New entry to polycyclic fused indoles via gold(I)-catalyzed cascade reaction

Article abstract of DOI:10.1002/asia.201201249

The gold standard: A gold-catalyzed cascade reaction sequence for the preparation of polycyclic fused indole cores exploits the ready availability and chemical flexibility of propargylic alcohols. The desired tri- or tetracyclic compounds are obtained in good yields with water as the only stoichiometric by-product. Copyright

Full text of DOI:10.1002/asia.201201249

COMMUNICATION
DOI: 10.1002/asia.201201249
New Entry to Polycyclic Fused Indoles via Gold(I)-catalyzed Cascade
Reaction
Michel Chiarucci,[a] Elia Matteucci,[a] Gianpiero Cera,[a] Giancarlo Fabrizi,[b] and
Marco Bandini*[a]
Polycyclic indoles are pivotal motifs in naturally occurring
compounds, featuring a plethora of pharmacological and
agrochemical applications.[1] Large chemical decoration and
nontrivial molecular scaffolds are common architectural pat-
terns in indole-based alkaloids that still trigger organic
chemists in developing chemically and economically effi-
cient and sustainable methodologies for their preparation.
Accordingly, efforts towards the synthesis of indole deriva-
tives by means of metal as well as metal-free catalysis are
growing exponentially.[2]
In this realm, N-fused indole alkaloids based on oxazino-
ACHTUNGTRENNUNG[4,3-a]indole and pyrazinoAHCTUNGTRENN[UGN 1,2-a]indole platforms have at-
tracted growing attention due to their peculiar activity as li-
gands for 5-HT2C receptor, antidepressant, and 5-HT4 recep-
tor antagonists.[3]
Scheme 1. a,b) State of the art in the synthesis azepino
c) Synthetic approach of this work.
ACHTUNGTREN[NGNU 4,3-a]indoles.
Very recently, the synthesis of densely functionalized oxa-
zinoindoles has been addressed independently by the groups
of Xiao[4a] and Gharpure[4b] through the N(1) or C(2) alkyla-
tion of the indole core with vinyl sulfonium salts and via Mi-
chael addition, respectively. Both of these elegant ap-
proaches required stoichiometric amounts of Lewis acid
(i.e., (CH3)3SiOTf) or base (i.e., KOH) along with the pre-
formed indole nucleus, with inevitable repercussions on the
availability of the acyclic precursors (Scheme 1a,b).
or electrophilic character, depending on the chemical sur-
rounding.[8] Based on this chemical flexibility, we envisioned
an unprecedented synthesis of densely functionalized tricy-
clic oxazinoACTHNUGTRNEUNG[4,3-a]indoles 2 by means of simultaneous con-
struction of the indole and the N(1)-C(2)-fused ring, starting
from readily available aniline diols 1. The mechanistic plan
would involve a cascade hydroamination/dehydrative syn-
thetic sequence, delivering water as the only stoichiometric
by-product of the entire process (Scheme 1c).
The choice of unprotected amine diol 1 as model sub-
strate significantly restricted the class of potentially useful
promoting agents. While Brønsted acids would presumably
form the unreactive corresponding ammonium salts, conven-
tional s Lewis acids could either be irreversibly deactivated
by coordination to the hard heteroatoms or would lead to
decomposition of the starting material if propargylic carbo-
We recently entered a new scientific program addressing
the simultaneous synthesis and functionalization of indole
cores assisted by the same catalytic species.[5] In particular,
we documented the combined use of p-activated alcohols[6]
(i.e., propargylic alcohols) and gold catalysis[7] as a powerful
tool towards this goal. Tertiary propargylic alcohols,[8] when
allowed to react in the presence of mild metal bifunctional
p/s acids,[9] can act as electrophiles towards the C C triple
À
bond, and the alcoholic group can exert either nucleophilic
[a] Dr. M. Chiarucci, E. Matteucci, G. Cera, Prof. M. Bandini
Department of Chemistry “G. Ciamician”, Alma Mater Studiorum—
University of Bologna
À
cations are formed before the hydroamination of the C C
triple bond takes place. Consequently, p acid late-transition-
metal species appeared to be the promoters of choice for
the titled transformation.
In Table 1, we summarize some of the results obtained
during the optimization of the catalytic system, choosing 1a
as the model substrate. As expected, organic Brønsted acids
such as HNTf2 and pTsOH were not efficient, leading to
rapid decomposition of 1a (Table 1, entries 1,2). Similar
via Selmi 2, 40138 Bologna (Italy)
Fax : (+39)051-2099456
E-mail: marco.bandini@unibo.it
[b] Prof. G. Fabrizi
Dipartimento di Chimica e Tecnologia del Farmaco
Sapienza, Universitꢀ di Roma
P.le A. Moro, 5, Rome (Italy)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201201249.
chemical outcomes were recorded with InACTHNURTGNENG(U OTf)3 (Table 1,
1
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
Marco Bandini et al.
Table 1. Optimization of the catalytic system for the cascade synthesis of
oxazino
[4,3-a]indole 2a.[a]
time was required (48 h) but synthetically useful yield
(76%) was still obtained (Table 1, entry 16).
ACHTUNGTRENNUNG
Having established the optimal reaction conditions, we ex-
amined the generality of the methodology in terms of sub-
strate scope. A range of tertiary and secondary propargylic
alcohols (1b–s) were synthesized and subjected to the
double ring-closing process under the assistance of [XPho-
sAuNTf2] (5 mol%, toluene, 508C). A collection of tricyclic
and tetracyclic products is summarized in Scheme 2.
Entry Catalyst
T [8C]/t [h] Yield 2a/2a’ [%][b]
1[c]
2[c]
3[c]
4[c]
5[d]
6
HNTf2
pTsOH
AgNTf2
25/3.5
25/3.5
25/3.5
25/3.5
25/3.5
25/3.5
25/3.5
–/–
–/–
9/–
–/–
In
N
G
–/–
AuCl3/AgOTf
[PPh3AuNTf2]
[JhonPhosAu]ACHTUNGTRENNUNG(SbF6)ACHTUNGTRENNUNG
16/–
–/11
–/57
55/–
71/–
62/–
33/–
–/24
–/25
92/–
76/–
7[d]
8
9
ACHTUNGTRENNUNG[AuIPrCl]/AgNTf2
10
[XPhosAuNTf2]
25/3.5
25/3.5
11[e]
12[e]
13[e]
14[e]
15
[XPhosAuCl]/AgOTf
[XPhosAuCl]/AgOTs
[XPhosAuCl]/AgSbF6
[XPhosAuCl]/AgBF4
[XPhosAuNTf2]
25/3.5
25/3.5
25/3.5
50/3.5
50/48
16[f]
[XPhosAuNTf2]
[a] All the reactions were carried out under nitrogen atmosphere in anhy-
drous toluene with 5 mol% loading of catalyst, unless otherwise speci-
fied. [b] Yields of produce isolated after flash chromatography. [c] Sub-
stantial decomposition of 1a was observed. [d] Starting material 1a was
recovered untouched. [e] XPhosAuCl was formed in situ. [f] With
1 mol% of XPhosAuNTf2. Tf=CF3SO2, Ts=p-CH3C6H4SO2, JhonPhos=
(2-biphenyl)di-tert-butylphosphine,
XPhos=dicyclohexyl[2’,4’,6’-tris(1-
methylethyl)[1,1’-biphenyl]-2-yl]phosphine, IPr=1,3-bis(diisopropylphe-
nyl)imidazol-2-ylidene.
Scheme 2. Scope of the reaction (reaction conditions: XPhosAuNTf2
(5 mol%), toluene, 508C, 3.5 h, unless otherwise specified). All substrates
were obtained as racemic mixtures. [a]: IPrAuCl/AgNTf2 (5 mol%) as
the catalyst.
entry 4), while AgNTf2 furnished 2a but in unsatisfactory
manner (yield=9%, Table 1, entry 3).
Some improvements were recorded in the presence of
AuCl3/AgOTf (5 mol%), which led to 2a in 16% yield
along with the recovery of untouched 1a (Table 1, entry 6).
This finding prompted us to investigate less heterophilic
[AuI] species, and the results are listed in Table 1, entries 7–
A range of oxazinoindoles (2b–h) were obtained in good
to excellent yields (53–96%). In particular, aliphatic (2d–g)
and aromatic groups (2b,c,h,) on the carbynol carbon atom
were adequately tolerated along with electron-withdrawing
and electron-donating groups on the aniline ring. Moreover,
the oxazinyl ring was also substituted with aryl and alkyl
groups, leading to compounds 2i–k in good yields (67–
98%).
16.
Well-defined
silver-free
Gagosh complex
[PPh3AuNTf2][10]
and
[JhonPhosAu]
A
(CH3CN)[11]
showed some aptitude in triggering the initial 5-endo-dig
cyclization (2a’, 11% and 57%, respectively), but the
second ring-closing event was not promoted at all (Table 1,
Remarkably, the methodology was efficiently applied to
the levulinate derivative 1g that led to the key precursor of
antidepressant AY-23,673 in 71% yield (1008C, 3.5 h).[15] Im-
entries 7,8). To our delight, cationic gold carbene species
[12]
[AuIPrCl]/AgNTf2
and commercially available [XPho-
sAuNTf2][13] provided 2a in 55% and 71% yield, respective-
ly, at room temperature after 3.5 h reaction time (Table 1,
entries 9,10).
The role of the counterion was then investigated, and con-
trol experiments (Table 1, entries 11–14) clearly show the su-
perior efficiency of NTf2 in comparison to OTf, OTs, and
SbF6.[14] Finally, we were pleased that the reaction at 508C
in the presence of silver-free [XPhosAuNTf2] (5 mol%)
gave quantitative conversion within 3.5 h, providing 2a in
92% yield (Table 1, entry 15). The loading of the catalyst
was also reduced to 1 mol%; in this case longer reaction
portant building blocks in medical chemistry such as tetra-
hydro-[1,4]oxazepino
G
a
seven-
membered ring fused at the N(1)/C(2) positions (2l–o),
were also accessible through the present protocol in good
yield (53–92%).
&
&
2
Chem. Asian J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

www.chemasianj.org
Marco Bandini et al.
Finally, we explored the gold-catalyzed cascade reaction
with secondary propargylic alcohols. Despite the higher re-
action temperature required (1008C), the corresponding tar-
geted compounds 2p–r were obtained in reasonable yields
(56–82%). On the contrary, primary propargylic alcohol 1s
was recovered untouched, thus suggesting that ionic species
À
are involved in the final C O bond-forming event of the
catalytic cycle (see below).
The synthetic potential of the present methodology was
underlined further by the realization of a triple-cascade re-
action using 1t as starting material. The use of 5 mol%
[AuIPrNTf2] led to the one-pot regioselective 5-endo-dig hy-
À
droamination of the C C triple bond, alkoxylation of the
carbynol carbon atom, and 6-exo-trig hydroindolination of
the olefin in satisfying yield (69%, Scheme 3). The three
consecutive gold-assisted ring-closing events (i.e., hydroami-
nation, alkoxylation, hydroindolination) open an efficient
access to tetracyclic indolyl scaffolds of type 2t, featuring
concomitant N(1), C(2) and C(3) functionalization.
Scheme 4. a) Tentative reaction mechanism for the [AuI]-catalyzed cas-
cade reaction. b,c) Control experiments that support the SN1 mechanism
via intimate ion pairs. NR=no reaction.
Scheme 3. Triple cascade reaction promoted by gold.
Scheme 4a shows a tentative reaction mechanism. In par-
ticular, after the initial regioselective gold-triggered forma-
tion of the indolyl core (5-endo-dig hydroamination of the
triple bond), protodeauration of the resulting indolyl-3-Au
intermediate A[16] would lead to intermediate 2’. In support
of this pathway, indolyl diol 2’ was isolated in some cases
from the crude reaction mixtures as a minor compound, and
when treated in toluene in the presence of freshly added
[XPhosAuNTf2] (5 mol%), the corresponding tricyclic com-
pound 2a was isolated in quantitative yield. At this stage,
several mechanistic channels could be hypothesized.
In particular, dehydration and consequent oxyalkylation
of the olefin can be excluded due to the thermodynamically
unfavorable dehydration of C2-indole tertiary alcohols at
508C. Moreover, when the probe ene-ino compound 3 was
subjected to gold catalysis, only partially cyclized 2-vinyl
indole 4 was isolated in 80% yield (Scheme 4b) Interesting-
ly, compound 4 was recovered untouched even after pro-
longed reaction (48 h), thus demonstrating the unfeasibility
of the gold-catalyzed hydroalkoxylation of the electron-rich
carbon–carbon double bond.
To more thoroughly investigate the second cyclization
event, enantiomerically enriched alcohols 1a (ee=60%)
and 1q (ee=81%, see the Supporting Information) were al-
lowed to react in the presence of [XPhosAuNTf2] (5 mol%).
Unexpectedly, under optimal conditions (i.e., toluene,
508C), significant but not complete racemization was re-
corded in both cases (Scheme 4c).[17,18] This evidence can be
rationalized with a “not-pure” SN1-type mechanism[19] in the
presence of ion pairs of type B. Accordingly, we recently re-
ported on the “folding effect” played by the counterion in
the enantioselective gold-catalyzed alkylation of indoles
with allylic alcohols.[20] Here, the negatively charged species
(i.e., OTfÀ) binds simultaneously to the acidic protons of the
substrate. Analogous situation could be envisioned in the
À
present C O bond-forming process (see intermediate 2’),
with direct impact of the gold counterion on the geometry
of the incoming electrophilic center.[21]
À
To confirm the presence of ion pairing in the C O bond-
forming event, a highly coordinating solvent (i.e., THF) was
3
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ

www.chemasianj.org
Marco Bandini et al.
Eur. J. Org. Chem. 2005, 2647–2656; c) J. Muzart, Eur. J. Org.
Chem. 2007, 3077–3089; d) M. Bandini, M. Tragni, Org. Biomol.
Chem. 2009, 7, 1501–1507; e) E. Emer, R. Sinisi, M. G. Capdevila,
D. Petruzziello, F. D. Vincentiis, P. G. Cozzi, Eur. J. Org. Chem.
2011, 647–666; f) B. Biannic, A. Aponick, Eur. J. Org. Chem. 2011,
6605–6617; g) M. Bandini, G. Cera, M. Chiarucci, Synthesis 2012,
504–512.
utilized, leading to the desired product in lower yield (39%)
and to complete racemization. This speculative interpreta-
tion is also in agreement with the decreasing reactivity trend
recorded with the propargylic alcohols (tertiary>secon-
dary>primary) that also tends to exclude a mixed SN1–SN2
reaction profile.[22]
[7] For recent monographs and reviews of the use of gold complexes in
homogeneous catalysis, see: a) Modern Gold Catalyzed Synthesis
(Eds.: A. S. K. Hashmi, F. D. Toste), Wiley-VCH, Weinheim, 2012;
b) Gold Catalysis: An Homogeneous Approach (Eds.: F. D. Toste, V.
Michelet), Imperial College Press, London, 2012; c) M. Rudolph,
A. S. K. Hashmi, Chem. Soc. Rev. 2012, 41, 2448–2462; d) T.
de Haro, C. Nevado, Synthesis 2011, 2530–2539; e) M. Bandini,
Chem. Soc. Rev. 2011, 40, 1358–1367.
In conclusion, an unprecedented gold-catalyzed cascade
reaction sequence for the preparation of polycyclic fused
indole cores is reported. The methodology exploits the
ready availability and chemical flexibility of propargylic al-
cohols, leading to the desired tri- or tetracyclic compounds
in good yields and delivering water as the only stoichiomet-
ric by-product. Efforts towards the realization of an enantio-
selective variant of the protocol are currently underway and
will be presented in due course.
[8] E. B. Bauer, Synthesis 2012, 1131–1151.
[9] Y. Yamamoto, J. Org. Chem. 2007, 72, 7817–7831.
[10] N. Mꢆzailles, L. Ricard, F. Gagosz, Org. Lett. 2005, 7, 4133–4136.
[11] a) C. Nieto-Oberhuber, S. Lꢇpez, A. M. Echavarren, J. Am. Chem.
Soc. 2005, 127, 6178–6179; b) E. Herrero-Gꢅmez, C. N. Oberhuber,
S. Lꢅpez, J. Benet-Buchholz, A. M. Echavarren, Angew. Chem.
2006, 118, 5581–5585; Angew. Chem. Int. Ed. 2006, 45, 5455–5459.
[12] a) N. Marion, S. P. Nolan, Chem. Soc. Rev. 2008, 37, 1776–1782;
b) S. Dꢃez-Gonzꢄlez, N. Marion, S. P. Nolan, Chem. Rev. 2009, 109,
3612–3672; c) S. P. Nolan, Acc. Chem. Res. 2011, 44, 91–100; d) S.
Gaillard, C. S. J. Cazin, S. P. Nolan, Acc. Chem. Res. 2012, 45, 778–
787.
Acknowledgements
Acknowledgements is made to Progetto FIRB “Futuro in Ricerca” Inno-
À
vative sustainable synthetic methodologies for C H activation processes,
(MIUR, Rome) and Universitꢀ di Bologna.
[13] G. C. Fortman, S. P. Nolan, Organometallics 2010, 29, 4579–4583.
[14] For recent studies on the role of counterions in gold(I)-catalyzed or-
ganic transformations, see: a) H. Schmidbaur, A. Schier, Z. Natur-
forsch. B 2011, 66, 0329–0350; b) D. Wang, R. Cai, S. Sharma, J.
Jirak, S. K. Thummanapelli, N. G. Akhmedov, H. Zhang, X. Liu,
J. L. Petersen, X. Shi, J. Am. Chem. Soc. 2012, 134, 9012–9019.
[15] M. Ding, F. He, M. A. Poss, K. L. Rigat, Y.-K. Wang, S. B. Roberts,
D. Qiu, R. A. Fridell, M. Gao, R. G. Gentles, Org. Biomol. Chem.
2011, 9, 6654–6662, and references cited therein.
[16] A. S. K. Hashmi, T. D. Ramamurthi, F. Rominger, Adv. Synth. Catal.
2010, 352, 971–975.
[17] For a recent review on gold-catalyzed chiral transfer and memory
effects, see: N. T. Patil, Chem. Asian J. 2012, 7, 2186–2194.
[18] At the present it is not possible to discriminate between partial in-
version or retention of configuration.
[19] a) I. P. Beletskaya, Russ. Chem. Rev. 1975, 44, 1067–1090; b) J. M.
Harris, Prog. Phys. Org. Chem. 1974, 11, 89–173; c) D. J. Raber,
J. M. Harris, P. v. R. Schleyer in Ions and Ion Pairs in Organic Reac-
tions, Vol. 2 (Ed.: M. Szwarc), Wiley, New York, 1974, pp. 247–374.
[20] M. Bandini, A. Bottoni, M. Chiarucci, G. Cera, G. Miscione, J. Am.
Chem. Soc. 2012, 134, 20690–20700.
[21] a) D. Zuccaccia, L. Belpassi, F. Tarantelli, A. Macchioni, J. Am.
Chem. Soc. 2009, 131, 3170–3171; b) N. Salvi, L. Belpassi, D. Zuc-
caccia, F. Tarantelli, A. Macchioni, J. Organomet. Chem. 2010, 695,
2679–2686.
Keywords: domino reactions · heterocycles · homogeneous
catalysis · gold
[1] a) W. G. Sumpter, F. M. Miller in Natural Products Containing
Indole, Wiley, New York, 2008, chap. VIII.
[2] Synthesis of indoles: a) S. Cacchi, G. Fabrizi, Chem. Rev. 2005, 105,
2873–2920; b) K. Krꢂger, A. Tillack, M. Beller, Adv. Synth. Catal.
2008, 350, 2153–2167; c) J. Barluenga, F. Rodrꢃguez, F. J. FaÇanꢄs,
Chem. Asian J. 2009, 4, 1036–1048; d) S. Cacchi, G. Fabrizi, Chem.
Rev. 2011, 111, PR215–PR283; e) R. Vicente, Org. Biomol. Chem.
2011, 9, 6469–6480; functionalization: f) M. Bandini, A. Eichholzer,
Angew. Chem. 2009, 121, 9786–9824; Angew. Chem. Int. Ed. 2009,
48, 9608–9644; g) S.-L. You, Q. Cai, M. Zheng, Chem. Soc. Rev.
2009, 38, 2190–2201; h) L. Joucla, L. Djakovitch, Adv. Synth. Catal.
2009, 351, 673–714; i) G. Bartoli, G. Bencivenni, R. Dalpozzo,
Chem. Soc. Rev. 2010, 39, 4449–4465; j) M. Shiri, Chem. Rev. 2012,
112, 3508–3549.
[3] a) C. A. Demerson, G. Santroch, L. G. Humber, J. Med. Chem. 1975,
18, 577–580; b) C. Farina, S. Gagliardi, P. Misiano, P. Celestini, F.
Zunino, PCT Int. Appl., WO 2005105213A2 20051110, 2005.
[4] a) J. An, N.-J. Chang, L.-D. Song, Y.-Q. Jin, Y. Ma, J.-R. Chen, W.-J.
Xiao, Chem. Commun. 2011, 47, 1869–1871; b) S. J. Gharpure,
A. M. Sathiyanarayanan, Chem. Commun. 2011, 47, 3625–3627.
[5] G. Cera, S. Piscitelli, M. Chiarucci, G. Fabrizi, A. Goggiamani, R. S.
Ramꢅn, S. P. Nolan, M. Bandini, Angew. Chem. 2012, 124, 10029–
10033; Angew. Chem. Int. Ed. 2012, 51, 9891–9895.
[22] R. A. Sneen, Acc. Chem. Res. 1973, 6, 46–53.
Received: December 30, 2012
Published online: && &&, 0000
[6] Fore recent reviews on the use of p alcohols in organic synthesis,
see: a) J. Muzart, Tetrahedron 2005, 61, 4179–4212; b) Y. Tamaru,
&
&
4
Chem. Asian J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

COMMUNICATION
Domino Reactions
Michel Chiarucci, Elia Matteucci,
Gianpiero Cera, Giancarlo Fabrizi,
Marco Bandini*
&&&& — &&&&
New Entry to Polycyclic Fused Indoles
via Gold(I)-catalyzed Cascade Reac-
tion
The gold standard: A gold-catalyzed
cascade reaction sequence for the
preparation of polycyclic fused indole
cores exploits the ready availability
and chemical flexibility of propargylic
alcohols. The desired tri- or tetracyclic
compounds are obtained in good yields
with water as the only stoichiometric
by-product.
5
&
&
Chem. Asian J. 2013, 00, 0 – 0
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
These are not the final page numbers! ÞÞ