Electronic Nature of Ketone Directing Group as a Key to Control C-2 vs C-4 Alkenylation of Indoles

Article abstract of DOI:10.1021/acs.orglett.6b02698

A novel mode of achieving site selectivity between C-2 and C-4 positions in the indole framework by altering the property of the ketone directing group is disclosed. Methyl ketone, as directing group, furnishes exclusively C-2 alkenylated product, whereas trifluoromethyl ketone changes the selectivity to C-4, indicating that the electronic nature of the directing group controls the unusual choice between a 5-membered and a 6-membered metallacycle. The screening of other carbonyl-derived directing groups reveals that strong and weak directing groups exhibit opposite selectivity. Experimental controls and deuteration experiments lend support to the proposed mechanism.

Full text of DOI:10.1021/acs.orglett.6b02698

Letter
pubs.acs.org/OrgLett
Electronic Nature of Ketone Directing Group as a Key To Control C‑2
vs C‑4 Alkenylation of Indoles
Veeranjaneyulu Lanke, Kiran R. Bettadapur, and Kandikere Ramaiah Prabhu*
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560 012, Karnataka, India
*
S Supporting Information
ABSTRACT: A novel mode of achieving site selectivity between C-
and C-4 positions in the indole framework by altering the property
2
of the ketone directing group is disclosed. Methyl ketone, as directing
group, furnishes exclusively C-2 alkenylated product, whereas
trifluoromethyl ketone changes the selectivity to C-4, indicating
that the electronic nature of the directing group controls the unusual
choice between a 5-membered and a 6-membered metallacycle. The screening of other carbonyl-derived directing groups reveals
that strong and weak directing groups exhibit opposite selectivity. Experimental controls and deuteration experiments lend
support to the proposed mechanism.
C−H activation has emerged as a primary mode o1f
functionalization for a wide variety of organic molecules.
Although a vast number of activation methods with a variety of
metal catalysts, ligands, reaction conditions, reagents, etc., have
been discovered, achieving selective functionalization still
Functionalized indoles are an important class of heterocyclic
compounds that are present in a variety of natural products and
drugs. Among them, 4-substituted indoles, known as privileged
5
frameworks, are attractive synthetic targets, as they are the
6
backbone of ergot alkaloids. Functionalization of indole at C-3 is
2
well addressed, while the functionalization at the C-2 and C-7
remains challenging. Even with the use of directing groups,
7
positions has received attention only in recent years. Selective
there are difficulties in selective activation especially when two or
more C−H bonds are present, in the vicinity of the directing3
group, with a similar propensity to undergo C−H activation.
Further, the presence of multiple functional groups in a molecule
leads to at least two different sites for C−H activation. The
approaches for addressing this issue of selective activation
include the use of special ligands, different catalytic systems,
functionalization of C-4 by using a directing group at C-3, in the
presence of a much more reactive C-2 position, is challenging
because the C-4 activation has to proceed via an uncommon 6-
8a,b
membered metallacycle,
whereas the activation at C-2 can7
proceed via an easier to obtain 5-membered metallacycle.
Developing such divergent and selective C−H functionaliza-
tions, between C-2 and C-4, on the indole framework can lead to
easy and short synthetic routes for natural, unnatural, and
4
additives, and sometimes by substrate control (Scheme 1). Only
5
limited examples of such selectivity can be found in the literature,
and the principles for the origin of such selectivity remain
biologically active compounds. On the basis of our previous
8a
work on C-4 functionalization of indole, we focused our efforts
to flip the selectivity toward C-2 using a directing group at C-3
itself. We hypothesized that the electronic factors of the directing
group (strong or weak coordinating, electron rich or poor) may
in some manner influence the selectivity of C−H activation.
Accordingly, we carried out work to establish a switch in the site
selectivity governed by the nature of the directing group. Our
optimization studies began with the trifluoromethyl ketone
derivative of indole, 1a, and methyl acrylate, 3a, as model
substrates being subjected to previously established conditions
for C−H activation with Ru(II) catalyst (entry 1, Table 1) to
furnish the requisite C-4-functionalized product 4a exclusively in
4a
4h
unclear. However, the works of Bedford and Houk have
demonstrated some elegant methods for obtaining insight into
the origin of selectivity.
Scheme 1. Concept of the Reaction
4
0% NMR yield. Concurrently, when methyl ketone 2a was used
instead of 1a, under similar conditions, we found that it furnished
C-2-functionalized product 5a exclusively in 50% NMR yield
(
entry 2, Table 1). Thus, we obtained two different products 4a
and 5a under the same reaction conditions, indicating that Ru(II)
Received: September 8, 2016
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.6b02698
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
Table 1. Optimization Studies
c
yield (%)
b
entry
1a/2a
3a (equiv)
Ru/Rh (mol %)
additive (equiv)
4a
4x
5a
5x
SM
1
2
3
4
5
6
7
8
9
1a−COCF3
2a−COCH3
1a−COCF3
2a−COCH3
1a−COCF3
2a−COCH3
1a−COCF3
2a−COCH3
1a−COCF3
1a−COCF3
2a−COCH3
2a−COCH3
4
4
2
2
4
2
4
2
4
2
2
2
Ru (5)
Ru (5)
Ru (5)
Ru (5)
Ru (5)
Ru (5)
Rh (5)
Rh (5)
Rh (2.5)
Rh (2.5)
Ru (5)
Ru (5)
40
0
38
0
50
52
72
15
0
0
0
0
35
31
87
0
0
0
30
0
AcOH (10)
AcOH (10)
36
0
0
0
d
90 (85)
10
11
12
85
d,e
AcOH (0.5 mL)
82 (80)
e
AcOH (1 mL)
76
b
a
Conditions: 1a (0.3 mmol), DCE (2 mL), AgSbF (4 times the mol % of Ru/Rh cat. used), Cu(OAc) ·H O (1 equiv). Ru = [Ru(p-cymene)Cl ] ;
Rh = [RhCp*Cl ] . H NMR yield, using terephthaldehyde as internal standard. Isolated yield in parentheses. Reaction at 80 °C
6
2
2
2 2
c
1
d
e
2
2
switches between two different sites via the formation of either a
-membered or a 5-membered ruthenacycle by merely changing
Scheme 2. Substrate Scope
6
the electronic factors of the directing group. In order to optimize
this selectivity, the amount of acrylate was reduced, and we found
that it did not lead to any significant changes in the yields (entries
3
and 4). Further, acetic acid was used as an additive, and it was
found that it still furnished exclusive products but increased the
formation of product from 2a (72%, entry 6); it was also found
that it disfavored the yield of the product from 1a (31%, entry 5).
Switching from Ru(II) to Rh(III) led to a noticeable change in
the reaction yield. Rh(III) was found to prefer the electron-
deficient COCF and increased the yield up to 87% (entry 7);
3
however, it drastically reduced the yield with methyl ketone
substrate, probably due to the promotion of enolization of α-
protons (entry 8). However, unable to realize good yields with
identical conditions for both 1a and 2a, we were compelled to
optimize the two reactions separately and establish two
independent conditions for the alkenylation reaction (see the
SI for full details of optimization studies). The C-4 alkenylation
with 1a was eventually optimized with 2.5 mol % of Rh(III)
catalyst leading to an isolated yield of 85% (entry 9), while the C-
2
alkenylation was optimized with 2a as the substrate using
Ru(II) catalyst and AcOH as cosolvent (80%, entry 11).
Through the optimization, the other position-coupled products
(4x and 5x) were never detected. In addition, the starting
materials underwent extensive decomposition in the process of
the reaction and were rarely observed after workup.
The substrate scope of the reaction was explored with various
indole derivatives and olefins to furnish the corresponding C-4-
and C-2-alkenylated products in good to excellent yields
Further, we explored various carbonyl-derived directing
groups and examined its effect on the selectivity of the reaction
Scheme 3). First, we explored simple derivatives of ketone itself
(Scheme 2). Electron-rich indoles were tolerated in the reaction
(
for both C-4 as well as C-2 activation reactions. Aliphatic acrylate
derivatives, namely, methyl, ethyl, and n-butyl, and aromatic
derivatives, such as phenyl, anisyl, and naphthyl, were found to
furnish good to excellent yields (Scheme 2). Useful olefin
partners such as styrene derivatives and vinyl phosphonates were
also found to react smoothly to afford the corresponding styrene
and phosphonate products.
and found that pivaloyl as directing group at C-3, under the
standard conditions (Rh and Ru), yielded double-alkenylated
(C-2 and C-4) product in 54% and 50% yield, respectively (6a,
Scheme 3). The C-3 benzoyl as directing group resulted in C-2
selectivity, but the product was double alkenylated on both the
indole as well as the phenyl ring (13% and 39%, 7a, Scheme 3).
We found that amide as a directing group furnished exclusively
B
DOI: 10.1021/acs.orglett.6b02698
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Selectivity with Other Carbonyl Directing Groups
Scheme 5. Control Experiments
C-2 alkenylation in excellent yields (9a, 9b, Scheme 3), while
ester as a directing group furnished exclusively C-4 alkenylation,
8a
albeit in moderate yields (10a, 10b). In our previous report, we
demonstrated that the use of aldehyde as a directing group leads
to C-4 alkenylation (11a). When carboxylic acid was used as the
directing group, a complex mixture was obtained.
Deuterium incorporation studies were carried out to shed light
on the site of metallacycle formation. Compounds 1a and 2a
were independently subjected to standard conditions with both
Rh and Ru, along with either D O or AcOD as the deuterium
2
source (Scheme 4). We found that with 1a as the substrate C-4
Scheme 4. Deuterium Incorporation Studies
source and found no trace of deuteration (eq 3, Scheme 5).
These controls indicate that COCF cannot function as a
3
directing group to activate the vicinal C−H bonds. However, it
appears that it can only assist those C−H bonds that have
proclivity toward electrophilic metalation. This hypothesis was
further confirmed by the inability of COCF to even function-
3
alize C-2 in indole. Accordingly, we blocked C-4 in 1a with either
Br or phenyl substituents. Certainly, due to the low electron
density of C-2, no expected product was formed (eq 4, Scheme
a
b
Deuterium source was used in place of AcOH. Deuteration of acetyl
was observed.
5
). In addition, 4-bromoindole was subjected to deuteration, and
deuteration was 23 times more successful than C-2 deuteration
as expected, no deuteration incorporation was observed on the
C-2 position (eq 5, Scheme 5). Further, when 4a (the alkenylated
product) was subjected to deuteration, it did not furnish even
trace amounts of deuteration at C-2 (eq 5). Once again, this can
be rationalized by the further deactivation (i.e., lessening of
electron density) of the indole moiety by the conjugated ester at
when D O was used as deuterium source, while 2a resulted in
2
equal amounts of C-2 and C-4 deuteration. This indicates that
COCF favors the C-4 metallacycle, whereas COCH has no
3
3
such preferences. Moreover, exchanging the metal catalysts
(
Scheme 4) did not lead to any change in the site of C−H
activation, indicating that the observed selectivity is independent
of the metal used.
C-4. However, when 5a (with COCH ) was subjected to
deuteration, it was able to activate the C-4 position (eq 6), as
3
First, we found a previous report for the alkenylation of
acetophenone under almost similar conditions (eq 1, Scheme 5).
However, no report for the trifluoromethyl derivative was found.
When the trifluoromethyl derivatives of pyrene or benzene were
used under the established conditions, Rh or Ru did not provide
any trace of the requisite product (eq 2, Scheme 5). Since no
product formed does not necessarily mean absence of C−H
evidenced by 54% deuteration. Since COCH was able to activate
3
the C-4 position, we thought if the C-2 is blocked then surely
alkenylation should take place at C-4. To our surprise, the
product 5a, when subjected to alkenylation again with the Ru
catalyst, was unable to furnish the alkenylation at C-4 (Scheme
6), in spite of the activation that it can accomplish. This clearly
indicates that although C−H activation at C-4 is very feasible
9
activation, we subjected the pyrene derivative to deuteration
with COCH as directing group, the insertion process appears to
3
under both conditions (Rh and Ru) using D O as deuterating
be nonfeasible with Ru catalyst. However, with Rh catalyst, we
2
C
DOI: 10.1021/acs.orglett.6b02698
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 6. Consecutive Alkenylation
thank Dr. A. R. Ramesha (RL Fine Chem) for useful discussions.
V.L. and K.R.B. thank CSIR, New Delhi, for fellowships.
REFERENCES
■
(
1) (a) Trost, B. M. Acc. Chem. Res. 2002, 35, 695. (b) Alberico, D.;
Scott, M. E.; Lautens, M. Chem. Rev. 2007, 107, 174. (c) Bru
̈
ckl, T.;
Baxter, R. D.; Ishihara, Y.; Baran, P. S. Acc. Chem. Res. 2012, 45, 826.
d) McMurray, L.; O’Hara, F.; Gaunt, M. J. Chem. Soc. Rev. 2011, 40,
885.
2) (a) Wencel-Delord, J.; Dro
011, 40, 4740. (b) Ackermann, L. Chem. Rev. 2011, 111, 1315.
(
1
(
̈
ge, T.; Liu, F.; Glorius, F. Chem. Soc. Rev.
observed the formation of C-4 alkenylation. As expected, when
2
4
a was subjected to further alkenylation, with Ru or Rh, no
(
(
c) Ritleng, V.; Sirlin, C.; Pfeffer, M. Chem. Rev. 2002, 102, 1731.
d) Padala, K.; Jeganmohan, M. Org. Lett. 2011, 13, 6144. (e) Pimparkar,
reaction was observed. The control experiments indicate that
COCF is effective for direct activation only at C-4 in indole and
3
S.; Padala, K.; Jeganmohan, M. Proc. Indian Natl. Sci. Acad., Part A 2014,
0, 999. (f) Colby, D.; Tsai, A.; Bergman, R.; Ellman, J. Acc. Chem. Res.
2012, 45, 814.
3) (a) Lyons, T. W.; Sanford, M. S. Chem. Rev. 2010, 110, 1147.
not C-2. One possible explanation for these observations is that
8
COCF does not direct C−H activation by precoordination to
3
the metal but instead only stabilizes a metal−arene σ complex
that forms due to electrophilic metalation. Hence, the reason for
C-4 activation in indole must surely have to do with the electron
(
(b) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc. Chem. Res. 2012,
45, 788. (c) Neufeldt, S. R.; Sanford, M. S. Acc. Chem. Res. 2012, 45, 936.
(d) Arockiam, P. B.; Bruneau, C.; Dixneuf, P. H. Chem. Rev. 2012, 112,
1
0
density at C4.
5
879.
In summary (Scheme 7), we have presented a directing group
dependent site-selective C−H activation and functionalization of
(
4) (a) Bedford, R. B.; Durrant, S. J.; Montgomery, M. Angew. Chem.,
́
Int. Ed. 2015, 54, 8787. (b) Martínez, A. M.; Echavarren, J.; Alonso, I.;
Rodríguez, N.; Arrayas, R. G.; Carretero, J. C. Chem. Sci. 2015, 6, 5802.
Scheme 7. Summary of Observations
(c) Liu, X.; Li, X.; Liu, H.; Guo, Q.; Lan, J.; Wang, R.; You, J. Org. Lett.
2
015, 17, 2936. (d) Liu, X.; Li, G.; Song, F.; You, J. Nat. Commun. 2014,
5
, 5030. (e) Huckins, J. R.; Bercot, E. A.; Thiel, O. R.; Hwang, T. L.; Bio,
M. J. Am. Chem. Soc. 2013, 135, 14492. (f) Guimond, N.; Gouliaras, C.;
Fagnou, K. J. Am. Chem. Soc. 2010, 132, 6908. (g) Rakshit, S.;
Grohmann, C.; Besset, T.; Glorius, F. J. Am. Chem. Soc. 2011, 133, 2350.
(
h) Neufeldt, S. R.; Jimen
K. N. J. Am. Chem. Soc. 2015, 137, 9843.
5) (a) Cacchi, S.; Fabrizi, G. Chem. Rev. 2005, 105, 2873.
b) Kochanowska-Karamyan, A. J.; Hamann, M. T. Chem. Rev. 2010,
10, 4489.
6) (a) Horwell, D. C. Tetrahedron 1980, 36, 3123. (b) Kozikowski, A.
́ ́
ez-Oses, G.; Huckins, J. R.; Thiel, O. R.; Houk,
(
(
1
(
indole using Ru(II) or Rh(III) catalysts. The use of
trifluoroacetyl as directing group has been disclosed for the
first time. The reaction is found to be tolerant to a wide variety of
indole derivatives and olefin partners like acrylates, styrenes, and
vinyl phosphonates. Examining different carbonyl-based direct-
ing groups, we found that weak coordinating directing groups
provide exclusive C-4 alkenylated products, while strong
coordinating groups provide C-2 products. Further work
regarding the origin of selectivity is being carried out in our
laboratory.
P.; Ishida, H. J. Am. Chem. Soc. 1980, 102, 4265. (c) Oppolzer, W.;
Grayson, J. I.; Wegmann, H.; Urrea, M. Tetrahedron 1983, 39, 3695.
(
3
d) Hatanaka, N.; Ozaki, O.; Matsumoto, M. Tetrahedron Lett. 1986, 27,
169. (e) Sandtorv, A. H. Adv. Synth. Catal. 2015, 357, 2403.
7) (a) Zhang, Y.; Zheng, J.; Cui, S. J. Org. Chem. 2014, 79, 6490.
b) Ikemoto, H.; Yoshino, T.; Sakata, K.; Matsunaga, S.; Kanai, M. J. Am.
(
(
Chem. Soc. 2014, 136, 5424. (c) Song, Z. Q.; Samanta, R.; Antonchick, A.
P. Org. Lett. 2013, 15, 5662. (d) Li, B.; Ma, J.; Xie, W.; Song, H.; Xu, S.;
Wang, B. J. Org. Chem. 2013, 78, 9345. (e) Sollert, C.; Devaraj, K.;
Orthaber, A.; Gates, P.; Pilarski, L. Chem. - Eur. J. 2015, 21, 5380.
(f) Yang, L.; Zhang, G.; Huang, H. Adv. Synth. Catal. 2014, 356, 1509.
(
5
2
g) Yang, X.-F.; Hu, X.-H.; Feng, C.; Loh, T.-P. Chem. Commun. 2015,
1, 2532. (h) Tang, C.-Y.; Tao, Y.; Wu, X. - Y.; Sha, F. Adv. Synth. Catal.
014, 356, 609. (i) Gong, B.; Shi, J.; Wang, X.; Yan, Y.; Li, Q.; Meng, Y.;
ASSOCIATED CONTENT
Supporting Information
■
*
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.6b02698.
Xu, H. E.; Yi, W. Adv. Synth. Catal. 2014, 356, 137. (j) Shi, J.; Yan, Y.; Li,
Q.; Xu, H. E.; Yi, W. Chem. Commun. 2014, 50, 6483. (k) Zhang, L.; Qiu,
R.; Xue, X.; Pan, Y.; Xu, C.; Wang, D.; Wang, X.; Xu, L.; Li, H. Chem.
Commun. 2014, 50, 12385. (l) Zhang, L.-Q.; Yang, S.; Huang, X.; You, J.;
Song, F. Chem. Commun. 2013, 49, 8830. (m) Xu, L.; Zhang, C.; He, Y.;
Tan, L.; Ma, D. Angew. Chem., Int. Ed. 2016, 55, 321.
Experimental procedure and characterization data for all
compounds (PDF)
X-ray crystallographic data for compounds 4a and 5a
(
CIF)
(8) (a) Lanke, V.; Prabhu, K. R. Org. Lett. 2013, 15, 6262. (b) Liu, Q.;
Li, Q.; Ma, Y.; Jia, Y. Org. Lett. 2013, 15, 4528.
AUTHOR INFORMATION
Corresponding Author
E-mail: prabhu@orgchem.iisc.ernet.in.
(9) Trifluoromethylacetophenone is lost when subjected to high
vacuum. Hence, deuterium labeling has been done with pyrene only.
■
*
(
10) For electron density calculations and prediction of pK values, see
a
the Supporting Information.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by SERB (NO.SB/S1/OC-56/2013),
New Delhi, Indian Institute of Science, and RL Fine Chem. We
■
D
DOI: 10.1021/acs.orglett.6b02698
Org. Lett. XXXX, XXX, XXX−XXX