Discovery of novel furanone derivatives as potent Cdc7 kinase inhibitors

Article abstract of DOI:10.1016/j.ejmech.2017.02.030

Cdc7 is a serine-threonine kinase and plays a conserved and important role in DNA replication, and it has been recognized as a potential anticancer target. Herein, we report the design, synthesis and structure-activity relationship of novel furanone derivatives as Cdc7 kinase inhibitors. Compound 13 was identified as a strong inhibitor of Cdc7 with an IC₅₀ value of 0.6?nM in the presence of 1?mM ATP and showed excellent kinase selectivity. In addition, it exhibited slow off-rate characteristics, which may offer advantages over known Cdc7 inhibitors in its potential to yield prolonged inhibitory effects in?vivo. Compound 13 potently inhibited Cdc7 activity in cancer cells, and effectively induced cell death.

Full text of DOI:10.1016/j.ejmech.2017.02.030

Accepted Manuscript
Discovery of novel furanone derivatives as potent Cdc7 kinase inhibitors
Takayuki Irie, Tokiko Asami, Ayako Sawa, Yuko Uno, Mitsuharu Hanada, Chika
Taniyama, Yoko Funakoshi, Hisao Masai, Masaaki Sawa
PII:
S0223-5234(17)30095-8
10.1016/j.ejmech.2017.02.030
EJMECH 9226
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 14 December 2016
Revised Date: 9 February 2017
Accepted Date: 11 February 2017
Please cite this article as: T. Irie, T. Asami, A. Sawa, Y. Uno, M. Hanada, C. Taniyama, Y. Funakoshi,
H. Masai, M. Sawa, Discovery of novel furanone derivatives as potent Cdc7 kinase inhibitors, European
Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.02.030.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Discovery of novel furanone derivatives as potent
Cdc7 kinase inhibitors
Takayuki Irie,*,† Tokiko Asami,† Ayako Sawa,† Yuko Uno,† Mitsuharu Hanada,†
Chika Taniyama,‡Yoko Funakoshi,‡ Hisao Masai,§ and Masaaki Sawa†
†Research and Development, Carna Biosciences, Inc., 3F BMA, 1-5-5 Minatojima-Minamimachi,
Chuo-ku, Kobe 650-0047, Japan
‡Research and Development Department, SBI Biotech Co., Ltd., Izumi Garden Tower 18F,
1-6-1 Roppongi, Minato-ku, Tokyo 106-6018, Japan
§Department of Genome Medicine, Tokyo Metropolitan Institute of Medical Science, 2-1-6
Kamikitazawa, Setagaya-ku, Tokyo 156-8506, Japan

ACCEPTED MANUSCRIPT
Highlights
ò
ò
ò
ò
Novel furanone derivatives were discovered as potent Cdc7 inhibitors and the structure-
activity relationships were discussed.
Compound 13 inhibited Cdc7 in a time-dependent fashion, and showed slow off-rate
kinetics.
The slow off-rate of compound 13 correlates with a prolonged down-regulation of MCM2
phosphorylation in tumor cells.
Compound 13 potently inhibited Cdc7 activity in cancer cells, and effectively induced cell
death.
Abstract
Cdc7 is a serine-threonine kinase and plays a conserved and important role in DNA replication,
and it has been recognized as a potential anticancer target. Herein, we report the design,
synthesis and structure-activity relationship of novel furanone derivatives as Cdc7 kinase
inhibitors. Compound 13 was identified as a strong inhibitor of Cdc7 with an IC50 value of 0.6
nM in the presence of 1 mM ATP and showed excellent kinase selectivity. In addition, it
exhibited slow off-rate characteristics, which may offer advantages over known Cdc7 inhibitors
in its potential to yield prolonged inhibitory effects in vivo. Compound 13 potently inhibited
Cdc7 activity in cancer cells, and effectively induced cell death.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Graphical abstract
Keywords
Kinase inhibitors; Cell cycle; Apoptosis; Anticancer activity; Furanone; Slow binding

ACCEPTED MANUSCRIPT
1. Introduction
Cdc7 is a serine-threonine kinase and plays a conserved and important role in DNA replication
[1-3]. In the initiation of S-phase in cell cycle, an activation subunit Dbf4 (also known as ASK)
binds to Cdc7 to activate its kinase activity, which leads to the phosphorylation of MCM [4].
This step is crucial for the initiation of DNA replication by recruiting other replication factors
including Cdc45 to pre-replicative complex (pre-RC). Recently, significant interest has arisen in
Cdc7 as a potential anticancer target [5,6], since the expression level of Cdc7 is frequently
elevated in various cancer cell lines and human tumor tissues [7,8]. In addition, Cdc7
overexpression has been associated with poor prognosis in various types of cancers [9,10].
In spite of the indisputable positive impact of Cdc7 as a drug target, there have been only a few
known classes of Cdc7 inhibitors reported so far, probably due to the unique structural feature of
Cdc7 protein including the presence of two large polypeptide segments which interrupt the
kinase domains at conserved locations [11]. Nevertheless, several Cdc7 inhibitors have been
reported to date (Figure 1), since the first nano-molar Cdc7 inhibitor (1a) was reported by
Vanotti et al. in 2008 [12-18]. However, most of the inhibitors only show moderate cellular
activity [12,13,15,16], probably due to via competition with ATP as Cdc7 has a relatively low
Km for ATP (Kmapp = 2.8 uM) compared with other kinases. To overcome the weak cellular
activity, we set out to find a novel Cdc7 inhibitor that is insensitive or resistant to high ATP
concentration.
In this report, we describe the design and syntheses of novel furanone-based inhibitors of Cdc7
kinase for anticancer treatment as exemplified by 3. These analogs displayed high inhibitory
potency for Cdc7 even in the presence of high ATP concentration. The selected compound was

ACCEPTED MANUSCRIPT
subjected to kinetic studies to characterize mode of inhibition. In addition, the effects on the
cancer cell were studied and discussed.
O
O
O
NH2
HCl
O
NH
H
N
NH
N
N
H
N
N
N
H
N
Cl
H2N
1a
1b
1c
O
O
Cl
F
NH
N
N
N
N
N
N
N
S
N
N
N
N
N
N
O
N
H
HN
H
H
HN
HN
HN
OMe
2a
2b
2c
2d
O
O
O
O
CO2Et
N
R1
R2
R3
O
N
H
HN
HN
3
Figure 1. Chemical structures of representative Cdc7 inhibitors (1a, IC50 = 10 nM [12];
1b, IC50 = 22 nM [13]; 1c, IC50 = 3.4 nM [14]), azaindole class of Cdc7 inhibitors (2a,
IC50 = 9 nM [15]; 2b, Ki = 0.2 nM [16]; 2c, Ki = 0.9 nM [17]; 2d, IC50 = 1.1 nM [18]),
and furanone-based inhibitors (3 as an initial compound) reported in this manuscript.
2. Results and discussion
2.1 Identification and characterization of compound 3
Cdc7 has a relatively low Km for ATP (Kmapp = 2.8 µM) compared with other kinases, and we
thought this high affinity for ATP might be a bottleneck to generate an effective inhibitor in vivo
,

ACCEPTED MANUSCRIPT
because an ATP competitive inhibitor would be competed out by the high concentration of ATP
in the cell to lose its potency in vivo [19]. Therefore we have implemented a high-throughput
screening (HTS) over our in-house compound library in the presence of 100 µM ATP which
corresponds to 36-times higher than the Km value, expecting to find hit compounds that are
insensitive or resistant to high ATP concentration. A hit compound having 7-azaindole group
was identified from the HTS, and the subsequent Hit-to-Lead study led to a novel furanone
analog 3 as the starting point of lead optimization. This 7-azaindole substructure is shared with
some of Cdc7 inhibitors (2a to 2d) as shown in Figure 1. The 7-azaindole is a well-known
privileged fragment as a hinge binding group [20,21], which would be responsible for the high
inhibitory activities against Cdc7 through bidentate hydrogen bonds with kinase hinge region.
Compound 3 has a characteristic double bond between 7-azaindole ring and furanone ring. To
predict the stereochemistry of our furanone analog and gain the structural insights on the binding
mode of compound 3 at the Cdc7 active site, a modeling study was performed using a homology
model developed based on a known crystal structure of the closely related kinase CK2 [12,22]
(the crystal structure of Cdc7 was published during the course of this study [23]). As shown in
Figure 2, we found that only the Z-isomer of 3 was able to be docked into the model. Ermoli et
al. reported a series of Cdc7 inhibitor having 7-azaindole ring, and they also concluded Z-
configuration of the double bond between 7-azaindole ring and core structure was important for
the inhibitory activity [15]. As indicated in the model, the 7-azaindole moiety was predicted to
form bidentate hydrogen bonds with the backbone amides in the hinge region (NH with carbonyl
O of Pro135 and N7 with NH of Leu137) (Figure 2a). The carbonyl O of the furanone ring and
the ester group were predicted to interact with the side chain amino group of Lys90, a catalytic
residue well conserved among protein kinases [24]. Interestingly, a possible intramolecular

ACCEPTED MANUSCRIPT
hydrogen bond in compound 3 between carbonyl O of the ester and the NH of the aniline group
was predicted by this model as an energetically favorable conformation. As shown in Figure 2b,
the aniline and ester moieties were exposed to the solvent accessible regions, providing
opportunities for further modifications. On the other hand, the model suggests the presence of
very little space around the 7-azaindole moiety.
a
b
Figure 2. Predicted binding mode of compound 3 (orange) in the ATP binding pocket of Cdc7.
Interactions between compound 3 and Cdc7 (a). The surface model was generated from Cdc7
protein (b).
2.2 Structure-activity relationship studies of furanone derivatives
To study the structure-activity relationship (SAR) of the furanone analogs more clearly,
enzymatic assays of Cdc7 were performed at 5 µM ATP concentration. Under this low ATP
concentration, the initial compound 3 showed 21-fold stronger inhibitory potency (IC50 = 3.3
nM) than that at 100 µM ATP concentration, suggesting that the compound is an ATP-
competitive inhibitor.

ACCEPTED MANUSCRIPT
Initially, we synthesized compounds having different X and R1 group to examine the impact of
the furanone ring and azaindole ring on the potency. As shown in Table 1, replacing O with S
resulted in a complete loss of potency (compound 4). The introduction of Cl at the 5-position in
the azaindole ring (5) deteriorated its inhibitory potency by one order of magnitude as predicted
from the binding model (IC50 = 44 nM). Incorporation of phenol (6) or methylenedioxyphenyl
(7) groups as hinge binders resulted in a dramatic decrease in potency (IC50 = 1200 and 3200
nM, respectively). These results suggest that the furanone ring and the unsubstituted azaindole
ring are important for the inhibitory activity.
Table 1. Effects of furanone and azaindole modification on Cdc7 inhibitory activities
R1
X
O
Cdc7 IC50
Cpd
(nM)a
3.8
3
S
O
O
>10,000
44
4
5
6
1,200

ACCEPTED MANUSCRIPT
O
3,200
7
a IC50 values are reported as the mean of duplicated measurements in the presence of 5 µM of
ATP. See Experimental Section for the assay details.
Because the ester was expected to form important hydrogen bond with Lys90, an SAR study
for the ester group was performed (Table 2). As expected, compounds having hydrogen bond
accepting oxygen such as OMe (8) and OH (9) maintained potency (IC50 = 3.3 and 9.9 nM,
respectively). However, significant decrease in potency was observed when hydrogen bond
donor functional groups were installed (10 and 11). Introduction of piperidine (12) resulted in a
modest decrease in potency (IC50 = 890 nM), suggesting the piperidine ring could interact with
Cdc7 via Van der Waals contact. Generally esters are labile and can be easily hydrolyzed to the
corresponding carboxylic acid. To our surprise, the ethyl ester of 3 was very stable even under
strong basic conditions, and it required an extremely strong basic condition (50% w/v KOH aq.,
reflux) to hydrolyze this ester. Therefore, we decided to keep the ester group at this position.
Table 2. Effects of R2 on Cdc7 inhibitory activities
Cpd
3
R2
OEt
OMe
OH
Cdc7 IC50 (nM)a
3.8
3.3
9.9
8
9

ACCEPTED MANUSCRIPT
NH2
NHEt
>10,000
1,600
890
10
11
12
1-Piperidinyl
a
IC50 values are reported as the mean of duplicated measurements in the presence of 5 µM of
ATP. See Experimental Section for the assay details.
The effects of substituents R4 at the benzene ring are shown in Table 3. Generally, various
aromatic ring substitutions were well tolerated, and para- and ortho- positions were more
favorable for potency (13 – 18). Substitution of the hydrogen atom with electron-withdrawing
groups at the ortho-position (13 and 19) resulted in slight increases in potency compared to
compound 3. Secondary substitution at the para-position did not affect IC50 values (23 vs 19 and
24 vs 21).
Compound 3 displayed moderate metabolic stability in human liver microsome, but it was
rapidly metabolized in mice (Table 3). We hypothesized that the unsubstituted benzene ring of
compound 3 could be metabolized rapidly, and thus metabolic stabilities of substituted benzene
derivatives were evaluated. As shown in Table 3, the ortho-Cl substitution (13) dramatically
improved metabolic stabilities in both human and mice liver microsomes (parent compound
remaining after 30 min: 67% and 58%, respectively).

ACCEPTED MANUSCRIPT
Table 3. Effects of substitution at benzene ring on Cdc7 inhibitory activities and liver
microsomes stabilities
HLMb
%
MLMb
%
R4
Cdc7
Cpd
IC50 (nM)a
H
2-Cl
3.8
2.4
12
23
67
-
7.9
58
-
3
13
14
15
16
17
18
19
20
21
22
23
24
3-Cl
4-Cl
6.7
12
-
-
2-MeO
3-MeO
4-MeO
2-F
-
-
29
-
-
4.9
2.4
4.3
3.8
5.9
2.5
3.9
33
30
52
53
-
5.1
10
25
24
-
4-F
2-Me
4-Me
2,4-di-F
2,4-di-Me
57
63
12
30
a IC50 values are reported as the mean of duplicated measurements in the presence of 5 µM of
ATP. See Experimental Section for the assay details.
b Remaining % of parent compounds after 30 min treatment of liver microsomes. HLM and
MLM: human and mouse liver microsomes.
Replacement of the benzene ring with other heterocycles could provide an insight into
structural requirement of the binding site (Table 4). 3- And 4-pyridines (26 and 27) showed

ACCEPTED MANUSCRIPT
potency comparable to that of the benzene analog 3 (IC50 = 8.0 and 3.5 nM, respectively).
However, 2-pyridine 25 exhibited a significantly reduced potency (IC50 = 30 nM), maybe due to
a decrease in hydrophobic interactions at this position with Cdc7. Replacing with other
heteroaryl groups (28 – 31) reduced the potency by 3 – 10-fold, except 6-indazole analog 31
(IC50 = 4.0 nM). Insertion of alkyl chains between phenyl ring and NH resulted in dramatic
decreases in potency (32 and 33).
Table 4. Effect of R3 position on Cdc7 inhibitory activities
R3
Cdc7
Cpd
IC50 (nM)a
Ph
2-Py
3.8
30
8.0
3.5
18
10
36
4.0
23
30
3
25
26
27
28
29
30
31
32
33
3-Py
4-Py
5-Pyrimidinyl
3-Pyrazolyl
6-Quinolinyl
6-Indazoyl
Benzyl
Phenethyl
a IC50 values are reported as the mean of duplicated measurements in the presence of 5 µM of
ATP. See Experimental Section for the assay details.

ACCEPTED MANUSCRIPT
2.3 ATP dependency and pre-incubation effects of 13 on Cdc7 inhibition
As compound 13 showed improved metabolic stabilities in mice and human, this compound
was selected for further evaluations. ATP competition assay was performed to determine
whether compound 13 is ATP competitive or not. Dose response curves of compound 13 in the
presence of various ATP concentrations are shown in Figure 3a. Compound 13 displayed ATP-
dependent proportional decreases in inhibitory potencies for Cdc7, suggesting that it is an ATP
competitive inhibitor. The increase in ATP concentrations resulted in a dramatic increase in IC50
values of 13 (IC50 = 191 nM in the presence of 1 mM ATP). It has been reported that some
kinase inhibitors show time-dependent inhibition due to slow on- and/or off-rate for binding [25-
28]. Therefore, we have examined a time dependency of Cdc7 inhibition by 13 with pre-
incubation method. Namely, compound 13 was pre-incubated for 30 minutes with Cdc7 prior to
the addition of the substrates mixture to start enzymatic reactions. Surprisingly, the dose
response curves in the presence of various ATP concentrations were not changed, and the
inhibitory potencies were sustained even in the presence of 1 mM ATP when 13 was pre-
incubated with the enzyme (Figure 3b). On the other hand, other chemotype of compound 1b did
not show such pre-incubation effects (data not shown), suggesting that this pre-incubation effect
was unique to our furanone-based inhibitor.

ACCEPTED MANUSCRIPT
b)
a)
100
80
60
40
20
0
100
80
60
40
20
0
0.0001
0.001
0.01
0.1
0.0001
0.001
0.01
0.1
13 (µM)
13 (µM)
Figure 3. Dose response curves of compound 13 for Cdc7 inhibition in the presence of various
ATP concentrations: 5 µM (closed circle), 25 µM (open circle), 100 µM (closed triangle), 1000
µM (open triangle). The enzymatic assays were performed without (a) or with pre-incubation (b).
2.4 Comparison of dissociation rates from Cdc7
To better understand the observed time dependent inhibition, compound 13 was subjected to
the rapid dilution assay [26] to study dissociation rate, and compared with compound 1b. The
inhibitor-enzyme mixture was rapidly diluted into the assay buffer containing the substrate and a
high concentration of ATP, and then the phosphorylated product formation was monitored over
500 min of time course. As shown in Figure 4, the curve of the product formation in the presence
of compound 1b was almost the same as that of vehicle control. On the other hand, Cdc7
produced the phosphorylated product with much reduced rate after treatment with compound 13.
We estimated the off-rate of compound 13 from Cdc7 (koff) to be 3.5 × 10-3 min-1 (t1/2 = 197
min). These results suggested that compound 13 has a unique inhibitory mechanism that binds to
Cdc7 in a reversible fashion but has a very slow off-rate.

ACCEPTED MANUSCRIPT
1000
800
600
400
200
0
0
100
200
300
400
500
Time (min)
Figure 4. Rapid dilution assay for Cdc7 inhibitors. Recovery of enzymatic activity was
monitored by formation of the phosphorylated product. DMSO control (closed circle), compound
1b (open triangle), compound 13 (open square).
2.5 Kinase selectivity of compound 13
To investigate the kinase selectivity at physiological ATP levels, compound 13 was screened
over a panel of 121 kinases in the presence of 1 mM ATP with preincubation method.
Compound 13 showed an excellent selectivity profile with only 7% of the panel inhibited (>50%
inhibition at 0.1 µM). IC50 values of hit kinases were determined to confirm the selectivity of
compound 13. As shown in Table 5, cross-reactivity with CLKs were observed albeit with
selectivity in favor of Cdc7 (8-fold). In addition, the assays demonstrated that compound 13 has
fair to good selectivity for other off-target hit kinases, with 61 to 655-fold selectivity.
Table 5. Kinase selectivity of compound 13
Kinases
Cdc7
IC50 (nM)a
0.6
Fold selectivityb
1.0

ACCEPTED MANUSCRIPT
CLK1
CLK2
GSK3α
GSK3
DYRK1B
Erk2
5.0
5.3
37
8.3
8.8
61
51
85
53
88
61
102
129
463
665
PIM1
77
Erk1
278
p70S6K
399
a IC50 values are reported as the mean of duplicated measurements. b Data are expressed as fold
selectivity of the IC50 value for each kinase versus Cdc7.
2.6 Cellular effects of compound 13
The results of the studies for the enzyme inhibition and the kinase selectivity profiling
confirmed that compound 13 was a potent and selective inhibitor of Cdc7, and thus this
compound was subjected to cellular experiments. To investigate the effects of selective Cdc7
inhibition in cancer cells, we evaluated the ability of compound 13 to inhibit phosphorylation of
MCM2 [29,30] and the functional consequences in Colo-205 cells. As shown in Figure 5a,
compound 13 potently inhibited phosphorylation at Ser53 of MCM2 even at 0.1 µM. It has been
reported that knockdown of Cdc7 was shown to cause cell death in cancer cells, because it
results in cell cycle progression through a defective S phase and the subsequent DNA strand
breakage that leading to p53-independent apoptotic cell death [31,32]. In 13-treated cells, the
DNA strand breakages were observed in response to the MCM2 inhibition, by analyzing
phospho-H2AX ( -H2AX), a well-established marker of DNA double strand breaks [33]. Dose-
dependent increases in cleaved PARP, the apoptosis marker [34] were also observed, suggesting
the observed cell death was elicited by selective inhibition of Cdc7 by 13. Compound 13-treated
cells were then subjected to cell cycle analysis (Figure 5b). DNA contents in Colo-205 cells were

ACCEPTED MANUSCRIPT
analyzed by flow cytometry after 48 hours of compound treatment. Marked decreases in G1
phase peak and significant increase in sub-G1 population were observed after treatment with
0.33 and 1 µM compound 13, as indicative of cell death [35]. This is consistent with the previous
report that Cdc7 knockdown in cancer cells induced DNA double strand breaks and cell death
[36]. Anti-proliferation activity of compound 13 was determined also in Colo-205 cells.
Treatment of Colo-205 cells with compound 13 led to reduction of the cell proliferation with an
IC50 value of 0.17 µM, which was consistent with the results of apoptosis marker studies. On the
other hand, weaker effect in HEL299, derived from a human embryo lung tissue, was observed
(IC50 = 1.2 µM), suggesting that compound 13 has good therapeutic window.
Figure 5. Effects of Cdc7 inhibition by compound 13 in Colo-205 cells. a) Western blot analysis
of extracts prepared from Colo-205 cells at 48 h after treatment with DMSO control (0 µM) or

ACCEPTED MANUSCRIPT
compound 13 at indicated concentrations. b) Flow chemistry analysis of Colo-205 cells treated
for 48 h with DMSO control (0 µM) and the indicated concentration of compound 13. DNA
content was measured by FACS after propidium iodide staining.
Next, we investigated the recovery of substrate phosphorylation in cells after inhibitor washout
to see whether the slow-off rate of inhibitor affects the recovery of phosphorylation. Colo-205
cells were treated for 48 hours with 13, and then, the inhibitor was washed out with compound-
free media. The cells were lysed at the indicated time, and the phosphorylation status of MCM2
was analyzed by western blotting. As shown in Figure 6, the phosphorylation of MCM2 in 13-
treated cells was inhibited even 24 hours after washout. These results indicate that the slow off-
rate observed in 13 may reflect the duration of inhibitor efficacy in vivo
Unfortunately, the half-life of compound 13 after intravenous administration in mice was only
0.38 h, which hampered further investigations of this compound in vivo
.
.
DMSO
13 (1 µM)
8 24
After washout (h)
pMCM2(Ser53)
0
2
4
8
24
0
2
4
MCM2
Figure 6. Recovery of MCM2 phosphorylation at Ser53 after treatment with 13. Colo-205 cells
were treated with or without compounds for 48 h. After inhibitor washout with compound-free
media, the cells were lysed at the indicated time and subjected to western blot analysis.

ACCEPTED MANUSCRIPT
2.7 Chemistry
The hit compound 3 and its thiophenone analog 4 were prepared by Knoevenagel condensation
reaction between 7-azaindole-3-aldehyde 34 and commercially available Knoevenagel donors
(35 or 36, respectively) using piperidine as catalyst (Scheme 1A). For the exploration of hinge
binders R1, the corresponding aldehydes were employed as Knoevenagel acceptors for the
condensation reaction with furanone 35 to yield the desired compounds (5 - 7) as shown in
Scheme 1B.
Scheme 1a
O
O
CHO
(a)
CO2Et
CO2Et
(A)
+
X
X
N
N
N
N
H
HN
HN
H
34
35 (X=O)
36 (X=S)
3
4
(X=O)
(X=S)
O
O
(a)
CO2Et
CO2Et
R1
R1-CHO
(B)
+
O
O
HN
35
HN
Cl
O
R1 =
O
HO
N
N
H
5
6
7
a Reagents and conditions: (a) piperidine, EtOH, reflux (yields: 48% for 3, 93% for 4, 0.4 – 5%
for 5 – 7).
The general procedure for the replacement of R2 by various groups is shown in Scheme 2.
Methyl ester 8 was obtained analogously to ethyl ester 3 from the furanone intermediate 39

ACCEPTED MANUSCRIPT
which was prepared from methyl 4-chloroacetoacetate 37 and phenyl isocyanate 38 (Scheme
2A). Compound 9 (Scheme 2B) was obtained by hydrolysis of compound 3 under extremely
strong basic condition (50% w/v KOH aq., reflux). Surprisingly the ethyl ester in 3 was very
stable even under conventional ester hydrolysis conditions such as acidic (2N HCl aq./EtOH,
60ºC) or basic (2N NaOH aq./EtOH, 60ºC) conditions. The amide analogs (10 - 12) were
prepared directly by aminolysis of ethyl ester 3 with appropriate amines (Scheme 2C).
Scheme 2a
O
O
(A)
O
O
(a)
(b)
CO2Me
CO2Me
+
OCN
Cl
O
O
N
O
HN
HN
N
H
37
38
39
8
O
CO2Et
(C)
(B)
O
N
N
H
HN
(d)
(c)
O
O
O
3
Amines
(R2H)
R2
COOH
O
N
O
N
N
HN
N
HN
H
H
2
9
10
11
R = NH2 ( ), NHEt ( ),
12
1-piperidinyl (
)
a Reagents and conditions: (a) Et3N, petroleum ether-EtOAc (10:1), RT; (b) 7-azaindole-3-
aldehyde, piperidine, -PrOH, reflux (yield: 43% in two steps); (c) 50% w/v KOH aq., EtOH,
i
reflux (yield: 37%) ; (d) amines, EtOH, microwave or heating, 80 - 150ºC (yields: 9– 43%).
The modification of the R3 moiety was performed by two methods (method A and B).
Commercially available isocyanate was treated with ethyl 4-chloroacetoacetate 40 to furnish
furanone intermediate 43 in high yield (method A). Alternatively, furanone 43 was obtained by
one-pot methodology from diethyl malonate 41 (method B). Namely, diethyl malonate 41 was

ACCEPTED MANUSCRIPT
treated with sodium hydride, and allowed to react with chloroacetyl chloride 42, followed by
treating with an appropriate amine to give the corresponding furanone intermediate 43 in one-pot
reaction. Completion of the synthesis was achieved as described in Scheme 1A to afford the
desired products (13–33). In some cases, acidic conditions were employed in Knoevenagel
condensation to achieve a faster reaction, and the product was isolated as the hydrochloride salt
(32 and 33).
The furanone analogs synthesized in this report were isolated as a single isomer by NMR
spectra unless otherwise noted in the Experimental Section. Based on the
Knoevenagel condensation reaction [37] and the modeling study discussed above, we expected
Z-form preference in
that compounds synthesized here would have
or ) of them have not been determined.
Z
-forms, although the absolute stereochemistries
(E
Z
Scheme 3a
(via method A)
R3 =
2-chlorophenyl (13), 3-chlorophenyl (14),
4-chlorophenyl (15), 2-methoxyphenyl (16),
3-methoxyphenyl (17), 4-methoxyphenyl (18),
2,4-difluorophenyl (23), benzyl (32)
O
O
R3-NCO
+
method A
(a)
Cl
O
40
O
O
(c)
CO2Et
R3
CO2Et
R3
O
O
N
HN
N
HN
(b)
H
43
(via method B)
R3 =
O
CO2Et
method B
R3-NH2
+
+
Cl
2-fluorophenyl (19), 4-fluorophenyl (20),
2-methylphenyl (21), 4-methylphenyl (22),
2,4-dimethylphenyl (24), 2-pyridyl (25),
3-pyridyl (26), 4-pyridyl (27),
Cl
CO2Et
42
41
5-pyrimidinyl (28), 3-pyrazolyl (29),
30
33
)
31
6-quinolinyl ( ), 6-indazoyl ( ),
phenethyl (

ACCEPTED MANUSCRIPT
a
Reagents and conditions: (a) Et3N, ether-EtOAc (10:1), RT (yields: 30 - 80%); (b) 1)
diethylmalonate, NaH, THF, reflux ; 2) chloroacetyl chloride, 10ºC and then 40 - 50ºC, 3)
R3NH2, 10ºC and then RT (yields: 2.0 - 33%); (c) 7-azaindole-3-aldehyde, piperidine, EtOH,
reflux (13 – 31, yields: 2.4 – 74%) or 7-azaindole-3-aldehyde, 2M HCl in EtOH, reflux (32 and
33 as HCl salts, yields: 56 and 32% respectively).
3. Conclusion
We have discovered a novel furanone-based Cdc7 inhibitor through the HTS campaign in the
presence of high concentration of ATP. Subsequent optimization efforts led to the ortho-chloro
benzene analog 13, which exhibited strong inhibitory potency for Cdc7 with improved metabolic
stabilities in human and mouse liver microsomes. The detailed kinetic studies of compound 13
elucidated the slow-on/off rate characteristics that may offer advantages over known Cdc7
inhibitors in its potential to cause a more lasting inhibitory effect on Cdc7 in vivo. In fact, the
slow dissociation rate of compound 13 is reflected in delayed recovery of MCM2
phosphorylation in drug-treated cells after washout of the compound (data not shown).
Furthermore, compound 13 showed extremely potent inhibitory activity for Cdc7 even in the
presence of 1 mM ATP, which is close to physiological environment. Generally, an ATP
competitive inhibitor easily loses its inhibitory effect under such conditions. Selective inhibition
of Cdc7 by compound 13 in cancer cells led to an increase of -H2AX, indicative of DNA
double-strand breaks, and the dramatic increase of cleaved PARP as well as increased sub-G1
population, consistent with the induction of apoptosis. Our further effort to optimize the PK
profiles of this series of the compounds will be reported in due course.
4. Experimental section
4.1 Chemistry

ACCEPTED MANUSCRIPT
General: Reagents and solvents were purchased from commercial sources and used without
further purification. All reactions involving air- or moisture-sensitive reagents were performed
under a nitrogen atmosphere unless otherwise noted. Microwave reactions were run in a Biotage
Initiator set to normal power at the indicated temperature and were performed in sealed
microwave reaction vessels. Silica gel chromatography refers to the use of an automated medium
pressure liquid chromatography system (Teledyne ISCO or Yamazen Corp.) using prepacked
silica gel cartridges with UV detection at 254 nm. Preparative reverse-phase HPLC (prep-HPLC)
was performed on a Waters Autopurification system (dual triggered by target mass and UV 254
nm) using Imtakt Unison US-C18, 5 µm, 50 mm × 20 mm I.D. column, eluting with a binary
solvent system A and B using a gradient elution (A, 10 mM formic acid aq.; B, 10 mM formic
acid in MeOH). All yields reported are isolated yield after removal of residual solvents. The
purity of a purified compound was determined using Shimazu Prominence HPLC system by UV
detection (215 nm) with collecting MS spectra (100 - 800 m/z scan) of the target peak. The
separation method is shown as following. Column: Imtakt Cadenza, 3 µm, 50 mm × 2.0 mm I.D.
Mobile phase: acetonitrile in water (10 mM formic acid) from 10% to 90% with total 5 min run.
1
Flow rate: 0.5 mL/min. H -NMR spectra were recorded on Burker Ultra Shield™ 400 Plus;
chemical shifts ( ) are reported relative to a signal of TMS. Data format of NMR spectra is as
follows: chemical shift ( ppm), multiplicity (s = singlet, br. s. = broad singlet, d = doublet, t =
triplet, q = quartet, dd = doublet of doublets, dt = doublet of triplets, m = multiplet or
overlapping), coupling constant (Hz), and integration. The synthesized compounds for biological
assay have over 95% purity, otherwise noted in each synthesis experimental below.

ACCEPTED MANUSCRIPT
4.1.1 General Procedure of Knoegenavel Reactions (GP-I): Ethyl 5-((1H-pyrrolo[2,3-
b]pyridin-3-yl)methylene)-4-oxo-2-(phenylamino)-4,5-dihydrofuran-3-carboxylate (3).
To a solution of 7-azaindole-3-carboxaldehyde (34, 0.10 g, 0.70 mmol) and ethyl 4-oxo-2-
(phenylamino)-4,5-dihydrofuran-3-carboxylate (35, 0.18 g, 0.70 mmol) in ethanol (3.0 mL),
piperidine (0.083 mL, 0.84 mmol) was added at ambient temperature. The mixture was heated at
reflux for 12 h. After cooling to ambient temperature, the precipitate was collected by filtration,
washed successively with cold ethanol and isopropyl ether, and then dried to afford 3 (0.13 g,
1
48%) as a colorless solid. H NMR (400 MHz, DMSO-d6) 12.30 (br. s., 1H), 10.58 (s, 1H),
8.23 (dd,
J
= 1.3, 4.6 Hz, 1H), 7.98 (d,
J
= 7.3 Hz, 1H), 7.79 (d,
J
= 2.8 Hz, 1H), 7.52 - 7.64 (m,
4H), 7.43 - 7.51 (m, 1H), 6.80 - 6.99 (m, 2H), 4.28 (q,
13C NMR (101 MHz, DMSO-d6)
J
= 7.0 Hz, 2H), 1.30 (t, = 7.0 Hz, 3H).
J
176.18, 171.32, 163.98, 149.20, 144.20, 142.90, 135.62,
131.12, 129.66, 128.82, 127.74, 126.14, 118.37, 117.01, 106.95, 103.04, 88.31, 59.64, 40.41,
40.20, 39.99, 39.78, 39.58, 15.03. MS m/z 376 [M + H]+.
4.1.2
Ethyl
5-((1H-pyrrolo[2,3-b]pyridin-3-yl)methylene)-4-oxo-2-(phenylamino)-4,5-
dihydrothiophene-3-carboxylate (4)
Compound 4 was synthesized from 7-azaindole-3-carboxaldehyde (34) and ethyl 4-oxo-2-
(phenylamino)-4,5-dihydrothiophene-3-carboxylate (36) utilizing similar reaction conditions as
described in GP-I as an yellow solid (257 mg, 93%). 1H NMR (400 MHz, DMSO-d6) 12.44 (s,
1H), 11.27 (s, 1H), 8.34 (dd,
J
= 4.7, 1.5 Hz, 1H), 8.28 (dd,
J
= 8.1, 1.6 Hz, 1H), 7.89 (d,
J
= 0.7
Hz, 1H), 7.72 (d,
Hz, 1H), 4.29 (q,
J
J
= 2.8 Hz, 1H), 7.60 - 7.50 (m, 4H), 7.48 - 7.39 (m, 1H), 7.22 (dd,
= 7.1 Hz, 3H). MS m/z 392 [M + H]+.
J = 7.9, 4.7
= 7.1 Hz, 2H), 1.31 (t,
J

ACCEPTED MANUSCRIPT
4.1.3 Ethyl 5-((5-chloro-1H-pyrrolo[2,3-b]pyridin-3-yl)methylene)-4-oxo-2-(phenylamino)-
4,5-dihydrofuran-3-carboxylate (5)
Step 1. To a solution of 5-chloro-7-azaindole (0.50 g, 3.3 mmol) in acetic acid (5.0 mL),
hexamethylenetetramine (0.69 g, 4.9 mmol) was added at ambient temperature. The mixture was
heated at reflux for 8 h. After cooling to ambient temperature, the reaction mixture was diluted
with water, extracted with ethyl acetate twice. The organic layer was washed successively with
water and brine, dried over sodium sulfate and concentrated. The residue was purified by
chromatography on silica gel (hexane/ethyl acetate) to afford 5-chloro-7-azaindole-3-
carboxyaldehyde (0.13 g, 22%) as a colorless solid. 1H NMR (400 MHz, DMSO-d6) 10.01 (s,
1H), 9.77 (br. s, 1H), 8.61 (d, J = 2.0 Hz, 1H), 8.37 (d, J = 2.1 Hz, 1H), 7.99 (d, J = 2.6 Hz, 1H).
MS m/z 181 [M + H]+.
Step 2. Compound 5 was synthesized from 5-chloro-7-azaindole-3-carboxaldehyde prepared in
Step 1 and ethyl 4-oxo-2-(phenylamino)-4,5-dihydrofuran-3-carboxylate (35) as an yellow solid
(2.0 mg, 4.6%), utilizing similar reaction conditions as described in GP-I with prep-HPLC
purification. 1H NMR (400 MHz, DMSO-d6) 12.52 (br. s., 1H), 10.58 (s, 1H), 8.42 (d,
Hz, 1H), 8.26 (d, = 2.3 Hz, 1H), 7.71 (d, = 2.5 Hz, 1H), 7.55 - 7.63 (m, 2H), 7.51 (t,
J
J
= 2.3
= 7.8
J
J
Hz, 2H), 7.35 - 7.44 (m, 1H), 7.00 (s, 1H), 4.28 (q, J = 7.0 Hz, 2H), 1.30 (t, J = 7.2 Hz, 3H). MS
m/z 410 [M + H]+.
4.1.4 Ethyl 5-(4-hydroxy-3-methylbenzylidene)-4-oxo-2-(phenylamino)-4,5-dihydrofuran-3-
carboxylate (6)
Compound 6 was synthesized from 4-hydroxy-3-methylbenzaldehyde and ethyl 4-oxo-2-
(phenylamino)-4,5-dihydrofuran-3-carboxylate (35) as a brown solid (2.9 mg, 4.8%, purity

ACCEPTED MANUSCRIPT
91%), utilizing similar reaction conditions as described in GP-I with prep-HPLC purification. 1H
NMR (400 MHz, DMSO-d6) 9.88 (s, 1H), 7.48 - 7.59 (m, 4H), 7.42 (d,
J = 2.2 Hz, 1H), 7.34 -
7.40 (m, 1H), 7.27 (dd,
J
= 8.5, 2.3 Hz, 1H), 6.73 (d,
J
J
= 8.3 Hz, 1H), 6.43 (s, 1H), 4.25 (q, J =
7.1 Hz, 2H), 2.98 - 3.04 (m, 1H), 2.00 (s, 3H), 1.28 (t,
= 7.1 Hz, 3H). MS m/z 366 [M + H]+.
4.1.5 Ethyl 5-(benzo[d][1,3]dioxol-5-ylmethylene)-4-oxo-2-(phenylamino)-4,5-dihydrofuran-3-
carboxylate (7)
Compound 7 was synthesized from piperonal and ethyl 4-oxo-2-(phenylamino)-4,5-
dihydrofuran-3-carboxylate (35) as a colorless solid (0.25 mg, 0.44%), utilizing similar reaction
conditions as described in GP-I with prep-HPLC purification. 1H NMR (400 MHz, DMSO-d6)
10.65 (s, 1H), 7.53 - 7.59 (m, 2H), 7.45 - 7.52 (m, 2H), 7.31 - 7.41 (m, 1H), 7.16 - 7.22 (m, 2H),
6.91 (d,
J = 8.0 Hz, 1H), 6.48 - 6.58 (m, 1H), 6.05 (s, 2H), 4.25 (q, J = 7.1 Hz, 2H), 1.28 (t, J =
7.0 Hz, 3H). MS m/z 380 [M + H]+.
4.1.6
Methyl
5-((1H-pyrrolo[2,3-b]pyridin-3-yl)methylene)-4-oxo-2-(phenylamino)-4,5-
dihydrofuran-3-carboxylate (8)
Step 1. Methyl 4-oxo-2-(phenylamino)-4,5-dihydrofuran-3-carboxylate (39). To a mixed
solution of methyl 4-chloroacetoacetate (3.7 mL, 30 mmol) and phenyl isocyanate (3.6 g, 33
mmol) in petroleum ether/ethyl acetate (50 mL/5.0 mL) that was cooled in an ice bath,
triethylamine (4.8 g, 34 mmol) was added dropwise. The reaction mixture was stirred at ambient
temperature for 1 h. The reaction suspension was diluted with water, 2N hydrochloric acid
solution, and ether, and the solid was collected by filtration and washed successively with water
and diethyl ether, and then dried to afford the titled compound (6.3 g, crude) as a colorless solid.
MS m/z 234 [M + H]+.

ACCEPTED MANUSCRIPT
Step 2. Compound 8 was synthesized from 7-azaindole-3-carboxaldehyde and methyl 4-oxo-2-
(phenylamino)-4,5-dihydrofuran-3-carboxylate (39) prepared in Step 1 utilizing similar reaction
conditions as described in GP-I as a colorless solid (31 mg, 43%). 1H NMR (400 MHz, DMSO-
d6) 12.29 (br. s., 1H), 10.59 (s, 1H), 8.23 (dd,
J
= 1.4, 4.6 Hz, 1H), 7.98 (d,
J
= 7.5 Hz, 1H),
7.77 (d,
7.9 Hz, 1H), 3.77 (s, 3H). MS m/z 362 [M + H]+.
J = 2.5 Hz, 1H), 7.52 - 7.64 (m, 4H), 7.44 - 7.51 (m, 1H), 6.93 (s, 1H), 6.89 (dd, J = 4.6,
4.1.7 5-((1H-Pyrrolo[2,3-b]pyridin-3-yl)methylene)-4-oxo-2-(phenylamino)-4,5-dihydrofuran-
3-carboxylic acid (9)
To a solution of compound 3 (50 mg, 0.13 mmol) in ethanol (1.0 mL), 50% potassium
hydroxide solution (0.5 mL, 0.13 mmol) was added at ambient temperature. The mixture was
heated at reflux for 1 h. After cooling to ambient temperature, the precipitate was collected by
filtration, and washed with ethanol. The crude material was dissolved in water (0.5 mL) and
tetrahydrofuran (0.5 mL), and then 2M hydrochloric acid (0.023 mL, 0.045 mmol) was added
and the mixture was stirred for 30 min. The precipitate was collected by filtration, washed
successively with water and diethyl ether, and dried to afford compound 9 (0.012 g, 26%) as a
pale yellow solid. 1H NMR (400 MHz, DMSO-d6) 12.11 (br. s., 2H), 8.21 (dd,
1H), 8.06 (d, = 7.8 Hz, 1H), 7.73 (d, = 2.3 Hz, 1H), 7.38 - 7.49 (m, 2H), 7.32 (d,
= 4.6, 7.9 Hz, 1H), 6.66 (s, 1H). NH proton of the aniline is
J
= 1.3, 4.5 Hz,
= 7.5 Hz,
J
J
J
2H), 7.19 - 7.27 (m, 1H), 6.91 (dd,
J
missing. MS m/z 348 [M + H]+.
4.1.8 5-((1H-Pyrrolo[2,3-b]pyridin-3-yl)methylene)-4-oxo-2-(phenylamino)-4,5-dihydrofuran-
3-carboxamide (10)

ACCEPTED MANUSCRIPT
The suspension of compound 3 (20 mg, 0.053 mmol) in 7 M ammonia in MeOH (1 ml, 7.00
mmol) and THF (0.7 ml) was heated under MW irradiation at 80ºC for 15 min. The resulting
pale yellow solution was concentrated in vacuo. The residue was suspended in CHCl3-MeOH
and applied to silica gel chromatography (CHCl3/MeOH = 100/0 to 90/10, two peaks). The
fractions of an earlier peak were collected and concentrated in vacuo to afford compound 10 (1.7
1
mg, 9.2 %, purity 90%) as a colorless solid. H NMR (400 MHz, DMSO-d6)
δ
12.22 (s, 1H),
= 4.6, 1.6 Hz, 1H), 8.14 (s, 2H), 7.96 (s,
= 7.9, 4.7 Hz, 1H), 6.75 (d, = 0.6 Hz,
10.11 (s, 1H), 8.46 (dd, J = 8.0, 1.6 Hz, 1H), 8.30 (dd, J
1H), 7.47 - 7.55 (m, 2H), 7.35 - 7.46 (m, 3H), 7.19 (dd,
J
J
1H). MS m/z 347 [M + H]+.
4.1.9
5-((1H-Pyrrolo[2,3-b]pyridin-3-yl)methylene)-N-ethyl-4-oxo-2-(phenylamino)-4,5-
dihydrofuran-3-carboxamide (11)
The suspension of compound 3 (20 mg, 0.053 mmol) in 70% w/v ethylamine in H2O (50 l,
0.62 mmol) and THF (1.0 ml) was heated at 80ºC for 7 h. The reaction mixture was concentrated
in vacuo and purified by prep-HPLC to afford compound 11 (0.84 mg, 3.2 % yield, regioisomer
1
mixture = 3 : 1) as a colorless solid. H NMR (400 MHz, DMSO-d6)
δ 12.20 (br. s., 0.75H),
11.64 (br. s., 0.25H), 10.07 (br. s., 0.75H), 9.78 (br. s., 0.25H), 8.39 - 8.55 (m, 1H), 8.29 (dd,
J =
4.5, 1.3 Hz, 0.75H), 8.17 - 8.25 (m, 0.25H), 7.93 (s, 0.75H), 7.67 (s, 0.25H), 7.20 - 7.50 (m, 6H),
7.18 (dd,
J
= 7.9, 4.6 Hz, 0.25H), 7.08 (dd,
J
= 7.8, 4.8 Hz, 0.25H), 6.70 (br. s., 1H), 2.75 - 3.10
(m, 2H), 1.07 (t,
J
= 7.2 Hz, 3H). MS m/z 375 [M + H]+.
4.1.10
2-((1H-Pyrrolo[2,3-b]pyridin-3-yl)methylene)-5-(phenylamino)-4-(piperidine-1-
carbonyl)furan-3(2H)-one (12)