Bioorganic & Medicinal Chemistry Letters
Discovery of potent, reversible MetAP2 inhibitors via fragment based
drug discovery and structure based drug design—Part 2
a
a
a
Christopher McBride a, , Zacharia Cheruvallath , Mallareddy Komandla , Mingnam Tang ,
Pamela Farrell b, J. David Lawson c, Darin Vanderpool b, Yiqin Wu b, Douglas R. Dougan d,
Artur Plonowski b, Corine Holub b, Chris Larson b
⇑
a Medicinal Chemistry, Takeda California, United States
b Biological Sciences, Takeda California, United States
c Computational Sciences, Takeda California, United States
d Structural Biology, Takeda California, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Methionine aminopeptidase-2 (MetAP2) is an enzyme that cleaves an N-terminal methionine residue
from a number of newly synthesized proteins. This step is required before they will fold or function cor-
rectly. Pre-clinical and clinical studies with a MetAP2 inhibitor suggest that they could be used as a novel
treatment for obesity. Herein we describe the discovery of a series of pyrazolo[4,3-b]indoles as reversible
MetAP2 inhibitors. A fragment-based drug discovery (FBDD) approach was used, beginning with the
screening of fragment libraries to generate hits with high ligand-efficiency (LE). An indazole core was
selected for further elaboration, guided by structural information. SAR from the indazole series led to
the design of a pyrazolo[4,3-b]indole core and accelerated knowledge-based fragment growth resulted
in potent and efficient MetAP2 inhibitors, which have shown robust and sustainable body weight loss
in DIO mice when dosed orally.
Received 1 March 2016
Revised 22 April 2016
Accepted 23 April 2016
Available online 25 April 2016
Keywords:
Methionine aminopeptidase-2
MetAP2
Metalloprotease
Fragment-based drug discovery
FBDD
Ó 2016 Elsevier Ltd. All rights reserved.
Structure-based drug design
SBDD
Pyrazolo[4,3-b]indoles
In our previous communication1, we described the rationale for
our interest in MetAP2 as a target to treat obesity and disclosed an
indazole fragment hit with moderate affinity (1, MetAP2 pIC50 of
5.6), and a high LE and lipophilic ligand efficiency (LLE) due to its
small size and low lipophilicity. In that communication we identi-
fied interactions vital to potency, most importantly those at 6-posi-
tion of the indazole core, which can generate gains in affinity of
>100 fold over the unsubstituted core (Fig. 1, compound 2 vs 3).
In this Letter, we disclose the structure based design of a new
scaffold as part of our efforts in identifying multiple chemotypes
for our program. This led to the identification of several potent
pyrazolo[4,3-b]indoles, as selective, orally bioavailable, and reversi-
ble MetAP2 inhibitors.
methionine of the endogenous ligands for MetAP2. Filling this
pocket yields the aforementioned potency gains of up to 100-fold.
Installing an aryl group at the 4 position of the indazole core fills
a hydrophobic cleft between Tyr444 and His-339 and boosts
potency (5–10 fold).
In parallel to our work on the indazoles, we sought a novel
chemical scaffold and designed a pyrazolo[4,3-b]indole core which
we hypothesized could satisfy all of the key interactions seen in the
indazole series (Fig. 2, indazole core overlaid with model of pyra-
zolo[4,3-b]indole core). By utilizing the same pyrazole warhead
we expected to maintain the interactions with the metal ions in
the active site of MetAP2. The size of the tricyclic core would cause
the phenyl ring to protrude even deeper into the lipophilic pocket,
though still leaving room for small substitutions around the ring (7-
fluorinated example shown). Additionally, this new scaffold offered
the opportunity for substitution at N4, which would point in nearly
the same vector as the indazole 4-substitutions.
The co-crystal structure of an indazole1 bound to MetAP2
showed the N2-nitrogen of the indazole ring coordinates with
one of the active site metal ions while the N1-nitrogen makes a
water mediated interaction with the other active site metal. The
6-position substituent on the indazole ring is oriented to fill the
adjacent hydrophobic site that is typically filled by the terminal
Previous work by Pudlo et al. utilized a thermally decomposed
azide for nitrene insertion and ring formation of fused aromatic
indole heterocycles.2 We utilized a similar strategy to access our
target molecules. A representative synthesis for this series is
⇑
Corresponding author.
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.