Improved synthesis of 4-/6-substituted 2-carboxy-1H-indole-3-propionic acid derivatives and structure-activity relationships as GPR17 agonists

Article abstract of DOI:10.1039/c3md00309d

The orphan G protein-coupled receptor GPR17 was shown to be involved in myelin repair and has been proposed as a novel drug target for the treatment of brain and spinal cord injury and for multiple sclerosis. Recently, 3-(2-carboxy-4,6-dichloro-indol-3-yl

Full text of DOI:10.1039/c3md00309d

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Improved synthesis of 4-/6-substituted 2-carboxy-
1H-indole-3-propionic acid derivatives and
structure–activity relationships as GPR17 agonists†
a
Cite this: DOI: 10.1039/c3md00309d
Younis Baqi, Samer Alshaibani,b Kirsten Ritter,b Aliaa Abdelrahman,b
*
c
c
b
*
Andreas Spinrath, Evi Kostenis and Christa E. Muller
¨
The orphan G protein-coupled receptor GPR17 was shown to be involved in myelin repair and has been
proposed as a novel drug target for the treatment of brain and spinal cord injury and for multiple
sclerosis. Recently, 3-(2-carboxy-4,6-dichloro-indol-3-yl)propionic acid (MDL29,951, 1a) was discovered
and characterized as a potent synthetic GPR17 agonist. In the present study we substantially optimized
the preparation of 1a, which is carried out via Japp–Klingemann condensation of 3,5-
dichlorophenyldiazonium chloride and deprotonated 2-(ethoxycarbonyl)cyclopentanone yielding
phenylhydrazone derivative 5a followed by Fischer indole (diaza-Cope) rearrangement.
A robust
synthesis of 1a (75% yield) was developed to allow upscaling of the procedure. The developed method
was applied to the synthesis of a series of 10 derivatives, eight of which represent new compounds.
Biological evaluation in calcium mobilization assays using 1321N1-astrocytoma cells recombinantly
expressing the human GPR17 provided first insights into their structure–activity relationships. 3-(2-
Carboxy-4,6-dibromo-indol-3-yl)propionic acid (1b) showed similar potency to 1a and represents the
most potent synthetic GPR17 agonist described to date with an EC50 value of 202 nM.
Received 15th October 2013
Accepted 13th November 2013
DOI: 10.1039/c3md00309d
www.rsc.org/medchemcomm
effects.2,4,7,8 Recently, Hennen et al. identied 3-(2-carboxy-4,6-
dichloro-indol-3-yl)propionic acid (MDL29,951, 1a, Fig. 1) as a
Introduction
synthetic agonist for GPR17 and characterized it broadly in
recombinant and native cells.2 Compound 1a showed high
potency at GPR17 in the nanomolar range; the determined EC50
value was found to be dependent on the assay system and the
receptor expression level.2
The described synthesis of 1a provides only moderate overall
yields. For extended studies of GPR17 using 1a as a tool
compound, and for setting up a high-throughput (HTS)
screening assay to identify GPR17 antagonists gram amounts of
the agonist are required. Therefore the goal of the present study
was to develop a signicantly improved synthetic procedure for
1a by carefully studying and optimizing the critical reaction
steps. Furthermore, the new method was to be applied to the
preparation of analogs to study their structure–activity rela-
tionships (SARs).
The orphan G protein-coupled receptor 17 (GPR17) belongs to
the large family of rhodopsin-like class A G protein-coupled
receptors (GPCRs). It is coupled to inhibition of adenylate
cyclase via Gi proteins resulting in decreased intracellular cAMP
levels, and to Gq proteins which activate phospholipase C
leading to IP3-mediated intracellular calcium release.1,2 GPR17
was recently shown to be involved in myelin repair and has
therefore been proposed as a novel drug target for the treatment
of multiple sclerosis, brain and spinal cord injury, and neuro-
degenerative diseases.2–6 Thus, the development of GPR17
modulators is of high pharmacological relevance. Several
compounds have been postulated as physiological agonists of
GPR17, including cysteinylleukotrienes (CysLTs) C4 and D4,
UDPglucose, UDPgalactose, and UDP.1 However, several groups,
including ours, were unable to conrm the described
aDepartment of Chemistry, Faculty of Science, Sultan Qaboos University, PO Box 36,
Postal Code 123, Muscat, Oman. E-mail: baqi@squ.edu.om; Fax: +968 2414 1469;
Tel: +968 2414 2347
bPharma-Zentrum Bonn, Pharmazeutisches Institut, Pharmazeutische Chemie I,
¨
Universitat Bonn, Bonn, Germany. E-mail: christa.mueller@uni-bonn.de; Fax: +49
228 73 2567; Tel: +49 228 73 2301
cInstitute of Pharmaceutical Biology, Section Molecular-, Cellular-, and
Pharmacobiology, University of Bonn, Bonn, Germany
† Electronic supplementary information (ESI) available: Analytical data of the
di-ethyl ester indole derivatives (9a–j).See DOI: 10.1039/c3md00309d
Fig. 1 The first reported synthetic GPR17 agonist.2
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
had not been reproducible by three different experienced
chemists in our laboratory using the previously reported
method.
Results and discussion
Syntheses
The complete optimized synthetic procedure for indole
derivative 1a as well as its analogs 1b–j is depicted in Scheme 2
and Scheme 3. In the rst step, the substituted anilines (2) are
diazotized by adding sodium nitrite in the presence of hydrogen
chloride at low temperature (0–5 ꢀC) to yield diazonium salts 3,
followed by the addition of an aqueous mixture of 2-(ethox-
ycarbonyl)cyclopentanone (4) and KOH. The resulting mixture
is treated with ice and stirred at 40 ꢀC until the phenylhydrazine
mono-ethyl ester 5a–j is formed (the reaction is monitored by
TLC). Carboxylic acids 5a–j are then reuxed in ethanol at 100
ꢀC in the presence of sulfuric acid to obtain the di-ethyl esters of
phenylhydrazones 8a–j, which are subsequently cyclized to the
corresponding indole di-ethyl ester derivatives 9a–j. Finally,
compounds 9a–j are saponied to yield the desired indoles 1a–j
containing two carboxylic acid functions (Scheme 3). This
improved method was applied to the synthesis of 10 analogous
indole derivatives, of which eight were new, not previously
described compounds (1b, c, e, f–j, Table 1). Besides 4,6-
disubstituted indole derivatives (1a–e), which were obtained
starting from symmetrically 3,5-disubstituted anilines, 4- and 6-
monosubstituted derivatives (1g–j), were prepared starting from
meta-mono-substituted anilines. Symmetrically substituted
aniline derivatives produced a single product (1a–f, Table 1),
while meta-mono-substituted anilines gave two isomers: 4- and
6-substituted indole derivatives (1g–j, Table 1). The separation
of the two formed isomers was achieved on the di-ester stage by
silica gel column chromatography of indoles 9g/9h, and 9i/9j,
respectively, using 20% ethyl acetate in cyclohexane as an
eluent. All nal products were analyzed by 1H- and 13C-NMR
spectroscopy, elemental analysis and high-performance liquid
chromatography (HPLC) coupled with electrospray ionization
(ESI) mass spectrometry (MS). Purity as determined by HPLC-
ESI-MS was in all cases greater than 95%.
The synthetic pathway to indole derivative 1a and its analogs
starts with a Japp–Klingemann reaction of aniline derivatives 2
with 2-(ethoxycarbonyl)cyclopentanone (4) to obtain the corre-
sponding phenylhydrazone derivatives 5. Subsequent heating in
the presence of a strong acid (Fischer indole synthesis, a special
type of the (diaza)-Cope rearrangement) leads to ring closure
producing the indole mono-ethyl esters 6, which can be
saponied to yield the desired products 1 (Scheme 1).2,9 The
published procedures for the preparation of 1a suffer from low
yields.2,9 Therefore, we systematically studied the low-yielding
Fischer indole cyclization of dichlorophenylhydrazone deriva-
tive 5a using four different protocols (Scheme 2). At rst we
applied the procedure published by Salituro et al.9 using
concentrated sulfuric acid in reuxing ethanol, which in our
hands resulted in a low yield of 10.5% (Scheme 2).2 By applying
polyphosphoric acid in reuxing toluene, or in acetic acid,
respectively, formation of the expected product was not
observed at all. Finally we applied para-toluenesulfonic acid in
reuxing ethanol,9 and again only a low yield of 6a was obtained
(9%, Scheme 2).
Since we suspected that the free carboxylate function of 5a
might be involved in side-reactions we decided to convert it to
an ester group prior to Fischer indole cyclization. This simple
intermediate reaction step, combined with optimized cycliza-
tion reaction conditions, led to a dramatically increased yield of
indole 6a. Thus, mono-ester 5a was converted to its di-ethyl
ester 8a, which was subsequently cyclized to the indole deriva-
tive 9a according to the following conditions (Scheme 3):
compound 8a was added to a p-toluenesulfonic acid (p-TSA)
solution in toluene, which had previously been dried by means
of a Dean–Stark apparatus in reuxing toluene for ca. 1 h (for
more details see the Experimental part). The solution was
heated for 5 h under the same conditions. The ester functions of
indole 9a were subsequently hydrolyzed. With these modica-
tions, an overall yield of the GPR17 agonist 1a of 75% was
obtained starting from the corresponding aniline derivative
(3,5-dichloroaniline), as compared to only 10.5% in our
previous study,2 or 40% reported by Salituro et al.,9 a yield that
Biological evaluation
The compounds were investigated for their potential to induce
calcium mobilization in 1321N1 astrocytoma cells stably
Scheme 1 General Japp–Klingemann/Fischer indole synthesis pathway for the preparation of indole derivatives 1.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Scheme 2 Different cyclization methods of compound 5a.
Scheme 3 Improved procedure for the synthesis of 3-(2-carboxy-indol-3-yl)propionic acid derivatives.
transfected with the human GPR17 using the calcium-chelating electronegativity of the 4,6-substituents appears to be required
uorescent dye Oregon Green®.
for potency (compare e.g. 1d and 1e).
Introducing a small substituent (uoro) in position 5 of lead
structure 1a signicantly decreased potency by >150-fold
(compare 1a with 1f, Table 1). This means that a substituent in
position 5, at least one with a high electronegativity, is hardly
tolerated by the receptor. Finally, derivatives substituted either
in position 4 or 6 of the indole moiety with a halogen atom (Br,
I) were investigated. It was found that a bulky halogen atom in
position 6 was very important for high potency. The larger
iodine atom was superior to a bromine atom in that position
(compare 1h and 1j). In contrast, indole derivatives 1g and 1i,
which were only substituted in position 4 but not in position 6,
were only weak agonists at GPR17. Based on a series of 10 indole
derivatives it can be recognized that their SARs as GPR17
agonists are steep, and small modications can modulate
potency and even abolish their activity.
Structure–activity relationships
GPR17 has been shown to be coupled to different G proteins,1,2
including Gq, which mediates the release of inositol tri-
sphosphate (IP3). IP3 binds to ryanodine receptors on the
endoplasmic reticulum and triggers the opening of calcium
channels resulting in an increase in intracellular calcium levels.
In 1321N1 astrocytoma cells recombinantly expressing
moderate levels of the human GPR17 an EC50 value of 330 nM
was determined for 1a in calcium mobilization assays (see Table
1). The Hill slope of the curve for 1a as well as for all other
investigated indole derivatives was not signicantly different
from 1.
In the present study initial SARs were obtained for a set of 10
compounds. Different tri- and tetra-substituted indole deriva-
tives were investigated (see Table 1). Analogs of the 4,6-dichloro-
substituted indole derivative 1a, in which the chlorine atoms
were replaced by bromine or iodine atoms, or by methoxy or
triuoromethoxy groups, showed the following rank order of
Experimental part
General
potency: dibromo (1b) > dichloro (1a) > diiodo (1c) > di-(tri- All materials were used without prior purication. Thin-layer
uoromethoxy) (1e) [ dimethoxy (1d). These results indicate chromatography was performed using TLC aluminum sheets
that the size of the substituents is important for receptor silica gel 60 F254. Synthesized compounds were visualized under
interaction. The dibromo-substituted indole showed a some- UV light (254 nm). 1H- and 13C-NMR data were measured in
what higher potency (EC50 ¼ 202 nM) than the dichloro- DMSO-d6 as a solvent. Chemical shis are reported in parts per
substituted lead structure 1a, although the difference was not million (ppm) relative to the deuterated solvent (DMSO-d6), d
statistically signicant. However, larger substituents were less 1H: 2.49 ppm, 13C: 39.7 ppm, coupling constants J are given in
well tolerated and led to reduced potency. Besides sterical Hertz and spin multiplicities are given as s (singlet), d (doublet),
effects, electronic effects are also important: high t (triplet), q (quartet), sext (sextet), m (multiplet), br (broad). The
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Table 1 Synthesized indole derivatives (1a–j), overall yields, melting points, and potency to stimulate GPR17
Product
Aniline derivative
Product (1a–j)
m.p. (ꢀC)
Yielda (%)
EC50 ꢁ SEMb (mM)
0.331 ꢁ 0.087
1a
275–277
75
1b
1c
1d
1e
1f
259–260
280–282
239–241
249–251
292–294
235–237
234–235
63
0.202 ꢁ 0.063
4.79 ꢁ 0.877
[100c
35
36
45
17.1 ꢁ 4.12
52.0 ꢁ 5.50
45.5 ꢁ 8.66
2.24 ꢁ 0.68
65
1g
23 (69% total)
1h
46 (69% total)
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Table 1 (Contd.)
Product
Aniline derivative
Product (1a–j)
m.p. (ꢀC)
Yielda (%)
EC50 ꢁ SEMb (mM)
1i
246–248
30 (90% total)
36.8 ꢁ 1.81
1j
230–232
60 (90% total)
0.715 ꢁ 0.128
a
b
c
Total isolated yield. Potency to induce calcium mobilization in 1321N1 astrocytoma cells transfected with the human GPR17. Also no
antagonistic activity was observed.
purities of isolated products were determined by ESI-mass
spectra obtained on an HPLC-MS instrument (LC-MS) using the
same procedure as previously published.10 The purity of the
compounds was determined at 254 nm. For further proof of
purity of the nal compounds (1a–j), elemental analysis and 1H-
NMR were determined. Melting points were measured on a
melting point apparatus and are uncorrected.
General procedure B
Synthesis of compounds 8a–j. Compound 5a–j (10 mmol),
obtained from general procedure A, was dissolved in absolute
ethanol (100 ml) followed by the addition of concentrated
sulfuric acid (2.7 ml, 50.5 mmol, 5.1 equiv.). The mixture was
ꢀ
then allowed to reux for 1 h at 100 C. Then the ethanol was
evaporated and the residue was treated with 100 ml of ice water.
The aqueous solution was extracted with dichloromethane (3 ꢂ
50 ml); the organic layer was dried over magnesium sulfate,
ltered and concentrated. The residue was puried by silica gel
column chromatography using 20% of ethyl acetate in cyclo-
hexane yielding a white solid in 85–100% yield.
General procedures for the synthesis of the products 1a–j
General procedure A
(i) Preparation of benzenediazonium salt derivatives (mixture
I). To a well stirred suspension of aniline derivatives (10 mmol)
in 16.6 ml aq. HCl (5 M) at 0–5 ꢀC was dropwise added a solution
of sodium nitrite (1.38 g, 20 mmol, 2 equiv.) in 8 ml water,
previously cooled to 0–5 ꢀC in an ice bath. The addition of
sodium nitrite solution was slow, in order to keep the temper-
ature of the mixture below 8 ꢀC. The resulting orange-red
General procedure C
Synthesis of indole di-ethyl esters (9a–j). A mixture of p-tolue-
nesulfonic acid (2.954 g, 15 mmol, 1ꢀ.5 equiv.) and 100 ml of dry
toluene was reuxed for 1 h at 140 C; water was continuously
removed by means of a Dean–Stark trap. Subsequently, 10
mmol of the starting material 8a–j dissolved in a minimum
amount of dry toluene (ca. 15 ml) was added and the mixture
was reuxed for 5 h. Then it was allowed to cool down to rt, and
toluene was removed under reduced pressure and the residue
was dissolved in dichloromethane and washed with water. The
organic layer was dried over magnesium sulfate, ltered, and
evaporated to dryness. The residue was puried by silica gel
column chromatography using 20% of ethyl acetate/cyclo-
hexane as eluent.
ꢀ
mixture was stirred at 0–5 C for additional 20 min in an ice
bath.
(ii) Preparation of 2-(ethoxycarbonyl)cyclopentanone anion
(mixture II). 2-(Ethoxycarbonyl)cyclopentanone (2.512 ml, 1.344
g, 15 mmol) was dissolved in ethanol (4.2 ml) and cooled to 0–5
ꢀC. Then, a potassium hydroxide solution (5.040 ꢀg, 90 mmol, 6
equiv.) in water (5 ml) previously cooled to 0–5 C was added
dropwise within ca. 30 min in order to keep the temperature
below 8 ꢀC. The mixture turned to a white-milky appearance,
General procedure D
ꢀ
Saponication of indole diethyl esters 9a–j yielding the nal
products 1a–j. Compound 9a–j (10 mmol) was dissolved in 25 ml
tetrahydrofuran (THF) with stirring at rt. Then a solution of 1.26
g of lithium hydroxide trihydrate (3 equiv.) in 25 ml water was
added and the resulting mixture was le to stir at rt for 24 h.
Aer completion of the reaction, THF was removed under
reduced pressure, the pH was adjusted to 4–5, and the product
was extracted with diethyl ether (3 ꢂ 30 ml). The organic layers
were dried over magnesium sulfate, ltered, and evaporated to
dryness to yield the nal products (1a–j) as solids in excellent
isolated yield (95–100%).
and the nal mixture was stirred at 0–5 C for further 30 min.
(iii) Synthesis of compounds 5a–j. Ice (50 g) was added to
mixture II with stirring at 0–5 ꢀC in an ice bath, followed by the
ꢀ
addition of mixture I, and stirring continued for 1 h at 40 C.
The combined mixtures were then let to cool to rt and the pH
was subsequently adjusted to 4–5 by adding 1 M aq. HCl. The
desired product was extracted with diethyl ether (3 ꢂ 50 ml).
The combined organic layers were collected, dried over
magnesium sulfate, ltered, and the ltrate was evaporated to
dryness yielding a gummy material (95–100%). This material
was used without further purication for the next step.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
(dd, 1H, J ¼ 7.5 Hz, 6-H); 7.25 (dd, 1H, J ¼ 1.8 Hz, 5-H); 7.42 (dd,
1H, J ¼ 1.8 Hz, 7-H); 11.82 (s, 1H, NH); 12.49 (b, 2H, 2CO2H).
13C-NMR (DMSO-d6) d 19.9 (C-20), 36.4 (C-10), 112.5 (C-7); 114.4
(C-4); 121.2 (C-3); 124.2 (C-2); 124.3 (C-5); 125.4 (C-6); 125.8 (C-
3a); 137.6 (C-7a); 163.0 (20-CO2H); 173.8 (2-CO2H). Elemental
analysis for C12H10BrNO4: calcd; C, 46.18; H, 3.23; N, 4.49,
Analytical data of the synthesized products 1a–j
3-(2-Carboxyethyl)-4,6-dichloro-1H-indole-2-carboxylic acid
(1a). 1H-NMR (DMSO-d6) d 2.47 (m, 2H, 20-H); 3.49 (m, 2H, 10-H);
7.14 (d, 1H, 2J ¼ 1.8, 5-H); 7.38 (d, 1H, 2J ¼ 1.8, 7-H); 11.95 (s, 1H,
13
NH); 12.70 (b, 2H, 2CO2H). C-NMR (DMSO-d6) d 20.21 (C-20),
36.24 (C-10), 111.38 (C5), 120.65 (C7), 121.06 (C4), 122.24 (C6),
126.55 (C3a), 127.35 (C7a) 128.75 (C2) 137.41 (C3), 162.68 (20-
CO2H), 173.69 (2-CO2H). LC-MS (m/z): 300 [M]ꢃ. Purity (LC-MS):
98.5%.
+ +
found C, 46.33; H, 3.59; N, 4.58%. LC-MS (m/z): 329 [M + NH4 ] ,
312 [M]+, 312 [M]ꢃ. Purity (LC-MS): 99.1%.
3-(2-Carboxyethyl)-6-bromo-1H-indole-2-carboxylic acid (1h).
1H-NMR (DMSO-d6) d 2.50 (m, 2H, 20-H); 3.23 (m, 2H, 10-H); 7.16
(dd, 1H, J ¼ 1.8 Hz, 5-H); 7.54 (dd, 1H, J ¼ 1.8 Hz, 7-H); 7.64 (dd,
1H, J ¼ 8.5 Hz, 4-H); 11.57 (s, 1H, NH); 12.54 (b, 1H, 2CO2H).
13C-NMR (DMSO-d6) d 19.9 (C-20); 35.2 (C-10); 114.9 (C-7); 117.6
(C-4); 121.6 (C-3); 122.5 (C-6, C-5); 125.0 (C-2); 126.2 (C-3a); 136.8
(C-7a); 163.0 (20-CO2H); 174.0 (2-CO2H). LC-MS (m/z): 329 [M +
3-(2-Carboxyethyl)-4,6-dibromo-1H-indole-2-carboxylic acid
(1b). 1H-NMR (MeOH-d4): d 2.67 (t, 2H, 3J ¼ 8.4 Hz, CH2); 3,71 (t,
2H, 3J ¼ 8.4 Hz, CH2); 7.42 (d, 1H, 4J ¼ 1.6 Hz, C7–H); 7,62 (d, 1H,
4J ¼ 1.6 Hz, C5–H). 13C-NMR (MeOH-d4): d 21.4 (C-20); 37.9 (C-10);
115.9 (C-7); 117.0 (C-4); 118.8 (C-6); 123.9 (C-3); 125.5 (C-2); 127.9
(C-3a); 128.2 (C-5); 139.6 (C-7a); 164.6 (20-CO2H); 170.1 (2-CO2H).
LC-MS (m/z): 392.0 [M]+; 390.0 [M]ꢃ. Purity (LC-MS): 98.8%.
+ +
NH4 ] , 312 [M]+, 312 [M]ꢃ. Purity (LC-MS): 98.9%.
3-(2-Carboxyethyl)-4-iodo-1H-indole-2-carboxylic acid (1i).
1H-NMR (DMSO-d6) d 2.49 (m, 2H, 20-H); 3.35 (m, 2H, 10-H); 6.94
(t, 1H, J ¼ 8 Hz, 6-H); 7.45 (d, 1H, J ¼ 8.2 Hz, 5-H); 7.55 (d, 1H, J
¼ 8.2 Hz, 7-H); 11.72 (s, 1H, NH); 12.60 (b, 2H, 2CO2H). 13C-NMR
(DMSO-d6) d 19.00 (C-20); 36.54 (C-10); 86.3 (C-4); 113.1 (C-7);
121.7 (C-3); 125.8 (C-3a); 125.9 (C-6); 126.5 (C-2); 131.7 (C-5);
137.0 (C-7a); 163.0 (20-CO2H); 173.70 (2-CO2H). Elemental
analysis for C12H10INO4: calcd: C, 40.14; H, 2.81; N, 3.90, found:
C, 40.58; H, 3.04; N, 3.85%. LC-MS (m/z): 377 [M + NH4+]+, 360
[M]+, 358 [M]ꢃ. Purity (LC-MS): 98.8%.
3-(2-Carboxyethyl)-4,6-diiodo-1H-indole-2-carboxylic
acid
(1c). 1H-NMR (DMSO-d6) d 2.48 (m, 2H, 20-H); 3.49 (m, 2H, 10-H);
7.78 (dd, 2H, J ¼ 1.7 Hz, 5-H, 7-H); 11.80 (s, 1H, NH); 12.68 (b,
13
2H, 2CO2H). C-NMR (DMSO-d6) d 18.9 (C-20); 36.4 (C-10), 88.3
(C-4); 89.5 (C-6); 121.4 (C-7); 122.0 (C-3); 126.2 (C-2); 126.4 (C-
3a); 137.8 (C-7a); 138.1 (C7); 162.7 (20-CO2H); 173.58 (2-CO2H).
+ +
LC-MS (m/z): 503 [M + NH4 ] , 486 [M]+, 484 [M]ꢃ. Purity (LC-
MS): 97.9%.
3-(2-Carboxyethyl)-4,6-dimethoxy-1H-indole-2-carboxylic acid
3
(1d). 1H-NMR (DMSO-d6) d 2.43 (t, 2H, J ¼ 8.3 Hz, H-10), 3.34
3-(2-Carboxyethyl)-6-iodo-1H-indole-2-carboxylic acid (1j).
1H-NMR (DMSO-d6) d 2.49 (m, 2H, 20-H); 3.22 (m, 2H, 10-H); 7.31
(dd, 1H, J ¼ 1.8 Hz, 5-H); 7.50 (d, 1H, J ¼ 8.5 Hz, 4-H); 7.74 (d,
1H, J ¼ 1.8 Hz, 7-H); 11.52 (s, 1H, NH); 12.53 (b, 2H, 2CO2H).
13C-NMR (DMSO-d6) d 19.9 (C20); 35.2 (C10); 89.6 (C-6); 121.0 (C-
7); 121.6 (C-3); 122.7 (C-4); 124.6 (C-2); 126.5 (C-3a); 127.9 (C-5);
137.3 (C-7a); 163.0 (20-CO2H); 174.0 (2-CO2H). LC-MS (m/z): 377
(t, 2H, 3J ¼ 8.3 Hz, H-20), 3.74, 3.81 (2 s, each 3H, OCH3), 6.11
4
4
(d, 1H, J ¼ 2.0 Hz, H-5), 6.40 (d, 1H, J ¼ 2.0 Hz, H-7), 11.15
(s, 1H, NH), 12.24 (s, 2H, CO2H). 13C-NMR (DMSO-d6) d 21.3
(C-10), 36.0 (C-20), 55.3, 55.4 (2C, OCH3), 86.7 (C-5), 92.1 (C-7),
112.3 (C-3a), 121.6 (C-3), 122.6 (C-2), 138.2 (C-7a), 155.9 (C-4),
159.2 (C-6), 163.1, 174.2 (2C, CO2H). Elemental analysis for
C
14H15NO6: calcd: C 57.34; H 5.16; N 4.78; O 32.73; found: C
+ +
[M + NH4 ] , 360 [M]+, 358 [M]ꢃ. Purity (LC-MS): 97.8%.
57.47; H 5.22; N 4.61%. LC-MS (m/z): 294.30 [M]+, 292.25 [M]ꢃ.
Purity (LC-MS): 98.9%.
3-(2-Carboxyethyl)-4,6-di(triuoromethyl)-1H-indole-2-carboxy- Biological evaluation
1
lic acid (1e). H-NMR (DMSO-d6) d 2.43 (m, 2H, 20-H); 3.33
1321N1 astrocytoma cells stably transfected with the human
(m, 2H, 10-H); 7.67 (s, 1H, 1H, 5-H); 8.07 (s, 1H, 7-H); 12.67
(s, 1H, NH); 12.92 (b, 2H, 2CO2H). 13C-NMR (DMSO-d6) d 20.0
(C-20); 35.0 (C-10); 114.6 (C-4); 115.2 (C-6); 119.5 (C-7); 120.9,
121.8, 122.1, 122.3 (CF3–C-4); 122.7 (C-2); 123.1 (C-3); 123.3 (C-
3a); 123.5, 123.8, 124.9, 125.2 (CF3–C-6); 129.1 (C-5); 136.2 (C-
7a); 160.7 (20-CO2H); 172.0 (2-CO2H). Elemental analysis for
GPR17 were used for uorimetric measurement of intracellular
calcium release induced by the test compounds in analogy to
previously described procedures.2,11
Conclusions
C
14H9F6NO4: calcd: C, 45.54; H, 2.46; N, 3.79, found: C, 45.94;
In conclusion, we have developed a fast, efficient, and high-
yielding procedure for the synthesis of 3-(2-carboxy-4,6-
dichloro-indol-3-yl)propionic acid (1a, MDL29,951), which will
allow the preparation of multi-gram amounts of this recently
discovered potent GPR17 agonist. The developed method was
used for the total synthesis of 10 indole derivatives, eight of
which are new compounds (1b, c, e, f–j), while two derivatives
(1a, d) had been previously reported.9 Compound 1b (3-(2-car-
boxy-4,6-dibromo-indol-3-yl)propionic acid) showed the highest
potency of the tested compound series at human GPR17
expressed in 1321N1 astrocytoma cells with an EC50 value of 202
nM. Steep SARs have been discovered for this class of
compounds and further, more extensive exploration is
H, 2.54; N, 3.82%. LC-MS (m/z): 387 [M + NH4+]+, 370 [M]+, 368
[M]ꢃ. Purity (LC-MS): 98.9%.
3-(2-Carboxyethyl)-4,6-dichloro-4-uoro-1H-indole-2-carboxy-
1
lic acid (1f). H-NMR (DMSO-d6) d 2.48 (m, 2H, 20-H); 3.48
2
(m, 2H, 10-H); 7.49 (dd, 1H, J ¼ 6 Hz, 7-H); 11.98 (s, 1H, NH);
12.72 (b, 2H, 2CO2H). 13C-NMR (DMSO-d6) d 20.1 (C-20); 36.1 (C-
10); 112.5 (C-7); 113.0, 113.2, 117.6, 117.7, 121.3, 121.4 (C-5);
122.3 (C-4); 127.7 (C-6); 132.5 (C-3); 147.2 (C-3a); 149.0 (C-2);
+ +
162.5 (20-CO2H), 173.6 (2-CO2H). LC-MS (m/z): 337 [M + NH4 ] ,
320 [M]+, 318 [M]ꢃ. Purity (LC-MS): 98.6%.
3-(2-Carboxyethyl)-4-bromo-1H-indole-2-carboxylic acid (1g).
1H-NMR (DMSO-d6) d 2.49 (m, 2H, 20-H); 3.54 (m, 2H, 10-H); 7.11
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
warranted. New ligands for GPR17 are of great interest as
pharmacological tools and as potential drug candidates.
U. Guerrini, E. Tremoli, P. Rosa, S. Cuboni, C. Martini,
A. Buffo, M. Cimino and M. P. Abbracchio, PLoS ONE,
2008, 3, e3579.
4 Y. Chen, H. Wu, S. Wang, H. Koito, J. Li, F. Ye, J. Hoang,
S. S. Escobar, A. Gow, H. A. Arnett, B. D. Trapp,
N. J. Karandikar, J. Hsieh and Q. R. Lu, Nat. Neurosci.,
2009, 11, 1398.
Acknowledgements
Y. B. is grateful for an SQU grant (IG/SCI/CHEM/13/02). This
study was supported by the German Federal Ministry of
Education and Research (BMBF) within the BioPharma – Neu-
roallianz consortium and by UCB Pharma GmbH, Monheim,
Germany.
5 M. Fumagalli, S. Daniele, D. Lecca, P. R. Lee, C. Parravicini,
R. D. Fields, P. Rosa, F. Antonucci, C. Verderio,
M. L. Trincavelli, P. Bramanti, C. Martini and
M. P. Abbracchio, J. Biol. Chem., 2011, 286, 10593.
6 H. Franke, C. Parravicini, D. Lecca, E. R. Zanier, C. Heine,
K. Bremicker, M. Fumagalli, P. Rosa, L. Longhi,
N. Stocchetti, M. G. De Simoni, M. Weber and
M. P. Abbracchio, Purinergic Signalling, 2013, 9, 451–462.
7 A. Maekawa, B. Balestrieri and Y. Kanaoka, Proc. Natl. Acad.
Sci. USA, 2009, 106, 11685–11690.
Notes and references
1 P. Ciana, M. Fumagalli, M. L. Trincavelli, C. Verderio,
P. Rosa, D. Lecca, S. Ferrario, C. Parravicini, V. Capra,
P. Gelosa, U. Guerrini, S. Belcredito, M. Cimino, L. Sironi,
E. Tremoli, G. E. Rovati, C. Martini and M. P. Abbracchio,
EMBO J., 2006, 25, 4615.
8 A.-D. Qi, K. Harden and R. A. Nicholas, J. Pharmacol. Exp.
Ther., 2013, 347, 38–46.
2 S. Hennen, L. Peters, H. Wang, A. Spinrath, N. Merten,
¨
S. Blattermann, R. Akkari, R. Schrage, K. Simon,
9 F. G. Salituro, B. L. Harrison, B. M. Baron, P. L. Nyce,
K. T. Stewart, J. H. Kehne, H. S. White and I. A. McDonald,
J. Med. Chem., 1992, 35, 1791.
¨
R. Schroder, D. Schulz, C. Vermeiren, K. Zimmermann,
S. Kehraus, C. Drewke, A. Pfeifer, G. M. Konig, K. Mohr,
¨
¨
¨
M. Gillard, C. E. Muller, Q. R. Lu, J. Gomeza and 10 Y. Baqi and C. E. Muller, Nat. Protoc., 2010, 5, 945–953.
¨
11 P. Hillmann, G. Y. Ko, A. Spinrath, A. Raulf, I. von Kugelgen,
E. Kostenis, Sci. Signaling, 2013, 6, ra93.
¨
S. C. Wolff, R. A. Nicholas, E. Kostenis, H. D. Holtje and
3 D. Lecca, M. L. Trincavelli, P. Gelosa, L. Sironi, P. Ciana,
M. Fumagalli, G. Villa, C. Verderio, C. Grumelli,
¨
C. E. Muller, J. Med. Chem., 2009, 52, 2762–2775.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.