Fragment-based discovery of potent ERK2 pyrrolopyrazine inhibitors

Article abstract of DOI:10.1016/j.bmcl.2015.08.048

A fragment-based lead discovery approach was used to discover novel ERK2 inhibitors. The crystal structure of N-benzyl-9H-purin-6-amine 1 in complex with ERK2 elucidated its hinge-binding mode. In addition, the simultaneous binding of an imidazole molecule adjacent to 1 suggested a direction for fragment expansion. Structure-based core hopping applied to 1 led to 5H-pyrrolo[3,2-b]pyrazine (3) that afforded direct vectors to probe the pockets of interest while retaining the essential hinge binding elements. Utilizing the new vectors for SAR exploration, the new core 3 was quickly optimized to compound 39 resulting in a greater than 6600-fold improvement in potency.

Full text of DOI:10.1016/j.bmcl.2015.08.048

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Fragment-based discovery of potent ERK2 pyrrolopyrazine inhibitors
a
b
c
b
c
Daniel J. Burdick a, , Shumei Wang , Christopher Heise , Borlan Pan , Jake Drummond , JianPing Yin ,
⇑
Lauren Goeser a, Steven Magnuson a, Jeff Blaney d, John Moffat b, Weiru Wang c, , Huifen Chen
d,
⇑
⇑
a Department of Discovery Chemistry, Genentech, Inc., South San Francisco, CA 94080, United States
b Department of Biochemical Pharmacology, Genentech, Inc., South San Francisco, CA 94080, United States
c Department of Structural Biology, Genentech, Inc., South San Francisco, CA 94080, United States
d Department of Computational Chemistry, Genentech, Inc., South San Francisco, CA 94080, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A fragment-based lead discovery approach was used to discover novel ERK2 inhibitors. The crystal struc-
ture of N-benzyl-9H-purin-6-amine 1 in complex with ERK2 elucidated its hinge-binding mode. In addi-
tion, the simultaneous binding of an imidazole molecule adjacent to 1 suggested a direction for fragment
expansion. Structure-based core hopping applied to 1 led to 5H-pyrrolo[3,2-b]pyrazine (3) that afforded
direct vectors to probe the pockets of interest while retaining the essential hinge binding elements.
Utilizing the new vectors for SAR exploration, the new core 3 was quickly optimized to compound 39
resulting in a greater than 6600-fold improvement in potency.
Received 12 June 2015
Revised 7 August 2015
Accepted 14 August 2015
Available online xxxx
Keywords:
ERK2
Pyrrolopyrazine
Fragment-based
Extracellular-regulated kinase
Ó 2015 Elsevier Ltd. All rights reserved.
The Extracellular-regulated kinases ERK1/MAPK2 and ERK2/
MAPK1 are the central effector nodes of the Ras/Raf/MEK path-
way.1 These kinases are responsible for transducing mitogenic
and pro-survival signals from growth factor receptors as well as
oncogenic mutant forms of Ras and Raf, and are therefore critical
mediators of oncogenic signaling in a large proportion of cancers
and tumor types. Downstream effects of ERK1/2 signaling relevant
to cancer include activation of the mitogenic transcription pro-
gram, regulation of protein synthesis though the Rsk kinases, and
pro-survival signaling though the Bad/Bim pathway. The validity
of the Ras/ERK pathway as a target for cancer therapeutics has
been demonstrated with inhibitors of Raf and MEK kinases.2 Since
ERK1/2 are the sole catalytic substrates of MEK1/2, it is likely that
small-molecule inhibitors of ERK would be equally as effective as
MEK inhibitors while presenting opportunities for novel chemical
matter and differential effects on pathway feedback and crosstalk.
In order to find novel ERK2 inhibitors, we screened a small frag-
ment library using an NMR-based saturation transfer difference
(STD) assay3 and using a surface plasmon resonance (SPR) assay.4
Herein we describe our efforts to improve the fragment hits
obtained in the screen which led to the discovery of a novel and
potent compound 39.
The synthesis of compounds 7–21 and 22–39 was accomplished
as outlined in Scheme 1. Commercially available boronic acids or
their pinacol esters were coupled to 2-bromo-5H-pyrrolo[3,2-b]
pyrazine 4 using standard Suzuki cross coupling methods to gener-
ate the C2 substituted analogues 7–21.5 Treatment of the pyridine
analogue 14 or the pyrazole analogue 17 with NIS in acetone
cleanly provided the 7-iodo intermediates, which were then con-
verted to compounds 40 and 41 by tosyl protection. These interme-
diates were subjected to standard Suzuki cross coupling methods
with a number of different boronates followed by base promoted
removal of the tosyl groups to generate analogues 22–39.
We screened a small 635 member fragment library by using
NMR-STD and SPR. The NMR-STD assay yielded 54 primary hits
and the SPR assay resulted in 78 hits. Thirteen compounds over-
lapped in both the SPR and NMR screens. Subsequent SPR KD mea-
surements of these hits identified compounds 1 and 2 as the most
ligand efficient hits (LE)6 (Fig. 1A).
To elucidate the binding interactions of fragments 1 and 2 with
ERK2 we determined their co-crystal structures with un-phospho-
rylated ERK2 protein. As illustrated in Figure 1B and C, compounds
1 and 2 bound in the ATP binding cleft and formed three hydrogen-
bonds with the backbone of the hinge region comprising residues
Asp106 though Met108. In addition, we observed some extra elec-
tron density in a pocket adjacent to the fragment, which was too
large to be interpreted as a structural water. After reviewing the
protein purification procedure, we interpreted this density as an
imidazole molecule that had co-purified with the protein. Interest-
⇑
Corresponding authors.
E-mail addresses: burdick.dan@gene.com (D.J. Burdick), wang.weiru@gene.com
(W. Wang), chen.huifen@gene.com (H. Chen).
http://dx.doi.org/10.1016/j.bmcl.2015.08.048
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Burdick, D. J.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.048

2
D. J. Burdick et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Tos
N
purine
7- azaindole
H
N
5H-pyrrolo[3,2-b]pyrazine (3)
H
N
N
N
H
N
N
N
N
a
b, c
d, e
e
e
e
g
Back
Pocket
g
g
n
HN
n
R1
X
n
i
HN
HN
i
i
N
R1
H
R1
H
N
N
H
Br
N
R2
I
N
N
N
N
4
7 - 21
40, 41
22 - 39
Minimal
pharmacophores
Structural
diversity
NHBn
N
X
Scheme 1. (a) [PdCl2(dppf)]CH2Cl2, aryl-B(OH)2 or pinacol ester, 1 M Na2CO3, CH3-
CN; (b) NIS, acetone; (c) NaH, TosCl, DMF; (d) [PdCl2(dppf)]CH2Cl2, appropriate aryl-
B(OH)2 or pinacol ester, 1 M Na2CO3, CH3CN; (e) 1 M NaOH, MeOH, THF.
Imidazole
region
Figure 2. Proposed core hopping strategy: replacement of the purine with the
minimal pharmacophore 7-azaindole would provide the needed hinge hydrogen
bonding interactions as well as providing new vectors to probe the imidazole region
and the back pocket.
ingly, the imidazole is within hydrogen-bond distance to the side
chains of Asp111 and Lys114, indicative of the potential for engag-
ing additional hydrogen bonds to the protein by linking the frag-
ments (Fig. 1).
We then investigated whether fragment expansion into the imi-
dazole-binding site can enhance the potency. Analysis of crystal
structures (Fig. 1) indicated that although both compounds 1 and
2 form the hydrogen bond interactions with the hinge, the purine
core of compound 1 is more co-planar with the imidazole ring. This
co-planarity may allow substitutions at positions 2 and 6 of the
purine core to reach the imidazole binding site. Compound 2 was
deemed less interesting because it sits well above the plane of
the imidazole ring and it was not obvious how to engage the imi-
dazole binding pocket; thus we focused our attention on 1.
Initial modification of the purine along the accessible vectors at
C2 and C6 did not lead to a significant improvement in potency and
ligand efficiency rapidly decreased (data not shown). It quickly
became clear that an alternate ring system that provided access
to different vectors was needed.
ment hit 1 (Table 1); however, it is approximately twice as ligand
efficient. We determined the crystal structure of 3 bound to ERK2
to confirm the binding mode. As shown in Figure 3, compound 3
engages the hinge through both classical and C–H type hydrogen
bonds with Asp106 and Met108, and again indicated that substitu-
tion at C2 and C7 can potentially generate additional interactions
with Asp111, Lys114 and Gln105. Encouraged by these results,
we tested a small series of commercially available 5H-pyrrolo
[3,2-b]pyrazines in the SPR assay. A bromine at C2 of the 5H-pyr-
rolo[3,2-b]pyrazine core was tolerated (4) while halide substitu-
tion at C7 (5 and 6) improved potency and preserved good ligand
efficiency. These initial results gave us confidence that we could
expand SAR along these vectors and we embarked on a medicinal
chemistry campaign to further improve the 5H-pyrrolo[3,2-b]pyr-
azine series.
Alternative cores were considered that would maintain a two or
three point hinge interactions like the initial fragment 1; reach the
back pocket to form additional hydrogen bond interactions; and
make an additional contact in the imidazole fragment region. An
additional consideration to move away from the heavily explored
purine scaffold, is that purines are the basis of many drug discov-
ery campaigns and finding novel inhibitors may prove to be
difficult.7
After determining the key pharmacophore features and consid-
ering a number of different cores, we proposed changing the purine
core of the screening hit to 5H-pyrrolo[3,2-b]pyrazine (3, Fig. 2).
This core is less represented in the literature relative to the purine.8
Furthermore this scaffold retains the same hydrogen bonding
donor/acceptor pair available for interacting with the hinge of
the protein and it has the proper pKa range at N7 to ensure that
it acts as a hydrogen bond acceptor. In addition, this core has car-
bon atoms at the proper positions to furnish vectors that point
toward the back side of ATP binding pocket as well as the region
where the imidazole was observed in the crystal structure.
The un-substituted 5H-pyrrolo[3,2-b]pyrazine core 3 was eval-
uated by SPR and was found to be slightly less active than the frag-
We then explored different groups at the 2 position of pyrrolo
[3,2-b]pyrazine (Table 2). Addition of a phenyl group (7) increased
the activity by more than 4 fold (Table 2), whereas a bromo group
(4) at this position showed no appreciable change in binding affin-
ity (Table 2).
These results suggested aryl substitutions would be productive
at the C2 position. To further explore the C2 position, we made a
series of analogues substituted with aryl and small heteroaryl rings
to mimic the imidazole that we observed in the initial crystal
structures. Analysis of the aryl analogues tested revealed that meta
substitutions (8 and 9) were not tolerated most likely due to steric
clashes with Glu109 and Asp111. para substitutions (10 and 11)
showed marginal improvement over the phenyl analogue regard-
less of the electronic nature of the functional group. The 4-fluoro
(12) and 4-hydroxy compounds (13) were the most potent of this
set of phenyl analogues, which indicated that a small hydrogen
bond accepting group may be advantageous and would interact
with Lys114. The pyridyl analogues, which contain a hydrogen
bond acceptor in the ring, were also active. The para-pyridyl (14)
and 2-fluoropyridin-4-yl (15) analogues were more active and
had higher ligand efficiencies than the meta analogue (16) possibly
NH2
B.
C.
A. HN
Q105
Q105
N
N
O
E106
2.7
3.0
E106
2.8
3.1
N
L107
L107
M108
NH
N
K54
K54
Cl
M108
3.2
3.2
1
2
Compound SPR KD ( M)
LE
µ
D111
K114
1
2
247
940
0.29
0.38
D111
K114
Figure 1. (A) Structure and activity of two potent and ligand efficient fragments found in both the NMR STD and SPR screens. (B) Crystal structure of N-benzyl-9H-purin-6-
amine (1) (PDB ID: 4QP1) and (C) 5-chlorobenzo[d]oxazol-2-amine (2) (PDB ID: 4QP2) in complex with ERK2. Both fragments form multiple hydrogen bonds (including non-
classical C–H type) with the kinase hinge as shown by green dashed lines (distances are in angstroms). The crystallographic imidazole fragments are colored in magenta.
Please cite this article in press as: Burdick, D. J.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.048

D. J. Burdick et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
Table 1
pyrrolo[3,2-b]pyrazine. Molecular modeling of potential inhibitors
in the binding pocket of the protein indicated that a 3-pyridyl or 5-
pyrimidyl moiety at the 7 position of the pyrrolo[3,2-b]pyrazine
core should place the ring nitrogen within hydrogen bonding dis-
tance of the gatekeeper residue Gln103. Compounds 22 and 23
(Table 3) were, respectively, 29-fold and 5.3-fold more potent than
the pyridyl analogue 14. In a similar fashion, compounds 24 and 25
were 21-fold and 5.6-fold more potent than compound 17. Based
on the activity of compound 22, a series of substituted 3-pyridyl
and pyrazole analogues were made at C7 and their activities were
compared to compound 22. Small electron donating groups in the
4 position of the 3-pyridyl moiety (26–28) were approximately
equipotent to the parent pyridyl (22). Introduction of an elec-
tron-withdrawing group (29) increased potency over 2-fold pre-
sumably due to the change in pKa of the pyridyl nitrogen making
it a better hydrogen bond acceptor and the introduction of a van
der Waals interaction between the chlorine and Tyr 36.
The crystal structure of 24 in ERK2 (Fig. 4A) confirmed our mod-
eling hypothesis. The pyrrolo[3,2-b]pyrazine core and pyrazole
binds to the hinge and front pocket like compound 17 and the 3-
pyridyl indeed binds to the back pocket with the pyridyl nitrogen
forming a hydrogen bond to the catalytic Lys54. In addition, a
water molecule is also within hydrogen bond distance of the pyr-
idyl nitrogen, which may offer additional potency improvement.
There is also a notable difference in the location and orientation
of Tyr36 in the P-loop. Tyr36, which packs against the alpha C helix
in the crystal structure of 24 (Fig. 4A) has been pulled in with its
aromatic ring slightly stacking over the 3-pyridyl moiety. The C6
position of the pyridine is 3.6 and 4.3 angstroms away from atoms
in Tyr36 and Asp167, respectively, and thus limits the size of func-
tional groups that can be tolerated in this region. The structure also
revealed that some space is available towards the sugar and alpha
phosphate binding region.
Activities of fragment screening hit 1 and initial core vector analogues
Compound
Structure
SPR KD
(
lM)
LE
H
N
N
1
247
0.29
N
N
NHBn
H
N
N
3
4
343
319
0.53
0.48
N
H
N
N
Br
N
H
N
N
5
6
97
0.55
0.49
N
Br
H
N
N
125
Br
N
I
due to better hydrogen bonding with Lys114. Heterocycles that
present a hydrogen bond donor/acceptor combination were also
well tolerated (17–21). Compound 17 had the best potency and
ligand efficiency and is greater than 30 fold more potent than the
parent 2-bromo analogue (4). Pyrazole 17 displays both a hydrogen
bond donor and acceptor and was 4-fold more potent than the
matched pair compound 18, which has the hydrogen bond donor
masked by a methyl group. Compound 17 was the most potent
and ligand efficient analogue of this series and it is apparent in
the co-crystal structure (Fig. 3) that the pyrazole makes favorable
hydrogen bond and polar interactions with both Asp111 and
Lys114 and occupies the space where the crystallographic imida-
zole was observed. The methyl substituted analogues (19 and 20)
were less potent than the un-substituted pyrazole even though
they contain the same hydrogen bond donor and acceptor as 17.
The reduction in potency can be attributed to less optimal dihedral
angles imposed by the methyl groups. Indazole 21 was five times
less potent than the pyrazole 17 presumably because it extends
beyond the interactions of Asp111 and Lys114.
Replacement of the 3-pyridyl group at C7 (22) with a 5-pyrazole
moiety (30) decreased inhibitor activity two fold whereas a slight
increase was obtained when the 3-pyridyl was replaced with a 4-
pyrazole group (31). The potency decrease of 30 could be a result
of a less optimal hydrogen bond between the pyrazole N2 and
Lys54 due to tautomerization. In compound 31, on the other hand,
the pyrazole is symmetric and would present a hydrogen bond
acceptor proximal to Lys54 regardless of tautomerization. In addi-
tion, we could substitute at the NH of the 4-pyrazole to not only
Having established two favorable C2 moieties, 4-pyridyl and
4-pyrazole, we turned our attention to substitutions at C7 of the
Table 2
SAR of position 2 of 5H-pyrrolo[3,2-b]pyrazines
H
N
N
N
R1
E106
K54
L107
Compound
R1
IC50 active ERK (
lM)
LE
4
7
8
9
Br
Ph
>200
50.1
>200
>200
44.1
89.7
29.7
12.4
11.7
21.6
71.5
5.9
<0.50
0.39
<0.28
<0.32
0.33
0.32
0.39
0.42
0.45
0.40
0.38
0.51
0.43
0.41
0.43
0.34
E71
M108
3-Benzamide
3-Hydroxyphenyl
4-Benzamide
4-Methoxyphenyl
4-Fluorophenyl
4-Hydroxyphenyl
4-Pyridyl
2-Fluoropyridin-4-yl
3-Pyridyl
4-Pyrazole
N-Me-4-pyrazole
3,5-Dimethyl-4-pyrazole
3-Methyl-4-pyrazole
6-Indazole
10
11
12
13
14
15
16
17
18
19
20
21
D111
K114
22.5
14
22.2
31.9
Figure 3. Crystal structure of compound 3 in complex with ERK2 (yellow, PDB ID:
4QP6) overlaid with the structure of compound 17 in complex with ERK2 (cyan,
PDB ID: 4QP7). Intermolecular hydrogen bonds (including non-classical C–H type)
between 17 and ERK2 are shown in green dashed lines.
Please cite this article in press as: Burdick, D. J.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.048

4
D. J. Burdick et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 3
Structure activity relationship of C2 4-pyridyl analogues
H
N
N
N
R1
R2
Compound
R1
R2
H
IC50 active ERK (
l
M)
LE
14
22
23
17
24
25
26
27
28
29
30
31
4-Pyridyl
4-Pyridyl
4-Pyridyl
4-Pyrazole
4-Pyrazole
4-Pyrazole
4-Pyridyl
4-Pyridyl
4-Pyridyl
4-Pyridyl
4-Pyridyl
4-Pyridyl
11.7
0.407
2.2
0.45
0.42
0.37
0.51
0.45
0.41
0.43
0.45
0.41
0.42
0.41
0.44
3-Pyridyl
5-Pyrimidyl
H
3-Pyridyl
5-Pyrimidyl
6-Methoxypyridin-3-yl
6-Aminopyridin-3-yl
6-Methylpyridin-3-yl
6-Chloropyridin-3-yl
5-Pyrazole
5.9
0.28
1.05
0.252
0.468
0.634
0.182
0.94
0.354
4-Pyrazole
reduce the desolvation penalty but also introduce additional inter-
actions towards the sugar phosphate binding region. Simple
methyl alkylation (32) of the nitrogen resulted in a potency
increase and not a decrease; this suggests that the NH of the pyra-
zole is not making any particular interaction with the protein
(Table 4). Ethylation of the pyrazole (33) led to a two-fold improve-
ment in potency compared to compound 31 and extending to a
propyl (34) resulted in another doubling of potency.
A crystal structure of 34 in ERK2 (Fig. 4B) was determined and
as expected the pyrrolopyrazine core binds to the kinase hinge
with the 4-pyridyl ring at C2 facing solvent and the C7 4-pyrazole
binding in the back pocket. The N2 acceptor atom of the 4-pyrazole
is 4.5 Å from the Lys54 tertiary amine and is likely interacting with
Lys54 through a water-mediated hydrogen bond (the water is not
resolved) while the propyl on N1 binds in a polar region near the
sugar phosphate group of ATP. Similar to the crystal structure of
24 (Fig. 4A), Tyr36 of the P-loop adopts the same conformation
and forms close hydrophobic interactions with the N1-propyl tail
(34). In an attempt to induce a Tyr36-out conformation, which
would give more room for expansion, we decided to pursue a
few larger substitutions. The 2-isobutyl substituted analog (35)
was slightly less potent than 31. The hydroxyl propyl (36) analog
did not improve potency, but on the contrary led to a potency drop
of 9-fold, possibly due to desolvation penalties. The morpholine
ethyl substitution (37) was not well tolerated and led to 6-fold
potency drop compared to compound 31. A phenethyl substitution
However, to our great satisfaction, the benzyl substitution (39)
resulted in a 10-fold increase in potency compared to compound
31. Compound 39 was the first of the pyrrolopyrazine series reach-
ing below 50 nM potency in the ERK2 biochemical assay, and rep-
resents an excellent starting point for further optimization. A
crystal structure of 39 in complex with ERK2 was determined
and indeed the P-loop adopts a Tyr36-out conformation (Fig. 4C).
Tyrosine 36 in this structure has swung out to open up more space
under the P-loop such that the phenyl ring in 39 binds where the
Tyr36 ring occupied in the pulled in conformation
(Fig. 4A and B). The flexibility of the P-loop will allow further opti-
mization of the benzyl moiety to improve potency and other prop-
erties of this promising molecule.
Purine inhibitor (1) was discovered in a NMR-STD and SPR frag-
ment screen. X-ray crystallography of early fragments revealed an
unexpected imidazole molecule in the vicinity of the kinase hinge.
Exploitation of the imidazole and back pocket of the binding site
could not be achieved with the initial purine scaffold. Core hopping
to a 5H-pyrrolo[3,2-b]pyrazine core provided efficient vectors to
reach both the imidazole region and the back pocket while keeping
the correct donor and acceptor atoms to form hydrogen bonds with
the hinge. Beneficial interactions were formed with Lys114 and
possibly Asp111 in the front pocket with either a 4-pyridyl or 4-
pyrazole substitution on the 5H-pyrrolo[3,2-b]pyrazine core at
C2. Further substitutions at C7 showed that a hydrogen bond could
be formed with Lys54 and Tyr36 could be moved into a position
under the P-loop giving a greater enhancement in potency.
(38) proved to be very detrimental, resulting in a 2 lM IC50.
A.
B.
C.
Y36
Q105
K54
Y36
E106
I31
A52
L107
M108
Y36
K114
T110
D111
Figure 4. (A) Crystal structure of compound 24 (green) in complex with ERK2 (PDB ID: 4QP8). (B) Crystal structure of compound 34 (magenta) in complex with ERK2 (PDB ID:
4QP9). (C) Crystal structure of compound 39 (cyan) in complex with ERK2 (PDB ID: 4QPA). Tyr36 in P-loop is colored in yellow. Intermolecular hydrogen bonds between
compounds and ERK2 are shown in green dashed lines.
Please cite this article in press as: Burdick, D. J.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.048

D. J. Burdick et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
Table 4
further optimization and the results of this effort will be disclosed
in a subsequent Letter.
The effect of substitution on 7-(1H-pyrazol-4-yl)-2-(pyridin-4-yl)-5H-pyrrolo[2,3-b]
pyrazine analogues
Acknowledgments
H
N
N
N
We thank the Analytical, Biopharmacology, Compound Man-
agement, NMR, Pharmacokinetics, and Structural Biology col-
leagues for their contributions. Crystallographic data were
collected at the Advanced Light Source (ALS). ALS is supported by
the Director, Office of Science, Office of Basic Energy Sciences, of
the U.S. Department of Energy under Contract No. DE-AC02-
05CH11231.
R2
N
Compound
R2
IC50 ERK2 LC3K (
l
M)
LE
N
NH
31
0.354
0.44
N
N
N
N
32
33
0.072a
0.149
0.47
0.43
Supplementary data
Supplementary data (synthetic procedures, assay protocols and
crystallographic refinement statistics) associated with this article
can be found, in the online version, at http://dx.doi.org/10.1016/j.
bmcl.2015.08.048.
N
N
34
35
36
0.071
0.091
0.623
0.43
0.41
0.36
N
N
References and notes
N
N
OH
1. Wong, K. K. Recent Pat. Anticancer Drug Discov. 2009, 4, 28.
2. Montagut, C.; Settleman, J. Cancer Lett. (Shannon, Ireland) 2009, 283, 125.
3. Mayer, M.; Meyer, B. J. Am. Chem. Soc. 2001, 123, 6108.
4. Erlanson, D. A. In Fragment-based Drug Discovery and X-ray Crystallography;
Topics in Current Chemistry; Springer: Berlin, Germany, 2011; Vol. 317, p 1.
5. Suzuki, A. J. Organomet. Chem. 2002, 653, 83.
N
N
37
0.416
0.31
N
O
6. (a) Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430; (b)
Abad-Zapatero, C. Expert Opin. Drug Discov. 2007, 2, 469.
N
N
38
39
2.0
0.18
0.38
7. (a) Song, Y.; Xu, Q.; Bauer, S. M.; Jia, Z. J.; Mehrotra, M.; Pandey, A. PCT Int. Appl.
2009, WO 2009131687 A2.; (b) Huang, W.; Zhu, X.; Wang, Y.; Azam, M.; Wen,
D.; Sundaramoorthi, R.; Thomas, M. R.; Liu, S., et al. J. Med. Chem. 2009, 52, 4743;
c Green, S.; Frame, S.; Fleming, I. PCT Int. Appl. 2008, WO 2008122779 A1.; (d)
Hauser, D. R. J.; Scior, T.; Domeyer, D. M.; Kammerer, B.; Laufer, S. A. J. Med.
Chem. 2007, 50, 2060.
N
N
0.037
a
8. (a) Gelbard, H. A.; Dewhurst, S.; Goodfellow, V. S.; Wiemann, T. PCT Int. Appl.
2010, WO 2010068483 A2.; (b) Graczyk, P.; Bhatia, G. S. PCT Int. Appl. 2010, WO
2010015803 A1.; (c) Dubois, D. J.; Elworthy, T. R.; Hendricks, R. T.; Hermann, J.
C.; Kondru, R. K.; Lou, Y.; Owens, T. D.; Smith, D. B. U.S. Pat. Appl. Publ. 2009, US
20090215785 A1.; (d) Arnold, W. D.; Bounaud, P.; Chen, C.; Eastman, B.;
Gosberg, A.; Gradl, S. N.; Hopkins, S.; Li, Z.; McDonald, I.; Sprengeler, P. A.; et al.
PCT Int. Appl. 2008, WO 2008124850 A1.
Value in the OMNIA ERK2 assay. Values for compounds have the same rank
order.
Ultimately, a greater than 6600-fold increase in potency and an
increase in ligand efficiency were achieved after moving from pur-
ine 1 to compound 39. Compound 39 became the starting point for
Please cite this article in press as: Burdick, D. J.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.08.048