The discovery of novel benzofuran-2-carboxylic acids as potent Pim-1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.03.030

Novel benzofuran-2-carboxylic acids, exemplified by 29, 38 and 39, have been discovered as potent Pim-1 inhibitors using fragment based screening followed by X-ray structure guided medicinal chemistry optimization. The compounds demonstrate potent inhibition against Pim-1 and Pim-2 in enzyme assays. Compound 29 has been tested in the Ambit 442 kinase panel and demonstrates good selectivity for the Pim kinase family. X-ray structures of the inhibitor/Pim-1 binding complex reveal important salt-bridge and hydrogen bond interactions mediated by the compound's carboxylic acid and amino groups.

Full text of DOI:10.1016/j.bmcl.2011.03.030

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3050–3056
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The discovery of novel benzofuran-2-carboxylic acids as potent Pim-1 inhibitors
Yibin Xiang a, , Bradford Hirth a, Gary Asmussen b, Hans-Peter Biemann b, Kimberly A. Bishop e,
⇑
Andrew Good a, Maria Fitzgerald c, Tatiana Gladysheva b, Annuradha Jain e, Katherine Jancsics a, Jinyu Liu b,
Markus Metz a, Andrew Papoulis a, Renato Skerlj a, J. David Stepp b, Ronnie R. Wei d
a Department of Medicinal Chemistry, Drug and Biomaterial R&D, Genzyme Corp., 153 Second Avenue, Waltham, MA 02451, USA
b Department of In Vitro Biology, Drug and Biomaterial R&D, Genzyme Corp., 153 Second Avenue, Waltham, MA 02451, USA
c Department of DMPK, Drug and Biomaterial R&D, Genzyme Corp., 153 Second Avenue, Waltham, MA 02451, USA
d Department of Therapeutic Protein Research, Genzyme Corp., 1 Mountain Road, Framingham, MA 01701, USA
e Department of Therapeutic Protein Discovery, Genzyme Corp., 49 New York Ave, Framingham, MA 01701, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Novel benzofuran-2-carboxylic acids, exemplified by 29, 38 and 39, have been discovered as potent Pim-1
inhibitors using fragment based screening followed by X-ray structure guided medicinal chemistry opti-
mization. The compounds demonstrate potent inhibition against Pim-1 and Pim-2 in enzyme assays.
Compound 29 has been tested in the Ambit 442 kinase panel and demonstrates good selectivity for
the Pim kinase family. X-ray structures of the inhibitor/Pim-1 binding complex reveal important salt-
bridge and hydrogen bond interactions mediated by the compound’s carboxylic acid and amino groups.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 8 February 2011
Revised 7 March 2011
Accepted 9 March 2011
Available online 16 March 2011
Keywords:
Enzyme,
Kinase
Inhibitor
Benzofuran
X-ray structure
Structure based design
Salt-bridge
Hydrogen bond interaction
Synthesis
Carboxylic acid
Amine
Basicity
SPR
Medicinal chemistry
Fragment screening
The Pim-1 serine/threonine kinases and its two other family
members, Pim-2 and Pim-3 are associated with several oncogenic
processes.1 The pim-1 proto-oncogene was initially identified as a
frequent site of integration for slowly transforming Maloney mur-
ine leukemia virus in murine T cell lymphomas.2 Recently, Pim-1
associated regulation of the c-myc, antiapoptotic protein BAD
and p27kip1 have also been described.3 Pim-1 overexpression is
observed in a range of human lymphomas and acute leukemia,4
whereas Pim-2 overexpression is associated with chronic lympho-
cytic leukemia and non-Hodgkin lymphoma.5 Leukemias driven by
the more common type kinases which are activated by phosphor-
ylation.7 Another structural feature of interest is the presence of an
unusual proline in the hinge region of ATP binding site, that not
only expands the ATP pocket but also alters the kinase hinge
hydrogen bonding motif. Such unique features implicate the po-
tential of finding small molecules which may selectively inhibit
the enzyme by binding to the ATP binding pocket of the Pim kinase
family.
The pleiotropic roles of Pim family kinases in proliferation and
survival pathways have attracted interest of many investigators
to the discovery of inhibitors as potential therapeutics. In the last
a few years, a number of potent Pim-1 inhibitors have been re-
ported in the literature.8 Figure 1 illustrates examples of these
compounds which include substituted fused triheterocycles (1
and 2), a disubstituted pyrazine (3), a disubstituted pyridine (6)
and disubstituted fused diheterocycles (4, 5, 7 and 8).
mutationally-activated FLT3 comprise
a proposed indication
because Pim kinases are transcriptionally activated by FLT3.6
Structural evidence corroborates this mode of activation as Pim ki-
nases appear to be constitutively activated enzymes in contrast to
⇑
As part of a Genzyme drug discovery program aimed at identi-
fying protein kinase inhibitors as potential therapeutics for cancer
Corresponding author.
E-mail address: yibin.xiang@genzyme.com (Y. Xiang).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.03.030

Y. Xiang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3050–3056
3051
N
N
O
N
O
NH
O
O
S
N
NH2
F
COOH
N
N
O
N
N
N
Cl
H
Br
N
Me
Me2N
Me
Cl
3
1
4
2
Ki = 6.3 nM
Ki = 5 nM
IC50 = 46 nM
IC50 = 91 nM
O
N
CN
COOH
H
N
N
N
OH HN
HN
N
N
N
H
N
N
N
H
N
N
N
N
Me
OCF3
CF3
Br
8
6
7
5
IC50 = 10 nM
IC50 = 50 nM
Ki = 21 nM
IC50 = 120 nM
Figure 1. Inhibitors of Pim-1 kinase and the reported IC50 or Ki values.8
Surface Plasmon Resonance (SPR) at 75 lM followed by determina-
Y
tion of steady-state binding affinities of the hits to confirm bind-
ing.9 The active compounds were then evaluated in a Pim-1
biochemical assay by measuring IC50 values.10 Among many active
compounds identified through this process, a few derivatives of
benzofuran-2-carboxylic acid were of special interest. As illus-
trated in Figure 2, compound 9 did not exhibit potent inhibition
of Pim-1, however, derivatives of 9 with a bromo substituent at
the 5-position (10 and 11) demonstrated a significant 12–20-fold
increase in activity. The ligand efficiency (LE) of compounds 10
and 11 is 0.54 and 0.48, respectively (Fig. 2), which are attractively
high.14 To the best of our knowledge, no benzofuran-2-carboxylic
acid derivatives have been reported in the literature as potent
Pim-1 inhibitors. Compounds 10 and 11 were soaked into Pim-1
crystals and the structures of the binding complexes were deter-
mined (Fig. 3).11 In both structures, the bromo group of 10 or 11
is sitting in a hydrophobic pocket created by hinge P125 and sur-
rounded by A65, R122, L44 and L174. This hydrophobic interaction
likely provides a significant contribution to Pim-1 binding and
inhibition because analog 9, which lacks the bromo substituent
W
O
X
9
Y = H, X = H,
10 Y = Br, X = H,
W = COOH; IC50 = 119 uM
W = COOH; IC50 = 8.5 uM
LE: 0.45
LE: 0.54
LE: 0.48
11
Y = Br, X = OMe, W = COOH; IC50 = 5.8 uM
52 Y = Br, X = OMe, W = CN;
No inhibition at 25 uM
Figure 2. Pim-1 inhibitors identified by SPR screening.9 The IC50s were determined
in a Pim-1 biochemical assay.10
treatment, we selected Pim-1 as a target of interest. Our strategy
involved using fragment based screening to find early hit mole-
cules, which would then be optimized by X-ray structure guided
medicinal chemistry optimization to improve their potency and
selectivity. In the pursuit of finding Pim-1 inhibitors, approxi-
mately 1800 fragment molecules (molecular weights in the range
of 150–300 D) were screened for binding activity to Pim-1 using
a
b
R122
L44
A65
R122
A65
K67
K67
E89
E89
F187
W2
W2
P125
L44
P125
F187
D186
D186
L174
L174
Figure 3. (a) X-ray structure of compound 10 binding to the ATP site of Pim-1. (b) X-ray structure of compound 11 binding to the ATP site of Pim-1. In both structures, the
bromo groups of 10 and 11 are located in the hydrophobic pocket surrounded by A65, R122, L44 and L174, and the 2-carboxylic acid group forms the salt-bridge interactions
with K67 and hydrogen bond interactions through the water molecule with E89 and D186. The orientation of the benzofuran core of compound 10 and 11 are almost 180°
opposite to each other.

3052
Y. Xiang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3050–3056
Table 1
Pim-1 and Pim-2 enzymatic inhibition by 5-substituted benzofuran-2-carboxylic acids
R
COOH
H2N
O
a
a
R
IC50
(lM)
IC50 (lM)
Pim-1
5.3
Pim-2
LEb
Pim-1
1.9
Pim-2
LEb
N
12
13
F3C–
13.5
0.45
19
20
1.2
0.33
N
H
N
N
N
N
F3CO–
6.1
7.8
NT
7.1
0.42
0.39
0.46
0.12
0.59
0.37
0.37
H2N
H2N
N
H
N
14
15
16
21
22
23
0.054
N
H
N
N
N
N
H2N
7.4
5.4
11.9
2.9
0.39
0.40
0.20
0.21
0.11
0.15
0.36
0.37
N
H
N
N
N
HN
Me
N
N
H
N
N
N
N
N
H2N
17
18
0.38
6.0
0.16
5.8
0.34
0.31
24
25
0.053
0.45
0.021
0.10
0.37
0.38
N
H
N
H
HN
HN
N
N
N
HN
N
N
H
a
Concentration of the testing compounds to inhibit 50% activity of the target enzymes.10 The values of IC50 are means of at least two independent experiments.
Ligand efficiency.14
b
A65
R122
K67
A65
K67
R122
P125
E89
W2
E89
L44
W2
F187
P125
L44
F187
L174
W1
D186
N172
W1
D186
L174
D128
D128
E171
E171
Figure 4. (a) X-ray structure of compound 21 binding to the ATP site of Pim-1. The pyrazine sits in the position which was partially occupied by the bromo group of
compound 10 (see Fig. 3a). The terminal amino group of the trans-diaminocyclohexanyl moiety forms a salt-bridge interaction with D128 and hydrogen bond interactions
with E171 mediated by a water molecule; (b) X-ray structure of compound 38 binding to the ATP site of Pim-1. The terminal amino group of the cis-diaminocyclohexanyl
moiety forms salt-bridge interactions with D128 and E17 and a hydrogen bond interaction with N172 through a water molecule, while the 5-bromo substituent sits in the
hydrophobic pocket near the hinge region.
at the 5-position, demonstrates only limited inhibitory activity
(Figure 2). The 2-carboxylic acid group forms a salt-bridge interac-
tion with K67 and hydrogen bond interactions (mediated by a
water molecule) to the amino acids E89 and D186. All the three
amino acids are highly conserved among many protein kinases.
The importance of this interaction is demonstrated by compound

Y. Xiang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3050–3056
3053
52. It contains a cyano group at the 2-position rather than carbox-
ylic acid group, which is thus rendered inactive at a concentration
of 25 lM (Fig. 2). It is interesting to note from the X-ray crystal
In view of this knowledge, two approaches were proposed and
examined in an attempt to improve the potency of these early hits:
replacing either the 5-bromo substituent of compound 10 or the 7-
methoxy group of compound 11 with an extended linker fragment
containing a basic moiety to capture additional interactions with
acidic amino acid residues in the ribose binding region, while
maintaining the benzofuran core and 2-carboxylic acid group un-
changed. The potential attribution of hydrogen bond interactions
with the acidic amino acid residues in the ribose binding region
was reported by Qian and his coworkers in their publication in
2009.8c
Table 1 illustrates analogs derived from compound 10 and their
activities against Pim-1 and Pim-2. As expected, the 5-position of
the benzofuran core can tolerate other hydrophobic groups, such
as trifluoromethyl (12) or trifluoromethoxy (13) moieties evi-
denced by that these compounds are equipotent to 10. Replacing
the 5-bromo group with a nitrogen containing heterocycle, such
structures that the orientation of the benzofuran core of com-
pounds 10 and 11 is inversed almost 180°, while the bromo group
remains sitting in almost the same position within the hydropho-
bic pocket. The structural features of the binding complexes of 10
and 11 and the SAR of 9, 10, 11 and 52 lead to several key consid-
erations: (a) hydrophobic interactions between a substituent at the
5-position of the benzofuran and the hydrophobic pocket of the
Pim-1 hinge region may contribute significantly to binding; (b)
the salt-bridge interaction mediated by the 2-carboxylic acid group
to K67 and hydrogen bond interactions with E89 and D186 could
define another important binding site; (c) there is little tolerability
for additional substituents near the gate keeper site and (d) a po-
tential space exits for structural expansion at the ribose binding
region.
Table 2
Pim-1 and Pim-2 enzymatic inhibition by 5-bromo-7-substituted benzofuran-2-carboxylic acids
Br
COOH
O
R
a
a
R
IC50
(
lM)
R
IC50 (lM)
Pim-1
0.060
Pim-2
LEb
Pim-1
0.22
Pim-2
LEb
O
O
O
O
HN
26
27
28
0.25
0.47
0.35
0.47
35
36
37
1.3
0.40
N
HN
HN
H
N
O
4.9
12.5
0.33
0.98
2.7
7.0
1.3
0.38
0.33
N
O
HN
O
0.071
N
H
HN
H
N
H
N
HN
29
30
0.020
0.051
0.19
0.35
0.50
0.46
38
39
0.001
0.019
0.004
0.066
0.59
0.50
H2N
O
H
N
HN
N
H2N
O
H
N
31
4.5
21.1
0.54
0.33
0.46
40
2.3
2.4
6.8
6.1
0.37
0.37
HO
HO
H
N
H
N
32
0.040
41
HN
O
33
34
0.15
0.26
0.34
0.43
0.43
N
H
HN
HN
H
N
0.055
O
a
Concentration of the testing compounds to inhibit 50% activity of the target enzymes.10 The values of IC50 are means of at least two independent experiments.
Ligand efficiency.14
b

3054
Y. Xiang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3050–3056
the crystal structures, an observation consistent with the afore-
O
B
mentioned modest activities of compounds 14–16 (Table 1).
Analogs of compounds 11 were also prepared and their Pim-1
and Pim-2 inhibitory activities are summarized in Table 2. These
analogs were designed to carry a basic moiety at the 7-position lin-
ker substituent in an attempt to capture interactions with D128
and E171. Indeed, this structural modification resulted in signifi-
cant increase of the potency in comparison to parent compound
11. The nature of the linker does not seem to be very important,
as demonstrated by comparing the activities of compound 26 to
29 and 33, or compound 30 to 32 and 34 (Table 2). On the other
hand, the basicity of the nitrogen center of the terminal amine is
a significant factor impacting the potency, as evidenced by com-
paring the IC50 values of compounds 26 and 30 (terminal piperi-
dine, pKa ꢀ11) to compounds 27 and 31 (terminal pyridine,
pKa ꢀ5). Replacing the amino group with a hydroxyl (e.g., 38 to
40, 39 to 41) diminishes the activity which again highlights the
importance of the salt-bridge interactions to D128 and E171 med-
iated by the terminal amino group (see Fig. 4).
Br
a
O
CO2Et
CO2Et
O
O
42
43
b
Ar
Ar
c, d
CO2H
CO2Et
O
14-25
O
A
Scheme 1. Reagents and conditions: (a) bis(pinacolato)diboron, Pd(OAc)2, tricy-
clohexylphosphine, CsF, ACN, reflux (75%); (b) aryl halide, PdCI2(PPh3)2, K2CO3,
DMF/H2O, 100 °C (yields varied); (c) LiOH, THF/CH3OH/H2O; (d) TFA, CH2CI2
(whenever the amino group of ‘Ar’ is protected by Boc).
In addition of the salt-bridge interactions with D128 and E171,
the X-ray structures of the bound complex of 38 (Fig. 4)11 also re-
veals a hydrogen bond interactions with N172 via a water bridge,
though it is difficult to assess the importance of this interaction
for the binding affinity. The position of the 5-bromo group of 38 al-
most coincides with the 5-bromo group of 11. The concise and rigid
cis-diamino-cyclohexanyl ring looked possessing an ideal shape to
position the terminal amino residue in the right distance and ori-
entation, allowing for strong interactions with D128 and E171.
The spatial arrangement could likely be the cause which makes
this relatively small size molecule (38), with a molecular weight
of 353, capable of strong binding and inhibition of both Pim-1
as 3-pyridine (14), 2-pyrazine (15) or 4-N-methylpyrazole (16),
attempting to capture possible interactions to the amino acid res-
idues in the hinge region, did not improve the activity. However, a
significant increase of potency was achieved when a basic moiety
was appended to these heterocylces (17–25). An appropriate dis-
tance between the terminal amino group and the benzofuran core
appears critical to achieve high potency, likely due to optimal
interactions to D128 and E171 at the ribose binding region. This
is evidenced by comparing the potency of compound 17 to 18, as
well as compound 25 to 21 and 22. An X-ray crystal structure of
21 bound to the Pim-1 (Fig. 4a)11 indicates an salt-bridge interac-
tion with D128 and water molecule mediated hydrogen bond inter-
actions with E171 by the terminal amino group. The pyrazine
moiety of 21 is shown to be sitting in the position partially occu-
pied by the bromo substituent of compound 10, though it leans
more towards the hinge site. No direct interaction of the pyrazine
with amino acid residues in the hinge region could be identified in
and Pim-2 as demonstrated by IC50’s of 0.001 lM and 0.004 lM,
respectively (Table 2). The LE of compound 38 is 0.59 (Table 2),
which is significantly increased from the LE of the original hit mol-
ecule 11 (0.48, Fig. 2).
Compounds 12 and 13 of Table 1 were prepared according to lit-
erature procedures,12 while the remaining compounds (14–25) in
Br
Br
Br
g, d
CO2CH3
a, b
CO2CH3
CO2H
O
O
O
O
O
45
OH
OH
OCH3
10
44
c
h, e, d
Br
Br
CO2CH3
CO2H
e
Br
O
O
COOH
NH
O
O
O
R
n
O
O
B
d, e, f
N
N
27
; n=1
35
31; n=2
Br
CO2H
O
O
n
26; n=1, X=NH, Y=CH2
28
; n=1, X=CH2, Y=NH
30; n=2, X=NH; Y=CH2
Y
X
Scheme 2. Reagents and conditions: (a) BBr3, CH2CI2; (b) CH3OH, cat. H2SO4, reflux (quant., two steps); (c) ROH, DEAD, PPh3, EtOAc; (d) TFA, CH2CI2; (e) LiOH, THF/CH3OH/
H2O (>90%); (f) HCI, H2O; (g) t-butyl bromoacetate, K2CO3, DMF (97%); (h) N-Boc-piperazine, EDC, CH2CI2 (79%).

Y. Xiang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3050–3056
3055
O
Br
Br
H
CO2Et
a
O
OH
NO2
47
NO2
46
b
HO2C
N
Boc
BocHN
O
n
Br
D
33
; n=0
38; cis
CO2Et
34; n=1
39
; trans
g, e, f
O
c, d, e, f
NH2
48
c, d, e
OHC
O
n
E
c, e, f
40; cis
41
N
Boc
; trans
OH
n = 0,1
29; n=0
32; n=1
Scheme 3. Reagents and conditions: (a) diethyl bromomalonate, K2CO3, 2-butanone, reflux; (b) Fe, NH4CI, EtOH/H2O; (c) ketone/aldehyde, cat. AcOH, NaBH(OAc)3, DCE; (d)
cis/trans chromatographic resolution; (e) LiOH, THF/CH3OH/H2O; (f) TFA, CH2CI2; (g) carboxylic acid, HATU, DIEA, DMF (>70%).
H2N
n
n = 0,1
C
O
N
Boc
Br
Br
H
b, c
a
36; n=0
37
CO2t-Bu
; n=1
O
OH
CO2R
49
d, e
CO2H
51
(R=H)
50
(R=CH3)
Scheme 4. Reagents and conditions: (a) CH3OH, cat. H2SO4, reflux (47%); (b) di-t-butyl bromomalonate, K2CO3, 2-butanone, reflux (50%); (c) LiOH, THF/CH3OH/H2O (16%); (d)
amine, HATU, DIEA, DMF (80/25%); (e) TFA, CH2CI2 (100/42%.).
Table 1 were prepared as shown in Scheme 1. For example, ethyl 5-
bromo-benzofuron-2-carboxylate (42) was converted to a boronic
ester (43), which then was coupled with a variety of aryl halides to
afford intermediate A. Upon hydrolysis with aqueous lithium
hydroxide and, if necessary, deprotection of the Boc-amino group
with acid, the desired compounds 14–25 were obtained. The com-
pounds listed in Table 2 were prepared using the methods outlined
in Schemes 2–4. For instance, in Scheme 2 compound 10 was con-
verted to 44 by demethylation with BBr3, followed by acid cata-
lyzed esterification with methanol. Mitsunobu reaction of 44
with selected alcohols produced intermediate B, from which the
desired acids (26–28, 30, 31) were prepared by conventional chem-
ical transformations. The intermediate 44 was also used to prepare
compound 35. Thus, O-alkylation with t-butylbromoacetate fol-
lowed by t-butyl cleavage afforded intermediate 45, which was
then subjected to EDC mediated amide formation, basic hydrolysis
and, finally, Boc-deprotection to produce the target compound
(35). Scheme 3 illustrates the synthesis of compounds 29, 32–34,
38–41 from a common intermediate 48. The commercially avail-
able 5-bromo-2-hydroxy-3-nitrobenzaldehyde (46) was reacted
with diethyl bromomalonate to form 47, which was reduced to
48 by iron powder and ammonium chloride in aqueous ethanol.
Standard reductive amination or amide coupling was used to at-
tach the tailpiece followed by ester hydrolysis and/or amine depro-
tection completed the syntheses. The compounds 36 and 37 were
prepared from the known intermediate 5-bromo-3-formyl-2-
hydroxybenzoic acid (49)13 using the following sequence as illus-
trated in Scheme 4: (a) conversion to methyl ester 50; (b) reaction
with t-butyl bromomalonate followed by selective hydrolysis to
form 51; (c) HATU mediated amide formation by coupling with
N-Boc protected piperidinyl amines C and (d) simultaneous depro-
tection of the N-Boc and the t-butyl ester functionalities with TFA.
The active compounds have also been tested in a panel of
in vitro ADME assays. Results for the three most potent Pim-1
inhibitors are summarized in Table 4. All three compounds were
Table 3
Ambit kinase selectivity data of compound 29a
Ambit gene symbol
% Control
PIM1
PIM3
9.6
13
MYO3B
CSNK2A1
PIM2
14
21
26
CSNK2A2
28
a
Compound 29 was tested in an Ambit 442
kinase panel at concentration of 100 nM.

3056
Y. Xiang et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3050–3056
Table 4
In vitro ADME data for selected Pim-1 inhibitors
b
c
d
d
d
d
d
Compds
MW
Sol., pH 7.4a
Microsomes rat CLint
Microsomes human CLint
1A2 IC50
2C9 IC50
2C19 IC50
2D6 IC50
3A4 IC50
29
38
39
353
353
353
>58
20.1
>58
14.1
10.0
8.1
9.9
10.2
11.2
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
a
b
c
Solubility measurement determined from DMSO stock at pH 7.4 in phosphate buffered saline, recorded as
Metabolic stability in rat liver microsomes. Intrinsic clearance reported as L/min/mg protein.
Metabolic stability in human liver microsomes. Intrinsic clearance reported as L/min/mg protein.
Cytochrome P450 inhibition using human liver microsome and selected chemical probe substrates, reported as
lg/mL.
l
l
d
lM.
Qian, K.; Wang, L.; Cywin, C. L.; Hickey, E.; Homon, C.; Jakes, S.; Kashem, M. A.;
Lee, G.; Leonard, S.; Li, J.; Magboo, R.; Mao, W.; Pack, E.; Peng, C.; Welzel, M.;
Wolak, J.; Morwick, T. J. Med. Chem. 2009, 52, 1814; (d) Pierce, A. C.; Jacobs, M.;
Stuver-Moody, C. J. Med. Chem. 2008, 52, 1972; (e) Chen, L. S.; Redkar, S.; Bearss,
D.; Wierda, W. G.; Gandhi, V. Blood 2009, 114, 4150; (f) Cheney, I. W.; Yan, S.;
Appleby, T.; Walker, H.; Vo, T.; Yao, N.; Hamatake, R.; Hong, Z.; Wu, J. Z. Bioorg.
Med. Chem. Lett. 2007, 17, 1679; (g) Grey, R.; Pierce, A. C.; Bemis, G. W.; Jacobs,
M. D.; Moody, C. S.; Jajoo, R.; Mohal, N.; Green, J. Bioorg. Med. Chem. Lett. 2007,
19, 3019; (h) Pogacic, V.; Bullock, A. N.; Fedorov, O.; Filippakopoulos, P.; Gasser,
C.; Biondi, A.; Meyer-Monard, S.; Knapp, S.; Schwaller, J. Cancer Res. 2007, 67,
6916.
inactive against a set of five CYP isozymes, demonstrated adequate
aqueous solubility and were moderately stable in rat and human
liver microsome metabolic stability assays.
To assess the selectivity of the inhibitors, compound 29 was
tested in an Ambit 442 kinase panel at a concentration of
100 nM, about five-fold higher than its IC50 value in the Pim-1 inhi-
bition assay. Under these conditions, compound 29 inhibits, in
addition to Pim-1, Pim-2 and Pim-3, only three non-Pim kinases
(MYO3B, CSNK2A1 and CSNK2A2) for >50% of the binding (Table
3), indicating the good selective nature of this structural class
against the Pim kinase family.
9. Pim-1 protein was chemically minimally biotinylated with EZ-link Sulfo-NHS-
LC-LC-biotin (Thermo Scientific) and removed of free biotin reagent by buffer
exchanging 10 times with Amicon 3K Ultracel Membrane Ultra Centrifugal
Filters (Millipore). All biosensor work was performed on
a Biacore T100
In summary, using fragment based screening against Pim-1 and
X-ray structure guided medicinal chemistry optimization, we have
identified novel benzofuran-2-carboxylic acid compounds which
potently inhibit Pim-1 and Pim-2. These inhibitors have good
selectivity as indicated by profiling compound 29 in an Ambit
442 kinase panel. The X-ray structures of these compounds bound
to Pim-1 revealed important hydrogen bond interactions involving
the 2-carboxylate group and the terminal amino group. The discov-
ery of potent and selective Pim kinase inhibitors provides useful
tools for the further evaluation of the role of this kinase family in
the context of disease states and possible interventions in the
treatment of cancer or other diseases.
instrument (GE Healthcare). NeutrAvidin (NA, Thermo Scientific) was amine
coupled to the surface of a CM5 sensor chip to 8000–12,000 RU by standard
methods. Biotinylated Pim-1 was captured onto the CM5-NA surface to 4000–
6000 RU. Compounds were individually screened for binding at
a final
concentration of 75 M at 4 °C. Steady-state affinities were determined for
l
select compounds using similar assay conditions. Raw sensorgrams were
double-referenced and solvent corrected with Biacore T100 Evaluation
Software.
10. Human Pim-1(0.6 nM), Pim-2 (1.2 nM) or Pim-3 (0.6 nM) enzyme was pre-
incubated with compound in a 100 mM Hepes pH 7.5 buffer containing 0.01%
Triton X-100, 10 mM MgCl2, 0.1% BSA, 1 mM DTT, 10
orthovanadate, 10 M beta-glycerophosphate and 1.25% DMSO for 15 min at
room temperature. The reaction was initiated by addition of 1 M peptide
substrate (5-FAM-RSRHSSYPAGT-CONH2, Caliper Life Sciences, MA) and
150 M ATP for Pim-1, 3 M ATP for Pim-2 or 25 M ATP for Pim-3 assay.
lM sodium
l
l
l
l
l
The reaction was incubated at room temperature for 45 min for Pim-1, 90 min
for Pim-2 and Pim-3 and terminated with a 100 mM Hepes pH 7.5 buffer
containing 100 mM EDTA, 0.02% Brij, 0.1% CR-3 and 0.36% DMSO. The reaction
References and notes
product was detected by one cycle run on
Sciences) using an off-chip mobility shift protocol.
a LabChip 3000 (Caliper Life
1. (a) Amaravadi, R.; Thompson, C. B. J. Clin. Invest. 2005, 115, 2618; (b) Magnuson,
N. S.; Wang, Z.; Ding, G.; Reeves, R. Future Oncol. 2010, 6, 1461; (c) Nawijn, M.
C.; Alendar, A.; Berns, A. Nat. Rev. Cancer 2011, 11, 23.
2. Cuypers, H. T.; Selten, G.; Quint, W.; Zijlstra, M.; Robanus-Manndag, E.; Boelens,
W.; van Wezenbeek, P.; Berns, A. Cell 1984, 37, 141.
3. (a) Verbeek, S.; van Lohuizen, M.; van der Valk, M.; Domen, J.; Kraal, G.; Berns,
A. Mol. Cell. Biol. 1991, 11, 1176; (b) Zhang, Y.; Wang, Z.; Li, X.; Magnuson, N. S.
Oncogene 2008, 27, 4809; (c) Li, S.; Xi, Y.; Zhang, H.; Wang, Y.; Wang, X.; Liu, H.;
Chen, K. Oncol. Rep. 2010, 24, 997; (d) Cen, B.; Mahajan, S.; Zemskova, M.;
Beharry, Z.; Lin, Y. W.; Cramer, S. D.; Lilly, M. B.; Kraft, A. S. J. Biol. Chem. 2010,
285, 29128.
11. Purified Pim-1 protein (aa 29–313) with a C-terminal His tag was concentrated
to 10–13 mg/ml in 20 mM HEPES, pH 8, 120 mM NaCl, 5 mM DTT. Pim-1 was
crystallized in 0.1 M imidazole, pH 6.4, 1–1.4 M sodium acetate anhydrous by
sitting drop vapor diffusion at 4 °C and reached maximum size after about
5 days. Pim-1 crystals were soaked with 1–10 mM compound in the well
solution overnight at 4 °C and flash frozen with 30% glycerol in the presence of
the compound. Diffraction data were collected at ALS503 (Advanced Light
Source at Lawrence Berkeley National Laboratory, Berkeley, CA) and processed
using HKL2000 and Scala. Crystals belong to space group P65, with a = 98 Å,
b = 98 Å, c = 80 Å,
a = 90.00, b = 90.00, c = 120.00. The structures were solved by
4. Amson, R.; Sigaux, F.; Przedborski, S.; Flandrin, G.; Givol, D.; Telerman, A. Proc.
Natl. Acad. Sci. U.S.A. 1989, 86, 8857.
molecular replacement using Molrep with PDB 1YWV as the search model.
Iterative manual model building was carried out with Coot, coupled with
refinement using Refmac5. For the crystal structures in complex with
compound 21 and 38, the Fo–Fc difference density indicates a peptide was
bound. A tentative four residue peptide was build into the density and refined
together with the Pim-1 structure. The X-ray coordinates of the structures of
the bound complexes have been deposited in the RCSB protein data bank. They
are 3R00 (structure a in Fig. 3), 3R01 (structure b in Fig. 3), 3R02 (structure b in
Fig. 4) and 3R04 (structure a in Fig. 4).
5. Cohen, A. M.; Grinblat, B.; Bessler, H.; Kristt, D.; Kremer, A.; Schwartz, A.;
Halperin, M.; Shalom, S.; Merkel, D.; Don, J. Leuk. Lymphoma 2004, 45, 951.
6. (a) Kim, K. T.; Baird, K.; Ahn, J. Y.; Meltzer, P.; Lilly, M.; Levis, M.; Small, D. Blood
2005, 105, 1759; (b) Kim, K. T.; Levis, M.; Small, D. Br. J. Haematol. 2006, 124,
500.
7. Qian, K. C.; Wang, L.; Hickey, E. R.; Studts, J.; Barringer, K.; Peng, C.; Kronkaitis,
A.; Li, J.; White, A.; Mische, S.; Farmer, B. J. Biol. Chem. 2005, 280, 6130.
8. (a) Tong, Y.; Stewart, K. D.; Thomas, S.; Przytulinska, M.; Johnson, E. F.;
Klinghofer, V.; Leverson, J.; McCall, O.; Soni, N. B.; Luo, Y.; Lin, N. H.; Sowin, T. J.;
Giranda, V. L.; Penning, T. D. Bioorg. Med. Chem. Lett. 2008, 18, 5206; (b) Tao, Z.
F.; Hasvold, L. A.; Leverson, J. D.; Han, E. K.; Guan, R.; Johnson, E. F.; Stoll, V. S.;
Stewart, K. D.; Stamper, G.; Soni, N.; Bouska, J. J.; Luo, Y.; Sowin, T. J.; Lin, N. H.;
Giranda, V. S.; Rosenberg, S. H.; Penning, T. D. J. Med. Chem. 2009, 52, 6621; (c)
12. Bernd, P.; Sonja, B.; Nielsen, R.; Kiuff, D.; Rimvall, K. Bioorg. Med. Chem. Lett.
2006, 16, 3162.
13. Ruell, J. A.; De Clerq, E.; Pannecouque, C.; Witvrouw, M.; Stup, T. L.; Turpin, J. A.;
Buchheit, R. W.; Cushman, M. J. Org. Chem. 1999, 64, 5858.
14. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.