (3 × 40 mL) and
re-extracted with diethyl ether [3 × 50 mL; acidic fraction (AF)]
after careful acidification. Phenolic compounds were separated
from the organic phase by washing with 0.1 mol/L NaOH (3
× 40 mL) and re-extracted with diethyl ether [3 × 50 mL;
phenolic fraction (PF)] after acidification. The remaining ether
phase [∼150 mL; neutral fraction (NF)] contained the neutral
metabolites. All extracts were concentrated to ∼3 mL by means
of a rotary evaporator at 37 °C.
Syn th esis of r,4-Dim eth yl-5-oxo-3-cycloh exen e-1-a ce-
tic Acid (M1). M1 was synthesized in a three-step sequence.
2-Methyl-5-(1-methyl-1-oxiranyl)-2-cyclohexen-1-one (8,9-
Epoxycarvone). The synthesis was carried out according to a
published procedure (15): MS (EI), m/z (relative intensity) 109
(100), 108 (86), 82 (47), 91 (47), 39 (44), 79 (43), 54 (38), 107
(36), 53 (34), 97 (30), 123 (22).
R,4-Dimethyl-5-oxo-3-cyclohexene-1-acetaldehyde. To a rap-
idly stirred solution of 5-(1-methyl-1-oxiranyl)-2-methyl-2-
cyclohexen-1-one (8,9-epoxycarvone, 3 mmol; 0.5 g) in diethyl
ether (50 mL) was added dropwise BF3 (0.5 mL) in diethyl
ether. The dark red reaction mixture was stirred for 30 min
at room temperature. The mixture was washed with saturated
NaHCO3 solution (10 mL), dried over Na2SO4, and evaporated
to dryness. Yield was 0.35 g (70%) of crude aldehyde, which
was immediately used for the next reaction: MS (EI), m/z
(relative intensity) 108 (100), 109 (90), 82 (42), 79 (35), 80 (34),
54 (27), 39 (24), 81 (23).
R,4-Dimethyl-5-oxo-3-cyclohexene-1-acetic Acid (Dihydrocar-
vonic Acid). The synthesis was performed following a general
procedure given in ref 16 using crude 2-(4-methyl-5-oxo-3-
cyclohexenyl)propanal (∼1 mmol; 166 mg): yield 120 mg (66%);
elemental composition (HRMS), C10H14O3; 1H NMR (CDCl3)
CH3-CH (1.23, d, 3H), CH3-Cd (1.79, s, 3H), CH2,CH,CH,CH2
(2.2-2.6, m, 6H), -CHdC (6.78, m, 1H), -COOH (11.47, s,
1H); MS (EI), m/z (relative intensity) 109 (100), 74 (26), 82
(25), 54 (17), 108 (16), 79 (13), 81 (13), 39 (11); 13C NMR
(CDCl3, mixture of two diastereoisomers) CH3-Cd (16.71/
16.82), CH3-CH (18.49 signals unresolved), -CH-COOHd
(32.08/33.09), -CH-CH2 (40.66/40.72), -CH2-CHd (44.06/
45.07), -CH2-CdO (46.61/46.78), CH3-Cd (138.49 signals
unresolved), -HCdC- (147.67/147.75), -COOH (183.56/
183.76), dC-CdO (202.40/202.48).
Syn th esis of r-Meth ylen e-4-m eth yl-5-oxo-3-cycloh ex-
en e-1-a cetic Acid (M2). M2 was synthesized in a five-step
sequence.
5-[1-(Chloromethyl)ethenyl]-2-methyl-2-cyclohexen-1-one. The
synthesis was carried out according to that in ref 17 starting
with S-(+)-carvone or R-(-)-carvone (10 mmol), and the
product was further used without purification: MS (EI), m/z
(relative intensity) 149 (100), 82 (78), 142 (69), 39 (57), 107
(48), 105 (44), 93 (41), 91 (40).
Acetic Acid [2-(4-methyl-5-oxo-3-cyclohexenyl)-2-propenyl]
Ester. The crude 5-[1-(chloromethyl)ethenyl]-2-methyl-2-cy-
clohexen-1-one (estimated 8 mmol) was dissolved in tert-butyl
methyl ether (40 mL). After the addition of tributyl hexadecyl
phosphonium bromide (0.1 mmol) and sodium acetate (20
mmol) dissolved in water (10 mL), the mixture was heated
under reflux (56 °C) for 10 h. The aqueous phase was removed,
and again tributyl hexadecyl phosphonium bromide (0.1 mmol)
and sodium acetate (20 mmol) in water (10 mL) were added.
Further heating for 10 h led to total conversion of 5-[1-
(chloromethyl)ethenyl]-2-methyl-2-cyclohexen-1-one. Purifica-
tion by column chromatography (17) yielded 1.65 g (79% yield
Der iva t iza t ion P r oced u r es. Trimethylsilylation. Tri-
methylsilylation was performed using N,O-bis(trimethylsilyl)-
trifluoroacetamide (BSTFA). BSTFA (50 µL) was added to the
ethereal solution (150 µL) of each fraction and the mixture
kept at room temperature for 60 min.
Ethylation. Iodoethane was used as ethylation agent. The
ethereal urine sample (150 µL) was reacted with iodoethane
(50 µL) in the presence of K2CO3 (10 mg) and tributyl
hexadecyl phosphonium bromide (10 mg) overnight at room
temperature.

Metabolism of Carvone
J. Agric. Food Chem., Vol. 49, No. 8, 2001 4071
F igu r e 1. Localization of major metabolites (M1a , M1b, M2, M3a , and M3b) by mass trace comparison in the acidic fraction
(both samples TMS-derivatized).
based on carvone) of the ester: MS (EI), m/z (relative intensity)
148 (100), 43 (57), 82 (47), 91 (46), 106 (42), 105 (41), 54 (28),
108 (26).
5-[1-(Hydroxymethyl)ethenyl]-2-methyl-2-cyclohexen-1-one (10-
Hydroxycarvone). The synthesis was carried out according to
that in ref 17. A yield of 1.0 g (86%) of 5-[1-(hydroxymethyl)-
ethenyl]-2-methyl-2-cyclohexen-1-one was obtained: MS (EI),
m/z (relative intensity) 148 (100), 39 (70), 106 (70), 41 (59), 53
(56), 54 (56), 91 (50), 82 (49), 105 (48), 133 (18), 135 (16), 166
(10).
R-Methylene-4-methyl-5-oxo-3-cyclohexene-1-acetaldehyde. The
synthesis was performed according to that in ref 16. 5-[1-
(Hydroxymethyl)ethenyl]-2-methyl-2-cyclohexen-1-one (1 mmol;
166 mg) was used. The resulting crude aldehyde was im-
mediately used for further synthesis without any purification.
No byproducts were present as checked by HRGC-MS: MS
(EI), m/z (relative intensity) 108 (100), 135 (61), 164 (53), 82
(40), 54 (38), 39 (35), 121 (25), 91 (25).
tion of 2-methyl-5-(1-methyl-1-oxiranyl)-2-cyclohexen-1-one (5
mmol; 0.83 g) in THF (10 mL; freshly distilled) was added a
mixture of H2SO4 (200 mg) in water (10 mL). After 2 h at room
temperature, no epoxide was detectable by HRGC-MS. The
reaction mixture was diluted with water (20 mL) and adjusted
first to pH 6 by adding NaHCO3 and finally acidified to pH 3
by dropwise addition of 85% H3PO4. The aqueous mixture was
washed with diethyl ether (2 × 20 mL) to remove acidic and
neutral byproducts. The combined organic phases were washed
with water (5 mL; adjusted to pH 3 with 85% H3PO4), and the
combined aqueous phases were evaporated under reduced
pressure to yield the desired diol together with Na2SO4 and
NaH2PO4. The diol was separated from the salts by washing
the residue with a mixture of ethanol and diethyl ether (20
mL total volume; 50:50, v/v). Column chromatography on a
diol phase (Bakerbond Diol, 40 µm, pentane/diethyl ether
gradient) gave pure 5-(1,2-dihydroxy-1-methylethyl)-2-methyl-
2-cyclohexen-1-one (uroterpenolone) as a viscous colorless
liquid: yield 0.75 g (81%); 1H NMR (DMSO-d6) CH3-Cd (1.01,
s, 3H), CH3-C-OH (1.66, s, 3H), CH2,CH2,CH-ring (2.1-2.5,
m, 5H), -CH2-OH (3.16-3.35, m, 2H), -OH (4.21, s, 1H),
-OH (4.59, s, 1H), -CHdC- (6.83, m, 1H); 13C NMR (DMSO-
d6) CH3-Cd [15.21/15.23 (DEPT +)], CH3-C-OH [21.63/21.82
(DEPT +)], -CH2-CHd [26.15/26.58 (DEPT -)], -CH2-Cd
O [38.41/38.93 (DEPT -)], -CH-C-OH [40.78/41.02 (DEPT
+)], -CH2-OH [67.12/67.18 (DEPT -)], HO-C-CH2-OH
[72.01/72.05 (DEPT 0)], CH3-Cd [133.61/133.69 (DEPT 0)],
-HCdC- [145.87/146.15 (DEPT +)], dC-CdO [199.53/199.78
(DEPT 0)]; MS (EI), m/z (relative intensity) 109 (100), 110 (47),
153 (42), 43 (27), 75 (21), 135 (20), 95 (15), 103 (15).
R-Methylene-4-methyl-5-oxo-3-cyclohexene-1-acetic Acid (Car-
vonic Acid). The oxidation of R-methylene-4-methyl-5-oxo-3-
cyclohexene-1-acetaldehyde to the corresponding acid was
performed according to the procedures described in refs 18 and
19. The crude aldehyde was used without further purification,
and 2-methylbutene was used as chlorine scavenger. The
target compound R-methylene-4-methyl-5-oxo-3-cyclohexene-
1-acetic acid (105 mg; yield based on the alcohol 58%) was
obtained as a colorless liquid: elemental composition (HRMS),
C
10H12O3; 1H NMR (CDCl3) CH3-Cd (1.80, s, 3H), CH2,CH2-
ring (2.4-2.7, m, 4H), CH-ring (3.25, m, 1H), dCH′H′′ (5.71,
m, 1H), dCH′H′′ (6.44, m, 1H), -CH2-CHd (6.77, s, 1H),
-COOH (9.34, s, 1H); MS (EI), m/z (relative intensity) 162
(100), 54 (78), 134 (70), 82 (69), 39 (65), 135 (55), 109 (52), 53
(45), 91 (40), 180 (43), 41 (34), 108 (26). Possibly due to rapid
polymerization, only uninterpretable 13C NMR data were
obtained.
NMR Sp ectr oscop y. NMR spectra were recorded on a
Bruker AM 360 (Bruker, Karlsruhe, Germany) in CDCl3
(unless otherwise stated) with TMS as internal standard (δ )
0 ppm).
High -Resolu tion Ga s Ch r om a togr a p h y (HRGC)-Ma ss
Sp ectr om etr y (MS). HRGC was performed using a DB-5
capillary column (30 m × 0.32 mm; J &W Scientific, Folsom,
Syn th esis of 5-(1,2-Dih yd r oxy-1-m eth yleth yl)-2-m eth -
yl-2-cycloh exen -1-on e (M3). To a magnetically stirred solu-

4072 J. Agric. Food Chem., Vol. 49, No. 8, 2001
Engel
CA) in a gas chromatograph type 8000 (ThermoQuest, Egels-
bach, Germany). Sample injection was performed using the
split technique at a temperature of 230 °C and a split ratio of
1:25. The volumes injected ranged from 1 to 20 µL. The
temperature during HRGC was held at 35 °C for 1 min and
raised to a final temperature of 230 °C at a rate of 6 °C/min.
Mass spectra were recorded on an MD 800 quadrupole mass
spectrometer (ThermoQuest) in the electron ionization (EI)
mode at 70 eV.
HRGC-High -Resolu tion Ma ss Sp ectr om etr y (HRMS).
HRGC was performed on a DB-5 capillary column (30 m ×
0.25 mm; J &W Scientific) in a gas chromatograph type 5890
(Hewlett-Packard, Heilbronn, Germany). Sample injection was
performed using the cold on-column technique. The volumes
injected ranged from 0.2 to 0.5 µL. The temperature during
HRGC was held at 35 °C for 1 min and raised to 230 °C at a
rate of 6 °C/min. Mass spectra were recorded with a Finnigan
MAT 95S (Finnigan, Bremen, Germany) mass spectrometer
in the EI mode at 70 eV, in the chemical ionization (CI) mode
at 115 eV with isobutane as reagent gas, and in high-resolution
(HR) mode using perfluorokerosene as internal standard.
RESULTS AND DISCUSSION
Urine samples of six volunteers (three males, three
females) were collected for 24 h before (control) and after
(test) ingestion of 1 mg of carvone/kg of body weight as
described under Materials and Methods. Because it can
be assumed that the metabolites of carvone have been
condensed with either sulfuric acid or glucuronic acid
during phase II reactions, the samples were treated with
glucuronidase and sulfatase to liberate the apolar
metabolites presumably formed in phase I reactions.
After extraction with diethyl ether and separation into
acidic, phenolic, and neutral compounds, each fraction
was analyzed by HRGC-MS either before or after
trimethylsilylation or ethylation.
By comparison of single mass traces in all fractions
of underivatized control and test urine samples, two
metabolites were localized in the NF, none in the PH,
and three in the AF. The same results were obtained
for the trimethylsilyl (TMS) derivatized samples, indi-
cating complete localization of all extractable major
metabolites. Only a small amount of unmetabolized
carvone was found in all test samples.
Detection of Meta bolites in th e AF . In the TMS
derivatizated AF the presence of metabolites was indi-
cated best by m/z 162, 153, and 146 (Figure 1). Due to
the very low amount of carvone ingested in this MICA
experiment, none of the metabolites in AF would have
been detectable by comparison of the TIC alone (lowest
pair of chromatograms in Figure 1).
Meta bolite M1. Due to coelution with other com-
pounds present in the AF, no clear mass spectrum of
M1 could be obtained from the underivatized test
sample. The mass spectrum obtained for M1-TMS (TMS
derivatized metabolite 1) showed characteristic signals
at m/z 131 and 146 (Figure 2). From HRMS the
structure of the ion m/z 146 was in agreement with
propionic acid-TMS, probably resulting from a McLaf-
ferty rearrangement of the molecular ion. The presence
of m/z 109 indicated the unchanged ring structure of
carvone. On the basis of these suggestions, the structure
of M1 was proposed as R,4-dimethyl-5-oxo-3-cyclohex-
ene-1-acetic acid.
F igu r e 2. Mass spectra of (A) R,4-dimethyl-5-oxo-3-cyclohex-
ene-1-acetic acid trimethylsilyl ester (M1-TMS), (B) R,4-
dimethyl-5-oxo-3-cyclohexene-1-acetic acid ethyl ester (M1-
ET), and (C) R,4-dimethyl-5-oxo-3-cyclohexene-1-acetic acid
(M1).
HRGC-MS (Figure 2). The base peak m/z 109 corre-
sponds to the 2-methyl-2-cyclohexenone part of the
molecule. Ethylation or trimethylsilylation of synthe-
sized M1 gave spectra identical with the ones of the
urine metabolite (data not shown).
Compared to the synthetic route published (14) a new,
more direct synthetic approach for the preparation of
R,4-dimethyl-5-oxo-3-cyclohexene-1-acetic acid was de-
veloped (Figure 3). By analogy to the known limonene
metabolite, dihydroperillic acid, M1 was assigned as
dihydrocarvonic acid.
Ethylation of M1 (M1-ET) resulted in the mass
spectrum shown in Figure 2. The McLafferty rearrange-
ment product m/z 102 was the base peak suggested to
stem from propanoic acid ethyl ester. To confirm the
structure, M1 was synthesized and directly analyzed by
Both diastereoisomers of M1 were obtained by syn-
thesis due to the new asymmetric center introduced at
C-8 of carvone (Figure 3). These could be separated by
HRGC; however, no preparative separation was per-
formed because both diastereoisomers were present in

Metabolism of Carvone
J. Agric. Food Chem., Vol. 49, No. 8, 2001 4073
F igu r e 3. Synthesis of R,4-dimethyl-5-oxo-3-cyclohexene-1-
acetic acid (M1): (a) MCPBA/CH2Cl2/room temperature, yield
80%; (b) BF3 × Et2O/room temperature, yield 70%; (c) NaOCl,
TEMPO, tributyl hexadecyl phosphonium bromide/CH2Cl2,
H2O/0 °C, yield 66%.
F igu r e 5. Synthesis of R-methylene-4-methyl-5-oxo-3-cyclo-
hexene-1-acetic acid (M2): (a) NaOCl, CO2 (solid)/H2O,
CH2Cl2/room temperature, yield 80%; (b) NaOAc, tributyl
hexadecyl phosphonium bromide/TBME, H2O/reflux, yield
quantitative; (c) K2CO3/H2O/room temperature, yield 86%; (d)
NaOCl, TEMPO, NaBr/CH2Cl2, H2O/0 °C; (e) NaClO2/tBuOH,
H2O, methylbutene/room temperature, yield (d + e) 58%.
F igu r e 4. Mass spectrum of R-methylene-4-methyl-5-oxo-3-
cyclohexene-1-acetic acid.
F igu r e 6. Mass spectrum of trimethylsilylated 5-(1,2-dihy-
droxy-1-methylethyl)-2-methyl-2-cyclohexen-1-one (M3-TMS).
human urine (Figure 1). Metabolite M1 was mentioned
once in the literature (14); however, no NMR or MS data
were given.
ylene signals decreased considerably within 24 h, lead-
ing to new signals in the range of 1-3 ppm.
Meta bolite M2. Because metabolite M2 was present
in higher concentrations, a mass spectrum of underiva-
tized M2 could be obtained directly from test urine
(Figure 4). By MS-CI a molecular weight of 180 was
confirmed. By HRMS the elemental composition was
determined to be C10H12O3. The loss of water (M - 18)
forming the base peak m/z 162, however, is quite
unusual for a carboxylic acid, which normally loses an
OH radical, that is, forming M - 17. Because only one
diastereoisomer of M2 (Figure 1) was detected, the
double bond of the side chain must still be present,
leaving two possible structures for M2: one as suggested
in Figure 4 and another with the carboxyl group at C-7.
The latter structure, however, can be ruled out due to
the intense fragment m/z 82 indicating the unchanged
R-methyl-R,â-unsaturated keto moiety of carvone.
To confirm these assumptions, the proposed structure
was synthesized (Figure 5) yielding identical GC reten-
tion times, MS-EI, and MC-CI spectra as M2 isolated
from test urine. M2 was, therefore, identified as R-meth-
ylene-4-methyl-5-oxo-3-cyclohexene-1-acetic acid and is
named carvonic acid. This is the first report of this
compound, which has not been described previously.
Metabolite M2 seems to be somewhat unstable,
because during NMR analysis the concentrated solution
of the acid appeared to polymerize because both meth-
Meta bolite M3. Although present in the AF after
standard workup, M3 surprisingly turned out to be
nonacidic. The fragment m/z 109 in M3-TMS (Figure
6) indicated the presence of the unchanged ring struc-
ture of carvone. The fragment m/z 153 suggested a
sterically hindered underivatized 1-hydroxyethyl group
attached to the ring. The intense fragment m/z 73
indicated the presence of at least one TMS group and,
consequently, of a second less hindered hydroxy group.
Combining all structural features, the structure of M3
was proposed to be 5-(1,2-dihydroxy-1-methylethyl)-2-
methyl-2-cyclohexen-1-one (Figure 6). Its occurrence in
the AF could be explained by its high water solubility,
thus being extracted from the ether phase by the
aqueous solution of NaHCO3 used for removing the
acids. Its nonacidic character was confirmed by washing
TF with 1 mol/L H3PO4 instead of the NaHCO3 solution,
leading to nearly complete removal of M3 compared to
only traces of the acidic metabolites M1 and M2.
To avoid the formation of chlorinated products during
the synthesis (Figure 7) of 5-(1,2-dihydroxy-1-methyl-
ethyl)-2-methyl-2-cyclohexen-1-one, hydration of 8,9-
epoxycarvone was performed with diluted H2SO4 in a
mixture of THF and water. The use of ethanol as
suggested in ref 20 has to be avoided because substan-
tial amounts of ethoxy derivatives are produced. A

4074 J. Agric. Food Chem., Vol. 49, No. 8, 2001
Engel
F igu r e 7. Synthesis of 5-(1,2-dihydroxy-1-methylethyl)-2-
methyl-2-cyclohexen-1-one: (a) MCPBA/CH2Cl2/room temper-
ature, yield 80%; (b) H2SO4/THF, H2O/room temperature, yield
81%; (c) HCl/acetone, H2O.
F igu r e 8. Proposed metabolism scheme leading to unconju-
gated primary metabolites of carvone in humans.
sample of 5-(1,2-dihydroxy-1-methylethyl)-2-methyl-2-
cyclohexen-1-one (M3, Figure 7) prepared according to
this method and derivatized with BSTFA showed iden-
tical GC retention times and MS-EI and MC-CI spectra
as M3-TMS isolated from test urine. M3 was, therefore,
identified as 5-(1,2-dihydroxy-1-methylethyl)-2-methyl-
2-cyclohexen-1-one. In analogy to the limonene metabo-
lite uroterpenol, uroterpenolone is suggested as a trivial
name.
The synthetic procedure published for M3 (15) starts
with hydrolysis of epoxidized carvone (2 in Figure 7)
using hydrochloric acid in aqueous acetone. However,
when we repeated this step, 5-(2-chloro-1-hydroxy-1-
methylethyl)-2-methyl-2-cyclohexen-1-one (3, Figure 7),
resulting from the addition of hydrochloric acid to the
epoxide, was formed as the main product (data not
shown). Additional evidence for the proposed structure
was obtained by reaction of 3 with base, leading to
recovery of the epoxide (2). This behavior is typical for
1-hydroxy-2-chloro compounds, whereas diols cannot be
converted into epoxides by simple treatment with a
base.
Furthermore, standard NMR techniques have to be
selected carefully to distinguish unequivocally between
5-(2-chloro-1-hydroxy-1-methylethyl)-2-methyl-2-cyclo-
hexen-1-one (3) and 5-(1,2-dihydroxy-1-methylethyl)-2-
methyl-2-cyclohexen-1-one (M3). In the case of H-D
exchange with a deuterated protic solvent such as D2O
or CD3OD, the remaining 1H and the observed 13C NMR
signals can be assigned to both compounds because the
chemical shifts differ only slightly and the OH signals
are absent. In the case of an aprotic deuterated solvent
such as CDCl3, the signal for OH protons is often broad,
preventing proper integration and thus clear discrimi-
nation between 3 and M3.
To prove the presence of two different hydroxy groups,
the NMR data were obtained from a sample dissolved
in DMSO-d6, yielding both OH signals as two sharp
singlets. It has to be mentioned that M3 is entirely
insoluble in CDCl3 and, therefore, the previously pub-
lished NMR data (15) obtained from a sample dissolved
in this solvent are questionable.
not further investigated; however, other diastereoiso-
mers of M4 and M5 were detectable to a minor extent,
indicating that metabolic reduction of carvone is not
stereospecific.
From the structures of the metabolites of carvone
identified in this study it is obvious that R-hydroxylation
or epoxidation of a double bond does not seem to be
possible if there is a carbonyl group in conjugation.
Therefore, in carvone, obviously only the double bond
in the side chain is oxidized. This behavior is consistent
with the chemical reactivity of both double bonds in
carvone toward epoxidation by peracids forming exclu-
sively the 8,9-epoxide. It cannot be excluded that the
epoxide is generated in vivo as a primary product,
followed by further oxidative or hydrolytic transforma-
tions. However, no epoxide was found in the urines.
In rabbits only 10-hydroxycarvone (6) was identified
as the key metabolite of (()-carvone. Furthermore, only
the (+)-isomer of 10-hydroxycarvone was detected after
the application of racemic carvone, suggesting different
metabolisms of (+)- and (-)-carvone or isomerization
of (-)-carvone to the (+)-isomer prior to metabolism. In
our experiment using much lower doses of carvone all
metabolites were identical after the application of either
S-(+)-carvone or R-(-)-carvone. Additionally, no quali-
tative differences were observed among the six persons.
Because the stereochemistry of the metabolites still has
to be determined, differences between R-(-)- and S-(+)-
carvone metabolism cannot be ruled out. However, it
seems more likely that stereochemistry at C4 of carvone
(1, Figure 3) is not affected by metabolism.
The alcohol, 10-hydroxycarvone, identified in rabbits
(6) might be an intermediate in humans. It may be
formed in the first stage of oxidative metabolism (Figure
8), being further transformed to carvonic acid (M2).
However, further oxidation seems to be very efficient
in humans, because 10-hydroxycarvone was not de-
tected. The observed differences between rabbits and
humans may also be explained by a simple concentra-
tion effect; the doses applied were 1 mg/kg of body mass
in this MICA experiment and 700 mg/kg in the case of
rabbits (6).
Meta bolites in th e NF . In the NF, two compounds
were identified, namely, carveol (M4) and dihydrocar-
veol (M5). The stereochemistry of both metabolites was
It remains to be determined whether the unsaturated
carboxylic acid M2 is finally reduced to M1 or if there

Metabolism of Carvone
J. Agric. Food Chem., Vol. 49, No. 8, 2001 4075
(7) McLean, S.; Foley, W. J .; Davies, N. W.; Brandon, S.;
Duo, L.; Blackman, A. J . Metabolic fate of dietary
terpenes from Eucalyptus radiata in common ringtail
possum. J . Chem. Ecol. 1993, 19, 1625-1643.
(8) Eriksson, K. A.; Levin, J . O. Identification of cis- and
trans-verbenol in human urine after occupational ex-
posure to terpenes. Int. Arch. Occup. Environ. Health
1990, 62, 379-383.
(9) Ko¨ppel, C.; Tenczer, J .; To¨nnesmann, U.; Schirop, T.;
Ibe, K. Acute poisoning with pine oilsmetabolism of
monoterpenes. Arch. Toxicol. 1981, 49, 73-78.
(10) Angerer, J .; Kra¨mer, A. Indoor air pollution with mono-
terpenes: Current knowledge and strategies for expo-
sure assessment (in German). Umweltmed. Forsch.
Prax. 1997, 1, 33-37.
(11) Ventura, P.; Schiavi, M.; Serafini, M.; Selva, A. Further
studies of trans-sobrerol metabolism: rat, dog and
human urine. Xenobiotica 1985, 15, 1317-1325.
(12) Eriksson, K. A.; Levin J . O.; Sandstro¨m, T.; Lindstro¨m-
Espeling, K.; Linden, G.; Stjernberg, N. L. Terpene
exposure and respiratory effects among workers in
Swedish joinery shops. Scand. J . Work Environ. Health
1997, 23, 114-120.
(13) Zlatkis, A.; Liebich, H. M. Profile of volatile metabolites
in human urine. Clin. Chem. 1971, 17, 592-594.
(14) Teresa, J . P.; Mateos, A. F.; Gonzalez, R. R. The
synthesis of 2-hydroxy-4-isopropyl-7-methoxy-1,6-di-
methyl-naphthalene, “chemical precursor” of the byssi-
notic agent from cotton. Tetrahedron Lett. 1982, 23,
3405.
is an independent pathway leading directly from car-
vone to M1 (Figure 8) as mimicked during synthesis
(Figure 3). In the first step during in vivo metabolism,
carvone might also be converted to the epoxide and after
rearrangement to the saturated aldehyde might be
finally oxidized to M1, making this route an alternative
to the reduction pathway starting with M2.
The neutral metabolites carveol (M4) and dihydro-
carveol (M5) might be formed by a competing 1,2- and
1,4-reduction of carvone (Figure 8). The enolized product
of the 1,4-reduction, dihydrocarvone, might be further
reduced to dihydrocarveol. However, even though no
exact quantification was performed, the amounts of M4
and M5 were apparently much lower than those of M1,
M2, and M3, suggesting that reductive metabolism is
only a side pathway.
The results suggest that low-dose metabolic experi-
ments in humans, as it is implied in MICA, are the most
realistic and thus the best model available for evalua-
tion of in vivo metabolism in humans. Further studies
with carvone site-specifically labeled with stable iso-
topes are in progress to gain more insight into the
oxidation mechanism.
ACKNOWLEDGMENT
I thank Bettina Fickert, Ines Otte, Valerie Rota,
Michael Czerny, and Magnus J ezussek for being patient
volunteers and Peter Schieberle for helpful comments
during revision of the manuscript. I gratefully acknowl-
edge the support of Edith Kaldyk (NMR), Oliver Frank
(NMR), Ines Otte (MS), and J ulius Scho¨nauer (MS).
(15) Pellegata, R.; Dosi, I.; Ventura, P.; Villa, M.; Lesma, G.;
Palmisano, G. Monoterpenoid chemistry. Part 3. Ste-
reoselective synthesis of the major oxygenated metabo-
lites of trans-sobrerol. Helv. Chim. Acta 1987, 70, 71-
78.
(16) Anelli, P. L.; Biffi, C.; Montanari, F.; Quici, S. Fast and
selective oxidation of primary alcohols to aldehydes or
to carboxylic acids and of secondary alcohols to ketones
mediated by oxammonium salts under two-phase condi-
tions. J . Org. Chem. 1987, 52, 2559-2562.
(17) Weinges, K.; Schwarz, G. Radical-type cyclization of
dienes, VI. Substrate-controlled asymmetric synthesis
of (3aS,6aR)-(+)-3,3a,6,6a-tetrahydro-2H-cyclopenta[b]-
furan-2-one (in German). Liebigs Ann. Chem. 1993,
811-814.
(18) Lindgren, B. O.; Nilsson, T. Preparation of carboxylic
acids from aldehydes (including hydroxylated benzal-
dehydes) by oxidation with chlorite. Acta Chem. Scand.
1973, 27, 888-890.
LITERATURE CITED
(1) Ventura, P.; Schiavi, M.; Serafini, S. The metabolism
of trans-sobrerol in the rat. Xenobiotica 1983, 13, 139-
146.
(2) Ventura, P.; Pellegata, R.; Schiavi, M.; Serafini, S.
Biotransformation of trans-sobrerol. III. Metabolites of
8-hydroxycarvonacetone in the rat. Xenobiotica 1986, 16,
317-323.
(3) Diliberto, J . J .; Srinivas, P.; Overstreet, O.; Usha, G.;
Burka, L. T.; Birnbaum, L. S. Metabolism of citral, an
R,â-unsaturated aldehyde, in male F344 rats. Drug
Metab. Dispos. 1990, 18, 866-875.
(4) Kouzi, S. A.; McChesney, J . D.; Walker, L. A. Identifica-
tion of four bilinary metabolites of the diterpene sclareol
in the laboratory rat. Xenobiotica 1993, 23, 621-632.
(5) Asakawa, Y.; Ishida, T.; Toyota, M.; Takemoto, T.
Terpenoid biotransformation in mammals IV. Biotrans-
formation of (+)-longifolene, (-)-caryophyllene, (-)-
caryophyllene oxide, (-)-cyclocolorenone, (+)-nootka-
tone, (-)-elemol, (-)-abietic acid and (+)-dehydroabietic
acid in rabbits. Xenobiotica 1986, 16, 753-767.
(6) Ishida, T.; Toyota, M.; Asakawa, Y. Terpenoid biotrans-
formation in mammals. V. Metabolism of (+)-citronellal,
(()-7-hydroxycitronellal, citral, (-)-perillaldehyde, (-)-
myrtenal, cuminaldehyde, thujone, and (()-carvone in
rabbits. Xenobiotica 1989, 19, 843-855.
(19) Bal, B. S.; Childers, W. E.; Pinnick, H. W. Oxidation of
R,â-unsaturated aldehydes. Tetrahedron 1981, 37, 2091-
2096.
(20) Thele, V.; Go¨ckeritz, D. Identification of 8,9-dihydroxy-
carvotanacetone and 8-hydroxycarvotanacetone in aque-
ous oxygen-containing carvone solutions (in German).
Pharmazie 1981, 36, 866-867.
Received for review February 7, 2001. Revised manuscript
received April 9, 2001. Accepted J une 12, 2001.
J F010157Q