Asymmetric synthesis of (+)-geranyllinaloisocyanide: Assignment of absolute stereochemistry

Article abstract of DOI:10.1021/ol200308m

The first nonracemic synthesis of (+)-geranyllinaloisocyanide, starting with (-)-lactic acid methyl ester, has been accomplished by exploiting a [3.3] sigmatropic rearrangement of allyl cyanate. The synthesis enables assignment of the S configuration of the C(3) isocyano substituted, quaternary stereogenic center in natural geranyllinaloisocyanide.

Full text of DOI:10.1021/ol200308m

ORGANIC
LETTERS
2011
Vol. 13, No. 10
2520–2523
Asymmetric Synthesis of (þ)-
Geranyllinaloisocyanide: Assignment of
Absolute Stereochemistry
Yoshiyasu Ichikawa,*,† Yasunori Matsuda,† Ken Okumura,† Mitsuhiro Nakamura,‡
Toshiya Masuda,‡ Hiyoshizo Kotsuki,† and Keiji Nakano†
Faculty of Science, Kochi University, Akebono-cho, Kochi 780-8520, Japan and Faculty
of Integrated Arts and Sciences, University of Tokushima, Tokushima 770-8502, Japan
ichikawa@kochi-u.ac.jp
Received February 1, 2011
ABSTRACT
The first nonracemic synthesis of (þ)-geranyllinaloisocyanide, starting with (ꢀ)-lactic acid methyl ester, has been accomplished by exploiting a
[3.3] sigmatropic rearrangement of allyl cyanate. The synthesis enables assignment of the S configuration of the C(3) isocyano substituted,
quaternary stereogenic center in natural geranyllinaloisocyanide.
In 1974, Scheuer and Burreson reported the isola-
tion of geranyllinaloisocyanide (1) from the marine
sponge Halichondria sp., collected by trawling at 200
m north of Oahu in Hawaii.1 Spectroscopic analysis
and degradation studies led to elucidation of the
unique structure of this marine natural product as an
isocyanide analogue of the jasmine constituent, ger-
anyllinalool (Figure 1).
absolute stereochemistry at C(3) and asymmetric synthesis
have not yet been explored.2 In this report, we present the
first asymmetric synthesis of 1 and assignment of the
absolute configuration of the quaternary stereogenic carbon
possessing the isocyano group.
Scheme 1. Retrosynthetic Analysis of (3S)-Geranyllinaloiso-
cyanide (1)
Figure 1. Unknown stereochemistry at C(3) of geranyllinaloiso-
cyanide.
Although geranyllinaloisocyanide is the first example of a
diterpene possessing an isocyano group, determination of its
Our retrosynthetic plan for the synthesis of geranyllina-
loisocyanide is shown in Scheme 1. Since the absolute
configuration at C(3) in 1 had not been defined, we
† Kochi University.
‡ University of Tokushima.
(1) (a) Burreson, B. J.; Scheuer, P. J. J. Chem. Soc., Chem. Commun.
1974, 1035. (b) Burreson, B. J.; Christophersen, C.; Scheuer, P. J.
Tetrahedron 1975, 31, 2015.
(2) For the synthesis of racemic 1, see: Ichikawa, Y.; Yamazaki, M.;
Isobe, M. J. Chem. Soc., Perkin Trans. 1 1993, 2429.
r
10.1021/ol200308m
Published on Web 04/14/2011
2011 American Chemical Society

arbitrarily selected (3S)-1 as the target. We envisioned
that (3S)-1 could arise from the R,R-dialkyl amino
aldehyde 2 via either the JuliaꢀKocienski or Wittig
olefination. Intermediate 2 in turn could be obtained by
the coupling reaction of a carbanion generated from the
sulfone 3 corresponding to the C(2ꢀ8) section of the
target, with geranyl bromide 4. Further disconnection
of 3 led to the synthon 5, whose quaternary stereocenter
could be created by using [3.3] sigmatropic rearrange-
ment attended by a transfer of chirality originating
from (ꢀ)-lactic acid methyl ester (6).3
Allyl alcohol 10 was then transformed to the allyl carbamate
11 by treatment with trichloroacetyl isocyanate followed by
hydrolysis of the resultant N-trichloroacetyl carbamate with
potassium carbonate in aq methanol. [1,3]-Chirality transfer
to install the nitrogen-substituent quaternary stereogenic
center was carried out by using [3,3] sigmatropic rearrange-
ment of the allyl cyanate.3c Specifically, dehydration of 11
employing a modified version of Appel’s conditions (PPh3,
CBr4, Et3N, CH2Cl2, ꢀ10 °C)5 generated the allyl cyanate
12 which spontaneously underwent [3.3] sigmatropic rear-
rangement to afford the corresponding allyl isocyanate 13.
After careful workup, the resultant isocyanate 13 was
immediately treated with lithium tert-butoxide in THF to
furnish 5a in 77% overall yield from 10.6
Scheme 2. Synthesis of Synthon 5 by Exploiting Sigmatropic
Rearrangement
Scheme 3. Synthesis of the C(2ꢀ8) Sulfone 3
The synthesis of synthon 5 began with protection of
the commercially available L-(ꢀ)-lactic acid methyl
ester (6) by treatment with ethyl vinyl ether in the
presence of pyridinium p-toluenesulfonate (PPTS) to
afford the ethoxyethyl ether 7 (Scheme 2). Diisobutyl-
aluminum hydride (DIBAL) reduction of 7 followed
by Wittig reaction with ethyl 2-(triphenylphospho-
ranylidene)propionate provided the R,β-unsaturated
ester 8 exclusively in 84% yield after chromatographic
purification. DIBAL reduction of the ester moiety in 8
followed by protection of the resulting allyl alcohol 9a
as a tert-butyldiphenylsilyl (TBDPS) ether gave rise to
9b. Careful removal of the ethoxyethyl ether group in
9b with PPTS in methanol furnished allyl alcohol 10 in
68% overall yield in three steps. It should be noted that
an initial route, using THP protection of 6, was found
to be complicated by the formation of the products
derived from carbocation intermediates (eq 1).4
Synthesis of the C(2ꢀ8) sulfone 3 began with protecting
group manipulation of 5a, involving desilylation with
TBAF, followed by acetonide formation using 2,2-di-
methoxypropane (DMP) to afford 14 in 75% yield
(Scheme 3). This protecting group change was necessitated
by the fact that DIBAL reduction of the ester shown in eq 2
was quite problematic. The aldehyde proton was not de-
tected by 1H NMR analysis of the products, whose struc-
tures were tentatively assigned to be the cyclized prducts.
(3) (a) Ichikawa, Y. Synlett 1991, 238. (b) Ichikawa, Y. Synlett 2007,
2927. (c) Ichikawa, Y.; Yamauchi, E.; Isobe, M. Biosci. Biotechnol.
Biochem. 2005, 69, 939. (d) Matsukawa, Y.; Isobe, M.; Kotsuki., H.;
Ichikawa, Y. J. Org. Chem. 2005, 70, 5339.
(4) The unusual nature of γ,γ-dialkyl allylic secondary alcohol
derivatives has been known to be strongly susceptible to solvolysis
reactions. See: Vernon, D. A. J. Chem. Soc. 1954, 423.
(5) Ichikawa, Y. J. Chem. Soc., Perkin Trans. 1 1992, 2135.
(6) (a) Kaiser, E. M.; Woodruff, R. A. J. Org. Chem. 1970, 35, 1198.
(b) Crowther, G. P.; Kaiser, E. M.; Woodruff, R. A.; Hauer, C. R.
Organic Synthesis, Coll. Vol. 6; John Wiley and Sons: New York; p 259.
Org. Lett., Vol. 13, No. 10, 2011
2521

Ozonolysis of 14 afforded the protected (R)-R-methy-
serinal 15,7 which was immediately subjected to Hor-
nerꢀWadsworthꢀEmmons olefination to furnish a 97:3
mixture of (E)- and (Z)-R,β-unsaturated ester 16 in 88%
yield. Catalytic hydrogenation of 16 followed by DIBAL
reduction of the ester and Wittig olefination provided (E)-
R,β-unsaturated ester 17 predominantly in 85% yield over
three steps. Transformation of 17 into allyl sulfone 3 was
then accomplished in three steps involving DIBAL reduc-
tion, mesylation, and displacement reaction with benzen-
sulfinic acid sodium salt in 77% yield.8
Scheme 5. Successful Coupling of 19 with 20 and Removal of
Sufone in 21
Scheme 4. Coupling of Sulfone 3 with Geranyl Bromide 4 To
Construct the C(8)ꢀC(9) Bond
Freshly prepared allyl bromide 19 was quickly reacted
with 4 equiv of sulfone carbanion prepared from 20 in
THF at ꢀ10 °C for 30 min, allowing the coupling product
21 to be isolated in 91% yield. Importantly, running the
reduction of allyl sulfone 21 with LiEt3BH in the presence
of PdCl2(dppp) at 0 °C afforded the desulfonylated pro-
duct 22 in 64% yield with no observable olefin scrambling
(1H NMR analysis of the product).10
Since olefination of R,R-dialkylaminoaldehydes was ex-
pected to be difficult due to the steric environment at the
neopentyl position [2 f (3S)-1, Scheme 1], we initially
investigated this crucial transformation by employing the
model aldehyde 23 (Table 1). Through the use of the one-
pot JuliaꢀKocienski olefination10 with1-phenyl-1H-tetra-
zol-5-yl (PT) sulfone and LiHMDS (entry A), the olefin 24
was obtained, albeit in a low yield (25%), along with
substantial amounts of byproduct tentatively assigned as
oxazolidinones 25. Although switching the base to
KHMDS or NaHMDS improved the yields to modest
levels (52 and 40%, respectively, entries B and C), we could
not suppress the formation of 25. Surprisingly, the Wittig
Our attempts to build the C(8)ꢀC(9) bond and to
remove the sulfone moiety are shown in Scheme 4. C(8)ꢀ
C(9) bond construction was achieved by treatment of 3
with n-butyllithium in THF at ꢀ10 °C to generate the
corresponding sulfone carbanion, which was then reacted
with freshly prepared geranyl bromide 4 (4 equiv) to
produce a diastereomeric mixture of the coupling products
18 in 71% yield. Removal of the sulfone from 18 was
examined by palladium-catalyzed reduction of allyl sul-
fone with lithium triethylborohydride (LiEt3BH). In spite
of a successful precedent in the literature,9 this process
under several conditions with varying palladium catalysts
(dppe, dppp, dppb, dphephos, dppf, and rac-BINAP
ligands) could not be satisfactorily accomplished, leading
always to significant amounts of olefin regio- and stereo-
chemical scrambling. Consequently, an alternative route
that switched the sulfone and bromide groups was
developed.
Table 1. Model Studies of the Olefination Process
In order to examine this new approach, allyl bromide 19
was prepared from 17 (Scheme 5) and sulfone 20 was
generated from geraniol.8 The conversion of 17 to 19 was
carried out via a three-step sequence, involving DIBAL
reduction, mesylation of the resulting allyl alcohol, and
treatment of the allyl mesylate with lithium bromide.
(7) For the syntheses of (S)- and (R)-N-Boc-N,O-isopropylidene-R-
methylserinals, see: (a) Alias, M.; Cativiela, C.; Diaz-de-Villegas, M.;
Galvez, J. A.; Lapena, Y. Tetrahedron 1988, 54, 14693. (b) Avenoza, A.;
Cativiela, C.; Corzana, F.; Peregrina, J. M.; Zurbano, M. J. Org. Chem.
1999, 64, 8220.
(8) Murakami, T.; Furusawa, K. Synthesis 2002, 479.
(9) Min, J.-H.; Lee, J.-S.; Yang, J.-D.; Koo, S. J. Org. Chem. 2003, 68,
7925.
(10) (a) Kotake, H.; Yamamoto, T.; Kinoshita, H. Chem. Lett. 1982,
1331. (b) Hutchins, R. O.; Learn, K. J. Org. Chem. 1982, 47, 4380. (c)
Trost, B. M.; Machacek, M. R.; Tsui, H. C. J. Am. Chem. Soc. 2005, 127,
7014.
entry
reagents
bases
temp time (h) yields of 24
A
B
C
D
E
F
CH3SO2ꢀPT LiHMDS ꢀ20 °C
CH3SO2ꢀPT NaHMDS ꢀ20 °C
3
24
19
3
25%
52%
40%
0%
CH3SO2ꢀPT KHMDS
ꢀ20 °C
Ph3PCH3Br LiHMDS rt
Ph3PCH3Br NaHMDS rt
1
34%
93%
Ph3PCH3Br KHMDS
rt
3
2522
Org. Lett., Vol. 13, No. 10, 2011

reaction of 23 was highly dependent on the countercation
of hexamethyldisilazane, which was used as a base to
generate the ylide from methyltriphenylphosphonium bro-
mide. In the case of LiHMDS (entry D), none of the
desired product 24 was produced, and the major product
was amide 26. Further screening of the bases showed that
NaHMDS afforded 24 in low yield (34%, entry E) along
with a considerable amount of 26. However, the use of
KHMDS provided 24 predominantly in good yield (93%,
entry F). Although no proof exists, we surmised that the
metal counterion in the carbamate anion (23: R = Li, Na,
and K) generated by deprotonation during olefination
and/or in the intermediate betaine determined the pathway
of the product distributions and yields.
The results of the optical rotation measurements demon-
strated that synthetic (3S)-1 is the dextrotatory enantiomer
25
with an [R]D value of þ27.8 (c 2.24 in CCl4) that is
21
the same sign as that reported for the natural product
([R]D þ15, c 2.8 in CCl4), thus confirming the absolute
configuration of natural geranyllinaloisocyanide as repre-
sented in structure (3S)-1.
Scheme 6. Completion of the Synthesis of (3S)-1
With the key intermediate 22 in hand and conditions for
terminal olefin construction established, the final stage for
the synthesis of (3S)-1 was investigated (Scheme 6). Clea-
vageof the acetonide in 22 withouthydrolysis of the N-Boc
group was carried out using scandium triflate in aq acet-
onitrile to afford 27 in 78% yield (based on recovered
starting material).7b Oxidation of the alcohol 27 with IBX
in DMSO provided aldehyde 2, which was immediately
subjected to Wittig olefination using the optimal condi-
tions established in Table 1 to furnish 28a in 88% yield.
Although Boc deprotection of 28a with TFA gave a
complex mixture of products, trimethylsilyl triflate in the
presence of 2,6-lutidine reported by Ohfune12 cleanly
afforded the amine, which was subsequently treated with
aceticformicanhydride togiverise to the formamide 28b in
74% yield in two steps. Finally, dehydration of formamide
28b with modification of Appel’s conditions5 completed
the synthesis of isocyanide (3S)-1.
In conclusion, we succeeded in the first asymmetric
synthesis of geranyllinaloisocyanide and assignment of
the S absolute stereochemistry at the C(3) quaternary
center in this natural product.
Acknowledgment. We are grateful for financial support
provided by Nagase Science and Technology Foundation.
The spectroscopic properties (IR, 1H and 13C NMR) of
the synthetic material matched those of the natural product.
Supporting Information Available. Experimental pro-
cedures and characterization data for all new compounds
and copies of NMR spectra. This material is avaibalble
free of charge via the Internet at http://pubs.acs.org.
(11) (a) Julia, M.; Paris, J.-M. Tetrahedron Lett. 1973, 14, 4833. (b)
Blakemore, P. R.; Cole, W. J.; Kochiensky, P.; Morley, A. Synlett 1998, 26.
(12) Sakaitani, M.; Ohfune, Y. J. Org. Chem. 1990, 55, 870.
Org. Lett., Vol. 13, No. 10, 2011
2523