Biosynthesis of menthofuran in Mentha x piperita: Stereoselective and mechanistic studies

Article abstract of DOI:10.1021/jf990439e

Mentha x piperita shoot tips and first leaf pair were fed with aqueous solutions of [²H₂]- and [²H₂]/[¹⁸O]-labeled pulegone. The essential oil was analyzed by solid phase microextraction and enantioselective multidimensional gas chromatography/mass spectrometry. After feeding experiments with labeled pulegone racemate, both labeled (S)-menthofuran and (R)-menthofuran were detectable simultaneously together with genuine (R)- menthofuran. It could be shown that both labeled pulegone enantiomers are converted by Mentha x piperita to the corresponding labeled menthofuran enantiomers, favoring the labeled analogue of the nongenuine (S)-pulegone. The oxygen in menthofuran is introduced by enzymatic oxidation of pulegone, as concluded from feeding experiments with mixed labeled [²H₂]/[¹⁸O]pulegone.

Full text of DOI:10.1021/jf990439e

4100
J. Agric. Food Chem. 1999, 47, 4100−4105
Biosyn th esis of Men th ofu r a n in Men th a × piper ita : Ster eoselective
a n d Mech a n istic Stu d ies
Sabine Fuchs, Silvia Zinn, Thomas Beck, and Armin Mosandl*
Institut fu¨r Lebensmittelchemie, J ohann Wolfgang Goethe-Universita¨t Frankfurt, Marie-Curie-Strasse 9,
D-60439 Frankfurt (Main), Germany
Mentha × piperita shoot tips and first leaf pair were fed with aqueous solutions of [2H2]- and [2H2]/
[18O]-labeled pulegone. The essential oil was analyzed by solid phase microextraction and enantio-
selective multidimensional gas chromatography/mass spectrometry. After feeding experiments with
labeled pulegone racemate, both labeled (S)-menthofuran and (R)-menthofuran were detectable
simultaneously together with genuine (R)-menthofuran. It could be shown that both labeled pulegone
enantiomers are converted by Mentha × piperita to the corresponding labeled menthofuran
enantiomers, favoring the labeled analogue of the nongenuine (S)-pulegone. The oxygen in
menthofuran is introduced by enzymatic oxidation of pulegone, as concluded from feeding
experiments with mixed labeled [2H2]/[18O]pulegone.
Keyw or d s: Menthofuran; stable isotope labeling; Mentha × piperita biosynthesis; solid-phase
microextraction (SPME); enantioselective multidimensional gas chromatography/ mass spectrometry
(enantio-MDGC/ MS)
INTRODUCTION
MATERIALS AND METHODS
Menthofuran is a characteristic monoterpene in Men-
tha with an undesirable influence on mint essential oil
quality. The oxidative bioconversion of pulegone leading
to menthofuran was assumed by Reitsema (1958) and
confirmed by Battaile and Loomis (1961), using incuba-
tion of very young leaf slices with labeled pulegone.
Although there have been several in vivo and in vitro
investigations on the conversion of pulegone into men-
thofuran in mammals (McClanahan et al., 1988; Moor-
thy et al., 1989; Nelson et al., 1992a,b; Madyastha and
Gaikwad, 1998), little recent research has been carried
out on the biosynthetic origin of menthofuran in Mentha
species. As in mammals, a cytochrome P-450 oxidase is
assumed to be involved in the biosynthesis of mentho-
furan. The cis-methyl group of the pulegone isopropyl-
idene moiety may be hydroxylated, followed by an
intramolecular cyclization and dehydration (Croteau
and Gershenzon, 1994; McCaskill and Croteau, 1997).
Genetic aspects of the menthofuran biosynthesis as well
as environmental effects on the menthofuran content
have been studied (Murray and Hefendehl, 1972; Bur-
bott and Loomis, 1967; Clark and Menary, 1980).
En a n tioselective Ga s Ch r om a togr a p h y (En a n tio-GC).
Enantioseparation of the synthesized labeled pulegone was
performed on an HP 5890 Series II gas chromatograph,
equipped with a duran glass capillary (30 m × 0.23 mm i.d.)
coated with 15% heptakis(2,3-di-O-methyl-6-O-tert-butyldi-
methylsilyl)-â-cyclodextrin in SE 52 (film thickness ) 0.1 µm).
Conditions were as follows: carrier gas, hydrogen, 130 kPa;
split, 30 mL/min; injector temperature, 250 °C; detector, FID,
260 °C; oven temperature 80 °C, raised at 1 °C/min to 200 °C
(30 min isothermal).
GC/MS. A Fisons Instruments GC 8065, coupled to a Fisons
Instruments MD 800 mass spectrometer, equipped with a
BPX-5 column (50 m × 0.32 mm i.d., film thickness ) 0.25
µm; SGE-Analytik, Weiterstadt, Germany) was employed for
the GC/MS analysis of the synthesized monoterpenes. Condi-
tions were as follows: carrier gas, helium, 109 kPa; split, 30
mL/min; injector temperature, 230 °C; oven temperature, 40
°C (5 min isothermal), raised at 2.5 °C/min to 320 °C (30 min
isothermal); ion source temperature, 200 °C; interface tem-
perature, 250 °C; mass range, 40-250 amu; EI, 70 eV. The
molecular ion (M+) and the fragmentation ions were given as
m/z with relative peak intensities to the base peak (percent).
En a n tio-MDGC/MS. A Siemens SiChromat 2, equipped
with two independent column oven programs and a live-T-
switching device, was used for the enantio-MDGC/MS analysis
of the SPME headspace extracts. The main column was
coupled to the transfer line of a Finnigan MAT ITD 800, using
an open split interface. Precolumn conditions were as follows:
duran glass capillary (30 m × 0.23 mm i.d.) coated with a 0.23
µm film of SE 52; carrier gas, hydrogen, 120 kPa; split, 8 mL/
min, injector temperature, 220 °C; detector, FID, 250 °C; oven
temperature, 50 °C (5 min isothermal), raised at 5 °C/min to
250 °C (49 min isothermal); cut times 24.5-28.5 min. Main
column conditions were as follows: duran glass capillary (30
m × 0.23 mm i.d.) coated with a 0.23 µm film of 50% octakis-
(2,3-di-O-butyryl-6-O-tert-butyldimethylsilyl)-γ-cyclodextrin in
OV-1701vi (Schmarr, 1992); carrier gas, hydrogen, 100 kPa;
oven temperature, 40 °C (30 min isothermal), raised at 0.5 °C/
Recently, enantioseparation by gas chromatography
was reported by Werkhoff et al. (1995). As expected,
they found enantiopure (R)-menthofuran in Mentha
piperita oil.
This paper reports on in vivo feeding experiments
with deuterium as well as deuterium and 18O-labeled
pulegone precursors using intact plant material of
Mentha × piperita. The essential oil of each single shoot
tip and first leaf pair was analyzed by solid-phase
microextraction (SPME) and enantioselective multidi-
mensional gas chromatography/mass spectrometry (enan-
tio-MDGC/MS).
10.1021/jf990439e CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/24/1999

Biosynthesis of Menthofuran in Mentha × piperita
J. Agric. Food Chem., Vol. 47, No. 10, 1999 4101
Sch em e 1. Syn th esis of Differ en t Deu ter iu m - a n d
moles (470 mg) of crude 6 was added to a solution containing
6 mL of CH3OH, 5 mL of H2O, and 4 mmol (92 mg) of sodium
metal. The solution was allowed to stand at room temperature
for 63 h. Crude 7 (1.4 mmol, 220 mg) was purified by flash
chromatography, yielding 0.3 mmol (48 mg) of pure 7: for
conditions see above; MS 158 (M+, 44), 140 (14), 115 (20), 112
(16), 81 (100), 70 (24); 1H NMR δ 1.00 (d, 3H, 7-H, J ) 6.5
Hz), 1.33 (m, 1H, 6a-H), 1.86 (m, 1H, 6e-H), 1.95-2.25 (m, 2H,
1-H, 2a-H), 2.24 (m, 1H, 5a-H), 2.49 (dd, 1H, 2e-H, J ) 11.0
Hz, J ) 2 Hz), 2,70 (dt, 1H, 5e-H, J ) 15.5 Hz, J ) 4.5 Hz).
Traces of 9-H and 10-H were found at δ 1.75 and 2.00,
respectively. Deuterium contents were [2H5] 25% and [2H6]
75%.
18O-La beled Ra cem ic P u legon es (6-8)
Synthesis of5-(R/S)-2-[[1-2H3]Methyl[2,2,2-2H3]ethylidene]-
5-m eth yl-[18O]cycloh exa n on e (d6-[18O]P u legon e) (8). Com-
pound 8 was synthesized according to the method of Vederas
(1980) with modifications. Six millimoles (93 mg) of crude 7
was added to a solution containing 1 mL of 0.3 N HCl/dry THF
and 80 µL of H218O (95.7% atom 18O, 0.6% atom 17O, 3.7% atom
16O; Euriso-top, Saint Aubin Cedex, France). After standing
at room temperature for 70 h, the solutions was diluted with
diethyl ether. The organic layer was extracted with brine and
dried over sodium sulfate. A second step was performed. The
resulting crude 8 was purified by preparative thin-layer
chromatography (TLC). Chromatographic conditions were as
follows: silica gel, 60 F 254 (Merck); mobile phase pentane/
diethyl ether 10:1 (v:v); detection, UV 254 nm; Rf 0.4; yield,
19 mg (0.1 mmol); MS 160 (M+, 35), 142 (14), 115 (19), 112
(17), 81 (100), 72 (36); 1H NMR δ 1.00 (d, 3H, 7-H, J ) 6.5
Hz), 1,13 (m, 1H, 6a-H), 1.86 (m, 1H, 6e-H), 1.95-2.05 (m, 2H,
1-H, 2a-H), 2.24 (m, 1H, 5a-H), 2.49 (dd, 1H, 2e-H, J ) 11 Hz,
J ) 2 Hz), 2.71 (dt, 1H, 5e-H, J ) 15.5 Hz, J ) 4.5 Hz).
Deuterium and 18O contents were [2H5,16O] 5%, [2H6,16O] 12%,
[2H5,18O] 35%, and [2H6,18O] 48%.
Sch em e 2. Syn th esis of (S)-Men th ofu r a n (3) Sta r tin g
fr om (S)-Citr on ellol (1)
min to 60 °C (0 min isothermal), raised at 2.5 °C/min to 120
°C; detector, ITD 800, transfer line 250 °C; open split interface,
250 °C; helium sweeping flow, 1 mL/min; ion trap manifold,
195 °C; EI, 70 eV.
1H Nu clea r Ma gn etic Reson a n ce (NMR). A Bruker ARX
300, at 300 MHz, was employed for recording the 1H NMR
spectra. CDCl3 was used as solvent. Abbrevations are as
follows: s, singlet; d, doublet; t, triplet; m, multiplet; J , spin-
spin coupling constant (Hz); a, axial; e, equatorial. The terpene
nomenclature, given in Schemes 1 and 2, was used for
assignment.
Syn th esis of (6S)-3,6-Dim eth yl-4,5,6,7-tetr a h yd r oben -
zo[b]fu r a n [(S)-Men th ofu r a n ] (3). Isopulegon (2) was syn-
thesized according to the method of Corey et al. (1976) with a
modified cleanup; menthofuran (3) was synthesized according
to the method of Friedrich and Bohlmann (1988). Pyridinium
chlorochromate (PCC; 8.9 mmol, 1.93 mg) was dissolved in 11
mL of dry methylene chloride. (S)-Citronellol (1; 430 mg, 2.7
mmol) (Fluka) was added. It was stirred for 36 h, pentane/
diethyl ether 1:1 (v:v) was added, and the suspension was
filtered (Celite). The solution was washed with 10% HCl, 10%
NaHCO3, and H2O and dried over sodium sulfate. The crude
product isopulegone (2) (1.5 mmol, 230 mg) was dissolved in
10 mL of dry benzene; 2.6 mmol (450 mg) of 85% m-
chloroperbenzoic acid was added, and the solution was stirred
for 24 h. The benzene was evaporated, and the residue was
taken up in petrol and filtered. After evaporation of the
solvent, 5.9 mL of methanol and 3.2 mL of 40% KOH solution
were added, and the solution was stirred for 4 h at room
temperature. Then diethyl ether was added to the solution.
After extraction with water (twice) and brine (once), the
solution was dried over sodium sulfate. Purification by flash
chromatography (conditions see above) yielded 0.07 mmol (11
Synthesis of5-(R/S)-2-[[1-2H3]Methyl[2,2,2-2H3]ethylidene]-
5-m eth yl-[6,6-2H2]cycloh exa n on e (d 8-P u legon e) (6). Pule-
gone was synthesized according to the method of Gibson
(1983). Three millimoles (470 mg) of (R)-pulegone and 3 mmol
(470 mg) of (S)-pulegone (both from Fluka, Deisenhofen,
Germany) were added to a solution containing 3 mL of D2O, 6
mL of CH3OD, and 4 mmol (92 mg) of sodium metal. The
mixture was heated under reflux for 1.5 h, subsequently cooled,
diluted with diethyl ether, and extracted twice with water and
once with brine. After drying over sodium sulfate, the diethyl
ether was removed in a vacuum. The crude product was used
for two further H/D-exchange reactions under identical condi-
tions. After three reaction steps, 280 mg of crude d8-pulegone
was purified by flash chromatography. Chromatographic
conditions were as follows: column diameter, 20 mm; silica
gel, 30-60 µm (Baker 7024-01); eluent pentane/diethyl ether
10:1 (v/v). After removal of the eluent, 0.07 mmol (11 mg) of
pure 6 was obtained: MS 160 (M+, 66), 142 (16), 117 (28), 114
(24), 82 (89), 70 (76); 1H NMR δ 1.00 (d, 3H, 7-H, J ) 6.5 Hz),
1.32 (m, 1H, 6a-H), 1.86 (m, 1H, 6e-H), 1.95 (m, 1H, 1-H), 2.25
(m, 1H, 5a-H), 2.70 (dt, 1H, 5e-H, J ) 15.5 Hz, J ) 4.5 Hz).
No traces of 2e-H, 2a-H, 9-H, and 10-H were found at δ 2.50,
1.8-2.0, 1.79, and 2.00, respectively (Tori et al., 1975; Bam-
bridge et al., 1995). Enantiomeric distribution was R/S ) 1/1
(enantio-GC). Deuterium contents were [2H7] 23% and [2H8]
77%, calculated by MS data. Enantiopure deuterium-labeled
[2H8]pulegone was synthesized starting from enantiopure
pulegone in one reaction step.
1
mg) of pure 3: MS 150 (M+, 38), 108 (100), 79 (29); H NMR
δ 1.07 (d, 3H, 7-H, J ) 6.5 Hz), 1.2-1.4 (m, 1H, 6-H), 1.8-1.9
(m, 2H, 1-H, 6-H) 1.92 (s, 3H, 9-H), 2.1-2.4 (m, 3H, 2-H, 5-H),
2.64 (dd, 1H, 2-H, J ) 11.0 Hz, J ) 5.0 Hz), 7.03 (s, 1H, 10-H)
(Friedrich and Bohlmann, 1988).
Syn th esis of (5S)-2-Isop r op ylid en -5-[2H3]m eth yl-[4,4-
2H2]cycloh exa n on e [d 5-(S)-P u legon e] (9). 9 was prepared
according to the method of Fuchs et al. (1999). Enantiomeric
distribution was S/R ) 92:8.
P la n t Ma ter ia l. Cuttings from Mentha × piperita were
kindly provided by the Botanical Garden of Frankfurt Uni-
versity. The plants selected for the experiments were 5-7 cm
high with 4-6 leaf pairs.
F eed in g Exp er im en ts. Precursor Solution. A solution of
the labeled pulegones (6-9) with a concentration of 0.1 mg/
mL was prepared by dissolving the same amounts of 6-9 and
Tween 20 in distilled water.
Synthesis of5-(R/S)-2-[[1-2H3]Methyl[2,2,2-2H3]ethylidene]-
5-m et h ylcycloh exa n on e (d 6-P u legon e) (7). Three milli-

4102 J. Agric. Food Chem., Vol. 47, No. 10, 1999
Fuchs et al.
F igu r e 1. Mass spectra of deuterium-labeled menthofurans obtained from feeding experiments with different deuterium-labeled
pulegones. Retro-Diels-Alder fragmentation is proposed.
(S)-Menthofuran (3) (Internal Standard, IST) Solution. The
solution contained 0.1 mg/mL Tween 20 and 0.01 mg/mL (S)-
menthofuran. For feeding experiments the respective precursor
solution and the internal standard solution were mixed (2:1,
v/v). In feeding experiments with d5-(S)-pulegone (9), only the
precursor solution was used. The shoot tip and first leaf pair
and 400 µL of the feeding solution were incubated in a sealed
2 mL vial at room temperature for 17-21 h in the dark. Blank
cultures were maintained with distilled water containing 0.1
mg/mL Tween 20. The plant material was then placed in a
new vial, sealed, and incubated for 1 h in the dark. Afterward,
the essential oil evaporating from the glandular trichomes as
well as the feeding solution was analyzed using headspace
SPME.
SP ME. An SPME fiber, coated with a 100 µm film of poly-
(dimethylsiloxane), installed in an SPME fiber holder for
manual use (both Supelco, Munich, Germany) was used for
headspace sampling. For thermal desorption, the SPME fiber
remained in the injector for 3 min, with the split valve being
opened after 2 min.
Labeled menthofurans were identified by their reten-
tion times on the precolumn and main column as well
as by their specific mass spectra. The mass spectra
showed a characteristic base peak, most probably due
to a retro-Diels-Alder reaction, as proved by the mass
spectra of different deuterium-labeled menthofurans
(Figure 1).
In first prescreening experiments with labeled race-
mic pulegone, small amounts of labeled racemic men-
thofuran were detected in the pure feeding solution by
SPME/enantio-MDGC/MS, despite exclusion of light and
without incubation of any plant material. Consequently,
photochemical reactions cannot be the cause for this
finding. However, the detection of small amounts of
labeled menthofuran racemate suggested a nonselective
autoxidation process. Furthermore, when using mixed
labeled pulegone, the 18O/16O ratio in pulegone and
corresponding menthofuran remained unchanged; how-
ever, when plant material was fed with mixed labeled
pulegone under identical experimental conditions, sig-
nificant differences were detectable. Enantiodiscrimi-
nation as well as a change in 18O/16O ratio in mentho-
furan took place, indicating an enzymatic process in the
plant. Therefore, the labeled menthofuran in the es-
sential oil analyzed consists mainly of biotransformed
pulegone, but a small amount of menthofuran from the
feeding solution also has to be taken into account. For
exact calculations, (S)-menthofuran (3) was chosen as
an ideal internal standard (IST), because this enanti-
omer (3) is not detectable in blank cultures of Mentha
× piperita. IST (3) was synthesized using oxidation and
cyclization of (S)-citronellol (1) to isopulegone (2). Ep-
oxidation and cyclization of an epoxyketone intermedi-
ate led to (S)-menthofuran (3) (Scheme 2).
RESULTS AND DISCUSSION
Feeding experiments with deuterium-labeled mono-
terpene precursors are powerful tools for in vivo studies
of monoterpene biosynthesis in plants. The essential oil
of the plants was analyzed by SPME and enantio-
MDGC/MS. This is a highly efficient and noninvasive
micro method for the evaluation of stereochemical
pathways during bioconversion reactions (Wu¨st et al.,
1996, 1998).
For the determination of the stereoselectivity in
menthofuran biogenesis, deuterium-labeled pulegone
precursors (6 and 7) were used. They are readily
available using hydrogen-deuterium-exchange reac-
tions and different reaction conditions. Deuterium and
18O-labeled pulegone (8) was synthesized using a special
exchange reaction (Scheme 1).
In feeding experiments with regioselectively deuter-
ated precursors (6 and 7) the corresponding menthofu-
ran enantiomers were calculated versus the internal
standard (S)-menthofuran (3). Thus, it was clearly
The procedure of mixed labeling proves whether the
oxygen-18 is lost during bioconversion.

Biosynthesis of Menthofuran in Mentha × piperita
J. Agric. Food Chem., Vol. 47, No. 10, 1999 4103
Ta ble 1. Ra tios of Deu ter iu m -La beled Men th ofu r a n
En a n tiom er s ver su s In ter n a l Sta n d a r d (S)-Men th ofu r a n
(3) in F eed in g Solu tion s a n d Cor r esp on d in g P la n tsa
Ta ble 2. En a n tiom er ic Distr ibu tion of Men th ofu r a n
Obta in ed fr om F eed in g Exp er im en ts w ith La beled
Ra cem ic P u legon esa
labeled (S)-
labeled (R)-
menthofuran (%) menthofuran (%)
labeled racemic pulegone
(6-8) (n ) 26,
58
42
σn-1 ) 5.38)
(R)-d8-pulegone [(R)-6]
(S)-d8-pulegone [(S)-6]
blank cultures
nd
>98
nd
>98
nd
nd
a
n, numbers of analysis; σn-1, standard deviation; nd, not
detectable.
experiments with deuterium and 18O-labeled pulegone
were carried out. The obtained mass spectra of men-
thofuran from feeding experiments with d6-pulegone (7)
and d6-[18O]pulegone (8) were identical, indicating that
the oxygen of the pulegone was lost during bioconver-
sion. The loss of oxygen-18 of compound 8 by hydrolysis
was excluded, as indicated by GC/MS analysis of 8 in
the blank feeding solution (deuterium and 18O con-
tents: [2H5,16O] 6%, [2H6,16O] 11%, [2H5,18O] 39%,
[2H6,18O] 43%). In the plant, no hydrolytic loss of 18O
labeling took place, as proved by the mass spectra of
menthone and isomenthone, also generated from d6-
[18O]pulegone (8). Their mass spectra showed the same
18O/16O ratio as the labeled pulegone precursor (Figure
3).
a
Ess. oil is essential oil.
Obviously, during bioconversion of labeled pulegone
a loss of deuterium occurs, proved by the mass spectra
of the generated menthofuran. After feeding labeled
pulegones (6-8), the mass spectra of corresponding
menthofurans showed high-intensity peaks at m/z 113
and 111, respectively, sometimes being the base peak.
The deuterium content of pulegones (6-8) in the feeding
solution remained almost unaffected [e.g., deuterium
content of feeding solution (6): [2H6] 2%, [2H7] 23%, [2H8]
75%], so the loss of the deuterium in the corresponding
labeled menthofurans cannot be explained as yet.
Our feeding experiments show that an enzymatic
process is responsible for the bioconversion of labeled
pulegone to labeled menthofuran. Photooxygenation can
be excluded because the feeding experiments were
carried out in the dark. Furthermore, autoxidative
processes are negligible, because no 18O-labeled men-
thofuran was detectable.
The enzymatic conversion of labeled pulegone into
labeled menthofuran in Mentha × piperita is not
specific, as both pulegone enantiomers are converted
into menthofuran with corresponding stereochemistry
at C-1. Furthermore, a discrimination in favor of labeled
(S)-pulegone, the analogue of the nongenuine (S)-
pulegone, was observed (Scheme 3).
From the literature, differences between (S)-pulegone
and (R)-pulegone in the metabolism in mammals are
known (Madyastha and Gaikwad, 1998). Both pulegone
enantiomers are metabolized to menthofuran, although
higher levels of menthofuran were found in the case of
(R)-pulegone. However, in this context, it must be taken
into account that both pulegone enantiomers do not
occur as genuine compounds in mammals.
F igu r e 2. Enantio-MDGC/MS analysis of the headspace
extract of M. × piperita fed with racemic d6-pulegone (7); (S)-
menthofuran (3) is used as internal standard (main column
chromatogram).
shown that the labeled menthofuran species could be
traced back almost exclusively to the bioconversion of
the corresponding pulegone precursors (Table 1).
The main column chromatogram (enantio-MDGC/MS)
of a feeding experiment using racemic d6-pulegone (7)
is shown in Figure 2.
At m/z 108 the genuine (R)-menthofuran (4) and the
IST (S)-menthofuran (3) are detected. The deuterium-
labeled menthofuran was detected at m/z 112. During
the bioconversion of labeled pulegone, an enantiodis-
crimination was observed. When a racemic mixture of
labeled pulegone was fed, labeled (S)-menthofuran, the
analogue of the nongenuine menthofuran enantiomer,
was preferably biosynthesized in Mentha × piperita. The
main result of all feeding experiments, using labeled
racemic pulegone, was 58% (S)-menthofuran to 42% (R)-
menthofuran (Table 2).
Feeding experiments with enantiopure d8-pulegone
[(R)-6, (S)-6] proved labeled (S)-menthofuran and (R)-
menthofuran, respectively, to be exclusively biosynthe-
sized from the corresponding pulegone precursor.
To determine whether the oxygen of the labeled
pulegone is the same as that in menthofuran, feeding
On the basis of the results obtained from our feeding
experiments, obviously an enzymatic oxidation of la-
beled pulegone (probably by a cytochrome oxidase) is
involved in the bioconversion of labeled pulegone into

4104 J. Agric. Food Chem., Vol. 47, No. 10, 1999
Fuchs et al.
F igu r e 3. Mass spectra of labeled menthofuran and labeled menthone/isomenthone obtained after feeding experiments with
d6-pulegone (7) and d6-[18O]-pulegone (8).
Sch em e 3. P r op osed P a th w a y for th e Biocon ver sion
CONCLUSIONS
of P u legon e to Men th ofu r a n in M. × piper ita
Feeding experiments with stable isotope labeled pule-
gone enantiomers proved that Mentha × piperita is able
to convert both pulegone enantiomers into the corre-
sponding (R)- and (S)-configured menthofurans. Fur-
thermore, when racemic pulegone precursors were
administered, a discrimination was observed, favoring
the labeled analogue of the nongenuine (S)-pulegone.
An enzymatic oxidation is involved in menthofuran
biosynthesis, as shown by incubation in the dark and
by loss of the pulegone oxygen.
ACKNOWLEDGMENT
We thank W. Girnus, Botanical Garden of Frankfurt
University, for cultivating the plant material.
LITERATURE CITED
Bambrige, K.; Clark, B. P.; Simskins, N. S. Regio- and
enantioselective enolisations of cyclic ketones using chiral
lithium amide bases. J . Chem. Soc., Perkin Trans. 1 1995,
20, 2535-2541.
Battaile, J .; Loomis, W. D. Biosytnhesis of Terpenes. II. The
Site and Sequence of Terpene Formation in Peppermint.
Biochim. Biophys. Acta 1961, 51, 545-552.
Burbott, A. J .; Loomis, W. D. Effects of Light and Temperature
on the Monoterpenes of Peppermint. Plant Physiol. 1967,
42, 20-28.
menthofuran. The oxygen introduced by oxidation of
pulegone remains in the menthofuran, whereas the
original oxygen of the pulegone carbonyl group is lost
during bioconversion. This finding is in agreement with
investigations in mammals using experiments carried
out under an 18O2 atmosphere, where complete incor-
poration of 18O2 into menthofuran was observed (Nelson
et al., 1992b).
Clark, R. J .; Menary, R. C. Environmental Effects on Pep-
permint (Mentha piperita L.). I. Effect of Daylength, Photon
Flux Density, Night Temperature and Day Temperature on
the Yield and Composition of Peppermint Oil. Aust. J . Plant
Physiol. 1980, 7, 685-692.
Corey, E. J .; Ensley, H. E.; Suggs, J . W. A Convenient
Synthesis of (S)-(-)-Pulegone from (-)-Citronellol. J . Org.
Chem. 1976, 41, 380-381.
The labeling pattern of the menthofuran biosynthe-
sized in Mentha × piperita does not contradict the
mechanism proposed for the bioconversion of pulegone
to menthofurane in mammals. Nelson et al. (1992b)
suggested a cytochrome P-450-catalyzed oxidation of the
cis-methyl group of the pulegone isopropylidene moiety,
subsequently cyclized and dehydrated. Therefore, fur-
ther investigations must be carried out to obtain more
details of menthofuran biosynthesis in Mentha × pip-
erita.
Croteau, R.; Gershenzon, J . Genetic Control of Monoterpene
Biosynthesis in Mints (Mentha: Lamiaceae). Recent Adv.
Phytochem. 1994, 28, 193-229.

Biosynthesis of Menthofuran in Mentha × piperita
J. Agric. Food Chem., Vol. 47, No. 10, 1999 4105
Friedrich, D.; Bohlmann, F. Total Synthesis of Various El-
emanolides. Tetrahedron 1988, 44, 1369-1392.
A Toxic Monoterpene. In Secondary-Metabolite Biosynthesis
and Metabolism; Petroski, R. J ., McCormick, S. P., Eds.;
Plenum Press: New York, 1992b; pp 287-296.
Reitsema, R. H. A Biogenetic Arrangement of Mint Species.
J . Am. Pharm. Assoc. 1958, 47, 267-269.
Fuchs, S.; Beck, T.; Burkardt, S.; Sandvoss, M.; Mosandl, A.
Biogenetic studies in Mentha × piperita. 1. Deuterium
Labeled Monoterpene Ketones: Synthesis and Stereoselec-
tive Analysis. J . Agric. Food Chem. 1999, 47, 3053-3057.
Gibson, T. Carbon-13 Magnetic Resonance Spectra and the
Stereochemistry of Some Substituted Bicyclo[2.2.2]hexanes.
Org. Magn. Reson. 1098, 21, 482-486.
Schmarr, H. G. Beitra¨ge zur on-line LC-GC Kopplung und
modifizierte Cyclodextrine als chirale stationa¨re Phasen in
der Kapillar-GC. Doctoral Thesis, University of Frankfurt/
Main, Germany, 1992.
Tori, K.; Horibe, I.; Shigemoto, H. 1H NMR Solvent Shifts
induced by Hexafluorobenzene in Monoterpenes. Tetrahe-
dron Lett. 1975, 26, 2199-2202.
Madyastha, K. M.; Gaikwad, N. W. Metabolic fate of S-(-)-
pulegone in rat. Xenobiotica 1998, 28, 723-734.
McCaskill, D.; Croteau, R. Prospects for the Bioengineering
of Isoprenoid Biosynthesis. In Advances in Biochemical
Engineering Biotechnology; Scheper, T., Ed.; Spinger: Ber-
lin, Germany, 1997; Vol. 55, p 107-146.
McClanahan, R. H.; Huitric, A. C., Pearson, P. G.; Desper, J .
C.; Nelson, S. D. Evidence for a Cytochrome P-450 Catalyzed
Allylic Rearrangement with Double Bond Topomerisation.
J . Am. Chem. Soc. 1988, 110, 1979-1981.
Moorthy, B.; Madyastha, P.; Madyastha, K. M. Metabolism of
a monoterpene ketone, R-(+)-pulegonesa hepatoxin in rat.
Xenobiotica 1989, 19, 217-224.
Vederas, J . C. Structural Dependence of 18O Isotope Shifts in
13C NMR Spectra. J . Am. Chem. Soc. 1980, 102, 374-376.
Werkhoff, P.; Brennecke, S.; Bretschneider, W.; Schreiber, K.
Enantioselektive GC-Analyse chiraler Riech- und Aromast-
offe, Teil 2. H&R Contact 1995, 64, 7-11.
Wu¨st, M.; Beck, T.; Dietrich, A.; Mosandl, A. On the Biogenesis
of Rose Oxide in Pelargonium Graveolens L’He´ritier and
Pelargonium radens H. E. Moore. Enantiomer 1996, 1, 167-
176.
Wu¨st, M.; Rexroth, A.; Beck, T.; Mosandl, A. Mechanistic
Aspects of the Biogenesis of Rose Oxide in Pelargonium
graveolens L’He´ritier. Chirality 1998, 10, 229-237.
Murray, M. J .; Hefendehl, F. W. Changes in Monoterpene
Composition of Mentha aquatica produced by Gene Substi-
tution from M. arvensis. Phytochemistry 1972, 11, 2469-
2474.
Nelson, S. D.; McClanahan, R. H.; Thomassen, D.; Gordon, W.
P.; Knebel, N. Investigations of mechanism of reactive
metabolite formation from (R)-(+)-pulegone. Xenobiotica
1992a , 22, 1157-1164.
Nelson, S. D.; McClanahan, R. H.; Knebel, N.; Thomassen, D.;
Gordon, W. P.; Oishi, S. The metabolism of (R)-(+)-pulegone,
Received for review April 30, 1999. Revised manuscript
received J uly 26, 1999. Accepted August 2, 1999. Financial
support from Deutsche Forschungsgemeinschaft (DFG) is
gratefully acknowledged.
J F990439E