Dispersion-Controlled Regioselective Acid-Catalyzed Intramolecular Hydroindolation of cis-Methindolylstyrenes to Access Tetrahydrobenzo[ cd]indoles

Article abstract of DOI:10.1021/acs.orglett.9b00043

Readily prepared cis-β-(α′,α′-dimethyl)-4′-methindolylstyrenes undergo acid-catalyzed intramolecular hydroindolation to afford tetrahydrobenzo[cd]indoles. Our experimental and computational investigations suggest that dispersive interactions between the indole and styrene preorganize substrates such that 6-membered ring formation is preferred, apparently via concerted protonation and C-C bond formation. When dispersion is attenuated (by a substituent or heteroatom), regioselectivity erodes and competing oligomerization predominates for cis substrates. Similarly, all trans-configured substrates that we evaluated failed to cyclize efficiently.

Full text of DOI:10.1021/acs.orglett.9b00043

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Dispersion-Controlled Regioselective Acid-Catalyzed Intramolecular
Hydroindolation of cis-Methindolylstyrenes To Access
Tetrahydrobenzo[cd]indoles
Xiao Cai,† Anargul Tohti,† Cristian Ramirez,† Hassan Harb,† James C. Fettinger,‡
Hrant P. Hratchian,† and Benjamin J. Stokes*,†
†Department of Chemistry & Chemical Biology, University of California, 5200 North Lake Road, Merced, California 95343, United
States
‡Department of Chemistry, University of California, One Shields Avenue, Davis, California 95616, United States
S
* Supporting Information
ABSTRACT: Readily prepared cis-β-(α′,α′-dimethyl)-4′-methindolylstyr-
enes undergo acid-catalyzed intramolecular hydroindolation to afford
tetrahydrobenzo[cd]indoles. Our experimental and computational inves-
tigations suggest that dispersive interactions between the indole and
styrene preorganize substrates such that 6-membered ring formation is
preferred, apparently via concerted protonation and C−C bond formation.
When dispersion is attenuated (by a substituent or heteroatom),
regioselectivity erodes and competing oligomerization predominates for
cis substrates. Similarly, all trans-configured substrates that we evaluated
failed to cyclize efficiently.
trans-configured starting materials and nondispersible cis-
ubstrate conformational biasing arising from steric
Sconstraints, such as those imposed by geminal dialkyl configured substrates (see below).
groups (i.e., the Thorpe−Ingold effect),1 are often used to
circumvent energetic barriers to bond formation in order to
prepare new and useful molecules. We recently developed an
intramolecular acid-catalyzed hydroarylation of β-(α′,α′-
dialkyl)benzylstyrenes (A, Scheme 1A), showing that a gem-
dialkyl group can be used to synthesize indanes efficiently
(B).2,3 We became interested in applying this design concept
to more complex and medicinally relevant substrates, namely
4-bromoindole-derived β-(α′,α′-dimethyl)-4′-methindolylstyr-
enes like C (Scheme 1B), which are rapidly prepared by
sequential enolate cross-coupling, Wittig, and benzyl protec-
tion reactions.4 Cyclization could occur at either C3 to afford
tetrahydrobenzo[cd]indole D or at C5 to afford
tetrahydrocyclopenta[e]indole E. Herein, we report that a
variety of 3-aryl-5,5-dimethyl-1,3,4,5-tetrahydrobenzo[cd]-
indoles (D) are efficiently prepared in good yield by treating
the cis-configured isomer of C with Brønsted acid catalysts
(Scheme 1B, top pathway). In contrast, trans-configured
substrates predominantly oligomerize (Scheme 1B, bottom
pathway). Our experimental and computational data suggest
that angle compression can induce a ground-state stabilization
of cis-substrates through dispersive interactions, presumably
between arene π systems,5 enabling a concerted protonation
and electrophilic attack to afford G, which rearomatizes to H
(Scheme 1C).6 A concerted mechanism avoids generating a
long-lived carbocation intermediate that would be more likely
to participate in competing intermolecular decomposition
pathways, which may explain the disparate cyclizing ability of
Products like H contain quaternary geminal dimethyl and
diarylmethine motifs, which are well-represented in natural
products and medicines,7 and contain the 5,5-dimethyl-1,3,4,5-
tetrahydrobenzo[cd]indole framework (in blue) common to
the important ambiguine and hapalindole natural product
families and closely related to lysergic acid.8 Despite significant
synthetic attention devoted to 5,5-dimethyl-1,3,4,5-
tetrahydrobenzo[cd]indoles, a general method for their
synthesis has not been reported. Target-oriented cyclization
strategies have included stoichiometric Lewis acid mediated
alkene hydroindolations using 7-methoxy-substituted 3-alke-
nylindoles9 or additions to carbonyls10 and intramolecular
Heck reactions of 4-bromoindoles.11
Our optimization revealed that cis-configured indole
analogues could be cyclized with good regioselectivity
(85:15) favoring tetrahydrobenzo[cd]indole by using arene-
containing Brønsted acid catalysts in non-Lewis basic polar-
izable aprotic solvents.4 Good yield was obtained after heating
for 24 h at 130 °C in the presence of 25 mol % of anhydrous
benzenesulfonic acid in toluene. Under the optimized
conditions, our evaluation of different N protecting groups
revealed their influence on the yield and regioselectivity of the
reaction (Table 1). The best result was obtained using cis-
configured benzyl-protected substrate 1a, which afforded 73%
isolated yield of major product 2a on 0.2 mmol scale (entry 1)
Received: January 4, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.9b00043
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
conversion (entry 8). Lastly, free N−H indole 1h simply
decomposed (entry 9).
Scheme 1. Geminal Dimethyl-Enabled Catalytic
Intramolecular Alkene Hydroarylation Reactions
We next evaluated the influence of functional groups on the
cyclization of N-benzylindolyl substrates (Scheme 2). Owing
Scheme 2. Scope of the Cyclization*
Table 1. Evaluation of Indole N-Protecting Groups*
*
Reactions were conducted on a 0.2 mmol scale in a closed vial unless
a
otherwise noted. Regioisomeric ratio determined by 1H NMR
analysis of the crude reaction. Reaction was conducted on a 1.0
*
Reactions employed pure cis-alkenyl starting materials unless
b
otherwise noted and were conducted on a 0.2 mmol scale in a closed
vial. The substrates were fully consumed in all cases. Unless otherwise
noted, yields refer to the isolated indicated major product, and
regioselectivities (2/3, indicated in parentheses) were determined by
c
mmol scale. Structure of 2e confirmed by X-ray crystallography.
d
e
Combined yield of inseparable regioisomers. Structure of 1h
confirmed by X-ray crystallography.
a
1H NMR analysis of the crude reaction. 80 mol % of PhSO3H was
b
c
used. Yield refers to an inseparable mixture of 2 and 3. Starting
d
material was a 65:35 mixture of cis and trans isomers. Starting
material was a 78:22 mixture of cis and trans isomers.
and a similar yield at 1.0 mmol scale (entry 2). The benzyl
group in 2a is readily deprotected in 91% yield.12 Another
readily deprotected indole, N-ethylsulfonyl analogue 1b,
afforded 2b in slightly reduced yield and selectivity (entry
3).13 Yields and regioselectivities for N-alkylindolyl substrates
were similar to those of 1a and include methyl (1c) ethyl (1d)
and isopropyl (1e) groups (entries 4−6). When an electron-
withdrawing tosyl protecting group was used, selectivity
diminished (1f, entry 7). Acetyl protection prevented substrate
to the generally good regioselectivity of the reaction (ranging
from 80:20 to >95:5), we were typically able to obtain the
fused tricyclic isomer in good or excellent yield. Para-
substituents on the styrene moiety, including Me, F, Cl, and
Br (1i−l), afforded fused tricycles in high yield, as did electron-
donating groups positioned meta, like methoxy (1m) and
methyl (1n). In contrast, stoichiometric amounts of
B
DOI: 10.1021/acs.orglett.9b00043
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Table 2. Calculated Enthalpies of Substrate Rotamers
Table 3. Influence of cis- and trans-Alkene Configuration on
the Cyclization of Methindolylstyrenes*
a
Gas-phase calculations of ΔΔH using B3PW91/6-311G(d) with the
GD3 empirical dispersion correction.
benzenesulfonic acid are required to obtain acceptable yields
when halogens are positioned meta to the alkene (1o−q); m-
bromo substrate 1q cyclized in 68% yield from an inseparable
65:35 mixture of cis and trans isomers. Beyond substituted
benzenes, we found that 2-naphthyl analogue 1r gave 76%
yield of 2r from a 78:22 mixture of cis and trans starting
stereoisomers, respectively. In our final variation of the alkene
aromatic substituent, a 2-thiophene analogue also afforded the
fused tricycle 2s in good yield. Shifting our focus to functional
group tolerance on the indole ring, a 2-methyl substituent was
well-tolerated, as were 6-chloro and 7-fluoro variants (2t−v,
respectively). The latter afforded the best yield and
regioselectivity that we observed in this study. Surprisingly,
adding electron-donating substituents to the 7 position
impacted the regioselectivity significantly, with 7-methyl
substrate 1w affording a 55% yield of an inseparable
regioisomeric mixture of 2w and 3w. Even more surprisingly,
formation of the 6-membered ring was completely prohibited
by the presence of a 7-methoxy substituentonly 3x was
isolated.
*
Reactions were conducted on a 0.2 mmol scale in a closed vial and
substrates were consumed in full in all cases. The major
a
decomposition pathway for trans substrates is oligomerization. Yield
and regioselectivity were determined by 1H NMR analysis of the
b
crude reaction mixture. 80 mol % of catalyst was used.
Scheme 3. Disparate Behavior of cis-β-Benzylstyrene and cis-
β-4′-Methindolylstyrene
The disparate regioselectivity outcomes for substrates 1v−x
imply a high degree of dependence on the electronic nature of
the indole ring. We hypothesized that this could be due to the
Thorpe−Ingold effect inducing overlap between the indole and
styrene, which would necessarily deconjugate the styrenyl
alkene. As an indirect measure of the distortion of the CC
bond, we computed the relative gas phase enthalpies (ΔΔH)
of the two rotamers (cis-1 and cis-1′) of 7-substituted indole
substrates that would lead to the respective products 2 or 3
using B3PW91/6-311G(d) with the GD3 empirical dispersion
correction (Table 2).14 (Note: the computed rotamer
enthalpies do not differ significantly without GD3 correction.4)
Fluorinated rotamer cis-1v is 6.66 kcal/mol more stable than
rotamer cis-1v′ (incidentally corresponding to an equilibrium
constant favoring cis-1v by a factor of over 4000 at 130 °C). As
R becomes more electron rich, dispersion is attenuated (ΔΔH
diminishes). Greater dispersion appears to facilitate alkene
protonation, hypothetically by attenuating the alkene’s
conjugation to the phenyl ring.
Combined with our DFT data, these configurational reactivity
differences suggest that dispersion-induced deconjugated
alkenes undergo concerted hydroindolation, since cis alkenes
resist isomerization to trans alkenes despite the cation-
promoting conditions. Incidentally, trans-A cleanly cyclizes to
indane B by increasing the reaction temperature to 130 °C
(Scheme 3A), perhaps due to sufficiently fast five-membered
ring formation.
We also found that benzothiophene analogue cis-4 cyclizes
with regioselectivity similar to indoles under identical
conditions, albeit a bit more sluggishly. In contrast, trans-4
cyclized with slightly improved yield and regioselectivity
compared to trans indoles (eq 1). In contrast, we did not
observe formation of the corresponding fused tricycle when
using benzofuran analogues (eq 2). Rather, cis-7 isomerized to
trans-7 at just 80 °C, with some formation of regioisomer 8
observed. This disparate behavior compared to indoles and
benzothiophenes is likely due to a combination of diminished
nucleophilicity at C3 and diminished dispersion (our DFT
calculations show ΔΔH = 3.16 kcal/mol for the two cis
rotamers of 7).4
In contrast to cis alkenes (odd-numbered entries, Table 3),
trans alkenes (even-numbered entries) afford uniformly low
yields and reduced regioselectivities, presumably due to the
absence of dispersion (see the Supporting Information for
trans enthalpies). Further, while nonindolic cis-β-benzylstyrene
cis-A isomerizes to trans-A in just 1 h at 80 °C (Scheme 3A),
indole analogue cis-1c is unreactive after 24 h (Scheme 3B).
In conclusion, we have developed a catalytic intramolecular
alkene hydroindolation to construct medicinally significant
C
DOI: 10.1021/acs.orglett.9b00043
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
REFERENCES
■
(1) For a recent review, see: Jung, M. E.; Piizzi, G. Chem. Rev. 2005,
105, 1735.
(2) Cai, X.; Keshavarz, A.; Omaque, J. D.; Stokes, B. J. Org. Lett.
2017, 19, 2626.
(3) For recent alkene hydroarylations, see: (a) Minami, Y.; Noguchi,
Y.; Hiyama, T. J. Am. Chem. Soc. 2017, 139, 14013. (b) Niharika, P.;
Satyanarayana, G. Eur. J. Org. Chem. 2018, 2018, 971. For relevant
reviews, see: (c) Rueping, M.; Nachtsheim, B. J.; Beilstein. Beilstein J.
Org. Chem. 2010, 6, 6. (d) Naredla, R. R.; Klumpp, D. A. Chem. Rev.
2013, 113, 6905. For recent electrophilic carbocyclization methods,
see: (e) Zhang, F.; Das, S.; Walkinshaw, A. J.; Casitas, A.; Taylor, M.;
Suero, M. G.; Gaunt, M. J. J. Am. Chem. Soc. 2014, 136, 8851. (f) Li,
A.; DeSchepper, D. J.; Klumpp, D. A. Tetrahedron Lett. 2009, 50,
1924.
(4) See the Supporting Information for details.
(5) For reviews on the role of dispersive aromatic interactions in
organic reactions, see: (a) Maji, R.; Wheeler, S. E. In Aromatic
Interactions: Frontiers in Knowledge and Application; The Royal Society
of Chemistry, 2017; pp 18−38. (b) Wagner, J. P.; Schreiner, P. R.
Angew. Chem., Int. Ed. 2015, 54, 12274. (c) Wheeler, S. E.; Bloom, J.
W. G. J. Phys. Chem. A 2014, 118, 6133. (d) Wheeler, S. E. Acc. Chem.
Res. 2013, 46, 1029. For new methods leveraging dispersive
intramolecular aromatic interactions, see: (e) Mishra, S.; Liu, J.;
Aponick, A. J. Am. Chem. Soc. 2017, 139, 3352.
(6) For examples of concerted cationic alkene functionalization
reactions, see: (a) Surendra, K.; Corey, E. J. J. Am. Chem. Soc. 2012,
134, 11992. (b) Williams, A. Concerted Organic and Bioorganic
Mechanisms; CRC Press: Boca Raton, 2000. (c) Haven, T.; Kubik, G.;
Haubenreisser, S.; Niggemann, M. Angew. Chem., Int. Ed. 2013, 52,
4016. (d) Ghigo, G.; Maranzana, A.; Tonachini, G. J. Org. Chem.
2013, 78, 9041. (e) Vidhani, D. V.; Cran, J. W.; Krafft, M. E.;
Alabugin, I. V. Org. Biomol. Chem. 2013, 11, 1624. (f) McCulley, C.
tetrahydrobenzo[cd]indoles from cis-methindolylstyrenes bear-
ing a benzylic gem-dimethyl group, putatively via a concerted
protonation and C−C bond formation. Empirical trends
(substrate isomerizability, regioselectivity outcomes, electronic
sensitivity, and temperature profile) and calculated ground-
state enthalpies suggest that the regioselectivity is dependent
on dispersive interactions between the styrene and indole,
which are forced into proximity by the gem dimethyl.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.9b00043.
́
H.; Geier, M. J.; Hudson, B. M.; Gagne, M. R.; Tantillo, D. J. J. Am.
Chem. Soc. 2017, 139, 11158.
(7) For leading reports, see: (a) Lian, Y.; Burford, K.; Londregan, A.
T. Tetrahedron 2015, 71, 9509. For a recent review, see: (b) Talele,
T. T. J. Med. Chem. 2018, 61, 2166.
(8) Maimone, T. J.; Ishihara, Y.; Baran, P. S. Tetrahedron 2015, 71,
3652.
(9) Dethe, D. H.; Sau, S. K.; Mahapatra, S. Org. Lett. 2016, 18, 6392.
(10) Muratake, H.; Natsume, M. Tetrahedron 1990, 46, 6331.
(11) Baran, P. S.; Maimone, T. J.; Richter, J. M. Nature 2007, 446,
404.
(12) Benzyl deprotection was accomplished following the procedure
developed by Deaton−Rewolinski and co-workers: Haddach, A. A.;
Kelleman, A.; Deaton-Rewolinski, M. V. Tetrahedron Lett. 2002, 43,
399.
Experimental details and characterization data (PDF)
Accession Codes
CCDC 1884246−1884247 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: bstokes2@ucmerced.edu.
ORCID
James C. Fettinger: 0000-0002-6428-4909
Hrant P. Hratchian: 0000-0003-1436-5257
Benjamin J. Stokes: 0000-0002-5982-654X
Notes
(13) For the development of the N-ethylsulfonyl indole protecting
group, see: Bashford, K. E.; Cooper, A. L.; Kane, P. D.; Moody, C. J.;
Muthusamy, S.; Swann, E. J. Chem. Soc., Perkin Trans. 2002, 1, 672.
(14) Frisch, M. J. et al. Gaussian Development Version, Revision I10;
Gaussian, Inc.: Wallingford, CT, 2016. (b) Becke, A. D. J. Chem. Phys.
1993, 98, 5648. (c) Wang, Y.; Perdew, J. P. Phys. Rev. B: Condens.
Matter Mater. Phys. 1991, 43, 8911. (d) Grimme, S.; Antony, J.;
Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010, 132, 154104. (e) Krishnan,
R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650.
(15) Cai, X.; Tohti, A.; Ramirez, C.; Harb, H.; Hratchian, H. P.;
Stokes, B. J. ChemRxiv 2018, DOI: 10.26434/chemrxiv.7397492.
The authors declare no competing financial interest.
A prior version of this work was previously published on
ChemRxiv.15
ACKNOWLEDGMENTS
■
This research was supported by the University of California,
Merced. Computing time on the Multi-Environment Com-
puter for Exploration and Discovery (MERCED) cluster is
supported by the National Science Foundation (No. ACI-
1429783). B.J.S. and H.P.H. thank the Hellman Fellows Fund
for research support.
D
DOI: 10.1021/acs.orglett.9b00043
Org. Lett. XXXX, XXX, XXX−XXX