A strategy to avoid anomalous O-alkylation of 4-hydroxyindole by diethyl bromomalonate

Article abstract of DOI:10.1002/jhet.5570420610

Alkylation of 4-hydroxyindole with diethyl bromomalonate under standard conditions (potassium carbonate-acetone) gave the expected indolyloxymalonate ester 3 together with the anomalous indolyloxy-2-bromomalonate 4, Depending on conditions, the ratio of t

Full text of DOI:10.1002/jhet.5570420610

A Strategy to Avoid Anomalous O-Alkylation of 4-Hydroxyindole by
Sep-Oct 2005
1101
Diethyl Bromomalonate
George Papageorgiou and John E. T. Corrie
*
National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, U.K.
Received February 22, 2005
Alkylation of 4-hydroxyindole with diethyl bromomalonate under standard conditions (potassium carbon-
ate−acetone) gave the expected indolyloxymalonate ester 3 together with the anomalous indolyloxy-2-bro-
momalonate 4. Depending on conditions, the ratio of the two products varied between 4:1 and 1:2.
Protection of the indole nitrogen by a carbobenzyloxy group eliminated formation of the anomalous product.
J. Heterocyclic Chem., 42, 1101 (2005).
Over the past several years we, and others have devel-
oped and applied the photochemistry of 1-acyl-7-nitroin-
dolines as a means to enable rapid (sub-millisecond) appli-
cation of the neuroactive amino acid L-glutamate onto liv-
ing neuronal cells [1,2]. At present, the 4-methoxy com-
pound 1 is becoming widely used for this purpose, and an
outline of its photochemical cleavage in aqueous solution
is shown in Scheme 1. Although the glutamate conjugate 1
is pharmacologically inactive, i.e. it has no measurable
binding to glutamate receptors prior to photocleavage,
similar conjugates with other neuroactive amino acids
such as GABA and glycine have been shown to bind to
their respective receptors and therefore to blunt the elec-
trophysiological response to photoreleased GABA or
glycine respectively [2b]. We have therefore been inter-
ested to modify the methoxy group to incorporate a center
with high negative charge such as in 2, in the hope of
developing a GABA precursor with such binding reduced
or eliminated. As an early intermediate in this projected
synthesis, we required the indolyloxymalonate 3. Several
aryloxymalonates are known and have normally been pre-
pared by alkylation of the appropriate phenol with a halo-
malonate ester [3,4]. Formation of diaryloxymalonates has
also been observed as a significant by-product in such
reactions, along with products formed by dimerization of
the halomalonate [3,4]. Application of this reaction with 4-
hydroxyindole gave further unexpected complexities and
we describe these results here, together with a means to
synthesize 3 without unwanted side reactions.
to use diethyl bromomalonate as it is significantly less
expensive than the chloro ester. An initial trial reaction used
p-cresol as a model phenol, and treatment with diethyl bro-
momalonate and potassium carbonate in acetone under
reflux gave the known [5] diethyl 4-methylphenoxymalonate
in excellent yield. However, when the same procedure was
applied to 4-hydroxyindole, either as above or with various
modifications of temperature and order of addition of
reagents, two main aromatic products in varying ratio were
always obtained. Tetraethyl 1,1,2,2-ethanetetracarboxylate
[3] was also isolated in low yield from a complex mixture
when the reaction was performed in refluxing butanone, but
was not found in reactions conducted at room temperature in
acetone. The two aromatic products were readily character-
ized as the required malonate 3 and its brominated analogue
4. The ratio of 3 to 4 (isolated yields) varied between 4:1 and
1:2, with formation of 4 apparently being favored by the use
of an excess of the alkylating agent. Trial modifications, such
as using Hunig’s base in acetonitrile or sodium bicarbonate
in DMF, only gave dark reaction mixtures from which no
recognizable product could be obtained.
Scheme 1
Overall photocleavage reaction of 1, showing release of L-glutamate.
So far as we can determine, only two previous examples
of compounds analogous to 4 have been described. The cor-
responding phenoxy compound was formed in low yield by
reaction of sodium phenoxide with diethyl dibromoma-
In the previous work noted above, both chloro- and bro-
momalonates have been used as alkylating agents. We chose

1102
G. Papageorgiou and J. E. T. Corrie
Vol. 42
lonate in ethanol [3] while an aryloxychloromalonate was
recently reported [6] as ~6% of the total product formed,
along with the expected aryloxymalonate, when a tyrosine
derivative was treated with diethyl chloromalonate and
potassium carbonate in acetone, conditions that are similar
to those used in our work above. The mode of formation of
the chloromalonate by-product was not further explored in
the previous report, although it was noted that the chloro
substituent could be removed by hydrogenolysis [6].
The commercial diethyl bromomalonate that we used con-
tained ~8% of the dibromo ester and we initially considered
whether the observed formation of 4 was attributable to the
presence of this impurity. Fractional distillation gave a pure
sample of the monobromo ester but the use of this purified
material did not reduce the yield of 4 compared to that
obtained when the unfractionated commercial material was
used. We have not tried to determine the mechanistic details
of the formation of 4 apart from performing two control
experiments. Use of the pure diethyl dibromomalonate under
the same alkylation conditions (i.e. potassium carbonate in
acetone under reflux) gave only a very dark mixture in which
the bromoester 4 could not be recognized. Thus the forma-
tion of 4 seems not to involve a simple nucleophilic displace-
ment on diethyl dibromomalonate and the very dark color of
these reaction mixtures suggests that redox processes involv-
ing the hydroxyindole are probably involved. A trial alkyla-
tion with diethyl chloromalonate gave a poor recovery of a
crude product in which the expected ester 3 was present in
low yield, together with more polar material.
the electron density of the hydroxyindole in order to sup-
press the putative redox chemistry. We considered that the
easiest means to achieve such deactivation of the indole
was by acylation of its nitrogen atom. The benzyloxycar-
bonyl (Cbz) group offers both stability under the alkyla-
tion conditions and capacity for later removal without per-
turbing the oxymalonate group. Introduction of this sub-
stituent required initial protection of the phenolic group as
its TDBMS ether 5, which was readily achieved with
TBDMS chloride−imidazole (Scheme 2). Subsequent acy-
lation of 5 with Cbz chloride followed a method previ-
ously applied to indole itself [7], and gave the Cbz deriva-
tive 6 in quantitative yield. In this reaction, an excess of
ethylenediamine was added when the acylation was com-
plete in order to remove excess Cbz chloride. All by-prod-
ucts could then be removed by simple, non-chromato-
graphic workup to give material suitable for use in the next
step without further purification. Removal of the silyl
ether with TBAF, buffered with acetic acid to avoid possi-
ble hydrolytic cleavage of the Cbz group, cleanly gave the
protected hydroxyindole 7. This compound has previously
been prepared in modest yield by photolysis of benzyl 2-
oxido-5-isoquinolinylcarbamate, followed by acid treat-
ment of the photolyzate and chromatographic separation of
products [8]. This latter route would not easily be suitable
for large-scale preparation.
Alkylation of 7 with diethyl bromomalonate gave 8 in
81% yield with no evidence of the brominated by-product,
and hydrogenolysis of the Cbz group proceeded cleanly to
give material identical with the sample of 3 previously
obtained. The overall yield from 4-hydroxyindole was
40% and the procedure was easily carried out on a rela-
tively large scale.
Our principal interest was to achieve an effective syn-
thesis of 3 and the most obvious strategy was to decrease
Scheme 2
These results confirm that N-acylation of 4-hydroxyin-
dole was successful in avoiding the unwanted formation of
the bromomalonate derivative 4. The precise mechanism
by which this is achieved remains unclear but it is evident
that the particular combination of 4-hydroxyindole and
diethyl bromomalonate is necessary to produce the anom-
alous by-product 4. In subsequent work, we have found
that simple monobromoacetates readily yield the expected
indolyloxyacetates upon treatment with unprotected 4-
hydroxyindole. Current work is directed to further elabora-
tion of the diester 3 to obtain the GABA conjugate 2.
These results and the pharmacological evaluation of 2 will
be reported in due course.
EXPERIMENTAL
1
H nmr spectra were determined on JEOL FX90Q or Varian
Unityplus 500 spectrometers in CDCl solution with tetramethyl-
3
silane as internal reference. Elemental analyses were carried out
by MEDAC Ltd., Surrey, U.K. Merck 9385 silica gel was used
for flash chromatography. Organic solvents were dried over
Reagents and conditions: (i) TBDMS-Cl, imidazole, CH Cl ; (ii) Cbz-
2
2
Cl, Bu NBr, powdered NaOH, CH Cl ; (iii) TBAF, HOAc, THF; (iv)
4
2
2
diethyl bromomalonate, K CO , (v) acetone; H , Pd-C, EtOH.
2
3
2

Sep-Oct 2005
A Strategy to Avoid Anomalous O-Alkylation of 4-Hydroxyindole
1103
anhydrous sodium sulfate and evaporated under reduced pres-
sure. Hexane solvent (bp 40-60°) was redistilled before use.
(m, 2H, Ar-H), 6.58-6.59 (m, 1H, H-3), 6.52 (dd, J = 7, 1.5 Hz, 1H,
H-5), 1.06 (s, 9H, C(CH ) ), 0.23 (s, 6H, Si(CH ) ).
3 3
3 2
Anal. Calcd. for C H NOSi: C, 67.97; H, 8.56; N, 5.66.
14 21
Reactions of 4-Hydroxyindole with Diethyl Bromomalonate.
Found: C, 67.60; H, 8.80; N, 5.57.
(i) A solution of 4-hydroxyindole (266 mg, 2 mmol) in acetone
(20 mL) was cooled to 0° under nitrogen and anhydrous potas-
sium carbonate (414 mg, 3 mmol) was added. A solution of com-
mercial diethyl bromomalonate (92%; 624 mg, 2.4 mmol) in ace-
tone (20 mL) was added dropwise over 1 h and the mixture was
stirred at 0-5° for 1 h, then at room temperature for further 5 h.
The very dark solution was filtered, the solid was washed with
acetone and the combined filtrates were evaporated. The residue
was dissolved in diethyl ether (50 mL), washed with 0.5 M aq.
sodium hydroxide and brine, dried and evaporated to a brown
viscous oil (535 mg). Flash chromatography (dichloromethane)
gave two products. The less polar material was diethyl 2-bromo-
2-(indol-4-yloxy)malonate 4 as white crystals (92 mg, 11%), mp
Benzyl 4-(tert-Butyldimethylsilyloxy)indole-1-carboxylate (6).
A stirred, ice-cold mixture of 5 (18.56 g, 75 mmol), tetrabutyl-
ammonium bromide (2.42 g, 7.5 mmol) and powdered sodium
hydroxide (4.0 g, 75 mmol) in dichloromethane (375 mL) was
treated dropwise with benzyl chloroformate (95% purity; 20.20
g, 112.5 mmol). The reaction mixture was stirred at room temper-
ature and the progress of the reaction was followed by TLC [ethyl
acetate−hexanes (1:9)]. Further aliquots of benzyl chloroformate
(each 20.20 g) were added after 1 h and 2 h, and the mixture was
stirred at room temperature for a total of 18 h, diluted with water
and extracted with dichloromethane. The combined organic
phases were washed with brine, dried and evaporated. The resid-
ual oil was dissolved in dry diethyl ether (250 mL), cooled in ice
and treated dropwise with a solution of ethylenediamine (47 mL,
355 mmol) in dry diethyl ether (150 mL) and stirred at room tem-
perature for 0.5 h. The mixture was washed with 0.5 M aq.
hydrochloric acid and brine and the organic phase was dried and
evaporated. After trituration with diethyl ether and cooling in ice,
some precipitated N,N-di-(benzyloxycarbonyl)ethylenediamine
was removed by filtration and 6 was isolated as a pale oil (28.62
g, 100%) which was used in the next step without further purifi-
1
92-93° (from diethyl ether−hexanes); H nmr (500 MHz): δ 8.24
(br s, 1H, NH), 7.19 (d, J = 8.1 Hz, 1H, H-7), 7.17 (t, J = 2.7 Hz,
1H, H-2), 7.07 (t, J = 8.1 Hz, 1H, H-6), 6.94 (d, J = 8.1 Hz, 1H,
H-5), 6.73-6.75 (m, 1H, H-3), 4.28-4.38 (m, 4H, OCH ), 1.23 (t,
2
J = 7.1 Hz, 6H, CH ).
3
Anal. Calcd. for C H BrNO : C, 48.76; H, 4.36; N, 3.79.
15 16
5
Found: C, 48.79; H, 4.43; N, 3.61.
The more polar compound was diethyl 2-(indol-4-yloxy)mal-
onate 3 as white crystals (260 mg, 45%), mp 52-53° (from diethyl
1
1
ether−hexanes); H nmr (500 MHz): δ 8.25 (br s, 1H, NH), 7.13
cation; H nmr (90 MHz): δ 7.78 (d, J = 7.2 Hz, 1H, H-7), 7.00-
(dd, J = 3.1, 2.6 Hz, 1H, H-2), 7.09 (d, J = 7.8 Hz, 1H, H-7), 7.05
(t, J = 7.8 Hz, 1H, H-6), 6.76-6.77 (m, 1H, H-3), 6.50 (d, J = 7.5
7.56 (m, 7H, Ar-H), 6.58-6.72 (m, 2H, Ar-H), 5.42 (s, 2H,
PhCH ), 1.03 (s, 9H, C(CH ) ), 0.22 (s, 6H, Si(CH ) ).
2
3 3
3 2
Hz, 1H, H-5), 5.35 (s, 1H, ArOCH), 4.28-4.37 (m, 4H, OCH ),
2
Benzyl 4-Hydroxyindole-1-carboxylate (7).
1.30 (t, J = 7.1 Hz, 6H, CH ).
3
A solution of 6 (28.62 g, 75 mmol) in tetrahydrofuran (375 mL)
containing acetic acid (4.5 g, 75 mmol) was treated at 0° with 1 M
tetrabutylammonium fluoride (75 mL, 75 mmol) and the mixture
was stirred at 0° for 40 min. The solvent was evaporated and the
residue was dissolved in diethyl ether (250 mL) and washed with
saturated aq. sodium bicarbonate and brine, dried and evaporated
to give a brown solid which was washed with cold hexanes to give
7 as white fluffy needles (14.13 g, 70%), mp 118-120° (from
dichloromethane−hexanes), (lit. [8] mp 121-124°).
Anal. Calcd. for C
Found: C, 61.91; H, 6.17; N, 4.68.
H NO : C, 61.85; H, 5.88; N, 4.81.
15 17 5
(ii) A solution of 4-hydroxyindole (266 mg, 2 mmol) in ace-
tone (20 mL) was cooled to 0° under nitrogen and anhydrous
potassium carbonate (414 mg, 3 mmol) was added. A solution of
commercial diethyl bromomalonate (92%; 520 mg, 2 mmol) in
acetone (20 mL) was added dropwise. After stirring for 5 h at
room temperature, more diethyl bromomalonate (92%; 520 mg)
was added and the mixture was stirred at room temperature
overnight. Work up and flash chromatography as above afforded
4 (360 mg, 45%) and 3 (123 mg, 21%).
Benzyl 4-[Di(ethoxycarbonyl)methoxy]indole-1-carboxylate (8).
The indole 7 (10.69 g, 40 mmol) was added to a suspension of
anhydrous potassium carbonate (8.29 g, 60 mmol) in acetone (400
mL) and the mixture was stirred at room temperature for 15 min.
Diethyl bromomalonate (11.47 g, 48 mmol) was added and the
mixture was heated under reflux for 17 h. The solid was filtered off
and washed with acetone and the combined filtrates were evapo-
rated. The residue was dissolved in diethyl ether (150 mL), washed
with 0.5 M aq. sodium hydroxide and brine, dried and evaporated
to give a brown viscous oil (16.87 g). After trituration with diethyl
ether, the precipitated white solid was filtered off and the filtrate
was evaporated. The residue was flash chromatographed [ethyl
acetate−hexanes (15:85)] to give more solid and the combined
solids were recrystallized to give 8 as white crystals (13.83 g,
(iii) A solution of 4-hydroxyindole (266 mg, 2 mmol) in acetone
(20 mL) was cooled to 0° under nitrogen and anhydrous potassium
carbonate (414 mg, 3 mmol) was added. A solution of freshly dis-
tilled pure diethyl bromomalonate (574 mg, 2.4 mmol) in acetone
(20 mL) was added dropwise and the mixture was stirred at room
temperature overnight. Work up and flash chromatography as
above afforded 4 (101 mg, 13%) and 3 (199 mg, 34%).
4-(tert-Butyldimethylsilyloxy)indole (5).
A solution of 4-hydroxyindole (13.31 g, 75 mmol) in dry
dichloromethane (400 mL) was treated with imidazole (8.17 g, 120
mmol) and tert-butyldimethylsilyl chloride (18.09 g, 120 mmol)
and the mixture was stirred at room temperature under nitrogen
overnight. The precipitated white solid was filtered off and washed
with dichloromethane and the combined filtrates were washed suc-
cessively with 0.5 M aq. hydrochloric acid, 0.5 M aq. sodium
hydroxide and brine, dried and evaporated to give 5 as white crys-
1
81%), mp 77-78° (from ethyl acetate−hexanes); H nmr (500
MHz) δ 7.86-7.92 (br d, 1H, H-7), 7.57 (d, J = 3.7 Hz, 1H, H-2),
7.47-7.49 (m, 2H, Ph-H), 7.36-7.43 (m, 3H, Ph-H), 7.20 (t, J = 8.1
Hz, 1H, H-6), 6.86 (d, J = 3.9 Hz, 1H, H-3), 6.63 (d, J = 8.1 Hz, 1H,
1
tals (21.03 g, 85%), mp 80-81° (from hexanes); H nmr (500 MHz):
H-5), 5.45 (s, 2H, PhCH ), 5.30 (s, 1H, ArOCH), 4.29-4.37 (m,
2
δ 8.06 (br s, 1H, NH), 7.08 (dd, J = 3.1, 2.4 Hz, 1H, H-2), 6.99-7.05
4H, OCH ), 1.30 (t, J = 7.1 Hz, 6H, CH ).
2
3

1104
G. Papageorgiou and J. E. T. Corrie
Vol. 42
and J. E. T. Corrie, Tetrahedron, 56, 8197 (2000).
Anal. Calcd for C
H NO : C, 64.93; H, 5.45; N, 3.29.
23 23 7
[2a] M. Canepari, G. Papageorgiou, J. E. T. Corrie, C. Watkins
and D. Ogden, J. Physiol., 533, 756 (2001); [b] M. Canepari, L.
Nelson, G. Papageorgiou, J. E. T. Corrie and D. Ogden, J. Neurosci.
Methods, 112, 29 (2001); [c] M. Matsuzaki, G. C. R. Ellis-Davies, T.
Nemoto, Y. Miyashita, M. Iino and M. Kasai, Nat. Neurosci., 4, 1086
(2001); [d] M. A. Smith, G. C. R. Ellis-Davies and J. C. Magee, J.
Physiol., 548, 245 (2003); [e] M. Canepari and D. Ogden, J.
Neurosci., 23, 4066 (2003); [f] G. Lowe, J. Neurophysiol., 90, 1737
(2003); [g] G. M. G. Shepherd, T. A. Pologruto and K. Svoboda,
Neuron, 38, 277 (2003); [h] M. Canepari, C. Auger and D. Ogden, J.
Neurosci., 24, 3563 (2004); [i] A. G. Carter and B. L. Sabatini,
Neuron, 44, 483 (2004).
Found: C, 65.06; H, 5.64; N, 3.21.
Diethyl 2-(Indol-4-yloxy)malonate (3).
A solution of 8 (8.51 g, 20 mmol) in ethanol (250 mL) was
mixed with 10% palladium on carbon (1.5 g) and hydrogenated at
atmospheric pressure for 1 h until hydrogen uptake ceased. The
catalyst was filtered off through a Celite bed and washed with
ethanol, and the filtrate was evaporated to give a brown oil.
Filtration through flash silica with ethyl acetate−hexanes (3:2) as
eluent, followed by trituration with diethyl ether−hexanes at -20°
gave 3 as white crystals (4.78 g, 82%), mp 52-53° (from diethyl
ether−hexanes) identical to the material described above.
[3] J. B. Niederl and R. T. Roth, J. Am. Chem. Soc., 62, 1154
(1940).
Acknowledgements.
[4] K. H. Takemura, M. Pulickal and F. O. Hoff, J. Org.
Chem., 36, 3646 (1971).
[5] W. A. West and B. J. Ludwig, J. Am. Chem. Soc., 74, 4466
(1952).
[6] S. D. Larsen, T. Barf, C. Liljebris, P. D. May, D. Ogg, T. J.
O’Sullivan, B. J. Palzuk, H. J. Schostarez, F. C. Stevens and J. E.
Bleasdale, J. Med. Chem., 45, 598 (2002).
We are grateful to Dr Ranjit Munasinghe for recording nmr
spectra and to the MRC Biomedical NMR Centre for access to
facilities. This work was supported in part by NIH Grant HL
19242.
REFERENCES AND NOTES
[7] A. C. Weedon and B. Zhang, Synthesis, 95 (1992).
[8] C. Kaneko, W. Okuda, Y. Karasawa and M. Somei, Chem.
Lett., 547 (1980).
[1a] G. Papageorgiou, D. C. Ogden, A. Barth and J. E. T.
Corrie, J. Am. Chem. Soc., 121, 6503 (1999); [b] G. Papageorgiou