Palladium-catalyzed decarboxylative vinylation of potassium nitrophenyl acetate: Application to the total synthesis of (±)-goniomitine

Article abstract of DOI:10.1002/anie.201209970

Merge and divert: The natural product (±)-goniomitine was synthesized by a method featuring two key steps: 1) fragment coupling to a functionalized cyclopentene by a novel palladium-catalyzed decarboxylative vinylation reaction and 2) an unprecedented one-pot integrated oxidation/reduction/cyclization (IORC) process to convert the substituted cyclopentene into the tetracyclic skeleton of goniomitine with high chemo-, regio-, and diastereoselectivity.

Full text of DOI:10.1002/anie.201209970

.
Angewandte
Communications
DOI: 10.1002/anie.201209970
Natural Product Synthesis
Palladium-Catalyzed Decarboxylative Vinylation of Potassium
Nitrophenyl Acetate: Application to the Total Synthesis of (Æ)-
Goniomitine**
Zhengren Xu, Qian Wang, and Jieping Zhu*
The class of monoterpene indole alkaloids, which is currently
comprised of more than 2000 members with broad structural
diversity and important bioactivities, has fascinated scientists
for over a century.[1] Goniomitine (1), a unique member of the
aspidosperma alkaloid family, was isolated from the root bark
of Gonioma malagasy by Husson and co-workers in 1987.[2]
The unprecedented octahydroindolo[1,2-a][1,8]naphthyridine
skeleton, together with a tryptophol moiety, rather than the
normal tryptamine fragment, resulted from the oxidative
skeletal rearrangement of vincadifformine. The interesting
molecular architecture and anti-proliferative activity of this
Scheme 1. Retrosynthesis of goniomitine.
natural product have made it a popular synthetic target.[3] To
date, five total syntheses have been reported from the groups
of Takano,[4] Pagenkopf,[5] Waser,[6] Mukai,[7] and Bach[8]
Although building 2,3-difunctionalized indole followed by
the stepwise formation of rings C and D is a common feature
of these syntheses, the synthetic routes have been significantly
shortened since the inaugural 28 step synthesis of Takano in
1991, thanks to the development of new reactions and
strategies for the construction of the key 2,3-disubstituted
indoles.
years.[12] Although the intramolecular decarboxylative allyla-
tion/benzylation of nitrophenyl acetates[13] and the intermo-
lecular arylation of nitrophenyl acetates are known,[14] we
noticed that vinyl triflates have never been employed as
coupling partners in decarboxylative couplings with aliphatic
carboxylic acids.[15] We began our studies by examining the
decarboxylative coupling of potassium 2-nitrophenylacetate
(5a) with cyclohex-1-en-1-yl trifluoromethanesulfonate (6a).
As shown in Table 1, only the protiodecarboxylation product
In connection with our continuous interest in the total
synthesis of indole alkaloids by late-stage construction of the
indole nucleus,[9,10] we thought to build the whole tetracyclic
scaffold of goniomitine from functionalized cyclopentene 2 by
a one-pot integrated oxidation/reduction/cyclization (IORC)
sequence (Scheme 1).[11] Although different multi-step syn-
theses of 2 could be envisaged,[9d] the as-yet unknown direct
decarboxylative coupling of two readily accessible building
blocks (the substituted potassium salt of o-nitrophenyl
acetate 3 and vinyl triflate 4) was deemed to be the method
of choice. This strategy, if realized, would allow us to
accomplish the total synthesis of goniomitine in only two
operations from the simple building blocks 3 and 4.
Table 1: Survey of reaction conditions for the decarboxylative coupling
reaction.[a]
Entry
Pd source
Ligand
Solvent
Yield [%][b]
1[c]
2
3
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
[{PdCl(allyl)}2]
X-Phos
X-Phos
X-Phos
X-Phos
X-Phos
S-Phos
tBu3P·HBF4
Cy3P·HBF4
Cy-JohnPhos
XantPhos[f]
–
mesitylene
mesitylene
diglyme
DMF
DMF
DMF
DMF
DMF
DMF
DMF
–
–
63
4
77(90)[d]
45
5[e]
6
71
43
58
60
51
54
66
The transition-metal-catalyzed decarboxylative coupling
of carboxylic acids (especially aryl, alkenyl, and alkynyl
carboxylic acids) has attracted much attention in recent
7
8
9
10
11
12
[g]
[PdCl2(dppf)]2
DMF
DMF
[Pd(Ph3P)4][g]
–
[*] Dr. Z. Xu, Dr. Q. Wang, Prof. Dr. J. Zhu
Ecole Polytechnique Fꢀdꢀrale de Lausanne
EPFL-SB-ISIC-LSPN, BCH 5304, 1015 Lausanne (Switzerland)
E-mail: jieping.zhu@epfl.ch
[a] Reaction conditions: 5a (0.12 mmol), 6a (0.1 mmol), Pd source
(2.0 mol%), ligand (6.0 mol%), in the given solvent (c 0.2m) at 1008C
for 2 h. [b] Yield of isolated product. [c] Run at 1508C. [d] Using 6a
(1.5 equiv). [e] Run at 408C for 16 h. [f] 3.0 mol%. [g] 4.0 mol%.
Cy-JohnPhos=(2-biphenyl)dicyclohexylphosphine, diglyme=diethylene
glycol dimethyl ether, dppf=1,1’-bis(diphenylphosphino)ferrocene,
S-Phos=2-dicyclohexylphosphino-2’,6’-dimethyoxylbiphenyl,
XantPhos=4,5-bis(diphenylphophino)-9,9-dimethylxanthene,
X-Phos=2-dicyclohexylphosphino-2’,4’,6’-triisopropylbiphenyl.
Homepage: http://lspn.epfl.ch
[**] We thank EPFL (Switzerland), the Swiss National Science Founda-
tion (SNSF) and the Swiss National Centres of Competence in
Research (NCCR) for financial support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201209970.
3272
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3272 –3276

Angewandte
Chemie
was obtained when the reaction was performed in mesitylene
at 1508C in the presence of [{PdCl(allyl)}2] and X-Phos
(entry 1).[14] Lowering the reaction temperature to 1008C did
reduce the rate of the protiodecarboxylation process, but
failed to produce the desired cross-coupling product. It was
subsequently found that solvent played an important role in
this coupling reaction. By simply switching the solvent to
diglyme (entry 3) or DMF (entry 4), the cross-coupling
product 7a was isolated in yields of 63% and 77%,
respectively. The yield of 7a was further increased to 90%
by using an excess of vinyl triflate (1.5 equiv, entry 4). Even at
408C, the reaction proceeded in DMF to provide 7a in 45%
yield, although a longer reaction time (16 h) was needed
(entry 5). Both monodentate and bidentate phosphine ligands
were effective in the coupling reaction (entries 6–12). Never-
theless, X-Phos stood out as the ligand of choice for the
present coupling reaction.
coupling product was detected with cyclooctenyl triflate.
When sterically congested vinyl triflates were used as
coupling partners, diglyme proved to be a better solvent
than DMF (7e and 7 f). The a-alkylated nitrophenyl acetates
participated in the reaction to provide the cross-coupling
products (7g–7k) in good yields. Functional groups such as
free hydroxy groups, tert-butyldimethylsilyl (TBS) ethers, and
double bonds, were tolerated. No diastereoselectivity was
observed in the coupling of a-alkylated nitrophenyl acetates
with chiral vinyl triflates, which could have important
mechanistic implications. The para-nitrophenyl acetate can
also be used as a coupling partner to produce the vinylated
product 7k in 54% yield.
Mechanistically, decarboxylation could occur either
before[16] or after the formation of the Csp3 Csp2 bond. To
À
gain insight into the reaction pathway, a deuterium labeling
experiment was performed. The reaction of vinyl triflate 6a
with potassium 2-deuterium-2-(2-nitrophenyl)-3-phenylpro-
panoate (8; > 95% D) under standard conditions afforded
the coupling product [D]7h in 50% yield (> 95% D) together
The scope of this reaction was then explored by varying
the carboxylates and vinyl triflates used under the optimized
conditions: [{PdCl(allyl)}2] (2.0 mol%), X-Phos (6.0 mol%),
DMF or diglyme, c = 0.2m, 1008C, 2 h (Scheme 2). The
reaction worked well with cyclopentenyl, cyclohexenyl, and
cycloheptenyl triflates, whereas only a trace amount of the
with the protiodecarboxylated product
9
(> 95% D,
Scheme 3a). The full preservation of deuterium in both
[D]7h and 9 indicated that the a-vinylation of 8 through its
Scheme 2. Reaction scope of the decarboxylative coupling. General
conditions: 5 (0.30 mmol), 6 (0.45 mmol), [{PdCl(allyl)}2] (2.0 mol%),
X-Phos (6.0 mol%), 1008C, 2 h, under argon, in the indicated solvent
(1.5 mL, 0.2m). Yields shown are of product after flash column
chromatography. [a] 5 (0.36 mmol), 6 (0.30 mmol) were used. [b]
[{PdCl(allyl)}2] (5.0 mol%) and X-Phos (15.0 mol%) were used. [c]
Yield obtained after in situ deprotection of the TBS ether by TBAF
(2.0 equiv) at room temperature for 4 h. TBS=tert-butyldimethylsilyl,
Tf =trifluoromethanesulfonate.
Scheme 3. Proposed mechanism and supportive evidence.
Angew. Chem. Int. Ed. 2013, 52, 3272 –3276
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3273

.
Angewandte
Communications
enolate form[17] followed by decarboxylation was not involved
in the present catalytic process.
of the primary hydroxy group, the 3-substituted cyclopent-2-
enone 15.[19] Conjugate addition of ethyl cuprate to 15
followed by trapping of the resulting enolate with the
Comins reagent[20] afforded vinyl triflate 4 in 82% yield. On
the other hand, potassium 4-(benzyloxy)-2-(2-nitrophenyl)-
butanoate (3) was synthesized from 16 by a three-step
sequence involving alkylation, hydrolysis, and salt formation.
The key decarboxylative coupling of 3 with vinyl triflate 4
proceeded smoothly to provide the desired product in 70%
yield. We subsequently found that, after completion of the
coupling reaction, adding tetrabutylammonium fluoride
(TBAF) to the reaction mixture directly afforded alcohol 18
in 70% yield as a mixture of two diastereomers in a 1:1 ratio.
Although the two diastereomers were separable, it was more
convenient to proceed with the synthesis using the mixture, as
the stereocenter at the benzylic position will disappear at the
end of the synthesis. Transformation of the hydroxy group
into an azido group under Mitsunobu conditions[21] set the
stage for the planned one-step conversion of 2 into the
tetracyclic skeleton of goniomitine. After an extensive survey
of reaction conditions, the one-pot integrated oxidation/
double reduction/triple cyclization (IORC) process was
realized as follows: Ozonolysis of 2 in methanol at À788C
in the presence of NaHCO3,[22] followed by the addition of
dimethyl sulfide (À788C, then at RT for 24 h) afforded the
ketoaldehyde. The addition of activated zinc and CaCl2
followed by heating the reaction mixture to reflux furnished
19 in 80% yield as the only diastereomer.[23] Deprotection of
the benzyl ether in the presence of the sensitive aminal group
was best realized with sodium naphthalenide to give (Æ)-
goniomitine (1) in 65% yield.[24] The spectroscopic data of the
synthetic goniomitine were identical to those reported in the
literature.[2,4–8]
Based on these experimental observations, a possible
reaction mechanism is depicted in Scheme 3b. Oxidative
addition of vinyl triflate to the Pd0 generated in situ affords
vinyl palladium(II) A which, upon ligand exchange with the
potassium salt, affords the carboxylic-acid-ligated intermedi-
ate B. Reductive elimination from intermediate C (Path a),
which is generated after extruding CO2 from B, then furnishes
the desired coupling product 7.
The following experimental results provided further
support to the proposed mechanism. First, a small amount
(ca. 6%) of reduction products 10 and 11, which result from
b-hydride elimination from intermediate C (Path b) were
isolated. That reductive elimination from C leading to 7
prevailed over b-hydride elimination under our reaction
conditions is remarkable considering that there were no
obvious structural constraints in C to retard the latter process.
Second, the reaction of potassium 2-methyl-2-(4-nitophenyl)-
propanoate (12) with vinyl triflate 6a afforded the cross-
coupling product 7l (32% yield, Scheme 3c), thus indicating
À
that the C C bond formation occurred after the extrusion of
CO2.
Having a reliable coupling method in hand, the total
synthesis of goniomitine was accomplished as shown in
Scheme 4. The reaction of the Normant Grignard reagent,[18]
prepared in situ from 3-chloropropan-1-ol (13), with
3-ethoxy-cyclopent-2-enone (14) afforded, after protection
The remarkable synthetic efficiency and selectivity in the
present one-pot multiple-bond-forming IORC process is
worthy of additional comment. First, the use of NaHCO3 as
an additive for ozonolysis is of the utmost importance to get
a high yield of 20. The ketoaldehyde 20 can be isolated in
60% yield and is fully characterized. However, partial
decomposition was observed during flash chromatographic
purification. Secondly, all issues of chemo-, regio- and
stereoselectivity were addressed in the one-pot reductive
triple-cyclization process. The optimal conditions should
allow us not only to chemoselectively reduce both the nitro
and the azido groups without touching the ketone, the
aldehyde, the iminium intermediate, and the final aminal
functions, but also to promote the desired regio- and
diastereoselective cyclizations. After many unsuccessful
trials using different hydrogenolysis conditions and metal/
additive combinations, it was found that refluxing a methanol
solution of 20 in the presence of zinc dust and calcium
chloride (CaCl2) afforded 19 directly and in excellent yield.
The presence of CaCl2 was crucial for the success of the
reaction, otherwise only the starting ketoaldehyde 20 was
recovered. On the other hand, attempts to reduce the ozonide
directly after ozonolysis under the same conditions (Zn/
CaCl2) resulted in complete decomposition. The cyclization
proceeded rapidly in a highly ordered fashion after reduction
as no intermediates or side products could be detected from
Scheme 4. Total synthesis of (Æ)-goniomitine. a) 13, CH3MgCl, THF,
À788C to RT; Mg, reflux; 14, reflux; 2n HCl. b) TBSCl, imidazole,
DMF, RT, 60% over 2 steps. c) EtMgBr, CuBr·Me2S, THF, À788C to
À408C; Comins reagent, À408C to RT, 24 h, 80%. d) Cs2CO3,
ICH2CH2OBn, DMF, 608C, 93%. e) 10% KOH, MeOH/THF (5:1).
f) tBuOK, EtOH, 91%. g) 3 (1.2 equiv), [{PdCl(allyl)}2] (5 mol%), X-
Phos (15 mol%), diglyme, 1008C, 2 h; TBAF, RT, 4 h, 70%. h) DPPA,
DIAD, Ph3P, THF, 08C to RT, 3 h, 72%. i) O3, NaHCO3, MeOH,
À788C; Me2S, À788C to RT; Zn, CaCl2, reflux, 2 h, 80%. j) Sodium
naphthalenide, THF, À208C, 15 min, 65%. Comins reagent=N-(5-
chloro-2-pyridyl)bis(trifluoromethanesulfonimide), DIAD=diisopropyl
azodicarboxylate, DPPA=diphenyl phosphoryl azide.
3274
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3272 –3276

Angewandte
Chemie
the reaction mixture at different time intervals. Furthermore,
the domino reductive triple cyclization was found to be
compatible with the ozonolysis conditions, and the one-pot
IORC process of 2 afforded 19 in 80% yield, which is much
higher than the overall yield of the stepwise process. In
Scheme 5 the reaction pathway of the IORC sequence is
Keywords: decarboxylative coupling · goniomitine ·
indole alkaloids · natural products · palladium
.
[1] For reviews, see: a) J. E. Saxton, Alkaloids 1998, 51, 1 – 197;
b) M. Ishikura, K. Yamada, T. Abe, Nat. Prod. Rep. 2010, 27,
ˇ
1630 – 1680; c) J. Hꢀjꢁcek, Collect. Czech. Chem. Commun. 2011,
76, 2023 – 2083, and references therein.
[2] L. Randriambola, J.-C. Quirion, C. Kan-Fan, H.-P. Husson,
Tetrahedron Lett. 1987, 28, 2123 – 2126.
[3] a) C. Hashimoto, H.-P. Husson, Tetrahedron Lett. 1988, 29,
4563 – 4566; b) G. Lewin, C. Schaeffer, Nat. Prod. Lett. 1995, 7,
227 – 234; c) G. Lewin, C. Schaeffer, P. H. Lambert, J. Org.
Chem. 1995, 60, 3282 – 3287; d) G. Lewin, C. Schaeffer, R.
Hocquemiller, E. Jacoby, S. Lꢂonce, A. Pierrꢂ, G. Atassi,
Heterocycles 2000, 53, 2353 – 2356; e) C. F. Gꢃrtler, E. Steckhan,
S. Blechert, J. Org. Chem. 1996, 61, 4136 – 4143. For a semi-
synthesis from vincadifformine, see: f) G. Lewin, G. Bernadat, G.
Aubert, T. Cresteil, Tetrahedron 2013, 69, 1622 – 1627.
[4] S. Takano, T. Sato, K. Inomata, K. Ogasawara, J. Chem. Soc.
Chem. Commun. 1991, 462 – 464.
[5] C. L. Morales, B. L. Pagenkopf, Org. Lett. 2008, 10, 157 – 159.
[6] F. De Simone, J. Gertsch, J. Waser, Angew. Chem. 2010, 122,
5903 – 5906; Angew. Chem. Int. Ed. 2010, 49, 5767 – 5770.
[7] M. Mizutani, F. Inagaki, T. Nakanishi, C. Yanagihara, I. Tamai,
C. Mukai, Org. Lett. 2011, 13, 1796 – 1799.
[8] L. Jiao, E. Herdtweck, T. Bach, J. Am. Chem. Soc. 2012, 134,
14563 – 14572.
Scheme 5. The one-pot integrated oxidation/reduction/cyclization
(IORC) process.
[9] For selected examples of monoterpene indole alkaloid synthesis
by late-stage indole formation, see: a) G. Stork, J. E. Dolfini, J.
Am. Chem. Soc. 1963, 85, 2872 – 2873; b) S. D. Knight, L. E.
Overman, G. Pairaudeau, J. Am. Chem. Soc. 1995, 117, 5776 –
5788; c) J. Bonjoch, D. Solꢂ, S. Garcꢁa-Rubio, J. Bosch, J. Am.
Chem. Soc. 1997, 119, 7230 – 7240; d) M. A. Toczko, C. H.
Heathcock, J. Org. Chem. 2000, 65, 2642 – 2645; e) S. A.
Kozmin, T. Iwama, Y. Huang, V. H. Rawal, J. Am. Chem. Soc.
2002, 124, 4628 – 4641; f) T. Ohshima, Y. Xu, R. Takita, S.
Shimizu, D. Zhong, M. Shibasaki, J. Am. Chem. Soc. 2002, 124,
14546 – 14547; g) L. Zu, B. W. Boal, N. K. Garg, J. Am. Chem.
Soc. 2011, 133, 8877 – 8879; h) Y. Miura, N. Hayashi, S. Yokosh-
ima, T. Fukuyama, J. Am. Chem. Soc. 2012, 134, 11995 – 11997;
i) L. McMurray, E. M. Beck, M. J. Gaunt, Angew. Chem. 2012,
124, 9422 – 9425; Angew. Chem. Int. Ed. 2012, 51, 9288 – 9291.
[10] T. Gerfaud, C. Xie, L. Neuville, J. Zhu, Angew. Chem. 2011, 123,
4040 – 4043; Angew. Chem. Int. Ed. 2011, 50, 3954 – 3957.
[11] For a review on domino processes, see: a) L. F. Tietze, G.
Brasche, K. M. Gericke, Domino Reactions in Organic Synthesis,
VCH, Weinheim, 2006; b) K. C. Nicolaou, D. J. Edmonds, P. G.
Bulger, Angew. Chem. 2006, 118, 7292 – 7344; Angew. Chem. Int.
Ed. 2006, 45, 7134 – 7186; for reviews on integrated reaction
sequences, see: c) C. Vaxelaire, P. Winter, M. Christmann,
Angew. Chem. 2011, 123, 3685 – 3687; Angew. Chem. Int. Ed.
2011, 50, 3605 – 3607; d) J.-I. Yoshida, K. Saito, T. Nokami, A.
Nagaki, Synlett 2011, 1189 – 1194.
[12] For reviews, see: a) N. Rodrꢁguez, L. J. Goossen, Chem. Soc. Rev.
2011, 40, 5030 – 5048; b) R. Shang, L. Liu, Sci. China Chem. 2011,
54, 1670 – 1687, and references therein.
[13] a) S. R. Waetzig, J. A. Tunge, J. Am. Chem. Soc. 2007, 129,
14860 – 14861; b) J. D. Weaver, A. Recio III, A. J. Grenning,
J. A. Tunge, Chem. Rev. 2011, 111, 1846 – 1913.
[14] R. Shang, Z. Huang, L. Chu, Y. Fu, L. Liu, Org. Lett. 2011, 13,
4240 – 4243.
[15] For the decarboxylative coupling of vinyl triflates with a,b-ynoic
acids, see: H. Kim, P. H. Lee, Adv. Synth. Catal. 2009, 351, 2827 –
2832.
summarized. Concomitant reduction of nitro and azido
groups followed by chemo- and regioselective indolization
and iminium formation would afford 23 through intermedi-
ates 21 and 22. Intramolecular attack of the indole nitrogen
on iminium from the face opposite to the ethyl substituent
would diastereoselectively produce the cis-fused aminal ring,
thus completing the construction of the tetracycle 19.
In summary, a seven step total synthesis of (Æ)-goniomi-
tine has been achieved through two key steps: 1) A novel
palladium-catalyzed decarboxylative coupling reaction
between potassium nitrophenyl acetates and vinyl triflates
for the rapid production of a functionalized cyclopentene that
incorporates all of the atoms required for the desired natural
product. Although this coupling reaction was developed for
the synthesis of monoterpene indole alkaloids, we believe that
it will also find application in the synthesis of other families of
natural products and medicinally relevant compounds. 2) A
one-pot multiple-bond-forming IORC process for converting
the substituted cyclopentene into the tetracyclic skeleton of
goniomitine. In this process, the oxidative scission of a double
bond, the chemoselective reduction of an azido and a nitro
À
group, and the concurrent formation of three C N bonds and
three rings took place with high regio-, chemo-, and diaste-
reoselectivity. The sequence completely reorganized the
skeleton of the starting material, thus allowing the conversion
of an easily accessible cyclopentene derivative into the
tetracyclic structure of goniomitine.
Received: December 13, 2012
Published online: February 7, 2013
[16] a) R. Shang, Z.-W. Yang, Y. Wang, S.-L. Zhang, L. Liu, J. Am.
Chem. Soc. 2010, 132, 14391 – 14393; b) R. Shang, D.-S. Ji, L.
Angew. Chem. Int. Ed. 2013, 52, 3272 –3276
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3275

.
Angewandte
Communications
Chu, Y. Fu, L. Liu, Angew. Chem. 2011, 123, 4562 – 4566; Angew.
Chem. Int. Ed. 2011, 50, 4470 – 4474; c) R. Shang, Z. Huang, X.
Xiao, X. Lu, Y. Fu, L. Liu, Adv. Synth. Catal. 2012, 354, 2465 –
2472; d) P. Y. Yeung, K. H. Chung, F. Y. Kwong, Org. Lett. 2011,
13, 2912 – 2915; e) Y.-S. Feng, W. Wu, Z.-Q. Xu, Y. Li, M. Li, H.-
J. Xu, Tetrahedron 2012, 68, 2113 – 2120.
121, 10264 – 10269; c) L. F. Tietze, H. Schirok, M. Wçhrmann,
Chem. Eur. J. 2000, 6, 510 – 518.
[20] D. L. Comins, A. Dehghani, Tetrahedron Lett. 1992, 33, 6299 –
6302.
[21] B. Lal, B. N. Pramanik, M. S. Manhas, A. K. Bose, Tetrahedron
Lett. 1977, 18, 1977 – 1980.
[22] S. L. Schreiber, R. E. Claus, J. Reagan, Tetrahedron Lett. 1982,
23, 3867 – 3870.
[23] R. J. Sundberg, W. J. Pitts, J. Org. Chem. 1991, 56, 3048 – 3054.
[24] a) K. D. Philips, J. P. Horwitz, J. Org. Chem. 1975, 40, 1856 –
1858; b) M. T. Crimmins, M. Shamszad, A. E. Mattson, Org.
Lett. 2010, 12, 2614 – 2617.
[17] F. Bellina, R. Rossi, Chem. Rev. 2010, 110, 1082 – 1146.
[18] G. Cahiez, A. Alexakis, J. F. Normant, Tetrahedron Lett. 1978,
19, 3013 – 3014.
[19] a) G. A. Molander, M. Hiersemann, Tetrahedron Lett. 1997, 38,
4347 – 4350; b) L. F. Tietze, H. Schirok, J. Am. Chem. Soc. 1999,
3276
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 3272 –3276