Catalytic enantioselective desymmetrisation as a tool for the synthesis of hodgkinsine and hodgkinsine B

Article abstract of DOI:10.1002/chem.201203150

Two palladium-catalysed amination protocols are deployed in the desymmetrisation of the complex dimeric alkaloid meso-chimonanthine. The power of these transformations is showcased in an efficient formal and total synthesis of the natural products hodgkinsine and hodgkinsine B, respectively. Two palladium-catalysed amination protocols are deployed in the desymmetrisation of the complex dimeric alkaloid meso-chimonanthine. The power of these transformations is showcased in efficient formal and total synthesis of the natural products hodgkinsine and hodgkinsine B (see figure), respectively. Copyright

Full text of DOI:10.1002/chem.201203150

DOI: 10.1002/chem.201203150
Catalytic Enantioselective Desymmetrisation as a Tool for the Synthesis of
Hodgkinsine and Hodgkinsine B
Robert H. Snell,[a] Matthew J. Durbin,[a] Robert L. Woodward,[b] and Michael C. Willis*[a]
Abstract: Two palladium-catalysed amination protocols are deployed in the de-
symmetrisation of the complex dimeric alkaloid meso-chimonanthine. The power
of these transformations is showcased in an efficient formal and total synthesis of
the natural products hodgkinsine and hodgkinsine B, respectively.
Keywords: alkaloids · desymmetri-
zation · heterocycles · natural prod-
ucts · total synthesis
Introduction
Pyrrolidinoindoline alkaloids have been the focus of chemi-
cal research for over seven decades,[1] and investigations into
their synthesis and reactivity have spearheaded the develop-
ment of numerous synthetic methodologies.[1,2] This family
of natural products plays host to the polypyrrolidinoindoline
alkaloids; a complex series of oligomeric compounds that
contain up to eight monomeric units (vatamidine).[3b] These
secondary metabolites are abundant in nature and have
been isolated from a multitude of sources.[4] The compounds
themselves often exhibit potent biological activity including
anti-bacterial,[5] anti-fungal,[5] anti-viral,[5] cytotoxic[5,6] and
analgesic[7] properties. The importance of this area of chem-
istry has been highlighted in several comprehensive review
articles.[3] Overman reported in depth the chemistry of struc-
turally similar compounds containing a linkage between
quaternary carbon atoms, such as hodgkinsine (2; Fig-
ACHTUNGTRENNUNG
ure 1).[3b] More recently, a review by Albericio and ꢀlvarez
looked more broadly at the structure and bioactivity of
these fascinating natural products.[3c] These alkaloids have
been the target of numerous synthetic campaigns including
recent achievements by Overman, Danishefsky, Movassaghi
and Baran.[8] Selected targets (1–7) are illustrated by
Figure 1.
Exploitation of symmetry as a synthetic strategy has been
a key research focus within the Willis group for several
years.[9,10] The inherent elements of symmetry possessed by
Figure 1. Selected polypyrrolidinoindoline alkaloids.
[a] Dr. R. H. Snell, Dr. M. J. Durbin, Dr. M. C. Willis
Department of Chemistry
the cyclotryptamine alkaloids shown in Figure 1 make them
attractive targets for synthesis, and despite numerous ele-
gant synthetic strategies,[3] our recent total synthesis of
hodgkinsine B (3)[9d] and Overmanꢁs synthesis of quadrige-
mine C (5)[8e] represent the only examples which utilise an
enantioselective desymmetrisation step to establish the ab-
solute stereochemical configuration of the target. In this ar-
ticle we elaborate on our initial synthesis, and also demon-
strate an alternative enantioselective desymmetrisation pro-
tocol.
University of Oxford
Chemistry Research Laboratory, Mansfield Road
Oxford, OX1 3TA (UK)
Fax : (+44)1865-285002
E-mail: michael.willis@chem.ox.ac.uk
[b] Dr. R. L. Woodward
AstraZeneca Global Medicines Development
Macclesfield Works, Hurdsfield Industrial Estate
Macclesfield, Cheshire, SK10 2NA (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203150.
16754
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16754 – 16764

FULL PAPER
Results and Discussion
recent report by Takayama and co-workers disclosed the
synthesis of chimonanthine in just two steps as a separable
mixture of regio- and diastereoisomers.[13] In our recent
communication we reported that by simple modifications to
this procedure we could scale the reaction to >40 g scale
and isolate the desired meso-isomer directly by crystalliza-
tion (Scheme 2).[9d]
Our synthetic strategies for the total synthesis of hodgkin-
sine focussed on the exploitation of the reactive secondary
amine groups (aminals) in a symmetry-breaking operation.
There are numerous examples of pro-chiral nucleophiles
being utilised in enantioselective desymmetrisation proce-
dures,[11] but these have generally involved alcohols;[9a,11g]
the increased nucleophilicity of amines renders them a more
challenging class of substrate.[11a,b] As a result, comparatively
few examples have been reported in the literature.[11] Sever-
al possible enantioselective reactions were investigated, in-
cluding acylation[11d,h] and Suzuki coupling reactions.[9b] This
report focuses on our two most successful desymmetrisation
approaches, namely
a
Buchwald–Hartwig cross cou-
pling[11e,f,h] and a Trost allylation.[11a,b,12] Retrosynthetic analy-
ses of hodgkinsine are shown in Scheme 1, in which the core
stereochemistry of the molecule is defined through an enan-
tioselective desymmetrisation.
Scheme 2. Chromatography-free synthesis of meso-chimonanthine (1).
Reagents and conditions: 1)
8
(1.00 equiv), PIFA (0.65 equiv),
CF3CH2OH, À308C, 5.5 h. 2) 9 (1.00 equiv), Red-Al (10.00 equiv), PhMe,
reflux, 16 h. Red-Al=sodium bis(2-methoxyethoxy)aluminiumhydride.
Enantioselective
desymmetrisation—Buchwald–Hartwig
cross coupling: With a highly efficient supply of multi-gram
quantities of meso-chimonanthine (1), the stage was set for
our key desymmetrisation step. Our interest in the Buch-
wald–Hartwig amination reaction prompted us to examine
its utility in desymmetrisation reactions. Minimal literature
precedent in the form of a kinetic resolution of a halocyclo-
phane reported by Rossen and Pye hinted that such a reac-
tion was possible.[14] 4-Bromoanisole was selected as a cou-
pling partner for our initial investigations as its oxidative
cleavage is well documented.[15] We quickly established reac-
tion conditions capable of forming the desired product, in
its racemic form, using PdACHTNUTRGNENG(U OAc)2, the mono-phosphine
ligand S-Phos (2-dicyclohexylphosphino-2’,6’-dimethoxybi-
phenyl), and KOtBu in toluene (Scheme 3). Alternative sol-
Scheme 1. Retrosynthetic strategy for the synthesis of hodgkinsine.
Synthesis of meso-chimonanthine (1): The synthetic chal-
À
lenge posed by the congested quaternary C-3a C-3a’ linkage
and four contiguous stereocentres of chimonanthine (1) has
meant that total synthesis of this target has received a great
deal of attention amongst the synthetic community. In addi-
Scheme 3. Desymmetrisation of meso-chimonanthine (1). Reagents and
conditions: Compound 1 (1.0 equiv), 4-bromoanisole (1.3 equiv), KOtBu
(1.5 equiv), PdACHTNGUTERNNU(G OAc)2 (5 mol%), S-Phos (6 mol%), PhMe, 508C, 16 h.
tion to other elegant enantioselective syntheses,[8a,b,f]
a
Chem. Eur. J. 2012, 18, 16754 – 16764
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16755

M. C. Willis et al.
Table 2. Aryl halide evaluation.[a]
vent and base selection generally gave a dramatic decrease
in yield.
After an initial lead for the enantioselective reaction em-
ploying (R)-BINAP (2,2’-bis(diphenylphosphino)-1,1’-bi-
naphthyl, 25% ee) (Table 1 entry 2), we undertook an exten-
sive evaluation of chiral ligands; selected examples are sum-
marised in Table 1.
Yield [%][b]
31
ee [%]
À
Ar X
Table 1. Chiral ligand evaluation.[a]
1
2
3
4
40
49
41
40
–
11
trace
Ligand
Yield [%][b]
ee [%]
1
2
3
4
5
6
7
8
9
S-Phos
(R)-BINAP
(R)-MOP
34
9
38
31
15
–
5
6
18
34
13
8
25
26
40
40
36
47
42
55
53
T
N
[a] meso-Chimonanthine (1; 1.0 equiv), Ar-X (1.3 equiv), KOtBu
(1.5 equiv), Pd(OAc)2 (5 mol%), (R,S)-PPFA (6 mol%), PhMe, 508C,
16 h. [b] Isolated yield, absolute configuration of products was not estab-
lished.
R
trace
48
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
A
U
44
A
U
45[d]
57[e]
10
A
CHTUNGTRENNUNG
As can be seen in Table 2, the substrate scope for the de-
symmetrisation was limited. Although electron-rich, elec-
tron-poor and vinyl substrates were tolerated in the reaction
when using the optimised catalyst system, electron-poor aryl
halides gave dramatically reduced reactivity (Table 2, en-
tries 4 and 5). Perhaps most surprising was the sharp de-
crease in yield on the switch from a methyl ether to an ethyl
ether (Table 2, entry 1 vs. 3). We concluded from our sub-
strate screen that the selected catalyst system was heavily
optimised for our original substrate. Consequently, we re-
peated the study using MOP as the ligand; unfortunately,
this delivered a similar set of data.
[a] meso-Chimonanthine (1, 1.0 equiv), 4-bromoanisole (1.3 equiv),
KOtBu (1.5 equiv), Pd(OAc)2 (5 mol%), ligand (6 mol%), PhMe, 508C,
16 h. [b] Isolated yield, absolute configuration of products was not estab-
lished. [c] Prepared using literature procedure.[16] [d] 24% Bis-arylated
product isolated. [e] 13% Bis-arylated product isolated.
ACHTUNGTRENNUNG
We observed that the bis-arylated material 11 was isolated
as a by-product in the more high yielding reactions (Table 1,
entries 8–10). These elevated yields appeared to correlate
with increased enantioselectivity and suggested that a kinet-
ic resolution may be in effect. To test this hypothesis, enan-
tioenriched product 10 (40% ee) was re-subjected to the re-
action conditions, but no appreciable increase in ee was ob-
served (Scheme 4; for the related study based on enantiose-
lective allylation, see Scheme 7).
Extensive investigations were carried out by examining
variation of solvent, base, catalyst loadings, temperature and
additives; none of these offered a significant enhancement
on our previous optimal conditions (Table 1, entry 10).
During investigations on the purification of the desym-
metrised material 10, we noted that the racemic product was
highly crystalline. This allowed us to increase the ee of our
product to approximately 99% by precipitation of the race-
mic material.
The investigation indicated that the BINAP-derived
ligand, MOP, and the ferrocene based ligands PPFA and
SL-J003-1 were most effective (Table 1, entries 3, 4 and 7;
structures of the ligands are given at the bottom of the
table). A publication by Hartwig and co-workers disclosed
that pre-complexes of the Josiphos family of ligands offered
increased efficiency in amination reactions.[16] Pleasingly, in
our hands we observed a similar boost in reactivity and se-
lectivity and were able to obtain enantioenriched product in
a 57% yield (Table 1, entry 10). Considering the complexity
of the substrate, this result compares very well to even the
simplest intramolecular systems reported in the litera-
ACHTUNGTRENNUNG
coupling partner to facilitate later cleavage of the protecting
group. It was hoped that alternative aryl halides would offer
an enhancement in yield and enantioselectivity (Table 2).
16756
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16754 – 16764

Natural Product Synthesis
FULL PAPER
gave at best, modest enantiodiscrimination (ca. 50% ee). A
more efficient reaction would be necessary for this desym-
metrisation approach to be a viable process.
In 2004, Taguchi and co-workers disclosed the desymmetr-
isation of simple, generally cyclic, diamines by means of a
Trost allylation.[11] Important observations from this study
included the need to attenuate the nucleophilicity of the
amine functionality with electron-withdrawing sulfonyl
groups. In addition, it was found that an increased distance
between the amine groups resulted in a significant drop in
enantioselectivity. meso-Chimonanthine (1), with four con-
tiguous stereocentres, four basic nitrogen centres and distant
secondary amines represented a challenging substrate for
such a reaction. Nevertheless, the potential utility of this re-
action for a synthesis of the high order polypyrrolidinoindo-
line alkaloids was clear.
When meso-chimonanthine (1) was exposed to the reac-
tion conditions of Taguchi, to our delight, we obtained allyl-
amine 16 with excellent enantioselectivity (99% ee). Howev-
er, the yield and scalability of the reaction were unpredicta-
ble; on occasion, only unreacted meso-chimonanthine (1)
could be recovered (Scheme 6).
Scheme 4. Attempted kinetic resolution.[17]
After our moderate success in the development of a de-
symmetrisation protocol, we chose to apply our methodolo-
gy to a formal synthesis of hodgkinsine (2). We hoped to in-
tercept the synthesis by Overman and co-workers by prepar-
ing the aryl iodide 15 used in their C-3a/C-7 bond-forming
Stille coupling. In the Overman synthesis, the racemic
iodide is used as a Heck coupling precursor and by means
of a diastereomeric resolution they were able to prepare
hodgkinsine (2) and hodgkinsine B (3).[8c,d] By employing an
enantioselective desymmetrisation, we hoped to be able to
generate the aryl halide 15 in an enantioenriched form, thus
removing the dependence on a resolution procedure.
For the formal synthesis we adopted the same ortho-lithia-
tion strategy used by Overman.[8c,d] A directing group (tert-
butyloxycarbonyl; Boc) was installed on the remaining free
amine of our desymmetrised material 10 to yield carbamate
12 in 83% yield. Gratifyingly, this transformation could also
be accomplished in a “one-pot” protocol directly from
meso-chimonanthine (1) in the same overall yield. Halogen-
ation at the C-7-position of 12 proceeded smoothly utilising
molecular iodine. The synthesis was completed with an un-
optimised deprotection sequence employing TMSOTf and
CAN to deliver aryl iodide 15, a key intermediate in the
Overman synthesis (Scheme 5).
Scheme 6. Desymmetrisation by means of a Trost allylation. Reagents
and conditions: 1) Compound 1 (1.0 equiv), allyl acetate (1.1 equiv),
KOtBu (2.0 equiv), (R,R)-DACT-Ph Trost ligand (7.3 mol%), [{Pd-
A
DACT-Ph Trost ligand=(1R,2R)-(+)-1,2-diaminocyclohexane-N,N’-bis(2-
diphenylphosphinobenzoyl).
A brief exploration of reaction conditions indicated that
by switching the base from KOtBu to Et3N,[18] the reliability
of the reaction could be dramatically improved. With this
Enantioselective desymmetrisation—Trost allylation: Our
asymmetric Buchwald–Hartwig protocol although effective,
Scheme 5. Formal synthesis of hodgkinsine. Reagents and conditions: 1) Compound 1 (1.0 equiv), 4-bromoanisole (1.3 equiv), Pd
[(R,S)-SL-J009-1]PdCl2 (6 mol%), KOtBu (1.5 equiv), PhMe, 508C, 16 h. 2) N-PMP-chimonanthine 10 (1.0 equiv), NaHMDS (2.2 equiv), Boc2O
(1.3 equiv), THF, RT, 16 h (“one pot”), compound 1 (1.0 equiv), 4-bromoanisole (1.2 equiv), Pd(OAc)2 (5 mol%), (R,S)-PPFA (6 mol%), KOtBu
ACHTUNGRTEN(NUNG OAc)2 (5 mol%),
AHCTUNGTRENNUNG
(1.5 equiv), PhMe, 658C, 16 h; Boc2O (1.5 equiv), NaHMDS (2.2 equiv), RT, 6 h. 3) N-PMP-N’-Boc-chimonanthine 12 (1.0 equiv), sBuLi (2.0 equiv),
TMEDA (3.0 equiv), I2 (2.0 equiv), Et2O, À788C, 30 min, 08C, 1 h. 4) N-PMP-N’-Boc-iodochimonanthine 13 (1.0 equiv), TMSOTf (2.2 equiv), CH2Cl2,
RT, 3 h. 5) N-PMP-iodochimonanthine 14 (1.0 equiv), MeCN [0.06m], CAN (3.0 equiv), H2O, 0 8C, 30 min. For completion of synthesis see Overman.[8c,d]
PMP=para-methoxyphenyl, HMDS=hexamethyldisilazane, TMEDA=N,N,N’,N’-tetramethylethylenediamine, TMS=trimethylsilyl, OTf=trifluoro-
ACHTUNGTRENNUNGmethanesulfonate, CAN=ceric ammonium nitrate.
Chem. Eur. J. 2012, 18, 16754 – 16764
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16757

M. C. Willis et al.
modification, we obtained consistent yields to the order of
55% with no compromise in enantioselectivity. This result
was attributed to the increased solubility of the amine base.
Finally, switching from a 1:1 PhMe/dioxane solvent system
to PhMe alone, increased the yield to 76%. We found that
the Trost allylation desymmetrisation protocol could be
easily performed on a multi-gram scale (3.0 g). Switching to
the naphthyl Trost ligand resulted in decreased conversion.
Similarly to the Buchwald–Hartwig strategy described
above, the major by-product from the reaction was bis-func-
tionalised material 17. Interestingly, we observed that the re-
action proceeded in high enantioselectivity even at elevated
temperatures (508C), although accompanied by increased
amounts of the bis-allylated compound 17 (Table 3).
C-7 Functionalisation—intramolecular a-arylation: The use
of enolates in palladium-catalysed arylation chemistry is
well documented.[22] In 2008 we reported the first example
of the direct coupling of an oxindole by means of a palladi-
um-catalysed a-arylation.[23] We hoped that this methodolo-
gy could be readily adapted for use in the synthesis of hodg-
kinsine B (3). The formation of quaternary carbon centres is
not a trivial undertaking and only a few simple examples
using the a-arylation of oxindoles have been reported.[22,23]
To overcome the predicted difficulty associated with the
direct formation of a quaternary chiral centre, we proposed
an intramolecular a-arylation facilitated by an amide tether.
We anticipated that the use of a chiral ligand would allow
the configuration of the newly formed stereocentre to be
catalyst-controlled.[23c] The resulting spirolactam could be
opened by a suitable nucleophile (e.g., MeNR2), and a final
deprotection/reductive amination cascade should yield the
natural product (Scheme 8).
Table 3. Enantioselective allylation.
Ligand
T
[8C]
16
ee
[%]
17
Yield[b]
[%]
Yield[b]
[%]
1[c]
2
3
N
0
0
50
76
56
62
>99
>99
>99
24
9
37
[a] Reagents and conditions: meso-Chimonanthine (1; 1.0 equiv), allyl
acetate (1.2 equiv), Et3N (2.0 equiv), [{PdCl2A(allyl)}2] (2.5 mol%, Pd),
CTHUNGTRENNUNG
ligand (3.8 mol%), PhMe. See Supporting Information for ligand struc-
tures. [b] Isolated yield. [c] 3 g scale.
In order to ascertain whether the high enantioselectivity
was the result of a kinetic resolution, we synthesised racemic
mono-allylated material 16, and re-subjected it to the de-
symmetrisation conditions (Scheme 7). The mono-allylated
Scheme 8. Intramolecular arylation retrosynthetic analysis.
The tethered “masked cyclotryptamine” fragment was
synthesised from N-benzylisatin 18 on gram scale without
chromatography by means of a facile Wittig reaction with
ylide 19 (Scheme 9). Following olefination, ester hydrolysis
of ester 20 (TFA) and alkene hydrogenation (H2, Pd/C) pro-
vided the key acid 22 in good yield.
Scheme 7. Kinetic resolution of 16.[17]
material recovered from the reaction had been enantio-
ACHTUNGTRENNUNGenriched to approximately 99% ee, thus confirming that a
As a simple model system, we elected to couple acid 22
with 7-haloindolines,[24] but regrettably the coupling was
hampered by the steric bulk of the ortho-halide. Investiga-
tion of numerous amide coupling agents identified that opti-
mal yields were obtained when the acid 22 was activated as
the acid chloride. The effect of steric bulk is highlighted by
the drop in yield on moving from Cl-Br-I (Scheme 10).
With the tethered oxindoles 23 and 24 in hand, we evalu-
ated various a-arylation conditions, which are summarised
kinetic resolution “proof reading” step was in part responsi-
ble for the high enantioselectivity of the reaction.[19]
The results presented above confirmed that desymmetri-
sation by means of Trost allylation represented a viable syn-
thetic method for the synthesis of polypyrrolidinoindoline
alkaloids, and so attention focused on the remaining syn-
thetic challenges posed by these complex targets.[20,21]
16758
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16754 – 16764

Natural Product Synthesis
FULL PAPER
the free amine group of 16 in a quantitative yield, on a
multi-gram scale with no chromatography. The same prod-
uct could be synthesised in a “one-pot” protocol by addition
of Boc2O to the crude reaction mixture after allylation. Lith-
iation of 27 with sBuLi and subsequent trapping of the aryl
lithium with dibromoethane afforded 28 in 61% yield (95%
based on recovered starting material, BRSM). Recovered 27
was re-subjected to the reaction conditions to produce an
additional crop of halogenated material. Finally, cleavage of
the directing group with TMSOTf yielded the desired aryl
halide 29. This whole sequence was performed in three-
steps from meso-chimonanthine (1) with only two chromato-
graphic operations on gram scale (Scheme 11). The 7’-chloro
analogue 30 was prepared in a similar fashion (see Support-
ing Information).[26]
Scheme 9. Synthesis of acid coupling partner. Reagents and conditions:
1) N-Benzylisatin 18 (1.05 equiv), 19 (1.0 equiv), PhMe, reflux, 1 h
À208C, 16 h. 2) TFA (20.0 equiv), 508C, 2 h. 3) 21 (1.0 equiv), Pd/C
(5 mol% Pd), H2 (balloon), MeOH, RT, 16 h.
Scheme 10. Synthesis of amide 23 and 24. Reagents and conditions: 1) in-
doline (1.0 equiv), acid 22 (1.2 equiv), oxalyl chloride (2.0 equiv), DMF
(cat), CH2Cl2, RT, 2 h. DMF=N’N-dimethylformamide.
by Table 4. Pleasingly, a catalyst system based on the
BINAP-derived ligand MOP gave rise to the desired spiro-
lactam 25 in a 61% yield (entry 5). In all successful reac-
tions, appreciable levels of by-product 26 were also isolated,
resulting from attack by the amide enolate versus the oxin-
dole enolate, followed by an oxidative step.
Following the success of the model system studies, we at-
tempted to apply the tethered a-arylation methodology to
the total synthesis of hodgkinsine B. The required 7-halochi-
monanthine portion was synthesised by a directed ortho-lith-
iation reaction.[25] A directing group (Boc) was installed on
Scheme 11. Synthesis of 7’-Br-N-allyl-chimonanthine 29. Reagents and
conditions: 1) Compound 1 (1.0 equiv), allyl acetate (1.2 equiv), Et3N
(2.0 equiv), [{PdACTHUNRGTNEUNG(allyl)Cl}2] (2.5 mol%, Pd), (R,R)-DACT-Ph Trost ligand
(3.8 mol%), PhMe, 08C, 1.5 h. 2) Compound 16 (1.0 equiv), Boc2O
(1.3 equiv), NaHMDS (2.2 equiv), THF, RT, 2 h. 3) 1 (1.0 equiv), allyl
acetate (1.2 equiv), Et3N (3.0 equiv), [{PdACTHNUTRGNEUNG(allyl)Cl}2] (5 mol%), (R,R)-
DACT-Ph Trost ligand (6 mol%), PhMe, 508C, 45 min. 4) Boc2O
(1.1 equiv), RT, 16 h. 5) Compound 27 (1.0 equiv), TMEDA (3.0 equiv),
sBuLi (2.5 equiv); dibromoethane (4.0 equiv), Et2O, À78 to 08C, 1.5 h. 6)
Compound 28 (1.0 equiv), TMSOTf (2.5 equiv), CH2Cl2, RT, 16 h.
Table 4. Spirolactam formation.
The attempted coupling of either the 7’-chloro- or bromo-
chimonanthine-based amines 29 or 30 using a variety of
amide-forming conditions resulted in no more than trace
amounts of the desired product 31 (Scheme 12). This lack of
reactivity was attributed to the dramatically increased steric
bulk associated with the natural product precursors when
compared to our previous model system. The investigation
of alternative a-arylation protocols was therefore required.
X
Ligand
Yield 25 [%][b]
Yield 26 [%][b]
1
2
3
4
5
6
Cl
Cl
Cl
Cl
Cl
Br
PEPPSI
S-Phos
X-Phos
48
28
41
33
61
42
33
37
33
28
28
13
.
PtBu3 HBF4
MOP
MOP
C-7 Functionalisation—intermolecular a-arylation: Our
second a-arylation strategy hinged upon an intermolecular
reaction with a tryptamine-derived oxindole coupling part-
[a] Reagents and conditions: Amide 23 or 24 (1.0 equiv), Cs2CO3
(1.1 equiv), Pd(OAc)2 (5 mol%), ligand (6 mol%), PhMe, 1008C, 16–
ACHTUNGTRENNUNG
À
24 h. [b] Isolated yield.
ner to form the key C-3a C-7 bond (Scheme 13).
Chem. Eur. J. 2012, 18, 16754 – 16764
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16759

M. C. Willis et al.
of the aryl halide and the presence of multiple basic amine
sites. The products from these reactions were isolated as
mixtures of diastereoisomers; however, they could conceiva-
bly be used in a synthesis of quadrigemine C (5) after an ap-
propriate asymmetric a-alkylation reaction (Scheme 15).[27]
Scheme 12. Unsuccessful amide coupling.
Scheme 15. Model a-arylation and proposed synthesis of quadrigemine C
(5). Reagents and conditions: meso-Dibromo-chimonanthine 34
.
(1.0 equiv), Pd
N
(2.2 equiv), Cs2CO3 (6.0 equiv), PhMe, 1008C, 3 h. The products were iso-
lated as a mixture of diastereoisomers.
Scheme 13. Intermolecular retrosynthetic strategy.
To investigate the viability of this reaction, we utilised a
meso-symmetric model. Substrate 34 was prepared in an
identical manner to the meso-diiodide used in the Overman
synthesis of quadrigemine C (Scheme 14).[8e]
This result prompted an attempt to form the required
quaternary centre directly by means of the same intermolec-
ular a-arylation methodology. As a model system, we select-
ed a benzyl/tosyl protected oxindole 41, derived from trypt-
We subjected meso-dibromide 34 to a variety of a-aryla-
ACHTUNGTRENaNUGN mine, which was synthesised in a three-step protocol from
tion conditions using several simple oxindoles. A catalyst
N-methyl tryptamine 38 (Scheme 16).
.
system consisting of PtBu3 HBF4 and Pd
N
We observed that the use of conventional indole oxidation
conditions on 40 (HCl, DMSO)[28] gave rise to none of the
desired oxindole 41. However, exposure of indole 40 to
NBS and water delivered the product in a modest 44%
yield (accompanied by appreciable levels of the 3-bromoox-
indole product; Scheme 16).[29] Nevertheless, with oxindole
41 in hand, the use of our a-arylation conditions gave a mix-
tion with Cs2CO3 was found to give optimal results. Pleas-
ingly, both N-benzyl and N-allyl oxindoles reacted in excel-
lent yields; an impressive result considering the steric bulk
Scheme 16. Synthesis of oxindole coupling partner 41. Reagents and con-
ditions: 1) N-Methyltryptamine 38 (1.0 equiv), tosylchloride (1.1 equiv),
Et3N (1.4 equiv), DMAP (9 mol%), CH2Cl2, 08C–RT, 6 h. 2) N-Methyl-
N-tosyltryptamine 39 (1.0 equiv), NaH (1.4 equiv), BnBr (1.1 equiv),
DMF, RT, 3 h. 3) N’-Bn-N-methyl-N-tosyl-tryptamine 40 (1.0 equiv),
NBS (2.4 equiv), H2O (4.0 equiv), tBuOH:THF (1:1), RT, 30 min.
DMAP=N,N-dimethylaminopyridine.
Scheme 14. Synthesis of meso-di-bromide 34. Reagents and conditions: 1)
Compound 1 (1.0 equiv), Boc2O (2.5 equiv), NaHMDS (4.5 equiv), THF,
RT, 2 h. 2) meso-bis-Boc-chimonanthine 32 (1.0 equiv), sBuLi (4.0 equiv),
TMEDA (6.0 equiv); dibromoethane (10.0 equiv), Et2O, À78 to 08C, 2 h.
3) meso-Dibromo-bis-Boc-chimonanthine 33 (1.0 equiv), TMSOTf
(6.0 equiv), CH2Cl2, RT, 16 h.
16760
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16754 – 16764

Natural Product Synthesis
FULL PAPER
ture of the mono- 43 and bis-arylated 42 products as a mix-
ture of diastereoisomers. The mono-product 43 appeared to
have been de-halogenated, presumably through a transfer
hydrogenation mechanism (Scheme 17).
proved effective in the reduction of any over-oxidised mate-
rial.
The allyl and Boc analogues were synthesised in a similar
manner (Scheme 19). Although bis-allyl oxindole 46 was iso-
lated in a modest 35% yield, we were able to prepare it di-
rectly from methyl-tryptamine (38). Gratifyingly, we were
able to synthesise the allyl/Boc analogue 45 on a multi-gram
scale in high yields without the need for chromatography.
Scheme 17. Model a-arylation. Reagents and conditions: meso-Dibromo-
Scheme 19. Synthesis of oxindoles 45 and 46. Reagents and conditions: 1)
N-Methyltryptamine 38 (1.0 equiv), KOtBu (2.4 equiv), allylbromide
(3.0 equiv), THF, RT, 16 h. 2) N,N’-Allyl-N-methyltryptamine (1.0 equiv),
NBS (2.4 equiv), H2O (4.0 equiv), tBuOH:THF (1:1), RT, 30 min. 3)
NaBH4 (6.0 equiv). 4) N-Methyltryptamine 38 (1.0 equiv), Boc2O
(1.2 equiv), Et3N (3.0 equiv), THF, RT, 1.5 h. 5) N-Boc-N-methyltrypta-
mine 48 (1.0 equiv), KOtBu (1.2 equiv), allylbromide (1.5 equiv), THF,
RT, 2 h. 6) N’-Allyl-N-Boc-N-methyltryptamine 49 (1.0 equiv), NBS
(2.4 equiv), H2O (4.0 equiv), tBuOH:THF (1:1), RT, 30 min. 7) NaBH4
(6.0 equiv).
.
chimonanthine 34 (1.0 equiv), Pd
N
(10 mol%), oxindole 41 (2.5 equiv), Cs2CO3 (6.0 equiv), PhMe, 1008C,
20 h. Mono- and bis-substituted products were isolated as inseparable
mixtures of diasteroisomers.[30]
In applying this result to the desymmetrised core of chi-
monanthine (16), we selected three oxindole coupling part-
ners (44–46), furnished with groups compatible with a host
of global deprotection strategies. In each example, the oxin-
dole nitrogen atom was masked by an allyl group (Figure 2).
Reaction of the oxindoles bearing tosyl and Boc protect-
ing groups on the pendent nitrogen with desymmetrised aryl
halide 29 gave the desired a-arylation products in good
yields (Table 5, entries 1 and 2). However, only trace
Table 5. a-Arylation substrates.
Figure 2. Proposed coupling partners.
The allyl/tosyl oxindole 44 was prepared by a simple
three-stage process, using NBS/H2O to effect the indole oxi-
dation (Scheme 18). A reductive workup using NaBH4
R
Compound
Yield [%][b]
1
2
3
Ts
Boc
allyl
50
51
52
44
77
trace
Scheme 18. Synthesis of oxindole 44. Reagents and conditions: 1) N-
Methyl-N-tosyltryptamine 39 (1.0 equiv), KOtBu (1.2 equiv), allylbro-
mide (1.5 equiv), THF, RT, 16 h. 2) N’-Allyl-N-methyl-N-tosyltryptamine
47 (1.0 equiv), NBS (2.4 equiv), H2O (4.0 equiv), tBuOH:THF (1:1), RT,
30 min. 3) NaBH4 (6.0 equiv). NBS=N-bromosuccinimide.
[a] Reagents and conditions: 29 (1.0 equiv), 44–46 (1.7 equiv), Cs2CO3
.
(6.0 equiv), Pd
ACHTUGNRTNE(NUNG OAc)2 (5 mol%), PtBu3 HBF4 (10 mol%), PhMe, 1008C,
2 h. [b] Isolated Yield.[31]
Chem. Eur. J. 2012, 18, 16754 – 16764
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16761

M. C. Willis et al.
amounts of product 52 were detected when the bis-allyl ox-
indole was used (Table 5, entry 3). In both successful exam-
ples, the product appeared to exist as a single diastereomer
1
(by H NMR spectroscopy); however, at this stage we were
unable to determine the relative stereochemistry of the new
quaternary centre. Attempts to switch the diastereoselectivi-
ty of the a-arylation step by variation of ligands proved un-
successful, although work is ongoing in this area.
Efficient, chromatography-free starting-material synthesis
and effective performance in our a-arylation chemistry re-
sulted in the selection of allyl/Boc oxindole 45, generating
product 51, for the continuation of the synthesis.
Scheme 21. Formation of final cyclotryptamine ring-system. Reagents and
conditions: 1) Compound 51 (1.0 equiv), TMSOTf (2.2 equiv), CH2Cl2,
RT, 16 h. 2) LiAlH4 (3.7 equiv), THF, reflux, 1.5 h.
Completion of the synthesis: A variety of methods exist for
the cleavage of allyl protecting groups from amines.[32] As a
model system, we chose to study the deprotection of N-allyl
chimonanthine 16. Numerous conditions were evaluated, in-
cluding Rh-,[32b] Ru-[32c] and Pd-catalysed[32d] isomerisations,
although unfortunately none of these gave the desired prod-
uct, possibly as a result of the presence of unprotected
amine functionality. The only conditions found to be effec-
tive for deprotection of our system involved either Na/
NH3
or KOtBu in DMSO at elevated temperatures.[32e]
[1c]
Scheme 22. Completion of the synthesis of hodgkinsine B (3). Reagents
and conditions: Compound 53 (1.0 equiv), Na (60 equiv), NH3(l), THF,
À788C, 2 h; NH4Cl quench.
The highly basic KOtBu/DMSO mixture facilitated isomeri-
sation to the enamine quantitatively in 4 h (Scheme 20). The
Dissolving metal reductions have been used to great
effect for reductive amination on similar substrates.[1c] With
this in mind, we attempted to eliminate the hydride reduc-
tion step from our synthesis. Treatment of compound 51
with TMSOTf exposed the pendant amine, and the crude
product was exposed to Na/NH3. The natural product was
generated in
a modest 32% yield (over two steps)
Scheme 20. Model de-allylation. Reagents and conditions: 1) N-Allylchi-
monanthine 16 (1.0 equiv), KOtBu (6.0 equiv), DMSO, 1008C, 3 h. 2)
HCl(aq) (1m), RT, 20 min then NaHCO3(aq).
(Scheme 23).[34]
enamine failed to hydrolyse under standard conditions;
however, brief exposure to dilute aqueous HCl regenerated
meso-chimonanthine (1) without the need for chromatogra-
phy. There was also no indication of degradation of the acid
sensitive chimonanthine core.[8a]
Armed with two possible de-allylation protocols, we sub-
jected a-arylated product 51 to a Boc-deprotection/reductive
amination sequence. Treatment of 51 with TMSOTf effi-
ciently cleaved the Boc group from the pendant amine and
subsequent intramolecular reductive amination with LiAlH4
gave bis-allyl-hodgkinsine B 53 in a 75% yield (over two
steps) (Scheme 21).
Scheme 23. Tandem global deprotection/reductive amination. Reagents
and conditions: 1) Compound 51 (1.0 equiv), TMSOTf (2.2 equiv),
CH2Cl2, RT, 16 h. 2) Na (100 equiv), NH3(l), THF, À788C, 2 h, NH4Cl
quench.
The stage was set for the final de-allylation, and to our
delight, reaction of bis-allyl-hodgkinsine B 53 with Na/NH3
gave the natural product hodgkinsine B (3) in an impressive
87% yield (Scheme 22). Surprisingly, given the results with
our model system, exposure of the natural product precursor
53 to KOtBu/DMSO resulted in immediate decomposi-
tion.[33]
Conclusion
In summary, we have completed a total synthesis of hodg-
kinsine B comprising just six isolated intermediates in its
longest linear sequence, and, in total, only four chromato-
graphic operations. To the best of our knowledge, chimonan-
16762
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16754 – 16764

Natural Product Synthesis
FULL PAPER
Scheme 24. Overview of total synthesis of hodgkinsine B.
Chem. Eur. J. 2011, 17, 1388; d) M. A. Schmidt, M. Movassaghi, Syn-
lett 2008, 313.
thine represents the most complex N-nucleophile used as a
substrate in a catalytic enantioselective allylic desymmetrisa-
tion; this single desymmetrisation reaction ultimately estab-
lished the configurations of all six stereocentres present in
the natural product. In the course of our investigation, we
also developed a catalytic enantioselective intermolecular
Buchwald–Hartwig amination process. A summary of the
complete synthesis of hodgkinsine B is shown in Scheme 24.
[4] a) E. F. L. J. Anet, G. K. Hughes, E. Ritchie, Aust. J. Chem. 1961, 14,
173; b) K. P. Parry, G. F. Smith, J. Chem. Soc. Perkin Trans. 1 1978,
1671; c) T. Tokuyama, J. W. Daly, Tetrahedron 1983, 39, 41; d) L.
Verotta, T. Pilati, M. Tatꢃ, E. Elisabetsky, T. A. Amador, D. S.
Nunes, J. Nat. Prod. 1998, 61, 392; e) V. Jannic, F. Guꢄritte, O. Lap-
rꢄvote, L. Serani, M. Martin, T. Sꢄvenet, P. Potier, J. Nat. Prod.
1999, 62, 838; f) L. Verotta, F. Peterlongo, E. Elisabetsky, T. A.
Amador, D. S. Nunes, J. Chromatogr. A 1999, 841, 165.
[5] H. A. Saad, S. H. El-Sharkawy, W. T. Shier, Planta Med. 1995, 61,
313.
[6] Y. Adjibadꢄ, H. Saad, B. Kuballa, J. P. Beck, T. Sꢄvenet, P. Cabalion,
R. Anton, J. Ethnopharmacol. 1990, 29, 127.
Experimental Section
[7] a) T. A. Amador, L. Verotta, D. S. Nunes, E. Elisabetsky, Planta
Med. 2000, 66, 770; b) L. Verotta, F. Orsini, M. Sbacchi, M. A.
Scheildler, T. A. Amador, E. Elisabetsky, Bioorg. Med. Chem. 2002,
10, 2133; c) D. D. Porto, A. T. Henriques, A. G. Fett-Neto, Open
Bioact. Compd. J. 2009, 2, 29; d) J. Sanz-Biset, J. Campos-de-la-
For details of the synthesis and chararcterisation of the compounds men-
tioned in this paper, see the Supporting Information.
ˇ
Cruz, M. A. Epiquiꢄn-Rivera, S. Canigueral, J. Ethnopharmacol.
2009, 122, 333.
[8] a) J. T. Link, L. E. Overman, J. Am. Chem. Soc. 1996, 118, 8166;
b) L. E. Overman, D. V. Paone, B. A. Stearns, J. Am. Chem. Soc.
1999, 121, 7702; c) J. J. Kodanko, L. E. Overman, Angew. Chem.
2003, 115, 2632; Angew. Chem. Int. Ed. 2003, 42, 2528; d) J. J. Ko-
danko, S. Hiebert, E. A. Peterson, L. Sung, L. E. Overman, V. M.
Linck, G. C. Goerck, T. A. Amador, M. B. Leal, E. Elisabetsky, J.
Org. Chem. 2007, 72, 7909; e) A. D. Lebsack, J. T. Link, L. E. Over-
man, B. A. Stearns, J. Am. Chem. Soc. 2002, 124, 9008; f) M. Movas-
saghi, M. A. Schmidt, Angew. Chem. 2007, 119, 3799; Angew. Chem.
Int. Ed. 2007, 46, 3725; g) J. Kim, J. A. Ashenhurst, M. Movassaghi,
Science 2009, 324, 238; h) T. Newhouse, P. S. Baran, J. Am. Chem.
Soc. 2008, 130, 10886; i) K. Foo, T. Newhouse, I. Mori, H. Takayama,
P. S. Baran, Angew. Chem. 2011, 123, 2768; Angew. Chem. Int. Ed.
2011, 50, 2716; j) K. M. Depew, S. P. Marsden, D. Zatorska, A. Za-
torski, W. G. Bornmann, S. J. Danishefsky, J. Am. Chem. Soc. 1999,
121, 11953.
Acknowledgements
This work was supported by the EPSRC, AstraZeneca and the University
of Oxford. The NMR, X-ray and mass spectrometry service at the Uni-
versity of Oxford are also thanked for their assistance. Abstract back-
ground image used with permission from Black Diamond Images: http://
blackdiamondimages.zenfolio.com/
[1] a) P. L. Julian, J. Pikl, J. Am. Chem. Soc. 1933, 55, 2105; b) P. L.
Julian, J. Pikl, D. Boggess, J. Am. Chem. Soc. 1934, 56, 1797; c) P. L.
Julian, J. Pikl, J. Am. Chem. Soc. 1935, 57, 539; d) P. L. Julian, J.
Pikl, J. Am. Chem. Soc. 1935, 57, 563; e) P. L. Julian, J. Pikl, J. Am.
Chem. Soc. 1935, 57, 755.
[2] D. Crich, A. Banerjee, Acc. Chem. Res. 2007, 40, 151.
[3] For reviews, see a) G. A. Cordell, J. E. Saxton, in The Alkaloids,
Vol. 20, Elsevier, Amsterdam, 1981, p. 10; b) A. Steven, L. E. Over-
man, Angew. Chem. 2007, 119, 5584; Angew. Chem. Int. Ed. 2007,
46, 5488; c) P. Ruiz-Sanchis, S. A. Savina, F. Albericio, M. ꢀlvarez,
[9] a) M. C. Willis, J. Chem. Soc. Perkin Trans. 1 1999, 1765; b) M. C.
Willis, L. H. W. Powell, C. K. Claverie, S. J. Watson, Angew. Chem.
2004, 116, 1269; Angew. Chem. Int. Ed. 2004, 43, 1249; c) S. J. Byrne,
Chem. Eur. J. 2012, 18, 16754 – 16764
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16763

M. C. Willis et al.
A. J. Fletcher, P. Hebeisen, M. C. Willis, Org. Biomol. Chem. 2010,
8, 758; d) R. H. Snell, R. L. Woodward, M. C. Willis, Angew. Chem.
2011, 123, 9282; Angew. Chem. Int. Ed. 2011, 50, 9116.
recent review of non-enzymatic kinetic resolution see: H. Pellissier,
Adv. Synth. Catal. 2011, 353, 1613.
[21] Reviews of asymmetric allylic alkylation: a) Z. Lu, S. Ma, Angew.
Chem. 2008, 120, 264; Angew. Chem. Int. Ed. 2008, 47, 258; b) B. M.
Trost, V. L. Van Ranken, Chem. Rev. 1996, 96, 395; c) B. M. Trost,
Pure. Appl. Chem. 1996, 68, 779.
[22] For recent reviews on catalytic carbonyl a-arylation, see: a) F. Belli-
na, R. Rossi, Chem. Rev. 2010, 110, 1082; b) C. C. C. Johansson, T. J.
Colacot, Angew. Chem. 2010, 122, 686; Angew. Chem. Int. Ed. 2010,
49, 676.
[10] For reviews relating to aspects of enantioselective desymmetrisation,
see: reference [19a] and a) ꢀ. Enrꢅquez-Garcꢅa, E. P. Kꢆndig, Chem.
Soc. Rev. 2012, 41, 7803; b) C. E. Mꢆller, P. R. Schreiner, Angew.
Chem. 2011, 123, 6136; Angew. Chem. Int. Ed. 2011, 50, 6012; c) E.
Garcꢅa-Urdiales, I. Alfonso, V. Gotor, Chem. Rev. 2011, 111, PR110;
d) M. D. Dꢅaz de Villegas, J. A. Gꢇlvez, P. Etayo, R. Badorrey, P.
Lꢈpez-Ram-de-Vꢅu, Chem. Soc. Rev. 2011, 40, 5564.
[11] a) O. Kitagawa, K. Yosumoto, M. Kohriyama, Y. Dobashi, T. Tagu-
chi, Org. Lett. 2004, 6, 3605; b) O. Kitagawa, S. Matsuo, K. Yosumo-
to, T. Taguchi, J. Org. Chem. 2006, 71, 2524; c) M. Anstiss, J. M. Hol-
land, A. Nelson, J. R. Titchmarsh, Synlett 2003, 1213; d) M. Anstiss,
A. Nelson, Org. Biomol. Chem. 2006, 4, 4135; e) K. Takenaka, N.
Itoh, H. Sasai, Org. Lett. 2009, 11, 1483; f) L. Porosa, R. D. Viirre,
Tetrahedron Lett. 2009, 50, 4170; g) R. S. Ward, Chem. Soc. Rev.
1990, 19, 1; h) F. O. Arp, G. C. Fu, J. Am. Chem. Soc. 2006, 128,
14264; i) K. Rossen, P. J. Pye, A. Maliakal, R. P. Volante, J. Org.
Chem. 1997, 62, 6462.
[12] Examples of achiral N-nucleophiles in desymmetrising palladium-
catalysed asymmetric allylic alkylations, see: a) B. M. Trost, A. Apo-
nick, J. Am. Chem. Soc. 2006, 128, 3931; b) B. M. Trost, D. E. Patter-
son, Chem. Eur. J. 1999, 5, 3279; c) B. M. Trost, D. E. Patterson, J.
Org. Chem. 1998, 63, 1339; d) G. Muchow, J. M. Brunel, M. Maffet,
O. Pardigon, G. Buono, Tetrahedron 1998, 54, 10435; e) B. M. Trost,
Z. Shi, J. Am. Chem. Soc. 1996, 118, 3037.
[23] a) M. J. Durbin, M. C. Willis, Org. Lett. 2008, 10, 1413; b) R. A.
Altman, A. M. Hyde, X. Huang, S. L. Buchwald, J. Am. Chem. Soc.
2008, 130, 9613; c) A. M. Taylor, R. A. Altman, S. L. Buchwald, J.
Am. Chem. Soc. 2009, 131, 9900; d) C. Mai, M. F. Sammons, T. Sam-
makia, Org. Lett. 2010, 12, 2306.
[24] Synthesised by a procedure based on that reported by Y. Yamada, S.
Arima, C. Okada, A. Akiba, Y. Kai, Y. Harigaya, Chem. Pharm.
Bull. 2006, 54, 788.
[25] V. Snieckus, Chem. Rev. 1990, 90, 879.
[26] This highly effective strategy was also used by Overman and co-
workers in their syntheses of polypyrrolidinoindoline alkaloids. See
reference [8b].
[27] For selected examples of a-alkylation of oxindoles see: a) D. L.
Hughes, U. H. Dolling, K. M. Ryan, E. F. Schoenewaldt, E. J. J. Gra-
bowski, J. Org. Chem. 1987, 52, 4745; b) T. B. K. Lee, G. S. K. Wong,
J. Org. Chem. 1991, 56, 872; c) B. M. Trost, M. U. Frederiksen,
Angew. Chem. 2005, 117, 312; Angew. Chem. Int. Ed. 2005, 44, 308.
[28] K. Szabꢈ-Pusztay, L. Szabꢈ, Synthesis 1979, 276.
[29] J. R. Fuchs, R. L. Funk, Org. Lett. 2005, 7, 677.
[13] a) H. Ishikawa, H. Takayama, N. Aimi, Tetrahedron Lett. 2002, 43,
5637; b) H. Takayama, Y. Matsuda, K. Masubuchi, A. Ishida, M. Ki-
tajima, N. Aimi, Tetrahedron 2004, 60, 893; c) Y. Matsuda, M. Kitaji-
ma, H. Takayama, Heterocycles 2005, 65, 1031.
[30] Due to difficulty seperating the mono- and bis-products, 42 and 43,
an accurate yield was not determined.
[14] For the first enantioselective Buchwald-Hartwig cross coupling see
reference [11h]. Subsequently the groups of Sasai and Viirre dis-
closed that they were able to perform an intramolecular desymmetr-
isation of malonamides via an asymmetric Buchwald–Hartwig cross
coupling with eeꢁs up to 70% see references [11e–f].
[15] a) A. Jarrahpour, M. Zarei, Synlett 2008, 381; b) P. Jacob, P. S. Call-
ery, A. T. Shulgin, N. Castagnoli, J. Org. Chem. 1976, 41, 3627.
[16] Q. Shen, J. F. Hartwig, Org. Lett. 2008, 10, 4109.
[31] The mass balance appeared to be made up by de-halogenated start-
ing material.
[32] a) S. Escoubet, S. Gastaldi, M. Bertand, Eur. J. Org. Chem. 2005,
3855; b) J. K. Stille, Y. Becker, J. Org. Chem. 1980, 45, 2139; c) B.
Alcaide, P. Almendros, J. M. Alonso, M. F. Aly, Org. Lett. 2001, 3,
3781; d) F. Garro-Helion, A. Merzouk, F. Guibꢄ, J. Org. Chem.
1993, 58, 6109; e) J. Hine, S. M. Linden, A. Wang, V. Thiagarajan, J.
Org. Chem. 1980, 45, 2821.
1
[17] Experiment carried out using HPLC, isolated yield not measured.
[18] For the application of amine bases in the Trost allylation see refer-
ences [12b–d].
[33] Observed in situ by H NMR spectroscopy.
[34] It was noted in a review by Overman that yields for this type of
transformation could be increased by the presence of tBuOH (see
ref. [3b]); however, when applied to our system there was no corre-
sponding improvement.
[19] For our use of a Trost allylation protocol in the kinetic resolution of
(Æ)-chimonanthine see Supporting Information.
[20] a) During our investigations a report by Hou and co-workers was
published on the kinetic resolution of indolines by means of a Trost
allylation: X. L. Hou, B. J. Zheng, Org. Lett. 2009, 11, 1789; b) for a
Received: September 5, 2012
Published online: November 30, 2012
16764
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16754 – 16764