Stereoselective Cascade Cyclizations with Samarium Diiodide to Tetracyclic Indolines: Precursors of Fluorostrychnines and Brucine

Article abstract of DOI:10.1002/chem.201900087

A series of γ-indolylketones with fluorine, cyano or alkoxy substituents at the benzene moiety was prepared and subjected to samarium diiodide-promoted cyclization reactions. The desired dearomatizing ketyl cascade reaction forming two new rings proceeded

Full text of DOI:10.1002/chem.201900087

A Journal of
Accepted Article
Title: Stereoselective Cascade Cyclizations with Samarium Diiodide
to Tetracyclic Indolines – Precursors of Fluorostrychnines and
Brucine
Authors: Hans-Ulrich Reissig, Christine Beemelmanns, Dominik Nitsch,
and Christoph Bentz
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201900087
Link to VoR: http://dx.doi.org/10.1002/chem.201900087
Supported by

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
Stereoselective Cascade Cyclizations with Samarium Diiodide to
Tetracyclic Indolines – Precursors of Fluorostrychnines and
Brucine
Christine Beemelmanns,[a,b] Dominik Nitsch, Christoph Bentz, Hans-Ulrich Reissig*
[a]
[a]
[a]
Dedicated to Professor Athanassios Giannis on the occasion of his 65th birthday
Abstract: A series of γ-indolylketones with fluorine, cyano or alkoxy
substituents at the benzene moiety was prepared and subjected to
samarium diioidide-promoted cyclizations. The desired dearomatizing
ketyl cascade reaction forming to new rings proceeded in all cases
with high diastereoselectivity, but with differing product distribution. In
most cases, the desired annulated tetracyclic compounds were
obtained in moderate to good yields, but as second product tetracyclic
spirolactones were isolated in up to 29% yield. The reaction rate was
influenced by the substituents at the benzene moiety of the substrate
as expected, with electron-accepting groups accelerating and
electron-donating groups decelerating the cyclization process. In case
of a difluoro-substituted γ-indolylketone a partial defluorination was
observed. The intermediate samarium enolate of the tetracyclic
products could be trapped by adding reactive alkylating agents as
electrophiles delivering products with quarternary carbons. In the case
subsequently investigated in detail under variation of the
[
5]
[6]
substituents at the benzene ring and of the spacer unit, the
type of carbonyl groups, or the reaction conditions.[ It was also
extended to naphthalene derivatives that allowed the preparation
of steroid analogs[8] and to heteroarenes.[9-11] Strong Lewis bases
such as hexamethylphosphoramide (HMPA) are required to
increase the redox potential to a level that allows the generation
of reactive samarium ketyl.[12] Tripyrrolidino phosphoric acid
triamide (TPPA) can often serve as a good substitute for the toxic
additive HMPA yielding satisfying conversions in many, but not in
all examined cases.[13]
7]
Of particular interest were the thoroughly studied indole
derivatives[10,14] since the resulting indolines are of importance as
intermediates for natural product syntheses or as privileged
structures of biologically active compounds.[15] We first started
with the simple N-alkylated or N-acylated indole derivatives A
(Scheme 1) that furnished under standard conditions the
expected tricyclic compounds B.[ A transition state TS1 with
minimized steric interactions of substituents[16] and ligands
explains the high stereoselectivity of the cyclization reactions.
of
a
dimethoxy-substituted tetracyclic cyclization product
a
a
subsequent reductive amination stereoselectively provided
10]
pentacyclic compound that was subsequently N-protected and
subjected to a regioselective elimination. The obtained functionalized
pentacyclic product should be convertible into the alkaloid brucine by
four well established steps. Overall, the presented report shows that
functionalized tetracyclic compounds with different substituents are
rapidly available with the samarium diiodide cascade cyclization as
crucial step. Hence analogs of the landmark alkaloid strychnine, e.g.
with specific fluorine substitutions, should be easily accessible.
Introduction
In 1999 a samarium diiodide-promoted cyclizations of γ-aryl
ketones was discovered by our group, providing bicyclic products
in good yields and with excellent diastereoselectivity,[1] This type
of transformation was new in samarium diiodide chemistry.[2,3] The
reaction proceeds under dearomatization[4] of the benzene ring
and hence it is a synthetically very valuable process that was
Scheme 1: Samarium diiodide-promoted cyclizations of N-acylated indole
derivatives A stereoselectively leading to tricyclic compounds B and proposed
transition state TS1.
2
Overall, two electrons (from SmI ) and two protons (from ROH)
are transferred to generate the cyclization product. Starting from
cyclic ketones, annulated tetracyclic products were isolated in
good yields and under control of the relative configuration of four
stereogenic centers. Precursors similar to A but bearing electron-
accepting substituents at C-3 of the indole moiety, also furnish the
expected tricyclic compounds under mild conditions and in high
yield.[14] Recent DFT calculations of these systems predicted the
correct stereochemical outcome of the cyclizations, but they also
[
[
a]
b]
Dr. C. Beemelmanns, Dr. D. Nitsch, Dr. C. Bentz, Prof. Dr. H.-U.
Reissig
Institut für Chemie und Biochemie, Freie Universität Berlin
Takustraße 3, 14195 Berlin, Germany
E-Mail: hans.reissig@chemie.fu-berlin.de
Dr. C. Beemelmanns
Leibniz-Institut für Naturstoff-Forschung und Infektionsbiologie
Beutenbergstraße 11a, 07745 Jena, Germany
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
showed that the electron-transfer steps might be more complex
than previously anticipated.[17]
was prepared in only three steps (stereoselective reductive
amination, N-protection and regioselective elimination). Rawal et
al. had converted PC1 into racemic strychnine in five steps.[20] To
confirm the configuration of our prepared sample of PC1 we
followed the Rawal route until we reached the hexacyclic iso-
strychnine skeleton and all analytical data was in accordance with
previously reported data. Overall, our approach to strychnine is
one of the shortest and most efficient for the synthesis of this
landmark natural product.[21]
Scheme 3: Route to strychnine via pentacyclic compound PC1 by a samarium
diiodide-promoted cascade cyclization of N-acylated indole derivative SM2 to
tetracyclic compound TC2 as key step.
Scheme 2: Samarium diiodide-promoted cascade cyclizations of N-acylated
indole derivatives SM1 to tetracyclic compound TC1 and spiro compound SL1
and trapping of intermediate samarium enolate D with bromoacetonitrile to
provide alkylated tetracyclic product TC2.
The high convergence and efficiency of our approach to
intermediate PC1 and finally to strychnine promoted us to study
the samarium diiodide-promoted cascade reactions of substituted
congeners of SM1 and SM2 that should lead to analogs of TC1 or
TC2. In this report, we concentrate on precursors with fluorine,
cyano or alkoxy substituents at positions C-4 to C-7 of the indole
moiety. Of particular interest was the compatibility of the
substituents under the employed reductive reaction conditions,
and the efficiency and the stereoselectivity of the cascade
cyclization. If successful, this approach would lead to interestingly
functionalized polycyclic compounds and it could open routes to
strychnine analogs with substituents in ring A, e. g. to naturally
not occurring compounds carrying fluorine substituents[ or to a
natural products such as brucine which is substituted by two
methoxy groups.[23]
With compounds that are related to A, but contain an additional
electrophilic functional group in the side chain, we could achieve
samarium diiodide-promoted cascade cyclizations in the absence
of proton sources. Starting material SM1 stereoselectively gave
the desired tetracyclic compound TC1 as major product together
with spirolactone SL1 as minor component (Scheme 2, equation
1
).[18] The ethoxycarbonyl group of intermediate dianion C is
intramolecularly attacked either by the benzylic carbanion to give
TC1 or by the tertiary alkoxide then leading to side product SL1.
Under the reaction conditions tetracyclic compound TC1 is
apparently deprotonated by the present samarium alkoxide to
generate enolate D that is either protonated during aqueous work-
up to give TC1 or it could be trapped by an alkylating agent such
as bromoacetonitrile to give tetracyclic compound TC2 with a
newly formed quarternary carbon (equation 2).
22]
Tetracyclic compound TC2 was obtained even more
efficiently when 3-cyanomethyl-substituted starting material SM2 Results and Discussion
was employed in the cascade reaction (Scheme 3). Under
optimized conditions SM2 furnished TC2 in 74% yield (by use of
TPPA: 45% yield).[19] As side product, in particular in the presence
of a proton source, spirolactone SL2 with a 3-exo-methylene unit
was isolated in up to 22%. This compound is formed by the
competing cyanide elimination at the stage of the carbanion
intermediate. Compound TC2 was an ideal precursor for a
synthesis of strychnine since the pentacyclic intermediate PC1
Synthesis of Starting Materials. The indole derivatives required
for the synthesis of SM3-SM14 were either commercially
available or prepared by literature methods (see Supporting
Information). As typical examples, the syntheses of indoles IN6
and IN13 are illustrated in Scheme 4. The preparation of 7-
fluoroindole IN6 was accomplished by
a modified Bartoli
reaction[24] that gave the desired compound in reasonable
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
Table 1. Synthesis of SM3-SM16 by N-acylation of the corresponding indole
derivatives IN3-IN16 (also see equation 3 of Scheme 4 for details).
quantity (equation 1). The synthesis of the so far unknown 3-
cyanomethyl-6-fluoro-substituted indole IN13 by a Mannich-type
reaction is shown in equation 2. Following a known protocol,[25]
first a gramine analog was prepared that was converted into IN13
by N-methylation and substitution employing cyanotrimethylsilane.
The overall yield of 66% was very satisfactory.
3
7
Method[
a]
Yield[b]
Indole
Derivative
R -R (hydrogen at
unlisted positions)
Product
(Time)
4
IN3
R = F
A, 2 d
A, 2 d
A, 2 d
A, 1 d
A, 1 d
B, 3 d
A, 10 d
B, 2 d
A, 3 d
B, 3 d
B, 2 d
B, 2 d
B, 2 d
B, 2d
SM3
70%
54%
61%
30%
24%
96%
49%
35%
57%
63%
11%
90%
57%
26%
5
IN4
R = F
SM4
6
IN5
R = F
SM5
7
IN6
R = F
SM6
5
7
IN7
R = F, R = F
SM7
5
IN8
R = CN
SM8
4
IN9
R = OMe
SM9
5
IN9
R = OMe
SM10
SM11
SM12
SM13
SM14
SM15
SM16
7
IN11
IN12
IN13
IN14
IN15
IN16
R = OBn
5
R = OMe, R6 = OMe
3
CN, R6 = F
R = CH
2
2
2
3
CN, R5 = CN
CN, R5 = OMe
R = CH
3
R = CH
R3 = CH
CN, R5 = OMe,
2
6
R = OMe
[
a] Method A: monoacid was coupled with indole employing DIC/DMAP;
Method B: in situ generation of monoacid chloride and treatment with indole
and base. [b] Yield of purified product.
Cyclization Experiments. First, the monofluoro-substituted
compounds SM3-SM6 were exposed to samarium diiodide (2.2
equiv.) in the presence of HMPA (10 equiv.) in tetrahydrofuran
(
THF) at room temperature (Scheme 5). These standard
conditions of cascade cyclization reactions provided in all cases
the expected annulated tetracyclic products TC3-TC6 in yields
between 24-42%, whereas spirolactones SL3-SL6 were isolated
in 14-29% yield, resulting in a reasonable overall mass balance
on the stage of purified products. The depicted configurations of
the isolated products are based on the NMR data and the close
analogy to the already published examples.[19b] A few minor
products were also isolated in low yield. Beside TC5 and SL5 diol
TC5´ was also formed in 9% yield, which is the result of a
stereoselective reduction of the carbonyl group of the
cyclohexanone subunit of TC5. Analogous subsequent reductions
by the applied slight excess of samarium diiodide were observed
during the cyclizations of SM1 and SM2[19] giving the
corresponding diols as side products usually in less than 10%
yield. In the other examples of Scheme 5 these diols were not
found, however their formation in low quantities cannot be
rigorously excluded.
Scheme 4: Syntheses of IN6 by a modified Bartoli protocol (1), of IN13 by a
Mannich-type reaction introducing the C-3 side chain (2) and N-acylations of
IN3-IN16 employing 4-oxo pimelic acid derivatives (3). (TMSCN =
cyanotrimethylsilane, TBAF = tetra-n-butylammonium fluoride, DIC = N,N´-
diisopropylcarbodiimide, DMAP = 4-(dimethylamino)pyridine.
For the subsequent N-acylation of the indole derivatives, 4-oxo
pimelic acid monoester[26] was either directly coupled using N,N´-
diisopropylcarbodiimide/4-(dimethylamino)pyridine (Method A).
Alternatively, the acid was first converted in situ into the
corresponding acid chloride and then treated with the indole
derivative in the presence of base (Method B).[18,19] The results of
equation 3 (Scheme 4) are summarized in Table 1 and show that
the reactions are generally very slow at room temperature. The
yields are scattering between 11% and 96%, being good to
excellent in many cases. However, there are also several
examples with low efficiency, in particular, if the indole derivative
bears substituents at C-7. Since most of the experiments were
performed only once, the differences in efficiency should not be
overestimated.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
Scheme 5: Cascade cyclizations of fluoro-substituted indole derivatives SM3-SM7 leading to tetracyclic compounds TC3-TC7 and spirolactones SL3-SL7 and
cyclizations of cyano- and alkoxy-substituted indole derivatives SM8-SM12 to TC8-TC12 and SL8-SL12.
The cyclization of 5,7-difluoro-substituted precursor SM7 requires
a separate discussion. Here, not only the expected compounds
TC7 and SL7 were isolated in 19% and 12% yield, respectively,
but also in 6% yield the spirolactone SL4 bearing only the fluorine
substituent at C-5 of the product (numbering of starting material).
In this case, we observed a relatively fast partial rearrangement
probably occurred during the tedious purification of the reaction
mixture by air oxidation of TC11. Although the cyclization of SM11
was not optimized the low yield of this transformation – possibly
induced by the 7-benzyloxy group[28,29] – excludes the use of
compound TC11 for an envisioned synthesis of vomicine type
strychnine alkaloids[30] that bear oxygen substituents at this
position of the indoline substructure. Finally, the reductive
cascade cyclization of 5,6-dimethoxy-substituted indole derivative
SM12 was investigated. Under standard conditions, tetracyclic
compound TC12 and spirolactone SL12 were isolated in 35% and
4% yield.
3
of TC7 into SL7 during recording the NMR spectra in CDCl ; this
observation will be discussed below. The defluorination at C-7
leading to SL4 may occur at different stages of the cascade
cyclization. One reason for the observed regioselectivity of
defluorination may be the proximity of the indole amide function
to C-7 allowing a coordination of samarium diiodide before
reduction. Related reductive process were earlier observed, when
simple chloro-substituted benzene derivatives of type A (see
Scheme 1) were cyclized.[6b] The two fluorine substituents of SM7
seem to facilitate its defluorination since it was not observed in
the other examples of Scheme 5[27] or in simple model compounds
A containing one fluoro substituent or a trifluoromethyl group.[6b]
The next group of experiments deals with substituents at the
indole benzene subunit, that are clearly electron-accepting such
as a cyano group or electron-donating like alkoxy groups
As a rough estimation of the overall reaction rate, we
recorded the time required for a color change of the purple
samarium diiodide solution to a beige-brownish suspension
during cyclizations of compounds SM8, SM10 and SM12.
Whereas with the electron-deficient substrate SM8 the color
disappeared already after ca. 20 seconds, it took four and five
minutes for the electron-rich precursors SM10 and SM12.
Although we are cautious to overestimate these “kinetics”, the
observations are in line with earlier results showing that
(hetero)arenes with electron-withdrawing substituents are much
better substrates for the reductive cyclization processes than
those with electron-donating substituent.[
(
Scheme 5). The cyclization of 5-cyano-substituted indolyl ketone
5,14]
SM8 afforded the expected tetracyclic product TC8 in 41% yield
and the spirolactone SL8 in 19% yield. Similarly good mass
balances were achieved in the cyclizations of SM9 and SM10 with
methoxy groups at C-4 or C-5, respectively, giving tetracycles
TC9 and TC10 as well as lactones SL9 and SL10. On the other
hand, a 7-benzyloxy substituent led to low efficiency of the
cyclization. Starting material SM11 provided only 17% of the
desired tetracyclic compound TC11 and as side product the
partially oxidized product TC11´ in 2% yield, where the two central
bridge-head hydrogens of TC11 are missing. This side reaction
As already mentioned above (Scheme 2) the reductive
cascade cyclization can be used to directly introduce additional
substituents at the bridge-head position via in situ formed
enolates such as F. This process was also examined with typical
substrates SM5, SM10 and SM12 (Scheme 6) and as
electrophiles bromoacetonitrile and allyl iodide were used. All
reactions proceeded smoothly and gave the expected tetracyclic
products TC13-TC16 in moderate yields between 35-48%. The
four compounds were formed stereoselectively and the new
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
substituent at the quarternary carbon was introduced cis to the
hydroxyl group.
cyclizations of 3-cyanomethyl-substituted indolyl ketones SM13-
SM16 (Scheme 7). 6-Fluoro-substituted precursor SM13 provided
the expected tetracyclic product TC13 in good yield together with
the exo-methylene spirolactone SL13´, a result of the elimination
of cyanide. The corresponding products of this side reaction were
not isolated in the three following cases. Electron-deficient 5-
cyano-substituted indolyl ketone SM14 furnished product TC17 in
57% yield, again demonstrating the higher efficacy of electron-
poor aromatic units in the cascade cyclization process. The two
alkoxy-substituted precursors SM15 and SM16 afforded the
expected tetracyclic compounds TC14 and TC16 in 40% and 30%
yield, respectively. In both cases, very little amounts of the
dealkylated spirolactones SL10 and SL12 were isolated. Their
formation will be explained in the mechanistic discussion. The
color change of reaction solutions required 20 seconds in the case
of SM14 and a few minutes with SM15 and SM16, again
confirming the above mentioned rate differences. Overall, the
examples of Scheme 7 show that these reactions are comparable
efficient in generating tetracyclic compounds with quarternary
carbon atoms than those depicted in Scheme 8 where the
cyanomethyl group was introduced as external electrophile.
Scheme 6: Cyclizations of substituted indole derivatives SM5, SM10 and
SM12 followed by trapping with alkylating agents leading to tetracycles TC13-
TC16.
In the transformations of methoxy-substituted precursor
SM10 spirolactone SL10 was isolated in ca. 10% yield. The
cyclization/alkylation of SM12 gave TC16 as expected, but the
oxidized tetracyclic compound TC12´ was also found in 20% yield.
For the synthesis of strychnine and its analogs our route requires
a cyanomethyl group at the bridgehead carbon next to the
carbonyl group of the tetracyclic compounds (Scheme 3). The
cyclization/alkylation cascades shown in Scheme 6 already
delivered products of this type. Alternatively, these compounds
could be prepared by starting with indolyl ketones already bearing
this substituent at C-3 of the indole moiety. In our earlier studies,
we observed an unexpected partial reductive removal of the
cyanomethyl group at the stage of the product. For instance, the
cyclization of SM2 not only led to the desired product TC2 but also
to the dealkylated compound TC1 isolated in ca. 10% yield.[19]
This reductive C-C bond cleavage should give a samarium
enolate such as intermediate F (see Scheme 2) and hence we
modified our cyclization protocol by addition of bromoacetonitrile
before aqueous work-up, in order to re-install the cyanomethyl
group. This slightly increased the overall efficiency of the reaction
leading to TC2. Under these conditions, we now studied the
Scheme 7: Cyclizations of 3-cyanomethyl-substituted indole derivatives
SM13-SM16 providing the tetracyclic products TC13-TC16.
Although the basic mechanistic details of the samarium diiodide-
mediated cyclizations of indolyl ketones had already been
[14,19]
discussed in earlier reports,
the formation of new side
products requires an explanation of the events as summarized in
Scheme 8. We propose for the precursors SM investigated in the
present study the “carbonyl-first” mechanism, assuming that the
first electron transferred from samarium diiodide is accepted by
the ketone moiety of the substrate.[31] The resulting samarium
ketyl E subsequently undergoes the first cyclization via transition
state TS3 to provide the stabilized radical
F with high
stereoselectivity. The transfer of the second electron from
samarium diiodide affords the crucial carbanionic intermediate G
which has two possibilities to react further. The intramolecular
attack at the ethoxycarbonyl group gives intermediate H that after
aqueous work-up furnishes the tetracyclic product TC as major
component. The stereoselectivity of the second cyclization (which
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
may be reversible) is steered by the already existing stereogenic
centers of the tricyclic precursor and provides the least congested
tetracyclic product.
Scheme 8: Mechanistic scenario in samarium diiodide-mediated cascade cyclizations of precursors SM to tetracyclic products TC and spirolactones SL or SL´
for clarity of presentation the HMPA ligands at samarium are not shown).
(
An alternative pathway converts dianion G into spirolactone SL
by attack of the tertiary alkoxide at the ethoxycarbonyl group via
intermediate J. Protonation gives the isolated spirolactone SL.
We have no arguments to exclude that intermediate H undergoes
a rearrangement to J by attack of the alkoxide oxygen to the
carbonyl group thereby cleaving the just formed C-C bond. A
related (acid or base catalyzed) rearrangement of product TC to
SL can also not been rigorously excluded (in one case this was
observed during NMR recording, see above) which may occur by
an attack of the hydroxyl oxygen to the (protonated) carbonyl
group and fragmentation under lactone formation. Nevertheless,
we assume that the major pathway to SL proceeds directly. For
compounds with a cyanomethyl group at the bridgehead, we
observed a C-C bond cleavage in up to 10% under standard
conditions leading to the isolation of dealkylated tetracyclic
product TC´. We propose an additional transfer of two electrons
by the excess of samarium diioidide to compound TC to provide
the samarium enolate I and the “anion” of acetonitrile as
fragments. As mentioned above, this enolate can successfully be
converted back to alkylated TC by addition of bromoacetonitrile
before aqueous work-up.[19] Another side reaction observed in two
cases was the formation of exo-methylene spirolactones SL´. An
elimination of cyanide at the stage of G explains this process.[32]
Synthesis of a Brucine Precursor. The examples of this
report demonstrate that several highly functionalized tetracyclic
indole derivatives are available in a stereoselective fashion by the
samarium diiodide-promoted ketyl cascade sequence, in several
examples even on a gram-scale. As described earlier by us, the
diastereoselective reduction of key compound TC2 (Scheme 3)
resulted in an ideal precursor for a very short formal total
synthesis of strychnine. Similarly, brucine, the dimethoxy
congener of strychnine, should be available using the same
straightforward series of functional group conversions.
We started with the exploration of the reductive amination
reaction of tetracyclic compound TC16 in the presence of an
excess of Raney Nickel. It stereoselectively furnished – via an
intermediate cyclic imine – the desired pentacyclic compound
PC2 in one step. Its subsequent protection as carbamate afforded
PC3 in 71% overall yield (Scheme 9). Again, the convex shape of
the intermediate imine favored an attack of the reducing agent
from the “bottom side” yielding exclusively diastereoisomer PC3.
As found previously,[
18,19]
the regioselective elimination of the
tertiary hydroxyl group was best achieved using the highly
reactive Martin’s sulfurane (Ph S[OC(CF Ph] at low
temperature.[33a] This protocol gave the desired compound PC4 in
2
3
)
2
2
)
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
high preference with 72% yield. Its regioisomer PC4’ was also
isolated in low amounts (1:1 mixture with PC4, combined yield
the synthesis of brucine precursor PC4 via TC16 involves only
five steps starting from IN12. Analogously to Rawal´s and our
route to strychnine, a few additional steps should very likely lead
to the hitherto never prepared natural product brucine in its
racemic form.
16%). Again, the preferential formation of regioisomer PC4 may
be explained by the suitable geometry with an antiperiplanar
alignment of the C-H bond and the C-OX moiety functioning as
leaving group. Application of Burgess´ reagent[33b] at higher
temperatures (70-100 °C) resulted in a lower selectivity. Overall,
2 3 2 2
Scheme 9: Conversion of compound TC16 into pentacyclic compound PC4 – a possible precursor of brucine. Martin´s sulfurane = Ph S[OC(CF ) Ph] .
far. Overall, the results underscore the efficacy and the flexibility
of samarium diiodide-promoted cyclizations to interesting
scaffolds suitable for natural product synthesis.[35]
Conclusions
Our results clearly demonstrate the high potential of
samarium diiodide-induced dearomatizing cascade reactions for
the rapid and stereoselective synthesis of indoline derivatives.[34] Experimental Section
Substituents in the benzene moiety of the precursors SM
influence the ratio of the two major products, the annulated
tetracyclic products TC and the spirolactones SL, but no simple
rule could be found that allows a prediction of their ratio. It has not
been investigated whether modifications of the reaction
conditions can lead to improved selectivities. On the other hand,
the substituents of SM influence the rate of the cyclization reaction
as expected, since electron-accepting groups accelerate and
electron-donating groups decelerate the process. The formation
of side-products isolated in low amounts could be plausibly
explained.
For general information, all experimental and analytical details see
Supporting Information.
General procedure for samarium diiodide-induced cyclizations of
indolyl ketones: The indole derivative SM (1.0 equiv.) was dissolved in
THF (16 mL per mmol indole) and argon was bubbled through the solution
for 10-20 min at r.t. The resulting solution was added in one portion to a
solution of SmI
the indicated temperature. After 30-60 min the reaction was quenched with
sat. aqueous solution of NaHCO , the organic phase was separated and
the aqueous phase was extracted three times with diethyl ether. The
4
combined ether extracts were washed with brine, dried with MgSO ,
2
(2.4 equiv.) in THF containing HMPA which was stirred at
3
The obtained spirolactones SL feature an interesting skeleton
with functional groups allowing further transformations whereas
the annulated tetracyclic compounds TC should be ideal starting
materials to prepare natural product analogs. Hence, a few steps
should convert mono- or difluorinated compounds TC3-TC7 or
TC13 into specifically fluorinated strychnine analogs. Cyclization
products TC14 or TC17 could be precursors of strychnine-type
compounds with methoxy or cyano groups in ring A of the alkaloid.
The reactions required to approach these target compounds
should be fully compatible with the substituents as evidenced by
the conversion of the dimethoxy-substituted derivative TC16 into
pentacyclic compound PC4 in three efficient steps. Four
additional steps are necessary to complete the total synthesis of
brucine, a strychnine relative not prepared by total synthesis so
filtrated and evaporated. The obtained residue was purified by column
chromatography on silica gel (hexanes/ethyl acetate).
Cyclization of SM3: According to GP3, indolyl ketone SM3 (266 mg, 0.83
mmol), SmI (20 mL, 2.0 mmol) and HMPA (1.64 g, 9.17 mmol) furnished
2
after work-up and column chromatography on silica gel (hexanes/ethyl
acetate 3:1, 1:1, 1:3, ethyl acetate) compounds TC3 (56 mg, 24%) and
SL3 (47 mg, 21%) as colorless solids.
(
1
(
3aS*,3bR*,11bS*) 11-Fluoro-3a-hydroxy-3,3a,3b,4,5,11b-hexahydro-
H-pyrido[3,2,1-jk]carba-zole-1,6(2H)-dione (TC3): M. p. > 180 °C
sublimation). 1H NMR (CDCl
+ CD OD 5:1, 400 MHz): δ = 1.82-1.92,
3
3
1
2
.96-2.08 (2m, 2H each, 3-H, 4-H), 2.17 (ddd, J = 2.8, 3.6, 14.8, Hz, 1H,
-H), 2.41 (ddd, J = 7.2, 11.8, 18.5 Hz, 1H, 5-H), 2.56 (ddd, J = 2.6, 7.2,
18.5 Hz, 1H, 5-H), 2.91 (td, J = 5.9, 14.8 Hz, 1H, 2-H), 3.97 (d, J = 7.9 Hz,
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
1
=
H, 11b-H), 4.26 (dd, J = 2.1, 7.9 Hz, 1H, 3b-H), 6.73 (t, JHH ≈ 8.4 Hz, JFH
8.4 Hz, 1H, 10-H), 7.12 (td, JHH = 8.1 Hz, JFH = 5.7 Hz, 1H, 9-H), 7.72 (d,
118.3, 136.6, 146.8, 150.7 (4s, Ar), 168.0, 206.1 (2s, C-6, C-1) ppm. IR
(ATR): 휈̃ = 3465-3255 (O-H), 3035-3010 (=C-H), 2920, 2850 (C-H), 2250
-
1
J = 8.1 Hz, 1H, 8-H) ppm; the signal for the OH group could not be
assigned unambiguously. 13C NMR (CDCl
+ CD OD 5:1, 101 MHz): δ =
9.9, 30.3, 33.4, 34.5 (4t, C-3, C-5, C-4, C-2), 48.6* (dd, JFC = 3.1 Hz, C-
1b), 66.3 (s, C-3a), 69.6 (d, C-3b), 111.7 (dd, JFC = 20.5 Hz, C-10), 112.7
(CN), 1780, 1730 (C=O), 1645 (C=C) cm . HRMS (ESI-TOF): calcd. for
+
+
+
3
3
C
19
H
20
N
2
O
5
: 357.1445 [M + H] , 379.1264 [M + Na] , 395.1004 [M + K] ;
2
1
found: 357.1468, 379.1288, 395.0971.
3
a-Hydroxy-9,10-dimethoxy-3,3a,4,5-tetrahydro-1H-pyrido[3,2,1-
(s, C-8), 114.9 (d, JFC = 21.6 Hz, C-11a), 130.3 (dd, JFC = 7.8 Hz, C-9),
1
jk]carbazole-1,6(2H)-dione (TC12´): H NMR (CDCl
3
, 700 MHz: δ = 2.11
143.4 (d, JFC = 7.5 Hz, C-7a), 159.9 (d, JFC = 248.4 Hz, C-11), 167.7 (s, C-
(
dt, J = 4.4, 13.5 Hz, 1H, 3-H), 2.23 (ddd, J = 4.6, 13.5 Hz, 1H, 4-H), 2.33-
2.38 (m, 2H, 3-H, 4-H), 2.58 (ddd, J = 2.0, 4.8, 17.6 Hz, 1H, 5-H), 2.80
ddd, J = 2.0, 4.6, 17.5 Hz, 1H, 2-H), 3.06 (ddd, J = 4.6, 13.4, 17.6 Hz, 1H,
6
), 206.9 (s, C-1) ppm; *the signal could not be detected in the recorded
1
3C NMR experiment due to the overlying signal of CD
3
OD, however, the
(
chemical shift and coupling constant could be unambiguously assigned in
a DEPT experiment. 19F NMR (CDCl
+ CD OD 5:1, 376 MHz): δ = -117.5
dd, J = 7.0, 8.8 Hz, 1F, 11-F) ppm. IR (film): 휈̃ = 3290 (O-H), 2980-2855
5
6
-H), 3.38 (ddd, J = 4.9, 13.4, 17.5 Hz, 1H, 2-H), 3.66 (s, 1H, OH), 3.89 (s,
H, OMe), 7.36 (s, 1H, 11-H), 7.76 (s, 1H, 8-H) ppm. 13C NMR (CDCl
, 176
3
3
3
(
-
1
MHz): δ = 30.4, 34.4, 34.5, 37.4 (4t, C-3, C-5, C-4, C-2), 56.12, 56.15 (2q,
OMe), 63.7 (s, C-3a), 99.3, 102.7, 114.4, 117.5, 129.0, 148.0, 148.5, 148.7
(
C-H), 1635 (C=O), 1610 (C=C) cm . HRMS (ESI-TOF): calcd. for
+
+
+
C
15 3
H14FNO : 276.1030 [M + H] , 298.0850 [M + Na] , 314.0589 [M + K] ;
(
3
2d, 6s, Ar), 168.7 (s, C-6), 194.6 (s, C-1) ppm. IR (ATR): 휈̃ = 3390 (O-H),
found 276.1033, 298.0859, 314.0577.
-
1
000-2835 (C-H), 1720, 1660 (C=O), 1650 (C=C) cm . HRMS (ESI-TOF):
+
+
(
2R*,9a’R*) 1’-Fluoro-7’,8’,9a’,10’-tetrahydro-3H,6’H-spiro[furan-2,9’-
calcd. for C17
H17NO
5
: 316.1179 [M + H] , 338.0999 [M + Na] , 354.0738
pyrido[1,2-a]indole]-5,6’-(4H)one (SL3): M. p. 172-174 °C. 1H NMR
[M + K] ; found: 316.1182 [M + H] , 338.1008 [M + Na] , 354.0775 [M +
K]+.
+
+
+
(CDCl
3
, 500 MHz): δ = 2.09-2.15, 2.28-2.40 (2m, 2H each, 3-H, 8’-H), 2.65
(ddd, J = 7.8, 10.6, 18.5 Hz, 1H, 7’-H), 2.67-2.79 (m, 2H, 4-H), 2.83 (ddd,
J = 2.6, 8.1, 18.5 Hz, 1H, 7’-H), 2.93 (dd, J = 10.1, 16.2 Hz, 1H, 10’-H),
.27 (dd, J = 8.9, 16.2 Hz, 1H, 10’-H), 4.61 (dd, J = 8.9, 10.1 Hz, 1H, 9a’-
H), 6.78 (td, JHH = 0.6, 8.2 Hz, JFH = 8.2 Hz, 1H, 2’-H), 7.20 (td, JHH = 8.2
(
2
3aR*,11aS*,11bR*,13aS*)
,3,10,11,11a,12,13,13a-octahydro-1H-pyrido[1,2,3-lm]pyrrolo[2,3-d]-
carbazol-9(11bH)-one (PC2): Raney-Nickel (1.00 g, ~500 wt% in H O)
11a-Hydroxy-5,6-dimethoxy-
3
2
13
Hz, JFH = 5.8 Hz, 1H, 3’-H), 7.90 (d, J = 8.2 Hz, 1H, 4’-H) ppm. C NMR
CDCl , 126 MHz): δ = 24.8, 26.6, 28.0, 29.8, 32.9 (5t, C-3, C-10’, C-4, C-
’, C-8’), 65.0 (d, C-9a’), 83.4 (s, C-2), 111.5 (dd, JFC = 19.8 Hz, C-2’),
12.6 (dd, JFC = 3.5 Hz, C-4’), 114.8 (d, JFC = 22.0 Hz, C-10a’), 130.0 (dd,
was washed several times with MeOH prior to use. The activated catalyst
was added to a solution of cyclization product TC16 (210 mg, 0.59 mmol)
in MeOH (25 mL). The solution was saturated with hydrogen for 1 h and
subsequently stirred for 4 d at r.t. under an atmosphere of hydrogen. The
catalyst was filtered off and the solvent was removed under reduced
pressure. Pentacyclic compound PC2 (177 mg, 87%) was obtained as
colorless oil that slowly solidified. Due to fast oxidation/decomposition it
(
3
7
1
J
FC = 7.8 Hz, C-3’), 144.5 (d, JFC = 7.8 Hz, C-4a’), 158.9 (d, JFC = 246.1
Hz, C-1’), 166.6 (s, C-6’), 174.8 (s, C-5) ppm. 19F NMR (CDCl
, 470 MHz):
δ = -117.5 (t, J ≈ 6.8 Hz, 1F, 1’-F) ppm. IR (film): 휈̃ = 2960-2860 (C-H),
3
-
1
1
2
2
780 (C=O), 1665 (C=C) cm . HRMS (ESI-TOF): calcd. for C15
H
14FNO
3
:
was used without purification. HRMS (ESI-TOF): calcd. for C19
3
24 2 4
H N O :
+
+
+
76.1030 [M + H] , 298.0850 [M + Na] , 573.1808 [2M + Na] ; found:
76.1042, 298.0863, 573.1831.
+
45.1814 [M + H] ; found: 345.1884.
(
2
[
3aR*,11aS*,11bR*,13aS*) Methyl 11a-Hydroxy-5,6-dimethoxy-9-oxo-
,3,32,9,10,11,11a,12,-13,13a-decahydro-1H-pyrido[1,2,3-lm]pyrrolo-
2,3-d]carbazole-1-carboxylate (PC3): To a solution of pentacyclic
compound PC2 (177 mg, 0.51 mmol) in CH Cl (50 mL) were added DMAP
N (101 mg, 1.00 mmol) and methyl chloroformate
100 mg, 1.24 mmol) at 0 °C. The mixture was stirred for 16 h at room
Cl
(20 mL), the
. After evaporation
Procedures for samarium diiodide-induced cyclizations and
subsequent trapping experiment: According to GP, the corresponding
indole derivative SM was reacted with SmI and stirred until the color of
2
the reaction solution turned from purple to brown. Then 1.0-10.0 eq. of the
corresponding alkylation reagent was added, the reaction mixture stirred
for a given time, and worked up as stated in GP.
2
2
(20 mg, 0.16 mmol), Et
3
(
temperature and quenched by addition of saturated aqueous NH
solution (20 mL). The product was extracted with CH Cl
organic phase washed with brine and dried with MgSO
4
2
2
Cyclization of SM12/Alkylation with bromoacetonitrile: According to
4
the procedures above, indolyl ketone SM12 (345 mg, 0.95 mmol), SmI
2
the crude mixture was purified by column chromatography on silica gel
(Hex/EA 2:1, 1:1, 1:2) yielding PC3 (171 mg, 87%) as colorless solid (m.
p. > 230°C (decomposition).
(22.9 mL, 2.29 mmol) and HMPA (1.71 g, 9.55 mmol) were stirred until the
color turned from purple to brownish. Then bromoacetonitrile (1.15 g, 9.55
mmol) was added. After 16 h at room temperature the mixture was worked
up and column chromatography on silica gel (hexanes/ethyl acetate 1:1,
1H NMR (700 MHz, CDCl
3
): δ = 1.59-1.69 (m, 1H, 12-H), 1.80 (m, 2H, 13-
H, 12-H), 1.88 (sbr, 1H, 13-H), 1.92-1.97 (m, 1H, 11-H), 2.09-2.15 (m, 1H,
11-H), 2.15 (ddd, J = 2.0, 9.1, 11.7 Hz, 1H, 3-H), 2.50-2.60 (m, 1H, 10-H),
2.61-2.69 (m, 1H, 10-H), 2.84-2.88 (m, 1H, 3-H), 3.39 (sbr, 13a-H), 3.60-
3.64 (m, 5H, OMe, 2-H), 3.80, 3.86 (2s, 3H each, OMe), 3.93 (s, 1H, 11b-
1:3, ethyl acetate) provided compound TC16 (143 mg, 42%) and
compound TC12´ (59 mg, 20%) as colorless solids. (In a second
experiment under almost identical conditions, TC16 and SL12 were
isolated in 36% and 7% yield).
H), 6.57, 7.87 (2s, 1H each, Ar) ppm. 13C NMR (175 MHz, CDCl
9.1, 30.4, 30.7, 35.9, 44.2 (6t, C-13, C-12, C-10, C-3, C-11, C-2), 52.2
br. s, C-3a), 52.4 (q, OMe), 56.2, 56.4 (2q, OMe), 62.4 (d, C-13a), 68.1 (s,
C-11a), 69.3 (d, C-11b), 102.3, 105.5 (2d, Ar), 130.5, 134.0, 146.6, 148.7,
55.6, 166.7 (6s, 4Ar, CO Me, C-9) ppm. IR (neat): 휈̃ = 3390 (O-H), 3010
=C-H), 2950-2835 (C-H), 1680-1600 (C=O), 1500 (C=C) cm . HRMS
3
): δ = 22.4,
2
(
(
3aS*,3a1R*,11bS*)
2-(3a-Hydroxy-9,10-dimethoxy-1,6-dioxo-
2
,3,3a,3a1,4,5,6,11b-octahydro-1H-pyrido[3,2,1-jk]carbazol-11b-
1
yl)acetonitrile (TC16): M. p. 140 °C. H NMR (500 MHz, CDCl
ddd, J = 3.7, 9.0, 14.3 Hz, 1H, 3-H), 2.03 (ddd, J = 7.4, 7.7, 14.3, 1H, 3-
H), 2.11-2.25 (m, 2H, 4-H), 2.34 (ddd, J = 5.2, 8.1, 18.8 Hz, 1H, 5-H), 2.57
ddd, J = 3.7, 7.4, 17.1 Hz, 1H, 2-H), 2.65 (m , 1H, 5-H), 2.82 (ddd, J = 7.7,
.0, 17.1 Hz, 1H, 2-H), 3.06, 3.11 (2d, J = 17.0 Hz, 1H each, CH CN), 3.84,
.92 (2s, 3H each, OCH ), 4.38 (d, J = 2.2 Hz, 1H, 3b-H), 6.61 (s, 1H, 11-
3
): δ = 1.92
1
2
(
-
1
(
(
+
+
ESI-TOF): calcd. for C21
Na]+; found: 425.1660, 827.3431. Calcd. (%) for
27.3474 [2M
26 2 2
H N O : 425.1683 [M + Na] , 441.1422 [M + K] ,
(
9
3
c
8
+
2
21 26 2 6
C H N O (402.4): C 62.67, H 6.51, N 6.96; found: C 62.58, H 6.63, N
3
7.08.
H), 7.88 (s, 1H, 8-H) ppm; the signal for the OH group could not be
assigned unambiguously. 13C NMR (126 MHz, CDCl
): δ = 27.2, 29.7, 30.9,
2.3, 34.4 (5t, CH CN, C-3, C-5, C-2, C-4), 55.7 (s, C-11b), 56.2, 56.3 (2q,
OMe), 68.4 (s, C-3a), 71.6 (d, C-3b), 101.0, 106.0 (2d, Ar), 117.1 (s, CN),
3
(
2
3aR*,11bS*,13aS*)
,3,3 ,9,10,11,13,13a-octahydro-1H-pyrido-[1,2,3-lm]pyrrolo[2,3-
Methyl
5,6-Dimethoxy-9-oxo-
3
2
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
d]carbazole-1-carboxylate (PC4): At - 15 °C a solution of compound PC3
Chem Rec. 2013, 13, 187–208; j) M. Szostak, M. Spain, D. J. Procter, Chem.
Soc. Rev. 2013, 42, 9155–9183; k) M. Szostak, N. J. Fazakerley, D. Parmar, D.
J. Procter, Chem. Rev. 2014, 114, 5959–6039; l) X. Just-Baringo, D. J. Procter,
Acc. Chem. Res. 2015, 48, 1263–1275; m) M. P. Plesniak, H.-M. Huang, D. J.
Procter, Nature Reviews 2017, 1, article 0077; for reviews describing work of
our group: n) M. Berndt, S. Gross, A. Hölemann, H.-U. Reissig, Synlett 2004,
2 2
(38 mg, 0.094 mmol) and triethylamine (100 mg, 0.99 mmol) in CH Cl (20
mL) was treated with Martin’s sulfurane (100 mg, 0.15 mmol, added in
portions). The mixture was stirred at this temperature for 16 h and then
quenched with sat. aqueous NaHCO
extracted with additional CH Cl (20 mL) and the organic phase was
washed with brine (5 mL), dried (MgSO ), filtered and concentrated in
3
solution (20 mL). The product was
2
2
422-438; o) C. Beemelmanns, H.-U. Reissig, Pure Appl. Chem. 2011, 83, 507-
18; p) C. Beemelmanns, H.-U. Reissig, Chem. Soc. Rev. 2011, 40, 2199-2210.
4
5
vacuo. Purification by column chromatography (silica gel, hexanes/ethyl
acetate 4:1, 3:1, 1:1) afforded PC4 (26 mg, 72%) and a mixture of PC4´
and PC4 (6 mg, 1:1, 16%) as colorless oils.
[
2
3] For related SmI -induced aryl carbonyl couplings, see: a) H.-G. Schmalz, S.
Siegel, J. W. Bats, Angew. Chem. Int. Ed. Engl. 1995, 34, 2383–2385; Angew.
Chem. 1995, 107, 2597–2599; b) H.-G. Schmalz, O. Kiehl, B. Gotov, Synlett
2
1H NMR (500 MHz, CDCl
): δ = 1.90-1.94 (m, 1H, 13-H*), 2.16-2.26 (m,
H, 3-H), 2.48-2.60 (m, 3H, 13-H*, 10-H, 11-H), 2.61-2.75 (m, 3H, 13-H*,
0-H, 11-H ), 3.57-3.63 (m, 1H, 2-H*, 13a-H*), 3.66 (s, 2H, CO Me*), 3.72-
.77 (m, 2H, 13a-H, CO Me*), 3.81 (s, 4H, OMe), 3.83-3.93 (m, 1H, 2-H*),
.90 (s, 3H, OMe), 4.39 (m , 1H, 11b-H), 5.71 (sbr, 1H, 12-H), 6.57 (m , 1H,
-H), 7.89 (sbr, 1H, 7-H) ppm; *signals are broadened or split due to two
3
002, 1253–1256; c) J.-S. Shiue, C.-C. Lin, J.-M. Fang, J. Org. Chem. 1997, 62,
643–4649; d) C.-W. Kuo, J.-M. Fang, Synth. Commun. 2001, 31, 877–892; for
2
1
3
3
4
4
2
related electrochemical reductive cyclizations, see: e) T. Shono, N. Kise, T.
Suzumoto, T. Morimoto, J. Am. Chem. Soc. 1986, 108, 4676–4677; f) N. Kise,
T. Suzumoto, T. Shono, J. Org. Chem. 1994, 59, 1407–1413; g) R. Gorny, H. J.
Schäfer, R. Fröhlich, Angew. Chem. Int. Ed. Engl. 1995, 34, 2007–2009; Angew.
Chem. 1995, 107, 2188–2191; b) J. Heinemann, H. J. Schäfer, R. Fröhlich, B.
Wibbeling, Eur. J. Org. Chem. 2003, 2919–2932.
2
c
c
observable rotamers, ratio ca. 3:1. 3C NMR (100 MHz, CDCl
1
3
): δ = 26.88,
6.93 (t, C-11 ), 27.1, 27.5 (t, C-13), 32.9 (t, C-10), 34.8, 36.3 (t, C-3), 43.1,
3.3 (t, C-2), 51.0 (s, C-3a), 51.6, 52.5 (q, CO Me), 56.1 (q, OMe), 56.53,
2
4
5
7
2
[4] Selected reviews dealing with synthetic applications of dearomatizing
6.59 (q, OMe), 58.9, 59.0 (d, C-13a), 63.4, 63.5 (d, C-11b), 100.9 (d, C-
), 106.3, 106.8 (d, C-4), 119.6, 119.8 (d, C-12), 127.4, 127.8 (s, Ar) 131.4
processes: a) L. N. Mander, Synlett 1991, 134–144; b) T. J. Donohoe, R. Garg,
C. A. Stevenson, Tetrahedron: Asymmetry 1996, 7, 317–344; c) T. Bach, Angew.
Chem. Int. Ed. Engl. 1996, 35, 729-730; d) A. R. Pape, K. P. Kaliappan, E. P.
Kündig, Chem. Rev. 2000, 100, 2917–2940; e) F. L. Ortiz, M. J. Iglesias, I.
Fernández, C. M. A. Sánchez, G. R. Gómez, Chem. Rev. 2007, 107, 1580–
1691; f) S. Quideau, L. Pouységu, D. Deffieux, Synlett 2008, 467–495; g) S. P.
Roche, J. A. Porco, Angew. Chem. Int. Ed. 2011, 50, 4068–4093; Angew. Chem.
(
1
s, Ar), 134.3, 134.6 (s, C-11b), 146.3, 146.4 (s, Ar), 149.1, 149.3 (s, Ar)
55.3, 155.7 (s, CO Me), 168.1, 168.3 (s, C-9) ppm; signals are broadened
or split due to two observable rotamers, ratio ca. 3:1. IR (neat): 휈̃ = 3100
2
-
1
(=C-H), 3005-2845 (C-H), 1700, 1650 (CO), 1500 (C=C) cm . HRMS (ESI-
+
+
24 2 5
TOF): calcd. for C21H N O : 385.1758 [M + H] , 407.1577 [M + Na] ;
2
011, 123, 4154–4179; h) C.-X. Zhuo, W. Zhang, S. L. You, Angew. Chem. Int.
found: 385.1805, 407.1632.
Ed. 2012, 51, 12662–12686; Angew. Chem. 2012, 124, 12834–12858; i) Q.
Ding, X. Zhou, R. Fan, Org. Biomol. Chem. 2014 12, 4807–4815; j) C.-X. Zhuo,
C. Zheng, S.-L. You, Acc. Chem. Res. 2014, 47, 2558–2573; k) N. Denizot, T.
Tomakinian, R. Beaud, C. Kouklovsky, G. Vincent, Tetrahedron Lett. 2015, 56,
4
1
Methyl (3aS*,5a1S*,13bR*)-11,12-Dimethoxy-8-oxo-1,2,4,5,5a1,7-
hexahydro-8H-pyrido[1,2,3-lm]pyrrolo[2,3-d]carbazole-3(3aH)-
carboxylate (PC4’): Partial assignments from 1H NMR data of crude
413–4429; l) X. Just-Baringo, D. J. Procter, Acc. Chem. Res. 2015, 48, 1263–
275; m) T. Wu, L. Zhang, S.-L. You, Chem. Soc. Rev. 2016, 45, 1570–1580;
reaction mixture with PC4: 1H NMR (500 MHz, CDCl
): δ = 3.85 (s, 3H,
, 1H, 11b-H), 5.77 (sbr, 1H, 12-H), 6.68
c
m , 1H, 4-H), 7.90 (sbr, 1H, 7-H) ppm.
3
OMe), 3.89 (s, 3H, OMe), 4.39 (m
c
n) M. Okumura, D. Sarlah, Synlett 2018, 29, 845–855; o) W. C. Wertjes, E. M.
Southgate, D. Sarlah, Chem. Soc. Rev. 2018, 47, 7996–8017.
(
[5] a) M. Berndt, H.-U. Reissig, Synlett 2001, 1290 – 1292; b) U. K. Wefelscheid,
M. Berndt, H.-U. Reissig, Eur. J. Org. Chem. 2008, 3635–3646.
Acknowledgements
[
6] a) Gross, H.-U. Reissig, Synlett 2002, 2027 – 2030; b) R. S. Kumaran, I.
Brüdgam, H.-U. Reissig, Synlett 2008, 991–994; c) A. Niermann, H.-U. Reissig,
Synlett 2011, 525–528; d) A. Niermann, J. E. Grössel, H.-U. Reissig, Synlett
2
Support of this work by the Deutsche Forschungsgemeinschaft,
the Fonds der Chemischen Industrie, the Studienstiftung des
Deutschen Volkes (fellowships for C.B.), and Bayer HealthCare is
most gratefully acknowledged. The help of C. Groneberg for
HPLC separations and the NMR service team of Dr. A. Schäfer
for numerous NOE experiments is also gratefully acknowledged.
013, 177–180; f) H. Mao, B.-X. You, L.-J. Zhou, T.-T. Xie, Y.-H. Wen, X. Lv,
X.-X. Wang, Org. Biomol. Chem. 2017,15, 6157–6166.
[7] For related systematic investigations of arene derivatives with electron-
withdrawing substituents, see: a) H. Ohno, S. Maeda, M. Okumura, R.
Wakayama, T. Tanaka, Chem. Commun. 2002, 316–317; b) H. Ohno, R.
Wakayama, S. Maeda, H. Iwasaki, M. Okumura, C. Iwata, H. Mikamiyama, T.
Tanaka, J. Org. Chem. 2003, 68, 5909–5916; c) H. Ohno, M. Okumura, S.
Maeda, H. Iwasaki, R. Wakayama, T. Tanaka, J. Org. Chem. 2003, 68, 7722–
Keywords: samarium diiodide • cascade reaction • alkaloids •
lactones • fluorinated heterocycles • dearomatization
7
732; d) H. Ohno, H. Iwasaki, T.Eguchi, T. Tanaka, Chem. Commun. 2004,
2228–2229; e) H. Iwasaki, T. Eguchi, N. Tsutsui, H. Ohno, T. Tanaka, J. Org.
Chem. 2008, 73, 7145–7152; f) H. Iwasaki, N. Tsutsui, T. Eguchi, H. Ohno, M.
Yamashita, T. Tanaka, Tetrahedron Lett. 2011, 52, 1770–1772.
References
[
8] a) M. Berndt, I. Hlobilová, H.-U. Reissig, Org. Lett. 2004, 6, 957–960, b) F.
[
1] a) C. U. Dinesh, H.-U. Reissig, Angew. Chem. Int. Ed. 1999, 38, 789–791;
Aulenta, M. Berndt, I. Brüdgam, H. Hartl, S. Sörgel, H.-U. Reissig, Chem. Eur.
J. 2007, 13, 6047–6062; c) U. K. Wefelscheid, H.-U. Reissig, Adv. Synth. Catal,
2008, 350, 65–69; d) U. K. Wefelscheid, H.-U. Reissig, Tetrahedron: Asymmetry
Angew. Chem. 1999, 111, 874–876; b) E. Nandanan, C. U. Dinesh, H.-U.
Reissig, Tetrahedron 2000, 56, 4267–4277.
2010, 21, 1601– 610.
[
2] First seminal publications: a) J.-L. Namy, H. B. Kagan, Nouv. J. Chim. 1977,
1
, 5–7; b) P. Girard, J.-L. Namy, H. B. Kagan, J. Am. Chem. Soc. 1980, 102,
[9] F. Aulenta, U. K. Wefelscheid, I. Brüdgam, H.-U. Reissig, Eur. J. Org. Chem.
2008, 2325–2335.
2693–2698; for selected reviews on samarium diiodide chemistry: c) G. A.
Molander, C. R. Harris, Chem. Rev. 1996, 96, 307–338; d) D. J. Edmonds, D.
Johnston, D. J. Procter, Chem. Rev. 2004, 104, 3371–3404; e) K. Gopalaiah, H.
B. Kagan, New J. Chem. 2008, 32, 607–637; f) D. J. Procter, R. A. Flowers II.,
T. Skrydstrup, in Organic Synthesis Using Samarium Diiodide: A Practical Guide,
Royal Society of Chemistry, Cambridge, UK, 2010; g) T. Honda, Heterocycles
[
10] a) S. Gross, H.-U. Reissig, Org. Lett. 2003, 5, 4305–4307; b) V. Blot, H.-U.
Reissig, Eur. J. Org. Chem. 2006, 4989–4992;
[
11] P. Chen, J. Wang, K. Liu, C. Li, J. Org. Chem. 2008, 73, 339–341.
2011, 83, 1–46; h) M. Szostak, D. J. Procter, Angew. Chem. Int. Ed. 2012, 51,
[12] The use of HMPA as additive strongly raises the reducing ability of SmI ; it
2
9238–9256; Angew. Chem. 2012, 124, 9372–9390; i) K. Gopalaiah, H. B. Kagan,
is therefore required for many ketyl coupling reactions, see: a) J. Inanaga, M.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
Ishikawa, M. Yamaguchi, Chem. Lett. 1987, 1485-1486; b) E. Prasad, R. A.
Flowers II, J. Am. Chem. Soc. 2002, 124, 6895-6899; c) A. Dahlén, G.
Hilmersson, Eur. J. Inorg. Chem. 2004, 3393–3403; d) R. A. Flowers II, Synlett
D. Gladow, H.-U. Reissig, J. Org. Chem. 2014, 79, 4492–4502; h) H. R. A. Golf,
A. Wiehe, H.-U. Reissig, Org. Lett. 2015, 17, 982−985.
2008, 1427-1439. c) S. Halder, S. Hoz, J. Org. Chem. 2014, 2682–2687.
[
23] (-) Brucine has been isolated together with its close relative strychnine in
the strychnos nux-vomica tree. It is slightly less toxic than strychnine,
commercially available and is used for separation of enantiomers or as chiral
ligand (or modified ligand) in asymmetric syntheses: see, for instance: W. J.
Kerr, D. M. Lindsay, E. M. Rankin, J. S. Scott, S. P. Watson, Tetrahedron Lett,
[
13] Whereas HMPA is known to be carcinogenic and teratogenic, TPPA very
likely does not show these harmful effects. a) C. E. McDonald, J. D. Ramsey,
D. G. Sampsell, J. A. Butler, M. R. Cecchini, Org. Lett. 2010, 12, 5178–5181; b)
M. Berndt, A. Hölemann, A. Niermann, C. Bentz, R. Zimmer, H.-U. Reissig, Eur.
J. Org. Chem. 2012, 1299–1302; c) C. E. McDonald, J. D. Ramsey, D. G.
Sampsell, L. A. Anderson, J. E. Krebs, S. N. Smith, Tetrahedron. 2013, 69,
2000, 41, 3229–3233; J.-Y. Li, H. Y. Kim, K. Oh, Org. Lett. 2015, 17, 1288−1291.
To the best of our knowledge no total synthesis of brucine has been reported
so far.
2
947–2953; d) for ureates as alternative additives, see: C. E. McDonald, J. R.
Ramsey, C. C. McAtee, J. R. Mauck, E. M. Hale, J. A. Cumens, J. Org. Chem.
016, 81, 5903–5014; e) also see ref. 7b. and 11c.
[
24] a) G. Bartoli, G. Palmieri, M. Bosco, R. Dalpozzo, Tetrahedron Lett. 1989,
2
30, 2129–2132; b) D. Dobson, A. Todd, J. Gilmore, Synth. Comm. 1991, 21,
611–617; c) A. P. Dobbs, M. Voyle, N. Whittall, Synlett 1999, 1594–1596.
[
14] a) V. Blot, H.-U. Reissig, Synlett 2006, 2763 – 2766; b) C. Beemelmanns,
H.-U. Reissig, Org. Biomol. Chem. 2009, 7, 4475–4480; c) C. Beemelmanns, V.
Blot, S. Gross, D. Lentz, H.-U. Reissig, Eur. J. Org. Chem. 2010, 2716–2732;
d) C. Beemelmanns, D. Lentz, H.-U. Reissig, Chem. Eur. J. 2011, 17, 9720–
[
25] C. Guillou, F. Kozielski, C. Labriere, F. Gueritte, S. Tcherniuk, D. Skoufias,
C. Thal, H.-P. Husson, WO 2010/150211 A2, 12/29/2010.
9730.
[26] 4-Oxo pimelic acid monoester is most easily prepared by a chemoselective
lipase-mediated monohydrolysis of the corresponding diester: a) F. Moriuchi, H.
Muroi, H. Aibe, Chem. Lett. 1987, 1141–1144; b) F. Moriuchi, H. Muroi, Y.
Yasushi, Patent No. JP 01132393, 1989.
[
15] For related SmI -induced indolyl carbonyl couplings, see: a) J.-S. Shiue,
2
M.-H. Lin, J.-M. Fang, J. Org. Chem. 1997, 62, 4643–4649; b) for related
electrochemical indolyl carbonyl couplings, see: N. Kise, T. Mano, T. Sakurai,
Org. Lett. 2008, 10, 4617–4620.
[27] Due to the moderate mass balance of these experiments we cannot
rigorously rule out the formation of defluorinated products. For an alternative
mechanism leading to defluorinated aromatic compounds see ref. 3b.
[
16] For stereoelectronic effect in radical cyclizations, see: a) A. L. J. Beckwith,
Tetrahedron 1981, 37, 3073–3100; b) D. C. Spellmeyer, K. N. Houk, J. Org.
Chem. 1987, 52, 959–974.
[28] The corresponding congener of SM11 with a 7-methoxy group provided not
the expected tetracyclic compound but only the spirolactone in 12% yield.
[
3
17] A. J. Achazi, D. Andrae, H.-U. Reissig, B. Paulus, J. Comput. Chem. 2017,
8, 2693–2700.
[29] C. Bentz, Dissertation, Freie Universität Berlin, 2015.
[30] R. Huisgen, Angew. Chem. 1950, 62, 527–534
[
1
18] a) C. Beemelmanns, S. Gross, H.-U. Reissig, Chem. Eur. J. 2013, 19,
7801–17808. For related SmI -induced cascade reaction, see: b) I.
a
2
[
31] For very electron-deficient indolyl ketones with electron-withdrawing
Shinohara, M. Yamada, H. Nagaoka, Tetrahedron Lett. 2003, 44, 4649–4652;
c) A. Kishida, H. Nagaoka, Tetrahedron Lett. 2008, 49, 6393–6397.
substituents at C-3 a “indole-first” mechanism, with electron-transfer to the
aromatic unit, has to be seriously considered (see discussion in ref. 18a, also
see general discussion in ref. 2f).
[
8
19] a) C. Beemelmanns, H.-U. Reissig, Angew. Chem. Int. Ed. 2010, 49, 8021–
025; Angew. Chem. 2010, 122, 8195–8199; b) C. Beemelmanns, H.-U. Reissig,
[
32] Although only isolated in minor quantities, the formation of spirolactones SL
Chem. Eur. J. 2015, 21, 8416–8425; c) Review: C. Beemelmanns, H.-U. Reissig,
Chem. Rec. 2015, 15, 872–885.
without group R from starting materials bearing this substituent (see Scheme 7)
is not easily explained. A reductive C-C bond cleavage could occur at the stage
of I or K.
[
[
20] V. H. Rawal, S. Iwasa, J. Org. Chem. 1994, 59, 2685–2686.
21] First seminal syntheses: a) R. B. Woodward, M. P. Cava, W. D. Ollis, A.
[33] For seminal publications on Martin’s sulfurane, see: a) J. C. Martin, R. J.
Arhart, J. Am. Chem. Soc. 1971, 93, 4327–4329; for applications of Burgess´
reagent, see: b) S. Khapli, S. Dey, D. Mal, J. Indian Inst. Sci. 2001, 81, 461; c)
C. Lamberth, J. Prakt. Chem. 2000, 342, 518–522.
Hunger, H. U. Daeniker, K. Schenker, J. Am. Chem. Soc. 1954, 76, 4749–4751;
b) P. Magnus, M. Giles, R. Bonnert, C. S. Kim, L. McQuire, A. Merritt, N. Vicker,
J. Am. Chem. Soc. 1992, 114, 4403–4405; c) S. D. Knight, L. E. Overman, G.
Pairaudeau, J. Am. Chem. Soc. 1993, 115, 9293–9294; reviews on strychnine
syntheses: d) J. Bonjoch, D. Solé, Chem. Rev. 2000, 100, 3455–3482; e) M.
Mori, Heterocycles 2010, 81, 259–292; f) J. S. Cannon, L. E. Overman, Angew.
Chem. Int. Ed. 2012, 51, 4288–4311; Angew. Chem. 2012, 124, 4362–4386;
syntheses since 2012: g) D. B. C. Martin, L. Q. Nguyen, C. D. Vanderwal, J.
Org. Chem. 2012, 77, 17–46. h) G. Jacquement, G. Maertens, S. Canesi, Chem.
Eur. J. 2015, 21, 7713–7715; i) L.-W. Feng, H. Ren, H. Xiong, P. Wang, L. Wang,
Y. Tang, Angew. Chem. Int. Ed. 2017, 56, 3055–3058; Angew. Chem. 2017,
[
34] For new dearomatizing cascade reactions leading to indoline derivatives,
see: a) S. Mühmel, D. Alpers, F. Hoffmann, M. Brasholz, Chem. Eur. J. 2015,
1, 12308–12312. b) D. Alpers, M. Gallhof, J. Witt, F. Hoffmann, M. Brasholz,
Angew. Chem. 2017, 129, 1423–1427; Angew. Chem. Int. Ed. 2017, 56, 1402–
406
2
1
[35] Selected reports on natural product syntheses employing samarium
diiodide: [a] T. Nakata, Chem. Soc. Rev. 2010, 39, 1955–1972; [b] K. Simon, J.
Weger, E. Schöttner, T. Lindel, Angew. Chem. 2012, 124, 11047–11050;
Angew. Chem. Int. Ed. 2012, 51, 10889–10892; [c] P. Gilles, S. Py, Org. Lett,
1
29, 3101–3104; j) G. S. Lee, G. Namkoong, J. Park, D. Y.-K. Chen, Chem. Eur.
J. 2017, 23, 16189–16193; k) for a seminal review on strychnine and its
structural elucidation, see: R. W. Hoffmann, Strychnine in Classical Methods in
Structure Elucidation of Natural Products, pages 191 – 205, Wiley-VHCA,
Zürich, 2018.
2012, 14, 1042–1045; [d] N. J. Fazakerley, M. D. Helm, D. J. Procter, Chem.
Eur. J.. 2013, 21, 6718–6723; [e] E. P. Farney, S. S. Feng, F. Schäfers, S. E.
Reisman, J. Am. Chem. Soc. 2018, 140, 1267–1270.
[
22] Selected reviews on fluorinated compounds: a) T. Hiyama, Organofluorine
Compounds: Chemistry and Applications; Springer: Berlin Heidelberg, 2000; b)
P. Kirsch, Modern Fluoroorganic Chemistry; Wiley-VCH: Weinheim, 2005; c) K.
Müller, C. Fäh, F. Diederich, Science 2007, 317, 1881–1886; e) D. O'Hagan,
Chem. Soc. Rev. 2008, 37, 308–319; f) T. Yamazaki, T. Taguchi, I. Ojima, In
Fluorine in Medicinal Chemistry and Chemical Biology; I. Ojima, Ed.; Wiley-
Blackwell: Chichester, 2009; pp 3–46. For recent contributions of our group: g)
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201900087
FULL PAPER
Entry for the Table of Contents
Natural Products
A series of γ-indolylketones with
fluorine, cyano or alkoxy substituents in
ring A was investigated in samarium
diiodide-promoted cascade cyclizations.
This process stereoselectively afforded
annulated tetracyclic indoline derivatives
together with spirolactones. The new
tetracyclic compounds are potential
precursors for specifically fluoro-
substituted analogs of strychnine. A
dimethoxy-substituted tetracyclic
compound was efficiently converted into
a potential precursor of brucine.
C. Beemelmanns, D. Nitsch, C. Bentz,
H.-U. Reissig,*…...… Page – Page
Stereoselective Cascade
Cyclizations with Samarium
Diiodide to Tetracyclic Indolines –
Precursors of Fluorostrychnines
and Brucine
This article is protected by copyright. All rights reserved.