Identification of the 5,5-dioxo-7,8-dihydro-6H-thiopyrano[3,2-d]pyrimidine derivatives as highly selective PDE4B inhibitors

Article abstract of DOI:10.1016/j.bmcl.2013.12.076

A PDE4B subtype selective inhibitor is expected to have a wider therapeutic window than non-selective PDE4 inhibitors. In this Letter, two series of 7,8-dihydro-6H-thiopyrano[3,2-d]pyrimidine derivatives and 5,5-dioxo-7,8- dihydro-6H-thiopyrano[3,2-d]pyri

Full text of DOI:10.1016/j.bmcl.2013.12.076

Bioorganic & Medicinal Chemistry Letters 24 (2014) 893–899
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of the 5,5-dioxo-7,8-dihydro-6H-thiopyrano[3,2-d]
pyrimidine derivatives as highly selective PDE4B inhibitors
b
a
a
a
a
Taiji Goto a,d, , Akiko Shiina , Takeshi Murata , Masato Tomii , Takanori Yamazaki , Ken-ichi Yoshida ,
Toshiharu Yoshino a, Osamu Suzuki a, Yoshitaka Sogawa a, Kiyoshi Mizukami c, Nana Takagi a,
Tomomi Yoshitomi a, Maki Etori a, Hiroshi Tsuchida a, Tsuyoshi Mikkaichi a, Naoki Nakao a,
Mizuki Takahashi a, Hisashi Takahashi a, Shigeki Sasaki d
⇑
a R&D Division, Daiichi Sankyo Co., Ltd, 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
b Quality & Safety Management Division, Daiichi Sankyo Co., Ltd, 3-5-1 Nihonbashihoncho, Chuo-ku, Tokyo 103-8426, Japan
c Corporate Strategy Division, Daiichi Sankyo Co., Ltd, 3-5-1 Nihonbashihoncho, Chuo-ku, Tokyo 103-8426, Japan
d Graduate School of Pharmaceutical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A PDE4B subtype selective inhibitor is expected to have a wider therapeutic window than non-selective
PDE4 inhibitors. In this Letter, two series of 7,8-dihydro-6H-thiopyrano[3,2-d]pyrimidine derivatives and
5,5-dioxo-7,8-dihydro-6H-thiopyrano[3,2-d]pyrimidine derivatives were evaluated for their PDE4B sub-
type selectivity using human PDE4B2 and PDE4D2 full length enzymes. To improve their PDE4B selectiv-
ity over PDE4D, we optimized the substituents on the pyrimidine ring and the side chain phenyl ring,
resulting in several derivatives with more than 100-fold selectivity for PDE4B. Consequently, we identi-
fied 2-(3-chloro-4-methoxy-phenyl)-5,5-dioxo-7,8-dihydro-6H-thiopyrano[3,2-d]pyrimidine derivative
54 as a highly selective PDE4B inhibitor, which had potent hPDE4B inhibitory activity with an IC50 value
of 3.0 nM and 433-fold PDE4B selectivity over PDE4D.
Received 10 September 2013
Revised 10 December 2013
Accepted 19 December 2013
Available online 25 December 2013
Keywords:
PDE4
PDE4B
PDE4D
Ó 2014 Elsevier Ltd. All rights reserved.
Selective inhibitor
Full length
COPD
Phosphodiesterase 4 (PDE4) is expressed in inflammatory and
immune cells and mediates a hydrolysis of cyclic adenosine mono-
phosphate (cAMP).1 Since an up-regulation of cAMP contributes to
a suppression of inflammatory responses, PDE4 has been estab-
lished as an attractive target for the treatment of inflammatory dis-
eases such as asthma, chronic obstructive pulmonary disease
(COPD), and so on.2 Since the 1980s, PDE4 inhibitors with a wide
variety of chemotypes have been developed as anti-inflammatory
agents, and Roflumilast3, a second generation PDE4 inhibitor, has
been approved for the treatment of severe COPD. Indeed, Roflumi-
last and the other PDE4 inhibitors demonstrated moderate efficacy
in COPD in the clinical studies, however, the maximum dose was
limited by the mechanism-associated side effects such as nausea,
emesis, and diarrhea.
evokes a desired anti-inflammatory effect.6 Consequently, a PDE4B
selective inhibitor is expected to have a wider therapeutic window
than non-selective PDE4 inhibitors, and several attempts to obtain
a selective PDE4B inhibitor have been reported.7 However, to our
knowledge, there are no compounds advancing into clinical study
as a selective PDE4B inhibitor.
Our challenge to acquire PDE4B selective inhibitors started with
a series of fused bicyclic sulfone derivatives that we previously
reported as novel PDE4 inhibitors.8 We initially evaluated the
selectivity for PDE4B over PDE4D of our representative compounds
(Fig. 1). Compound 1 exhibited good in vitro PDE4B inhibitory
activity (IC50 = 25 nM) and high selectivity (112 folds) over PDE4D
for recombinant human PDE4B (152–564) and PDE4D (78–508).
However, when full length human PDE4B2 and PDE4D2 were used
PDE4 isozymes are coded by four distinct genes to give four iso-
forms (PDE4A-D), and 22 splice variants of these four isoforms are
known.4 In the knockout studies, it is implied that PDE4D inhibi-
tion promotes undesired adverse effects5 and PDE4B inhibition
for the assay, the PDE4B inhibitory activity of 1 decreased
(IC50 = 150 nM) and the PDE4B selectivity dropped to 3.7 folds.
Similarly, sulfide 2 and amide 3 showed only moderate PDE4B
inhibition and low selectivity for the full length enzymes.
It is known that PDE4 enzymes contain unique regulatory do-
mains, called upstream conserved regions 1 and 2 (UCR1 and
UCR2), and the UCR2 or C-terminal regulatory domain can interact
with the catalytic domain of PDE4 enzymes to regulate enzyme
⇑
Corresponding author at: R&D Division, Daiichi Sankyo Co., Ltd, 1-2-58
Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan.
E-mail address: goto.taiji.kt@daiichisankyo.co.jp (T. Goto).
0960-894X/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.12.076

894
T. Goto et al. / Bioorg. Med. Chem. Lett. 24 (2014) 893–899
Table 1
O
S
S
O
N
In vitro inhibitory activity and selectivity of compounds 12–18
H
N
H
N
O
O
S
H
N
N
N
N
OH
OH
O
N
N
OH
R
1
2
hPDE4B (152-564) IC50 = 25 nM
hPDE4D (78-508) IC50 = 2800 nM
Selectivity (D/B) = 112
hPDE4B (152-564) IC50 = 29 nM
hPDE4D (78-508) IC50 = 1300 nM
Selectivity (D/B) = 45
No.
R
hPDE4B2 IC50 (nM) hPDE4D2 IC50 (nM) Selectivity (D/B)
F
F
12
880
57
3400
680
3.9
hPDE4B2 (full) IC50= 150 nM
hPDE4D2 (full) IC50= 550 nM
Selectivity (D/B) = 3.7
hPDE4B2 (full) IC50= 210 nM
hPDE4D2 (full) IC50= 1300 nM
Selectivity (D/B) = 6.2
13
14
12
O
S
O
N
H
N
120
120
30
1000
420
8.3
3.5
9.7
O
N
F
N
H
3
15
16
hPDE4B (152-564) IC50 = 22 nM
hPDE4D (78-508) IC50 = 260 nM
Selectivity (D/B) = 12
Cl
CN
O
290
hPDE4B2 (full) IC50= 70 nM
hPDE4D2 (full) IC50= 180 nM
Selectivity (D/B) = 2.6
17
18
11
280
25
Figure 1. In vitro profiles of compounds 1–3.
activity.9 Since the access of PDE4 inhibitors to the catalytic pocket
would be controlled by these helix-capped conformations, the
inhibitory activity and selectivity with full length enzymes were
different from those with truncated enzymes. We considered that
the in vitro result with full length enzymes would reflect an
120
1800
15
O
S
S
S
SH
Cl
S
S
OH
OH
Cl
a
b
d
c
e
O
O
O
O
O
+
N
N
N
N
N
N
O
O
O
O
O
O
S
OH
Cl
4
5
6
7
8
9
10
12 R = 2-fluorophenyl
19 R = 4-dimethylaminophenyl
S
S
R
H
N
H
N
13
14
15
16
17
18
3-fluorophenyl
4-fluorophenyl
4-chlorophenyl
4-cyanophenyl
20
21
22
23
4-acetylphenyl
f
g
h
5-(2-methoxypyridyl)
O
O
5-(2-dimethylaminopyridyl)
2,3-difluoro-4-methoxyphenyl
3-chloro-4-methoxyphenyl
N
N
N
N
O
OH
Cl
4-methoxyphenyl 24
11
3-methoxyphenyl
i
O
O
S
O
S
O
H
H
N
g
h
N
O
O
N
N
N
N
O
OH
R
Cl
25
26 R = 4-methoxyphenyl
33 R = 2-naphthyl
27
28
29
30
31
32
4-dimethylaminophenyl
4-acetylphenyl
34
35
36
37
4-biphenyl
2-(5-methoxynaphthyl)
2-(5-fluoronaphthyl)
2-benzothiophenyl
2-benzofuranyl
5-(2-methoxypyridyl)
5-(2-dimethylaminopyridyl)
2,3-difluoro-4-methoxyphenyl 38
3-chloro-4-methoxyphenyl 39
3-fluoro-4-methoxyphenyl
Scheme 1. Reagents and conditions: (a) NaOMe, NaI, MeOH, 65 °C; (b) NaOMe, toluene, 110 °C (98% from 5); (c) 2-methylisothiourea sulfate, KOH, MeOH; (d) AcOH, H2O,
110 °C (30% from 7); (e) POCl3, N,N-dimethylaniline, 90 °C (82%); (f) ethyl 2-(4-aminophenyl)acetate, DIPEA, EtOH, reflux (93%); (g) arylboronic acid or arylboronic acid
pinacol ester, PdCl2dppfÁCH2Cl2, Na2CO3, DME, H2O, 130 °C, microwave; (h) 1 N NaOH, MeOH, THF, 60 °C; (i) mCPBA, CH2Cl2 (79%).

T. Goto et al. / Bioorg. Med. Chem. Lett. 24 (2014) 893–899
895
in vivo response better than that with truncated enzymes. There-
fore, we used full length human PDE4B2 and PDE4D2 enzymes
for the in vitro assay to explore PDE4B subtype selective inhibitors
in this study.
derivatives (12–24). On the other hand, sulfide 11 was oxidized
with mCPBA to furnish sulfone intermediate 25, which was con-
verted to the sulfone derivatives (26–39) by the same procedure
used to obtain the aforementioned sulfide derivatives.
The synthetic study started with sulfide 2 as the lead com-
pound, because the PDE4B selectivity of 2 was superior to that of
3, even though the cell-based activity of 2 was lower than that of
3. Additionally, sulfone compounds which were supposed to be
more stable for metabolism than sulfide compounds were also
synthesized.
The target compounds (12–24, 26–39) were synthesized
according to the procedure outlined in Scheme 1. The initial reac-
tion of methyl 4-chlorobutanoate (4) with methyl thioglycolate (5)
gave diester 6, and a successive Dieckmann condensation reaction
provided ketoester 7. The pyrimidine ring was constructed by a
cyclization reaction of ketoester 7 with S-methylisothiourea and
the resultant 8 was treated with acetic acid to provide 2,4-dihy-
droxypyrimidine 9. After conversion of the hydroxyl groups to
chlorine atoms using phosphoryl chloride, ethyl 2-(4-aminophe-
nyl)acetate was introduced selectively to the 4-position of the
pyrimidine ring to afford 11. Introduction of substituents at the
2-position of 11 were performed by the Suzuki–Miyaura coupling
reaction with commercially available or prepared boron reagents,
and the hydrolysis of the ester gave the target sulfide ring
Initially, modification of the phenyl group at the 2-position of
the pyrimidine ring was conducted to assess the influence of sub-
stituents on selectivity (Table 1). In the case of fluorine substituted
derivatives (12–14), the 3-substitution (13) was favorable for both
PDE4B inhibition and selectivity. Replacement of the 4-fluorine
atom with a chlorine atom lowered the selectivity (15). On the
other hand, 16 with the 4-cyano group showed an increased inhib-
itory activity for PDE4B with 9.7-fold selectivity. Subsequently, the
4-methoxy substitution exhibited further improvement of PDE4B
inhibition with 25-fold selectivity (17). However, the 3-methoxy
substitution (18) weakened the PDE4B inhibitory activity com-
pared to the 4-methoxy substitution.
Considering the result that 4-methoxy derivative 17 was the
most potent inhibitor for PDE4B with the highest PDE4B selectivity
over PDE4D among these derivatives (12–18), we assumed that a
proton acceptor at the 4-position would be favorable for selective
inhibition of PDE4B. Therefore, we next explored derivatives
having proton acceptors at the 4-position. Table 2 summarizes
the results with the sulfide derivatives (17, 19–24) and the corre-
sponding sulfone derivatives (26–32).
Table 2
In vitro inhibitory activity and selectivity of compounds 17, 19–24, and 26–32
O
S
S
H
O
N
H
N
O
O
N
N
OH
N
N
OH
R
R
R
No.
hPDE4B2 IC50 (nM)
hPDE4D2 IC50 (nM)
280
Selectivity (D/B)
25
No.
hPDE4B2 IC50 (nM)
hPDE4D2 IC50 (nM)
1500
Selectivity (D/B)
34
17
11
26
27
44
O
19
20
38
34
790
21
29
38
10
2100
860
55
86
N
1000
28
O
N
21
22
140
140
7400
1600
53
11
29
30
250
56
3800
1200
15
21
O
N
N
F
F
23
24
140
71
2200
430
16
31
32
46
13
3400
1300
74
O
6.1
100
Cl
O

896
T. Goto et al. / Bioorg. Med. Chem. Lett. 24 (2014) 893–899
Table 3
Dimethylamino and acetyl derivatives (19, 20) showed good
PDE4B selectivity equal to that of 17, although their PDE4B inhib-
itory activity was slightly decreased. Exchanging the benzene ring
of 17 with a pyridine ring gave 5-(2-methoxy)pyridyl derivative 21
which exhibited higher selectivity (53 folds) than 17, whereas
5-(2-dimethylamino)pyridyl derivative 22 exhibited lower selec-
tivity (11 folds) than the corresponding benzene derivative 19.
Multiple substitution of the 4-methoxyphenyl moiety of 17 with
halogen atoms did not cause improvement of both PDE4B inhibi-
tory activity and selectivity (23, 24).
In vitro inhibitory activity and selectivity of compounds 33–38
O
S
O
H
N
O
N
N
OH
R
No.
R
hPDE4B2 IC50 (nM) hPDE4D2 IC50 (nM) Selectivity (D/B)
On the other hand, the sulfone derivative 26 which corresponds
to an oxidized compound of 17 had higher selectivity (34 folds)
than 17, although the PDE4B inhibitory activity was decreased.
Moreover, the dimethylamino derivative 27 showed 55-fold selec-
tivity and the acetyl derivative 28 displayed 86-fold selectivity
33
19
38
520
790
27
21
34
Table 4
In vitro inhibitory activity and selectivity of compounds 39, 44–49
O
S
O
N
R2
35
36
30
14
290
440
9.7
H
N
R3
O
O
F
N
OH
R1
31
F
O
No. R1 R2 R3 hPDE4B2 IC50 (nM) hPDE4D2 IC50 (nM) Selectivity (D/B)
S
39
44
45
46
H
F
F
H
H
F
H
H
H
H
H
F
10
4.6
16
5.2
64
49
9.9
1000
620
1300
580
100
135
81
112
20
37
38
8.3
400
48
55
F
47 Cl
H
H
H
1300
600
560
O
48 Cl Cl
49 Cl
12
57
20
1100
F
O
S
O
S
O
R2
R1
S
S
O
H
N
R3
H2N
F
NH
OH
OTf
OTf
O
S
a
N
N
b
N
N
c
f
d or e
N
N
N
N
OH
O
+
O
O
O
F
F
F
F
7
40
O
O
O
O
44 R1 = F, R2, R3 = H
45 R1, R2 = F, R3 = H
46 R1, R3 = F, R2 = H
47 R1 = Cl, R2, R3 = H
48 R1, R2 = Cl, R3 = H
41
42
43
49 R1 = F, R2 = Cl, R3 = H
O
S
O
S
S
S
H
O
O
N
H
N
H
N
h
f
Cl
g
N
F
i
F
F
O
O
O
N
N
N
N
N
N
N
O
O
OH
Cl
10
Cl
Cl
R
50
51
52 R = 4-methoxyphenyl
53
54
55
56
2,3-difluoro-4-methoxyphenyl
3-chloro-4-methoxyphenyl
2-naphthyl
2-benzothiophenyl
Scheme 2. Reagents and conditions: (a) NaOMe, MeOH (81%); (b) Tf2O, triethylamine, DMAP, CH2Cl2 (34%); (c) mCPBA, CH2Cl2 (79%); (d) substituted (4-aminophenyl)acetate,
DMF, 60 °C; (e) substituted (4-aminophenyl)acetate, Pd2(dba)3, xantphos, Cs2CO3, dioxane, 100 °C; (f) 1 N NaOH, MeOH; (g) methyl (4-amino-2-fluorophenyl)acetate,
Pd2(dba)3, xantphos, Na2CO3, dioxane, 70 °C (74%); (h) mCPBA, CH2Cl2 (98%); (i) arylboronic acid or arylboronic acid pinacol ester, PdCl2dppfÁCH2Cl2, Na2CO3, DME, H2O,
130 °C, microwave.

T. Goto et al. / Bioorg. Med. Chem. Lett. 24 (2014) 893–899
897
Table 5
Structure and profile of 44
O
O
S
H
N
F
O
N
N
OH
F
O
hPDE4B2
IC50 (nM)
hPDE4D2
IC50 (nM)
Selectivity
(D/B)
mTNF-
IC50 (nM)
a
LogD
(pH 7.4)
PAMPA Pe*(10À6 cm/sec)
(pH 7.4)
Mouse PK (10 mg/kg, p.o.)
Cmax
AUC
g h/mL)
T1/2(h)
(
l
(l
g/mL)
4.6
620
135
18
0.8
14.3
2.77
1.70
2.0
*
Permeability coefficient in parallel artificial membrane permeability assay.
Table 6
moderate selectivity (27 folds), while biphenyl derivative 34
showed decreased inhibitory activity with 21-fold selectivity. From
the previous results that the introduction of a methoxy group and a
fluorine atom improved selectivity, a methoxy group or a fluorine
atom was introduced to the naphthyl group (35, 36). Unexpectedly,
35 resulted in reduced activity and selectivity, and 36 showed sim-
ilar PDE4B inhibitory activity and selectivity to those of 33. On the
other hand, the benzothiophene and benzofuran groups (37, 38)
were preferable to the naphthyl group for PDE4B inhibition and
selectivity.
In vitro inhibitory activity and selectivity of compounds 52–56
O
S
O
N
H
N
F
O
N
OH
R
No.
R
hPDE4B2
IC50 (nM)
hPDE4D2
IC50 (nM)
Selectivity
(D/B)
mTNF-a
IC50 (nM)
We next focused on substituents for the phenyl group of the
side chain. The derivatives (44–49, 52–56) were synthesized as
shown in Scheme 2. Intermediate 7 was condensed with 3-flu-
oro-4-methoxybenzamidine 40 to form pyrimidine 41. The hydro-
xyl group of 41 was converted to the corresponding triflate (42),
and the sulfur atom was oxidized by mCPBA to give sulfone 43.
The side chains were introduced to 43 by a substitution reaction
or palladium catalyzed cross-coupling reaction with substituted
4-aminophenylacetate. Hydrolysis of the terminal ester provided
the target compounds (44–49). On the other hand, 52–56 were
synthesized from intermediate 10. Methyl 4-amino-2-fluorophe-
nylacetate was selectively introduced at the 4-position of pyrimi-
dine under a cross-coupling condition and subsequent oxidation
of the sulfur atom gave 2-chloropyrimidine intermediate 51. Sev-
eral aromatic rings were introduced at the 2-position by the Suzu-
ki–Miyaura coupling reaction and deprotection of the ester group
furnished the target compounds (52–56).
Table 4 summarizes the results of the derivatives with fluorine
and chlorine atoms on the phenyl group of the side chain. 2-Fluoro
derivative 44 and 2,6-difluoro derivative 46 showed higher PDE4B
inhibitory activity (IC50 = 4.6 and 5.2 nM) and selectivity (135 and
112 folds) than unsubstituted derivative 39, whereas 2,5-disubsti-
tuted derivative 45 failed to improve the selectivity. 2-Chloro
derivative 47 and 2,5-dichloro derivative 48 revealed to have only
low PDE4B selectivity (20 and 12 folds) with moderate PDE4B
inhibitory activity. Replacement of the 2-chlorine atom of 48 with
a fluorine atom showed a recovery of the selectivity to 57 folds
(49), but this selectivity was lower than that of the other fluoro
derivatives (44–46). From these results, the mono fluorine substi-
tuted side chain was considered to be appropriate for expressing
high PDE4B selectivity.
52
7.5
1300
173
171
433
137
158
96
O
F
F
53
54
55
56
8.2
3.0
5.9
3.8
1400
1300
810
131
11
O
Cl
O
44
S
600
11
with strong PDE4B inhibitory activity (IC50 = 10 nM). However, the
pyridine derivatives (29, 30) failed to improve the selectivity (15
and 21 folds) compared to the corresponding benzene compounds
(26, 27). Introduction of the halogen atoms to the 4-methoxy-
phenyl moiety of 26 had positive effects on the selectivity for the
sulfone derivatives (31, 32). Compounds 31 and 32 showed excel-
lent PDE4B selectivity (74 and 100 folds) retaining potent PDE4B
inhibitory activity in contrast to the decrease of potency and
selectivity for the corresponding sulfide derivatives (23, 24). These
results have indicated that the sulfone scaffold is appropriate for
selective PDE4B inhibition.
The in vitro profile and pharmacokinetic property of 44 are
shown in Table 5. Compound 44 showed high inhibitory activity
against lipopolysaccharide (LPS)-induced tumor necrosis factor al-
We further searched for a suitable substituent for the 2-position
of the pyrimidine core (Table 3). Naphthyl derivative 33 had good
PDE4B inhibitory activity with an IC50 value of 19 nM and
pha (TNF-a) production in mouse splenocytes with an IC50 value of
18 nM, and exhibited good exposure by oral administration at a
dose of 10 mg/kg in mice.

898
T. Goto et al. / Bioorg. Med. Chem. Lett. 24 (2014) 893–899
Whereas, the lack of this interaction with Phe196 of PDE4D might
cause the weakness of PDE4D inhibitory activity. These hypotheses
were supported by the result that amide derivatives of 39 showed
lower PDE4B inhibitory activity than 39.10 The 3-chloro-4-
methoxyphenyl moiety of 54, which is proximal to UCR2, may also
contribute to the stabilization of the capping conformation of
UCR2.
In summary, in the course of the exploration of novel PDE4B
selective inhibitors, it was revealed that in vitro PDE4B and PDE4D
inhibitory activities should be evaluated in the assay with full
length enzymes. Modification of the phenyl ring moiety at the
2-position of the pyrimidine ring in 7,8-dihydro-6H-thiopyr-
ano[3,2-d]pyrimidine derivative 2 was effective in improving the
PDE4B selectivity over PDE4D, which gave 4-methoxyphenyl
derivative 17 with 25-fold selectivity. Successive optimization of
the 2-position substituent, oxidation of the sulfur atom, and intro-
duction of the fluorine atom to the phenyl ring of the side chain
culminated in the identification of potent and highly selective
PDE4B inhibitor 54 with an IC50 value of 3.0 nM for hPDE4B2,
and 433-fold selectivity over hPDE4D2. Further investigation of
this series of potent and selective PDE4B inhibitors including an
in vivo model study and toxicity study has been reported in an-
other publication.11
Figure 2. Putative binding model of 54 to UCR2-PDE4B (PDB accession code: 3G45).
The compound model of 54 was generated according to our reported structure (PDB
accession code: 3W5E).
Since 44 showed a potential for an orally available agent in Ta-
ble 5, we next tried to replace the 3-fluoro-4-methoxyphenyl
group of 44 with other aromatic rings which were found to have
potent PDE4B inhibitory activity and high selectivity in Tables 1–
3, and the results are summarized in Table 6. The 4-methoxy
substituted derivatives (52–54) exhibited high PDE4B inhibitory
activity with single nano molar IC50 and had an excellent PDE4B
selectivity of more than 170 folds. Especially, 3-chloro-4-methoxy-
phenyl derivative 54 showed 433-fold selectivity, which was the
highest selectivity in all of the derivatives. Bicyclic derivatives
(55, 56) also showed single nano molar potency of PDE4B inhibi-
tion. However, 55 and 56 had lower selectivity (137 and 158 folds)
than the 4-methoxyphenyl derivatives (52–54).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.
12.076.
References
1. (a) Bender, A. T.; Beavo, J. A. Pharmacol. Rev. 2006, 58, 488; (b) Zhang, K. Y. J.;
Ibrahim, P. N.; Gillette, S.; Bollag, G. Exp. Opin. Ther. Targets 2005, 9, 1283; (c)
Torphy, T. J. Am. J. Respir. Crit. Care Med. 1998, 157, 351.
2. (a) Huang, Z.; Mancici, J. A. Curr. Med. Chem. 2006, 13, 3253; (b) Houslay, M.;
Schafer, P.; Zhang, K. Y. J. Drug Discovery Today 2005, 10, 1503; (c) Cazzola, M.;
Ciaprini, C.; Page, C. P.; Matera, M. G. Exp. Opin. Ther. Targets 2007, 11, 1273; (d)
Kodimuthali, A.; Jabaris, S. L.; Pal, M. J. Med. Chem. 2008, 51, 5471.
3. (a) Rabe, K. F. Br. J. Pharmacol. 2011, 163, 53; (b) Cazzola, M.; Picciolo, S.;
Matera, M. G. Exp. Opin. Pharm. 2010, 11, 441; (c) Giembycz, M. A.; Field, S. K.
Drug Des. Dev. Ther. 2010, 4, 147.
In addition, we evaluated inhibitory activity of each compound
against LPS-induced TNF-
a production in mouse splenocytes. In
this assay, the inhibitory activity correlated to the PDE4B2 inhibi-
tory activity of each compound. For example, 54 which exhibited
potent PDE4B2 inhibitory activity with an IC50 value of 3.0 nM also
4. (a) Houslay, M. D.; Baillie, G. S.; Maurice, D. H. Circ. Res. 2007, 100, 950; (b)
Houslay, M. D.; Adams, D. R. Biochem. J. 2003, 370, 1.
showed high inhibitory activity against TNF-
a production
5. (a) Robichaud, A.; Stamatiou, P. B.; Jin, S.-L. C.; Lachance, N.; MacDonald, D.;
Laliberte, F.; Liu, S.; Huang, Z.; Conti, M.; Chan, C.-C. J. Clin. Invest. 2002, 110,
1045; (b) Lehnart, S. E.; Wehrens, X. H. T.; Reiken, S.; Warrier, S.; Belevych, A.
E.; Hervey, R. D.; Richter, W.; Jin, S.-L. C.; Conti, M.; Marks, M. R. Cell 2005, 123,
25.
6. (a) Jin, S.-L. C.; Lan, L.; Zoudilova, M.; Conti, M. J. Immun. 2005, 175, 1523; (b)
Jin, S.-L. C.; Conti, M. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 7628; (c) Komatsu, K.
Lee, J.Y. Miyata, M. Lim, J.H. Jono, H. Koga, T. Xu, H. Yan, C. Kai, H. Li, J.D. Nat.
Commun. doi: 10.1038/ncomms2674.; (d) Manning, C. D.; Burman, M.;
Christensen, S. B.; Cieslinski, L. B.; Essayan, D. M.; Grous, M.; Torphy, T. J.;
Barnette, M. S. Br. J. Pharmacol. 1999, 128, 1393.
(IC50 = 11 nM). On the other hand, 53 which had the lowest
PDE4B2 inhibitory activity (IC50 = 8.2 nM) in Table 6 exhibited
the lowest inhibitory activity against TNF-
(IC50 = 131 nM). Of course, the PDE4D2 inhibitory activity also
would affect the inhibition of TNF- production. However, the
a
production
a
PDE4D2 inhibitory activity of these compounds was supposed to
be low enough to ignore. From these results, although the differ-
ence in inhibitory activity against TNF-
a production seemed to
7. (a) Donnell, A. F.; Dollings, P. J.; Butera, J. A.; Dietrich, A. J.; Lipinski, K. K.;
Ghavami, A.; Hirst, W. D. Bioorg. Med. Chem. Lett. 2010, 20, 2163; (b) Taltavull,
J.; Serrat, J.; Gracia, J.; Gavalda, A.; Andres, M.; Cordoba, M.; Miralpeix, M.;
Vilella, D.; Ryder, H.; Pages, L. J. Med. Chem. 2010, 53, 6912; (c) Naganuma, K.;
Omura, A.; Maekawara, M.; Saitoh, M.; Ohkawa, N.; Kubota, T.; Nagumo, H.;
Kodama, T.; Takemura, M.; Ohtsuka, T.; Nakamura, J.; Tsujita, R.; Kawasaki, K.;
Yokoi, H.; Kawanishi, M. Bioorg. Med. Chem. Lett. 2009, 19, 3174; (d) Kranz, M.;
Wall, M.; Evans, B.; Miah, A.; Ballantine, S.; Delves, C.; Dombroski, B.; Gross, J.;
Schneck, J.; Villa, J. P. Bioorg. Med. Chem. 2009, 17, 5336; (e) Odingo, J. O. Exp.
Opin. Ther. Pat. 2005, 15, 773.
8. Goto, T.; Shiina, A.; Yoshino, T.; Mizukami, K.; Hirahara, K.; Suzuki, O.; Sogawa,
Y.; Takahashi, T.; Mikkaichi, T.; Nakao, N.; Takahashi, M.; Hasegawa, M.; Sasaki,
S. Bioorg. Med. Chem. Lett. 2013, 23, 3325.
9. (a) Burgin, A. B.; Magnusson, O. T.; Singh, J.; Witte, P.; Staker, B. L.; Bjornsson, J.
M.; Thorsteinsdottir, M.; Hrafnsdottir, S.; Hagen, T.; Kiselyov, A. S.; Stewart, L.
J.; Gurney, M. E. Nat. Biotechnol. 2010, 28, 63; (b) Houslay, M. D.; Adams, D. R.
Nat. Biotechnol. 2010, 28, 38.
be somewhat drastic compared to that of PDE4B2 inhibition, we
concluded that the PDE4B inhibitory activity was a significant fac-
tor for the inhibition of TNF-a production and the selective PDE4B
inhibitor could be a potent anti-inflammatory agent.
In order to rationalize the high PDE4B selectivity of 54, we ana-
lyzed a binding model of 54 with PDE4B. Figure 2 shows a binding
model of 54 based on the crystal structure of amide compound
with truncated PDE4B (152–528)8 superimposed on the reported
crystal structure of PDE4B including UCR2.9 In this model, Tyr274
from the UCR2 helix extends into the active site and is positioned
between the substituent on 2-position of the pyrimidine ring and
the carboxylic acid moiety. The Tyr274 residue, which corresponds
to Phe196 in PDE4D, is considered to be the one of the key factors
to make the subtype selectivity. The hydroxyl group of Tyr274
seems to interact with carboxylic acid moiety of 54, which would
contribute to enhancement of the PDE4B inhibitory activity.
10. The amide compound A exhibited lower PDE4B inhibitory activity compare to
the corresponding carboxylic acid compound 39, which made the PDE4B
selectivity low (Figure 3). The change of the hydrogen bonding interaction with
Tyr274 and the lipophilicity of amide alkyl group might reduce the PDE4B

T. Goto et al. / Bioorg. Med. Chem. Lett. 24 (2014) 893–899
899
11. Suzuki, O.; Mizukami, K.; Etori, M.; Sogawa, Y.; Takagi, N.; Tsuchida, H.;
Morimoto, K.; Goto, T.; Yoshino, T.; Mikkaichi, T.; Hirahara, K.; Nakamura, S.;
Maeda, H. J. Pharmacol. Sci. 2013, 123, 219.
inhibitory activity of A. On the other hand, in the case of interaction with
PDE4D, a hydrophobic interaction of amide alkyl group with Phe196 was
considered to cause the gain of the PDE4D inhibitory activity.
O
O
S
O
S
O
H
N
H
N
O
O
N
N
N
N
OH
N
H
F
F
39
A
O
O
hPDE4B2(full) = 10
hPDE4D2(full) = 1000
D/B = 100
hPDE4B2(full) = 78
hPDE4D2(full) = 300
D/B = 3.8
Figure 3. Comparison of the in vitro activity and selectivity
between the carboxylic acid 39 and the amide A