A concise and versatile double-cyclization strategy for the highly stereoselective synthesis and arylative dimerization of aspidosperma alkaloids

Article abstract of DOI:10.1002/anie.201200387

Building cycles: A strategy for the concise, stereoselective synthesis of aspidosperma alkaloids and related structures via a common putative diiminium ion intermediate is reported. The approach enables the dimerization of aspidosperma-type structures at the sterically hindered C2 position. The intermediate is prepared in one step from the shown lactam through an electrophilic double-cyclization cascade (see scheme; Tf= trifluoromethanesulfonyl). Copyright

Full text of DOI:10.1002/anie.201200387

.
Angewandte
Communications
DOI: 10.1002/anie.201200387
Alkaloid Synthesis
A Concise and Versatile Double-Cyclization Strategy for the Highly
Stereoselective Synthesis and Arylative Dimerization of Aspidosperma
Alkaloids**
Jonathan William Medley and Mohammad Movassaghi*
The monoterpene indole alkaloids represent the largest
family of alkaloid natural products, the more than 2000
members of which show a broad range of chemical diversity
and potent biological activity.[1] The structural challenges
presented by this family have long been a source of interest,
resulting in the development of a variety of inventive
synthetic strategies to access various family members. The
biogenetically related natural alkaloids N-methylaspido-
spermidine (1), N-methylquebrachamine (2), and tabernae-
bovine (3) represent the aspidosperma subfamily of mono-
terpene indole alkaloids (Figure 1).[2–5] The dimeric alkaloid 3,
isolated from Tabernaemontana bovina in 1998,[2e] has a fas-
alkaloids (ꢀ)-1, (+)-2, and dimeric (+)-dideepoxytabernae-
bovine (4).
Inspired by precedence in biogenetically relevant dimer-
izations of other monoterpene indole alkaloids,[1,5] we postu-
lated the biogenesis of (+)-tabernaebovine (3) to involve the
ꢀ
late-stage union of two aspidosperma fragments at the C2
C15ꢀ linkage. This retrobiosynthetic analysis[7] prompted the
development of a regio- and diastereoselective arylation at C2
of pentacycle 5 (Scheme 1) en route to decacycle (+)-4. We
envisioned that a highly electrophilic diiminium ion 5 would
allow stereo- and regioselective transformations that provide
divergent access to dimeric decacycle (+)-4 as well as
monomeric aspidosperma alkaloids, such as (ꢀ)-1 and (+)-2
(Scheme 1). We expected that reduction of the diiminium ion
5 would afford (ꢀ)-1, whereas hydrative Grob fragmentation
followed by reduction would afford (+)-2. We hypothesized
that the diiminium ion 5 could be generated from lactam
(ꢀ)-7 through a novel, stereoselective double-cyclization
cascade. We envisioned a transformation that involves
spirocyclization of an electrophilically activated lactam inter-
Figure 1. Representative aspidosperma alkaloids.
ꢀ
cinating molecular constitution that exhibits a unique C2
C15’ linkage between two pentacyclic aspidosperma skele-
tons. While elegant strategies for the synthesis of other
dimeric monoterpene indole alkaloids have been reported,[5]
ꢀ
no synthetic solution to the distinctive C2 C15’ bond that is
present in 3 exists. As a outgrowth of our studies concerning
electrophilic amide activation,[6] we report a concise and
convergent strategy for the enantioselective synthesis of
[*] J. W. Medley, Prof. Dr. M. Movassaghi
Department of Chemistry, Massachusetts Institute of Technology
Cambridge, MA 02139 (USA)
E-mail: movassag@mit.edu
Homepage: http://web.mit.edu/movassag/www/index.htm
[**] We are grateful for financial support by NIH-NIGMS (GM074825).
We thank Justin Kim, Owen Fenton, and Dr. Peter Mꢀller for X-ray
crystal structure analysis of 17·2HCl.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201200387.
Scheme 1. Retrosynthetic analysis of the aspidosperma alkaloids.
4572
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4572 –4576

Angewandte
Chemie
mediate onto the 2-chloroindole, and subsequent interruption
of the Bischler–Napieralski reaction by cyclization of an
Attempts to hydrolyze amide (+)-9 to the corresponding
carboxylic acid were unsatisfactory because of competitive
lactone formation under either basic or acidic conditions.
Initial efforts to couple a-quaternary amide (+)-9 with
sulfonamide 8 through nucleophilic displacement of the
chloride atom at C8 proved inefficient; fast N!O acyl
transfer of amide (+)-9 led to intramolecular N alkylation.
This propensity of amide (+)-9, however, could be used to our
advantage for the synthesis of lactam (ꢀ)-7 and recovery of
the chiral auxiliary. Sequential treatment of sulfonamide 8
with potassium hydride and O-silyl-protected derivative
(+)-14 in DMF followed by heating to 1008C afforded N-
alkylated sulfonamide (+)-15 in 86% yield. Desulfonyl-
ation[16] of (+)-15 to the corresponding secondary amine
and in situ desilylation and heating in ethanol afforded the
key lactam (ꢀ)-7 in 95% yield and 94% ee.[9] This single-step
transformation, which occurs through a desilylation/N!O
acyl transfer/lactam cyclization cascade, also leads to efficient
recovery of the chiral auxiliary (ꢀ)-11 in 99% yield
(Scheme 2).[17]
We next focused on the development of a unified strategy
to access a versatile intermediate en route to alkaloids (ꢀ)-1,
(+)-2, and (+)-4. Electrophilic activation of lactam (ꢀ)-7 with
trifluoromethanesulfonic anhydride[6,18] initiated a double-
cyclization cascade leading to the versatile diiminium ion 5
(Scheme 3). Guided by our prior studies on the synthesis of
azaheterocycles by employing electrophilic amide activa-
tion,[6] we recognized that the optimal conditions for con-
version of lactam (ꢀ)-7 into diiminium ion 5 involve the use of
mildly basic additive 3-cyanopyridine in acetonitrile followed
by warming. This transformation relies on electrophilic
activation at C19 of lactam (ꢀ)-7 and rapid nucleophilic
spirocyclization at C12 affording the putative 2-chlorospiro-
indoleninium intermediate 6 (Scheme 1) that undergoes
addition at C2 by the vinyl group and loss of hydrogen
ꢀ
unactivated C3 C4 vinyl group onto the C2 position of the
putative 2-chlorospiroindoleninium intermediate 6. The over-
all stereochemical outcome of the process would be secured
from the resident stereochemistry of the quaternary center at
C5. The requisite lactam (ꢀ)-7 could be simplified by
N acylation and N alkylation transforms to 2-chlorotrypt-
amine sulfonamide 8 and a-quaternary amide (+)-9, the latter
of which could be synthesized diastereoselectively by alkyla-
tive quaternization of an amide enolate.
The concise enantioselective synthesis of the key trypt-
amine lactam (ꢀ)-7 is shown in Scheme 2. Regioselective
methylation of sulfonamide 10[8] via its disodium dianion
provided the requisite N1-methylated derivative in 91%
yield.[9] Subsequent treatment with N-chlorosuccinimide
afforded C2-chlorinated tryptamine 8 in 76% yield. The
quaternary stereogenic center at C5 that we envisioned would
enable stereocontrolled introduction of all stereocenters
found in alkaloids (ꢀ)-1, (+)-2, and (+)-4 was secured by
successive diastereoselective a alkylations of crotonamide
(+)-12. Acylation of (ꢀ)-pseudoephenamine (11)[10] with E-
crotonyl chloride gave the enamide (+)-12 in 97% yield.
Chemoselective g deprotonation of enamide (+)-12 with
lithium 2,2,6,6-tetramethylpiperidide in the presence of
lithium chloride,[11] followed by electrophilic trapping of the
resulting enolate with 3-chloro-1-iodopropane afforded a-
vinyl amide (+)-13 as a single diastereomer in 83% yield.
Inspired by Myersꢀ alkylative quaternizations[12] of pseudo-
ephedrine amides and precedent for a alkylation of a-methyl
crotonimides,[13] we reasoned that deprotonation of a-vinyl
amide (+)-13 would afford the corresponding enolate with
the sterically less-demanding vinyl group cis to the amide
nitrogen. Alkylation from the less-sterically shielded face of
the enolate[10,11] would secure the desired quaternary stereo-
center at C5. Gratifyingly, deprotonation of amide (+)-13
with lithium diisopropylamide in the presence of lithium
chloride, followed by electrophilic trapping with iodoethane
at ꢀ508C in the presence of N,N’-dimethylpropylene urea
provided the a-quaternary amide (+)-9 in 72% yield with an
excellent level of stereoselection (d.r. > 29:1).[9,14,15]
ꢀ
chloride. The ability to employ an unactivated C3 C4 olefin
as the nucleophile in the second cyclization is likely a result of
the enhanced electrophilicity at C2 of intermediate 6
imparted by the chlorine atom,[19] together with a high
resilience of 6 toward an undesired Pictet–Spengler rear-
rangement.[20,21] Consistent with the sensitivity of this double-
cyclization step, the use of less-basic 2-chloropyridine or
Scheme 2. Enantioselective synthesis of lactam (ꢀ)-7: a) NaH, DMF, 238C; MeI, 0!238C, 91%; b) N-chlorosuccinimide, MeCN, 238C, 76%;
c) E-crotonyl chloride, Et3N, THF, ꢀ30!238C, 97%; d) lithium 2,2,6,6-tetramethylpiperidide, LiCl, THF, 0!ꢀ788C; 3-chloro-1-iodopropane,
ꢀ78!08C, 83%; e) lithium diisopropylamide, LiCl, THF, ꢀ78!08C; N,N’-dimethylpropylene urea, ꢀ408C; EtI, ꢀ508C, 72%, d.r.>29:1;
f) triethylsilyl triflate, 2,6-lutidine, CH2Cl2, 238C, 100%; g) (+)-14, 8, KH, nBu4NI, DMF, 1008C, 86%; h) PhSH, K2CO3, DMSO, 238C; KOEt,
Et3N·3HF, EtOH, 858C, 95%, 94% ee, 99% recovery of (ꢀ)-11. DMF=N,N’-dimethylformamide, Ns=2-nitrobenzenesulfonyl, TES=triethylsilyl.
Angew. Chem. Int. Ed. 2012, 51, 4572 –4576
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4573

.
Angewandte
Communications
iminium ion at C19 of intermediate 5 would enhance the
electrophilicity of the iminium ion at C2 both inductively and
by reducing the steric bulk through the flattening of the
DE ring system (see Figure 1). Gratifyingly, treatment of
in situ generated diiminium ion 5 with 4-(N,N-dimethylami-
no)phenyl magnesium bromide at ꢀ408C for 30 seconds
followed by addition of Red-Al afforded hexacyclic C2 ani-
line adduct (ꢀ)-17 in 40% yield as a single diastereomer
(Scheme 3).[9] The steric congestion around C2 in adduct (ꢀ)-
17 is evidenced by the rotation barrier of approximately
20 kcalmolꢀ1 [9] around the C2 C23 s bond, as measured by
ꢀ
coalescence-temperature experiments in NMR spectroscopy.
Also, heating an acidic aqueous solution of diiminium ion 5 to
708C effected Grob fragmentation to give the tetracyclic
lactam (ꢀ)-18[23] in 57% yield in a single step from lactam
(ꢀ)-7. Platinum-catalyzed hydrogenation[24] of the C3 C4
ꢀ
olefin and subsequent carbonyl reduction at C19 with lithium
aluminium hydride at 608C provided (+)-N-methylquebrach-
24
24
amine (2; ½aꢁD ¼ + 102 (c = 0.22, CHCl3); Ref. [2a]: ½aꢁD ¼ +
110 (CHCl3)) in 82% yield over two steps.[9]
With particular interest in evaluating this chemistry as
a general entry to the synthesis of complex aspidosperma
alkaloids, we investigated a series of addition reactions at C2,
which are of relevance in synthetic planning. Importantly,
lactam (ꢀ)-18, which required mild activation conditions,
proved to be an excellent precursor to diiminium ion 5.
Treatment of lactam (ꢀ)-18 with the reagent combination
Tf2O/2-chloropyridine[6] in acetonitrile (238C, 10 min)
resulted in rapid stereo- and regioselective electrophilic
Scheme 3. Synthesis of aspidosperma alkaloids by interception of
diiminium ion 5: a) Tf2O, 3-cyanopyridine, MeCN, 858C; b) NaBH3CN,
THF, 50%; c) 4-(Me2N)-C6H4MgBr, ꢀ408C; Red-Al, 40%; d) trifluoro-
acetic acid, sodium trifluoroacetate, H2O, 708C, 57%; e) H2, Pt/C,
THF, 100%; f) H2, Pt/C, THF; g) LiAlH4, THF, 658C, 82% (two steps).
Thermal ellipsoids in crystal structure of 17·2HCl drawn at 50%
probability. Tf=trifluoromethanesulfonyl, Red-Al=sodium bis(2-
methoxyethoxy)aluminum hydride.
transannular spirocyclization to
5 en route to various
more-nucleophilic pyridine as the additive gave the
desired diiminium ion 5 with reduced efficiency as Table 1: Regio- and stereoselective transformations of lactam (ꢀ)-18.
evidenced by the presence of singly cyclized side
products, recovered starting material, and significant
decomposition.[22] The synthetic versatility of diimi-
nium ion 5 is illustrated by its conversion into
alkaloids (ꢀ)-1, (+)-2, and (ꢀ)-17 (Scheme 3).
In situ reduction of intermediate 5 with sodium
No. Nuc-1[a]
Nuc-2
R1
R2 Product Yield
[%][b]
cyanoborohydride furnished (ꢀ)-N-methyldehy-
droaspidospermidine (16) in 50% yield as a single
diastereomer. Catalytic hydrogenation of cis-alkene
(ꢀ)-16 with a carbon-supported platinum catalyst
quantitatively afforded (ꢀ)-N-methylaspidospermi-
1
2
3
4
NaBH3CN
nBu3SnH
–
H
H
H
(ꢀ)-16 95
NaBD4
Red-Al
–
D[c] (ꢀ)-19 94
H
(ꢀ)-17 76
(ꢀ)-20 76
[d]
24
–
dine (1; ½aꢁD ¼ꢀ23 (c = 0.17, CHCl3); Ref. [3d]:
25
½aꢁD ¼ꢀ23 (c = 1.1, CHCl3); Ref. [2c] for (+)-1:
20
½aꢁD ¼ + 24 (c = 1.25, CHCl3); Scheme 3). All spec-
5
Red-Al
H
H
(ꢀ)-21 59
troscopic data for our synthetically obtained (ꢀ)-
1 were consistent with those reported in the litera-
ture.[3d] The concise enantioselective synthesis of
lactam (ꢀ)-7 combined with the double-cyclization
strategy described above enables rapid access to
useful intermediates with highly reactive iminium
functions at C2 and C19.
6
7
NaHB(OMe3)
NaHB(OMe3)
(ꢀ)-22 92
(ꢀ)-23 79
H
H
8
9
NaHB(OMe3)
–
(ꢀ)-17 74
(ꢀ)-20 73
[d]
The unique reactivity of diiminium ion 5 is
demonstrated by its utility in an arylation at C2,
–
ꢀ
reminiscent of the C2 C15’ bond adjoining the two
halves of the complex natural alkaloid (+)-taber-
[a] Grignard reagents used at ꢀ408C. Other nucleophiles used at 238C. [b] Yield of
isolated single diastereomer. [c] 93% Deuterium incorporation at C19. [d] Isolated
naebovine (3, Figure 1). We reasoned that the vicinal as C19 iminium triflate. 2-ClPyr=2-chloropyridine.
4574
www.angewandte.org ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2012, 51, 4572 –4576

Angewandte
Chemie
C2 adducts 16–23 (Table 1). Use of sodium cyanoborohydride
afforded (ꢀ)-N-methyldehydroaspidospermidine (16) in 95%
yield (Table 1, entry 1), consistent with efficient generation of
the same electrophilic intermediate 5 accessed from lactam
(ꢀ)-7 (Scheme 3). The greater reactivity at C2 compared to
C19 of diiminium ion 5 can be used for regioselective addition
of the first nucleophile at C2.[25] For example, treatment of 5
with tributylstannane followed by introduction of sodium
borodeuteride afforded pentacycle (ꢀ)-19, which is deuter-
ated at C19, in 94% yield with no deuterium enrichment at C2
and 93% deuterium incorporation at C19 (Table 1, entry 2).
Notably, the C2-arylated product (ꢀ)-17 could be prepared
efficiently from lactam (ꢀ)-18 by using 4-(N,N-dimethylami-
no)phenyl magnesium bromide as the first nucleophile
followed by in situ reduction at C19 (Table 1, entry 3, 76%
yield). Alternatively, hexacyclic iminium triflate (ꢀ)-20 could
be isolated and reduced with sodium cyanoborohydride to
(ꢀ)-17 in a subsequent step (Table 1, entries 4 and 9). That
this reduction at C19 of pentacycle (ꢀ)-20 occurs in the
absence of an acidic additive is consistent with the spectro-
scopic data of (ꢀ)-20, thus revealing its iminium ion
structure.[9] It is notable that the rotation barrier of approx-
Scheme 4. Synthesis of (+)-dideepoxytabernaebovine (4): a) Tf2O, 2-
ClPyr, MeCN, 238C; (ꢀ)-16 (1.0 equiv), 858C, 80%; b) Red-Al, 08C,
76%; c) H2, Pt/C, THF, 84%.
methyldehydroaspidospermidine (16) as described above,
with sodium trimethoxyborohydride directly afforded prod-
uct (ꢀ)-25 from (ꢀ)-18 in 73% yield.[9] Apart from increasing
the electrophilicity of the vicinal C2 iminium ion, the
C19 iminium ion may be responsible for reducing the
nucleophilicity of the dimeric intermediate (+)-24, as no
oligomerized products could be observed, even when only
one equivalent of (ꢀ)-16 was employed as nucleophile.
We have developed a concise synthetic strategy to access
the aspidosperma-type molecular framework by employing
a double-cyclization cascade that results in up to three
contiguous stereogenic centers and forms up to three
imately 12 kcalmolꢀ1 [9] around the C2 C23 s bond in imi-
ꢀ
nium ion (ꢀ)-20 is significantly lower than in the reduced
product (ꢀ)-17 (see above), consistent with the aforemen-
tioned structural flattening effect of the iminium ion at C19.
ꢀ
The high electrophilicity of diiminium ion 5 allows C C bond
formation at C2 with highly hindered and mildly nucleophilic
species. Treatment of intermediate 5 with 2,6-dimethylphenyl
magnesium bromide followed by hydride reduction afforded
the highly congested xylene adduct (ꢀ)-21 in 59% yield
(Table 1, entry 5). The high degree of steric congestion
ꢀ
carbon carbon bonds with complete regio- and stereochem-
ical control in a single step. The use of the chiral auxiliary (ꢀ)-
11[10] was critical in enabling our concise and enantioselective
synthesis of the key intermediate (ꢀ)-7. The ability to use an
unactivated olefin as a pendant nucleophile minimizes the
need for functional group removal and allows for concise and
convergent access to complex aspidosperma alkaloids. We
have shown putative diiminium ion 5 to be a highly reactive
and versatile intermediate, allowing the rapid enantioselec-
tive total syntheses of (ꢀ)-N-methylaspidospermidine (1) and
(+)-N-methylquebrachamine (2) in eight and nine steps,
respectively, from E-crotonyl chloride and (ꢀ)-pseudoephen-
around C2 in (ꢀ)-21 is evidenced by the complete lack of
1
ꢀ
observable C2 C23 s-bond rotation on the H NMR time-
scale, even at 1408C. Reaction of 5 with 2-methallyltri-
methylsilane or 1-(tert-butyldimethylsilyloxy)-1-methoxye-
thene and subsequent hydride reduction afforded methallyl
adduct (ꢀ)-22 (Table 1, entry 6, 92% yield) and methyl
acetate adduct (ꢀ)-23 (Table 1, entry 7, 79% yield), respec-
tively. The utility of this strategy to access C2-arylated
derivatives is highlighted by a Friedel–Crafts reaction of 5
with N,N-dimethylaniline (238C, 90 min) and either in situ
reduction at C19 to provide C2-arylated amine (ꢀ)-17 or
isolation of the pentacyclic C19 iminium salt (ꢀ)-20 (Table 1,
entries 8 and 9, 74% and 73% yield, respectively).
ꢀ
amine (11), as well as the unprecedented C C bond
formations onto the highly congested C2 position of the
aspidosperma skeleton. The power of this synthetic strategy
ꢀ
With insight gained from these studies, in particular
entries 8 and 9 of Table 1, we sought to implement this
chemistry in effecting the dimerization of two pentacyclic
aspidosperma-type molecular frameworks at the challenging
has been demonstrated in the first example of a C2 C15’
dimerization of two aspidosperma-type systems, a complex
assembly drawing on biogenetic considerations of (+)-3, in
the synthesis of (+)-dideepoxytabernaebovine (4).
ꢀ
C2 C15’ linkage (Scheme 4). In the event, electrophilic
Received: January 15, 2012
Published online: March 12, 2012
activation of tetracyclic lactam (ꢀ)-18 followed by treatment
with equimolar (ꢀ)-N-methyldehydroaspidospermidine (16)
and heating to 858C afforded the decacyclic iminium triflate
(+)-24 in 80% yield. Subsequent reduction at C19 of (+)-24
gave (ꢀ)-didehydrodideepoxytabernaebovine (25), which
upon hydrogenation provided (+)-dideepoxytabernaebovine
(4) in 64% yield over two steps (Scheme 4). Alternatively,
in situ reduction at C19 of dimeric iminium ion (+)-24, which
was formed through the union of lactam (ꢀ)-18 with (ꢀ)-N-
Keywords: alkaloids · dimerization · enantioselectivity · indole ·
.
total synthesis
[1] a) J. E. Saxton, The Alkaloids, Chem. Biol. 1998, 51, 1; b) P. M.
Dewick in Medicinal Natural Products:
A Biosynthetic
Angew. Chem. Int. Ed. 2012, 51, 4572 –4576
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4575

.
Angewandte
Communications
Approach, 2nd ed., Wiley, Chichester, 2001, p. 350; c) S. E.
OꢀConnor, E. McCoy, Recent Adv. Phytochem. 2006, 40, 1;
d) S. E. OꢀConnor in Comprehensive Natural Products II, Vol. 1
(Eds.: L. Mander, H.-W. Liu), Elsevier, Amsterdam, 2010, p. 977.
Chen, Angew. Chem. 2009, 121, 7752; Angew. Chem. Int. Ed.
2009, 48, 7616.
[6] a) M. Movassaghi, M. D. Hill, O. K. Ahmad, J. Am. Chem. Soc.
2007, 129, 10096; b) M. Movassaghi, M. D. Hill, Org. Lett. 2008,
10, 3485; c) J. W. Medley, M. Movassaghi, J. Org. Chem. 2009, 74,
1341.
[7] M. Movassaghi, D. S. Siegel, S. Han, Chem. Sci. 2010, 1, 561.
[8] Y. Matsuda, M. Kitajima, H. Takayama, Org. Lett. 2008, 10, 125.
[9] See the Supporting Information for details.
[10] M. R. Morales, K. T. Mellem, A. G. Myers, Angew. Chem. Int.
Ed., Angew. Chem., 2012, 124, 4646 – 4649; Angew. Chem. Int.
Ed., 2012, 51, 4568 – 4571. We are grateful to Professor Myers for
providing us with a generous sample of (ꢀ)-pseudoephenamine
(11).
´
[2] a) J. Mokry, I. Kompis, L. Dubravkova, P. Sefcovic, Tetrahedron
Lett. 1962, 3, 1185; b) J. Mokry, I. Kompis, Lloydia 1964, 27, 428;
c) J. Mokry, I. Kompis, G. Spiteller, Collect. Czech. Chem.
Commun. 1967, 32, 2523; d) M. Zꢁches-Hanrot, J.-M. Nuzillard,
B. Richard, H. Schaller, H. A. Hadi, T. Sꢂvenet, L. Le Men-
Olivier, Phytochemistry 1995, 40, 587; e) T. P. Lien, C. Kamper-
dick, T. V. Sung, G. Adam, H. Ripperger, Phytochemistry 1998,
49, 1797; f) K. Merzweiler, T. P. Lien, T. V. Sung, H. Ripperger,
G. Adam, J. Prakt. Chem. 1999, 341, 69.
[3] For syntheses related to 1, see: a) N. Benchekroun-Mounir, D.
Dugat, J.-C. Gramain, H.-P. Husson, J. Org. Chem. 1993, 58,
6457; b) S. A. Kozmin, T. Iwama, Y. Huang, V. H. Rawal, J. Am.
Chem. Soc. 2002, 124, 4628; c) J. P. Marino, M. B. Rubio, G. F.
Cao, A. de Dios, J. Am. Chem. Soc. 2002, 124, 13398; d) H.
Ishikawa, G. I. Elliot, J. Velcicky, Y. Choi, D. L. Boger, J. Am.
Chem. Soc. 2006, 128, 10596; e) Y. Sasaki, D. Kato, D. L. Boger,
J. Am. Chem. Soc. 2010, 132, 13533; f) S. B. Jones, B. Simmons,
A. Mastracchio, D. W. C. MacMillan, Nature 2011, 475, 183; for
syntheses related to 2, see: Ref. [3b] and g) S. Takano, M.
Yonaga, K. Ogasawara, J. Chem. Soc. Chem. Commun. 1981,
1153; h) M. Node, H. Nagasawa, K. Fuji, J. Am. Chem. Soc. 1987,
109, 7901; i) O. Temme, S.-A. Taj, P. G. Andersson, J. Org. Chem.
1998, 63, 6007; j) M. Amat, O. Lozano, C. Escolano, E. Molins, J.
Bosch, J. Org. Chem. 2007, 72, 4431; k) S. J. Malcolmson, S. J.
Meek, E. S. Satterly, R. R. Schrock, A. H. Hoveyda, Nature
2008, 456, 933; l) E. S. Sattely, S. J. Meek, S. J. Malcolmson, R. R.
Schrock, A. H. Hoveyda, J. Am. Chem. Soc. 2009, 131, 943.
[11] a) A. G. Myers, B. H. Yang, H. Chen, L. McKinstry, D. J.
Kopecky, J. L. Gleason, J. Am. Chem. Soc. 1997, 119, 6496;
b) B. H. Yang, PhD thesis, Harvard University, 1997.
[12] D. A. Kummer, W. J. Chain, M. R. Morales, O. Quiroga, A. G.
Myers, J. Am. Chem. Soc. 2008, 130, 13231.
[13] T. Abe, T. Suzuki, K. Sekiguchi, S. Hosokawa, S. Kobayashi,
Tetrahedron Lett. 2003, 44, 9303.
[14] W. J. Chain, A. G. Myers, Org. Lett. 2007, 9, 355.
[15] A parallel strategy using (+)-pseudoephedrine as chiral auxiliary
did not provide products with higher than 6:1 d.r.
[16] T. Fukuyama, C.-K. Jow, M. Cheung, Tetrahedron Lett. 1995, 36,
6373.
[17] At partial conversion the uncyclized secondary C9-amine retains
the triethylsilyl group consistent with O-desilylation preceding
cyclization.
[18] A. B. Charette, M. Grenon, Can. J. Chem. 2001, 79, 1694.
ꢀ
[19] Use of C2 H derivatives of 7 gave no double-cyclization
products.
ˇ
[4] For a review, see a) J. Hꢃjꢄcek, Collect. Czech. Chem. Commun.
[20] a) D. Desmaꢅle, K. Mekouar, J. dꢀAngelo, J. Org. Chem. 1997,
62, 3890; b) Y. Yasui, H. Takeda, Y. Takemoto, Chem. Pharm.
Bull. 2008, 56, 1567.
[21] An elegant report involving a functional pendant nucleophile
(Ref. [4e]) requires additional steps for synthetic simplification.
[22] The use of less-electrophilic dehydrating agents (e.g., trifluoro-
acetic anhydride, POCl3, POBr3, the Hendrickson reagent or the
Burgess reagent) provided none of the desired cyclization
products.
2004, 69, 1681; for related syntheses, see Ref. [5] and: b) M. E.
Kuehne, D. E. Podhorez, T. Mulamba, W. G. Bornmann, J. Org.
Chem. 1987, 52, 347; c) K. Cardwell, B. Hewitt, M. Ladlow, P.
Magnus, J. Am. Chem. Soc. 1988, 110, 2242; d) S. Kobayashi, G.
Peng, T. Fukuyama, Tetrahedron Lett. 1999, 40, 1519; e) F. He, Y.
Bo, J. D. Altom, E. J. Corey, J. Am. Chem. Soc. 1999, 121, 6771;
f) S. Kobayashi, T. Ueda, T. Fukuyama, Synlett 2000, 883.
[5] a) P. Mangeney, R. Z. Andriamialisoa, N. Langlois, Y. Langlois,
P. Potier, J. Am. Chem. Soc. 1979, 101, 2243; b) J. P. Kutney,
L. S. L. Choi, J. Nakano, H. H. Tsukamoto, M. McHugh, C. A.
Boulet, Heterocycles 1988, 27, 1845; c) M. E. Kuehne, P. A.
Matson, W. G. Bornmann, J. Org. Chem. 1991, 56, 513; d) P.
Magnus, J. S. Mendoza, A. Stamford, M. Ladlow, P. Willis, J. Am.
Chem. Soc. 1992, 114, 10232; e) S. Yokoshima, T. Ueda, S.
Kobayashi, A. Sato, T. Kuboyama, H. Tokuyama, T. Fukuyama,
J. Am. Chem. Soc. 2002, 124, 2137; f) H. Ishikawa, D. A. Colby,
D. L. Boger, J. Am. Chem. Soc. 2008, 130, 420; g) H. Ueda, H.
Satoh, K. Matsumoto, K. Sugimoto, T. Fukuyama, H. Tokuyama,
Angew. Chem. 2009, 121, 7736; Angew. Chem. Int. Ed. 2009, 48,
7600; h) K. C. Nicolaou, S. M. Dalby, S. Li, T. Suzuki, D. Y.-K.
[23] For semisynthesis of a similar tetracyclic lactam, see P. Yates,
F. N. MacLachlan, I. D. Rae, Can. J. Chem. 1978, 56, 1052.
ꢀ
[24] Hydrogenation at 400 psi; the C3 C4 olefin in 3,4-dehydroque-
brachamine systems is resilient to hydrogenation, see F. E.
Ziegler, J. A. Kloek, P. A. Zoretic, J. Am. Chem. Soc. 1969, 91,
2342.
[25] The steric shielding of the C19 iminium ion inhibited the
addition of carbon nucleophiles at C19.
[26] CCDC 862060 (17·2HCl) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
4576
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4572 –4576