Stereoselective synthesis of chiral polycyclic indolic architectures through Pd0-catalyzed tandem deprotection/cyclization of tetrahydro-β-carbolines on allenes

Article abstract of DOI:10.1002/chem.201500273

Enantioenriched N-allyl tetrahydro-β-carbolines were prepared by chiral phosphoric acid-catalyzed Pictet-Spengler reactions. The compounds undergo Pd⁰-catalyzed cyclizations through a tandem deprotection/cyclization process. The regioselectivit

Full text of DOI:10.1002/chem.201500273

DOI: 10.1002/chem.201500273
Full Paper
&
Alkaloids
Stereoselective Synthesis of Chiral Polycyclic Indolic Architectures
through Pd0-Catalyzed Tandem Deprotection/Cyclization of
Tetrahydro-b-carbolines on Allenes
Valꢀrian Gobꢀ and Xavier Guinchard*[a]
Abstract: Enantioenriched N-allyl tetrahydro-b-carbolines
were prepared by chiral phosphoric acid-catalyzed Pictet–
Spengler reactions. The compounds undergo Pd0-catalyzed
cyclizations through a tandem deprotection/cyclization pro-
cess. The regioselectivity of the attack is controlled by the
chain length and by the substitution pattern of the allene
function. Products resulting from 5-exo- or 6-exo-attack were
obtained with diastereoisomeric ratio up to 95:5. Azepino-
pyrrido[3,4-b]indoles were obtained by 7-endo-cyclizations.
Introduction
Indole alkaloids display structural richness and a broad range
of biological activities, rendering them attractive targets for
both synthetic and biological purposes.[1] Some typical pat-
terns frequently encountered in such structures define privi-
leged scaffolds that are likely to provide success in the discov-
ery of novel bioactive compounds.[2] Chemists’ creativity has
led to many strategies for indole synthesis and functionaliza-
tion[2–3] but there is still an avenue for the discovery of novel
scaffolds. The indolo[2,3-a]quinolizidine moiety is found in
hundreds of natural products and holds a good position in
chemist’s efforts towards the development of bioactive com-
pounds.[4] The vast majority of natural products featuring the
indolo[2,3-a]quinolizidine moiety presents a cycle D substitut-
ed on positions 1, 2, and 3, generally fused with an additional
cycle.[5] In contrast, their 4-substituted congeners are encoun-
tered less frequently. (ꢀ)-Tangutorine,[6] (ꢀ)-normalindine,[7]
(ꢀ)-cadamine,[8] (ꢀ)-manadomanzamines,[9] or (ꢀ)-phoebe-
grandine E[10] are representative natural products of this family
(Figure 1). Few total syntheses of such natural products have
been reported, with the exception of normalindine,[11] demon-
strating the challenge of enantio- and diastereoselective syn-
thesis of such scaffolds.
Figure 1. Natural products featuring an indolo[2,3-a]quinolizidine skeleton
with substituent at the 4-position.
gler reactions by using either chiral thioureas[13] or chiral phos-
phoric acids.[14] Thioureas catalyze Pictet–Spengler reactions
with concomitant N-acylations[13a] and necessitate the use of
highly electron-rich tryptamines if N-acylation is to be avoi-
ded.[13b] This drawback was addressed recently by Seidel,
through the use of a conjugate-base-stabilized thiourea, lead-
ing to tetrahydro-b-carbolines in excellent enantiomeric excess
(ee). This method is mostly limited to aromatic aldehydes.[15]
On the other hand, chiral phosphoric acids, when applied to
gem-diester tryptamines[14a] or N-protected tryptamines[14b–d]
afford good conversion and ee values. In most cases, the trypt-
amine protecting group must be removed prior to further
functionalization, unless it has been conveniently designed to
be involved in further steps.[2,16] These methodologies have af-
forded elegant syntheses of enantioenriched polycyclic archi-
tectures.[17]
The synthesis of indolo[2,3-a]quinolizidine-containing com-
pounds often relies on a diastereoselective Pictet–Spengler re-
action.[12] Over the past few years, Jacobsen, List, Hiemstra, and
Wang have made several breakthroughs in this domain by de-
veloping Brønsted acid catalyzed enantioselective Pictet–Spen-
[a] V. Gobꢀ, Dr. X. Guinchard
Centre de Recherche de Gif
Institut de Chimie des Substances Naturelles, CNRS
1 Avenue de la Terrasse
91198 Gif-sur-Yvette (France)
E-mail: xavier.guinchard@cnrs.fr
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500273.
Chem. Eur. J. 2015, 21, 1 – 11
1
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 1. Synthesis of allenols 6a–l.
Entry
Conditions
X
R2
Yield [%][a]
1
2
3
4
5
6
7
8
9
i
i
i
i
CH2CH2
CH2
CH2CH2CH2
CH2O
CH2N(Boc)
CH2
CH2CH2
CH2CH2
CH2CH2
H
H
H
H
H
Ph
Ph
6a, 68
6b, 51
6c, 70
6d, 50
6e, 55
6 f, 81
6g/7g, 75 (79:21)
6h/7h, 84 (80:20)
6i–i’, 0
i
ii
ii
ii
ii
p-FC6H4
m-BrC6H4
p-ClC6H4
p-MeOC6H4
2-pyridine
Ph
Scheme 1. Synthetic strategy.
10
11
12
ii
ii
ii
CH2CH2
CH2CH2
CH2O
6j, 57
6k, 0
6l, 0
We recently described the synthesis of enantioenriched 4-
vinyl substituted indolo[2,3a]quinolizidines 1a (>93% ee)
through a direct palladium-catalyzed tandem deprotection/
cyclization from N-allyl tetrahydro-b-carbolines 2 (Scheme 1).[18]
In this work, the N-allyl protecting group of 3 was chosen for
its compatibility with the asymmetric Pictet–Spengler reaction
catalyzed by chiral phosphoric acids and in the perspective of
a one-pot deprotection/cyclization reaction sequence. In this
paper, we wish to report on the chemistry developed by using
these palladium-catalyzed cyclizations. In particular, we de-
scribe the synthesis of highly functionalized chiral heterocycles
bearing quaternary stereogenic centers 1a and heterocycles
1b, resulting from a novel, regioselective mode of cyclization.
The chemistry was extended to 1,3-disubstituted allenes, re-
sulting in the formation of vinyl substituted derivatives 1a (R2
¼ H). The results of mechanistic studies are presented that
provide information on the reaction mechanism of this
tandem cyclization process.
[a] Isolated yields. DIPA=diisopropylamine.
tion of 4-pentyn-1-ol or 5-hexyn-1-ol furnished allenes 6 f and
6g in 81 and 75% yields, respectively (Table 1, entries 6–7).
Aryl groups bearing halides were used and we found that, sur-
prisingly, whereas para-fluorobenzyl chloride led to the allene
in 84% yield, the m-Br and p-Cl analogues were inefficient in
this reaction (Table 1, entries 8–9). The electron-rich 4-methoxy-
benzyl chloride afforded allene 6j in 57% yield (Table 1,
entry 10). Notably, the use of 2-(chloromethyl)pyridine and 2-
(prop-2-yn-1-yloxy)ethanol failed to provide the corresponding
allene (Table 1, entries 11–12). As reported,[22] allenes 6 f–h are
often obtained as a mixture with their alkyne precursors 7 f–h.
In most cases, the allenes can be separated from the alkynes
by column chromatography. Interestingly, the allene 6j did not
contain the corresponding alkyne 7j, although the yield was
moderate. Both methods reported in Table 1 are compatible
with a free hydroxyl group, thus saving protection/deprotec-
tion steps.
Results and Discussion
Synthesis of Allenaldehydes 4
We initiated this study with the synthesis of a range of allenal-
dehydes 4. Numerous methods are available for the construc-
tion of the allene function,[19] among which we selected those
using an alkyne as starting material, because of the ready avail-
ability of 1,n-alkynols (see the Supporting Information). Crabbe
homologation[20] was used to convert 1,5-, 1,6-, or 1,7-alkynols
into terminal allenes 6, by using the convenient protocol de-
veloped by Ma[21] (Table 1, entries 1–5). Allenols 6a–e were ob-
tained in moderate yields that are typical for this type of classi-
cal chemical transformation (50–71%).
The Heck alkynylation developed by Buchwald[22,23] has been
applied here to prepare 7-arylhepta-5,6-dienals in good yields
by reaction of alkynes with benzyl chloride and [Pd(SPhos)2]
complex under basic conditions. The reaction conditions were
applied to a range of alkynes (Table 1, entries 6–12). The reac-
gem-Disubstituted allenes 6m–v were then prepared in two
steps. Addition of cuprate reagents to propargylic mesylates
8a–b afforded silylethers 9m–v,[24] which, upon deprotection
with fluoride ions,[25] secured the formation of allenols 6m–v.
Numerous Grignard reagents could be used successfully with
both protected 4-pentynol 8a and 5-hexynol 8b, furnishing
1,5- and 1,6-allenols (Table 2). We found that both aliphatic
and aromatic groups could be introduced successfully through
the corresponding Grignard reagents, including electron-rich
(Table 2, entry 9) and halogenated aryl groups (Table 2,
entry 10).
Allenols 6a–v were then submitted to Swern oxidation,[26]
which furnished the corresponding allenals 4a–v in mostly
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
2
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Asymmetric Pictet–Spengler Reactions
Table 2. Synthesis of gem-disubstituted allenols 6m–v.
The asymmetric Pictet–Spengler reaction was used as the key
step for the asymmetric construction of 1. Tryptamines previ-
ously reported in such process were protected either by sulfe-
nyl,[14b] benzyl,[14c] or 2-naphthylmethyl (NAP)[14d] groups. The
key point for our strategy was accordingly the compatibility of
the N-allyl protecting group[28] and the 1,n-allenals 4 with
phosphoric acid-catalyzed asymmetric Pictet–Spengler reac-
tions. In our initial studies,[18] we demonstrated the ability of
the allyl group to act as a protecting group for tryptamine in
Pictet–Spengler reactions and the compatibility of the allene
function with the reaction conditions. In particular, we found
that the spinol-derived[14d] chiral phosphoric acid 15 was the
best catalyst for the asymmetric Pictet–Spengler reaction of N-
allyl tryptamines 3 with allenals 4. Asymmetric catalytic Pictet–
Spengler reactions were accordingly performed by reaction of
N-allyl tryptamines 3a–c with the extended set of aldehydes 4
(3 equiv)[29] in the presence of the Spinol-derived catalyst 10
(2 mol%; Scheme 2). The corresponding tetrahydro-b-carbo-
lines 2a–s were obtained in essentially good yields and excel-
lent enantioselectivities (87–97% ee).[30] Overall, we found that
the length and the functionalization of the side chain have
little influence on the enantiomeric excesses. In addition, nei-
ther the substituent R3 present on the tryptamine moiety nor
Entry
n
R1
Yield of 9 [%][a]
Yield of 6 [%]
1
2
3
4
5
6
7
8
9
4
3
4
3
4
3
4
3
3
3
Me
Me
iPr
iPr
Ph
Ph
allyl
tBu
9m, 96
9n, 88
9o, 86
9p, 87
9q, 83
9r, 69
9s, 82
9t, 92
9u, 100
9v, 100
6m, 94
6n, 93
6o, 90
6p, 96
6q, 99
6r, 90
6s, 72
6t, 100
6u, 87
6v, 80
p-MeOC6H4
p-ClC6H4
10
Table 3. Swern oxidation of allenols 6 to allenaldehydes 4.
Entry
X
R1
R2
Yield of 4 [%][a]
1
2
3
4
5
6
7
8
CH2CH2
CH2
CH2CH2CH2
CH2O
CH2N(Boc)
CH2
CH2CH2
CH2CH2
CH2CH2
CH2CH2
CH2
CH2CH2
CH2
CH2CH2
CH2
H
H
H
H
H
H
H
H
H
Me
Me
iPr
iPr
Ph
Ph
allyl
H
H
H
H
4a, 90
4b, 83
4c, 89
4d, 61
4e, 88
4 f, 52
4g, 94
4h, 79
4j, 48
4m, 61
4n, 80
4o, 60
4p, 84
4q, 63
4r, 67
4s, 68
4t, 78
4u, 65
4v, 86
H
Ph
Ph
p-FC6H4
9
p-MeOC6H4
10
11
12
13
14
15
16
17
18
19
H
H
H
H
H
H
H
H
H
H
CH2CH2
CH2
CH2
tBu
p-OMeC6H4
p-ClC6H4
CH2
20
H
H
4w, 19
[a] Isolated yield.
good yields (Table 3). In some cases, the volatility of the alde-
hyde reduced the isolated yield. However, if precautions were
accordingly taken with the solvent evaporation process, good
yield could be maintained. Surprisingly, allenol 6w[27] furnished
the corresponding aldehyde 4w in low yield (19%; Table 3,
entry 20). In most cases, allenals 4 could be stored at ꢀ208C
for a few weeks. This set of aldehydes 4 constitutes a library
with substantial molecular diversity, including chain length,
substitution pattern of the allene function, and heteroatom
functionalization that will allow for the synthesis of novel het-
erocycles.
Scheme 2. Scope of the asymmetric Pictet–Spengler reaction.
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
3
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
the R1 group of the allene affected the enantioselectivity or
the yield of these reactions significantly. Notably, previously
unknown compounds bearing various R1 groups were ob-
tained in good yields and enantiomeric excesses. The only ex-
ception was compound 2s, for which the p-ClPh group low-
ered both the yield and the ee to 54 and 87%, respectively. In
addition, compound 2r, bearing an electron-rich aryl group
was not obtained, with the reaction leading to degradation
products. Similarly, aldehyde 4w did not afford compound 2t.
Pd0-Catalyzed Cyclizations
After establishing the synthesis of enantioenriched tetrahydro-
b-carbolines 2, their palladium-catalyzed cyclizations were then
investigated. In our initial studies,[18] we found that a catalytic
system composed of 5 mol% tetrakis(triphenylphosphine) pal-
ladium and dimethylbarbituric acid 11 as the allyl scavenger[31]
furnished, at a concentration of 0.027m, compound 12a in
88% yield as a 84:16 separable mixture of cis and trans diaste-
reoisomers (Scheme 3, Eq. 1). Additional extensive studies
Scheme 4. 6-exo-Cyclizations to indoloquinolizidines 12.
rings 12a–i through 6-exo-cyclization on the allene (Scheme 4).
Diastereoisomeric pairs of compounds were separated on silica
gel[32] and characterized independently by NMR techniques.
Indolo[2,3-a]quinolizidines substituted with a methoxy or fluo-
ride group, led to heterocycles 12b and 12c in good yields
and diastereoselectivities, supporting the conclusion that the
nature of the substituents on the indole has little influence.
Substrates displaying a heteroatom in the chain led to the cor-
responding vinyl-morpholine 12d and piperazine 12e in 37
and 78% yields, respectively.
We then focused on the cyclization of gem-disubstituted al-
lenes, potentially leading to piperidines with stereocontrolled
quaternary centers in the 4-position (Scheme 4). The reaction
of tetrahydro-b-carbolines 2j–m furnished the corresponding
cyclized compound 12 f–i through a 6-exo-cyclization pathway.
We noticed that the diastereoselectivity ratio increased with
the steric bulk of the R1 group. Whereas the less bulky allyl
group led to a moderate diastereoisomeric ratio (d.r.=62:38),
the diastereoselectivity increased to 79:21 when R1 =Me and
to 93:7 when R1 =iPr. The use of a substrate with a phenyl
group substituent led to 12i with complete diastereocontrol in
64% yield. Notably, in all cases, the main diastereoisomer pres-
ents a cis relationship between H12b and the smaller substitu-
ent of C4.
Scheme 3. Selective formation of tetracycle 12a or amine 15a.
showed that, depending on the reaction conditions, the forma-
tion of compound 12a was accompanied by substantial
amounts of olefin 13a, which resulted from intermolecular ad-
dition of monoallyl-DMBA 14 to the allene function (see pro-
posed mechanism). In addition, we found that the intermedi-
ate deprotected amine 15a could be obtained cleanly and se-
lectively by reducing the amount of catalyst and by perform-
ing the reaction in dimethyl sulfoxide (DMSO) within one hour
(Scheme 3, Eq. 2). The method provides efficient access to this
amine that will be further used for mechanistic studies.
The stereochemistry of the cis/trans diastereoisomers was as-
signed based on 2D NMR techniques (Figure 2). Compound
12a presents all relevant NOE correlations between H-4, H-6b,
H-12b, and H-2 characteristic for 1,3-diaxial interactions and
a NOE correlation between H-6a and the vinylic proton. Simi-
larly, the major diastereoisomer of 12h (R1 =iPr) shows correla-
tions between H-12b, H-6b, and the vinylic proton, and a corre-
lation between H-6a and the iPr group.
6-exo-Cyclizations
The enantioenriched tetrahydro-b-carbolines 2 were then sub-
mitted to the conditions previously established for the palladi-
um-catalyzed tandem allyl deprotection/cyclization process.
Compounds presenting a 1,6-relationship between the amine
and the allene function gave the corresponding six-membered
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
4
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
5-exo- and 7-endo-Cyclizations
spectively (Table 4, entry 4). When the concentration was in-
creased (0.1m), a mixture was again obtained (Table 4, entry 5),
whereas high dilution (0.01m) afforded the indolizidine 16 f
chemoselectively in 62% yield, as a sole diastereoisomer
(Table 4, entry 6). Similarly, by using phenyl-substituted allene
2q, the cyclization afforded a 25:75 mixture of 16g and 17g,
respectively, when the reaction was performed at 0.01m
(Table 4, entry 7), whereas concentrated conditions (0.1m) af-
forded the 7-endo-cyclization product 17g regioselectively in
76% yield (Table 4, entry 8). The latter results indicate that the
cyclization pathway is not only governed by the steric bulk of
The series of compounds 2d–f and
2n–s, possessing a 1,5-relationship
between the amine and the allene,
were successfully submitted to cyc-
lizations. Interestingly, the reaction
proceeded through either a 5-exo-
or 7-endo-cyclization pathway, de-
pending on the substitution pat-
tern of the allene and on the con-
Figure 2. Relevant NOESY cor-
relations for compounds 12a
and 12h.
Table 4. Regioselective 5-exo- or 7-endo-cyclizations to heterocycles 16 or 17.
centration of the reaction mix-
ture (Table 4). 5-exo-Cyclizations
furnished indolizino[8,7-b]indoles
16, whereas 7-endo-cyclizations
led
to
azepino[1’,2’:1,2]pyri-
do[3,4-b]indoles 17, both scaf-
folds being encountered in natu-
ral products with biological ac-
tivities.[33] Diastereoisomeric mix-
tures of 16 were separated on
silica gel and the corresponding
cis/trans derivatives were charac-
terized by 2D NMR techniques.
Allenes 2d–f underwent selec-
Entry
2
c [M] Product 16[a]
or 17
Ratio 16/
d.r.
Yield
[%][c]
17[b]
(16)[b]
2d
(R3 =H)
2e
75:25 80
79:21 67
73:27 79
1
0.01
100:0
(R3 =OMe)
2 f (R3 =F)
2
3
2n
2o
0.027
0.027
100:0
100:0
55:45 67
9:91 64
tive
5-exo-cyclizations
to
indolo[2,3-a]pyrrolizidines 16a–
c in good yields with diastereo-
isomeric ratios of approximately
75:25 (Table 4, entry 1). In con-
trast, with the indolo[2,3-a]qui-
nolizidines 12 series, the major
diastereoisomer obtained pres-
4
2p
0.027
50:50
>95:5 61
ents
a
trans-relationship be-
5
6
2p
2p
0.1
0.01
68:32
100:0
>95:5 81
>95:5 62
tween H-3 and H-11b. In these
cases, we found that dilution of
the reaction mixture to 0.01m
was beneficial to both the yield
and the diastereoselectivity. The
introduction of an R1 substituent
on the allene afforded five-mem-
bered rings 16d–e with quater-
nary centers in moderate to ex-
cellent diastereoselectivities in
the case of R1 =Me and R1 =iPr,
respectively (Table 4, entries 2–
3). The increase in the steric bulk
of the R1 group from iPr to tBu
promoted a different cyclization
pattern. With a tBu group, we
obtained a 1:1 mixture of com-
pounds 16 f and 17 f in 61%
yield, resulting from 5-exo-and 7-
endo-cyclization pathways, re-
7
8
2q
2q
0.01
0.1
25:75
0:100
>77:23 87
–
–
76
63
9
2s
0.1
0:100
[a] The major diastereomer is shown. [b] Ratio measured by 1H NMR spectroscopic analysis. [c] Yield is given for
pure, isolated compounds 16 or 17, or for both when mixtures were obtained.
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
5
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
the R2 substituent but also by its electronic nature. Replace-
ment of the phenyl substituent by a 4-chlorophenyl group af-
forded the 7-endo-cyclized product 17h in 63% yield with full
regioselectivity (Table 4, entry 9).
Table 5. Optimization for the one-pot deallylation/cyclization
Case of 1,3-Disubstituted Allenes
The Pictet–Spengler reactions reported above and subsequent
cyclizations to polycyclic compounds were accomplished at
first with monosubstituted and gem-disubstituted allenes. We
then investigated the case of 1,3-disubstituted allenes 4g–h.
The reaction of tryptamine 3a with aldehyde (+/ꢀ)-4g in the
presence of diphenyl phosphate furnished the corresponding
tetrahydro-b-carboline 2t in 70% yield, as a 1:1 mixture of two
diastereoisomers 2ta and 2tb (Scheme 5). Purification on silica
Entry
2
T [8C] 18a–d ratio Global cis/trans ratio Global
yield [%][d]
1
2
3
2ta–b[a] RT
2ta
2tb
33:40:27:0[b] 73:27
10:62:28:0[b] 72:28
68
69
50
RT
RT
66:7:27:0[c]
73:27
[a] 1:1 mixture. [b] Ratios were measured on the crude mixture by
1H NMR spectroscopic analysis. [c] Ratio measured on purified products.
[d] Isolated yield.
tween H-1 and H-2 were found to be characteristic for the Z
geometry (JH1,H2 =11.9 Hz) and E geometry (JH1,H2 =15.9 Hz) of
the double bond, allowing for the determination of the double
bond configurations in each product.
Asymmetric Pictet–Spengler reactions were then performed
between tryptamine 3a and aldehydes (+/ꢀ)-4g and (+/ꢀ)-
4h, in the presence of the chiral catalyst 10, affording the cor-
responding compounds in good yields (Scheme 6) as mixtures
of diastereoisomers. The enantiomeric excess determinations
were measured in a subsequent step.
Scheme 5. Pictet–Spengler reaction with racemic aldehyde 4g.
gel afforded both diastereoisomers separately, which were
characterized independently.
The enantioenriched compounds 2ta–b and 2ua–b were
then submitted to palladium-catalyzed cyclization, furnishing
mixtures of diastereoisomers 18a–c and 20a–c (Scheme 7,
step 1). In an effort to avoid the troublesome purification of
the diastereoisomeric mixtures, it was envisaged to submit the
crude cyclization mixtures to a reduction step of the double
bonds to simplify the purification of the compounds. Accord-
We then turned our attention to the cyclization of 2t under
the conditions established previously. The palladium-catalyzed
deprotection/cyclization was performed on either the mixture
of diastereoisomers 2ta and 2tb, or on each diastereoisomer,
independently. All the cyclizations furnished the desired piperi-
dine 18; the case is, however, complicated by the presence of
three out of four possible diastereoisomers 18a–d, due to the
geometry of the newly-formed double-bond. Indeed, when the
reaction was performed with a 1:1 mixture of 2ta/2tb, a separa-
ble mixture of three diastereoisomers 18a–c was obtained in
68% global yield (Table 5, entry 1) after a few hours at room
temperature. When the reaction was performed independently
with both diastereoisomers 2ta and 2tb, the ratio changed
dramatically. Whereas the diastereoisomer 2ta furnished 18b
as the main product (Table 5, entry 2), the cyclization of 2tb led
to 18a as the main compound, albeit with a lower global yield
(Table 5, entry 3). It should, however, be noted that despite the
complexity of the reaction mixtures, the global cis/trans ratio
([18a+18b]/18c ratio) remained very similar in all cases (ca.
73:27), irrespective of the geometry of the double bond. In ad-
dition, the fourth possible diastereoisomer 18d was never ob-
tained. It is consequently clear that the initial geometry of the
allene affects the stereochemical outcome of the reaction, al-
though it was previously reported that the chiral information of
the allene is lost in such a palladium-catalyzed process.[34]
The three diastereoisomers 18a–c were fully characterized
by 1D and 2D NMR methods to determine the cis/trans rela-
tionship between both stereogenic centers and the E/Z geom-
etry of the double bond. Measured coupling constants be-
Scheme 6. Enantioselective synthesis of tetrahydro-b-carbolines 2t–u.
Scheme 7. Cyclization/reduction strategy to compounds 19 and 21.
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
6
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
ingly, the crude mixtures resulting from the palladium-cata-
lyzed reactions were submitted to NaBH4, in the presence of
a catalytic amount of RuCl3 (Scheme 7, step 2).[35] We found
that a short purification to remove the DMBA derivatives was
necessary prior to the reduction step. The reductions were
then performed only on the cis diastereoisomers 18a,b and
20a,b.[36] This approach led rapidly to 19 and 21 in 42 and
46% yields, respectively (two steps). The enantiomeric excesses
of 19 and 21 were then found to be 93%, demonstrating the
compatibility of 1,3-disubstituted allene functions with asym-
metric Pictet–Spengler conditions.
Mechanistic Studies
Experiments were performed to gather more information on
the mechanism and to better understand the role of each
component in the reaction (Scheme 8). We performed in par-
ticular a set of experiments to understand and demonstrate
the pivotal role of DMBA in the reaction. The deprotected
amine 15a was reacted with a catalytic amount of Pd(PPh3)4,
leading to no reaction (Scheme 8, Eq. 1). In contrast, when the
same reaction was performed in the presence of three equiva-
lents of DMBA 11, the product 12a was obtained in 75% yield
in a 83:17 d.r., which is similar to that obtained from allyl-pro-
tected compound 2a (Scheme 8, Eq. 2). These results demon-
strate clearly the involvement of DMBA 11 in the cyclization
step. DMBA also plays a critical role in scavenging the allyl
cation prior to cyclization. We have observed that diallyl-DMBA
is the only DMBA-derivative formed during the reaction, show-
ing that 0.5 equivalent is enough to bring deallylation to com-
pletion. Indeed, the tandem deprotection/cyclization process
performed with 0.55 equivalent furnished the desired com-
pound 12a in 64% yield and in good diastereoselectivity
(Scheme 8, Eq. 3). Interestingly, when deuterated DMBA was
used in the presence of palladium catalyst from 2a or 15a, the
cyclized products 12a-D were obtained in excellent yields,
with 40% incorporation of deuterium on the vinyl proton
(Scheme 8, Eq. 4). This rate of incorporation is lowered by
scrambling phenomena between exchangeable protons of 2a
and 15a and the reactant.[37] All these experimental results ac-
count for the critical role of DMBA for both the deallylation, as
trapping agent, and the cyclization.
The tandem deprotection/cyclization process may conse-
quently be the result of two combined catalytic cycles A and B
Scheme 8. Mechanistic investigations.
Scheme 9. Postulated mechanism for the tandem deprotection/cyclization process.
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
7
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
in which DMBA 11 is involved. Cycle A is a classical Pd-cata-
lyzed allyl group removal,[38] and cycle B is the Pd-catalyzed ad-
dition of the intermediate deprotected amine to the allene
(Scheme 9).[39] The addition of pronucleophiles to allenes,[34,40]
including amines,[34,40c–e,g] catalyzed by Pd0 is a known atom-
economic[41] strategy for allylic functionalization from allenes. It
has been shown, in particular by Trost,[40l] to proceed via a hy-
dridopalladium species, the formation of which has been
shown to be facilitated by the adjunction of an acid cocata-
lyst.[42]
rapid and selective elaboration of complex, enantioenriched,
polycyclic indolic compounds.
Experimental Section
Detailed descriptions of experimental procedures, spectral data
and 1H and 13C NMR spectra of all new compounds are given in
the Supporting Information.
General procedure for the Swern oxidation: Oxalyl chloride
(1.5 equiv) was dissolved in CH2Cl2 in a flask under an argon atmos-
phere. The reaction mixture was then cooled at ꢀ788C and DMSO
(3 equiv) was added dropwise. Stirring was continued at ꢀ788C for
30 min, followed by dropwise addition of a solution of alcohol 6
(1 equiv) dissolved in CH2Cl2. The reaction mixture was further
stirred for 1 h and Et3N (6 equiv) was added. The reaction mixture
was then allowed to warm to RT. Water was added, the organic
layer was separated, and the aqueous phase was extracted twice
with CH2Cl2. The combined organic layers were dried with MgSO4,
filtered, and concentrated under vacuum. The crude mixture was
then purified by flash chromatography to give the desired product
4.
On the basis of our results and on reported precedents, the
mechanism shown in Scheme 9 can be suggested. Starting ma-
terial 2 enters in the first catalytic cycle and reacts with Pd0 to
generate a p-allyl species i, which, in turn, is trapped by DMBA
11, thus liberating the intermediate tetrahydro-b-carboline 15
and the monoallyl DMBA 14.[43] The reaction of Pd0 with the
proton donor DMBA 11 generates PdII-H iii,[40h] which adds to
the allene through hydropalladation, resulting in the formation
of p-allyl complex iv. Subsequent intramolecular allylic substi-
tution by the nucleophilic amine through either 6-exo- or 7-
endo-attack leads to the products 12, 16, or 17, depending on
the chain length. Intramolecular trapping of the p-allyl com-
plex also ensures regeneration of DMBA 11 and Pd0.
Competing intermolecular allylic substitution of monoallyl-
DMBA 14 on intermediate iv explains the formation of com-
pound 13. It becomes clear that this intermolecular process is
strongly favored under concentrated conditions, whereas
dilute conditions favor the intramolecular process. In addition,
when the cyclization is not favored, the p-allyl complex iv can
only evolve toward the DMBA addition product 13.
Typical procedure for 5-phenylhepta-5,6-dienal (4q): Prepared
according to the general procedure from 6q (550 mg,
2.921 mmol), oxalyl chloride (0.38 mL, 4.382 mmol), DMSO
(0.62 mL, 8.763 mmol) and Et3N (2.44 mL, 17.256 mmol) in CH2Cl2
(17 mL). Compound 4q was obtained after column chromatogra-
phy on silica gel (MTBE/petroleum ether, 5:95). Yield: 340 mg
(1.825 mmol, 63%); colorless oil; Rf =0.27 (MTBE/petroleum ether,
5:95); IR (neat): n˜ =2951, 2721, 1940, 1723, 1494, 1452, 855, 764,
1
696 cmꢀ1; H NMR (300 MHz, CDCl3): d=9.77 (t, J=1.6 Hz, 1H; H-1),
7.39–7.35 (m, 2H; 2ꢂArH), 7.33–7.28 (m, 2H; 2ꢂArH), 7.22–7.16
(m, 1H; ArH), 5.08 (t, J=3.3 Hz, 2H; 2ꢂH-7), 2.52 (td, J=7.3,
1.7 Hz, 2H; 2ꢂH-8), 2.49–2.43 (m, 2H; 2ꢂH-4), 1.90 ppm (q, J=
7.3 Hz, 2H; 2ꢂH-3); 13C NMR (75 MHz, CDCl3): d=208.7 (Cq, C6),
202.5 (CH, C1), 136.1 (Cq, CAr), 128.7 (CH, 2ꢂCAr), 127.0 (CH, CAr),
126.1 (CH, 2ꢂCAr), 104.4 (Cq, C5), 78.9 (CH2, C7), 43.6 (CH2, C2), 29.0
(CH2, C4), 20.5 ppm (CH2, C3).
The success of the reaction is consequently the result of
a very fine balance between inter- and intramolecular reactions
and exo- or endo-cyclization pathways.
General procedure for the enantioselective Pictet–Spengler re-
action: A mixture of Nb-a-allyl tryptamine 3 (1 equiv), catalyst 10
(0.02 equiv), and 4 ꢃ molecular sieves (0.23 g for 0.35 mmol of 3,
powdered) in toluene (1.5 mL for 0.1 mmol of 3) was stirred for
5 min at RT under an argon atmosphere. Aldehyde 4 (3 equiv) was
added and the mixture was stirred at 308C for 16 h. Upon comple-
tion of the reaction (TLC monitoring), the reaction mixture was fil-
tered over silica. The filtrate was concentrated under vacuum and
the residue was purified by flash chromatography to give the de-
sired product 2.
Conclusion
Numerous 1,n-allenaldehydes 4, possessing different substitu-
tion patterns, were successfully engaged in Brønsted acid-cata-
lyzed Pictet–Spengler reactions with N-allyltryptamines, leading
to functionalized tetrahydro-b-carbolines 2, bearing pendent
allene functions, in good yields and enantiomeric excess. A pal-
ladium-catalyzed strategy was used to trigger both the N-allyl
deprotection and the cyclization of the intermediate amine on
the allene function via a transient p-allyl-Pd intermediate. De-
pending on the chain length and substitution pattern of the
allene function, regioselective cyclizations occurred through 5-
exo-, 6-exo-, or 7-endo-mechanisms, affording various tetracy-
clic compounds. The mechanism of the tandem reaction con-
sists of two distinct catalytic cycles involving Pd0 through “self-
relay catalysis”.[44] This tandem process saves the deprotection
step and allows an atom-economical formation of the p-allyl.
In addition, the dimethyl barbituric acid used as the allyl scav-
enger is also pivotal in the generation of the PdII-H intermedi-
ate. Notably, the method has been applied to unprotected
indole derivatives. Taken together, this tandem process can be
considered as a step- and atom-economical process for the
Typical procedure for (S)-2-allyl-1-(4-phenylhexa-4,5-dien-1-yl)-
2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole (2l): Prepared accord-
ing to the general procedure from 3a (70 mg, 0.350 mmol), 4q
(196 mg, 1.050 mmol), 10 (5.2 mg, 0.007 mmol), and 4 ꢃ molecular
sieves (230 mg) in toluene (5.2 mL). Purification on silica gel
(EtOAc/petroleum ether, 5:95 to 10:90) afforded 2l. Yield: 96 mg
(0.261 mmol, 74%); yellow oil; Rf =0.15 (EtOAc/petroleum ether,
5:95). [a]2D6 = +0.5 (c 1.00, CHCl3); IR (neat): n˜ =3413, 3056, 2931,
1940, 1493, 1451, 1300, 1155, 1171, 996, 919, 851, 763, 740,
695 cmꢀ1; ee=95% {Chiralpak AD-H column; heptanes/IPA, 95:5 +
0.1% Et3N; 1 mLminꢀ1; l=277 nm; tR =6.49 (major), 8.24 min
1
(minor)}; H NMR (300 MHz, CDCl3): d=7.51 (s, 1H; H-9), 7.46 (dd,
J=7.1, 1.5 Hz, 1H; ArH), 7.39 (dd, J=7.3, 1.6 Hz, 2H; 2ꢂArH), 7.33–
7.26 (m, 3H; 3ꢂArH), 7.22–7.16 (m, 1H; ArH), 7.15–7.04 (m, 2H; 2ꢂ
ArH), 5.91 (dtt, J=16.9, 10.3, 6.5 Hz, 1H; H-17), 5.15–5.04 (m, 4H;
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
8
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
2ꢂH-18 and 2ꢂH-15), 3.68 (t, J=5.8 Hz, 1H; H-1), 3.26–3.17 (m,
3H; 2ꢂH-16 and H-3a), 2.93 (ddd, J=13.4, 5.2, 3.4 Hz, 1H; H-3b),
2.86–2.75 (m, 1H; H-4a), 2.59–2.44 (m, 3H; H-4b and 2ꢂH-12),
1.88–1.70 ppm (m, 4H; 2ꢂH-10 and 2ꢂH-11); 13C NMR (75 MHz,
CDCl3): d=208.9 (Cq, C14), 137.0 (CH, C17), 136.6 (Cq, CAr), 136.0 (Cq,
CAr), 135.5 (Cq, CAr), 128.7 (CH, 2ꢂCAr), 127.5 (Cq, CAr), 126.9 (CH, CAr),
126.3 (CH, 2ꢂCAr), 121.6 (CH, CAr), 119.5 (CH, CAr), 118.2 (CH, CAr),
117.4 (CH2, C18), 110.8 (CH, CAr), 108.2 (Cq, CAr), 105.0 (Cq, C13), 78.3
(CH2, C15), 56.5 (CH2, C16), 56.2 (CH, C1), 45.3 (CH2, C3), 34.0 (CH2,
C10), 29.8 (CH2, C12), 24.8 (CH2, C11), 18.2 ppm (CH2, C4); HRMS
(ESI): m/z calcd for C26H29N2 369.2335 [M+H]+; found 369.2331.
CAr), 127.2 (CH, C4), 127.2 (Cq, CAr), 126.2 (CH, 2ꢂCAr), 121.9 (CH,
CAr), 119.8 (CH, CAr), 118.6 (CH, CAr), 111.1 (CH, CAr), 108.8 (Cq, CAr),
61.7 (CH, C13b), 52.9 (CH2, C5), 51.0 (CH2, C7), 31.2 (CH2, C1), 30.5
(CH2, C2), 21.0 ppm (CH2, C8); HRMS (ESI): m/z calcd for C22H23N2
315.1861 [M+H]+; found 315.1875.
Synthesis of (4R,12bS)-4-phenethyl-1,2,3,4,6,7,12,12b-octahy-
droindolo[2,3-a]quinolizine (19): Tetrahydro-b-carboline 2t
(120 mg, 0.326 mmol, two diastereoisomers), Pd(PPh3)4 (18.7 mg,
0.016 mmol), and 1,3-dimethylbarbituric acid 11 (152 mg,
0.978 mmol) were introduced in a reaction flask that was purged
with argon and dissolved in CH2Cl2 (12 mL). The reaction mixture
was stirred at RT for 16 h. A saturated aqueous solution of NaHCO3
was added, the organic layer was separated, and the aqueous
phase was extracted twice with CH2Cl2. Combined organic layers
were dried with MgSO4, filtered, and concentrated under vacuum.
The crude mixture was then purified by flash chromatography
(EtOAc/heptane, 20:80 to 100% EtOAc) to give a mixture of 18a
and 18b (65 mg, 0.199 mmol, 61%) and 18c (11.7 mg, 0.036 mmol,
11%). The mixture of 18a and 18b (35 mg, 0.107 mmol) was intro-
duced in a reaction flask that was purged with argon and dissolved
in THF/H2O (8 mL, 3:1). RuCl3·3H2O (3.6 mg, 0.006 mmol) and
NaBH4 (11.5 mg, 0.305 mmol) were then added and the mixture
was stirred at 508C for 16 h. Upon completion of the reaction, the
mixture was cooled to RT and a saturated aqueous solution of
NaHCO3 was added. The organic layer was separated, the aqueous
phase was extracted twice with MTBE, and the combined organic
layers were dried with MgSO4, filtered, and concentrated under
vacuum. The crude mixture was then purified by flash chromatog-
raphy (EtOAc/heptane, 20:80 to 40:60) to afford cis-19. Yield:
24 mg (0.073 mmol, 69%); orange oil; Rf =0.25 (heptane/EtOAc,
65:35); [a]2D5 =ꢀ25.9 (c 1.00, CHCl3); IR (neat): n˜ =3409, 2928, 2854,
1574, 1509, 1465, 1301, 1259, 1216, 1158, 1059, 1036, 870, 820,
General procedure for the tandem Pd0-catalyzed deprotection/
cyclization: Tetrahydro-b-carboline
2
(1 equiv), Pd(PPh3)4
(0.05 equiv), and 1,3-dimethylbarbituric acid 11 (3 equiv) were in-
troduced in a reaction flask that was purged with argon. CH2Cl2
was then added and the mixture was stirred at 408C or RT for 6 h.
Upon completion of the reaction, the mixture was cooled to RT
and a saturated aqueous solution of NaHCO3 was added. The or-
ganic layer was separated and the aqueous phase was extracted
twice with CH2Cl2. The combined organic layers were dried with
MgSO4, filtered, and concentrated under vacuum. The crude mix-
ture was then purified by flash chromatography to give the desired
products 12, 16, 17, or 18.
Typical
procedure
for
(4S,12bS)-4-phenyl-4-vinyl-
1,2,3,4,6,7,12,12b-octahydroindolo[2,3-a]quinolizine (12i): Pre-
pared according to the general procedure using 2l (75 mg,
0.203 mmol), Pd(PPh3)4 (11.7 mg, 0.010 mmol), and DMBA 11
(95 mg, 0.609 mmol) in CH2Cl2 (7.5 mL). Purification on silica gel
(EtOAc/petroleum ether, 2.5:97.5) afforded 12i. Yield: 44 mg
(0.134 mmol, 66%); yellow oil; Rf =0.27 (EtOAc/petroleum ether,
2.5:97.5); [a]2D5 =ꢀ48.2 (c 1.00, CHCl3); IR (neat): n˜ =3421, 2057,
2928, 2845, 1597, 1446, 1379, 1301, 1215, 1109, 1032, 1005, 923,
737, 663 cmꢀ1 1H NMR (500 MHz, CD3CN): d=8.90 (s, 1H; H-12),
;
1
739, 703 cmꢀ1; H NMR (500 MHz, CDCl3): d=7.69 (s, 1H; ArH), 7.65
7.41 (d, J=7.6 Hz, 1H; ArH), 7.33–7.25 (m, 5H; 5ꢂArH), 7.20–7.16
(m, 1H; ArH), 7.06 (t, J=7.3 Hz, 1H; ArH), 7.00 (t, J=7.6 Hz, 1H;
ArH), 3.47 (dt, J=11.6, 3.7 Hz, 1H), 3.42 (d, J=9.5 Hz, 1H), 2.82–
2.64 (m, 4H), 2.49–2.44 (m, 1H), 2.38–2.33 (m, 1H), 2.10–2.06 (m,
1H), 1.92–1.88 (m, 2H), 1.72–1.69 (m, 1H), 1.56–1.48 ppm (m, 3H);
13C NMR (75 MHz, CDCl3): d=142.9 (Cq, CAr), 141.4 (Cq, CAr), 136.3
(Cq, CAr), 128.6 (CH, 2ꢂCAr), 128.5 (CH, 2ꢂCAr), 127.6 (Cq, CAr), 126.0
(CH, CAr), 121.5 (CH, CAr), 119.5 (CH, CAr), 118.3 (CH, CAr), 110.9 (CH,
CAr), 108.4 (Cq, CAr), 61.4 (CH, C4), 60.3 (CH, C12b), 45.6 (CH2, C6),
36.1 (CH2, C13), 31.6 (CH2, C14), 30.4 (CH2, C1), 29.6 (CH2, C3), 24.5
(CH2, C2), 22.3 ppm (CH2, C7); HRMS (ESI): m/z calcd for C23H27N2
331.2174 [M+H]+; found 331.2199.
(s, 1H; H-12), 7.45 (d, J=7.6 Hz, 1H; ArH), 7.33–7.28 (m, 2H; 2ꢂ
ArH), 7.24–7.20 (m, 1H; ArH), 7.11 (t, J=7.6 Hz, 1H; ArH), 7.06 (t,
J=7.7 Hz, 1H; ArH), 6.14 (dd, J=18.0, 11.4 Hz, 1H; H-13), 5.68 (dd,
J=11.2, 1.3 Hz, 1H; H-14a), 5.36 (dd, J=18.0, 1.2 Hz, 1H; H-14b),
3.95 (d, J=9.8 Hz, 1H; H-12b), 2.97–2.92 (m, 1H; H-6a), 2.82–2.75
(m, 1H; H-7a), 2.57–2.50 (m, 2H; H-6b and H-7b), 2.13–2.06 (m, 2H;
H-3a and H-1a), 1.91–1.84 (m, 1H; H-2a), 1.77–1.65 ppm (m, 3H; H-
1b, H-2b and H-3b); 13C NMR (125 MHz, CDCl3): d=148.3 (Cq, CAr),
136.9 (Cq, CAr), 136.3 (Cq, CAr), 135.7 (CH, C13), 128.4 (CH, CAr), 127.7
(Cq, CAr), 126.9 (CH, 2ꢂCAr), 126.8 (CH, 2ꢂCAr), 121.4 (CH, CAr), 119.6
(CH, CAr), 119.2 (CH2, C14), 118.3 (CH, CAr), 110.8 (CH, CAr), 109.2 (Cq,
CAr), 66.8 (Cq, C4), 54.7 (CH, C12b), 45.9 (CH2, C6), 38.5 (CH2, C1),
31.9 (CH2, C3), 22.5 (CH2, C7), 21.4 ppm (CH2, C2); HRMS (ESI): m/z
calcd for C23H25N2 329.2018 [M+H]+; found 329.2012.
Acknowledgements
(S)-3-Phenyl-2,5,7,8,13,13b-hexahydro-1H-azepino[1’,2’:1,2]pyri-
do[3,4-b]indole (17g): Prepared according to the general proce-
dure using 2q (37 mg, 0.104 mmol), Pd(PPh3)4 (6 mg, 0.055 mmol),
and DMBA 11 (49 mg, 0.312 mmol) in CH2Cl2 (1.0 mL). Purification
on silica gel (EtOAc/heptane, 50:50) afforded 17g. Yield: 25 mg
(0.079 mmol, 76%); yellow powder; Rf =0.26 (EtOAc/heptane,
50:50); [a]2D5 =ꢀ92.7 (c 1.00, CHCl3); IR (neat): n˜ =2922, 2849, 1483,
V.G. thanks the ICSN for financial support. The authors warmly
thank Dr. Angela Marinetti (ICSN) for her support and Lei Yang
(ICSN) for his contribution at the very beginning of the project.
1448, 1340, 1237, 1129, 1076, 1038, 909, 815, 742 cmꢀ1 1H NMR
;
Keywords: alkaloids · allenes · enantioselectivity · nitrogen
heterocycles · palladium
(300 MHz, CDCl3): d=7.73 (s, 1H; H-13), 7.48 (d, J=7.5 Hz, 1H;
ArH), 7.35–7.21 (m, 6H; 6ꢂArH), 7.16–7.06 (m, 2H; 2ꢂArH), 6.04
(td, J=5.7, 1.1 Hz, 1H; H-4), 4.22 (dd, J=10.4, 4.0 Hz, 1H; H-13b),
3.65 (dd, J=15.6, 5.1 Hz, 1H; H-5a), 3.52 (dd, J=16.0, 6.0 Hz, 1H;
H-5b), 3.27–3.19 (m, 1H; H-7a), 3.04–2.88 (m, 3H; H-7b, H-2a and
H-8a), 2.85–2.68 (m, 2H; H-2b and H-8b), 2.27–2.19 (m, 1H; H-1a),
2.06–1.94 ppm (m, 1H; H-1b); 13C NMR (75 MHz, CDCl3): d=144.0
(Cq, C3), 136.3 (Cq, CAr), 135.9 (Cq, CAr), 133.8 (Cq, CAr), 128.6 (CH, 2ꢂ
[1] a) W. Gul, M. T. Hamann, Life Sci. 2005, 78, 442–453; b) A. J. Kochanow-
ska-Karamyan, M. T. Hamann, Chem. Rev. 2010, 110, 4489–4497; c) A.
Rahman, A. Basha, Indole Alkaloids, Harwood Academic Publishers,
Reading, UK, 1999; d) M. J. Fisher, R. T. Backer, S. Husain, H. M. Hsiung,
J. T. Mullaney, T. P. O’Brian, P. L. Ornstein, R. R. Rothhaar, J. M. Zgombick,
K. Briner, Bioorg. Med. Chem. Lett. 2005, 15, 4459–4462.
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
9
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[2] G. Bartoli, G. Bencivenni, R. Dalpozzo, Chem. Soc. Rev. 2010, 39, 4449–
4465.
[21] a) J. Kuang, S. Ma, J. Org. Chem. 2009, 74, 1763–1765; b) J. Kuang, H.
Luo, S. Ma, Adv. Synth. Catal. 2012, 354, 933–944.
[3] a) G. R. Humphrey, J. T. Kuethe, Chem. Rev. 2006, 106, 2875–2911; b) S.
Cacchi, G. Fabrizi, Chem. Rev. 2011, 111, PR215–PR283; c) S. Cacchi, G.
Fabrizi, Chem. Rev. 2005, 105, 2873–2920; d) M. Bandini, A. Eichholzer,
Angew. Chem. Int. Ed. 2009, 48, 9608–9644; Angew. Chem. 2009, 121,
9786–9824.
[4] a) P. Mondal, N. P. Argade, J. Org. Chem. 2013, 78, 6802–6808; b) S. M.
Allin, C. I. Thomas, J. E. Allard, K. Doyle, M. R. J. Elsegood, Tetrahedron
Lett. 2004, 45, 7103–7105; c) J. Sun, L.-L. Zhang, C.-G. Yan, Tetrahedron
2013, 69, 5451–5459; d) M. Amat, M. M. M. Santos, O. Bassas, N. Llor, C.
Escolano, A. Gꢄmez-Esquꢀ, E. Molins, S. M. Allin, V. McKee, J. Bosch, J.
Org. Chem. 2007, 72, 5193–5201; e) O. Bassas, N. Llor, M. M. M. Santos,
R. Griera, E. Molins, M. Amat, J. Bosch, Org. Lett. 2005, 7, 2817–2820;
f) M. Perez, F. Arioli, G. Rigacci, M. M. M. Santos, A. Gꢄmez-Esquꢀ, C. Es-
colano, P. Florindo, C. Ramos, J. Bosch, M. Amat, Eur. J. Org. Chem. 2011,
3858–3863; g) C. Sanaboina, S. Jana, S. Chidara, B. Patro, G. B. Raolji, L.
Eppakayala, Tetrahedron Lett. 2012, 53, 5027–5029.
[22] C. H. Larsen, K. W. Anderson, R. E. Tundel, S. L. Buchwald, Synlett 2006,
2941–2946.
[23] V. Gobꢀ, P. Retailleau, X. Guinchard, Org. Lett. 2014, 16, 5438–5441.
[24] a) H. Tsukamoto, T. Matsumoto, Y. Kondo, Org. Lett. 2008, 10, 1047–
1050; b) C. Aubert, P. Garcia, Y. Harrak, L. Diab, P. Cordier, C. Ollivier, V.
Gandon, M. Malacria, L. Fensterbank, Org. Lett. 2011, 13, 2952–2955.
[25] Y. Kaburagi, Y. Kishi, Org. Lett. 2007, 9, 723–726.
[26] K. Omura, D. Swern, Tetrahedron 1978, 34, 1651–1660.
[27] See the Supporting Information for the synthesis of 6w.
[28] A 2-butenyl functionalized tryptamine was used with success by Hiem-
stra, resulting in a 86% ee. See Ref. [16a].
[29] The use of 3 equiv of aldehyde is widespread in asymmetric Pictet–
Spengler reaction catalyzed by chiral phosphoric acids. However, in the
case of expensive aldehydes it can be considered a limitation. We
found that 60 and 72% of aldehydes 4q and 4t could be recovered in
the course of the synthesis of 2l and 2p, respectively.
[5] S. E. O’Connor, J. J. Maresh, Nat. Prod. Rep. 2006, 23, 532–547.
[6] J. A. Duan, I. D. Williams, C. T. Che, R. H. Zhou, S. X. Zhao, Tetrahedron
Lett. 1999, 40, 2593.
[7] G. Massiot, P. Thepenier, M. J. Jacquier, L. Lemenolivier, R. Verpoorte, C.
Delaude, Phytochemistry 1987, 26, 2839–2846.
[8] R. T. Brown, C. L. Chapple, Tetrahedron Lett. 1976, 17, 1629–1630.
[9] J. Peng, J.-F. Hu, A. B. Kazi, Z. Li, M. Avery, O. Peraud, R. T. Hill, S. G.
Franzblau, F. Zhang, R. F. Schinazi, S. S. Wirtz, P. Tharnish, M. Kelly, S. Wa-
hyuono, M. T. Hamann, J. Am. Chem. Soc. 2003, 125, 13382–13386.
[10] K. Awang, M. R. Mukhtar, A. H. A. Hadi, M. Litaudon, J. Latip, N. R. Abdul-
lah, Nat. Prod. Res. 2006, 20, 567–572.
[11] a) B. C. Maiti, V. S. Giri, S. C. Pakrashi, Heterocycles 1990, 31, 847–850;
b) A. W. Rey, W. A. Szarek, D. B. Maclean, Heterocycles 1991, 32, 1143–
1151; c) M. Ohba, H. Kubo, T. Fujii, H. Ishibashi, M. V. Sargent, D. Arbain,
Tetrahedron Lett. 1997, 38, 6697–6700; d) M. Ohba, H. Kubo, H. Ishiba-
shi, Tetrahedron 2000, 56, 7751–7761; e) F. A. Davis, J. Y. Melamed, S. S.
Sharik, J. Org. Chem. 2006, 71, 8761–8766.
[30] For the absolute stereochemistry determination, see the Supporting In-
formation.
[31] H. Kunz, J. Mꢅrz, Angew. Chem. Int. Ed. Engl. 1988, 27, 1375–1377;
Angew. Chem. 1988, 100, 1424–1425.
[32] cis and trans diastereoisomers of piperidines 12 present strongly differ-
ent Rf on silica gel and are consequently easily separated.
[33] a) T.-S. Kam, K.-M. Sim, Phytochemistry 1998, 47, 145–147; b) J. i. Ko-
bayashi, M. Sekiguchi, S. Shimamoto, H. Shigemori, H. Ishiyama, A.
Ohsaki, J. Org. Chem. 2002, 67, 6449–6455; c) C.-E. Nge, C.-Y. Gan, Y.-Y.
Low, N. F. Thomas, T.-S. Kam, Org. Lett. 2013, 15, 4774–4777; d) R. T.
Brown, S. B. Fraser, Tetrahedron Lett. 1974, 15, 1957–1959; e) V. Samoy-
lenko, M. M. Rahman, B. L. Tekwani, L. M. Tripathi, Y.-H. Wang, S. I. Khan,
I. A. Khan, L. S. Miller, V. C. Joshi, I. Muhammad, J. Ethnopharmacol.
2010, 127, 357–367.
[34] S. Qiu, G. Liu, Y. Wei, Chem. Eur. J. 2009, 15, 2751–2754.
[35] P. K. Sharma, S. Kumar, P. Kumar, P. Nielsen, Tetrahedron Lett. 2007, 48,
8704–8708.
[12] a) A. Pictet, T. Spengler, Ber. Dtsch. Chem. Ges. 1911, 44, 2030–2036;
b) E. D. Cox, J. M. Cook, Chem. Rev. 1995, 95, 1797–1842; c) J. Stçckigt,
A. P. Antonchick, F. R. Wu, H. Waldmann, Angew. Chem. Int. Ed. 2011, 50,
8538–8564; Angew. Chem. 2011, 123, 8692–8719.
[36] The trans diastereoisomers with E double bond were obtained in small
amounts and were not engaged in the reduction.
[37] It was shown that 50% scrambling occurs after 1 h of reaction of 2a or
15a with DMBA-D2 in dichloromethane.
[13] a) M. S. Taylor, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126, 10558–
10559; b) R. S. Klausen, E. N. Jacobsen, Org. Lett. 2009, 11, 887–890.
[14] a) J. Seayad, A. M. Seayad, B. List, J. Am. Chem. Soc. 2006, 128, 1086–
1087; b) M. J. Wanner, R. N. S. van der Haas, K. R. de Cuba, J. H. van
Maarseveen, H. Hiemstra, Angew. Chem. Int. Ed. 2007, 46, 7485–7487;
Angew. Chem. 2007, 119, 7629–7631; c) N. V. Sewgobind, M. J. Wanner,
S. Ingemann, R. de Gelder, J. H. van Maarseveen, H. Hiemstra, J. Org.
Chem. 2008, 73, 6405–6408; d) D. Huang, F. Xu, X. Lin, Y. Wang, Chem.
Eur. J. 2012, 18, 3148–3152.
[15] N. Mittal, D. X. Sun, D. Seidel, Org. Lett. 2014, 16, 1012–1015.
[16] a) M. J. Wanner, R. N. A. Boots, B. Eradus, R. de Gelder, J. H. van Maar-
seveen, H. Hiemstra, Org. Lett. 2009, 11, 2579–2581; b) M. J. Wanner, E.
Claveau, J. H. van Maarseveen, H. Hiemstra, Chem. Eur. J. 2011, 17,
13680–13683; c) B. Herlꢀ, M. J. Wanner, J. H. van Maarseveen, H. Hiem-
stra, J. Org. Chem. 2011, 76, 8907–8912.
[17] a) Q. Cai, X. W. Liang, S. G. Wang, S. L. You, Org. Biomol. Chem. 2013, 11,
1602; b) C. A. Holloway, M. E. Muratore, R. L. Storer, D. J. Dixon, Org. Lett.
2010, 12, 4720; c) Q. Cai, X. W. Liang, S. G. Wang, J. W. Zhang, X. Zhang,
S. L. You, Org. Lett. 2012, 14, 5022; d) I. T. Raheem, P. S. Thiara, E. A. Pe-
terson, E. N. Jacobsen, J. Am. Chem. Soc. 2007, 129, 13404–13405;
e) E. A. Peterson, E. N. Jacobsen, Angew. Chem. Int. Ed. 2009, 48, 6328–
6331; Angew. Chem. 2009, 121, 6446–6449; f) I. P. Kerschgens, E. Clav-
eau, M. J. Wanner, S. Ingemann, J. H. van Maarseveen, H. Hiemstra,
Chem. Commun. 2012, 48, 12243–12245.
[38] T. H. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis, 4 ed.,
Wiley, Hoboken, 2007.
[39] R. Zimmer, C. U. Dinesh, E. Nandanan, F. A. Khan, Chem. Rev. 2000, 100,
3067–3125.
[40] a) Y. Yamamoto, M. Al-Masum, N. Asao, J. Am. Chem. Soc. 1994, 116,
6019–6020; b) Y. Yamamoto, M. Al-Masum, Synlett 1995, 969–970;
c) M. Al-Masum, M. Meguro, Y. Yamamoto, Tetrahedron Lett. 1997, 38,
6071–6074; d) M. Meguro, Y. Yamamoto, Tetrahedron Lett. 1998, 39,
5421–5424; e) L. Besson, J. Gorꢀ, B. Cazes, Tetrahedron Lett. 1995, 36,
3857–3860; f) M. Meguro, S. Kamijo, Y. Yamamoto, Tetrahedron Lett.
1996, 37, 7453–7456; g) L. Besson, J. Gorꢀ, B. Cazes, Tetrahedron Lett.
1995, 36, 3853–3856; h) B. M. Trost, V. J. Gerusz, J. Am. Chem. Soc.
1995, 117, 5156–5157; i) B. M. Trost, P. Y. Michellys, V. J. Gerusz, Angew.
Chem. Int. Ed. Eng. 1997, 36, 1750–1753; Angew. Chem. 1997, 109,
1837–1839; j) B. M. Trost, C. Jꢅkel, B. Plietker, J. Am. Chem. Soc. 2003,
125, 4438–4439; k) B. M. Trost, A. B. C. Simas, B. Plietker, C. Jꢅkel, J. Xie,
Chem. Eur. J. 2005, 11, 7075–7082; l) B. M. Trost, J. Xie, J. Am. Chem.
Soc. 2008, 130, 6231–6242; m) B. M. Trost, J. Xie, J. D. Sieber, J. Am.
Chem. Soc. 2011, 133, 20611–20622.
[41] B. M. Trost, Science 1991, 254, 1471–1477.
[42] B. M. Trost, Chem. Eur. J. 1998, 4, 2405–2412.
[43] Monoallyl DMBA 14 is likely to be better trapping agent than DMBA
itself, because diallyl DMBA is the only derivative found at the end of
the reaction.
[18] V. Gobꢀ, X. Guinchard, Org. Lett. 2014, 16, 1924–1927.
[44] N. T. Patil, V. S. Shinde, B. Gajula, Org. Biomol. Chem. 2012, 10, 211–224.
[19] a) K. M. Brummond, J. E. DeForrest, Synthesis 2007, 795–818; b) S. Yu, S.
Ma, Chem. Commun. 2011, 47, 5384–5418.
[20] a) P. Crabbe, H. Fillion, D. Andre, J. L. Luche, J. Chem. Soc. Chem.
Commun. 1979, 859–860; b) Alternatively, we also used the methods
derived from the Crabbe reaction: S. Searles, Y. Li, B. Nassim, M. T. R.
Lopes, P. T. Tran, P. Crabbe, J. Chem. Soc. Perkin Trans. 1 1984, 747–751.
Received: January 21, 2015
Published online on && &&, 0000
&
&
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
10
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Alkaloids
V. Gobꢀ, X. Guinchard*
&& – &&
Stereoselective Synthesis of Chiral
Polycyclic Indolic Architectures
through Pd0-Catalyzed Tandem
Deprotection/Cyclization of
The right fold: Enantioenriched N-allyl
tetrahydro-b-carbolines were prepared
by chiral phosphoric acid-catalyzed
Pictet–Spengler reactions. The com-
pounds undergo Pd0-catalyzed cycliza-
tions through a tandem deprotection/
cyclization process (see scheme). The re-
gioselectivity of the attack is controlled
by the chain length and by the substitu-
tion pattern of the allene function.
Products resulting from 5-exo- or 6-exo-
attack were obtained with diastereoiso-
meric ratios up to 95:5. Azepinopyrri-
do[3,4-b]indoles were obtained by 7-
endo-cyclizations.
Tetrahydro-b-carbolines on Allenes
Chem. Eur. J. 2015, 21, 1 – 11
www.chemeurj.org
11
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ