The structure of laricinolic acid and its biomimetic transformation into officinalic acid

Article abstract of DOI:10.1039/b001984o

Laricinolic acid (8), a new sesquiterpene of the drimane type, has been isolated from the wood-rotting fungus Laricifomes officinalis. Its structure was elucidated by spectroscopic means and confirmed via a correlation with the known drimenine derivative 13. Oxidation of 8 to 1, followed by a mild thermal treatment, furnished (-)-officinalic acid (4) in 65% yield. This transformation establishes the hitherto unknown absolute configuration of the latter. An independent correlation was achieved by pyrolysis of 4 which furnished (-)-dihydrooxoisodrimenine (14) of known absolute configuration.

Full text of DOI:10.1039/b001984o

View Article Online / Journal Homepage / Table of Contents for this issue
The structure of laricinolic acid and its biomimetic transformation
into officinalic acid
Bernhard Erb, Hans-Jürg Borschberg and Duilio Arigoni*
1
Laboratorium für Organische Chemie der ETH Zürich, Universitätstrasse 16, CH-8092 Zürich,
Switzerland
Received (in Cambridge, UK) 13th March 2000, Accepted 20th March 2000
Published on the Web 4th May 2000
Laricinolic acid (8), a new sesquiterpene of the drimane type, has been isolated from the wood-rotting fungus
Laricifomes officinalis. Its structure was elucidated by spectroscopic means and confirmed via a correlation with the
known drimenine derivative 13. Oxidation of 8 to 1, followed by a mild thermal treatment, furnished (Ϫ)-officinalic
acid (4) in 65% yield. This transformation establishes the hitherto unknown absolute configuration of the latter. An
independent correlation was achieved by pyrolysis of 4 which furnished (Ϫ)-dihydrooxoisodrimenine (14) of known
absolute configuration.
the crystallisation of 4 we noted the presence of a crystalline,
Introduction
optically active acid with the elemental composition C15H24O3,
mp 213 ЊC, [α]D = Ϫ30.8, for which we propose the trivial name
laricinolic acid. Comparison with the data on record for the so-
called “Substanz 12” (mp 215–218 ЊC, [α]D = Ϫ28.6), previously
isolated from the same source by Schulte et al.,4a suggests that
the two compounds may be identical, in spite of some minor
Some time ago, we reported a biomimetic synthesis of the
racemic form of officinalic acid (4) (Scheme 1),1 a metabolite
first isolated in 1893 by Jahns2 from the fungus Laricifomes
officinalis. Its structure was established many years later by
Epstein et al.3 through a single crystal X-ray analysis of a tetra-
hydro derivative. In the key step of our synthesis a hetero-
Diels–Alder reaction of the racemic forms of the enone acid 1
yielded a 5:1-mixture of the homodimeric enol ether 2 and
its heterodimeric counterpart 3, which underwent in situ
lactonisation to officinalic acid (4) and its diastereoisomer 6,
respectively.
1
deviations in their IR and H-NMR spectra. Unfortunately, a
reference sample of “Substanz 12” is no longer available for
direct comparison.4b
According to its IR and H-NMR spectra, laricinolic acid
1
contains a carboxy group and a secondary hydroxy group, as
well as three quaternary methyl groups and an exo-methylene
unit. The off-resonance decoupled 13C-NMR spectrum con-
firms the presence of these groups and, in addition, displays
signals for three methine units, four saturated methylene groups
and two quaternary carbon atoms.
The working hypothesis that the natural product represents a
new drimane sesquiterpene was substantiated through analysis
of the 1H-NMR spectrum of the corresponding methyl ester 9,
Results and discussion
For an unambiguous identification of our racemic synthetic
officinalic acid, a reference sample of enantiomerically pure 4
was isolated from a commercially available extract of Laricifo-
mes officinalis. During an examination of mother liquors from
Scheme 1
DOI: 10.1039/b001984o
J. Chem. Soc., Perkin Trans. 1, 2000, 2307–2309
This journal is © The Royal Society of Chemistry 2000
2307

View Article Online
Scheme 2
officinalic acid (4) is assembled from two enone units (1) of
opposite absolute configuration.
A second configurational correlation was later discovered by
the observation that officinalic acid (4), when heated above its
melting point, decomposes to a mixture containing 1 and the
known (Ϫ)-dihydrooxodrimenine (14).6 This transformation,
which provides independent proof for the absolute configur-
ation of 4, must involve an acid-catalysed opening of the
lactone ring, followed by a retro-Diels–Alder reaction of the
resulting enol ether 2 to provide 1 which then undergoes an
unusual7 5-endo-trig-lactonisation to 14.
Fig. 1 Torsional angles as calculated by PM3 (MOPAC) and observed
1H-NMR coupling constants of 9. Unlabelled substituents represent
hydrogen atoms.
which was corroborated by extensive decoupling experiments
(see Fig. 1). Specifically, the axially oriented C(7)-hydrogen of
9 is responsible for the appearance of a doublet of doublets,
centred at δ 4.06 ppm, whereas in the racemic compound
( )-10 (Scheme 2) with an axial arrangement of the hydroxy
group the signal of the equatorial C(7)-hydrogen was reported
as a broad singlet at δ 4.40 ppm.5
The problems of the relative and absolute configuration at
the C(5) and C(10) centres of laricinolic acid could be solved via
a two-step correlation with the known (ϩ)-7-oxodrimenine
(13).6 This transformation involved an acid-catalysed dehydra-
tion of 8 to give a mixture of drimenine (11) and isodrimenine
(12)† which was oxidised with chromic acid to give 13. As the
sign of the specific rotation of our sample was the same as that
reported by the Overton group, the absolute configuration of
laricinolic acid (8) must be as shown in Scheme 2. During the
transformation of 8 to 13 the original stereogenic centre at C(9)
of 8 is destroyed, but the correlations reported below provide
convincing evidence for the equatorial arrangement of the
carboxy group of 8.
The availability of 8 paved the way for a configurational
correlation with natural officinalic acid (4) through a biomi-
metic synthesis of the latter. To this end, compound 8 was
oxidised to the crystalline but rather unstable enone 1, which
was subjected without purification to mild thermal treatment
(3 h reflux in toluene). Chromatography of the resulting crude
product furnished a pure compound, identical in every respect
with the naturally occurring (Ϫ)-officinalic acid (4), thus estab-
lishing the hitherto unknown absolute configuration of this
compound. The yield obtained in the conversion of 1 into 4
(65% over both steps) is considerably higher than the one previ-
ously observed for the dimerisation of racemic 1 (46%); in
addition, no trace of a compound corresponding to iso-
officinalic acid (6) could be detected in this experiment, a find-
ing that can be taken as an indirect, but nevertheless compelling
argument in support of the claim that this stereoisomer of
Conclusion
The co-occurence of laricinolic acid (8) and officinalic acid (4)
in Laricifomes officinalis and the demonstration that 1, the
oxidised form of 8, undergoes specific dimerisation to 4 lends
credence to the hypothesis that this transformation is an excel-
lent mimic for the formation of 4 in the fungus. The specificity
and relative ease of this dimerisation raises the interesting
question of whether or not this process requires participation
of a specific “Diels–Alderase” in the fungus.
Experimental
Melting points were determined in sealed evacuated capillaries.
Specific rotations were measured in a 10 cm quartz tube on a
Perkin-Elmer polarimeter, model 141. [α]D has units of 10Ϫ1 deg
cm2 gϪ1. IR spectra were recorded on a Perkin-Elmer, model
125, as 3–5% solutions in chloroform, unless stated otherwise.
1H-NMR spectra were recorded in CDCl3 on a Bruker WM 300
spectrometer (300 MHz), unless stated otherwise. 13C-NMR
spectra were recorded in CDCl3 on a Bruker WM 300 spec-
trometer (75 MHz), unless stated otherwise. J values are given
in Hz. Mass spectra were recorded on a Hitachi Perkin-Elmer
spectrometer (model RMU-6A) using EI (70 eV); relative
intensities are given in % of the base peak.
Isolation of laricinolic acid (8)
Continuous Soxhlet extraction of 2.63 kg of pulverised Larici-
fomes officinalis (Siegfried AG, Zofingen, Switzerland) with
low-boiling petroleum ether for 10 days furnished 106.2 g of a
crude extract which was dissolved in 200 ml of petroleum ether
and kept at 4 ЊC for 2 weeks. The resulting precipitate (23.9 g)
was suspended in 200 ml of a warm toluene–ethyl acetate mix-
ture (10:1). The insoluble part (1.21 g) was discarded and the
residue of the evaporated solution was purified by chromato-
graphy using toluene–ethyl acetate (20:1 to 1:1) as eluent. The
fractions with rF = 0.26–0.52, toluene–ethyl acetate (1:1), were
pooled, resulting in 2.54 g of an amorphous material. Crystal-
† In keeping with the equatorial orientation of the hydroxy group,
compound 8 proved more resistant to dehydration than its epimeric
racemic counterpart 10.
2308
J. Chem. Soc., Perkin Trans. 1, 2000, 2307–2309

View Article Online
lisation from benzene–acetone gave 186 mg of crude 4 which
was recrystallised from CH2Cl2–acetone to give 78 mg of pure
4. The combined mother liquors were dissolved in ether and
extracted with aq. NaHCO3 to give 142 mg of an acidic fraction
which after crystallisation from hexane–acetone and recrystal-
lisation from CHCl3–acetone delivered 78 mg of pure 8, [α]D =
Ϫ30.8 (c, 1.12, EtOH); mp 213 ЊC (Found: C, 71.23; H, 9.39.
C15H24O3 requires C, 71.39; H 9.59%); νmax (KBr)/cmϪ1 3600–
2500, 2950, 2920, 2860, 1714, 1694, 1448, 1460, 1368, 1308,
1211, 1183, 1088, 905, 830, 688, 653; δH (d6-acetone) 0.86 (3H,
s, Me), 0.92 (3H, s, Me), 1.05 (3H, s, Me), 1.16–1.29 (3H, m,
including 1.22 (1H, dd, J 12.9, 2.2), 1.35 (1H, ddd, J 12.9, 12.0,
10.8), 1.40–1.53 (6H, m), 2.04 (1H, ddd, J 12.0, 5.6, 2.2), 2.73
(1H, br s), 4.04 (1H, br dd, J 10.8, 5.6), 4.90 (1H, m), 5.26 (1H,
m); δC (d6-acetone) 14.5 (q), 19.6 (t), 22.0 (q), 33.1 (t), 33.7 (q),
33.8 (s), 39.3 (t), 39.6 (s), 42.5 (t), 53.0 (d), 61.4 (d), 72.8 (d),
105.5 (t), 147.7 (s), 172.5 (s); m/z (EI) 252 (7%), 234 (41), 219
(25), 191 (16), 189 (10), 173 (12), 137 (56), 123 (100), 119 (24),
109 (48), 105 (23), 95 (39), 81 (45), 69 (79).
and m/z values were identical with those obtained previously for
racemic 13.1
Transformation of laricinolic acid (8) into officinalic acid (4)
To a solution of 23 mg (0.091 mmol) of 8 in 3.5 ml of acetone
were added 0.1 ml (5.9 equiv.) of Jones’ reagent at 0 ЊC. After
stirring for 5 min, excess reagent was destroyed by addition of a
few drops of EtOH. The mixture was poured onto crushed ice
and extracted 3 times with ether. The combined extracts were
washed with brine, dried (MgSO4), filtered and evaporated to
yield 24 mg of a crystalline residue (mp 233–235 ЊC, decomp.),
1
consisting of essentially pure 1 which showed UV, IR and H-
NMR data identical with those observed for racemic 1.1 This
material was dissolved in 3 ml of toluene and heated for 3 h at
reflux. The residue was recrystallised from CH2Cl2–acetone to
yield 15 mg (0.03 mmol, 65%) of 4 in the form of long needles,
mp 268 ЊC (mixed mp with natural 4 267 ЊC). The mother
liquors were treated with diazomethane and analysed by TLC,
toluene–ethyl acetate (20:1). The major compound was shown
to be 5 (rF = 0.28), and no methyl isoofficinalate (7) (rF = 0.21)
could be detected.
Methyl laricinolate (9)
A solution of 26.2 mg (0.1 mmol) of 8 in 5 ml of CH2Cl2 was
treated with a slight excess of a freshly distilled ethereal solu-
tion of diazomethane. The solvents were removed and the
residue distilled (95 ЊC/0.01 Torr) to yield 26 mg (0.098 mmol,
98%) of a colourless viscous oil, [α]D = Ϫ19.7 (c, 1.37, EtOH)
(Found: C, 71.89; H, 9.70. C16H26O3 requires C, 72.14; H
9.84%); νmax (CHCl3)/cmϪ1 3610, 2955, 2850, 1735, 1650, 1435,
1392, 1371, 1265, 1164, 1019, 908; δH 0.84 (3H, s, Me), 0.90 (3H,
s, Me), 1.05 (3H, s, Me), 1.10–1.28 (3H, m, including 1.13 (1H,
dd, J 12.8, 2.5)), 1.35 (1H, ddd, J 12.8, 12.2, 10.8), 1.42–1.70
(4H, m), 1.90 (1H, br s OH), 2.09 (1H, ddd, J 12.2, 5.7, 2.5),
2.72 (1H, br s), 3.66 (3H, s OMe), 4.06 (1H, br dd, J 10.8, 5.7),
4.90 (1H, m), 5.19 (1H, m); δC 14.2 (q), 18.9 (t), 21.7 (q), 33.1
(t), 33.3 (s), 33.4 (q), 38.6 (t), 39.3 (s), 41.8 (t), 51.1 (q), 52.6 (d),
61.1 (d), 72.8 (d), 105.4 (t), 145.8 (s), 171.4 (s); m/z (EI) 266
(12%), 248 (29), 233 (17), 205 (10), 189 (15), 173 (19), 137 (43),
129 (74), 123 (100), 119 (29), 109 (36), 97 (58), 81 (50), 69 (84).
Thermolysis of officinalic acid (4)
Crystalline 4 (17 mg, 0.034 mmol) was heated for 5 min in a
small glass tube under argon in a sublimation block, preheated
to 275–280 ЊC. TLC-analysis of the crude material indicated the
presence of some starting material, and of the known com-
pounds 1 and 14. Chromatography of the cooled mixture using
hexane–ethyl acetate (3:1) as eluent gave 6 mg (0.024 mmol,
35%) of 14 which was recrystallised from hexane to yield 4.2 mg
of colourless needles, [α]D = Ϫ98.6 (c, 0.43, benzene) (lit.6 Ϫ115
(c, 1.0, benzene)); mp 124 ЊC (lit.6 124–126 ЊC); νmax, δH, and
m/z values were identical with those obtained previously for
racemic 14.1
Acknowledgements
Dedicated to the memory of Professor Leslie Crombie.
Financial support from Novartis AG (formerly Sandoz AG) is
gratefully acknowledged.
Transformation of laricinolic acid (8) into 6-oxoisodrimenine (13)
To a solution of 38 mg (0.15 mmol) of 8 in 5 ml of 1,4-dioxane
were added 0.7 ml of 6 M H2SO4. After refluxing for 24 h the
cooled solution was treated with 30 ml of saturated aq.
NaHCO3 solution and the mixture extracted 3 times with ether.
The combined extracts were dried (MgSO4), filtered and evap-
orated to yield 22 mg of a neutral fraction, containing a mix-
ture of 11 and 12 (TLC-evidence). This mixture was dissolved
in 1.2 ml of AcOH and treated with 0.2 ml of Beckmann’s
mixture (10 g K2Cr2O7, 8.7 g conc. H2SO4 and 50 ml H2O).
After stirring at 25 ЊC for 16 h the mixture was poured onto
crushed ice and extracted with ether. Preparative thin layer
chromatography of the resulting crude mixture using hexane–
ethyl acetate (5:1) as eluent furnished 4.5 mg (0.018 mmol,
12%) of 13. Crystallisation from ether–hexane gave 3.6 mg of
colourless prisms, [α]D = ϩ41.6 (c, 0.14, benzene) (lit.6 ϩ52
References
1 B. Erb, H.-J. Borschberg and D. Arigoni, Croat. Chem. Acta, 1985,
58, 653.
2 E. Jahns, Arch. Pharm., 1883, 221, 260.
3 W. W. Epstein, F. W. Sweat, G. VanLear, F. M. Lovell and E. J. Gabe,
J. Am. Chem. Soc., 1979, 101, 2748.
4 (a) K. E. Schulte, G. Rücker and H. Fachmann, Arch. Pharm., 1969,
302, 965; (b) K. E. Schulte, University of Münster, Germany,
personal communication.
5 Y. Kitahara, T. Kato, T. Suzuki, S. Kano and M. Tanemura, Chem.
Commun., 1959, 342.
6 H. H. Appel, J. D. Connolly, K. H. Overton and (in part) R. P. M.
Bond, J. Chem. Soc., 1960, 4685.
7 (a) J. E. Baldwin, J. Chem. Soc., Chem. Commun., 1976, 734;
(b) J. E. Baldwin, Tetrahedron, 1982, 19, 2939.
(c, 1.89, benzene)); mp 112 ЊC (lit.6 112–113 ЊC); νmax, δH, λmax
,
J. Chem. Soc., Perkin Trans. 1, 2000, 2307–2309
2309