A unified strategy for the synthesis of 7-membered-ring-containing Lycopodium alkaloids

Article abstract of DOI:10.1021/ja507740u

A unique subset of the Lycopodium alkaloid natural products share a 7-membered-ring substructure and may potentially arise from a common biosynthetic precursor. To both explore and exploit these structural relationships, we sought to develop a unified biosynthetically inspired strategy to efficiently access these complex polycyclic alkaloids through the use of a cascade sequence. In pursuit of these goals, the first total synthesis of (+)-fastigiatine (2) was accomplished via a series of cascade reactions; we describe herein a full account of our efforts. Insight from these endeavors led to critical modifications of our synthetic strategy, which enabled the first total syntheses of (-)-himeradine A (1), (-)-lycopecurine (3), and (-)-dehydrolycopecurine (4), as well as the syntheses of (+)-lyconadin A (5) and (-)-lyconadin B (6). Our approach features a diastereoselective one-pot sequence for constructing the common 7-membered-ring core system, followed by either a biomimetic transannular Mannich reaction to access himeradine A (1), lycopecurine (3), and dehydrolycopecurine (4) or an imine reduction for lyconadins A (5) and B (6). This strategy may potentially enable access to all 7-membered-ring-containing Lycopodium alkaloids and provides additional insight into their biosynthetic origin.

Full text of DOI:10.1021/ja507740u

Article
pubs.acs.org/JACS
A Unified Strategy for the Synthesis of 7‑Membered-Ring-Containing
Lycopodium Alkaloids
Amy S. Lee, Brian B. Liau, and Matthew D. Shair*
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States
S
* Supporting Information
ABSTRACT: A unique subset of the Lycopodium alkaloid natural
products share a 7-membered-ring substructure and may
potentially arise from a common biosynthetic precursor. To
both explore and exploit these structural relationships, we sought
to develop a unified biosynthetically inspired strategy to efficiently
access these complex polycyclic alkaloids through the use of a
cascade sequence. In pursuit of these goals, the first total synthesis
of (+)-fastigiatine (2) was accomplished via a series of cascade
reactions; we describe herein a full account of our efforts. Insight
from these endeavors led to critical modifications of our synthetic
strategy, which enabled the first total syntheses of (−)-himeradine A (1), (−)-lycopecurine (3), and (−)-dehydrolycopecurine
(4), as well as the syntheses of (+)-lyconadin A (5) and (−)-lyconadin B (6). Our approach features a diastereoselective one-pot
sequence for constructing the common 7-membered-ring core system, followed by either a biomimetic transannular Mannich
reaction to access himeradine A (1), lycopecurine (3), and dehydrolycopecurine (4) or an imine reduction for lyconadins A (5)
and B (6). This strategy may potentially enable access to all 7-membered-ring-containing Lycopodium alkaloids and provides
additional insight into their biosynthetic origin.
INTRODUCTION
■
Background. The Lycopodium alkaloids are a diverse family
of complex natural products that have garnered long-standing
interest in synthetic organic chemistry due to their intricate
polycyclic structures.1 Recently, certain members have been
demonstrated to exhibit neurological bioactivity, most notably
huperzine A. A subset of the Lycopodium alkaloids contains a
common 7-membered-ring system, which results in unique
polycyclic core structures that distinguish them from other
members (Figure 1). Included in this subset are himeradine A
(1)2 and fastigiatine (2),3 both of which belong to the lycodine
structural class of Lycopodium alkaloids. Both 1 and 2 contain
an unprecedented pentacyclic core with a C4−C10 linkage,4
which adds considerable strain and complexity. Additionally,
himeradine A (1) contains in total 7 rings, 10 stereocenters, 3
potentially basic nitrogens, and a quinolizidine subunit
appended to the core via a methylene linker, rendering 1
arguably the most synthetically challenging Lycopodium
alkaloid. Lycopecurine (3)5 and related alkaloids also contain
a strained C4−C10 bond and an analogous core structure to 1
and 2 but differ in that the A ring is rearranged, which is
characteristic of members of the lycopodine class. Lyconadins A
(5) and B (6) are members of the miscellaneous class6 and
possess a unique pentacyclic ring system. Despite belonging to
different structural classes, the presence of the common C4−
C10 bond and resultant 7-membered-ring system in this subset
of Lycopodium alkaloids raises the question of whether (1) they
are derived biosynthetically from a common precursor and
subsequently diverge and (2) if a synthetic strategy mirroring
Figure 1. Himeradine A (1), fastigiatine (2), lycopecurine (3),
dehydrolycopecurine (4), and lyconadins A and B (5 and 6).
this process could be devised to efficiently access all such
members from a common intermediate.
We describe herein a detailed account of our synthesis of
fastigiatine (2)7 and report the first total syntheses of the
proposed structure of himeradine A (1), lycopecurine (3), and
Received: July 29, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Proposed Biosynthetic Pathway for 7-Membered-Ring-Containing Lycopodium Alkaloids
Scheme 2. Biosynthetically Inspired Retrosynthetic Analysis of 7-Membered-Ring-Containing Lycopodium Alkaloids
dehydrolycopecurine (4) as well as the syntheses of lyconadins
A and B (5 and 6) via a unified, biosynthetically inspired
strategy for constructing 7-membered-ring-containing Lyco-
podium alkaloids.
interception of the resultant imine by a hydride equivalent. Our
biosynthetic hypothesis could provide an alternative explan-
ation for the possible common origin of other 7-membered-
ring-containing Lycopodium alkaloids and their divergence from
the pentacyclic members 1 and 2. A biosynthetically inspired
strategy based upon our proposed biosynthesis could
potentially provide ready access to all such 7-membered-ring-
containing members. Hence, we decided to embark on the total
syntheses of several 7-membered-ring-containing Lycopodium
alkaloids, including fastigiatine (2), himeradine A (1), lyco-
pecurine (3), dehydrolycopecurine (4), and lyconadins A and B
(5 and 6).
Biosynthetic Hypothesis. The proposed biosynthetic
pathway for 7-membered-ring-containing Lycopodium alkaloids
is illustrated in Scheme 1. In MacLean’s original proposed
biosynthesis of the pentacyclic core of himeradine A (1) and
fastigiatine (2),3 it was postulated that lycodane skeleton 8,
which could arise from imine 7 via an intramolecular Mannich
reaction, could undergo oxidative functionalization at C10 to
afford tetracycle 9. A subsequent intramolecular enamine SN2
cyclization through a boat-like conformation could then
construct the key C4−C10 bond, delivering pentacycle 10,
the core skeleton common to 1 and 2.
Alternatively, we propose that the 7-membered ring could be
constructed via initial cyclization of phlegmarine derivative 11
to form tetracycle 12. Imine 12 (redrawn as 13) would then be
poised to undergo an intramolecular transannular Mannich
reaction to construct the C4−C13 bond in 10, which could
then result in himeradine A (1) or fastigiatine (2).
Alternatively, reduction of the C13-imine in 12 would allow
access to the core structures common to the lucidines8 (e.g.,
15) and lyconadins (e.g., 5 and 6). In order to access these
related alkaloids in MacLean’s original hypothesis, a retro-
Mannich reaction of 10 to 13 would be necessary, followed by
RESULTS AND DISCUSSION
■
Biosynthetically Inspired Retrosynthetic Analysis. Our
biosynthetically inspired retrosynthetic analysis for himeradine
A (1), fastigiatine (2), lycopecurine (3), dehydrolycopecurine
(4), and lyconadins A and B (5 and 6) is outlined in Scheme 2.
We envisioned installing the C4−C13 bond in 1, 2, and 3 from
iminium ion intermediate 16 via a biomimetic transannular
Mannich reaction to construct the strained core system. This
disconnection would test the plausibility of our proposed
biosynthesis, since at the outset it was unknown whether the
strained intermediate 16 could readily be accessed. Importantly,
16 (redrawn as 17) could potentially serve as a common
synthetic intermediate for the synthesis of other 7-membered-
B
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 3. Convergent Nucleophilic Cyclopropane Opening To Synthesize β-Carboxyimide 20 and Azide 30
a
Conditions: (a) cat. KH, 2-(trimethylsilyl)ethanol (TMSEOH), THF; (b) Boc2O, cat. 4-DMAP, Et3N, CH2Cl2, 85% (2 steps); (c) I2, PIFA, Py,
CH2Cl2; (d) cat. TMSOTf, 1,2-bis(trimethylsiloxy)ethane (BTSE), CH2Cl2, −78 → −20 °C, 68% (2 steps); (e) t-BuLi, Et2O, −78 °C; then 28, −78
°C; then 25, −78 → 0 °C, 93%; (f) Cs2CO3, 1-chloro-3-iodopropane, DMF; (g) NaN3, NaI, DMF, 65 °C; (h) TBAF, DBU, THF, 50 °C, 88% (3
steps).
ring-containing Lycopodium alkaloids. For example, reduction of
the C13-iminium ion followed by oxidative amination could
lead to the lyconadins in a biomimetic fashion.
nontransferable ligand. Subsequent exposure of cyclopropane
25 to 29 led to regioselective nucleophilic opening to afford β-
carboxyimide 20 in 93% yield. Carboxyimide 20 underwent
efficient alkylation with 1-chloro-3-iodopropane, and the
resultant primary chloride was displaced with NaN3. The
corresponding azide was then exposed to TBAF and DBU,
resulting in cleavage of the TMSE ester with concomitant
decarboxylation and in situ base-catalyzed epimerization at C4
to yield N-Boc-2-pyrrolidinone 30;16 the use of a TMSE ester
proved necessary as other ester derivatives could not be
selectively hydrolyzed at this stage. Next, the C6-carbon was
successfully introduced via regioselective addition of MeMgBr
to the C5-carbonyl of 30, furnishing hemiaminal 31 (Scheme
4). Dehydration of 31 with CSA occurred with simultaneous
ketal deprotection to provide the corresponding dihydro-
Enamine 17 could arise from β-ketoester 19 via a
diastereoselective 7-endo-trig intramolecular 1,4-conjugate
addition,9 forming the C6−C7 bond as well as the C7- and
C12-stereocenters. Concomitantly, the Nα−C5 and Nβ−C13
bonds could be introduced via condensation of Nα and Nβ
with the C5- and C13-carbonyls, respectively. We anticipated
that the stereoselectivity of these transformations would be
controlled by topological constraints or by stereoelectronically
favored axial attack anti to the C16-methyl group.9,10 At the
outset, the order of these potentially tandem bond-forming
events was considered flexible depending on judicious selection
of nitrogen protecting groups. Our original objective was to
achieve a one-pot cascade reaction to construct the pentacyclic
core in 1 and 2, and more importantly, we hoped to later
program this sequence to access all 7-membered-ring-
containing Lycopodium alkaloids. A convergent synthesis of
cascade precursor 19 can be accomplished via initial alkylation
of β-carboxyimide 20 with iodide 22 and subsequent addition
of the lithium enolate of tert-butylacetate 21 to the C5-
carbonyl. By varying the electrophilic alkylation partner of 20, a
variety of structurally diverse target members could potentially
be accessed. To explore our strategy, we initially targeted
fastigiatine (2), which culminated in its first total synthesis.7 A
detailed account of our efforts toward the synthesis of 2 will
first be presented.
Scheme 4. Synthesis of Dihydropyrrole 32 as a Potential
a
Cascade Precursor
Total Synthesis of (+)-Fastigiatine (2). Our synthesis of
fastigiatine (2) commenced with the syntheses of coupling
fragments cyclopropane 25 and vinyl iodide 27 (Scheme 3).
Transesterification of ethyl ester 23, derived in four steps from
(S)-epichlorohydrin,11 with 2-(trimethylsilyl)ethanol
(TMSEOH),12 followed by N-Boc protection, afforded cyclo-
propane 25. Cyclohexenone 26, derived in four steps from (R)-
pulegone, was first α-iodinated according to literature
protocol13 and then protected as a dioxolane to afford vinyl
iodide 27. Electrophilic cyclopropanes such as 25 have been
demonstrated to undergo nucleophilic opening in the presence
of suitable nucleophiles.14 Exploiting this reactivity, we sought
to couple a 2-cyclohexenone anion equivalent15 derived from
27 to cyclopropane 25. To this end, we first generated mixed
organocuprate 29 from vinyl iodide 27 via lithium−halogen
exchange with tert-butyllithium followed by transmetalation
with copper acetylide 28, where the acetylide functions as a
a
Conditions: (a) MeMgBr, TMEDA, THF, −78 → −20 °C; (b) CSA,
PhH; (c) recycle (CSA, PhH); (d) PPh3, THF/H2O, 60 °C, 67% (4
steps); (e) PPTS, EtOH, 80 °C, 80%.
C
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
pyrrole. Staudinger reduction of the azide functionality then
delivered primary amine 32.
Scheme 6. Installation of the Correct Nα−C5 Connectivity
a
via Inductive Deactivation of Nβ
Although the C−N connectivity in 32 did not correspond to
that found in fastigiatine (2), we hypothesized that the more
nucleophilic Nα-amine could exchange with the Nβ-carbamate
to provide imine 36. Imine 36 could then undergo the desired
7-endo-trig cyclization via tautomerization to the corresponding
exocyclic enamine, followed by a subsequent transannular aldol
reaction to establish the C4−C13 bond of tetracycle 33 or 34.
We originally assumed that 32 could not undergo 5-endo-trig
cyclization to form the incorrect C4−C7 bond since this would
violate Baldwin’s rules.17 However, heating 32 with PPTS in
EtOH resulted in exclusive formation of constitutional isomer
35.
Our proposed mechanism for the formation of 35 is depicted
in Scheme 5. Upon exposure to PPTS, formation of the
Scheme 5. 5-Endo-Trig Cyclization To Form Tetracycle 35
a
Conditions: (a) cat. Mg(ClO4)2, MeCN, 60 °C; (b) LiHMDS, THF;
NsCl, 0 °C → RT, 89% (2 steps); (c) LDA, t-BuOAc, THF, −78 °C;
then 41, −78 °C; (d) PPh3, PhH, 50 °C, 88% (2 steps); (e) PhSH,
Cs2CO3, DMF, 75%.
otherwise preferred endocyclic Nα−C5−C4 enamine,20 there-
by suppressing potential undesired 5-endo-trig cyclization. Only
a few steps remained in order to test the key cascade reaction:
(1) cleavage of the Nβ-Ns group, (2) deprotection of the C13-
ketal, and (3) condensation of Nβ with the resultant C13-
ketone to form the corresponding α,β-unsaturated imine, which
we anticipated would initiate the cascade sequence to form the
pentacyclic core of 2. However, upon cleavage of the Nβ-Ns
group, Nα and Nβ underwent rapid transamination to afford 5-
membered vinylogous urethane 44. On the basis of these
results, it was clear that the Nβ-Ns group would have to be
removed at a later stage.
Instead of cleaving the Nβ-Ns group, 43 was directly exposed
to aqueous HCl, which resulted in a formal [3+3]-cyclo-
addition20,21 to furnish tetracycle 51 as a single diastereomer in
92% yield (Scheme 7). Formation of 51 presumably occurred
via initial C13-dioxolane cleavage to provide enone 45, which
then underwent 7-endo-trig intramolecular 1,4-conjugate
addition to form the C6−C7 bond via stereoelectronically
favored axial attack anti to the C16-methyl group. Axial attack
of the enone syn to the C16-methyl group is disfavored due to a
developing 1,3-diaxial interaction in the resultant enol
intermediate between the C16-methyl group and the newly
formed C6−C7 bond. Next, stereoselective protonation of the
ensuing C12−C13 enol 49 occurred via stereoelectronically
favored axial protonation22 to provide ketone 50. Finally, a
transannular aldol reaction resulted in formation of the C4−
C13 bond to deliver tetracycle 51. Despite the efficiency of this
transformation, which selectively generated four contiguous
stereocenters, it was uncertain whether tetracycle 51 could
ultimately be converted to fastigiatine (2); this would require a
formal exchange between the C13-hydroxyl and Nβ in order to
access the pentacyclic core.
corresponding oxocarbenium ion from 32 would result in 5-
endo-trig cyclization anti to the C16-methyl group to establish
the C4−C7 bond in cis-5,5-bicyclic iminium ion intermediate
38. We speculated that formation of a charged oxocarbenium
intermediate allowed an exception to Baldwin’s rules, which
generally disfavors 5-endo-trig cyclizations. Next, Nα and Nβ
could undergo exchange to provide imine 39. Tautomerization
of 39 to exocyclic enamine 40 followed by a transannular aldol
reaction would then yield tetracycle 35. From these results it
was apparent that (1) the correct C−N connectivity was most
likely required for the cascade sequence and (2) 5-endo-trig
cyclization was facile.
Unfortunately, attempts to form the 6-membered imine with
the correct Nα−C5 connectivity (e.g., 36) were largely
unsuccessful, primarily due to an apparent strong thermo-
dynamic preference for the 5-membered-ring system containing
the incorrect Nβ−C5 bond. In order to reduce the propensity
of forming the undesired cyclic 5-membered-ring system, the
Nβ-Boc group was replaced with a 2-nitrobenzenesulfonyl (Ns)
group, which we speculated would inductively deactivate Nβ
and thus prevent Nα and Nβ exchange. This was accomplished
via chemoselective cleavage of the N-Boc group of 30 with
18
Mg(ClO4)2 followed by N-sulfonylation to yield N-Ns-2-
pyrrolidinone 41 (Scheme 6). Gratifyingly, addition of the
lithium enolate of tert-butyl acetate to the C5-carbonyl
provided β-ketoester 42 with Nβ completely disengaged from
C5. Staudinger reaction of 42 afforded vinylogous urethane 43
as an inconsequential ∼3:2 mixture of C4-epimers.19 Of note,
the C6-tert-butyloxycarbonyl was utilized to induce preferential
formation of the exocyclic Nα−C5−C6 enamine over the
Nonetheless, only a few steps remained in order to achieve a
synthesis of fastigiatine (2). To this end, alkylation of 51 with
MeI in the presence of K2CO3, followed by addition of PhSH,
delivered N-methylamine 52 (87%). Up until this point, there
was no evidence that 52 or other tetracyclic intermediates
existed in equilibrium with the corresponding tricyclic retro-
aldol isomers. However, submission of 52 to electrospray
D
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 7. Completion of the Total Synthesis of (+)-Fastigiatine via a Formal [3+3]-Cycloaddition
a
Conditions: (a) HCl, THF/H2O, 92%; (b) K2CO3, MeI, DMF, 0 °C → RT; then PhSH, 0 °C → RT, 87%; (c) CF3CH2OH, 80 °C, 85%; (d) p-
TsOH·H2O, PhH, 80 °C, 95%; (e) Ac2O, Et3N, CH2Cl2, 85%.
ionization mass spectrometry revealed an ion with mass [M −
H2O + H]+, potentially corresponding to the desired
dehydrated product and suggesting that the desired conversion
of tetracycle 52 to the pentacyclic core was possible.
Gratifyingly, heating 52 in deoxygenated 2,2,2-trifluoroethanol
(TFE) yielded pentacycle 55 in 85% yield, presumably via (1)
initial retro-aldol reaction to afford ketone 53, (2) condensation
of Nβ with the resultant C13-ketone to form iminium ion 54,
and (3) the key biomimetic transannular Mannich reaction to
construct the C4−C13 bond. The use of deoxygenated TFE, a
strong hydrogen-bond donor but not a strong acid such that it
would protonate Nβ or induce cleavage of the tert-
butyloxycarbonyl group, as solvent was crucial to the success
of this reaction. Other common solvents (e.g., EtOH, DMSO,
DMF, MeCN, PhMe, etc.) and additives (e.g., acids) failed to
yield 55. We speculated that under acidic conditions,
protonation of Nβ would disfavor formation of the positively
charged intermediates involved in the retro-aldol reaction.
Finally, heating 55 with p-TsOH resulted in removal of the tert-
butyloxycarbonyl group to furnish the corresponding imine,
which was subsequently acetylated to afford (+)-fastigiatine (2)
in 81% yield over 2 steps ([α]D24 = +375 (c 1.4, CHCl3)). This
constitutes the first total synthesis of fastigiatine (2), the
structure of which was unambiguously confirmed by single-
crystal X-ray diffraction analysis.
In summary, an efficient synthesis of fastigiatine (2) was
achieved. Our synthesis featured a convergent fragment
coupling via a nucleophilic cyclopropane opening and a
diastereoselective formal [3+3]-cycloaddition reaction. Addi-
tionally, an unusual retro-aldol, iminium ion formation, and
biomimetic transannular Mannich cascade allowed for the
construction of the strained pentacyclic core. The ease of the
transannular Mannich reaction suggests that imine 13 is a
feasible biosynthetic intermediate. However, despite the
efficiency of the formal [3+3]-cycloaddition and subsequent
retro-aldol−Mannich sequence, this strategy prohibited direct
isolation of an imine analog derived from an intermediate such
as 13 or 17 and thus prevented our ultimate objective of
developing a unified strategy to access other 7-membered-ring-
containing Lycopodium alkaloids. Indeed, reprogramming the
cascade sequence and thereby obviating the need to formally
exchange the C13-hydroxyl with Nβ would be necessary toward
achieving this goal. This objective was ultimately accomplished
in our total synthesis of the proposed structure of himeradine A
(1), which will be discussed next.
Total Synthesis of the Proposed Structure of
(−)-Himeradine A (1). In addition to sharing a pentacyclic
core with fastigiatine (2), himeradine A (1) contains a
quinolizidine subunit appended to the core via a methylene
linker. A synthesis of 1 has not been reported in the literature
to date. As discussed previously, we wanted to accomplish two
objectives in a synthesis of 1: (1) the development of a unifying
strategy that could permit access to all 7-membered-ring-
containing Lycopodium alkaloids and (2) a one-pot sequence
for the construction of the strained core system common to 1
and 2.
Our retrosynthetic analysis of the quinolizidine subunit of
himeradine A (1) is illustrated in Scheme 8. The desired β-
carboxyimide (20) alkylation partner 56 could be constructed
from allyl N-Boc piperidine 57 via a strategy involving a ring-
closing metathesis. N-Boc piperidine 57 contains four stereo-
centers; we planned to introduce the C6′-, C10′-, and C17-
stereocenters through substrate control. The C6′-stereocenter
in 57 could be installed via diastereotopic deprotonation at C6′
in N-Boc piperidine 58 followed by alkylation. The C10′- and
C17-stereocenters could be introduced by relaying the
stereochemical information of the single C8′-stereocenter in
N,O-methoxyacetal 61 via a diastereoselective N-acyliminium
ion Mannich reaction followed by a stereoselective ketone
reduction.
Our synthesis of himeradine A (1) commenced with the
lipase-mediated desymmetrization of 3-methylglutaric anhy-
dride with n-PrOH,23 providing carboxylic acid 63 in 98% yield
(93% ee) (Scheme 9). Chemoselective reduction of the
E
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
lactam, and a two-step reduction of the C6′-carbonyl to a
methylene (62% over 5 steps).
Scheme 8. Retrosynthesis of the Quinolizidine Subunit 56
Next, diastereotopic deprotonation of 70 at C6′ was
accomplished with sec-butyllithium in the presence of
TMEDA (Scheme 10).26 The conformation of lithiated N-
a
Scheme 10. Synthesis of N-Boc Piperidine Iodide 75
a
Scheme 9. Synthesis of N-Boc Piperidine 70
a
Conditions: (a) sec-BuLi, TMEDA, −78 °C; CuCN·2LiCl, −78 °C;
AllylBr, −78 °C → RT, 50% (97% brsm); (b) TBAF, THF, 0 °C →
RT, quant.; (c) TBSCl, 4-DMAP, Et3N, CH2Cl2, 97%; (d) DPPA,
PPh3, DEAD, THF, 82%; (e) TBAF, THF, 0 °C → RT, 96%; (f) I2,
PPh3, Imid, CH2Cl2, 96%.
Boc piperidine 71 is rigidly locked due to 1,3-allylic strain
minimization between the N-Boc group and the C10′-
stereocenter, which is further reinforced by the equatorial
C12′-methyl group. Equatorial lithiation is favored due to
chelation with the N-Boc group.26 Transmetalation with Cu(I),
followed by alkylation with allyl bromide, yielded allyl N-Boc
piperidine 72 as a single diastereomer in 50% yield (97% based
on recovered starting material). Attempted alkylation with
more functionalized electrophiles (e.g., electrophilic cyclo-
propanes, substituted allylic halides) was unsuccessful. Global
silyl deprotection, followed by selective protection of the
resultant primary hydroxyl, afforded TBS ether 73 (97%, 2
steps). Finally, Mitsunobu displacement of the secondary
hydroxyl group in 73, cleavage of the primary TBS ether, and
formation of the corresponding iodide yielded the desired
alkylation partner, iodide 75. Of note, attempted synthesis of
quinolizidine 56 or the mesylate or tosylate analogs as potential
alkylation partners was unsuccessful, presumably due to inter-
and intramolecular alkylation by the tertiary amine. We also
attempted to protect the quinolizidine tertiary amine as an
amide, but downstream efforts to reduce the tertiary amide
proved incompatible with the diverse array of functional groups
present.
a
Conditions: (a) Amano Lipase PS, n-PrOH, i-Pr2O, 98% (93% ee);
(b) BH3·SMe2, THF, 0 °C → RT; (c) SO3·Py, i-Pr2NEt, DMSO,
CH2Cl2, 0 °C, quant. (3 steps); (d) NH3/MeOH, 50 °C; (e) p-TsOH,
MeOH, 50 °C; (f) 65, BF3·OEt2, MeCN, −40 °C → RT, 43% (3
steps); (g) DIBAL-H, THF, −78 °C; (h) TBSCl, Imid, DMF, 98% (2
steps); (i) Boc2O, 4-DMAP, MeCN, 84%; (j) DIBAL-H, −78 → 0 °C;
(k) BF3·OEt2, Et3SiH, CH2Cl2, −78 °C, 75% (2 steps).
carboxylic acid in 63 with BH3·SMe2, followed by Parikh−
Doering oxidation of the resultant primary carbinol, furnished
aldehyde 64. Exposure of 64 to NH3/MeOH resulted in the
formation of a mixture of N,O-hydroxy and N,O-methoxy
acetals, which was subsequently heated with p-TsOH in MeOH
at 50 °C to afford N,O-methoxy acetal 61 as a mixture of
diastereomers. Next, exposure of 61 and silyl enol ether 65 to
BF3·OEt2 afforded lactam 66 as a single diastereomer via an N-
acyliminium ion Mannich reaction in 43% yield (3 steps).24
The diastereoselectivity of this reaction was controlled via
stereoelectronically favored axial attack by 65 at C10′ of the N-
acyliminium ion intermediate anti to the C12′-methyl group.
Next, chelate-controlled 1,3-syn reduction25 of ketone 66 with
DIBAL-H afforded alcohol 67 as a single diastereomer, which
was converted to N-Boc-piperidine 70 in four steps, involving
TBS protection of the C17-hydroxyl, Boc protection of the
β-Carboxyimide 20 underwent efficient alkylation with
iodide 75 to afford coupled product 76 in 94% yield (Scheme
11). Carboxyimide 76 was converted to β-ketoester 79 in four
steps via (1) cleavage of the TMSE ester with concomitant
decarboxylation, (2) chemoselective removal of the N-Boc
group with Mg(ClO4)2, (3) protection of the resultant
pyrrolidinone with a Ns group,27 and (4) addition of the
lithium enolate of tert-butylacetate to N-Ns-2-pyrrolidinone 78.
With 79 in hand, the key cascade sequence was next explored.
F
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
direct isolation of imine 83, which would enable access to other
7-membered-ring-containing Lycopodium alkaloids (vide infra).
Next, Staudinger reduction of 83 resulted in formation of the
corresponding vinylogous urethane, which underwent the key
biomimetic transannular Mannich reaction to form the C4−
C13 bond in situ, furnishing the corresponding hexacycle in
75% yield (Scheme 12). With the successful construction of the
Scheme 11. One-Pot Sequence for the Synthesis of the
Tricyclic Core of (−)-Himeradine A (1)
a
Scheme 12. Total Synthesis of the Proposed Structure of
a
(−)-Himeradine A (1)
a
Conditions: (a) PPh3, 8:1 THF/H2O, 70 °C, 75%; (b) Ac2O, Et3N,
CH2Cl2, 83%; (c) acrolein, Hoveyda−Grubbs catalyst II, CH2Cl2,
quant.; (d) H2, Pd/C, EtOAc; (e) NaBH4, EtOH, 0 °C; (f) p-TsCl,
Et3N, 4-DMAP, CH2Cl2, 70% (3 steps); (g) TBSOTf, 2,6-lut, CH2Cl2,
0 °C → RT; (h) TBAF, THF, 0 °C, 66% (2 steps); (i) 1:1 TFA/
CH2Cl2, quant.
a
Conditions: (a) 75, Cs2CO3, DMF, 94%; (b) TBAF, DBU, THF, 50
°C, 97%; (c) cat. Mg(ClO4)2, MeCN, 60 °C; (d) LiHMDS, THF;
NsCl, 0 °C → RT, 90% (2 steps); (e) LDA, t-BuOAc, THF, −78 °C;
then 78, −78 °C; (f) p-TsOH, 8:1 THF/H2O; (g) 2-tert-butyl-1,1,3,3-
tetramethylguanidine (Barton’s base), MeCN, 0 °C → RT; then
PhSH, K2CO3, 0 °C → RT, 84% (3 steps).
pentacyclic core realized, the resultant secondary amine was
then acetylated to provide amide 84 in 83% yield. Next, the
allyl N-Boc piperidine moiety in 84 was converted to the
desired quinolizidine subunit in a multistep sequence. First,
cross metathesis of the allyl group of 84 with acrolein was
conducted with Hoveyda−Grubbs catalyst II,28 which yielded
enal 85 (quant.). Hydrogenation of 85, followed by NaBH4
reduction of the aldehyde intermediate, and sulfonylation of the
resultant hydroxyl group, afforded tosylate 86 in 70% yield (3
steps). Two-step cleavage of the N-Boc group of 86, involving
initial treatment with TBSOTf followed by TBAF, resulted in in
situ cyclization via displacement of the tosylate group by the
resultant secondary amine to deliver heptacycle 87 (66%, 2
steps), which now contained the quinolizidine subunit.
As discussed previously, the formal [3+3]-cycloaddition
employed in the synthesis of fastigiatine (2) could not allow
access to an imine intermediate such as 17. This strategy was
thus not readily amenable toward the synthesis of other 7-
membered-ring-containing Lycopodium alkaloids, which was
one of our ultimate objectives.
Instead, a new reaction sequence was developed that enabled
the construction of the strained tricyclic core structure in imine
83 in a single pot. Instead of performing a Staudinger reaction
on 79 and forming the corresponding vinylogous urethane, the
C13-dioxolane was first removed upon exposure to p-TsOH to
provide enone 80. Next, 80 was converted to imine 83 in a one-
pot sequence involving (1) initial exposure of 80 to Barton’s
base to induce 7-endo-trig intramolecular cyclization to form the
C6−C7 bond via axial attack anti to the C16-methyl group, (2)
tautomerization of the resultant C12−C13 enol to secure the
C12-stereocenter through protonation from the convex face,
(3) cleavage of the N-Ns protecting group upon addition of
PhSH and K2CO3, and (4) in situ condensation of Nβ with the
C13-ketone to form imine 83. By forming the C6−C7 bond
first via the β-ketoester prior to cleavage of the N-Ns group, we
were able to circumvent both undesired formation of the Nβ−
C5 bond as well as the transannular aldol reaction. This allowed
Finally, cleavage of the tert-butyloxycarbonyl group with
concomitant decarboxylation occurred upon exposure to TFA,
yielding the proposed structure of (−)-himeradine A (1) as the
23
bis-TFA salt (quant., [α]D = −19 (c 0.3, MeOH)). With the
exception of the chemical shift of H10′, which is shifted upfield
by Δδ = 0.14 ppm in synthetic relative to natural himeradine A
1
(1), the H NMR, 13C NMR, COSY, NOESY, HSQC, and
HMBC spectra, as well as the specific rotation, are in excellent
agreement with the data reported for the natural product. It is
well documented that the protonation state of basic nitrogen
atoms can have a significant impact on proton chemical shifts.29
Hence, we carefully titrated the bis-TFA salt of himeradine A
G
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 13. Total Syntheses of Lycopecurine (3) and Dehydrolycopecurine (4)
a
Conditions: (a) benzyl 3-bromopropyl ether, Cs2CO3, DMF, quant.; (b) TBAF, DBU, 50 °C, 87%; (c) cat. Mg(ClO4)2, MeCN, 60 °C; (d)
LiHMDS, THF; NsCl, 0 °C → RT, 95% (2 steps); (e) LDA, t-BuOAc, THF, −78 °C; then 90, −78 °C; (f) HCl, 8:1 THF/H2O; (g) K2CO3, DMF;
then PhSH, 0 °C → RT, 84% (3 steps); (h) 1:1 TFA/CH2Cl2; (i) HCl, MeOH, 65 °C, 95% (2 steps); (j) H2, Pd/C, HCl, EtOH, 80%; (k) 3:1
AcOH/HBr; (l) K2CO3, MeCN, 50 °C, 59% (2 steps); (m) LiEt3BH, THF, 0 °C → RT, 66%.
(1) with NaOCD3 until the free base was obtained, and then
we titrated aliquots of TFA until the bis-TFA salt was
reisolated. We were unable to identically replicate the published
1H NMR spectrum of the bis-TFA salt of natural himeradine A,
and subsequent addition of PhSH, 91 underwent a 7-endo-trig
intramolecular cyclization to form the C6−C7 bond with
stereoselective C12-protonation, followed by in situ condensa-
tion of Nβ with the C13-ketone, to yield imine 92 (69%, 6
steps).
1
although the H NMR spectrum of the free base of synthetic
(1) and natural himeradine A appeared to match.30 Varying the
concentration and temperature did not affect the proton
chemical shift. Without an authentic sample of 1, we were
unable to unambiguously verify that the proposed structure of
himeradine A (1) corresponded to the correct structure of
natural himeradine A. This constitutes the first total synthesis of
the proposed structure of himeradine A (1).
Treatment of 92 with TFA resulted in cleavage of the tert-
butyloxycarbonyl group with concomitant decarboxylation.
Upon heating with HCl in MeOH,31 ketone 93 underwent a
key biomimetic transannular Mannich reaction to establish the
C4−C13 bond in tetracycle 94 (95%, 2 steps). Cleavage of the
benzyl ether group in 94 via hydrogenolysis afforded alcohol
95. Alcohol 95 was then treated with 25% HBr in glacial AcOH,
resulting in bromination of the hydroxyl group and formation
of the corresponding ammonium bromide salt. This inter-
mediate underwent cyclization upon heating with K2CO3 via
displacement of the bromide by the secondary amine to yield
In summary, a biosynthetically inspired and convergent
strategy for the synthesis of himeradine A (1) was achieved.
Noteworthy transformations include a diastereoselective one-
pot sequence for constructing the core system shared by 1 and
other Lycopodium alkaloids, and a key biomimetic transannular
Mannich reaction for accessing the pentacyclic core. As
discussed previously, isolation of imine 83 (Scheme 11)
could potentially allow access to other 7-membered-ring-
containing Lycopodium alkaloids through a unifying strategy,
which will be the topic of the upcoming sections.
Total Syntheses of (−)-Lycopecurine (3) and
(−)-Dehydrolycopecurine (4). We decided to first embark
on the syntheses of lycopecurine (3) and dehydrolycopecurine
(4) using our biosynthetically inspired, unifying strategy
(Scheme 13). Lycopecurine (3) was isolated in 1969 from
Lycopodium alopecuroides,5a,b and dehydrolycopecurine (4) was
obtained from Lycopodium inundatum5c in 1971. To date, a
synthesis of either 3 or 4 has not been reported in the literature.
To this end, β-carboxyimide 20 underwent efficient
alkylation with benzyl 3-bromopropyl ether to provide alkylated
product 88 (quant.). Next, 88 was converted in five steps to
enone 91 by utilizing an analogous reaction sequence employed
in our synthesis of himeradine A (1). Using a modified protocol
for our one-pot sequence involving initial exposure to K2CO3
22
dehydrolycopecurine (4) (59%, 2 steps, [α]D = −69 (c 0.34,
MeOH)). Diastereoselective reduction of the ketone with
22
LiEt3BH then delivered lycopecurine (3) (66%, [α]D = −19
(c 0.14, MeOH)). The structure of lycopecurine (3) was
unambiguously confirmed via single-crystal X-ray diffraction
analysis of the HBr salt of 3. Our report constitutes the first
total syntheses of lycopecurine (3) and dehydrolycopecurine
(4).
Total Syntheses of (+)-Lyconadin A (5) and (−)-Lyco-
nadin B (6). Lyconadin A (5) was isolated in 2001 from the
club moss Lycopodium complanatum by Kobayashi and co-
workers6a and exhibited modest cytotoxicity against murine
lymphoma L1210 cells (IC50 = 0.46 μg/mL) and human
epidermoid carcinoma KB cells (IC50 = 1.7 μg/mL). Lyconadin
A (5) contains a pentacyclic core, including a 2-pyridone ring, 6
stereocenters, and a tertiary amine. Lyconadin B (6), isolated in
2006, contains a dihydropyridone instead of a pyridone subunit
but is otherwise identical to 5.6b Both 5 and 6 have been shown
to enhance expression of nerve growth factor (NGF) in
1321N1 human astrocytoma cells. Due to their complex
H
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Scheme 14. Total Syntheses of (+)-Lyconadin A (5) and (−)-Lyconadin B (6)
a
Conditions: (a) acrylonitrile, n-Bu4NOH, MeCN, 0 °C → RT; (b) TBAF, DBU, 50 °C, 78% (2 steps); (c) cat. Mg(ClO4)2, MeCN, 60 °C; (d)
LiHMDS, THF; NsCl, 0 °C → RT, 64% (2 steps); (e) LDA, t-BuOAc, THF, −78 °C; then 98, −78 °C; (f) HCl, 8:1 THF/H2O; (g) K2CO3, DMF;
then PhSH, 0 °C → RT, 63% (3 steps); (h) NaBH(OAc)3, AcOH, CH2Cl2, 0 °C → RT; (i) 1:1 TFA/CH2Cl2, 59% (2 steps); (j) LiHMDS; I2,
THF, −78 → 0 °C, 73%; (k) NH3/MeOH, 120 °C, 57%; (l) neat, 160 °C, 57%.
polycyclic architecture and promising biological activity, the
lyconadins have been the subject of synthetic efforts by various
laboratories.9,32 We envisioned that our unifying strategy could
be applied toward the syntheses of lyconadins A (5) and B (6)
via reduction of an iminium ion intermediate such as 17 instead
of application of a transannular Mannich reaction.
Our synthesis commenced with the alkylation of β-
carboxyimide 20 with acrylonitrile to afford nitrile 96 (Scheme
14). Carboxyimide 96 was converted to enone 99 in five steps
by employing an analogous reaction sequence utilized in our
syntheses of himeradine A (1) and lycopecurine (3). Successful
application of our one-pot protocol involving initial addition of
K2CO3 followed by PhSH then yielded tricyclic imine 100 in
63% yield (3 steps). Chemoselective reduction of the imine
functionality in 100 with NaBH(OAc)3 and AcOH occurred
smoothly to furnish amine 101.
Next, treatment of 101 with TFA resulted in loss of the tert-
butyloxycarbonyl group with concomitant decarboxylation to
provide ketone 102 in 59% yield (2 steps). Formation of the
corresponding enolate of 102 via deprotonation by LiHMDS,
followed by addition of I2, directly afforded tertiary amine 103
in 73% yield. In Sarpong’s synthesis of lyconadin A (5), an
analogous transformation was employed for the construction of
the C−N bond involving deprotonation with n-butyllithium,
followed by addition of I2.32a,b Formation of 103 could have
occurred via two possible mechanisms: (1) initial α-iodination
of the ketone from the convex face and subsequent intra-
molecular SN2 displacement of the iodide by the secondary
amine or (2) oxidative enolate amination to form the C−N
bond. Finally, heating nitrile 103 in an NH3/MeOH solution33
in a sealed vessel at 120 °C resulted in dihydropyridone
formation to deliver (−)-lyconadin B (6) (57%, [α]D21 = −102
(c 0.5, MeOH)). During conversion of 103 to lyconadin B (6),
we observed the formation of trace quantities of lyconadin A
(5) in the crude 1H NMR.34 The formation of lyconadin A (5)
presumably occurred via autoxidation in the presence of trace
oxygen. Heating lyconadin B (6) neat at 160 °C under an
atmosphere of air delivered lyconadin A (5) in 57% yield
([α]D22 = +37 (c 0.12, MeOH)). In conclusion, the biomimetic
syntheses of lyconadins A (5) and B (6) have been
accomplished via our unified strategy.
CONCLUSION
■
The successful synthesis of several diverse 7-membered-ring-
containing Lycopodium alkaloids was accomplished through the
development of a unifying, biosynthetically inspired strategy.
The first total syntheses of the proposed structure of
himeradine A (1), fastigiatine (2), lycopecurine (3), and
dehydrolycopecurine (4), as well as the syntheses of lyconadins
A (5) and B (6) were achieved. Our efforts commenced with
the total synthesis of fastigiatine (2) in 2010,7 a detailed
account of which was discussed. A convergent fragment
coupling via a nucleophilic cyclopropane opening was
developed to form the C11−C12 bond and yield the versatile
synthetic intermediate β-carboxyimide 20 (vide infra). Prob-
lems associated with attaining the correct C−N connectivity
were solved with the use of a strongly electron-withdrawing Ns
group, which allowed for a successful diastereoselective formal
[3+3]-cycloaddition. An unusual retro-aldol, iminium ion
formation, and transannular Mannich reaction sequence then
furnished the pentacyclic core of 2. However, despite the high
overall efficiency of the cascade sequences developed, the
synthetic route for fastigiatine (2) was not amenable to the
synthesis of other 7-membered-ring-containing Lycopodium
alkaloids.
A unifying strategy was finally realized in the first total
synthesis of the proposed structure of himeradine A (1)
(Scheme 12), which contains 7 rings and 10 stereocenters. A
stereoselective synthesis of the quinolizidine subunit was
accomplished via substrate stereocontrol. A one-pot sequence
for the construction of the strained core system of 1, common
to other 7-membered-ring-containing Lycopodium alkaloids, was
achieved. The isolation of imine intermediate of type 83, the
direct product of the one-pot sequence, would eventually allow
us to access other Lycopodium alkaloid members. A key
biomimetic transannular Mannich reaction then provided the
pentacyclic core of himeradine A (1).
I
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Our unifying strategy was successfully applied toward the
first total syntheses of lycopecurine (3) and dehydrolyco-
pecurine (4), which also featured the key transannular Mannich
reaction. Our one-pot sequence was also successfully utilized in
the concise syntheses of lyconadins A (5) and B (6), which
involved reduction of an imine intermediate derived from our
one-pot sequence instead of an intramolecular Mannich
reaction to construct the core system. The plausibility of our
biosynthetic hypothesis that 7-membered-ring-containing Lyco-
podium alkaloids can arise from a common imine precursor was
further validated by the successful syntheses of 5 and 6.
In summary, we have developed a unifying, biosynthetically
inspired strategy for the synthesis of structurally diverse 7-
membered-ring-containing Lycopodium alkaloids. Through
simple variation of the alkylation partners of versatile
intermediate 20, we were able to divergently access six such
members with diverse carbon skeletons (Scheme 15). We
ACKNOWLEDGMENTS
■
A.S.L. is thankful for an NDSEG and NSF predoctoral
fellowship. B.B.L. acknowledges an NSF predoctoral fellowship.
We thank Professor Hiroshi Morita for providing us with
additional NMR spectra. Dr. Shao-Liang Zheng is acknowl-
edged for assistance with X-ray crystallography. Dr. Shaw
Huang is acknowledged for assistance with NMR experiments.
A.S.L. and B.B.L. would like to thank Dr. Justin Kim for
reviewing this manuscript.
REFERENCES
■
(1) For reviews, see: (a) Kobayashi, J.; Morita, H. In The Alkaloids;
Cordell, G. A., Ed.; Academic Press: New York, 2005; Vol. 61, pp 1−
57. (b) Ma, X.; Gang, D. R. Nat. Prod. Rep. 2004, 21, 752−772.
(c) Ayer, W. A.; Trifonov, L. S. In The Alkaloids; Cordell, G. A., Brossi,
A., Ed.; Academic Press: New York, 1994; Vol. 45, pp 233−266.
(2) For isolation, see: Morita, H.; Hirasawa, Y.; Kobayashi, J. J. Org.
Chem. 2003, 68, 4563−4566.
(3) For isolation, see: (a) Gerard, R. V.; MacLean, D. B.; Fagianni,
R.; Lock, C. J. Can. J. Chem. 1986, 64, 943−949. (b) Gerard, R. V.;
MacLean, D. B. Phytochemistry 1986, 25, 1143−1150.
Scheme 15. A Unified Biosynthetically Inspired Strategy for
the Synthesis of 7-Membered-Ring-Containing Lycopodium
Alkaloids
(4) The positional numbering system we employed for himeradine A
(1) is different from that utilized in the original isolation report (see
ref 2) in order to facilitate comparison with related Lycopodium
alkaloids, which share a common numbering scheme. See Supporting
Information for more details.
(5) For isolation, see: (a) Ayer, W. A.; Altenkirk, B.; Masaki, N.;
Valverde-Lopez, S. Can. J. Chem. 1969, 47, 2449−2455. (b) Ayer, W.
A.; Masaki, N. Can. J. Chem. 1971, 49, 524−527. (c) Braekman, J. C.;
Hootele, C.; Ayer, W. A. Bull. Soc. Chim. Belg. 1971, 80, 83−90.
(6) For isolation, see: (a) Kobayashi, J.; Hirasawa, Y.; Yoshida, N.;
Morita, H. J. Org. Chem. 2001, 66, 5901−5904. (b) Ishiuchi, K.;
Kubota, T.; Hoshino, T.; Obara, Y.; Nakahata, N.; Kobayashi, J. Bioorg.
Med. Chem. 2006, 14, 5995−6000.
(7) (a) Liau, B. B.; Shair, M. D. J. Am. Chem. Soc. 2010, 132, 9594−
9595. (b) Liau, B. B. Ph.D. Thesis, Harvard University, 2013.
(8) Tori, M.; Shimoji, T.; Shimura, E.; Takaoka, S.; Nakashima, K.;
Sono, M.; Ayer, W. A. Phytochemistry 2000, 53, 503−509.
(9) For a similar intramolecular 7-endo-trig cyclization, see:
(a) Beshore, D. C.; Smith, A. B., III. J. Am. Chem. Soc. 2007, 129,
4148−4149. (b) Beshore, D. C.; Smith, A. B., III. J. Am. Chem. Soc.
2008, 130, 13778−13789.
(10) (a) Allinger, N. L.; Riew, C. K. Tetrahedron Lett. 1966, 12,
1269−1272. (b) Stork, G.; Kretchmer, R. A.; Schlessinger, R. H. J. Am.
Chem. Soc. 1968, 90, 1647−1648.
(11) Medda, A. K.; Lee, H.-S. Synlett 2009, 6, 921−924 and
references therein.
envision that this strategy could potentially be readily applied to
the synthesis of all such 7-membered-ring-containing Lyco-
podium alkaloids in order to further enable their biological
profiling.
(12) Nicolaou, K. C.; Hwang, C.-K.; Duggan, M. E.; Nugiel, D. A.;
Abe, Y.; Reddy, K. B.; DeFrees, S. A.; Reddy, D. R.; Awartani, R. A.;
Conley, S. R.; Rutjes, F. P. J. T.; Theodorakis, E. A. J. Am. Chem. Soc.
1995, 117, 10227−10238.
(13) Linghu, X.; Kennedy-Smith, J. J.; Toste, F. D. Angew. Chem., Int.
Ed. 2007, 46, 7671−7673 and references therein.
ASSOCIATED CONTENT
■
S
* Supporting Information
(14) For a review on electrophilic cyclopropanes, see: Danishefsky, S.
Acc. Chem. Res. 1979, 12, 66−72.
1
Experimental procedures, spectroscopic data, copies of H and
13C NMR spectra, and X-ray structure of lycopecurine (3). This
material is available free of charge via the Internet at http://
pubs.acs.org.
(15) Majetich, G.; Leigh, A. J.; Condon, S. Tetrahedron Lett. 1991, 32,
605−608.
(16) Knobloch, E.; Bruckner, R. Synlett 2008, 12, 1865−1869.
̈
(17) (a) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734−736.
(b) Johnson, C. D. Acc. Chem. Res. 1993, 26, 476−482.
(18) Stafford, J. A.; Brackeen, M. F.; Karanewsky, D. S.; Valvano, N.
L. Tetrahedron Lett. 1993, 34, 7873−7876.
AUTHOR INFORMATION
■
Corresponding Author
(19) Lambert, P. H.; Vaultier, M.; Carrie,
5352−5356.
(20) Movassaghi, M.; Chen, B. Angew. Chem., Int. Ed. 2007, 46, 565−
568.
́
R. J. Org. Chem. 1985, 50,
shair@chemistry.harvard.edu
Notes
The authors declare no competing financial interest.
J
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(21) (a) Schumann, D.; Naumann, A. Liebigs Ann. Chem. 1983, 220−
225. (b) Gosh, S. K.; Buchanan, G. S.; Long, Q. A.; Wei, Y.; Al-Rashid,
Z. F.; Sklenicka, H. M.; Hsung, R. P. Tetrahedron 2008, 64, 883−893.
(22) House, H. O.; Tefertiller, B. A.; Olmstead, H. D. J. Org. Chem.
1968, 33, 935−942.
(23) Marcoux, D.; Binschadler, P.; Speed, A. W. H.; Chiu, A.; Pero, J.
̈
E.; Borg, G. A.; Evans, D. A. Org. Lett. 2011, 13, 3758−3761 and
references therein.
(24) (a) Vink, M. K. S.; Schortinghuis, C. A.; Luten, J.; Maarseveen, J.
H.; Schoemaker, H. E.; Hiemstra, H.; Rutjes, F. P. J. T. J. Org. Chem.
2002, 67, 7869−7871. (b) Suzuki, H.; Aoyagi, S.; Kibayashi, C.
Tetrahedron Lett. 1995, 36, 935−936.
(25) Josephsohn, N. S.; Snapper, M. L.; Hoveyda, A. H. J. Am. Chem.
Soc. 2004, 126, 3734−3735.
(26) (a) Beak, P.; Lee, W. K. J. Org. Chem. 1990, 55, 2578−2580.
(b) Beak, P.; Lee, W. K. J. Org. Chem. 1993, 58, 1109−1117.
(c) Dieter, R. K.; Oba, G.; Chandupatla, K. R.; Topping, C. M.; Lu, K.;
Watson, R. T. J. Org. Chem. 2004, 69, 3076−3086. (d) Guerrero, C.
A.; Sorensen, E. J. Org. Lett. 2011, 13, 5164−5167.
(27) Unfortunately, performing an exchange of protecting groups
from Boc to Ns was necessary because attempted nucleophilic
cyclopropane opening of the corresponding Ns-protected derivative of
cyclopropane 25 was unsuccessful.
(28) (a) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H.
J. Am. Chem. Soc. 2003, 125, 11360−11370. (b) Movassaghi, M.;
Tjandra, M.; Qi, J. J. Am. Chem. Soc. 2009, 131, 9648−9650.
(29) For examples, see: (a) Altman, R. A.; Nilsson, B. L.; Overman,
L. E.; Alaniz, J. R.; Rohde, J. M.; Taupin, V. J. Org. Chem. 2010, 75,
7519−7534. (b) Martin, C. L.; Overman, L. E.; Rohde, J. M. J. Am.
Chem. Soc. 2008, 130, 7568−7569.
1
(30) The H NMR spectrum of the free base of himeradine A (1)
was obtained via private communication with Professor Hiroshi
Morita. Please see Supporting Information, p S53.
(31) (a) Heathcock, C. H.; Kleinman, E.; Binkley, E. S. J. Am. Chem.
Soc. 1978, 100, 8036−8037. (b) Heathcock, C. H.; Kleinman, E. F.;
Binkley, E. S. J. Am. Chem. Soc. 1982, 104, 1054−1068.
(32) (a) Bisai, A.; West, S. W.; Sarpong, R. J. Am. Chem. Soc. 2008,
130, 7222−7223. (b) West, S. P.; Bisai, A.; Lim, A. D.; Narayan, R. R.;
Sarpong, R. J. Am. Chem. Soc. 2009, 131, 11187−11194. (c) Nishimura,
T.; Unni, A. K.; Yokoshima, S.; Fukuyama, T. J. Am. Chem. Soc. 2011,
133, 418−419. (d) Nishimura, T.; Unni, A. K.; Yokoshima, S.;
Fukuyama, T. J. Am. Chem. Soc. 2013, 135, 3243−3247. (e) Yang, Y.;
Haskins, C. W.; Zhang, W.; Low, P. L.; Dai, M. Angew. Chem., Int. Ed.
2014, 53, 3922−3925.
(33) Hogenauer, K.; Baumann, K.; Mulzer, J. Tetrahedron Lett. 2000,
̈
41, 9229−9232.
(34) A similar observation was made by Fukuyama and co-workers.
See ref 32d.
K
dx.doi.org/10.1021/ja507740u | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX