Total synthesis of valerenic acid, a potent gabaa receptor modulator

Article abstract of DOI:10.1021/ol9000137

The first total synthesis of the sesquiterpenoid valerenic acid, a constituent of Valeriana officinalis, is described. The compound is a potent modulator of the GABAA receptor and may thus be useful in the treatment of various dysfunctions of the central

Full text of DOI:10.1021/ol9000137

ORGANIC
LETTERS
2009
Vol. 11, No. 5
1151-1153
Total Synthesis of Valerenic Acid, a
Potent GABAA Receptor Modulator
Ju¨rgen Ramharter and Johann Mulzer*
UniVersity of Vienna, Institute of Organic Chemistry, Wa¨hringer Strasse 38,
1090 Wien, Austria
johann.mulzer@uniVie.ac.at
Received January 5, 2009
ABSTRACT
The first total synthesis of the sesquiterpenoid valerenic acid, a constituent of Valeriana officinalis, is described. The compound is a potent
modulator of the GABAA receptor and may thus be useful in the treatment of various dysfunctions of the central nervous system. The synthesis
is enantio-, diastereo-, and regiocontrolled and utilizes an enyne-RCM, a metal-coordinated Diels-Alder reaction, a hydroxy-directed Crabtree
hydrogenation, and a Negishi methylation as key steps.
γ-Aminobutyric acid (GABA) is the principal inhibitory
neurotransmitter in the mammalian brain and mediates its
action by interaction with GABA type A (GABAA) receptors
representing a ligand-gated chloride channel, second mes-
senger linked GABAB receptors that are coupled to Ca2+-
and K+-channels via G-proteins and GABAC receptors.1
There is clear evidence for specific expression of certain
GABAA receptor subtypes (with different subunit composi-
tions) in different regions in the central nervous system
(CNS).2 The subunit composition determines the GABA-
sensitivity and the pharmacological properties of the GABAA
receptor.3 Searching for molecules selective for a certain
GABAA receptor subtype is, therefore, a promising approach
to develop new therapeutics.
Valerenic acid (1), isolated from the roots of Valeriana
officinalis,4 has recently been shown to selectively modulate
GABAA receptor subtypes.5 These data suggest that 1 is a
subunit-specific allosteric modulator of GABAA receptors and
as a drug, might be applicable against a variety of CNS
dysfunctions, such as panic disorders, hyperactivity, poor
motor coordination, learning deficits, spontaneous epileptic
seizures, and abnormal facial development. Therefore, as
more extensive and elaborate structural variations are envis-
aged, an efficient total synthesis of 1 appeared of importance.
This intention seemed all the more worthwhile to us as,
despite some fruitless efforts in the past,6 no total synthesis
of 1 has been reported so far.
(1) (a) Barnard, E. A.; Skolnick, P.; Olsen, R. W.; Mohler, H.; Sieghart,
W.; Biggio, G.; Braestrup, C.; Bateson, A. N.; Langer, S. Z. Pharmacol.
ReV 1998, 50, 291–313. (b) Bormann, J. Trends Pharmacol. Sci. 2000, 21,
16–19.
(2) (a) Korpi, E. R.; Grunder, G.; Luddens, H. Prog. Neurobiol. 2002,
67, 113–159. (b) Pirker, S.; Schwarzer, C.; Wieselthaler, A.; Sieghart, W.;
Sperk, G. Neuroscience 2000, 101, 815–850.
(4) Stoll, A.; Seebeck, E. Justus Liebigs Ann. Chem. 1957, 603, 158–
168.
(5) Khom, S.; Baburin, I.; Timin, E. N.; Hohaus, A.; Sieghart, W.;
Hering, S. Mol. Pharmacol. 2002, 69, 640–649.
(3) (a) Hevers, W.; Luddens, H. Mol. Neurobiol. 1998, 18, 35–86. (b)
Sieghart, W.; Sperk, G. Curr. Top. Med. Chem. 2002, 2, 795–816. (c)
Boileau, A. J.; Li, T.; Benkwitz, C.; Czajkowski, C.; Pearce, R. A.
Neuropharmacology 2003, 44, 10031012.
(6) (a) Joseph, K. T.; Krishna, R. G. S. Tetrahedron 1967, 23, 3215–
3220. (b) Bohlmann, F.; Lonitz, M. Chem. Ber. 1980, 113, 2410–2423. (c)
Baudouy, R.; Sartoretti, J.; Choplin, F. Tetrahedron 1983, 39, 3293–3305.
10.1021/ol9000137 CCC: $40.75
Published on Web 01/29/2009
2009 American Chemical Society

Scheme 2
.
Stereochemical Implications of the IMDA Reaction
of 8
Figure 1. Structural features of valerenic acid (1).
Structurally, 1 is a sesquiterpene featuring an indanyl core
with three stereogenic centers and two double bonds (Figure
1). A particular challenge in a total synthesis lies in the
diaxial arrangement7 of two onring substituents, which had
to be installed without subsequent epimerization.6b
Scheme 1. Retrosynthetic Analysis of 1
The IMDA reaction of ester-linked trienes such as 8 is
known to proceed exclusively via the so-called trans () exo)
transition state 9a (Scheme 2).10,11 Because of great distortion
and steric hindrance the electronically normally preferred cis
() endo) transition state 9b is clearly disfavored and, thus,
product 10 should be formed. On the other hand, this
outcome would be contra-thermodynamic, as 10, according
to STO 3-21G calculations, is 13.2 kcal·mol-1 higher in
energy than product 11.
Our synthesis (Scheme 3) started with readily available
racemic 3-OTBS-oct-1-en-6-yne (7),12 which smoothly un-
derwent RCM to cyclopentene 12 with Grubbs’ first genera-
tion catalyst, if the reaction was carried out under an ethylene
atmosphere (Mori’s conditions).13 Several other conditions
such as using an argon atmosphere instead of ethylene or
PtCl2 as catalyst led to no conversion or decomposition. In
situ deprotection and acylation furnished acrylic ester 8 as
the envisaged IMDA substrate. However, all attempts to
achieve cycloaddition under thermal conditions just led to
decomposition. After this failure we switched to a metal-
coordinated Diels-Alder reaction14 of alcohol 6 and methyl
acrylate with MgBr2 as the template. This reaction furnished
lactone 11 stereoselectively.15 The transition state (13) of
this Diels-Alder reaction now resembles the electronically
favored endo-arrangement 9b and furnishes lactone 11 as
the thermodynamically more stable product, so that all
disadvantages of the covalently tethered acrylate are now
reversed.
From the retrosynthetic perspective (Scheme 1), an obvious
route to 1 might start from a pulegone-derived cyclohexanone
such as 2 and attach the cyclopentene ring via alkylation to
3 and olefination/ring closing metathesis (RCM) to 4.
However, enolate additions to 2 are known8 to favor the
trans-diastereomer, which would generate 4 with the wrong
configuration at the ring juncture. Therefore, we decided to
prepare the bicyclic core from an acyclic precursor 7 via an
enyne-RCM9-IMDA (IMDA ) intramolecular Diels-Alder
addition) sequence via 6 to generate lactone 5. The relative
configuration at C-7/7a should be created via hydroxy-
directed catalytic hydrogenation. As the ultimate steps of our
synthesis we considered a Negishi-coupling to introduce the
methyl substituent at C-3, whereas the exocyclic enoate
appendage at C-4 should be formed by an E-selective Wittig
olefination. This plan had two main unknowns: (a) the
stereochemical outcome of the IMDA reaction generating 5
and (b) the possibility to achieve the correct relative
configuration at C-7/7a with respect to C-4.
The synthesis was continued (Scheme 4) by installing the
side chain at C-4 via reduction to the lactol 14 (not isolated)
(10) Strekowski, L.; Kong, S.; Battiste, M. A. J. Org. Chem. 1988, 53,
901–904.
(11) Cayzer, T. N.; Paddon-Row, M. N.; Moran, D.; Payne, A. D.;
Sherburn, M. S.; Turner, P. J. Org. Chem. 2005, 70, 5561–5570.
(12) Adrio, J.; Rodr´ıguez Rivero, M.; Carretero, J. C. Angew. Chem.,
Int. Ed. 2000, 39, 2906–2909.
(13) Kinoshita, A.; Sakakibara, N.; Mori, M. J. Am. Chem. Soc. 1997,
119, 12388–12389.
(7) Cf. the crystal structure of valerenolic acid: Birnbaum, G. L.; Findlay,
J. A.; Krepinsky, J. L. J. Org. Chem. 1978, 43, 272–276.
(8) Touriya, Z.; Santelli-Rouvier, C.; Santelli, M. J. Org. Chem. 1993,
58, 2686–2693.
(14) Barriault, L.; Thomas, J. O. D.; Clement, R. J. Org. Chem. 2003,
68, 2317–2323.
(15) The relative configuration of 11 was secured via 2D NMR
techniques (see the Supporting Information).
(9) Diver, S. T.; Giessert, A. J. Chem. ReV. 2004, 104, 1317–1382.
1152
Org. Lett., Vol. 11, No. 5, 2009

Scheme 3
.
Preparation of Lactone 11a
Scheme 4.
Completion of the Synthesisa
a Abbreviations: DCM ) dichloromethane, DIPEA ) diisopropylethy-
lamine.
and Wittig reaction to form enoate 15. The free OH group
was now used as an anchoring group for Crabtree’s catalyst,16
which delivered dihydrogen in a syn fashion to generate
hydrindane 16 stereo- and chemoselectively. Oxidation to
ketone 17 was followed by deprotonation to the thermody-
namically more stable enolate, formation of vinyl triflate 18
(not isolated), and Negishi coupling17 with dimethyl zinc.
From this sequence the tetrasubstituted olefin 19 was
obtained with >90% regioselectivity. Base-catalyzed ester
hydrolysis led to rac-1. After that, the synthesis was repeated
with enantiomerically pure (S)-718 to provide the natural
enantiomer (-)-1. All analytical data of rac-1 and (-)-1 were
in full agreement with those of an authentic sample (see the
Supporting Information).
a Abbreviations: DCM ) dichloromethane, IBX ) 2-iodoxybenzoic acid.
the initial stereochemical information introduced with the
carbinol center in compound 7 is used in a relay-manner to
control the stereogenic centers in the target molecule. After
that it is discarded in favor of an endocyclic double bond.
The route is flexible and should be applicable to a variety
of derivatives, suitable for SAR investigations.
Acknowledgment. We thank Hanspeter Ka¨hlig, Susanne
Felsinger, and Lothar Brecker, all at University of Vienna,
Institute of Organic Chemistry, for NMR analysis, as well
as Sophia Khom and Steffen Hering, both at University of
Vienna, Department of Pharmacology and Toxicology, for
fruitful discussions, and the Austrian Science Fund (FWF)
for financial support (project P 21241-N19).
In conclusion we have achieved a concise stereo- and
regiocontrolled synthesis of valerenic acid (1) in racemic (8
steps from 7, 26% overall yield) and optically pure form
(13 steps from (R)-glycidol, 8% overall yield). Remarkably,
Supporting Information Available: Experimental data
and analytical characterization for all new compounds
provided. This material is available free of charge via the
Internet at http://pubs.acs.org.
(16) Crabtree, R. H.; Davis, M. W. J. Org. Chem. 1986, 51, 2655–
2661.
(17) (a) Marshall, J. A.; Zou, D. Tetrahedron Lett. 2000, 41, 1347–
1350. (b) Review: Negishi, E.; Hu, Q.; Huang, Z.; Qian, M.; Wang, G.
Aldrichim. Acta 2005, 38, 71–92.
(18) (S)-7 was prepared via a stereo-unambiguous route from (R)-
glycidol in five steps and 32% overall yield (see the Supporting Information).
OL9000137
Org. Lett., Vol. 11, No. 5, 2009
1153