Preparation and optimization of pyrazolo[1,5-a]pyrimidines as new potent PDE4 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2015.11.093

A new series of pyrazolo[1,5-a]pyrimidines exemplified by compound 1, has been identified with moderate activity (IC₅₀ = 165 nM), following GSK256066 rescaffolding. Compound 1 optimization at positions 2, 3, 6 and 7 gave compound 10 with high in vitro activity (IC₅₀ = 0.7 nM). Modeling studies based on the PDB structure 3GWT with compound 5 showed the expected overlay with the carboxamide, the aryl moiety and the sulfone. Cyclisation of the primary amide to the 5 position of the pyrazolo[1,5-a]pyrimidines scaffold afforded compounds 15 and 16 with 200-fold enhancement in activity and cellular potency.

Full text of DOI:10.1016/j.bmcl.2015.11.093

Bioorganic & Medicinal Chemistry Letters 26 (2016) 454–459
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Preparation and optimization of pyrazolo[1,5-a]pyrimidines as new
potent PDE4 inhibitors
Jacques Le Roux a, Caroline Leriche b, Philippe Chamiot-Clerc b, John Feutrill c, Frank Halley a,
,
⇑
David Papin a, Nathalie Derimay d, Christelle Mugler d, Claudine Grépin b, Laurent Schio a
a Sanofi Research Center, 13 Quai Jules Guesde, 94400 Vitry-sur-Seine, France
b Sanofi Ophthalmology, 17 Rue Moreau, 75012 Paris, France
c SYN|thesis Med Chem, 399 Royal Parade, Parkville, VIC 3052, Australia
d Sanofi Open innovation Access Plateform, 16 Rue d’Ankara, 67000 Strasbourg, France
a r t i c l e i n f o
a b s t r a c t
Article history:
A new series of pyrazolo[1,5-a]pyrimidines exemplified by compound 1, has been identified with mod-
erate activity (IC50 = 165 nM), following GSK256066 rescaffolding. Compound 1 optimization at positions
2, 3, 6 and 7 gave compound 10 with high in vitro activity (IC50 = 0.7 nM). Modeling studies based on the
PDB structure 3GWT with compound 5 showed the expected overlay with the carboxamide, the aryl moi-
ety and the sulfone. Cyclisation of the primary amide to the 5 position of the pyrazolo[1,5-a]pyrimidines
scaffold afforded compounds 15 and 16 with 200-fold enhancement in activity and cellular potency.
Ó 2015 Elsevier Ltd. All rights reserved.
Received 30 October 2015
Revised 24 November 2015
Accepted 25 November 2015
Available online 26 November 2015
Keywords:
Carboxamide
PDE4 inhibitor
cAMP
Phosphodiesterase-4 (PDE4) inhibitors prevent cyclic adenosine
monophosphate (cAMP) hydrolysis to adenosine monophosphate
(50AMP) which is involved in pro-inflammatory cytokines secre-
tion. PDE4 inhibitors have been extensively studied as anti-inflam-
matory drugs since the discovery of Rolipram in the early 1990s.1,2
Many compounds have been developed and evaluated in the clinic
for depression, asthma, chronic obstructive pulmonary disease
(COPD), dermatitis, and rheumatoid arthritis.3 The list of indica-
tions extended to multiple sclerosis (MS), Crohn’s disease,4 psoria-
sis, ulcerative colitis, rhinitis, cutaneous lupus and sarcoidosis.5,6
PDE4 inhibitors also found application in the central nervous sys-
tem (CNS) area as cognition enhancer in Alzheimer’s and Parkin-
son’s diseases, and anxiety.7 Recently PDE4 inhibitors have been
used in the potential treatment of the dry eye condition.8 Dry
eye is a multifactorial disease of the tears and ocular surface that
results in symptoms of discomfort, and visual disturbance. It is
accompanied by increased osmolarity of the tear film and inflam-
mation of the ocular surface.9 Elevated levels of inflammatory
cytokines play a prominent role in dry eye disease, contributing
to the dry eye symptoms and the disease process itself.9,10
Although two PDE4 inhibitors, Roflumilast and Apremilast have
been recently approved for COPD and psoriasis indications, respec-
tively, efforts to find new drug candidates with different scaffolds
and indications are still ongoing in the pharmaceutical industry
(see Fig. 1).
This Letter describes the discovery and the optimization of a
novel series of pyrazolo[1,5-a]pyrimidines for the potential treat-
ment of the dry eye condition. Rescaffolding of GSK256066
(IC50 = 0.01 nM)11,12 led to the preparation of 7-(3-methoxy-
phenylamino)-3-methyl-pyrazolo[1,5-a]pyrimidine-6-carboxylic
acid amide compound 1 which was optimised to compound 10 (see
Fig. 2).
Overlay between compound 5 and GSK256066 (PDB entry
3GWT)12 show the superposition of the amide, 3-methoxy aryl
and sulfone groups (see Fig. 3). The Gln443 (PDE4B) interacts with
the heterocycle nitrogen and the amide via a water molecule, and a
second hydrogen bond is formed between the amide and the car-
bonyl of Asn395.
The first compound made as a proof of concept was compound
1 which was found to be moderately active on PDE4 with an
IC50 = 165 nM (Table 1), and selective towards PDE3 (IC50 > 10
results not shown).
lM,
A cellular assay measuring the production of cAMP in human
SH-SY5Y cell lines was set up and compound 1 was found inactive
(effect <50% @ 10 lM, see Table 1). Replacement of the anisole with
a simple phenyl as in compound 2, and methylation of the primary
amide as in compounds 3, led to a loss in activity as seen in the
quinoline series.11,12 Introduction of the 3,5-dichloro-phenyl as in
compound 4 improved IC50 to 16 nM, although no cellular activity
⇑
Corresponding author. Tel.: +33 (0)1 58 93 36 14; fax: +33 (0)1 58 93 86 14.
E-mail address: frank.halley@sanofi.com (F. Halley).
http://dx.doi.org/10.1016/j.bmcl.2015.11.093
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

J. Le Roux et al. / Bioorg. Med. Chem. Lett. 26 (2016) 454–459
455
O
Cl
O
O
O
N
N
S
O
O
O
O
O
N
H
O
NH
Cl
N
H
O
O
Rolipram
O
Roflumilast
Apremilast
F
F
Figure 1. Structures of Rolipram, Roflumilast and Apremilast.
O
HN
N
O
S
O
O
HN
O
O
HN
N
O
O
O
7
O
6
1
O
O
N
H2N
N
N
H2N
N
H2N
N
S
2
N
N
4
5
3
10
GSK256066
Figure 2. Structures of compound 1 and GSK256066.
Figure 3. Overlay of compound 5 (pink) and GSK256066 (orange).
was detected. The synthetic pathway of compounds 1–4 is shown
in Scheme 1. Commercially available 3-amino-4-methyl pyrazole
was reacted with ethyl (ethoxymethylene)cyanoacetate in acetic
acid at reflux for 4 h to give 7-amino-3-methyl-pyrazolo[1,5-a]
pyrimidine-6-carboxylic acid ethyl ester in 60% yield. Subsequent
hydrolysis with sodium hydroxide followed by Ullmann coupling
with the appropriate iodo-aryl and amidification of the acid led
to compounds 1–4. Compound 7 was also synthesized following
this scheme, starting from the commercially available 3-amino-4-
ethyl pyrazole. To improve further the potency of these com-
pounds, the methyl sulfone moiety was introduced at position 2
(see compound 5 in Fig. 3). This small structural change required
to revisit entirely the synthetic scheme, since attempts to function-
alize this position, through the introduction of a halogen, failed.
The preparation of the sulfones, compounds 5 and 6, were
undertaken in 5 steps starting from 2-methyl-3,3-bis-methylsul-
fanyl-acrylonitrile13 which was reacted with hydrazine hydrate
in ethanol under reflux to provide the key intermediate 3-amino-
4-methyl-5-methylsulfanyl-pyrazole (Scheme 2). This intermedi-
ate was cyclized with ethyl (ethoxymethylene)cyanoacetate in
refluxing ethanol to form the 7-amino-3-methyl-2-methylsul-
fanyl-pyrazolo[1,5-a]pyrimidine-6-carboxylic acid ethyl ester,
which was oxidized to the sulfone with hydrogen peroxide in
acetic acid, coupled with the suitable aryl iodide in Ullmann condi-
tions and finally treated with ammonia under microwave condi-
tions to give the amides 5 and 6. Compounds 5 and 6 did show
an activity improvement of 6- and 2-fold over compounds 1 and
4, respectively. Also for the first time, observable cellular activity
was detected in the micromolar range with IC50s of 1820 nM and
1500 nM, respectively (Table 1).
Those promising results motivated the preparation of the corre-
sponding arylsulfones. Introduction of the aryl groups implicated
the preparation of the intermediate thiol from the methyl sulfide
through a Pummerer rearrangement of the corresponding sulfoxide

Table 1
Biochemical and cellular activity, PSA and calculated solubility of pyrazolopyrimidine compounds 1–1618
Compd
Structure
MW
PDE4B1a
IC50 (nM)
cAMPb release
IC50 (nM)
PSAc (Å2)
Solc pH 7.4
l
M
Compd
Structure
MW
476
PDE4B1a
IC50 (nM)
cAMPb release
IC50 (nM)
PSAc (Å2)
Solc pH 7.4
lM
Cl
N
O
HN
N
O
O
O
HN
Cl
O
S
N
1
297
165
>10,000
>10,000
>10,000
95
226
9
7
5600
629
128
0.3
H2N
N
H2N
N
N
O
O
HN
N
O
O
S
O
N
HN
N
N
2
3
267
311
1950
85
81
290
249
10
11
509
547
0.7
0.2
157
148
8
1
H2N
N
N
N
H2N
N
N
O
Cl
O
HN
N
O
O
O
O
HN
N
Cl
O
S
N
>10,000
130
N
H
N
N
H2N
N
N
O
Cl
N
O
O
HN
N
O
O
HN
Cl
H2N
4
5
336
375
16
26
>10,000
85
11
38
12
13
403
442
5
6
270
507
137
128
22
S
O
H2N
N
N
HN
N
Cl
N
O
O
S
N
Cl
O
S
O
HN
N
H2N
N
1820
137
1
H2N
N
O
O
Cl
N
Cl
O
HN
N
Cl
O
S
O
HN
N
Cl
S
6
7
8
414
311
437
7
1500
5400
243
128
95
2
14
15
16
408
401
440
0.2
480
97
123
114
4
N
H2N
H2N
H2N
N
N
N
HN
HN
N
O
O
O
O
HN
N
O
HN
N
O
O
38
112
0.1
8
32
N
N
N
S
O
Cl
HN
N
O
O
S
N
O
HN
N
Cl
O
S
4
137
5
0.03
5
2
N
HN
N
O
O
a
b
c
IC50 values in nM are reported as the mean from at least 2 independent experiments.19
Inhibition of cAMP hydrolysis in human SH-SY5Y cell lines.20
PSA (Å2) and solubility at pH 7.4 were calculated with ACD v12. Advanced Chemistry Development, Inc., Toronto, ON, Canada, www.acdlabs.com.

J. Le Roux et al. / Bioorg. Med. Chem. Lett. 26 (2016) 454–459
457
R2
R2
OEt
Cpd R1 R2
R3
O
HN
N
O
HN
N
OH
NH2
O
NH2
EtO2C
N
R2
1 = Me 3-MeO H
H
N
I
N
N
2 = Me H
3 = Me 3-MeO Me
4 = Me 3,5-Cl
7 = Et 3-MeO H
H
N
HO
N
N
HN
R3
N
O
N
O
N
CN
(i)
H
H2N
(iii)
(iv)
(ii)
N
N
R1
R1
R1
R1
R1
Scheme 1. (i) 3-Amino-4-alkyl pyrazole, ethyl (ethoxymethylene)cyanoacetate (1 equiv), AcOH reflux, 4 h, (60%); (ii) NaOH (10 equiv), EtOH, reflux, 4 h, (62%); (iii) ArI
(3 equiv), CuI (1 equiv), proline (1 equiv) K2CO3 (3 equiv), DMF, microwave 110 °C, 2 h, (5–35%); (iv) 1-hydroxybenzotriazole (HOBT, 3 equiv), 1-ethyl-3-(3-dimethylamino-
propyl)carbodiimide (EDCI, 3 equiv), R3NH2 (excess), DMF, rt, 2 h, (8–33%).
OEt
R1
NC
R1
O
HN
N
NH2
R1
NH2
Cpd R1
NH2
HN
N
CO2Et
EtO2C
I
H2N
(i)
O
S
EtO2C
O
H
N
N
S
N
H2N
N
O
N
N
EtO2C
O
N
5
6
=
=
3-OMe
3,5-Cl
O
N
N
S
N
S
(v)
(iii)
(ii)
S
S
O
(iv)
S
H2N
N
N
CN
Scheme 2. (i) Hydrazine hydrate (10 equiv), ethanol, reflux, 72 h, (36%); (ii) ethyl (ethoxymethylene)cyanoacetate (1 equiv), AcOH reflux, 1 h, (43%); (iii) hydrogen peroxide
(5 equiv), AcOH, 70 °C, 16 h, (80%); (iv) ArI (3 equiv), CuI (1 equiv), proline (0.7 equiv), K2CO3 (3 equiv), DMF, microwave, 110 °C, 2 h, (34%); (v) ammonium hydroxide
(excess), THF, 50 °C, 48 h, (14–15%).
with the mixed acetic trifluoroacetic anhydride14 followed by
hydrolysis with hydrochloric acid. The aryl group was subsequently
introduced via a second Ullmann reaction with the corresponding
aryl iodide, and the arylsulfide obtained, oxidized to the sulfone
with hydrogen peroxide in acetic acid. The ethyl ester was finally
converted to the primary amide in one step with formamide and
sodium methoxide in N,N-dimethylformamide at 100 °C. The syn-
thesis of the aryl sulfones 8, and 10 is shown in Scheme 3.
Introduction of the phenylsulfone in compound 8 improved
activity 6-fold when compared to compound 5 and a similar effect
was obtained in cell potency. This improvement was not seen on
the other hand with compounds 9 and 6, and compound 9 was
even less active in the cellular assay (Table 1). Finally the dimethyl
carboxamide was introduced since this substituent led to a 10-fold
potency increase in the case of the quinoline series.12 This strategy
did work partially for compounds 10 and 11, with IC50 reaching
sub-nanomolar levels, but the cellular potency remained
above 100 nM. Synthesis of compounds 9 and 11 was performed
following a similar sequence but replacing the ethyl ethoxymethy-
lene cyanoacetate with ethoxymethylene malononitrile, achieving
a better yield at the first condensation stage (71% instead of 43%).
The Ullmann and the oxidation steps where inverted and the more
convergent oxidation step was performed first. The nitrile used as a
‘masked’ or protected primary amide was hydrolyzed at the last
step with a commercially available dimethylphosphinito platinum
catalyst,15 since mild conditions using hydrogen peroxide in
dimethyl sulfoxide at room temperature16 did not work in that
case (see Scheme 4).
Position 3 was also optimized and introduction of slightly larger
ethyl (compound 7, Scheme 1) and isopropyl (compounds 12 and
13) groups led to a 3- to 5-fold increase in activity, when compar-
ing to compounds 1, 5 and 6, respectively, delineating a small
hydrophobic pocket around this position, but the cellular potency
remained in the 100 nM range. The synthesis of compounds 12
and 13 is shown in Scheme 5. The first step involved the prepara-
tion of the 2-isopropyl-3,3-bis-methylsulfanyl-acrylonitrile,
Cpd R1
O
HN
N
O
HN
N
O
HN
N
O
S
HN
N
O
S
HN
N
O
S
O
HN
N
O
EtO2C
N
8 = H
NH2
EtO2C
(vi)
H2N
(vii)
N
O
N
O
10 = CON(Me)2
EtO2C
(ii)
(iii)
(iv)
N
S
EtO2C
(i)
O
N
N
EtO2C
N
O
EtO2C
N
N
N
S
N
N
N
N
SH
(v)
S
R1
R1
N
R1
Scheme 3. (i) 3-Iodo-anisole (5 equiv), CuI (1 equiv), proline (1 equiv), K2CO3 (5 equiv), DMF, 140 °C, 16 h, (85%); (ii) NaIO4 (3 equiv), methanol/water (50/50; v/v), 50 °C,
16 h, (65%); (iii) acetic anhydride (20 equiv), trifluoroacetic acid (20 equiv), 50 °C, 16 h, (74%); (iv) HCl aq 10% in ethanol, reflux, 1 h (100%); (v) iodo-aryl (3 equiv), CuI
(1 equiv), proline (1 equiv), K2CO3 (5 equiv), DMF, 140 °C, 16 h, (30–49%); (vi) hydrogen peroxide (20 equiv), AcOH, 70 °C, 6 h, (46–53%); (vii) formamide (15 equiv), MeONa
(20 equiv), DMF, 100 °C, 3 h, (20%).
Cl
N
Cl
N
Cl
N
Cl
N
Cpd R1
OEt
NC
O
HN
N
Cl
HN
N
Cl
NH2
9 = H
HN
N
Cl
HN
N
Cl
NH2
11 = CON(Me)2
O
NC
(iii)
NC
(ii)
H
N
NC
N
O
O
(iv)
(v)
CN
(i)
NC
O
NC
O
(vi)
SH
(viii)
H2N
N
N
N
N
N
N
O
N
S
N
S
S
S
(vii)
S
H2N
S
N
N
R1
R1
Scheme 4. (i) Ethoxymethylene malononitrile (1.2 equiv), ethanol, reflux, 3 h, (71%); (ii) NaIO4 (1.3 equiv), methanol/water (50/50; v/v), 50 °C, 16 h, (80%); (iii) iodo-3,5-
dichlorobenzene (3 equiv), CuI (1 equiv), proline (1 equiv), K2CO3 (5 equiv), DMF, 140 °C, 16 h, (28%); (iv) acetic anhydride (40 equiv), trifluoroacetic acid (40 equiv), 50 °C,
16 h, (43%); (v) HCl aq 10% in ethanol, reflux, 1 h, (80%); (vi) iodo-aryl (3 equiv), CuI (1 equiv), proline (1 equiv) K2CO3 (5 equiv), DMF, 140 °C, 16 h, (33–54%); (vii) m-CPBA
(3 equiv), dichloroethane, 50 °C, 0.5 h, (80–90%); (vii) hydrogen bis(dimethylphosphinito-kP) platinum (0.2 equiv), ethanol/water (3/1; v/v), 80 °C, 3 h, (37–48%).

458
J. Le Roux et al. / Bioorg. Med. Chem. Lett. 26 (2016) 454–459
R1
R1
OEt
O
HN
N
NC
NH2
HN
N
Cpd R1
S
NH2
N
O
N
H
N
NC
(iv)
NC
O
O
N
N
N
H2N
(vi)
N
O
N
CN
(iii)
O
NC
N
O
N
12 = 3-OMe
13 = 3,5-Cl
N
S
S
S
S
N
S
(i)
(ii)
H2N
S
(v)
N
N
Scheme 5. (i) HMDSLi (2 equiv), CS2 (1 equiv), MeI (2 equiv), THF, ꢀ78 °C to rt, 2 h, (98%);8 (ii) hydrazine hydrate (10 equiv), butanol, 100 °C, 16 h, (73%); (iii)
ethoxymethylene malononitrile (1.2 equiv), acetic acid, reflux, 1 h, (42%); (iv) m-CPBA (3 equiv), dichloroethane, 50 °C, 0.5 h, (100%); (v) Ar-I (5 equiv), CuI (1 equiv), proline
(1 equiv) K2CO3 (5 equiv), DMF, 140 °C, 16 h, (52–55%); (vi) hydrogen bis(dimethylphosphinito-kP) platinum (0.2 equiv), ethanol/water (3/1; v/v), 80 °C, 3 h, (41–70%).
H
R1
S
R1
N
N
(1 eq.)
S
CN
O
NH2
HN
N
O
HN
N
O
O
H2N
Cpd R1
H
N
CN
O
O
HO
O
S
N
N
N
HN
N
HN
N
HN
N
15 = 3-OMe
16 = 3,5-Cl
O
NH
S
EtO2C
(v)
(iii)
(ii)
NH2
(iv)
(i)
CN
N
O
Cpd 14, R1 = 3,5-di-Cl
EtO2C
Scheme 6. (i) SOCl2 (1 equiv), DIPEA (2 equiv), dichloromethane, ꢀ20 °C to rt, 0.5 h, (11%); (ii) t-BuOK (3.75 equiv), THF, ꢀ78 °C to rt, 2 h (58%); (iii) POCl3, 120 °C, 0.5 h, (52%);
(iv) Ar-I (3 equiv), CuI (1 equiv), proline (1 equiv) K2CO3 (5 equiv), DMF, 140 °C, 16 h, (13–16%); (v) m-CPBA (3 equiv), dichloroethane, 50 °C, 0.5 h, (32–41%).
following the same procedure depicted in the literature for the
dithioacetals (Scheme 2) but using isopentylnitrile in place of
propionitrile.11
with the amide, the bis-aryl compound being the major product.
Compounds 15 and 16 showed more than 100-fold improvement
in activity as compared to compounds 5 and 6, respectively. In
the cellular assay a similar improvement was noticed, reaching
the single digit nanomolar potency. The methyl thioether deriva-
tive 14, showed an IC50 of 0.2 nM, although cellular potency was
high at 480 nM. Noteworthy, the tricyclic compounds exhibit bet-
ter ligand efficiency (LE) score as compared to the phenylsulfones,
and have molecular weights compatible with further potential
optimisation for generating drug like molecules.
Recently the synthesis and biological evaluation of 5-car-
bamoyl-2-phenylpyrimidine derivatives as novel and potent
PDE4 inhibitors was published,17 where a primary carboxamide
was cyclized to a secondary amide leading to a 50 fold increase
in cellular potency. To apply this strategy to our series, the whole
synthesis (depicted in Scheme 6) was revisited. The 2,4-dioxo-3-
piperidinecarbonitrile key intermediate was prepared in two
steps from ethyl-3-amino propionate, coupling with cyano acetic
acid followed by cyclisation with potassium tert-butoxide in
tetrahydrofuran (yields not optimized). Condensation with the
3-amino-4-methyl-5-methylsulfanyl-pyrazole (Scheme 2) afforded
the tricyclic template from which the final compounds could be
prepared by subsequent Ullmann coupling and methyl sulfide oxi-
dation to the sulfone with meta-chloroperbenzoic acid (m-CPBA).
The Ullmann step was low yielding in that case, due to competition
A
new pyrazolo[1,5-a]pyrimidine scaffold with moderate
affinity on PDE4 has been identified, which was improved to reach
sub-nanomolar IC50 level in the biochemical assay in three
successive 6-fold optimization rounds. From compound
1
(IC50 = 165 nM), the introduction of the methyl sulfone gave
compound 5 with a 6-fold increase in activity (IC50 = 26 nM).
Replacement of the methyl with a phenyl group gave compound
7 with a 6-fold gain in activity, (IC50 = 4 nM). Finally substitution
Figure 4. Overlay of compounds 5 (pink) and 15 (sky blue).

J. Le Roux et al. / Bioorg. Med. Chem. Lett. 26 (2016) 454–459
459
of the phenyl with a 3-N,N-dimethylcarboxamide gave compound
References and notes
10 with a 6-fold gain in activity (IC50 = 0.7 nM). This steep gain in
activity did not follow in the cellular assay, which even worsened
on the last optimisation round. One likely explanation being the
high PSA of 157 Å2 for compound 10 well above the 140 Å2 recom-
mendation for good cell penetration.21 Another important param-
eter to consider for activity is aqueous solubility (Table 1), which is
observed especially for the less soluble 3,5-dichloro-phenyl series.
Indeed, when going from compound 6 to compound 9 a jump in
activity was expected, but compound 9 displayed the same
biochemical activity as compound 6 and a 4-fold lower cellular
activity. This loss in activity could be related to the low solubility
observed with compound 9, this effect being only partially
restored with the N,N-dimethyl benzamide compound 11. Cyclisa-
tion of the primary amide afforded a new tricyclic scaffold with
200-fold improvement in biochemical and cellular activity (when
comparing compounds 5 and 15, for example). This gain in activity
could be explained by the favorable combination of two effects:
freezing of the bioactive conformation on the one hand, directing
the NH amide towards the carbonyl of Asn395, and displacing
the water molecule on the other hand (see Fig. 4). Indeed this
water molecule displacement has already been shown to occur
with another fused bicyclic 4-amino-2-phenylpyrimidine deriva-
tive (PDB code 3W5E).22 This water molecule is usually displaced
by the alkoxy group of the former PDE4 inhibitors, of which
Rolipram is the prototypical example,23 the X-ray structure of
Rolipram bound to PDE4B (PDB code 1RO6) shows the replace-
ment of the water molecule with the methoxy group of
Rolipram.24 WaterMap25,26 is a powerful tool for calculating the
energetics of structural water molecules in apo or holo struc-
tures.27 WaterMap analysis performed on structure PDB code
3GWT12 identified this water molecule, close to Gln443 as having
1. Beavo, J. A.; Reifsnyder, D. H. Trends Pharmacol. Sci. 1990, 11, 150.
2. Zhu, J.; Mix, E.; Winblad, B. CNS Drug Rev. 2001, 7, 387.
3. Norman, P. Expert Opin. Ther. Pat. 1998, 8, 771.
4. Dyke, H. J.; Montana, J. G. Expert Opin. Investig. Drugs 2002, 11, 1.
5. Odingo, O. A. Expert Opin. Ther. Pat. 2005, 7, 773.
6. Pagès, L.; Gavaldà, A.; Lehner, M. D. Expert Opin. Ther. Pat. 2009, 19, 1501.
7. Maurice, D. H.; Ke, H.; Ahmad, F.; Wang, Y.; Chung, J.; Manganiello, V. C. Nat.
Rev. Drug Disc. 2014, 13, 290.
8. Sadrai, Z.; Stevenson, W.; Okanobo, A.; Chen, Y.; Dohlman, T. H.; Hua, J.;
Amparo, F.; Chauhan, S. K.; Dana, R. Invest. Ophthalmol. Vis. Sci. 2012, 53, 3584.
9. Moore, J. E.; Vasey, G. T.; Dartt, D. A.; McGilligan, V. E.; Atkinson, S. D.; Grills, C.;
Lamey, P. J.; Leccisotti, A.; Frazer, D. G.; Moore, T. C. B. Invest. Ophthalmol. Vis.
Sci. 2011, 52, 6174.
10. Stern, M. E.; Schaumburg, C. S.; Pflugfelder, S. C. Int. Rev. Immunol. 2013, 32, 19.
11. Woodrow, M. D.; Ballantine, S. P.; Barker, M. D.; Clarke, B. J.; Dawson, J.; Dean,
T. W.; Delves, C. J.; Evans, B.; Gough, S. L.; Guntrip, S. P.; Holman, S.; Holmes, D.
S.; Kranz, M.; Lindvaal, M. K.; Lucas, F. S.; Neuf, M.; Ranshaw, L. E.; Solanke, Y.
E.; Somers, D. O.; Ward, P.; Wiseman, J. O. Bioorg. Med. Chem. Lett. 2009, 17,
5261.
12. Lunniss, C. J.; Cooper, A. W. J.; Eldred, C. D.; Kranz, M.; Lindvall, M.; Lucas, F. S.;
Neu, M.; Preston, A. G. S.; Ranshaw, I. E.; Redgrave, A. J.; Robinson, J. E.; Shipley,
T. J.; Solanke, Y. E.; Somers, D. O.; Wiseman, J. O. Bioorg. Med. Chem. Lett. 2009,
19, 1380.
13. Dieter, R. K. J. Org. Chem. 1981, 46, 5031.
14. de Lucchi, O.; Miotti, U.; Modena, G. Org. React. 2004, 40, 157.
15. Jiang, X. B.; Minnaard, A. J.; Feringa, B. L.; de Vries, J. G. J. Org. Chem. 2004, 69,
2327.
16. Katritzky, A. R.; Pilarski, B.; Urogdi, L. Synthesis 1989, 12, 949.
17. 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. 2013, 21, 7025.
18. All compounds were characterized by LC/MS and 1H NMR analysis.
19. The ability of compounds to inhibit catalytic activity of PDE4B and PDE3
(human recombinant from Sf9 cell lines) was determined by Fluorescence-
capillary-electrophoresis Assay in 384-well format. Fluorescently labeled
analogues of cAMP was used as PDE substrate and due to their different
charge, substrate and product appeared at different times in the detection
window. Substrate turnover can thus be determined as the ratio of the product
peak over the sum of substrate and product peak.
20. Inhibition of cAMP hydrolysis was measured by HTRF (homogeneous time-
resolved fluorescence) from isolated human neuroblastome cell lines (SH-
SY5Y) using a competitive immunoassay between native cAMP produced by
cells and the cAMP labelled with the dye d2. The tracer binding is visualized by
a high and positive
D
G = +7.51 Kcal molꢀ1. Displacement of this
‘unhappy’ water molecule would therefore lead to a significant
gain in activity.28 Indeed it has been shown in serine proteases,
that unstable water displacement with protease inhibitors can
lead to major improvement in potency superior to hundredfold.29
The other route of optimisation was the position 3, where the
introduction of the isopropyl group led to another 6-fold gain
in activity when comparing compound 5 and compound 12
(IC50 = 5 nM), this activity increase was also seen in the cellular
assay. Thus combination of the 3-isopropyl and cyclisation of the
amide might lead to a cellular potency close to 1 nM in this assay.
Considering the relatively low molecular weight of compound
15 many new combinations can be envisaged, such as the intro-
duction of solubilizing groups at positions 2, 3, and 7, contributing
to the opening of new opportunities towards the development of
potent drug candidates. Compound 16, confirmed the interest of
the series, with a potent biochemical activity (IC50 = 0.03 nM) and
a cellular potency of 5 nM.
a
Mab anti-cAMP labelled with cryptate. The specific signal is inversely
proportional to the concentration of cAMP in the sample.
21. Clark, D. E.; Pickett, S. D. Drug Discovery Today 2000, 5, 49.
22. 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.
23. Huai, Q.; Wang, W.; Sun, Y.; Kim, H.-Y.; Liu, Y.; Ke, H. Structure 2003, 11, 865.
24. Xu, R. X.; Rocque, W. J.; Lambert, M. H.; Vanderwall, D. E.; Luther, M. A.; Nolte,
R. T. J. Mol. Biol. 2004, 337, 355.
25. Abel, R.; Young, T.; Farid, R.; Berne, B. J.; Friesner, R. A. J. Am. Chem. Soc. 2008,
130, 2817.
26. Young, T.; Abel, R.; Kim, B.; Berne, B. J.; Friesner, R. A. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 808.
27. Snyder, P. W.; Mecinovic, J.; Moustakas, D. M.; Thomas, S. W.; Harder, M.;
Mack, E. T.; Lockett, M. R.; Heroux, A.; Sherman, W.; Whitesides, G. M. Proc.
Natl. Acad. Sci. U.S.A. 2011, 108, 17889.
28. Michel, J.; Tirado-Rives, J.; Jorgensen, W. L. J. Am. Chem. Soc. 2009, 131, 15403.
29. Abel, R.; Salam, N. K.; Shelley, J.; Farid, R.; Friesner, R. A.; Sherman, W.
ChemMedChem 2011, 6, 1049.