Dual nucleophilic catalysis with DABCO for the N-methylation of indoles

Article abstract of DOI:10.1021/jo0266644

DABCO is an extremely active catalyst for the methylation of indoles in conjunction with dimethyl carbonate (DMC). This green chemistry is highly effective and produces N-methylindoles in nearly quantitative yields. The reaction sequence consists of competing alkylation and acylation pathways and involves 1,4-diazabicyclo[2.2.2]octane (DABCO) dually as a nucleophilic catalyst, ultimately resulting in a single product: the N-methylated indole.

Full text of DOI:10.1021/jo0266644

Du a l Nu cleop h ilic Ca ta lysis w ith DABCO for th e N-Meth yla tion of
In d oles
Wen-Chung Shieh,* Steven Dell, Andrew Bach, Oljan Repicˇ, and Thomas J . Blacklock
Chemical and Analytical Development, Novartis Institute for Biomedical Research,
East Hanover, New J ersey 07936
wen.shieh@pharma.novartis.com
Received November 5, 2002
DABCO is an extremely active catalyst for the methylation of indoles in conjunction with dimethyl
carbonate (DMC). This green chemistry is highly effective and produces N-methylindoles in nearly
quantitative yields. The reaction sequence consists of competing alkylation and acylation pathways
and involves 1,4-diazabicyclo[2.2.2]octane (DABCO) dually as a nucleophilic catalyst, ultimately
resulting in a single product: the N-methylated indole.
In tr od u ction
sulfate.6 Improved processes have been reported recently
utilizing nontoxic reagents, however, under harsh and
noncatalytic conditions: (1) DMC/DMF/K2CO3/130 °C7
and (2) dimethyl oxalate/DMF/t-BuOK/153 °C.8 Thus,
development of an efficient, safe, and environmentally
friendly method of methylation constitutes an important
challenge. Recently, our lab discovered that the “green”
reagent dimethyl carbonate (DMC) can also function
effectively for the N-methylation of indoles and benzimi-
dazoles under milder conditions if DBU is employed as
the catalyst.9 The only drawback to this method is that
it demands the use of stoichiometric amounts of DBU to
achieve reasonable process efficiency. Hence, we had a
need to develop a more active catalyst and thereby a more
efficient process. Herein, we report a novel methodology
for the methylation of the indole nitrogen in nearly
quantitative yields using a catalytic amount of DABCO
under mild conditions.
Methylation of nitrogen atoms is a versatile biological
process. For example, methylation of cytosine in CpG
dinucleotide governs eukaryotic gene control.1 Epine-
phrine, a catecholamine neurotransmitter, is formed by
the methylation of norepinephrine catalyzed by a trans-
methylase that utilizes S-adenosylmethionine as a meth-
yl donor.2 Nature is sophisticated and adroit in methyl-
ation: it utilizes enzymes in a highly efficient way even
if the process involves unactivated methyl donors. De-
spite the importance of methylation, chemists seem to
lack an efficient methodology in this area. We depend
largely on noncatalytic processes that employ highly
reactive and toxic methylating agents. N-Methylindole
is important within nature as the common nucleus of
indole alkaloids that exhibit a wide array of biological
activities. For example, affinisine has been shown to
produce delayed intention tremors, marked CNS depres-
sant activity, ataxia, hypothermia, and bradypnea,3 and
a synthetic pentacyclic indole has revealed antitumor
properties.4 Common approaches to the synthesis of
N-methylindoles typically require a two-step protocol: (1)
formation of an active indole anion by a stoichiometric
amount of strong base (NaH, KHMDS, KTB, or NaOH)
and (2) reaction of the resulting anion with a toxic and
hazardous reagent such as methyl iodide5 or dimethyl
Resu lts a n d Discu ssion
DABCO, a tertiary amine base with weak basicity (pKa
) 8.7),10 has been used as a catalyst for the Baylis-
Hillman reaction.11 It has also been employed as a
catalyst for the substitution of chloropurine with alco-
hols.12 On the basis of these precedents, we considered
DABCO as a potential catalyst that could react with
DMC to form an activated methylating reagent. The
effectiveness of DABCO serving as a catalyst for the
* Corresponding author.
(1) Zubay, G. Regulation of Gene Expression in Eukaryotes. In
Biochemistry, 3rd ed.; Wm. C. Brown Publishers: Dubuque, Iowa, 1993;
Chapter 30.
(2) Stryer, L. Excitable Membranes and Sensory Systems. In
Biochemistry, 2nd ed.; W. H. Freeman and Company: New York, 1981;
Chapter 37.
(6) For leading references using Me2SO4, see: Ottoni, O.; Cruz, R.;
Alves, R. Tetrahedron 1998, 54, 13915.
(7) J iang, X.; Tiwari, A.; Thompson, M.; Chen, Z.; Cleary, T. P.; Lee,
T. B. Org. Proc. Res. Dev. 2001, 5, 604.
(3) Liu, X.; Wang, T.; Xu, Q.; Ma, C.; Cook, J . M. Tetrahedron Lett.
2000, 41, 6299.
(4) Christophe, S.; Kuehm-Caube`re, C.; Renard, P.; Pfeiffer, B.;
Caube`re, P. Tetrahedron Lett. 1998, 39, 9431.
(8) Bergman, J .; Norrby, P.-O.; Sand, P. Tetrahedron 1990, 46, 6113.
(9) Shieh, W.-C.; Dell, S.; Repicˇ, O. Org. Lett. 2001, 3, 4279.
(10) Frenna, V.; Vivona, N.; Consiglio, G.; Spinelli, D. J . Chem. Soc.,
Perkin Trans. 2 1985, 1865.
(5) For leading references using MeI, see: (a) Hilton, S. T.; Ho, T.
C. T.; Pljevaljcic, G.; Schulte, M.; J ones, K. J . Chem. Soc., Chem.
Commun. 2001, 209. (b) Hilton, S. T.; Ho, T. C. T.; Pljevaljcic, G.; J ones,
K. Org. Lett. 2000, 2, 2639. (c) Tratrat, C.; Giorgi-Renault, S.; Husson,
H.-P. J . Org. Chem. 2000, 65, 6773. (d) de Koning, C. B.; Michael, J .
P.; Rousseau, A. L. J . Chem. Soc., Perkin Trans. 1 2000, 1705.
(11) (a) Yu, C.; Liu, B.; Hu, L. J . Org. Chem. 2001, 66, 5413. (b)
Kaye, P. T.; Robinson, R. S. Synth. Commun. 1996, 26, 2085. (c)
Drewes, S. E.; Emslie, N. D.; Karodia, N. Synth. Commun. 1990, 20,
1915.
(12) Linn, J . A.; McLean, E. W.; Kelly, J . L. J . Chem. Soc., Chem.
Commun. 1994, 913.
10.1021/jo0266644 CCC: $25.00 © 2003 American Chemical Society
Published on Web 02/07/2003
1954
J . Org. Chem. 2003, 68, 1954-1957

Dual Nucleophilic Catalysis with DABCO
TABLE 1. Effect of Ba se on th e Meth yla tion Ra te of
5-Br om oin d olea
TABLE 2. DABCO-Ca ta lyzed N-Meth yla tion of In d oles
w ith DMC
entry
base
none
Bu3N
DMAP
DBU
% 1
% 3b
% 4b
1
2
3
4
5
100
100
14
9
0
0
73
84
0
0
0
12
6
DABCO
0
>99
a
All reactions were conducted on a 1 g (5 mmol) scale in 1 mL
of DMF and 10 mL of DMC using 0.1 equiv of base at 90 °C.
Product distributions were determined by HPLC analysis of the
b
reaction mixture after 5 h. Identity of product was confirmed by
1H and 13C NMR and MS.
methylation of indoles with DMC under mild conditions
was demonstrated for the reaction of 5-bromoindole (1).
In the presence of 0.1 equiv of DABCO, the catalytic
process afforded the N-methyl-5-bromoindole (4) in nearly
quantitative yield within 5 h (Table 1, entry 5). In
contrast, with use of 0.1 equiv of DMAP or DBU, the
same reaction affords the methylcarbamate (3) as the
major component of the reaction mixture (Table 1, entries
3 and 4). This demonstrated that DABCO is a more
effective catalyst in comparison to either DMAP or DBU.
It should be noted that without any catalyst or in the
presence of Bu3N, the reaction generated no product at
all (Table 1, entries 1 and 2).
We examined the structure-reactivity profile for a
series of indoles, the results of which are summarized in
Table 2. In all cases, isolated yields were excellent and
ranged from 95 to 99%.13 Varying the electronic proper-
ties of the substituent on the aromatic ring from electron-
donating (6) to electron-withdrawing (7) enhances the
methylation rate (entries 1 through 4). Increasing the
steric hindrance around the nitrogen atom by introducing
a phenyl group at the 2-position (9) slows the reaction
when compared to the unsubstituted indole (entries 2 and
6). When an additional carboxylic acid functionality was
present (10), 1.1 equiv of DABCO was employed, where
the additional 1 equiv of base functioned as a proton
scavenger. Under these conditions, methyl N-methyl-
indole-2-carboxylate (10b, Chart 1) was isolated in excel-
lent yield (entry 7). Hence, the system provided an
advantage: concurrent methylation of both the indole
nitrogen and a carboxylic acid group can be accomplished.
This is superior to an analogous two-step process.5a The
protocol also worked efficiently for carbazole 11.
a
General procedure: a reaction flask was charged with sub-
strate (1 g), DABCO (0.1 equiv), DMF (1 mL), and DMC (10 mL).
The mixture was heated to 94-95 °C, and the reaction was
monitored by HPLC until trace or no starting substrate was
b
detected (reaction time). Identity of the methylated products was
confirmed by 1H and 13C NMR and MS. c Isolated yield based on
d
starting substrate. Same as procedure in footnote a, except 1.1
equiv of DABCO was charged to the reaction mixture. e Isolated
yield of 10b, methyl N-methylindole-2-carboxylate.
CHART 1
A plausible mechanism consisting of two competing
pathways, acylation and alkylation, in the catalytic cycle
(Scheme 1) is proposed for the methylation of indoles.
As mentioned before, a combination of indole and DMC
is initially inert. However, methylation of indole with
DMC was highly efficient in the presence of DABCO
(Table 1, entries 1 and 5). This implied that DMC must
be transformed by DABCO to an activated species that
can react with indole. It is reasonable to assume that in
the catalytic cycle, DABCO functions as a nucleophilic
catalyst reacting with DMC to generate an ion pair A.14
Deprotonation of indole 12 with the methoxide ion of A
could lead to an indole-containing ion pair B. We ob-
(13) All N-methylated indoles are known. For the methylated indole
1, see: Soll, R. M.; Park, J . A.; Rimele, T. J .; Heaslip, R. J .; Wojdan,
A.; Oshiro, G.; Grimes, D.; Asselin, A. Eur. J . Med. Chem. 1990, 25,
191. For methylindoles 5, 7, and 11, see: ref 8. For indole 6, see: ref
7. For indole 8, see: Tamura, Y. Adachi, M.; Kawasaki, T.; Yasuda,
H.; Kita, Y. J . Chem. Soc., Perkin Trans. 1 1980, 1132. For indole 9,
see: ref 9. For indole 10, see: Kitano, M.; Kojima, A.; Nakano, K.;
Miyagishi, A.; Noguchi, T.; Ohashi, N. Chem. Phar. Bull. 1999, 47,
1538.
J . Org. Chem, Vol. 68, No. 5, 2003 1955

Shieh et al.
SCHEME 1. P la u sible Ca ta lytic Cycle for In d ole
Meth yla tion
functions as a nucleophilic catalyst to regenerate ion pair
B and eventually afford methylated indole 14. The
relative rates of acylation and alkylation for the sub-
stituent series (B, R ) Br, H, OCH3, or NO2) are
consistent with the hardness of the nucleophiles. The
preference of acylation over alkylation for 5-methoxy-
indolyl anions (B, R ) OCH3) during the reaction may
result from a relatively faster reaction of the 5-methoxy-
indolyl anion (a hard nucleophile) with the polarized sp2
carbonyl carbon of the carbamate (a hard electrophile)
than with the sp3 methyl carbon of the carbamate (a
relatively softer electrophile).15 On the other hand, the
electron-deficient 5-nitroindolyl anion behaves as rela-
tively a softer nucleophile due to the stabilization of its
negative charge by delocalization of the charge density
around nitrogen through the π system, facilitating alkyl-
ation leading to the formation of N-methylindole 14.
Anion stability appears qualitatively to control the rate
of the alkylation step. This accounts for the formation of
iminostilbene carbamate (15b,16 Chart 1) as the dominant
product (90% isolated yield, 8% SM) from the carbazole
analogue iminostilbene (15), under the same conditions:
0.1 equiv DABCO/DMC/94-95 °C. Apparently, the non-
planar butterfly form of 1517 inhibits delocalization of the
negative charge on nitrogen into the aromatic rings.
served that the relative rates of methylation for 5-sub-
stituted indoles correlated to the electronegativity of the
substituents and were determined in the order of: NO2
> Br > H > OCH3 (Table 2). In each individual methyl-
ation reaction, formation of significant amounts of a
stable intermediate was observed, as indicated by HPLC
analyses. These intermediates were isolated and deter-
mined to be the corresponding indole carbamates (13, R
) Br, H, OCH3, or NO2) as supported by 1H and 13C NMR,
MS, and elemental analyses. The relative rates of car-
bamate 13 formation were also dependent on the elec-
tronegativity of the substituents and were determined
to be in the order of OCH3 > H > Br > NO2. It was
experimentally verified that these carbamate intermedi-
ates were transformed into the methylated products 14
during the course of the reaction.
It is noteworthy that when the individual indole
carbamates (13, R ) Br, H, OCH3, or NO2) were heated
to 94-95 °C in DMC in the absence of DABCO, no
corresponding N-methylated indole 14 was produced
(Scheme 1). This eliminates the possibility that a ther-
mally induced rearrangement of carbamates 13 was
responsible for the formation of N-methylindole 14.
However, when a catalytic amount of DABCO (0.1 equiv)
was added to the same system, complete conversion to
the methylated indole 14 occurred with the observed
relative rate R ) NO2 > Br > H > OCH3. Observation of
small amounts (<10%) of “free” indole in the reaction
mixture by HPLC during the course of conversion pro-
vided qualitative evidence for the existence of ion pair
B, which could generate indole and DABCO when it is
quenched with HPLC mobile phase. These results sug-
gested that ion pair B could be involved in the indole
methylation reaction. One can envision the existence of
two competing pathways: an acylation step leading to
the indole carbamate and an alkylation step that is
responsible for the methylindole formation (Scheme 1).
The acylation step may be reversible wherein DABCO
Con clu sion
In summary, we developed the first catalytic process
for the N-methylation of the indole family. This protocol
is efficient and affords excellent yields under mild condi-
tions for a variety of indoles. A plausible multistep
mechanism is proposed in which an alkylation pathway
competes with a reversible acylation pathway and DAB-
CO serves twice as a nucleophilic catalyst in the catalytic
cycle. Mechanistic data that are available at this point
do not permit quantitative verification of this hypothesis.
Further mechanistic studies are required to elucidate the
details of the mechanistic pathway(s) leading to the
methylation product.
Exp er im en ta l Section
All reagents and solvents were purchased from commercial
suppliers and used without further purification. All N-meth-
ylindoles 1 and 5-11 are known compounds.13 The identity of
the methylated products was confirmed by 1H and 13C NMR
and MS spectra. Solvents used were of technical grade. 1H and
13C NMR spectra were recorded at 300 and 75 MHz, respec-
tively. DCI used a robotic probe (Scientific Instrument Ser-
vices) with reagent-grade ammonia gas. HPLC analyses were
performed on a Waters HPLC system with a 996 PDA detector
and Waters Symmetry Shield RP8 column (5 µm, 3.9 × 150
mm) at 40 °C.
Gen er a l P r oced u r e for th e Meth yla tion of In d oles
w ith DMC. A reaction flask was charged with an indole
(15) For the hard-soft acid-base concept and definition, see: (a)
Carey, F. A.; Sundberg, R. J . Nucleophilic Substitution. In Advanced
Organic Chemistry, 2nd ed.; Plenum Press: New York, 1984; Part A,
Chapter 5. (b) Ho, T.-L. Hard and Soft Acids and Bases Principle in
Organic Chemistry; Academic Press: New York, 1977.
(16) Compound 15b: IR (KBr) 1711 (s) cm-1 1H NMR (CDCl3) δ
;
7.30-7.41 (m, 8H), 6.92 (s, 2H), 3.65 (s, 3H); 13C NMR (CDCl3) δ 155.5,
134.2, 132.1, 130.6, 130.2, 129.1, 128.5, 127.4, 53.1; EIMS m/z 251 (M+).
Anal. Calcd for C16H13NO2: C, 76.48; H, 5.21; N, 5.57. Found: C, 77.00;
H, 5.14; N, 5.55.
(14) A nucleophilic catalysis mechanism involving DBU was pro-
posed for the esterification of carboxylic acid with DMC. Shieh, W.;
Dell, S.; Repicˇ, O. J . Org. Chem. 2002, 67, 2188.
(17) Reetz, M. T.; Hutte, S.; Goddard, R.; Minet, U. J . Chem. Soc.,
Chem. Commun. 1995, 275.
1956 J . Org. Chem., Vol. 68, No. 5, 2003

Dual Nucleophilic Catalysis with DABCO
substrate (1 g), 1,4-diazabicyclo[2.2.2]octane (DABCO, 0.1
equiv), DMF (1 mL), and dimethyl carbonate (10 mL). The
mixture was heated to 94-95 °C, and the reaction was
monitored by HPLC until a total conversion to the respective
methylindole was detected. The reaction was cooled to room
temperature and diluted with EtOAc (50 mL) and H2O (50
mL). The organic layer was separated and washed in sequence
with H2O (50 mL), 10% aqueous citric acid (2 × 50 mL), and
H2O (4 × 50 mL). The organic layer was dried over Na2SO4,
filtered, and concentrated under vacuum to give the corre-
sponding N-methylindole.
Gen er a l P r oced u r e for t h e P r ep a r a t ion of In d ole
Ca r ba m a tes. A solution of indole or 5-substituted indoles (1
g) and 10 mL of dimethyl carbonate at 94 °C was instanta-
neously charged with a solution of of 1,4-diazabicyclo[2.2.2]-
octane (DABCO, 0.1 equiv) in 1 mL of DMF. The reaction was
stirred at 94 °C for a specified period of time as follows: 1 (10
min), 5 (10 min), 6 (10 min), and 7 (5 min). The reaction
mixture was rapidly cooled to 23 °C and concentrated under
vacuum. The resulting crude product was purified by flash
chromatography (silica gel, 9:1 hexanes/EtOAc).
1H), 7.02 (d, J ) 3.9 Hz, 1H), 4.18 (s, 3H); 13C NMR (acetone-
d6) δ 151.0, 144.2, 138.5, 130.8, 129.2, 119.6, 117.5, 115.4,
108.8, 54.3; (DCI-NH3) m/z 220 (M+). Anal. Calcd for
C10H8N2O4: C, 54.55; H, 3.66; N, 12.72. Found: C, 54.40; H,
3.65; N, 12.64.
5-Br om oin d ole-1-ca r boxylic Acid Meth yl Ester (13c,
1
R ) Br ). White solid: mp 63-65 °C; IR (KBr) 1741 cm-1; H
NMR (CD3CN) δ 8.23 (d, J ) 8.8 Hz, 1H), 7.94 (d, J ) 2.0 Hz,
1H), 7.84 (d, J ) 3.8 Hz, 1H), 7.61 (dd, J ) 8.8, 2.0 Hz, 1H),
6.80 (d, J ) 3.6 Hz, 1H), 4.19 (s, 3H); 13C NMR (CDCl3) δ 150.7,
133.6, 132.0, 126.8, 126.6, 123.2, 116.1, 115.2, 106.6, 53.4;
(DCI-NH3) m/z 253 (M+). Anal. Calcd for C10H8BrNO2: C,
47.27; H, 3.17; N, 5.51; Br, 31.45. Found: C, 47.29; H, 3.16;
N, 5.41; Br, 31.43.
5-Meth oxyin d ole-1-ca r boxylic Acid Meth yl Ester (13d ,
R ) OCH3). White solid: mp 82-85 °C; IR (KBr) 1729 (s)
1
cm-1; H NMR (CD3CN) δ 8.02 (d, J ) 9.0 Hz, 1H), 7.60 (d, J
) 3.7 Hz, 1H), 7.11 (d, J ) 2.6 Hz, 1H), 6.93 (dd, J ) 9.0, 2.6
Hz, 1H), 6.58 (dd, J ) 3.7, 0.4 Hz, 1H), 3.98 (s, 3H), 3.82 (s,
3H); 13C NMR (CD3CN) δ 155.8, 151.0, 131.1, 129.4, 126.0,
115.2, 112.7, 107.3, 103.1, 54.8, 53.2; (DCI-NH3) m/z 206 (M
+ H)+. Anal. Calcd for C11H11NO3: C, 64.56; H, 5.18; N, 8.00.
Found: C, 64.37; H, 5.27; N, 6.80.
In d ole-1-ca r boxylic Acid Meth yl Ester (13a , R ) H).
Colorless oil: IR (KBr) 1741 (s) cm-1 1H NMR18a (CDCl3) δ
;
Gen er a l P r oced u r e for th e Con ver sion of In d ole Ca r -
ba m a tes in to N-Meth ylin d oles. A solution of indole car-
bamate or 5-substituted indole carbamate (100 mg) and 1 mL
of dimethyl carbonate at 94 °C was instantaneously charged
with a solution of of 1,4-diazabicyclo[2.2.2]octane (DABCO, 0.1
equiv) in 100 µL of DMF. The reaction mixture was stirred at
94 °C and analyzed by HPLC periodically until the conversion
to the corresponding N-methylindole was complete.
8.18 (d, J ) 7.9 Hz, 1H), 7.53-7.58 (m, 2H), 7.32 (ddd, J )
7.2, 7.2, 1.1 Hz, 1H), 7.23 (ddd, J ) 7.7, 7.7, 1.1 Hz, 1H), 6.57
(dd, J ) 3.8, 0.8 Hz, 1H), 4.00 (s, 3H); 13C NMR18b (CDCl3) δ
151.5, 135.2, 130.4, 125.5, 124.5, 123.0, 121.0, 115.1, 108.1,
53.7; (DCI-NH3) m/z 176 (M + H)+. Anal. Calcd for C10H9-
NO2: C, 68.56; H, 5.18; N, 8.00. Found: C, 68.83; H, 5.17; N,
8.06.
5-Nitr oin d ole-1-ca r boxylic Acid Meth yl Ester (13b, R
) NO2). Off-white solid: mp 162-165 °C; IR (KBr) 1734 cm-1
;
1H NMR (acetone-d6) δ 8.64 (d, J ) 2.2 Hz, 1H), 8.41 (d, J )
9.2 Hz, 1H), 8.30 (dd, J ) 9.2, 2.3 Hz, 1H), 7.97 (d, J ) 3.8 Hz,
Ack n ow led gm en t. We thank Dr. J effrey McKenna,
Dr. M. Girgis, and Professor R. Boeckman for useful
discussions. We are grateful to Mr. L. Hargiss and Mr.
R. Beveridge for their assistance in obtaining MS and
NMR spectra.
(18) (a) Ito, Y.; Kobayashi, K.; Seko, N.; Saegusa, T. Bull. Chem.
Soc. J pn. 1984, 57, 73. (b) Begtrup M.; Elguero, J .; Faure, R.; Camps,
P.; Estopa´, C.; Ilavsky´, D.; Fruchier, A.; Marzin, C.; Mendoza, J . Magn.
Reson. Chem. 1988, 26, 134.
J O0266644
J . Org. Chem, Vol. 68, No. 5, 2003 1957