Oxidative Dearomatization of 4,5,6,7-Tetrahydro-1H-indoles Obtained by Metal- and Solvent-Free Thermal 5-endo-dig Cyclization: The Route to Erythrina and Lycorine Alkaloids

Article abstract of DOI:10.1002/chem.201600273

A facile one-pot approach based on a thermally induced metal- and solvent-free 5-endo-dig cyclization reaction of the amino propargylic alcohols in combination with Dess-Martin periodinane-promoted oxidative dearomatization of 4,5,6,7-tetrahydroindole intermediates provides an efficient and robust access to 5,6-dihydro-1H-indol-2(4H)ones. Green, relatively mild and operationally simple characteristics of the synthetic sequence are the major advantages, which greatly amplify the developed methodology. The utility of obtained indolones as unified key precursors is demonstrated by the application of these products to the formal total syntheses of a whole pleiad of Erythrina- and Lycorine-type alkaloids, namely (±)-erysotramidine, (±)-erysotrine, (±)-erythravine, (±)-γ-lycorane, and abnormal erythrinanes (±)-coccoline and (±)-coccuvinine.

Full text of DOI:10.1002/chem.201600273

DOI: 10.1002/chem.201600273
Full Paper
&
Organic Synthesis
Oxidative Dearomatization of 4,5,6,7-Tetrahydro-1H-indoles
Obtained by Metal- and Solvent-Free Thermal 5-endo-dig
Cyclization: The Route to Erythrina and Lycorine Alkaloids
Ivan A. Andreev, Nina K. Ratmanova, Anton M. Novoselov, Dmitry S. Belov, Irina F. Seregina,
and Alexander V. Kurkin*[a]
Dedicated to the academician N. S. Zefirov on the occasion of his 80th birthday
Abstract: A facile one-pot approach based on a thermally in-
duced metal- and solvent-free 5-endo-dig cyclization reaction
of the amino propargylic alcohols in combination with
Dess–Martin periodinane-promoted oxidative dearomatiza-
tion of 4,5,6,7-tetrahydroindole intermediates provides an ef-
ficient and robust access to 5,6-dihydro-1H-indol-2(4H)ones.
Green, relatively mild and operationally simple characteristics
of the synthetic sequence are the major advantages, which
greatly amplify the developed methodology. The utility of
obtained indolones as unified key precursors is demonstrat-
ed by the application of these products to the formal total
syntheses of a whole pleiad of Erythrina- and Lycorine-type
alkaloids, namely (Æ)-erysotramidine, (Æ)-erysotrine, (Æ)-er-
ythravine, (Æ)-g-lycorane, and abnormal erythrinanes (Æ)-
coccoline and (Æ)-coccuvinine.
lol and (Æ)-chuangxinmycin,[4a] (À)-goniomitine,[4b] arcyria-
cyanin A,[4c] 6,7-secoagroclavine,[4d] psilocin and psilocybin[4e]
via the intermediary of 4-hydroxyindole[4f,g] and, finally, (Æ)-g-ly-
corane.[4h]
Introduction
Erythrina- and Lycorine-type tetracyclic alkaloids constitute two
vast classes of natural products. Members of these families dis-
play remarkable biological activities and a number of pharma-
cological effects; Erythrina scaffolds exhibit sedative, hypnotic,
hypotensive, antiasthmatic, spasmolytic, antineoplastic, diuret-
ic, and central nervous system (CNS) activities,[1] while Lycorine-
type alkaloids exhibit anticancer, acetylcholinesterase (AChE)
inhibitory, antiviral, antibacterial, anticonvulsant and antide-
pressant activities.[2] An equal if not exciding interest has been
attracted to their challenging polycyclic cores, as a plethora of
synthetic approaches have been designed over the past
50 years.
According to the literature, both natural product classes can
be retrosynthetically traced back to a common intermediate,
5,6-dihydro-1H-indol-2(4H)one 3. In the case of a Lycorine
framework: n=1, X=Hal[3a,b] and with respect to an Erythrina
core: n=2, X=H[3a,c–e] (Scheme 1). These indolones, in turn,
could be envisioned as oxygenated forms of 4,5,6,7-tetrahy-
droindoles 2, which are extensively employed as highly func-
tionalized indole precursors—valuable intermediates in the nu-
merous syntheses of alkaloids and drugs, such as S-(À)-pindo-
[a] I. A. Andreev, N. K. Ratmanova, A. M. Novoselov, D. S. Belov,
Dr. I. F. Seregina, Dr. A. V. Kurkin
Department of Chemistry, Lomonosov Moscow State University
Leninsky Gory 1/3, 119991 Moscow (Russia)
E-mail: kurkin@direction.chem.msu.ru
Scheme 1. Application of 4,5,6,7-tetrahydro-1H-indoles as valuable inter-
mediates in a variety of total syntheses and their possible transformation to
5,6-dihydro-1H-indol-2(4H)ones—key precursors to both Erythrina- and Ly-
corine-type alkaloids.
Supporting information and the ORCID identification number(s) for the au-
thor(s) of this article are available on the WWW under http://dx.doi.org/
10.1002/chem.201600273.
Chem. Eur. J. 2016, 22, 7262 – 7267
7262
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Despite the fact that 4,5,6,7-tetrahydroindoles can be readily
converted into indoles, it was rather unclear if they could also
be easily transformed into the requisite indolones. Neverthe-
less, our synthetic efforts en route to the alkaloids mentioned
above should have started from the effective assembly of the
required heterocycles 2.
Table 1. Synthesis of 4,5,6,7-tetrahydroindoles 2 by metal- and solvent-
free thermally induced 5-endo-dig cyclization.[a]
Results and Discussion
Entry Amino
propargylic
n (0–2); R; R’
4,5,6,7-
Tetra-
hydro-
indole 2
Yield
[%][b]
Recently, we have developed a two-step synthetic methodolo-
gy that leads to 4,5,6,7-tetrahydroindoles possessing a wide-
range of substituents both at C2 or nitrogen atoms (including
chiral moieties).[5a,b] This approach bears a general character
and is readily scalable to generate sufficient amounts of mate-
rial on a gram-scale in good to excellent yields (Scheme 2, the
alcohol 1
1
2
3
4
5
6
7
8
1a
1a’
1b
1b’
1c
1d
1e
1 f
1; Ph; Ph
1; Vyn; Ph
1; Ph; H
1; Vyn; H
1; Ph; TMS
0; 4-F-Ph; H
0; H; H
2a
73 (66)
75 (87)
90 (54)
77 (82)
0[d] (0)[e]
81
26[g]
83
94
88
2a’
2b
2b’
2c[c]
2d[f]
2e
1; CONH2; H
2; 3,4-di-OMe-Ph; H
2; 4-OMe-Ph; H
2; 3,4-OCH2O-Ph; H
1; 6-Br-3,4-OCH2O-Ph;
H
2 f
9
1g
1h
1i
2g
2h
2i
2j[h]
10
11
12
85
58
1j
13
1k
2; 2-I-4,5-di-OMe-Ph; H 2k[h]
68
[a] N2 atm, 1 h, 160–1708C (internal oil bath temperature). These are gen-
eral conditions, unless otherwise stated. [b] Isolated yield after flash
column chromatography. The yield in parentheses was obtained perform-
ing cyclization under previously disclosed conditions:[5a,c] Pd(OAc)2
(1 mol%), MeCN (0.1m), reflux, 1 h. [c] Reaction conditions: >1 h;
>2008C. [d] Only unchanged 1c was obtained. [e] Only desilylated tetra-
hydroindole 2b was obtained. [f] The reaction time was increased to 3 h
at 1808C. [g] Deviations from the general conditions both in reaction
temperature or time led to decreased yields or complete decomposition.
[h] The reaction time was increased to 2 h.
Scheme 2. Synthetic pathways leading to 4,5,6,7-tetrahydroindoles. dba=di-
benzylideneacetone.
upper path). Alternatively, tetrahydroindoles 2 can be obtained
directly through Pd(OAc)2-promoted 5-endo-dig cyclization
(Scheme 2, the middle path).[5a,c] Unfortunately, crude yields of
such transformations in the case of C2-unsubstituted com-
pounds (R’=H) are not excellent. Moreover, the overall yields
are even worse when the low oxidative stabilities of these
labile substances and the following chromatographic purifica-
tions are taken into account.
All of the starting amino alcohols 1 were produced according
to previously reported procedures on a gram-scale.[6]
An interesting observation was noticed during the pro-
longed storage of amino propargylic alcohol 1b (R=Bn, R’=
H), both as a solution in CH2Cl2 or neat. TLC and H NMR spec-
According to the table above, the majority of the 4,5,6,7-tet-
rahydroindoles were obtained in good to excellent yields. In
the case of N-benzyl-substituted compounds, the outcomes of
the thermal cyclization reactions were better than those using
palladium (Table 1, entries 1 and 3), while N-allyl-substituted
compounds demonstrated contrary results (entries 2 and 4).
The low reactivity of silyl amino propargylic alcohol 1c
(entry 5) compared to other alkynes was anticipated and in full
agreement with previous reports.[7] The low cyclization yield of
1e (entry 7) was also expected as is always the case with N-un-
substituted amino propargylic alcohols.[5a] The moderate yield
of 2j (entry 12) could be attributed to steric factors. Neverthe-
less, the reaction demonstrated general character and didn’t
require any solvents or metal catalysts and produced water as
a major byproduct.
1
troscopic analyses of the samples showed considerable
amounts of tetrahydroindole 2b (R=Bn, R’=H). Intrigued by
this fact, we wondered if this transformation could be applica-
ble on a preparative scale and whether the reaction conditions
could be mild enough to decrease reaction times from months
to minutes to furnish 4,5,6,7-tetrahydroindoles 2 in appreciable
yields (Scheme 2, the lower path).
Indeed, simple heating of neat 1b in a vial at 1408C (exter-
nal oil bath temperature) produced water in the first few mi-
nutes. After approximately 30 min, the TLC control showed
almost complete consumption of the starting material. Further
verification of the conditions revealed the necessity of exclud-
ing air from the reaction media. Generally, to achieve full con-
version on the majority of the amino propargylic alcohols 1,
temperatures of 160–1708C were employed for a period of
1 to 3 h. The results of the novel thermally induced 5-endo-dig
cyclization reactions on various substrates are listed in Table 1.
Although the cyclization proceeded without a metal, we
wanted to exclude the influence of transition-metal salts on
the reaction pathway. Trace metal analysis was performed by
ICP-MS of aliquots (1–2 mg) taken after the completion of the
reactions, which showed that the levels of major transition
Chem. Eur. J. 2016, 22, 7262 – 7267
www.chemeurj.org
7263
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
metals (Cu, Zn, Pd, Ag, Pt, Au) capable of promoting 5-endo-
dig cyclization reactions were lower than the detection limit.[8]
Other metals were detected in units of ppm, emphasizing and
rigorously demonstrating the metal-free nature of designed
cyclization.[9] Moreover, the reaction takes place only above
1408C, whereas in the presence of PdII the cyclization proceed-
ed in refluxing acetonitrile (~808C) or even Et2NH (~558C) at
relatively small concentrations (up to 1 mm of catalyst).[5a–c]
Taking into the account that the thermally induced cyclization
is also solvent-free, the concentration of both substrate and
adventitious metals should have risen tremendously forcing
the reaction to proceed at much lower temperatures. While
these facts are only speculation, they additionally solidify our
initial hypothesis of the absence of trace metal impurities,
which could interfere with the course of the thermal cycliza-
tion.
tioned work of Smith et al.,[11a] oxidation of 2-substituted N-
alkyl pyrroles under similar conditions with 2.5 equivalents of
DMP afforded 5-hydroxy-g-lactam, while the obtainment of the
elimination product required an additional treatment of the in-
termediate reaction mixture with BF3·OEt2/Et3SiH. In our case,
the product 3 was formed without any additional manipula-
tions. The plausible explanation of this fact could be due to
the liberation of AcOH during the oxidation, which was
enough to trigger the elimination process. Moreover, the iden-
1
tity of H NMR spectra of the crude and isolated 3b further
supported our assumption that the elimination took place ex-
actly during the oxidation, but not upon the chromatographic
purification on silica gel. Though a minor probability of dehy-
droxylation during the work-up procedure persisted, the TLC
control of the reaction mixture before and after the basic
aqueous treatment eliminated the final uncertainty as the
major product spot corresponded with 3b completely. While
the exact mechanism of the oxidation is not elaborated yet,
the one proposed by Smith and co-workers[11a] seems to be
reasonable and applicable in our instance with the exception
of the product outcome. With the optimized one-pot cycliza-
tion/oxidation conditions in hand, we then examined the
scope of this novel sequence on amino propargylic alcohols
1 f–k (Table 2).
After having established permanent and reliable access to 2-
unsubstituted tetrahydroindoles 2, the next goal was to find
suitable oxidation conditions for these labile substrates.[10]
Based on a study published by Smith et al.,[11] tetrahydroin-
doles 2 were envisioned as suitable starting materials to ach-
ieve oxidized pyrroles, oxypyrrolinones 3, as the key intermedi-
ates in the alkaloid syntheses. These intermediates could be
easily transformed into target tetracyclic cores by either radi-
cal- or cation-promoted intramolecular cyclization reactions.
Aiming to develop a one-pot approach to minimize losses,
tetrahydroindole 2b was subjected to oxidation directly upon
thermal cyclization without purification. To our delight, the
portionwise treatment of crude 2b in CH2Cl2 with Dess–Martin
periodinane (DMP) at 08C afforded the desired 5,6-dihydro-1H-
indol-2(4H)one 3b[12] in impure form in 56% yield after chro-
matographic purification. Optimization of the reaction condi-
tions revealed a number of features. The reaction temperature
during the slow portionwise addition of DMP should be ap-
proximately 08C (especially during the scale-up of experi-
ments) because overheating during the exothermic oxidation
reaction considerably affects the purity of the product. Revers-
ing the order of addition does not alter the yield, but it is
much less convenient than the direct addition. The final con-
centration of DMP in the reaction mixture could be as large as
1m, and lower concentrations result in slightly decreased
yields. Additives such as MgSO4, NaHCO3, and trifluoroacetic
acid (TFA) particularly decrease the effectiveness of the oxida-
tion reaction. Finally, the application of chromatographically
isolated 2b is unnecessary and also has a negative influence
on the reaction outcome. This observation could be explained
by the presence of one equivalent of water in the crude cycli-
zation product, which results in the formation of acetoxyiodi-
nane oxide upon the reaction with DMP. Acetoxyiodinane
oxide is well known to be more active and efficient in promot-
ing oxidation.[13] Consequently, varying the solvent from di-
chloromethane to acetonitrile and finally to benzotrifluoride
(PhCF3)[14] resulted in improvements in both of the purity and
yield (68%) of 3b.[15]
Table 2. One-pot syntheses of 5,6-dihydro-1H-indol-2(4H)ones 3 by suc-
cessive metal- and solvent-free thermally induced 5-endo-dig cyclization
followed by Dess–Martin periodinane-promoted oxidation of the pyrrole
cores.[a]
Entry Amino
propargylic
R’=H; n (1 or 2); R
5,6-Dihydro- Yield [%][b]
1H-indol-
alcohol 1
2(4H)one 3
1
2
3
4
5
6
7
1b
1 f
1g
1h
1i
1; Ph
1; CONH2
3b
3 f
3g
3h
3i
68
0[c]
2; 3,4-di-OMe-Ph
2; 4-OMe-Ph
2; 3,4-OCH2O-Ph
1; 6-Br-3,4-OCH2O-Ph 3j
2; 2-I-4,5-di-OMe-Ph 3k
79[d] (20)
78[d] (60)
50
1j
1k
47[d] (30)
48
[a] N2 atm, 1 h, 160–1708C (internal oil bath temperature). These were
general conditions, unless otherwise stated. After cooling to r.t., the
crude 2 was dissolved in PhCF3 and treated portionwise with DMP
(2.5 equiv) at 08C. [b] Isolated two-step yield after flash column chroma-
tography. The yield in parentheses was obtained after additional recrys-
tallization from the appropriate solvent (see the Supporting Information).
[c] DMF (20 vol%) was added to increase the solubility of 2 f. A complex
mixture of products was obtained. [d] Average yield of the reaction.
The treatment of crude 2 f with DMP produced a complex
mixture of products, which contained no required product ac-
cording to LCMS analysis. Presumably, primary alkyl amide re-
acted with the polyvalent iodine reagent through a Hofmann
The obtainment of 5,6-dihydro-1H-indol-2(4H)ones 3 de-
served a special attention as no formation of aroyloxy- or hy-
droxy- product 3’ was detected. According to the above-men-
Chem. Eur. J. 2016, 22, 7262 – 7267
www.chemeurj.org
7264
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
rearrangement even at 08C.[16] Nevertheless, the oxidation of
the other tetrahydroindoles proceeded smoothly and furnished
the desired indolones 3g–k in moderate to good yields. The
reasons for the poor scalability of DMP-oxidation on a scale
larger than 10 mmol are the subject of further investigations.
The acquirement of 3g, 3i, and 3j completed the formal
total syntheses of a whole pleiad of alkaloid structures,[17]
namely (Æ)-3-demethoxy-1,2-dihydroerysotramidine and (Æ)-3-
conditions, whereas numerous elaborated cyclization ap-
proaches led to unsatisfying results.[15,22] It should be noted
that this cyclization reaction eliminates the necessity of an ad-
ditional electron-donating directing group that was employed
in the first total synthesis of (Æ)-coccoline by Ju-ichi et al., thus
obviating at least two low-yielding steps for the deprotective
cleavage of this auxiliary.[21a] The obtained tetracyclic erythrina
skeleton 5h was smoothly converted into a,b,g,d-unsaturated
diene amide 6h upon treatment with DBU in refluxing p-
xylene for approximately 20 h in 76% yield according to previ-
ously disclosed conditions.[3d]
demethoxy-1,2-dihydroerythraline,[3e]
(Æ)-erysotramidine,[3d]
which could be further transformed into (Æ)-erysotrine[18a] and
(Æ)-erythravine,[18b] and finally, (Æ)-g-lycorane[3b] (Scheme 3). To
achieve another formal synthesis of the rearranged alkaloid tet-
rahydroerythraline, hexahydroapoerysopine dimethyl ether[19]
indolone 3k was subjected to intramolecular ligand-free Heck
cyclization under Jeffery conditions.[20] Unfortunately, regard-
less of variations in reaction conditions, erythrinane 5k (56%
yield) was obtained as an inseparable mixture of unassigned
After having established a robust and facile method to
access the desired tetracyclic framework, only one simple but
not easy and unevident step, namely the oxidation of the C3-
allylic position, was sufficient for the completion of the formal
total syntheses of the abnormal Erythrina-type alkaloids (Æ)-
coccuvinine and (Æ)-coccoline.[21] Reproducing Padwa’s condi-
tions[3d] exploited in the synthesis of (Æ)-erysotramidine verba-
tim proved to be ineffective in terms of the tremendously long
reaction times and low conversion of the starting material.
Thus, exposing 6h to 10 equivalents of SeO2 in a refluxing mix-
ture of dioxane and formic acid[3d] for 8 h resulted in <3%
conversion of the starting material according to LCMS analysis.
To improve both conversion and reaction times, we performed
an extensive screening of reagents and conditions generally
employed for allylic oxidation in the field of total synthesis.[15]
This exhaustive optimization showed that the employment of
any alternative reagents was not helpful. Thus, we returned to
utilizing selenium dioxide oxidation to carefully revise and im-
prove the literature conditions. Several further experiments es-
tablished the scope and limitations of the process and demon-
strated that the temperature should not greatly exceed 1008C,
while the reaction rate was majorly increased as the concentra-
tion of 6h rose. Conducting solid-phase ball milling-assisted
oxidation disappointingly led to undesired decomposition
1
regioisomeric alkenes in ~4:1 ratio (according to H NMR and
LCMS analyses) instead of the expected apoerysopine core in-
termediate 5k’.
To achieve another formal syntheses[21] of (Æ)-coccoline and
(Æ)-coccuvinine, we focused our attention on the transforma-
tion of indolone 3h into the tetracyclic intermediate 6h
(Scheme 4). Bromocyclization under conditions developed by
Padwa[3d] did not result in the required cyclization product 5h.
Instead, an inseparable mixture of vinyl bromide 4h and bro-
mohydrin 4h’(desMe) was obtained in 77% crude yield. An ex-
tensive search for conditions leading to the requisite 5h re-
vealed that a two-step procedure could be successfully uti-
lized. First, indolone 3h was converted into either vinyl bro-
mide 4h or methoxy bromide 4h’ according to the judicious
choice of solvent, temperature and even the order of addition
of NBS. Next, 4h and 4h’ were both subjected to acid-mediat-
ed cyclization; 4h’ in neat trifluoromethanesulfonic acid afford-
ed the best results on a gram-scale under very mild reaction
Scheme 3. 5,6-Dihydro-1H-indol-2(4H)ones 3 as the key intermediates en route to Erythrina- and Lycorine-type tetracyclic alkaloids.
Chem. Eur. J. 2016, 22, 7262 – 7267 www.chemeurj.org ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7265

Full Paper
Scheme 4. Formal syntheses of (Æ)-coccoline and (Æ)-coccuvinine: a) epoxide (1m), 4-MeO-phenethyl amine (2 equiv), LiClO4 (1.5 equiv), MeCN, 55–608C,
16 h; b) 160–1708C, neat, N2 atm, 1 h; c) 2h (without isolation), PhCF3, Dess–Martin periodinane (2.5 equiv, 1m final concentration) portionwise addition at
08C!r.t., 1 h; d) NBS (1.1 equiv), MeCN (0.1m), r.t., 1 h; e) for 4h: NBS (1.1 equiv), MeOH (0.1m), 25–308C, 30 min, 4h/4h’ꢀ10:1; for 4h’: NBS (1.1 equiv),
MeOH (0.1m), 08C!r.t., 30 min, 4h’/4h > 100:1; f) 4h or 4h’ (0.1m), HClO4 (70 wt.-%, aq.), r.t., 30 min; g) 4h’ (0.33m), TfOH, À308C!r.t., 15-20 min; h) 5h
(0.2m), DBU (10 equiv), p-xylene, N2 atm, reflux, 20 h; i) 6h (0.5m in dioxane) in a sealed vial with a rubber septum, N2 atm, SeO2 (portionwise addition,
1 equiv per hour starting from 2 equiv), AcOH (10 equiv), 110 8C (oil bath), 12–14 h; j) see ref. [19b]; k) see ref. [19b]. DBU=1,8-diazabicyclo[5.4.0]undec-7-ene;
DMP=1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benziodoxol-3-(1H)-one (Dess–Martin periodinane); NBS=N-bromosuccinimide.
products, though the reaction profile at the beginning was
promising. Finally, we established an optimal set of conditions.
The allylic oxidation was performed in dioxane in a sealed vial
under a nitrogen atmosphere at a concentration of 6h as high
as 0.5m at 110 8C (internal oil bath temperature). The highest
conversion (approx. 88%) was achieved after 12 h of a portion-
wise (1 equiv every hour starting from 2 equiv) addition of
SeO2 and afforded 7h in 17% yield after flash chromatography.
The spectral characterization data for 7h was in accordance
with that obtained by Sano and co-workers.[21b,23] Thus, the
formal total syntheses of (Æ)-coccoline and (Æ)-coccuvinine
were accomplished in only seven and eight steps, respectively,
from the commercially available epoxide 1-ethynyl-7-oxabicy-
clo[4.1.0]heptane, which provided the most facile access to
these abnormal erythrinanes to date.
ic oxidation conditions for 6h will not go unnoticed, making
the approach based on the late-stage installation of OH-
groups to C3-positions much more viable and applicable for
the final assembly of other erythrinanes.
Experimental Section
Experimental procedures, product characterization data, and
copies of the 1H and 13C NMR spectra are presented in the Sup-
porting Information.
Acknowledgements
We acknowledge and appreciate the financial support of the
Russian Foundation for Basic Research (RFBR), Russia (project
nos. 14-03-31685, 14-03-01114). We thank M. A. Andreeva for
drawing the coral tree for the graphical abstract.
Conclusion
To summarize, we successfully designed a novel one-pot two-
step approach leading to 5,6-dihydro-1H-indol-2(4H)ones 3
based on green and uncatalyzed thermally induced cyclization
of amino propargylic alcohols 1 followed by DMP oxidation of
4,5,6,7-tetrahydroindole 2 intermediates. First of all, we have
made another use of tetrahydroindoles as worth-while synthet-
ic intermediates. On the other hand, we have extended the
range of applications of pyrroles as suitable substrates for the
oxidative dearomatization reactions, which culminate in syn-
thetically useful intermediates[24] and shown the unique oxidiz-
ing power of hypervalent iodine reagents, which suggests that
they could be reasonable metal-free alternatives to conven-
tional reagents.[25] The array of synthesized indolones 3, in
turn, served as unified key intermediates in the total syntheses
of alkaloids representing the Erythrina and Lycorine families.
The facile obtainment of 7h was the missing link en route to
(Æ)-coccoline and (Æ)-coccuvinine, which further extended our
approach to synthesizing abnormal Erythrina alkaloids. We
hope that our exhaustive work on the optimization of the allyl-
Keywords: 4,5,6,7-tetrahydroindoles
chemistry · thermally induced cyclization · total synthesis
·
alkaloids
·
green
[1] J. F. Grove, E. Reimann, S. Roy in Progress in the Chemistry of Organic
Natural Products, Vol. 88 (Eds.: W. Herz, H. Falk, G. W. Kirby), Springer,
Wien, 2007, pp. 2–62.
[2] M. He, C. Qu, O. Gao, X. Hu, X. Hong, RSC Adv. 2015, 5, 16562–16574.
[3] a) J. H. Rigby, M. Qabar, J. Am. Chem. Soc. 1991, 113, 8975–8976; b) J.
Cassayre, S. Z. Zard, Synlett 1999, 501–503; c) J. Cassayre, B. Quiclet-
Sire, J.-B. Saunier, S. Z. Zard, Tetrahedron Lett. 1998, 39, 8995–8998;
d) H. I. Lee, M. P. Cassidy, P. Rashatasakhon, A. Padwa, Org. Lett. 2003, 5,
5067–5070; e) A. Mondon, J. Zander, H.-U. Menz, Justus Liebigs Ann.
Chem. 1963, 667, 126–140.
[4] a) H. Ishibashi, S. Akamatsu, H. Iriyama, K. Hanaoka, T. Tabata, M. Ikeda,
Chem. Pharm. Bull. 1994, 42, 271–276; b) C. L. Morales, B. L. Pagenkopf,
Org. Lett. 2008, 10, 157–159; c) M. Murase, K. Watanabe, T. Yoshida, S.
Tobinaga, Chem. Pharm. Bull. 2000, 48, 81–84; d) N. Hatanaka, O. Ozaki,
M. Matsumoto, Tetrahedron Lett. 1986, 27, 3169–3172; e) O. Shirota, W.
Hakamata, Y. Goda, J. Nat. Prod. 2003, 66, 885–887; f) D. B. Repke, W. J.
Ferguson, D. K. Bates, J. Heterocycl. Chem. 1977, 14, 71–74; g) M. Matsu-
moto, N. Watanabe, Heterocycles 1984, 22, 2313–2316; h) S. R. Angle,
J. P. Boyce, Tetrahedron Lett. 1995, 36, 6185–6188.
Chem. Eur. J. 2016, 22, 7262 – 7267
www.chemeurj.org
7266
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[5] a) I. A. Andreev, D. S. Belov, A. V. Kurkin, M. A. Yurovskaya, Eur. J. Org.
Chem. 2013, 649–652; b) I. A. Andreev, D. Manvar, M. L. Barreca, D. S.
Belov, A. Basu, N. L. Sweeney, N. K. Ratmanova, E. R. Lukyanenko, G.
Manfroni, V. Cecchetti, D. N. Frick, A. Altieri, N. Kaushik-Basu, A. V.
Kurkin, Eur. J. Med. Chem. 2015, 96, 250–258; c) I. A. Andreev, I. O. Ryzh-
kov, A. V. Kurkin, M. A. Yurovskaya, Chem. Heterocycl. Compd. 2012, 48,
715–719; d) D. S. Belov, N. K. Ratmanova, I. A. Andreev, A. V. Kurkin,
Chem. Eur. J. 2015, 21, 4141–4147.
[17] For recent approaches to Erythrina scaffolds, see the following referen-
ces: a) D. Kalaitzakis, T. Montagnon, E. Antonatou, G. Vassilikogiannakis,
Org. Lett. 2013, 15, 3714–3717; b) D. Kalaitzakis, T. Montagnon, E. Anto-
natou, N. Bardají, G. Vassilikogiannakis, Chem. Eur. J. 2013, 19, 10119–
10123; c) D. Kalaitzakis, A. Kouridaki, D. Noutsias, T. Montagnon, G. Vas-
silikogiannakis, Angew. Chem. Int. Ed. 2015, 54, 6283–6287; Angew.
Chem. 2015, 127, 6381–6385; d) C. L’Homme, M.-A. MØnard, S. Canesi,
J. Org. Chem. 2014, 79, 8481–8485; e) D. Mostowicz, M. Dygas, Z.
Kałuz˙a, J. Org. Chem. 2015, 80, 1957–1963; f) M. Paladino, J. Zaifman,
M. A. Ciufolini, Org. Lett. 2015, 17, 3422–3425; g) M. He, C. Qu, B. Ding,
H. Chen, Y. Li, G. Qiu, X. Hu, X. Hong, Eur. J. Org. Chem. 2015, 3240–
3250.
[6] For representative general synthetic procedures, see the Supporting In-
formation and references listed in ref. [4].
[7] S. R. Walker, M. L. Czyz, J. C. Morris, Org. Lett. 2014, 16, 708–711.
[8] The complete analysis of the ICP-MS data is presented in the support-
ing information. Generally the detection limit was lower than 2 ppm (in
some cases even lower by 1–2 orders). See the Supporting Information
to find out the precise details on every measured metal.
[18] a) T. Sano, J. Toda, N. Kashiwaba, T. Ohshima, Y. Tsuda, Chem. Pharm.
Bull. 1987, 35, 479–500; b) M. H. Sarragiotto, P. A. da Costa, A. J. Mar-
saioli, Heterocycles 1984, 22, 453–456.
[9] For recent examples of thermally induced transition metal-free cycliza-
tions, see the following references: a) D. Chandra Mohan, S. N. Rao, S.
Adimurthy, J. Org. Chem. 2013, 78, 1266–1272; b) D. Chandra Mohan,
N. B. Sarang, S. Adimurthy, Tetrahedron Lett. 2013, 54, 6077–6080;
c) S. T. Heller, T. Kiho, A. R. H. Narayan, R. Sarpong, Angew. Chem. Int. Ed.
2013, 52, 11129–11133; Angew. Chem. 2013, 125, 11335–11339; d) P. R.
Golubev, A. S. Pankova, M. A. Kuznetsov, J. Org. Chem. 2015, 80, 4545–
4552.
[10] Though chromatographically stable, almost all of the 4,5,6,7-tetrahy-
droindoles were rapidly oxidized on air. Thus, the acidophobic proper-
ties of these compounds required the development of a one-pot ap-
proach to generate 5,6-dihydro-1H-indol-2(4H)ones. See the Supporting
Information for comparison of the 1H NMR spectra for 2j in slightly
acidic CDCl3 and [D6]acetone.
[11] a) J. K. Howard, C. J. T. Hyland, J. Just, J. A. Smith, Org. Lett. 2013, 15,
1714–1717; For recent examples of pyrrole oxidative dearomatization
reactions that are most similar to our study, see b) A. S. Demir, F. Aydo-
gan, I. M. Akhmedov, Tetrahedron: Asymmetry 2002, 13, 601–605; c) D.
Lubriks, I. Sokolovs, E. Suna, Org. Lett. 2011, 13, 4324–4327; d) P. Kan-
charla, W. Lu, S. M. Salem, J. X. Kelly, K. A. Reynolds, J. Org. Chem. 2014,
79, 11674–11689.
[12] The analytical data were consistent with those reported in the follow-
ing papers: a) M. Giannangeli, L. Baiocchi, J. Heterocycl. Chem. 1982, 19,
891–895; b) H. Ishibashi, M. Higuchi, M. Ohba, M. Ikeda, Tetrahedron
Lett. 1998, 39, 75–78; c) J. Cassayre, B. Quiclet-Sire, J.-B. Saunier, S. Z.
Zard, Tetrahedron 1998, 54, 1029–1040.
[19] a) H. Iida, T. Takarai, C. Kibayashi, J. Chem. Soc. Chem. Commun. 1977,
644–645; b) H. Iida, T. Takarai, C. Kibayashi, J. Org. Chem. 1978, 43,
975–979; c) Y.-M. Zhao, P. Gu, Y.-Q. Tu, H.-J. Zhang, Q.-W. Zhang, C.-A.
Fan, J. Org. Chem. 2010, 75, 5289–5295.
[20] a) T. Jeffery, Tetrahedron 1996, 52, 10113–10130; b) J. H. Rigby, R. C.
Hughes, M. J. Heeg, J. Am. Chem. Soc. 1995, 117, 7834–7835; c) A.
Padwa, M. A. Brodney, S. M. Lynch, J. Org. Chem. 2001, 66, 1716–1724.
[21] a) M. Ju-ichi, Y. Fujitani, Y. Ando, Chem. Pharm. Bull. 1981, 29, 396–401;
b) T. Sano, J. Toda, N. Maehara, Y. Tsuda, Can. J. Chem. 1987, 65, 94–98;
c) D. S. Bhakuni, S. Jain, Tetrahedron 1980, 36, 3107–3114.
[22] Y. Tsuda, Y. Sakai, A. Nakai, M. Kaneko, Y. Ishiguro, K. Isobe, J. Taga, T.
Sano, Chem. Pharm. Bull. 1990, 38, 1462–1472.
[23] However, the authors did not provide some of the signals in the ali-
phatic region of the 1H NMR spectrum, and the 13C NMR spectrum was
completely absent.
[24] For more examples of pyrrole oxidation, see the following review and
references therein: J. K. Howard, K. J. Rihak, A. C. Bissember, J. A. Smith,
Chem. Asian J. 2016, 11, 155–167.
[25] For recent applications of iodine(V) reagents, see the following referen-
ces: a) K. C. Nicolaou, P. S. Baran, Y.-L. Zhong, K. Sugita, J. Am. Chem.
Soc. 2002, 124, 2212–2220, and 3 following parts; b) U. Ladziata, V. V.
Zhdankin, ARKIVOC (Gainesville, FL, U.S.) 2006, 9, 26–58. For iodine(III)
reagents: c) R. Narayan, S. Manna, A. P. Antonchick, Synlett 2015, 1785–
1803; d) S. Murarka, A. P. Antonchick, Top. Curr. Chem. 2015, 356-361, 1–
30; e) R. Narayan, K. Matcha, A. P. Antonchick, Chem. Eur. J. 2015, 21,
14678–14693.
[13] S. D. Meyer, S. L. Schreiber, J. Org. Chem. 1994, 59, 7549–7552.
[14] A. Ogawa, D. P. Curran, J. Org. Chem. 1997, 62, 450–451.
[15] For full and detailed optimization, see the Supporting Information.
[16] a) A. S. Radhakrishna, M. E. Parham, R. M. Riggs, G. M. Loudon, J. Org.
Chem. 1979, 44, 1746–1747; b) R. M. Moriarty, C. J. Chany II, R. K. Vaid,
O. Prakash, S. M. Tuladhar, J. Org. Chem. 1993, 58, 2478–2482.
Received: January 20, 2016
Published online on April 14, 2016
Chem. Eur. J. 2016, 22, 7262 – 7267
www.chemeurj.org
7267
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim