Redox divergent synthesis of fawcettimine-type lycopodium alkaloids

Article abstract of DOI:10.1002/chem.201403163

A new approach for synthesis of fawcettimine-type Lycopodium alkaloids is described. A divergent strategy was achieved by applying stereoselective Diels-Alder reaction followed by redox-controlled elaboration. Eventually, (-)-8-deoxyserratinine, (+)-fawcettimine, (-)-lycopoclavamine-A, (-)-serratine, (-)-lycopoclavamine-B and (-)-serratanidine were successfully accessed. A class reunion: The divergent synthesis of fawcettimine-type Lycopodium alkaloids is reported. The present synthetic study features stereoselective Diels-Alder reactions to construct carbon cores of alkaloids. Subsequently, (-)-8-deoxyserratinine, (+)-fawcettimine, (-)-lycopoclavamine-A, (-)-serratine, (-)-lycopoclavamine-B and (-)-serratanidine were synthesized from the common synthetic intermediate via stereoselective reduction or oxygenation.

Full text of DOI:10.1002/chem.201403163

DOI: 10.1002/chem.201403163
Full Paper
&
Natural Product Synthesis
Redox Divergent Synthesis of Fawcettimine-Type Lycopodium
Alkaloids
Hisaaki Zaimoku and Tsuyoshi Taniguchi*[a]
Abstract: A new approach for synthesis of fawcettimine-
type Lycopodium alkaloids is described. A divergent strategy
was achieved by applying stereoselective Diels–Alder reac-
tion followed by redox-controlled elaboration. Eventually,
(À)-8-deoxyserratinine, (+)-fawcettimine, (À)-lycopoclav-
amine-A, (À)-serratine, (À)-lycopoclavamine-B and (À)-serra-
tanidine were successfully accessed.
Introduction
cently provided a platform for synthetic chemists. Many ele-
gant approaches to construct complex cyclic structures of
these alkaloids have been reported.[1,8] On the other hand,
most of the synthetic studies focused on (À)-8-deoxyserrati-
nine, (+)-fawcettimine, and their skeletal derivatives (defined
as “Standard”) because easily available chiral cyclohexenones
1, which is often prepared from (+)-pulegone, could be an
origin of the 15-(R)-methyl group in these alkaloids.[1e,f] Conse-
quently, they are “skeletal divergent synthesis” of these stan-
dard alkaloids. In contrast, synthetic studies of irregular types
of the alkaloids, such as (À)-lycopoclavamine-A having an “un-
usual methyl group”, “highly oxygenated” (À)-serratine and
(À)-serratanidine, are sparse.[8a,9,10] This is probably because the
strategy using convenient material 1 is not applicable to syn-
thesis of these alkaloids unlike that of (À)-8-deoxyserratinine
and (+)-fawcettimine. In such cases, stereocontrolled introduc-
tion of methyl and hydroxy groups is required. Inubushi and
co-workers reported the results of their pioneer study on the
synthesis of (Æ)-serratinine in 1974,[11] but insufficient stereo-
control was unavoidable at that time. Quite recently, we re-
ported the first syntheses of (Æ)-serratine and related alkaloids
using a stereocontrolled strategy based on Diels–Alder reaction
between a 2-alkynylcyclopetenone and (E)-1-(trimethylsilyloxy)-
3-methylbutadiene.[12]
More than 300 alkaloids have been isolated from the Lycopodi-
um genus to date, and they have unique and complex cyclic
structures.[1] In addition, recent studies have revealed that
some of them have remarkable biological activity.[1] These facts
imply that Lycopodium alkaloids are potentially a rich vein of
medicinal seeds. Lycopodium alkaloids consist of some sub-
classes. The fawcettimine class, which contains over 80 alka-
loids, recently received much attention in the field of chemis-
try. (+)-Fawcettimine is a representative alkaloid in this class
and bears a cis-fused [4.3.0]bicyclo structure and an azonane
ring with a quaternary carbon (Figure 1, Core B).[2] (À)-8-Deoxy-
serratinine also belongs in the same class because its skeleton
is built up by connecting a nitrogen atom of the azonane ring
with a C-4 carbon (Figure 1, Core A). In addition, many miscel-
laneous skeletal derivatives are known. Thus, fawcettimine-
type alkaloids can be classified by differences in functional
groups (position, number or stereochemistry). In category A,
there are four related alkaloids, (À)-8-deoxyserratinine,[3] (À)-
serratinine[4] (not shown in Figure 1, see Scheme 8), (À)-serra-
tine[4] and (À)-serratanidine,[5a,d,6] and they differ in the number
or position of hydroxyl groups. Some newcomers relevant to
(+)-fawcettimine were recently isolated. (À)-Lycopoclavamine-
A is an atypical example of fawcettimine-type Lycopodium alka-
loids because it has a methyl group of which the stereochem-
istry is opposite to that of other related alkaloids.[7] (À)-Lycopo-
clavamine-B has a tertiary hydroxyl group on the C-15 position
as does (À)-serratine.[7] In any case, an oxygen functional
group (hydroxyl or ketone group) on a C-13 position emerge
as a common component in most of fawcettimine-type alka-
loids. Indeed, fawcettimine-type Lycopodium alkaloids have re-
In this paper, we present a concept of redox divergent syn-
thesis of fawcettimine-type Lycopodium alkaloids by expanding
our previous synthetic strategy. The present work addresses
representative alkaloids consisting of major core structure A or
B to plainly demonstrate this concept (Figure 1). Our synthetic
study can be placed in “redox divergent synthesis” of fawcetti-
mine-type alkaloids complementing precedent reports of “skel-
etal divergent synthesis”. We herein describe that this strategy
is a general method to access various fawcettimine-type
alkaloids.
[a] Dr. H. Zaimoku, Dr. T. Taniguchi
School of Pharmaceutical Sciences, Institute of Medical
Pharmaceutical and Health Sciences, Kanazawa University
Kakuma-machi, Kanazawa, 920-1192 (Japan)
E-mail: tsuyoshi@p.kanazawa-u.ac.jp
Results and Discussion
An outline of our synthetic strategy is shown below. Diastereo-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201403163.
selective Diels–Alder reaction between (E)-1-(trimethylsilyloxy)-
Chem. Eur. J. 2014, 20, 9613 – 9619
9613
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
cyclopentenone derivatives are known to work as good dieno-
philes of diastereoselective Diels–Alder reactions.[13] Although
chiral 4-hydroxy-2-cyclopentenones are easily available and
could be used as a starting material of 8,[15] kinetic resolution
of a racemic 2-substituted 4-hydroxy-2-cyclopentenone deriva-
tive using a lipase was chosen as a simple and reliable method
to prepare it on a multigram scale.[16] An iodine atom was in-
troduced to a 2-position of racemic protected 4-hydroxy-2-cy-
clopentenone 4[17] by treatment with iodine (1.5 equiv) in the
presence of N,N-dimethyl-4-aminopyridine (DMAP, 20 mol%)
and triethylamine (2.0 equiv),[18] and the crude product 5 was
subjected to Sonogashira conditions (2 mol% of catalysts) with
N-propargylphthalimide (1.5 equiv) to give racemic 2-alkynyl-4-
hydroxy-2-cyclopentenone 6 in good yield (Scheme 1).[19] A 2-
alkynyl group is essential to improve the poor reactivity of 2-
substituted cyclopetenones in the Diels–Alder reaction.[20]
When compound 6 was exposed to vinyl acetate (3.0 equiv)
and Lipase PS Amano SD (30 wt%), optical resolution was
readily achieved to provide (S)-4-hydroxy-2-cyclopentenone 7
and (R)-4-acetoxy-2-cyclopentenone 8. HPLC analysis of a small
amount of sample taken from the reaction mixture indicated
high enantiomeric excess values of 7 (>95% ee) and 8 (94–
95% ee). The mixture of 7 and 8 was subjected to Mitsunobu
conditions with acetic acid and converged to 8 in good yield
and an optical purity (>95% ee). When a mixture of dienophile
8 and (E)-1-(trimethylsilyloxy)-3-methylbutadiene (90% geo-
metrical purity, 2.0 equiv) was stirred for 48–72 h at room tem-
perature, the Diels–Alder reaction readily proceeded to form
cycloadduct 9, though heating was required in the reaction of
dienophile having no acetoxy group.[12] Probably, an acetoxy
group activated the dienophile by an inductive effect. Subse-
quent treatment with diisopropylethylamine (1.5 equiv) set off
elimination of the acetoxy group to afford chiral bicyclo[4.3.0]
compound 10 in good yield. Cycloaddition should occur from
the less hindered b-face of 8,[13] and facial selectivity could be
estimated to be good because the optical purity of 10 did not
significantly decrease (91% ee). The optical purity of 10 could
be improved by recrystallization (>99% ee: minor enantiomer
not detected).
Figure 1. Target natural products and the concept of synthesis. TMS=trime-
thylsilyl.
Hydrogenation of compound 10 with 10% palladium carbon
was a key step for redox-divergent synthesis (Table 1). A work-
up procedure with acidification followed hydroxylation of com-
pound 10 to remove a trimethylsilyl group. Tracing the reac-
3-methylbutadiene and chiral cyclopentenone 2 (X=OPG;
PG=protecting group) could be designed to obtain chiral
bicyclo[4.3.0] compound 3 (Figure 1).[13] Stereoselective reduc-
tion of the olefin of compound 3 would each provide a saturat-
ed bicyclo compound having an (R) or (S)-methyl group on
a cyclohexane ring, which corresponds to partial structures of
(À)-8-deoxyserratinine or (À)-lycopoclavamine-A, respectively.
We observed palladium-catalyzed olefin isomerization of a race-
mic derivative of compound 3 (X=H) in a previous study.[12]
Therefore, hydroxyl groups of highly oxygenated alkaloids
such as (À)-serratine and (À)-serratanidine could be introduced
via isomerization of 3 and following stereoselective oxygena-
tion of an olefin. Construction of an azonane skeleton would
be achievable by a common established procedure, and final
elaboration would lead to the corresponding alkaloids.[14]
1
tion (20 wt% of 10% Pd/C) by H NMR analysis indicated that
alkyne and cyclopentenone moieties were saturated faster
than was a cyclohexene moiety. When compound 10 was ex-
posed to hydrogenation conditions for a limited time (24 h) in
the presence of a small catalytic amount of 10% Pd/C, cyclo-
hexenol product 11 was isolated as a major product along
with isomerized cyclohexenol product 12 and thoroughly satu-
rated product 13 bearing an (S)-methyl group (entry 1).[21] The
above NMR experiment also indicated that isomerization of
the cyclohexene moiety was promoted in the presence of
20 wt% of the catalyst. In fact, isomerized product 12 could be
obtained as a major product by increasing the catalytic
amount of 10% Pd/C to 20 wt% and prolonging the reaction
time (120 h) (entry 2). Further increase of the catalytic amount
First, we set out to prepare chiral cyclopentenone 8 to test
the diastereoselective Diels–Alder reaction. Chiral 4-hydroxy-2-
Chem. Eur. J. 2014, 20, 9613 – 9619
www.chemeurj.org
9614
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Synthesis of 14a and 14b. Ac=acetyl; Tf=trifluoromethanesul-
fonyl.
the presence of cesium carbonate (2.0 equiv) at low tempera-
ture (08C) and provided diol 16 in good yield. Desulfurization
of 16 with Raney nickel (W-2) followed by acetylation of two
hydroxyl groups gave diacetate 14c having a protected terti-
ary hydroxyl group. Another oxygen atom could be introduced
Scheme 1. Preparation of chiral dienophile 8 and Diels–Alder reaction.
DMAP=N,N-dimethyl-4-aminopyridine; Phth=phthaloyl; DMEAD=
di-2-methoxyethyl azodicarboxylate.
into epoxide 15 by
a combination of cesium acetate
(1.0 equiv) and acetic acid (2.0 equiv), and protection of the
two remaining hydroxyl groups by acetyl groups gave triace-
tate 14d (Scheme 3).
(50 wt%) provided saturated product 13 as a major product
(entry 3). The reaction under high pressured hydrogen gas pro-
moted production of 13 (entry 4). Although selectivity of these
reactions was not perfect, all of the products obtained could
be used for synthesis of each alkaloid.
Saturated compound 14a bearing an (R)-methyl group was
obtained by hydroxyl group-directed diastereoselective hydro-
genation of cyclohexenol compound 11 by Crabtree’s catalyst
(1 mol%)[22] and the following bismuth triflate-catalyzed
(5 mol%) acetylation of the hydroxyl group.[23] A hydroxyl
group of saturated compound 13 bearing an (S)-methyl group
was also protected by an acetyl group to give compound 14b
(Scheme 2).
Sharpless diastereoselective oxidation of compound 12 ex-
clusively gave the corresponding epoxide 15 in good yield.[24]
Addition of thiophenol (2.0 equiv) to epoxide 15 proceeded in
Scheme 3. Synthesis of 14c and 14d by stereoselective introduction of
oxygen functional groups. acac=acetylacetone; TBHP=tert-butyl hydroper-
oxide; MS=molecular sieves; DMF=N,N-dimethylformamide.
Table 1. Hydrogenation of 10.
Since we succeeded in obtaining four synthetic intermedi-
ates by elaborating desired functional groups from common
synthetic intermediate 10, we next set out to construct an azo-
nane skeleton from each compound. A common procedure
could be applied to this transformation basically (Table 2).
Phthaloyl groups of compounds 14a–d were removed by
treatment with an excessive amount of methylamine, and 2-ni-
trobenzenesulfonamides (nosylamides) 17a–d were obtained
by subsequent treatment with 2-nitrobenzenesulfonyl chloride
(2.0 equiv) after quick removal of excess methylamine by
simple evaporation.[25] Reproducibility of these reactions was
somewhat sensitive to reaction scale because the correspond-
ing cyclic imines were partially produced by evaporation after
removal of the phthaloyl group. Introduction of a three-carbon
Entry
10% Pd/C [wt%]
t [h]
Yield [%][a]
12
11
13
1
2
3
4[b]
10
20
50
20
24
59
24
13
52
34
4
120
120
36
17
40
62
[a] Isolated yield. [b] 50 atm of H2.
Chem. Eur. J. 2014, 20, 9613 – 9619
www.chemeurj.org
9615
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 2. Construction of the azonane ring. Ns=2-nitrobenzenesulfonyl;
Thx=thexyl; Ts=p-toluenesulfonyl.
Scheme 4. Total synthesis of (À)-8-deoxyserratinine. IBX=2-iodoxybenzoic
acid; DMSO=dimethylsulfoxide.
Entry
Yield [%][a]
19 (2 steps)
17
21 (2 steps)
sponding diketone (8-deoxy-13-dehydroserratinine) was hardly
detected in this reaction, the diketone would have been ob-
tained if another stronger oxidizer had been used.[29] Despite
many reports of synthesis of (À)-8-deoxyserratinine, this obser-
vation has been overlooked because synthesis of (À)-8-deoxy-
serratinine has usually been achieved by chemoselective re-
duction of (À)-8-deoxy-13-dehydroserratinine.[1e,f] (À)-8-Deoxy-
serratinine can become a precursor of other related alkaloids,
such as (+)-fawcettimine, (+)-fawcettidine and (+)-lycoflexine,
via (À)-8-deoxy-13-dehydroserratinine.[1e,f,30] This implies that
our synthetic route leads to various related alkaloids. Optical
rotation data for synthesized (À)-8-deoxyserratinine ([a]1D8 =
À10.2 (c=0.58, EtOH)) were consistent with previously report-
ed values ([a]D17 =À17.2 (c=1.01, EtOH);[3] [a]1D8 =À15.6 (c=
1.00, EtOH);[30] [a]D21 =À12.0 (c=0.1, EtOH)[1e]), and this result
proved that absolute configurations of all chiral synthesized
compounds were correct.
a
b
c
58–85
77–92
58
54–63
57–66
58
38–50
49–64
52
d
65–83
44–46
36–46
[a] Isolated yield.
unit to a ketone group of compounds 17a–d was achieved by
an indium (5 mol%)-catalyzed addition reaction of an allyl alu-
minum reagent prepared by allyl bromide (6.0 equiv) and
metal aluminum (6.0 equiv) and provided homoallyl alcohols
18a–d as single isomers (stereochemistry not determined).[26]
Terminal alkenes of compounds 18a–d were converted to pri-
mary alcohols by hydroboration with thexylborane (ca.
4 equiv), and the following oxidative treatment with sodium
perborate (10 equiv) gave 1,4-diols 19a–d. Elimination of terti-
ary hydroxyl groups of compounds 19a–d readily proceeded
by treatment with p-toluenesulfonic acid (20 mol%) or Amber-
lyst 15 to give cyclopentene derivatives 20a–d.[27] An azonane
ring of tricyclic compounds 21a–d was constructed by intra-
molecular amination between primary alcohols and nosyl
amides of compounds 20a–d under Mitsunobu conditions
Treatment of compound 21b with m-chloroperbenzoic acid
(mCPBA, 2.0 equiv) predominantly gave b-epoxide 25. Elimina-
tion of epoxide 25 with p-toluenesulfonic acid (1.0 equiv) pro-
duced allyl alcohol 26 as
a single geometric isomer
(Scheme 5). In our previous study, we found that a similar ep-
oxide to compound 25 gave the corresponding allyl alcohol
stereospecifically under the same reaction conditions.[12] By
comparing with the previous results, it was presumed that 26
had a (Z)-olefin. After exchanging the Ns group on the nitro-
gen atom of compound 26 for a tert-butoxycarbonyl (Boc)
group, subsequent removal of the acetoxy group gave diol 27.
Oxidation of two hydroxyl groups of 27 with IBX (5.0 equiv) af-
forded diketone 28, which was successively exposed to an ex-
cessive amount of trifluoroacetic acid (TFA) to remove the Boc
group. 1H NMR analysis of the crude product obtained from
the reaction mixture showed production of TFA salt 29 and
gradual transformation into another product. A significant
change in the chemical shift of alkene C-H peaks was observed
(29: d=5.80, new product: d=6.95). After complete disap-
pearance of 29, purification of the crude product by amino-
with
1.5 equiv).[25,28,29]
di-2-methoxyethyl
azodicarboxylate
(DMEAD,
End games to lead to each alkaloid from four azonane com-
pounds 21a–d are shown in Schemes 4–7. Alcohol 22 pro-
duced by hydrolysis of compound 21a was subjected to
Sharpless epoxidation conditions to give b-epoxide 23 exclu-
sively (Scheme 4). When crude epoxide 23 was treated with
thiophenol (2.0 equiv) and potassium carbonate (2.0 equiv),
tetracyclic compound 24 was produced by a sequence of
transformation including removal of a nosyl group and cycliza-
tion onto an epoxide. Interestingly, we found that chemoselec-
tive oxidation of a cyclopentanol moiety of diol 24 was suc-
cessful by using 2-iodoxybenzoic acid (IBX, 2.0 equiv) and di-
rectly afforded (À)-8-deoxyserratinine. Although the corre-
Chem. Eur. J. 2014, 20, 9613 – 9619
www.chemeurj.org
9616
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
silica gel (CHROMATOREX NH-DM1020) afforded lycopoclav-
amine-A. All physical data including the optical rotation data
([a]2D3 =À56.3 (c=0.58, CHCl3)) of synthesized lycopoclavamine-
A accorded with those of the natural product ([a]2D5 =À48.8
(c=0.11, CHCl3).[7] In this point, we suspected that a (Z)-olefin
of compound 28 was isomerized into an (E)-olefin by TFA, and
the following cyclization produced lycopoclavamine-A. If salt
29 originally had an (E)-olefin, the following cyclization would
occur faster.[31] Actually, transformation of salt 29 into another
product was very slow (48 h), and it would therefore include
not only the cyclization step but also the isomerization step.
Scheme 6. Total synthesis of (À)-serratine and (À)-lycopoclavamine-B.
Scheme 7. Total synthesis of (À)-serratanidine. Oxone=potassium peroxy-
monosulfate; EDTA=ethylenediaminetetraacetate.
We also made efforts to synthesize (À)-serratinine from our
synthetic intermediate 12 (Scheme 8). A hydroxyl group of
compound 12 was protected by
a tert-butyldimethylsilyl
group, and epoxidation of the obtained product with mCPBA
gave a-epoxide 32 exclusively. Treatment of epoxide 32 with
boron·trifluoride diethyl etherate (2.0 equiv) caused a stereo-
specific 1,2-hydride shift to produce a-methylcyclohexanone
33. Subsequent treatment of the reaction mixture with sodium
cyanoborohydride (2.0 equiv) in one pot gave desired alcohol
34 as a minor diastereomer due to a hydride approach from
a concave face. Stereochemistry of a hydroxyl group of major
isomer 35 could be inverted by Mitsunobu reaction with 4-ni-
trobenzoic acid to provide a 4-nitrobenzoate derivative of
36.[34] Although the current procedure to introduce the hydrox-
yl group is still circuitous, the possibility of synthesis of (À)-ser-
ratinine has been shown. Further study will be continued to
improve the procedure to introduce the secondary alcohol and
complete the total synthesis of (À)-serratinine.
Scheme 5. Total synthesis of (À)-lycopoclavamine-A. mCPBA=m-chloroper-
benzoic acid; Boc=tert-butoxycarbonyl; TFA=trifluoroacetic acid.
Transformation of 21c into (À)-serratine and (À)-lycopoclav-
amine-B was performed according to the procedure described
by us previously (Scheme 6).[12] Optical rotation values of syn-
thesized compounds ((À)-serratine: [a]2D9 =À7.9 (c=0.073,
EtOH); (À)-lycopoclavamine-B: [a]2D3 =À14.0 (c=0.087, CHCl3))
were consistent with those of natural products ((À)-serratine:
[a]2D2 =À15.0 (c=1.02, EtOH);[5c] [a]2D4 =À11.8 (c=0.39,
EtOH);[5d] (À)-lycopoclavamine-B: [a]2D5 =À13.9 (c=0.06,
CHCl3)[7]).
Epoxidation of highly oxygenated compound 21d with
mCPBA produced undesired a-epoxide as a main product
along with b-epoxide 30 (a/b ca. 80:20). Fortunately, we found
that Shi epoxidation of 21d provided b-epoxide 30 as a main
product.[32,33] Removal of the Ns group of 30 set off intramolec-
ular addition of the resultant secondary amine to the epoxide
to give tetracyclic compound 31 having seven stereocenters.
Finally, after oxidation of alcohol 31 with IBX (2.0 equiv), re-
moval of three acetyl groups afforded (À)-serratanidine for
which physical data including optical rotation data ([a]2D0 =
À34.7 (c=0.17, EtOH)) were identical to those of the natural
product ([a]D12:5 =À52.0 (c=1.01, EtOH);[6] [a]2D6 =À31.0 (c=
0.049, EtOH)[5d]) (Scheme 7).
Conclusion
We have achieved the total synthesis of (À)-8-deoxyserratinine,
(+)-fawcettimine (formal), (À)-lycopoclavamine-A, (À)-serratine,
(À)-lycopoclavamine-B and (À)-serratanidine. The main carbon
core was enantioselectively constructed on the basis of Diels–
Alder reaction, and plural synthetic intermediates were derived
from the product by stereoselective introduction of desired
functional groups. The present synthetic study is a robust ap-
proach to fawcettimine-type Lycopodium alkaloids by applying
the redox divergent strategy. The outstanding aspect of the
synthetic route was emphasized by the successful synthesis of
Chem. Eur. J. 2014, 20, 9613 – 9619
www.chemeurj.org
9617
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Keywords: asymmetric synthesis · cycloaddition · divergent
synthesis · natural products · total synthesis
[1] Recent reviews of the Lycopodium alkaloids: a) X. Ma, D. R. Gang, Nat.
Prod. Rep. 2004, 21, 752–772; b) J. Kobayashi, H. Morita, in The Alkaloids
(Ed.: G. A. Cordell), Vol. 61, Academic, New York, 2005, pp. 1–57; c) Y.
Hirasawa, J. Kobayashi, H. Morita, Heterocycles 2009, 77, 679–729; d) M.
Kitajima, H. Takayama, Top. Curr. Chem. 2012, 309, 1–31; e) X. Wang, H.
Li, X. Lei, Synlett 2013, 1032–1043; f) R. A. Murphy, R. Sarpong, Chem.
Eur. J. 2014, 20, 42–56.
[2] R. H. Burnell, J. Chem. Soc. 1959, 3091–3093.
[3] Y. Inubushi, H. Ishii, B. Yasui, T. Harayama, M. Hosokawa, R. Nishino, Y.
Nakahara, Yakugaku Zasshi 1967, 87, 1394–1404 (in Japanese).
[4] a) Y. Inubushi, H. Ishii, B. Yasui, M. Hashimoto, T. Harayama, Tetrahedron
Lett. 1966, 7, 1537–1549; b) Y. Inubushi, H. Ishii, B. Yasui, T. Harayama,
Tetrahedron Lett. 1966, 7, 1551–1559; c) Y. Inubushi, H. Ishii, B. Yasui, M.
Hashimoto, T. Harayama, Chem. Pharm. Bull. 1968, 16, 82–91; d) Y. Inu-
bushi, H. Ishii, B. Yasui, M. Hashimoto, T. Harayama, Chem. Pharm. Bull.
1968, 16, 92–100.
[5] a) Y. Inubushi, Y. Tsuda, H. Ishii, T. Sano, M. Hosokawa, T. Harayama, Ya-
kugaku Zasshi 1964, 84, 1108–1113 (in Japanese); b) Y. Inubushi, H. Ishii,
T. Harayama, Chem. Pharm. Bull. 1967, 15, 250–252; c) Y. Inubushi, T.
Harayama, M. Akatsu, H. Ishii, Y. Nakahara, Chem. Pharm. Bull. 1968, 16,
2463–2470. The detailed spectroscopic data of (À)-serratine and (À)-
serratanidine have been reported by Takayama and co-workers: d) K.
Katakawa, A. Nozoe, N. Kogure, M. Kitajima, M. Hosokawa, H. Takayama,
J. Nat. Prod. 2007, 70, 1024–1028.
Scheme 8. The synthetic study toward (À)-serratinine. TBS=tert-butyldime-
thylsilyl.
[6] Y. Inubushi, T. Harayama, M. Akatsu, H. Ishii, Y. Nakahara, Chem. Pharm.
Bull. 1968, 16, 561–563.
[7] K. Katakawa, H. Mito, N. Kogure, M. Kitajima, S. Wongseripipatana, M.
Arisawa, H. Takayama, Tetrahedron 2011, 67, 6561–6567.
(À)-serratanidine because it was accessible to derivatives
having dense stereocenters. In short, many previous synthetic
studies of them pursued elegance of skeletal construction,
whereas we demonstrated how structurally close natural prod-
ucts and various derivatives could be generally synthesized.
The meaning of natural product synthesis is currently di-
verse and somewhat wandering.[35] In addition, the reliability of
product yield has been vacillating in recent studies on organic
synthesis.[36] In such situations, at least, it is crucial to synthe-
size more natural products and derivatives systematically as
long as they are potential candidates of pharmaceuticals.[37]
The potential of Lycopodium alkaloids is still unknown because
their ability as pharmaceuticals has not been investigated suffi-
ciently and new derivatives are currently being isolated. We
believe that our synthetic study will serve as a motif of deriva-
tive synthesis when Lycopodium alkaloids will be taken up as
important targets in pharmaceutical science.
[8] Recent examples of total synthesis of fawcettimine-type Lycopodium al-
kaloids: a) C. Zeng, C. Zheng, J. Zhao, G. Zhao, Org. Lett. 2013, 15,
5846–5849; b) N. Itoh, T. Iwata, H. Sugihara, F. Inagaki, C. Mukai, Chem.
Eur. J. 2013, 19, 8665–8672; c) K. Xu, B. Cheng, Y. Li, T. Xu, C. Yu, J.
Zhang, Z. Ma, H. Zhai, Org. Lett. 2014, 16, 196–199.
[9] Synthetic studies of (Æ)-serratine: a) G. Luedtke, T. Livinghouse, J. Chem.
Soc. Perkin Trans. 1 1995, 2369–2371; b) J. Cassayre, F. Gagosz, S. A.
Zard, Angew. Chem. 2002, 114, 1861–1863; Angew. Chem. Int. Ed. 2002,
41, 1783–1785.
[10] Inubushi and co-workers reported transformation of (À)-serratine into
(À)-serratanidine, see ref. [6].
[11] a) T. Harayama, M. Ohtani, M. Oki, Y. Inubushi, J. Chem. Soc. Chem.
Commun. 1974, 827–828; b) T. Harayama, M. Ohtani, M. Oki, Y. Inubushi,
Chem. Pharm. Bull. 1975, 23, 1511–1515.
[12] H. Zaimoku, H. Nishide, A. Nishibata, N. Goto, T. Taniguchi, H. Ishibashi,
Org. Lett. 2013, 15, 2140–2143.
[13] Z. Pudukulathan, S. Manna, S.-W. Hwang, S. P. Khanapure, J. A. Lawson,
G. A. FitzGerald, J. Rokach, J. Am. Chem. Soc. 1998, 120, 11953–11961.
[14] In light of the reproducibility of results, we tried to avoid describing
only “champion data” in the product as much as possible in the present
paper. The range of yields was shown when the reaction was run a few
times. On the other hand, it means that the reaction could be per-
formed only once when no range of yields was shown. See the Sup-
porting Information for details.
[15] a) M. Asami, Tetrahedron Lett. 1985, 26, 5803–5806; b) M. Suzuki, T. Ka-
wagishi, A. Yanagisawa, T. Suzuki, N. Okamura, R. Noyori, Bull. Chem.
Soc. Jpn. 1988, 61, 1299–1312; c) S. P. Khanapure, N. Najafi, S. Manna,
J.-J. Yang, J. Rokach, J. Org. Chem. 1995, 60, 7548–7551; d) C.-T. Chang,
S. H. Jacobo, W. S. Powell, J. A. Lawson, G. A. FitzGerald, D. Pratico, J.
Rokach, Tetrahedron Lett. 2005, 46, 6325–6328.
Experimental Section
All experimental details including procedures, characterization and
copies of spectra are available in the Supporting Information.
Acknowledgements
[16] a) A. Rodrꢁguez, M. Nomenl, B. W. Spur, J.-J. Godfroid, Eur. J. Org. Chem.
1999, 2655–2662; b) J. P. Henschke, Y. Liu, X. Huang, Y. Chen, D. Meng,
L. Xia, X. Wei, A. Xie, D. Li, Q. Huang, T. Sun, J. Wang, X. Gu, X. Huang, L.
Wang, J. Xiao, S. Qiu, Org. Process Res. Dev. 2012, 16, 1905–1916.
[17] Although compound 4 seemed to be commercially available from sev-
eral suppliers, we prepared it from cheap 2-furfuryl alcohol in two steps
according to known procedures: a) P. R. Hamann, A. Wissner, Synth.
Commun. 1989, 19, 1509–1518; b) B. S. Morgan, D. Hoenner, P. Evans,
The authors are thankful to Professor Dr. Hiromitsu Takayama
(Chiba University) for providing us copies of NMR spectra of
the natural products and for helpful discussions. This research
was supported by a Grant-in-Aid for Scientific Research from
the Ministry of Education, Culture, Sports, Science and Technol-
ogy of Japan (MEXT).
Chem. Eur. J. 2014, 20, 9613 – 9619
www.chemeurj.org
9618
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
S. M. Roberts, Tetrahedron: Asymmetry 2004, 15, 2807–2809; c) K. Ul-
brich, P. Kreitmeier, O. Reiser, Synlett 2010, 2037–2040.
[18] M. E. Krafft, J. W. Cran, Synlett 2005, 1263–1266.
[19] a) K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16,
4467–4470; b) R. Chinchilla, C. Nꢂjera, Chem. Rev. 2007, 107, 874–922.
[20] S.-J. Min, G. O. Jones, K. N. Houk, S. J. Danishefsky, J. Am. Chem. Soc.
2007, 129, 10078–10079.
[21] Prolonging the reaction time (>72 h) gave the three products in lower
selectivity (12: 10–20%, 13: 26–37%, 14: 37–38%) when 10 wt% of
10% Pd/C was employed.
[28] T. Sugimura, K. Hagiya, Chem. Lett. 2007, 36, 566–567.
[29] A. Nakayama, N. Kogure, M. Kitajima, H. Takayama, Org. Lett. 2009, 11,
5554–5557.
[30] H. Ishii, B. Yasui, R. Nishino, T. Harayama, Y. Inubushi, Chem. Pharm. Bull.
1970, 18, 1880–1888.
[31] X.-M. Zhang, Y.-Q. Tu, F.-M. Zhang, H. Shao, X. Meng, Angew. Chem.
2011, 123, 4002–4005; Angew. Chem. Int. Ed. 2011, 50, 3916–3919.
[32] Y. Tu, Z.-X. Wang, Y. Shi, J. Am. Chem. Soc. 1996, 118, 9806–9807.
[33] Y.-R. Yang, L. Shen, J.-Z. Huang, T. Xu, K. Wei, J. Org. Chem. 2011, 76,
3684–3690.
[22] R. H. Crabtree, M. W. Davis, J. Org. Chem. 1986, 51, 2655–2661.
[23] A. Orita, C. Tanahashi, A. Kakuda, J. Otera, J. Org. Chem. 2001, 66, 8926–
8934.
[34] S. F. Martin, J. A. Dodge, Tetrahedron Lett. 1991, 32, 3017–3020.
[35] a) K. Sanderson, Nature 2007, 448, 630–631; b) J. Mulzer, Nat. Prod. Rep.
2014, 31, 595–603.
[24] K. B. Sharpless, R. C. Michaelson, J. Am. Chem. Soc. 1973, 95, 6136–
6137.
[36] M. Wernerova, T. Hudlicky, Synlett 2010, 2701–2707.
[37] Our example: Y.-Y. Li, Y.-Y. Wang, T. Taniguchi, T. Kawakami, T. Baba, H.
Ishibashi, N. Mukaida, Int. J. Cancer 2010, 127, 474–484.
[25] a) T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 1995, 36, 6373–
6374; b) T. Kan, T. Fukuyama, Chem. Commun. 2004, 353–359.
[26] K. Takai, Y. Ikawa, Org. Lett. 2002, 4, 1727–1729.
[27] a) L. M. T. Frija, C. A. M. Afonso, Tetrahedron 2012, 68, 7414–7421; b) R.
Pal, T. Sarkar, S. Khasnobis, ARKIVOC (Gainesville, FL, U.S.) 2012, 570–
609.
Received: April 19, 2014
Published online on July 7, 2014
Chem. Eur. J. 2014, 20, 9613 – 9619
www.chemeurj.org
9619
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim