The discovery and structure-activity relationships of indole-based inhibitors of glutamate carboxypeptidase II

Article abstract of DOI:10.1016/j.bmcl.2010.10.109

A series of N-substituted 3-(2-mercaptoethyl)-1H-indole-2-carboxylic acids were synthesized as inhibitors of glutamate carboxypeptidase II (GCPII). Those containing carboxybenzyl or carboxyphenyl groups at the N-position exhibited potent inhibitory activity against GCPII. These indole-based compounds represent the first example of achiral GCPII inhibitors and demonstrate greater tolerance of the GCPII active site for ligands with significant structural difference from the endogenous substrate, N-acetyl-aspartylglutamate.

Full text of DOI:10.1016/j.bmcl.2010.10.109

Bioorganic & Medicinal Chemistry Letters 20 (2010) 7222–7225
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The discovery and structure–activity relationships of indole-based inhibitors
of glutamate carboxypeptidase II
Brian Grella a, Jessica Adams a, James F. Berry b, Greg Delahanty a,b, Dana V. Ferraris a,b, Pavel Majer a,
Chiyou Ni a, Krupa Shukla b,c, Scott A. Shuler b, Barbara S. Slusher a,b,c, Marigo Stathis b,
Takashi Tsukamoto a,b,c,
⇑
a Eisai Research Institute, Baltimore, MD 21224, USA
b Brain Science Institute, Johns Hopkins University, Baltimore, MD 21205, USA
c Department of Neurology, Johns Hopkins University, Baltimore, MD 21205, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of N-substituted 3-(2-mercaptoethyl)-1H-indole-2-carboxylic acids were synthesized as inhibi-
tors of glutamate carboxypeptidase II (GCPII). Those containing carboxybenzyl or carboxyphenyl groups
at the N-position exhibited potent inhibitory activity against GCPII. These indole-based compounds
represent the first example of achiral GCPII inhibitors and demonstrate greater tolerance of the GCPII
active site for ligands with significant structural difference from the endogenous substrate,
N-acetyl-aspartylglutamate.
Received 25 August 2010
Revised 19 October 2010
Accepted 21 October 2010
Available online 26 October 2010
Keywords:
Glutamate carboxypeptidase II
Ó 2010 Elsevier Ltd. All rights reserved.
Inhibitor
Indole
Metallopeptidase
Thiol
O
Glutamate carboxypeptidase II (GCP II, EC 3.4.17.21) catalyzes
the extracellular hydrolysis of the neuropeptide N-acetyl-aspar-
tylglutamate (NAAG) to N-acetyl-aspartate and glutamate. Since
the hydrolysis of NAAG is believed to be one of the major sources
of glutamate in the nervous system, inhibition of GCP II has gained
HN
I
R'
R
O
considerable attention as
a strategy to suppress glutamate
excitotoxicity leading to neurological disorders. Thus, substantial
efforts have been made to identify potent and selective GCP II
inhibitors.1 Some of these inhibitors have shown efficacy in a vari-
ety of animal models of neurological diseases associated with
glutamate excitotoxicity including stroke,2 amyotrophic lateral
sclerosis (ALS),3 neuropathic pain,4 and diabetic neuropathy.5 Fur-
thermore, an orally available GCPII inhibitor, 2-(3-mercaptopropyl)
pentanedioic acid (2-MPPA), was tested for its safety in clinical
studies and was generally well-tolerated at plasma exposures
equivalent to those exhibiting efficacy in animal models of neuro-
pathic pain.6
HS
HO2C
N
H
N
H
CO2H
CO2H
2-MPPA (R = CH2CO2H)
DCIBzL (R' = CH2CO2H)
CMBA (R = 3-Carboxyphenyl) P1'-Allyl analog (R' = CH=CH2)
Like 2-MPPA, most of the earlier series of GCPII inhibitors
incorporate a glutarate moiety in the P10 position to take advantage
of the glutamate recognition site of GCPII. Increasing efforts are
currently being devoted for designing P10 substituents capable of
improving potency and drug-like properties. For example, our
group investigated the effect of P10 side chain modification on
GCP II inhibitory potency using 2-MPPA (IC50 = 90 nM) as a tem-
plate and found that several more lipophilic analogs, including
3-(2-carboxy-5-mercaptopentyl)-benzoic acid (CMBA), inhibit
GCP II in a more potent manner (IC50 = 15 nM) than 2-MPPA.7
These compounds have also shown improved in vivo potency in
the rat chronic constriction injury (CCI) model of neuropathic pain
by oral administration. Pomper’s group synthesized a variety of
The preclinical efficacy of GCPII inhibitors in multiple animal
models prompted us and other groups to explore new GCPII inhib-
itors with improved drug-like properties.
⇑
Corresponding author. Tel.: +1 410 614 0982; fax: +1 410 614 0659.
E-mail address: ttsukamoto@jhmi.edu (T. Tsukamoto).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.10.109

B. Grella et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7222–7225
7223
GCPII inhibitors containing a bioisostere for the P10 glutamate
using a potent urea-based GCPII inhibitor, DCIBzL (Ki = 0.01 nM),
as a template.8 From this extensive SAR study, they identified sev-
eral glutamate-free inhibitors with Ki values below 20 nM, includ-
ing P10 allyl substituted analog. Although the lack of a carboxyl
group on the P10 side chain resulted in the reduction of potency,
the X-ray crystal structure of GCPII in complex with the allyl ana-
log indicates that the allyl side chain minimizes the effect by con-
tributing to the binding primarily via non-polar interactions with
the side chains of Phe209, Leu261, and Leu428.8
These findings prompted us to explore more drastic structural
changes at the P10 position of GCPII inhibitors in an attempt to fur-
ther improve their drug-like molecular properties. Our molecular
design strategy lies in introducing an aromatic ring system as a
core backbone, which allows multiple substitutions without gener-
ating any chiral centers. To this end, we chose indole-2-carboxylic
acid as a core ring system and incorporated 2-sulfanylethyl group
into the 3-position. The new scaffold possesses a reduced number
of rotatable bonds and increased lipophilicity compared to CMBA.
In addition, various P10 side chains can be readily explored as sub-
stituents at the 1-postion to establish structure–activity relation-
ships (SAR) in this series. In this Letter, we describe the design,
synthesis, and biological evaluation of N-substituted 3-(2-sulfanyl-
ethyl)-1H-indole-2-carboxylic acid, representing the first achiral
GCPII inhibitors with IC50 values in the nanomolar range.
NH
NH
CO2H
NH
CO2CH3
b
a
O
O
HO
HO
NH
1
2
3
d
c
NH
CO2CH3
R'
CO2H
AcS
HS
4
5
e
R'
R
R
N
N
f
CO2CH3
CO2H
AcS
(R = H, R' = H);
HS
6a-h
7a
7a-h
6a
(R = R" = H)
6b (R = 2-CO2CH3, R' = H); 7b (R = 2-CO2H, R' = H)
6c (R = 3-CO2CH3, R' = H); 7c (R = 3-CO2H, R' = H)
6d
7d
(R = 4-CO2H, R' = H)
(R = 4-CO2CH3, R' = H);
6e (R = 2-Br, R' = 5-CO2CH3); 7e (R = 2-Br, R' = 5-CO2H)
6f (R =3-tert -Bu, R' = 5-CO2CH3); 7f (R = 3-tert-Bu, R' = 5-CO2H)
6g
7g
(R = 4-Br, R' = 3-CO2H)
(R = 4-Br, R' = 3-CO2CH3);
6h (R = 2-CO2CH3, R' = OCH3); 7h (R = 2-CO2H, R' = OCH3)
R
R
N
Scheme 1. Reagents and conditions: (a) 3 N KOH–THF, rt, 4.5 h; (b) concd H2SO4,
MeOH, reflux, overnight; (c) (i) PPh3, DIAD, THF, 0 °C, 30 min; (ii) 3 and AcSH, THF,
0 °C, 2 h, 63% from 1; (d) degassed 0.5 N KOH, THF, rt, 20 h, 68% (e) (i) NaH, DMF,
ꢀ15 °C, 15 min; (ii) ArCH2Br, ꢀ15 °C, then rt, overnight; (f) degassed 1 N KOH–THF,
rt, 24 h; 88% for 7c from 4.
HS
CO2H
CO2H
HS
Introduction of indole into GCPII inhibitors
H3CO2C
HO2C
As shown in Scheme 1, a majority of compounds were synthe-
sized using 3,4-dihydropyrano[3,4-b]indol-1(9H)-one 1 as a start-
ing material. The lactone was opened under basic conditions to
provide 3-(2-hydroxyethyl)-1H-indole-2-carboxylic acid 2, which
was subsequently converted to the corresponding methyl ester 3.
Mitsunobu reaction with thioacetic acid afforded the thioester 4.
The compound 4 was either hydrolyzed to give 3-(2-mercapto-
ethyl)-1H-indole-2-carboxylic acid 5 or alkylated at its N-position
with various benzyl bromides to provide 6a–h. Base-mediated
hydrolysis of 6a–h gave N-substituted 3-(2-mercaptoethyl)-
1H-indole-2-carboxylic acids 7a–h.
In order to assess the significance of a thiol group as a zinc-
binding group, we synthesized a few analogs in which the thiol
of compound 7c is replaced with other functional groups. As out-
lined in Scheme 2, N-alkylation of 1 with methyl 3-(bromo-
methyl)benzoate followed by hydrolysis gave 9. Compound 7c
was methylated at the sulfhydryl group using dimethyl sulfate to
give the S-methyl derivative 10. N-Alkylation of methyl 3-(2-meth-
oxy-2-oxoethyl)-1H-indole-2-carboxylate 119 with methyl 3-(bro-
momethyl)benzoate followed by hydrolysis gave 13.
As illustrated in Scheme 3, N-phenylation of 4 was carried out
by Cu(OAc)2 mediated coupling of phenylboronic acid to 4.10 The
poor yield (33%) of N-phenyl derivative 14 is presumably due to
the existence of a sulfur atom in the substrate which could poison
the catalysis. Base-mediated hydrolysis of 14 afforded 15 in 93%
yield. A similar coupling approach failed to produce N-3-carboxy-
phenyl derivative 16, which would ultimately lead to 17. Thus,
we redesigned our synthetic path to 17 as outlined in Scheme 4.
Ethyl 3-(2-ethoxy-2-oxoethyl)-1H-indole-2-carboxylate 1811 was
first coupled with methyl 3-bromobenzoate at its N-position to
b
a
1
N
N
O
CO2H
O
c
HO
HO2C
8
9
HO2C
N
N
CO2H
CO2H
HS
H3CS
H3CO2C
7c
10
HO2C
e
d
NH
N
N
CO2Me
CO2Me
CO2H
MeO2C
MeO2C
HO2C
11
12
13
Scheme 2. Reagents and conditions: (a) (i) NaH, DMF, ꢀ15 °C, 15 min; (ii) methyl
3-(bromomethyl)benzoate, ꢀ15 °C, then rt, overnight, 85%; (b) NaOH, THF–MeOH,
rt, overnight, 90%; (c) dimethyl sulfate, 3 N NaOH, 50 °C, 1 h, 25%; (d) (i) NaH, DMF,
ꢀ15 °C, 15 min; (ii) methyl 3-(bromomethyl)benzoate, ꢀ15 °C, then rt, overnight,
70%; (e) NaOH, THF–MeOH, rt, overnight, 84%.

7224
B. Grella et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7222–7225
Table 1
Inhibition of GCPII by indole-based compounds
N
N
Compd
Structure
IC50 (nM)a
b
a
4
CO2CH3
CO2H
X
c
AcS
N
HS
15
14
NH
CO2H
5
1200 300
HS
CO2CH3
N
CO2H
7a ((R = R0 = H)
12,000 5000
22 21
R'
7b (R = 2-CO2H, R0 = H)
7c (R = 3-CO2H, R0 = H)
22 10
94 12
CO2CH3
CO2H
R
7d (R = 4-CO2H, R0 = H)
AcS
HS
16
17
7e (R = 2-Br, R0 = 5-CO2H)
7f (R = 3-t-Bu, R0 = 5-CO2H)
7g (R = 4-Br, R0 = 3-CO2H)
7h (R = 2-CO2H, R0 = 5-OCH3)
54
9
34 24
140 10
93 41
N
Scheme 3. Reagents and conditions: (a) Cu(OAc)2, phenylboronic acid, pyridine,
dichloromethane, rt, 2 days, 33%; (b) 0.6 N KOH–dioxane, rt, overnight, 93%; (c)
Cu(OAc)2, (3-methoxycarbonylphenyl)boronic acid, pyridine, dichloromethane, rt.
CO2H
HS
9 (X = CH2OH)
10 (X = CH2SCH3)
24,000 11,000
20,000 6000
HO2C
CO2CH3
b
CO2H
13 (X = CO2H)
1700 1000
N
a
NH
CO2Et
EtO2C
N
N
CO2H
CO2Et
CO2H
X
20
EtO2C
19
HO2C
18
15 (Y = H)
17 (Y = CO2H)
6100 1000
22
Y
4
CO2H
CO2H
N
N
N
c
d
e
CO2H
CO2H
CO2H
HS
HO
N
22
a
21
EtO2C
Values are means SD of at least three experiments.
CO2CH3
CO2CH3
g
N-substituents bind in the S10 pocket which consists of positively
charged residues. Position of the carboxyl group (ortho-, meta-, or
para-) had only marginal effects on the GCPII inhibitory potency.
This could be explained by the rotational flexibility contained in
the benzyl group coupled with the relatively mobile positively
charged residues at the S10 pocket.13 Additional substituent to
the phenyl group made little change in IC50 values except for
para-bromo substitution as shown by compound 7g.
Substitution of the thiol group of 7c with either hydroxyl group
(compound 9) or methyl sulfide group (compound 10) resulted in
the substantial loss of inhibitory potency. Even substitution by a
carboxyl group (compound 13), which is known to serve as a
zinc-binding group, led to a significant increase in IC50 value. These
findings underscore the critical role of the thiol group of 7c in the
binding to the active site zinc atom(s) of GCPII.
Consistent with the effect of N-benzylation, N-phenyl substi-
tuted derivative 15 exhibited significantly lower inhibitory po-
tency. Incorporation of a carboxyl group into the meta-position
(compound 17), however, resulted in increased potency compara-
ble to those of potent N-benzyl series. It is worth noting that the
entire skeletal frame of CMBA is embedded in compound 17 with
an increasing log D value due to the added lipophilicity (from
ꢀ2.79 for CMBA to ꢀ0.11 for compound 17 at pH 7.4).14 Further-
more, elimination of the chiral center makes the achiral indole-
based inhibitors more attractive platform from which a practical
drug candidate can be developed. The nearly identical potency of
CMBA (15 nM) and 17 (22 nM) demonstrates unexpectedly high
degree of tolerance to the structural changes exhibited by the GCPII
active site. One could take advantage of this feature and expand the
structural diversity of GCPII inhibitors by exploring various con-
formationally constrained ring systems.
h
f
N
16
17
CO2CH3
CO2CH3
HO
TsO
24
23
Scheme 4. (a) methyl 3-bromobenzoate, K2CO3, CuBr, N-methylpyrrolidine, 170 °C,
overnight, 73%; (b) 3.2 N NaOH, THF–MeOH, rt, overnight; (c) concd H2SO4, ethanol,
rt, 2.5 h, 94% from 19; (d) LiBH4, DME (containing 3% MeOH), reflux, 3 h; (e) CH2N2,
MeOH, rt, 2 h; (f) TsCl, triethylamine, dichloromethane, rt, 48 h, 63% from 21; (g)
AcSK, DMF, rt, overnight; (h) 0.5 N KOH–THF, rt, 24 h; 85% from 24.
give N-3-carboxyphenyl derivative 19. Hydrolysis of all the ester
groups followed by re-esterification of the aliphatic carboxyl group
gave 21. Reduction of the ester group, followed by re-esterification
of the remaining carboxyl groups afforded 23. The hydroxyl group
was then converted to a tosylate ester and substituted by potas-
sium thioacetate to afford 16. Hydrolysis of all the ester groups
gave the desired final product 17.
Inhibitory potencies of the indole-based compounds were eval-
uated using N-acetyl-L L-glutamate as a substrate
-aspartyl-[3H]-
and a purified recombinant GCP II.12 The results are summarized
in Table 1.
N-Unsubstituted indole derivative 5 inhibited GCPII with an
IC50 value of 1.2 lM. N-Benzyl derivative 7a exhibited 10-fold in-
crease in IC50 value. However, incorporation of a carboxyl group
into the benzene ring dramatically enhanced the inhibitory po-
tency as shown by compounds 7a–h. The significant improvement
in potency by the addition of a carboxyl group suggests that these

B. Grella et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7222–7225
7225
In summary, we have identified a novel series of GCPII inhibi-
References and notes
tors based on the indole scaffold. These compounds represent the
first achiral inhibitors reported for GCPII and offer a substantial
advantage over chiral GCPII inhibitors in terms of development
feasibility. In addition, the use of synthetic intermediate 4 allows
us to generate a wide variety of N-benzyl analogs in two steps
and further optimize the structure to maximize potency. Further-
more, the breadth of synthetic methods available for 4-, 5-, 6-,
and/or 7-sustituted indole-2-crabxylic acid should facilitate SAR
studies with respect to the benzene ring portion. Additional bind-
ing affinity gained through this part of the molecules might allow
us to explore alternative zinc groups and bioisosteres for the two
carboxylic groups without significant loss of inhibitory potency.
These steps may lead to the discovery of a new generation of GCPII
inhibitors with much improved drug-like molecular properties.
1. Tsukamoto, T.; Wozniak, K. M.; Slusher, B. S. Drug Discovery Today 2007, 12,
767.
2. Slusher, B. S.; Vornov, J. J.; Thomas, A. G.; Hurn, P. D.; Harukuni, I.; Bhardwaj, A.;
Traystman, R. J.; Robinson, M. B.; Britton, P.; Lu, X. C.; Tortella, F. C.; Wozniak, K.
M.; Yudkoff, M.; Potter, B. M.; Jackson, P. F. Nat. Med. 1999, 5, 1396.
3. Ghadge, G. D.; Slusher, B. S.; Bodner, A.; Canto, M. D.; Wozniak, K.; Thomas, A.
G.; Rojas, C.; Tsukamoto, T.; Majer, P.; Miller, R. J.; Monti, A. L.; Roos, R. P. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100, 9554.
4. Majer, P.; Jackson, P. F.; Delahanty, G.; Grella, B. S.; Ko, Y. S.; Li, W.; Liu, Q.;
Maclin, K. M.; Polakova, J.; Shaffer, K. A.; Stoermer, D.; Vitharana, D.; Wang, E.
Y.; Zakrzewski, A.; Rojas, C.; Slusher, B. S.; Wozniak, K. M.; Burak, E.; Limsakun,
T.; Tsukamoto, T. J. Med. Chem. 2003, 46, 1989.
5. Zhang, W.; Murakawa, Y.; Wozniak, K. M.; Slusher, B.; Sima, A. A. J. Neurol. Sci.
2006, 247, 217.
6. van der Post, J. P.; de Visser, S. J.; de Kam, M. L.; Woelfler, M.; Hilt, D. C.; Vornov,
J.; Burak, E. S.; Bortey, E.; Slusher, B. S.; Limsakun, T.; Cohen, A. F.; van Gerven, J.
M. Br. J. Clin. Pharmacol. 2005, 60, 128.
7. Majer, P.; Hin, B.; Stoermer, D.; Adams, J.; Xu, W.; Duvall, B. R.; Delahanty, G.;
Liu, Q.; Stathis, M. J.; Wozniak, K. M.; Slusher, B. S.; Tsukamoto, T. J. Med. Chem.
2006, 49, 2876.
Acknowledgment
8. Wang, H.; Byun, Y.; Barinka, C.; Pullambhatla, M.; Bhang, H. E.; Fox, J. J.;
Lubkowski, J.; Mease, R. C.; Pomper, M. G. Bioorg. Med. Chem. Lett. 2010, 20, 392.
9. Shou, X.; Miledi, R.; Chamberlin, A. R. Bioorg. Med. Chem. Lett. 2005, 15, 3942.
10. Mederski, W. W. K. R.; Lefort, M.; Germann, M.; Kux, D. Tetrahedron 1999, 55,
12757.
This work was in part supported by the Johns Hopkins Brain
Science Institute through the NeuroTranslational program.
11. Gray, N. M.; Dappen, M. S.; Cheng, B. K.; Cordi, A. A.; Biesterfeldt, J. P.; Hood, W.
F.; Monahan, J. B. J. Med. Chem. 1991, 34, 1283.
Supplementary data
12. Rojas, C.; Frazier, S. T.; Flanary, J.; Slusher, B. S. Anal. Biochem. 2002, 310, 50.
13. Mesters, J. R.; Barinka, C.; Li, W.; Tsukamoto, T.; Majer, P.; Slusher, B. S.;
Konvalinka, J.; Hilgenfeld, R. EMBO J. 2006, 22, 1375.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.10.109.
14. Log D values were calculated using ACD/PhysChem Suite, version 12.0.