A chemoenzymatic, preparative synthesis of the isomeric forms of p-menth-1-en-9-ol: Application to the synthesis of the isomeric forms of the cooling agent 1-hydroxy-2,9-cineole

Article abstract of DOI:10.1002/ejoc.200701010

A preparative-scale synthesis of the four p-menth-1-en-9-ol isomers 2a-5a has been achieved by means of two chemoenzymatic processes. Both synthetic pathways start from the enantiomeric forms of limonene that are converted into p-mentha-1,8-dien-9-al isomers 12 and 15. The baker's yeast mediated reduction of the latter aldehydes afforded compounds 3a and 5a, respectively, with very high enantioselectivity. Moreover, chemical reduction of 12 and 15 gives the mixtures of enantiopure diastereoisomers 2a/3a and 4a/5a, respectively. PPL (Porcine pancreas lipase) mediated resolution of the latter mixtures followed by fractionating crystallization of derivatives 2b-5b allowed the enantio- and diastereoisomerically pure alcohols 2a-5a to be obtained. Compounds 2a-5a have then been used as starting materials for the preparation of four isomers of the cooling agent 1-hydroxy-2,9-cineole (6-9). Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200701010

FULL PAPER
DOI: 10.1002/ejoc.200701010
A Chemoenzymatic, Preparative Synthesis of the Isomeric Forms of
p-Menth-1-en-9-ol: Application to the Synthesis of the Isomeric Forms of the
Cooling Agent 1-Hydroxy-2,9-cineole
Stefano Serra,*[a] Claudio Fuganti,[a] and Francesco G. Gatti[b]
Keywords: Biotransformations / Enzyme catalysis / Reduction / Baker’s yeast / Natural products
A preparative-scale synthesis of the four p-menth-1-en-9-ol
isomers 2a–5a has been achieved by means of two chemoen-
zymatic processes. Both synthetic pathways start from the en-
antiomeric forms of limonene that are converted into p-men-
tha-1,8-dien-9-al isomers 12 and 15. The baker’s yeast medi-
ated reduction of the latter aldehydes afforded compounds
3a and 5a, respectively, with very high enantioselectivity.
Moreover, chemical reduction of 12 and 15 gives the mix-
tures of enantiopure diastereoisomers 2a/3a and 4a/5a,
respectively. PPL (Porcine pancreas lipase) mediated resolu-
tion of the latter mixtures followed by fractionating crystalli-
zation of derivatives 2b–5b allowed the enantio- and diaste-
reoisomerically pure alcohols 2a–5a to be obtained. Com-
pounds 2a–5a have then been used as starting materials for
the preparation of four isomers of the cooling agent 1-hy-
droxy-2,9-cineole (6–9).
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
one in the cyclohexene ring and the other in the side
chain.[1] Since many products of this kind show biological
activity, which is strictly related to their absolute configura-
tion, their syntheses need a high degree of stereocontrol. It
is worthy to note that the formation of the C(4)–C(8) bond
with simultaneous creation of two new asymmetric centres
is a demanding synthetic reaction. The use of the four p-
menth-1-en-9-ol isomers 2a–5a as common chiral building
blocks is an attractive alternative to the use of the above-
mentioned stereoselective approach.
Introduction
A great number of natural compounds show the general
p-menthane structure 1 (Scheme 1). The most well known
are the p-menthane terpenes and the sesquiterpenes of the
bisabolene family that widely occur in nature as a constitu-
ents of a large number of essential oils.
In view of the ready availability of both enantiomers of
limonene, the specific transformation of the latter terpenes
into the alcohols 2a–5a was first investigated in 1966[2] and
was improved a few years later.[3] The method consists of
the hydroalumination or hydroboration of the C(8)=C(9)
bond of limonene, followed by oxidation of the correspond-
ing organoalane or organoborane derivatives, respectively,
to afford the p-menth-1-en-9-ol isomers. These reactions
proceed with high regioselectivity and very low (or no) dia-
stereoselectivity, and the mixtures of enantiopure diastereo-
isomers are not easily separable. Fractionating crystalli-
zation of the corresponding 3,5-dinitrobenzoyl ester allows
the large-scale preparation of alcohols 3a and 4a, whereas
alcohols 2a and 5a can be obtained in very low yields (3–
4%) only by a tedious process based on the use of mother
liquors that are derived from the crystallization of 3b and
4b. Until now, this path remains the most simple and direct
method of preparing the above-mentioned building blocks.
As a consequence of this fact, many syntheses of natural
products are based on the use of p-menth-1-en-9-ol as a
mixture of enantiopure diastereoisomers,[4] and only few of
them have employed the pure isomers.[5]
Scheme 1. Structures and numbering of p-menth-1-en-9-ol isomers
2–5.
Among these compounds, the most challenging are those
that contain two contiguous secondary stereogenic centres:
[a] C.N.R. Istituto di Chimica del Riconoscimento Molecolare,
Via Mancinelli 7, 20131 Milano, Italy
Fax: +39-2-2399-3080
E-mail: stefano.serra@polimi.it
[b] Dipartimento di Chimica, Materiali ed Ingegneria Chimica del
Politecnico,
Via Mancinelli 7, 20131 Milano, Italy
Eur. J. Org. Chem. 2008, 1031–1037
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1031

S. Serra, C. Fuganti, F. G. Gatti
FULL PAPER
As part of a project aimed at the preparation of a new
cooling agent with the p-menthane structure,[6] we recently
discovered that 1-hydroxy-2,9-cineole (6–9, Scheme 2) exhi-
bits a definite cooling effect. In order to start with a valu-
able structure–activity relationship (SAR) study, we needed
to evaluate all the latter isomers in their pure form. Since
the isomers of 1-hydroxy-2,9-cineole have been synthesized
starting from the p-menth-1-en-9-ol isomers,[5a] we focused
our attention on the large-scale preparation of the latter
alcohols. Moreover, our previous interest in the use of bio-
catalysis for the enantioselective synthesis of terpene-[7] and
sesquiterpene[8] derivatives led us to develop a chemoen-
zymatic approach to the compounds 2–5.
Scheme 3. Synthesis of the p-menth-1,8-dien-9-al enantiomers 12
and 15.
satisfactory, although the metalation was not completely re-
gioselective. Indeed, also the perillic alcohol was afforded
in about 10% of the p-menthenol mixture. Luckily, perillal-
dehyde is easily separable from p-mentha-1,8-dien-9-al by
chromatography. Therefore, manganese dioxide oxidation
of the impure alcohols 11 and 14, followed by purification
afforded aldehydes 12 and 15, respectively. The latter com-
pounds were then submitted to microbial reduction
(Scheme 4).
Scheme 2. Structures of the 1-hydroxy-2,9-cineole isomers (6–9).
Herein, we report the accomplishment of our plan by
means of two different synthetic strategies. The first one
afforded compounds 3a and 5a and is based on the baker’s
yeast mediated diastereoselective reduction of the
C(8)=C(10) bond of p-mentha-1,8-dien-9-al isomers 12 and
15. The second strategy afforded the four isomers 2a–5a by
means of the lipase-mediated resolution of the mixtures of
enantiopure diastereoisomers 2a/3a and 4a/5a.
Scheme 4. Baker’s yeast mediated synthesis of 3a and 5a.
Results and Discussion
As mentioned above, limonene enantiomers are cheap
The experimental conditions were the same for each sub-
and commercially available enantiopure starting materials strate (see Experimental Section), and we decided to per-
for the synthesis of alcohols 2a–5a. We selected aldehydes form the baker’s yeast reduction in the presence of a nonpo-
12 and 15 (Scheme 3) as precursors for the latter alcohols. lar resin. This technique allowed a large-scale reduction,
We envisaged that the diastereoselective reduction of the since high concentration of substrate (7 g L–1) was achieved
disubstituted double bond will be achieved by microbial re- and the workup procedure was simplified. The results of
duction. It is known that the baker’s yeast reduction of acti- the biotransformations were very good. The reduction of
vated di- and trisubstituted olefins proceeds with high the conjugated double bond by baker’s yeast gave high
enantioselectivity.[8b–8d,9] Preliminary studies, aimed at the yields of saturated alcohols (67–59% of isolated products),
synthesis of the bisabolene sesquiterpenes, have shown that and the diastereoisomeric ratio of the products was inde-
3-(4-methylcyclohex-3-enyl)-but-2-enal isomers are good pendent of the enantiomeric form of the substrates. The
substrates for this type of transformations.[8c] In order to new chiral centre was formed with high preference for the
verify the applicability of the latter approach, we prepared (R) absolute configuration, which confirms the general
a large amount of the two p-mentha-1,8-dien-9-al isomers.
trend of the enantioselectivity for the reduction of this type
The selective metalation at C(10) of (+)-limonene 10 and of aldehydes.[9] Since the microbial transformation gave also
(–)-limonene 13 was performed by using a n-butyllithium/ allyl alcohols (11 and 14) and unreacted aldehydes, the en-
TMEDA complex.[10] According to a method described in tire reaction mixture was submitted to treatment with man-
the patent literature, the allyllithium intermediates were ganese dioxide. Thus, after chromatographic separation, the
treated with trimethylborate, and the obtained organobor- saturated alcohols were isolated, and the starting aldehydes
ane derivatives were oxidized in situ with hydrogen peroxide could be resubjected to microbial reduction. Overall, the
to afford alcohols 11 and 14, respectively.[11] The yields were process allows the diastereoselective, large-scale preparation
1032
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1031–1037

Chemoenzymatic Synthesis of the Isomeric Forms of p-Menth-1-en-9-ol
of alcohol 3a (97% de) and alcohol 5a (96% de) from alde- Table 1. Results of the enzyme-mediated acetylation of p-men-
thane-9-ol 2a/3a and 4a/5a.
hyde 12 and 15, respectively.
However, we are interested in preparing all the isomeric
forms of the latter alcohols. Our previous study on the li-
pase-mediated resolution of p-menthan-3,9-diols[7d] showed
that the acetylation of the 9-hydroxy group is seldom dia- 2a/3a CRL
stereoselective. Therefore, we decided to prepare the mix-
Entry Enzyme Acetylated product
Enantiomer ratio Conversion
configuration; de [%][a] (E)[b]
PPL
4R,8S; 73 17
4R,8R; 25 1.9
0.54
0.40
0.49
PS
4R,8S; 29
2.3
tures of enantiopure diastereoisomers 2a/3a and 4a/5a,
respectively, in order to submit them to enzyme-mediated
resolution.
PPL
4a/5a CRL
PS
4S,8S; 79
4S,8R; 21
4S,8S; 38
18
1.8
3.1
0.48
0.50
0.48
[a] Chiral GC analysis. [b] E = ln[1 – cϫ(1 + dep)]/ln[1 – cϫ(1 –
dep)].
The latter alcohols were prepared by a two-step pro-
cedure starting again from the p-mentha-1,8-dien-9-al iso-
mers (Scheme 5). The aldehydes 12 and 15 were treated with
triethylsilane in the presence of a catalytic amount of Wilk-
inson’s catalyst.[12] The reduction of the conjugated double
bond followed by the in situ hydrolysis of the obtained tri-
ethylsilylenolether intermediates afforded aldehydes 16 and
17, respectively. The latter compounds were purified by
chromatography and then reduced with sodium borohyd-
ride to give two mixtures of enantiopure diastereoisomers
2a/3a and 4a/5a, respectively.
for the 2-arylpropan-1-ol (primary alcohols)[13] – both show
the preferential acetylation of the (S) enantiomer by PPL
with a higher selectivity.
Taking advantage of these experimental findings, we per-
formed the resolution of the two mixtures of enantiopure
diastereoisomers 2a/3a and 4a/5a with PPL as catalyst
(Scheme 6).
Scheme 5. Regioselective chemical reduction of aldehydes 12 and
15.
Each of the two mixtures of alcohols was treated with
vinyl acetate in tBuOMe solution in the presence of dif-
ferent lipases [PPL (Porcine pancreas lipase) type II, CRL
(Candida rugosa lipase) type VII and lipase PS (Pseudomo-
nas cepacia)]. The reactivity of each substrate towards the
irreversible acetylation was tested by monitoring the prod-
uct distribution at regular time intervals by chiral GC
analysis. The results of this study are collected in the Table 1
Scheme 6. PPL-mediated resolution of the two mixtures of alcohols
2a/3a and 4a/5a.
The enzymatic acetylation was interrupted at about 50%
and present some interesting observations. All the lipases conversion and acetates 2c and 4c were separated from the
mediate the acetylation of the alcohols, but the diastereo- unreacted alcohols 3a and 5a, respectively. The achieved
selectivity is strongly dependent on the type of lipase used. new set of derivatives was further manipulated in order to
PPL showed a higher selectivity with a preference for the increase its diastereoisomeric purity. Acetates 2c and 4c
conversion of the (8S) isomer. In contrast, lipase PS showed were hydrolyzed, thus all alcohols were converted into the
low selectivity, whereas CRL converted the (8R) isomer, al- corresponding 3,5-dinitrobenzoyl esters. Two crystalli-
though with poor selectivity. Noteworthy, the absolute con- zations from hexane afforded derivative 2b–5b with a dia-
figuration of the stereocentre at position 4 did not affect stereoisomeric excess up to 97%. Finally, hydrolysis of the
the stereochemical course of the acetylation.
latter compounds gave the desired set of enantio- and dia-
Interestingly, these results are in partial contrast with stereoisomerically pure alcohols 2a–5a. The described pro-
those previously obtained for the acylation of the very cedures compare favourably with previously reported syn-
closely related substrate p-menthane-3,9-diol,[7d] where li- theses and allow the preparation of the title compounds in
pase PS showed a higher enantioselectivity. Unexpectedly, larger scales and in high isomeric purities.
the present study displays some similarity between the ki-
In accord with our first pursuit, we employed these
netic resolution of p-menth-1-en-9-ol and those described building blocks for the synthesis of the 1-hydroxy-2,9-cine-
Eur. J. Org. Chem. 2008, 1031–1037
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1033

S. Serra, C. Fuganti, F. G. Gatti
FULL PAPER
ole isomers (Scheme 7). It is known[5a] that the epoxidation
of p-menth-1-en-9-ol affords a 1:1 mixture of epoxides and,
because of steric reasons, only one of these two diastereo-
isomers is prone to acid-catalyzed cyclization to give cineole
derivatives. Although this process allows the preparation of
the target compounds in a yield below 50%, it has the ad-
vantage to be direct and does not require demanding reac-
tions. Moreover, we found that the isolation of the mixture
of epoxides is not necessary because quenching with so-
dium sulfite already catalyzes the cyclization.
mercial quality. Lipase from Porcine pancreas type II (Sigma,
147 units/mg), lipase from Candida rugosa type VII (Sigma,
1150 units/mg) and lipase from Pseudomonas cepacia (Amano
Pharmaceuticals Co., Japan, 30 units/mg) were used in this work.
Fresh baker’s yeast (DSM Bakery Ingredients Italy s.p.a.) was
bought in a local market and used without further manipulations.
TLC: Merck silica gel 60 F254 plates. Column chromatography
(CC): silica gel. GC-MS analyses: HP-6890 gas chromatograph
equipped with a 5973 mass detector, with a HP-5MS column
(30 mϫ0.25 mm, 0.25 µm firm thickness; Hewlett Packard) with
the following temperature program 60 °C (1 min) – 6 °C/min –
150 °C (1 min) – 12 °C/min – 280 °C (5 min); carrier gas, He; con-
stant flow 1 mL/min; split ratio, 1/30; tR given in min: tR(2a–5a)
13.2, tR(2b–5b) 29.5, tR(2c–5c) 16.3, tR(6, 9) 13.1, tR(7, 8) 14.0,
tR(11, 14) 13.0, tR(12, 15) 11.1, tR(16, 17) 13.4; mass spectra: m/z
(rel. %). Chiral GC analyses: DANI-HT-86.10 gas chromatograph;
enantiomer excesses determined on a CHIRASIL DEX CB-Col-
umn with the following temperature program 60 °C (0 min) – 2 °C/
min – 80 °C (0 min) – 1 °C/min – 85 °C (0 min) – 0.2 °C/min –
88 °C (0 min) – 25 °C/min – 180 °C (2 min); tR given in min: tR(2c)
24.7, tR(3c) 25.0, tR(4c) 23.8, tR(5c) 24.1. Optical rotations: Jasco-
DIP-181 digital polarimeter. 1H- and 13C NMR spectra: CDCl3
solutions at room temperature; Bruker-AC-400 spectrometer at 400
and 100 MHz, respectively; chemical shifts in ppm relative to in-
ternal SiMe4 (=0 ppm), J values in Hz. The diastereoisomeric ex-
cesses of alcohols 3a–5a and of the esters 3b–5b were determined
by 1H NMR spectroscopic analysis by using the signals arising
from the C(10) methyl group (doublet) for the determination of the
diastereoisomeric purities. IR spectra were recorded on a Perkin–
Scheme 7. Synthesis of the 1-hydroxy-2,9-cineole isomers (6–9).
Thus, overall, the latter two-step process could be per-
formed in just “one pot”. According to this observation, we
treated the set of alcohols 2a–5a with m-chloroperbenzoic,
acid and after complete epoxidation, we added sodium sul-
fite to the reaction mixture. After a few hours, workup and
chromatographic separation afforded the 1-hydroxy-2,9-cin-
eole isomers (6–9), which were further purified by crystalli-
zation from hexane. Spectroscopic data of these compounds
are in good agreement with those previously reported,[5a]
whereas optical rotation measurements showed higher val-
ues, which clearly demonstrates the good isomeric purity of
the starting alcohols. Evaluation of the cooling properties
of compounds 6–9 are still in progress and will be reported
in due course.
Elmer 2000 FT-IR spectrometer; films; ν in cm–1. Melting points
˜
were measured on a Reichert apparatus, equipped with a Reichert
microscope, and are uncorrected. Microanalyses were determined
on an analyzer 1106 from Carlo Erba.
Procedure for Preparation of p-Mentha-1,8-dien-9-al: To a cooled
(0 °C) and well-stirred solution of 1.2  n-butyllithium in hexane
(834 mL, 1 mol) was added dropwise under nitrogen N,N,NЈ,NЈ-
tetramethylethylenediamine (108 g, 1.1 mol). To the resulting yel-
low solution, (+)-limonene (205 g, 1.5 mmol) was added slowly, and
the mixture was stirred overnight at room temperature. The dark
red solution of metalated limonene obtained was then cooled to
–78 °C, and trimethyl borate (115 g, 1.1 mol) was added dropwise.
The reaction was warmed to –30 °C and then treated over a 2 h
period with 30% aqueous hydrogen peroxide solution (170 mL,
1.5 mol). The mixture was then quenched with water (200 mL) and
diluted with diethyl ether (400 mL). The layers were separated, and
the aqueous phase was extracted with diethyl ether (2ϫ200 mL).
The combined organic solution was washed with brine, dried
(Na2SO4) and evaporated in vacuo. The residue was distilled to give
recovered (+)-limonene (89 g, 0.65 mol) and a 9:1 mixture of p-
mentha-1,8-dien-9-ol 11 and (+)-perillic alcohol (83 g, 545 mmol).
The latter oil was dissolved in CHCl3 (300 mL) and treated with
MnO2 (200 g, 2.3 mol), and the mixture was stirred at reflux for
8 h. The residue obtained upon filtration and evaporation of the
CHCl3 phase was purified by column chromatography with hexane/
diethyl ether (95:5–8:2) as eluent to give pure (+)-p-mentha-1,8-
dien-9-al 12 (67.9 g, 453 mmol, 45% based on n-butyllithium) and
(+)-perillaldehyde (6.5 g, 43 mmol).
Conclusions
In summary, we have developed an efficient methodology
for the large-scale preparation of the four isomers of p-
ment-1-en-9-ol. The starting materials are the two enantio-
meric forms of p-mentha-1,8-dien-9-al that are, in turn, pre-
pared from the cheap and commercially available limonene
enantiomers. Baker’s yeast mediated reduction of the latter
aldehydes directly provides the (8R) isomers in very good
yields and in a stereoselective fashion. In addition, the com-
bination of the lipase-mediated resolution of the mixture of
enantiopure diastereoisomers with fractionating crystalli-
zation allows the preparation of all isomeric forms of the
target alcohols in high purity.
A first use of these chiral building blocks is in the prepa-
ration of four isomers of 1-hydroxy-2,9-cineole. Moreover,
we will exploit their use for the preparation of natural bisa-
bolane sesquiterpenes.
The procedure described above was repeated with (–)-limonene to
afford (–)-p-mentha-1,8-dien-9-al 15 (69.1 g, 461 mmol, 46% based
on n-butyllithium) and (–)-perillaldehyde (6.2 g, 41 mmol).
Experimental Section
General Remarks: All moisture-sensitive reactions were carried out
under a static atmosphere of nitrogen. All reagents were of com-
(R)-p-Mentha-1,8-dien-9-al (12): [α]2D0 = +76 (c = 1.5, CHCl3),
ref.[14] [α]2D0 = +88 (c = 0.03, CHCl ). IR (neat): ν = 1694, 1622,
˜
3
1034
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1031–1037

Chemoenzymatic Synthesis of the Isomeric Forms of p-Menth-1-en-9-ol
1
1437, 941, 796 cm–1. H NMR (400 MHz, CDCl3): δ = 9.55 (s, 1
gen with Et3SiH (28.7 mL, 180 mmol) and (Ph3P)3RhCl (0.5 g,
H), 6.26 (s, 1 H), 6.00 (s, 1 H), 5.43–5.38 (m, 1 H), 2.76–2.67 (m, 0.54 mmol). The mixture was stirred and heated at 50 °C for 2 h.
1 H), 2.23–2.03 (m, 2 H), 1.98–1.74 (m, 3 H), 1.66 (s, 3 H), 1.52
(dddd, J = 12.7, 11.3, 10.6, 5.5 Hz, 1 H) ppm. 13C NMR (100 MHz,
CDCl3): δ = 194.7, 154.6, 133.7, 133.1, 120.0, 31.4, 30.6, 29.8, 27.5,
After cooling to room temperature, the reaction mixture was di-
luted with THF (100 mL), water (30 mL) and 5% aqueous HCl
(10 mL). The stirring was continued for a further 2 h, then the mix-
23.4 ppm. MS (EI): m/z (%) = 150 (15) [M+], 135 (13), 122 (100), ture was diluted with water (500 mL) and extracted with diethyl
107 (36), 91 (41), 79 (74), 67 (63), 53 (31), 39 (26). C10H14O
(150.22): calcd. C 79.96, H 9.39; found C 79.85, H 9.40.
(S)-p-Mentha-1,8-dien-9-al (15): [α]2D0 = –73.9 (c = 1.5, CHCl3). IR,
1H NMR, 13C NMR and MS data in accordance with those of 12.
C10H14O (150.22): calcd. C 79.96, H 9.39; found C 79.80, H 9.40.
ether (3ϫ150 mL). The combined organic phase was washed with
brine (100 mL), dried (Na2SO4) and concentrated under reduced
pressure. The residue was purified by chromatography by using
hexane/diethyl ether (95:5–9:1) as eluent to afford aldehyde 16
(20.1 g, 132 mmol). The latter oil was dissolved in methanol
(60 mL) and treated at 0 °C with NaBH4 (2.4 g, 63 mmol) whilst
stirring. After complete reduction of 16 (1 h, TLC monitoring), the
reaction mixture was diluted with diethyl ether (250 mL) and
treated with 5% aqueous HCl (100 mL). The layers were separated,
and the aqueous phase was extracted with diethyl ether
(2ϫ100 mL). The combined organic solution was washed with
brine, dried (Na2SO4) and evaporated in vacuo. The residue was
distilled (b.p. 70–75 °C/0.5 Torr) to give a 1:1 mixture (NMR analy-
sis, 96% chemical purity by GC) of alcohols 2a and 3a (18.8 g,
122 mmol, 73%).
Baker’s Yeast Reduction of Aldehydes 12 and 15: A 10-L open cylin-
drical glass vessel equipped with a mechanical stirrer was charged
with tap water (5 L) and glucose (300 g). Fresh baker’s yeast
(1.5 kg) was added in small pieces to the stirred mixture, and the
fermentation process was allowed to proceed for 2 h. Aldehyde 12
(50 g, 333 mmol), adsorbed on the resin AMBERLITE XAD 1180
(Rohm and Haas Company, 250 g), was added in one portion. Vig-
orous stirring was continued for 4 d at room temperature. During
this time, more baker’s yeast (250 g) and glucose (100 g) were added
after 24 and 48 h since fermentation began. The resin was then
separated by filtration through a sintered glass funnel (porosity 0,
Ͼ160 µm), and the water phase extracted again with further resin
(50 g). The combined resin crops were extracted with diethyl ether
(4ϫ150 mL), and the organic solution was washed with brine. The
dried organic phase (Na2SO4) was concentrated under reduced
pressure to give an oil (57 g). The latter was dissolved in CHCl3
(200 mL) and treated with MnO2 (100 g, 1.15 mol), and the mixture
was stirred at reflux for 8 h. The residue obtained upon filtration
and evaporation of the CHCl3 phase was purified by column
chromatography using hexane/diethyl ether (9:1–3:1) as eluent to
give recovered aldehyde 12 (10.1 g, 67 mmol) and alcohol 3a
(34.5 g, 224 mmol, 67%).
The procedure described above was repeated with aldehyde 15
(25 g) to afford a 1:1 mixture (NMR analysis, 95% chemical purity
by GC) of alcohols 4a and 5a (19.3 g, 125 mmol, 75%).
PPL-Mediated Separation of the Diastereoisomers 2a/3a and 4a/4b:
A mixture of alcohols 2a/3a (15 g, 97 mmol), PPL (5 g), vinyl ace-
tate (15 mL) and tBuOMe (70 mL) was stirred at room tempera-
ture, and the formation of the acetate was monitored by TLC
analysis. The reaction was stopped at about 54% conversion (GC
analysis) by filtration of the enzyme and evaporation of the solvent
at reduced pressure. The residue was purified by chromatography
using hexane/diethyl ether (95:5–8:2) as eluent to give acetate 2c
(9.9 g, 50 mmol, 51%, 73% de by chiral GC analysis) and unre-
acted alcohol 3a (6.6 g, 43 mmol, 44%, 86% de by chiral GC analy-
sis of the corresponding acetate). The acetate was then dissolved in
methanol (10 mL) and treated with KOH (5 g, 89 mmol) in meth-
anol (30 mL) whilst stirring at room temperature until no more
starting material was detected by TLC analysis. The mixture was
diluted with water (80 mL) and extracted with diethyl ether
(3ϫ80 mL). The organic phase was washed with brine, dried
(Na2SO4) and concentrated in vacuo. The residue was treated with
pyridine (30 mL) and a solution of 3,5-dinitrobenzoyl chloride
(13 g, 56 mmol) in CH2Cl2 (60 mL). After complete conversion of
the starting alcohol, the mixture was diluted with water (300 mL)
and extracted with CH2Cl2 (2ϫ200 mL). The combined organic
phase was washed with aqueous NaHCO3 (5% solution), brine and
then dried (Na2SO4). Concentration at reduced pressure gave an
oil, which was purified by CC (hexane/Et2O, 9:1) and crystallized
twice from hexane to give pure ester 2b (10.8 g, 31 mmol, 62%).
The procedure described above was repeated with aldehyde 15
(50 g) to afford recovered aldehyde 15 (10.9 g, 73 mmol) and
alcohol 5a (30.5 g, 198 mmol, 59%).
(4R,8R)-p-Menth-1-en-9-ol (3a): 98% chemical purity (GC), 97%
de (by NMR analysis), [α]2D0 = +106 (c = 1, CHCl3), ref.[3a] [α]2D0
=
+104 (neat), ref.[3c] [α]2D0 = +106.79 (neat). IR (neat): ν = 3340,
˜
1450, 1378, 1040, 799 cm–1. 1H NMR (400 MHz, CDCl3): δ = 5.39–
5.35 (m, 1 H), 3.64 (dd, J = 10.6, 5.1 Hz, 1 H), 3.49 (dd, J = 10.6,
6.5 Hz, 1 H), 2.08–1.88 (m, 3 H), 1.83–1.68 (m, 2 H), 1.64 (s, 3 H),
1.61–1.50 (m, 2 H), 1.46 (br. s, 1 H), 1.33 (dddd, J = 12.7, 11.7,
11, 5.5 Hz, 1 H), 0.91 (d, J = 6.5 Hz, 3 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 133.9, 120.6, 66.4, 40.2, 35.2, 30.7, 27.7,
27.2, 23.4, 13.2 ppm. MS (EI): m/z (%) = 154 (23) [M+], 136 (21),
121 (56), 107 (49), 94 (100), 79 (67), 67 (62), 55 (29), 41 (25).
C10H18O (154.25): calcd. C 77.87, H 11.76; found C 77.80, H 11.80.
(4S,8R)-p-Menth-1-en-9-ol (5a): 97% chemical purity (GC), 96%
de (by NMR analysis), [α]2D0 = –95 (c = 1, CHCl3), ref.[3a] [α]2D5
=
In a similar manner, alcohol 3a (6.5 g, 42 mmol) was treated with
pyridine (30 mL) and a solution of 3,5-dinitrobenzoyl chloride
(11 g, 48 mmol) in CH2Cl2 (50 mL). After complete conversion of
the starting alcohol, workup and chromatographic purification, the
corresponding 3,5-dinitrobenzoyl ester was obtained, which was
crystallized twice from hexane to give pure 3b (10.4 g, 30 mmol,
71%).
–94 (c = 0.92, CHCl ). IR (neat): ν = 3335, 1451, 1378, 1042,
˜
3
798 cm–1. 1H NMR (400 MHz, CDCl3): δ = 5.39–5.34 (m, 1 H),
3.64 (m, 1 H), 3.49 (m, 1 H), 2.08–1.87 (m, 3 H), 1.87–1.69 (m, 2
H), 1.64 (s, 3 H), 1.59–1.46 (m, 2 H), 1.34 (br. s, 1 H), 1.25 (dddd,
J = 12.6, 11.4, 10.9, 5.7 Hz, 1 H), 0.93 (d, J = 6.3 Hz, 3 H) ppm.
13C NMR (100 MHz, CDCl3): δ = 133.9, 120.7, 66.3, 39.9, 35.3,
30.6, 29.8, 25.5, 23.3, 13.6 ppm. MS (EI): m/z (%) = 154 (28) [M+],
136 (23), 121 (63), 107 (54), 94 (100), 79 (64), 67 (52), 55 (23), 41
(16). C10H18O (154.25): calcd. C 77.87, H 11.76; found C 77.85, H
11.75.
(4R,8S)-p-Menth-1-en-9-yl 3,5-Dinitrobenzoate (2b): 99% chemical
purity by GC, up to 98% de by NMR analysis, m.p. 74–75 °C.
[α]2D0 = +46 (c = 1.5, CHCl3), ref.[3a] [α]2D0 = +41.85 (c = 0.965,
CHCl ). IR (nujol): ν = 1722, 1633, 1545, 1346, 721 cm–1. 1H NMR
˜
3
Chemical Reduction of Aldehydes 12 and 15: A solution of aldehyde
12 (25 g, 167 mmol) in dry THF (40 mL) was treated under nitro-
(400 MHz, CDCl3): δ = 9.19 (br. t, J = 2 Hz, 1 H), 9.12 (br. d, J
= 2 Hz, 2 H), 5.36 (br. s, 1 H), 4.46 (dd, J = 10.8, 5.4 Hz, 1 H),
Eur. J. Org. Chem. 2008, 1031–1037
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1035

S. Serra, C. Fuganti, F. G. Gatti
FULL PAPER
4.32 (dd, J = 10.8, 7.1 Hz, 1 H), 2.12–1.76 (m, 6 H), 1.64 (s, 3 H),
of 2b. C17H20N2O6 (348.35): calcd. C 58.61, H 5.79, N 8.04; found
1.64–1.52 (m, 1 H), 1.35 (dt, J = 11.5, 6 Hz, 1 H), 1.07 (d, J = C 58.80, H 5.80, N 8.00.
6.8 Hz, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 162.6, 148.7,
A sample of the ester 4b (9.1 g, 26 mmol) was hydrolyzed with
134.2, 134.1, 129.3, 122.2, 120.2, 70.2, 36.8, 35.9, 30.4, 29.5, 25.7,
23.3, 14.2 ppm. MS (EI): m/z (%) = 348 (1) [M+], 195 (17), 149
(14), 136 (30), 121 (32), 107 (25), 94 (100), 79 (25). C17H20N2O6
(348.35): calcd. C 58.61, H 5.79, N 8.04; found C 58.75, H 5.80, N
8.10.
KOH (2.5 g, 44 mmol) in refluxing methanol (40 mL). After
workup and bulb-to-bulb distillation (oven temperature 75–80 °C/
0.5 Torr), pure alcohol 4a (3.8 g, 24.7 mmol, 95%) was obtained.
According to the procedure described above, hydrolysis of a sample
of the ester 5b (8 g, 23 mmol) afforded pure 5a (3.25 g, 21.1 mmol,
92%).
(4R,8R)-p-Menth-1-en-9-yl 3,5-Dinitrobenzoate (3b): 99% chemical
purity by GC, up to 98% de by NMR analysis, m.p. 95–96 °C.
[α]2D0 = +37.8 (c = 1.5, CHCl3), ref.[3a] [α]2D0 = +36.7 (c = 0.77,
(4S,8S)-p-Menth-1-en-9-ol (4a): 98% chemical purity by GC, up to
98% de by NMR analysis, [α]2D0 = –104 (c = 1, CHCl3), ref.[3a]
[α]2D0 = –103.1 (c = 4.79, benzene). IR, 1H NMR, 13C NMR and
MS data in accordance with those of alcohol 3a obtained by the
baker’s yeast reduction procedure. C10H18O (154.25): calcd. C
77.87, H 11.76; found C 77.65, H 11.70.
CHCl ). IR (nujol): ν = 1721, 1632, 1546, 1345, 719 cm–1. 1H NMR
˜
3
(400 MHz, CDCl3): δ = 9.20 (br. t, J = 2 Hz, 1 H), 9.12 (br. d, J
= 2 Hz, 2 H), 5.37 (br. s, 1 H), 4.48 (dd, J = 10.8, 5.5 Hz, 1 H),
4.31 (dd, J = 10.8, 7.1 Hz, 1 H), 2.12–1.91 (m, 4 H), 1.91–1.74 (m,
2 H), 1.64 (s, 3 H), 1.64–1.52 (m, 1 H), 1.41 (dt, J = 11.6, 6 Hz, 1
H), 1.05 (d, J = 6.8 Hz, 3 H) ppm. 13C NMR (100 MHz, CDCl3):
δ = 162.6, 148.7, 134.2, 134.0, 129.3, 122.2, 120.2, 70.3, 36.9, 35.8,
30.5, 27.9, 27.0, 23.3, 13.8 ppm. MS (EI): m/z (%) = 348 (1) [M+],
195 (19), 149 (15), 136 (34), 121 (33), 107 (30), 94 (100), 79 (25).
C17H20N2O6 (348.35): calcd. C 58.61, H 5.79, N 8.04; found C
58.80, H 5.80, N 8.00.
(4S,8R)-p-Menth-1-en-9-ol (5a): 98% chemical purity by GC, up to
97% de by NMR analysis, [α]2D0 = –97 (c = 1, CHCl3), ref.[3a] [α]2D0
1
= –94.0 (c = 0.92, CHCl3). IR, H NMR, 13C NMR and MS data
in accordance with those of alcohol 5a obtained by the baker’s
yeast reduction procedure. C10H18O (154.25): calcd. C 77.87, H
11.76; found C 77.95, H 11.75.
A sample of the ester 2b (8.7 g, 25 mmol) was hydrolyzed with
KOH (2.5 g, 44 mmol) in refluxing methanol (40 mL). After
workup and bulb-to-bulb distillation (oven temperature 75–80 °C/
0.5 Torr), pure alcohol 2a (3.65 g, 23.7 mmol, 95%) was obtained.
Procedure for the Preparation of 1-Hydroxy-2,9-cineole (6–9) Start-
ing from Alcohols 2–5: A solution of alcohol 2a (3 g, 19.5 mmol) in
CH2Cl2 (50 mL) was treated with MCPBA (6.67 g of a 75% wet
acid, 29 mmol) whilst stirring at 0 °C until no more 2a was detected
by TLC analysis (3 h). The reaction mixture was then treated with
a saturated aqueous solution of NaHSO3 (20 mL) and stirred at
room temperature for 3 h. The mixture was diluted with water
(80 mL) and extracted with CH2Cl2 (2ϫ80 mL). The organic layer
was washed in turn with saturated NaHCO3 solution (100 mL) and
brine (100 mL), and dried (Na2SO4) and concentrated under re-
duced pressure. The residue was purified by CC (hexane/Et2O, 9:1)
and crystallized from hexane to give pure cineole 6 (0.98 g,
5.8 mmol, 30%).
According to the procedure described above, the hydrolysis of a
sample of the ester 3b (6.3 g, 18 mmol) afforded pure 3a (2.55 g,
92%).
(4R,8S)-p-Menth-1-en-9-ol (2a): 98% chemical purity by GC, up to
98% de by NMR analysis [α]2D0 = +102 (c = 1, CHCl3), ref.[3a]
[α]2D0 = +97.27 (c = 1.033, CHCl3), ref.[3c] [α]2D0 = +103.1 (neat). IR,
1H NMR, 13C NMR and MS data in accordance with those of
alcohol 5a. C10H18O (154.25): calcd. C 77.87, H 11.76; found C
77.80, H 11.75.
(1R,2R,4R,8S)-1-Hydroxy-2,9-cineole (6): 98% chemical purity by
(4R,8R)-p-Menth-1-en-9-ol (3a): 98% chemical purity by GC, up to
98% de by NMR analysis, [α]2D0 = +106 (c = 1, CHCl3), IR, 1H
NMR, 13C NMR and MS data in accordance with those of alcohol
3a obtained by the baker’s yeast reduction procedure. C10H18O
(154.25): calcd. C 77.87, H 11.76; found C 77.75, H 11.71.
GC, m.p. 82–83 °C. [α]2D0 = –79.1 (c = 1, CHCl3), ref.[5a] [α]2D0
=
–65.5 (c = 1.78, CHCl ). IR (neat): ν = 3325, 3267, 1457, 1372,
˜
3
1212, 998, 898 cm–1. H NMR (400 MHz, CDCl3): δ = 3.63 (dd, J
1
= 11.4, 6.6 Hz, 1 H), 3.54 (br. d, J = 4.5 Hz, 1 H), 3.31 (t, J =
11.4 Hz, 1 H), 2.01–1.90 (m, 2 H), 1.87–1.71 (m, 3 H), 1.54 (br. s,
1 H), 1.43–1.32 (m, 2 H), 1.21 (s, 3 H), 1.20 (br. s, 1 H), 0.85 (d, J
= 6.9 Hz, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 74.9, 71.4,
66.5, 33.5, 31.7, 30.8, 28.1, 27.9, 21.4, 18.4 ppm. MS (EI): m/z (%)
= 170 (9) [M+], 152 (11), 137 (4), 134 (5), 123 (6), 110 (100), 97
(90), 79 (8), 71 (19), 55 (12), 43 (26). C10H18O2 (170.25): calcd. C
70.55, H 10.66; found C 70.60, H 10.65.
The separation procedure described above was repeated starting
from the mixture of alcohol 4a and 5a (17 g, 110 mmol). The PPL-
mediated acetylation was stopped at about 48% conversion to give
acetate 4c (10.1 g, 51 mmol, 46%, 79% de by chiral GC analysis)
and unreacted alcohol 5a (8.3 g, 54 mmol, 49%, 73% de by chiral
GC analysis of the corresponding acetate). Acetate 4c was hy-
drolyzed, and both the obtained alcohol and alcohol 5a were then
converted to the corresponding 3,5-dinitrobenzoyl esters, which
were then crystallized twice from hexane to give pure 4b (11.5 g,
33 mmol, 65%) and 5b (10.4 g, 30 mmol, 55%), respectively.
The procedure described above was repeated starting from alcohol
3a (2.8 g, 18.2 mmol), 4a (3.1 g, 20.1 mmol) and 5a (3.5 g,
22.7 mmol) to give cineole 7 (0.9 g, 5.3 mmol, 29%), 8 (0.95 g,
5.6 mmol, 28%) and 9 (1.04 g, 6.1 mmol, 27%), respectively.
(4S,8S)-p-Menth-1-en-9-yl 3,5-Dinitrobenzoate (4b): 98% chemical
purity by GC, up to 97% de by NMR analysis, m.p. 94–95 °C.
[α]2D0 = –33 (c = 1, CHCl3), ref.[3a] [α]2D0 = –34.0 (c = 3.27, CHCl3).
IR, 1H NMR, 13C NMR and MS data in accordance with those
of 3b. C17H20N2O6 (348.35): calcd. C 58.61, H 5.79, N 8.04; found
C 58.75, H 5.75, N 7.99.
(1R,2R,4R,8R)-1-Hydroxy-2,9-cineole (7): 99% chemical purity by
GC, m.p. 92–93 °C. [α]2D0 = –32.5 (c = 1, CHCl3), ref.[5a] [α]2D0
=
–22.4 (c = 1.19, CHCl ). IR (neat): ν = 3311, 3254, 1444, 1372,
˜
3
1236, 1068, 989, 913, 827 cm–1. H NMR (400 MHz, CDCl3): δ =
1
3.68 (dd, J = 11.8, 7.3 Hz, 1 H), 3.52 (t, J = 11.8 Hz, 1 H), 3.44
(br. s, 1 H), 2.19–2.05 (m, 2 H), 1.85–1.60 (m, 6 H), 1.35 (br. s, 1
H),1.28 (s, 3 H), 0.85 (d, J = 6.9 Hz, 3 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 74.6, 71.2, 68.2, 36.1, 33.8, 30.3, 30.0, 28.4,
21.8, 15.4 ppm. MS (EI): m/z (%) = 170 (8) [M+], 152 (7), 134 (3),
137 (3), 123 (5), 110 (100), 97 (43), 71 (18), 55 (7), 43 (21).
(4S,8R)-p-Menth-1-en-9-yl 3,5-Dinitrobenzoate (5b): 98% chemical
purity by GC, up to 98% de by NMR analysis, m.p. 72–73 °C.
[α]2D0 = –43 (c = 1, CHCl3), ref.[3a] [α]2D0 = –39.91 (c = 1.152, CHCl3).
IR, 1H NMR, 13C NMR and MS data in accordance with those
1036
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 1031–1037

Chemoenzymatic Synthesis of the Isomeric Forms of p-Menth-1-en-9-ol
[5] a) R. M. Barman, A. C. Garner, Aust. J. Chem. 1995, 48, 1301–
1309; b) K. Yoshihara, T. Sakai, T. Sakan, Chem. Lett. 1978,
433–434; c) E. E. van Tamelen, R. J. Anderson, J. Am. Chem.
Soc. 1972, 94, 8225–8228.
[6] D. Joulain, C. Fuganti, S. Serra, A. Vecchione, International
Patent WO2005/1210 58 A1, 2005.
[7] a) E. Brenna, C. Fuganti, F. G. Gatti, M. Perego, S. Serra, Tet-
rahedron: Asymmetry 2006, 17, 792–796; b) S. Serra, C. Fu-
ganti, Helv. Chim. Acta 2004, 87, 2100–2109; c) S. Serra, E.
Brenna, C. Fuganti, F. Maggioni, Tetrahedron: Asymmetry
2003, 14, 3313–3319; d) S. Serra, C. Fuganti, Helv. Chim. Acta
2002, 85, 2489–2502.
C10H18O2 (170.25): calcd. C 70.55, H 10.66; found C 70.63, H
10.67.
(1S,2S,4S,8S)-1-Hydroxy-2,9-cineole (8): 97% chemical purity by
GC, m.p. 91–92 °C. [α]2D0 = +30.9 (c = 1, CHCl3). IR, 1H NMR,
13C NMR and MS data in accordance with those of 7. C10H18O2
(170.25): calcd. C 70.55, H 10.66; found C 70.75, H 10.62.
(1S,2S,4S,8R)-1-Hydroxy-2,9-cineole (9): 98% chemical purity by
GC, m.p. 80–81 °C. [α]2D0 = +78.4 (c = 1, CHCl3). IR, 1H NMR,
13C NMR and MS data in accordance with those of 6. C10H18O2
(170.25): calcd. C 70.55, H 10.66; found C 70.70, H 10.65.
[8] a) E. Brenna, C. Dei Negri, C. Fuganti, F. G. Gatti, S. Serra,
Tetrahedron: Asymmetry 2004, 15, 335–340; b) C. Fuganti, S.
Serra, J. Chem. Soc. Perkin Trans. 1 2000, 3758–3764; c) C.
Fuganti, S. Serra, J. Chem. Soc. Perkin Trans. 1 2000, 97–101;
d) C. Fuganti, S. Serra, A. Dulio, J. Chem. Soc. Perkin Trans.
1 1999, 279–282.
[1] M. C. Pirrung, A. T. Morehead in The Total Synthesis of Natu-
ral Products, Vol. 10 (Ed.: D. Goldsmith), John Wiley & Sons,
New York, 1997, pp. 45–77.
[2] K. H. Schulte-Elte, G. Ohloff, Helv. Chim. Acta 1966, 49,
2150–2157.
[3] a) B. A. Pawson, H.-C. Cheung, S. Gurbaxani, G. Saucy, J.
Am. Chem. Soc. 1970, 92, 336–343; b) J. F. Blount, B. A. Paw-
son, G. Saucy, J. Chem. Soc. C 1969, 715; c) G. Ohloff, W.
Giersch, K. H. Schulte-Elte, Helv. Chim. Acta 1969, 52, 1531–
1536.
[9] S. Serra, C. Fuganti, Tetrahedron: Asymmetry 2001, 12, 2191–
2196.
[10] R. J. Crawford, W. F. Erman, C. D. Broaddus, J. Am. Chem.
Soc. 1972, 94, 4298–4306.
[11] D. E. Chastain, N. Mody, G. Majetich, US Patent 5574195,
1996.
[12] I. Ojima, T. Kogure, Organometallics 1982, 1, 1390–1399.
[4] a) P. S. Baran, T. J. Maimone, J. M. Richter, Nature 2007, 446,
404–408; b) R. M. Cravero, M. Gonzales-Sierra, G. R. La ba- [13] a) A. Abate, E. Brenna, C. Dei Negri, C. Fuganti, S. Serra,
die, Helv. Chim. Acta 2003, 86, 2741–2753; c) S. Reichert, A.
Mosandl, J. High Resolut. Chromatogr. 1999, 22, 631–634; d)
H. Guth, Helv. Chim. Acta 1996, 79, 1559–1571; e) T. Duvold,
G. W. Francis, D. Papaioannou, Tetrahedron Lett. 1995, 36,
3153–3156; f) I. Blank, W. Grosch, W. Eisenreich, A. Bacher,
J. Firl, Helv. Chim. Acta 1990, 73, 1250–1257; g) D. R. Wil-
liams, J. G. Phillips, J. Org. Chem. 1981, 46, 5452–5455.
Tetrahedron: Asymmetry 2002, 13, 899–904; b) A. Basak, A.
Nag, G. Bhattacharya, S. Mandal, S. Nag, Tetrahedron: Asym-
metry 2000, 11, 2403–2407.
[14] S. Saeidnia, A. R. Gohari, N. Uchiyama, M. Ito, G. Honda, F.
Kiuchi, Chem. Pharm. Bull. 2004, 52, 1249–1250.
Received: October 25, 2007
Published Online: December 13, 2007
Eur. J. Org. Chem. 2008, 1031–1037
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1037