Phosphodiesterase inhibitors. Part 1: Synthesis and structure-activity relationships of pyrazolopyridine-pyridazinone PDE inhibitors developed from ibudilast

Article abstract of DOI:10.1016/j.bmcl.2011.04.021

Ibudilast [1-(2-isopropylpyrazolo[1,5-a]pyridin-3-yl)-2-methylpropan-1-one] is a nonselective phosphodiesterase inhibitor used clinically to treat asthma. Efforts to selectively develop the PDE3- and PDE4-inhibitory activity of ibudilast led to replacemen

Full text of DOI:10.1016/j.bmcl.2011.04.021

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3307–3312
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Phosphodiesterase inhibitors. Part 1: Synthesis and structure–activity
relationships of pyrazolopyridine–pyridazinone PDE inhibitors developed
from ibudilast
Robert W. Allcock a, Haakon Blakli a, Zhong Jiang a, Karen A. Johnston a, Keith M. Morgan a,
Georgina M. Rosair a, Kazuhiko Iwase b, Yasushi Kohno b, David R. Adams a,
⇑
a Chemistry Department, School of Engineering and Physical Sciences, Heriot-Watt University, Riccarton, Edinburgh EH14 4AS, United Kingdom
b Discovery Research Laboratories, Kyorin Pharmaceutical Co., Ltd, 2399-1, Nogi, Nogi-machi, Shimotsuga-gun, Tochigi 329-0114, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 February 2011
Revised 2 April 2011
Accepted 6 April 2011
Available online 13 April 2011
Ibudilast [1-(2-isopropylpyrazolo[1,5-a]pyridin-3-yl)-2-methylpropan-1-one] is a nonselective phospho-
diesterase inhibitor used clinically to treat asthma. Efforts to selectively develop the PDE3- and PDE4-
inhibitory activity of ibudilast led to replacement of the isopropyl ketone by a pyridazinone heterocycle.
Structure–activity relationship exploration in the resulting 6-(pyrazolo[1,5-a]pyridin-3-yl)pyridazin-
3(2H)-ones revealed that the pyridazinone lactam functionality is a critical determinant for PDE3-inhib-
itory activity, with the nitrogen preferably unsubstituted. PDE4 inhibition is strongly promoted by
introduction of a hydrophobic substituent at the pyridazinone N(2) centre and a methoxy group at
C-70 in the pyrazolopyridine. Migration of the pyridazinone ring connection from the pyrazolopyridine
30-centre to C-40 strongly enhances PDE4 inhibition. These studies establish a basis for development of
potent PDE4-selective and dual PDE3/4-selective inhibitors derived from ibudilast.
Keywords:
Cyclic 30,50-nucleotide phosphodiesterase
(PDE)
Ibudilast
PDE3 inhibitors
PDE4 inhibitors
Respiratory disease
Ó 2011 Elsevier Ltd. All rights reserved.
The cyclic 30,50-nucleotide phosphodiesterases (PDEs) play piv-
otal roles in shaping spatial and temporal gradients of the signal-
ling second messengers, cAMP and cGMP, mediating their
hydrolysis to 50-AMP and 50-GMP, respectively. There are 11 PDE
gene families, with enzymes belonging to the PDE3, PDE4 and
PDE5 families having received greatest attention to date as targets
for chemotherapeutic interventions. Inhibitors of cGMP-specific
PDE5 have been successfully deployed in the treatment of male
erectile dysfunction and pulmonary arterial hypertension.1 Iso-
forms of the PDE3 and PDE4 families regulate cAMP levels, with
the PDE3 inhibitor, milrinone (Fig. 1), used clinically for acute man-
agement of cardiogenic shock that is unresponsive to alternative
treatments.2 Development of PDE3 inhibitors was largely curtailed
in the 1990s when clinical data emerged that linked their long-
term application in heart failure patients to reduced survival
rates.3 Despite these adverse data, recent years have seen a resur-
gence of interest in PDE3 as a target allied to efforts to develop iso-
form-selective PDE inhibitors with improved pharmacological
profiles.4 Efforts to harness the therapeutic potential of PDE4
inhibitors have been substantially focused on anti-inflammatory
applications in asthma, chronic obstructive pulmonary disease
Figure 1. Structures of selected PDE3- and PDE4-inhibitors.
(COPD), atopic dermatitis and rheumatoid arthritis,5 but the cogni-
tive enhancing properties of PDE4 inhibitors have also received
attention.5c,6 Roflumilast (Fig. 1) recently became the first PDE4-
selective inhibitor to gain regulatory approval (for use within the
European Union) and is indicated for severe COPD.7 The anti-
inflammatory activity of PDE4 inhibitors as a potential treatment
for asthma is augmented by additional, if modest, bronchodilatory
activity.8 PDE3 inhibitors are reportedly9 superior to PDE4 inhibi-
tors in mediating airways smooth muscle relaxation, however,
leading some groups to explore the therapeutic potential of dual
PDE3/4 inhibitors.10
⇑
Corresponding author.
E-mail address: D.R.Adams@hw.ac.uk (D.R. Adams).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.04.021

3308
R. W. Allcock et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3307–3312
Table 1
PDE-inhibitory activity of compounds from Figures 2 and 3 and Schemes 1–3
Compound
Inhibition IC50
PDE4a
(lM)
PDE3a
PDE5a
Ibudilast
KCA-965 (1)
299
89
5.0
9.0
103
44
KCA-1312 (1R)
3.8
13
21
KCA-1313 (1S)
2
3
4
>400
110
72
>400
7.0
26
15
8.9
3.2
16
40
114
>400
200
11
30
2.0
73
16
39
2.5
16
17
1.7
29
>400
115
1.38
2.58
0.055
2.61
0.12
>400
78
79
>400
145
>400
>400
>400
156
>400
350
>400
>400
225
>400
65
13
14
15
16
17
18
27
28
29
30
31
32
33
34
41
42
43
44
45
69
210
NIb
NIc
260
>400
>400
>400
>400
NIb
2.75
41.5
39.6
33.5
Figure 2. Ibudilast and prototypical compounds of a new PDE-inhibitory series of
6-(pyrazolo[1,5-a]pyridin-3-yl)pyridazin-3(2H)-ones.
>400
>400
NIb
Ibudilast (Fig. 2) is a nonselective PDE inhibitor used in Japan to
treat allergies and asthma. The compound inhibits PDE3, PDE4 and
PDE5 isoforms with Ki values ranging from 3.3 to 78 lM, but it also
>100
6.4
inhibits most other PDE families in the micromolar range.11 Efforts
to selectively develop the PDE4- or dual PDE3/4-inhibitory activity
of ibudilast, as part of an ongoing respiratory disease programme,
led to replacement of the isopropyl ketone by the pyridazinone
subunit of KCA-965 (1, Fig. 2).12 Introduction of the pyridazinone
was prompted by the observation that this heterocycle features
in a number of PDE3 inhibitors, such as imazodan13 and CI-930,14
as well as the dual PDE3/4 inhibitor, zardaverine15 (Fig. 1). Synthe-
sis of the prototypical pyrazolopyridine–pyridazinone, KCA-965,
and its enantiomers (KCA-1312, KCA-1313) has been reported pre-
viously.12 We have also recently described the development of
concise synthetic routes for the construction of a series of 6-(pyr-
azolo[1,5-a]pyridin-3-yl)pyridazin-3(2H)-ones related to KCA-
965.16 This synthetic programme facilitated a detailed structure–
activity relationship survey for the compound series, and we sum-
marise here the most important findings of that survey in relation
to in vitro PDE3-, PDE4- and PDE5-inhibitory activity (assayed
using the core catalytic domains from the PDE3A, PDE4B and
PDE5A isoforms using previously reported protocols11b). The re-
sults of this survey provide a basis for development of potent
PDE4-selective and dual PDE3/4-selective inhibitors derived from
ibudilast.
The unsubstituted parent dihydropyridazinone and pyridazi-
none core structures 2 and 3 (Fig. 2) exhibited only modest levels
of PDE-inhibitory activity (Table 1). Introduction of the 5-methyl
group in KCA-965 (1) led to a slight enhancement in activity. Inter-
estingly, whereas both 5-methyldihydropyridazinone enantiomers
(1R and 1S) exhibited comparable activity against PDE4, the (S)-
enantiomer was essentially inactive against PDE3 and PDE5. A sec-
ond striking observation was that introduction of the 5-methyl
group in pyridazinone 3, affording 4, severely compromised activ-
ity against all three PDEs (Table 1). This loss of activity caused by
methyl substitution at the sp2-hybridised 5-position in 4 suggested
a probable requirement for a near coplanar ring arrangement in the
inhibitor’s enzyme-bound conformation with each enzyme, and
the presence of the in-plane 5-methyl group would incur a signif-
icant energy penalty for adoption of such a conformation. These
considerations pointed to an enzyme-bound topology for the
dihydropyridazinone in which the 5-methyl group may assume a
pseudoaxial position, orientated approximately orthogonal to the
near coplanar core of the two heterocycles. To test this hypothesis
we prepared and evaluated a number of torsionally-constrained
NIb
28.9
a
Enzyme assays were performed using the core catalytic domains from the
PDE3A, PDE4B and PDE5A isoforms according to previously reported protocols.11b
Data reported are the mean of at least three experiments.26
b
No inhibition up to 100
No inhibition up to 400
l
l
M.
M.
c
analogues (Schemes 1 and 2), with coplanarity enforced by intro-
duction of bridging rings from the (dihydro)pyridazinone 5-posi-
tion to the pyrazolopyridine at either C-20 or C-40. In this way we
were able to assess the impact of locking the diazine functionality
orientated proximal to either the pyrido ring (Scheme 1, Series 1,
13–18) or to the pyrazole ring (Scheme 2, Series 2, 27 and 28).
Compounds 13–18 were prepared as summarised in Scheme 1.
The known tricycle 9 has been prepared by condensation of 1-
aminopyridinium iodide with cyclohexane-1,3-dione.17 In our
hands this reaction was low-yielding (6–12%) and failed altogether
for some N-aminopyridinium salts substituted at the C-2 position
(e.g., 1-amino-2-methoxypyridinium mesitylenesulfonate). We
developed an alternative route to 9 using the cycloaddition
reaction between alkynoate 6 and pyridine N-imine, the latter gen-
erated in situ by treatment of 1-aminopyridinium mesitylenesulf-
onate with K2CO3 in DMF. Dieckmann ring closure of cycloadduct 7
afforded tricyclic ketoester 8 which was subsequently hydrolysed
with concomitant decarboxylation to give 9.18 At this point the
route branched in order to access torsionally-constrained ana-
logues of the parent 5-methyldihydropyridazinone (1) and the
unsubstituted analogue (2). Thus, treatment of ketone 9 with base
followed by bromoacetate afforded ester 11. Closure of the dihyd-
ropyridazinone ring was accomplished by reaction with hydrazine
to give 13. Dihydropyridazinone 13 was converted into pyridazi-
none 17 by a standard bromination–dehydrobromination proce-
dure. Insertion of an additional alkylation step into the route (9
to 10) prior to the bromoacetate reaction allowed installation of
a methyl group at the 5-position of the dihydropyridazinone ring
subsequently formed in 15. The impact of pyridazinone ring
N(2)-alkylation was assessed by preparing the N-benzyl analogues
(14, 16 and 18).
Compounds 27 and 28, with the pyridazinone ring torsionally-
constrained by a tether to the pyrazolopyridine C-40 centre (Series

R. W. Allcock et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3307–3312
3309
Scheme 1. Reagents and conditions: (a) n-BuLi (2.4 equiv), THF, ꢀ78 °C, then ClCO2Me; (b) MeOH, HCl (cat.) (68% two steps); (c) 1-aminopyridinium mesitylenesulfonate,
K2CO3, DMF, 0 °C (49%); (d) LiHMDS, THF, ꢀ78 °C, (8, 70% from 7); (e) LiOH, THF, H2O (9, 40% from 8); (f) LiHMDS, THF, ꢀ78 °C, then MeI (10, 78% from 9); (g) LiHMDS, THF,
ꢀ78 °C, then BrCH2CO2Me (11, 80% from 9; 12, 74% from 10); (h) H2N–NH2, AcOH, EtOH, H2O, reflux (13, 80% from 11; 15, 51% from 12); (i) BnBr, Cs2CO3, DMF (14, 24% from
13; 18, 53% from 17); (j) BnBr, NaH, THF (16, 90% from 15); (k) Br2, HBr, AcOH, H2O, reflux (17, 92% from 13).
Scheme 2. Reagents and conditions: (a) O-mesitylenesulfonyl hydroxylamine, CH2Cl2 (62%); (b) methyl 4-methylpent-2-ynoate, K2CO3, DMF, 0 °C (66% as 3:2 mixture of 6-
hydroxymethyl and 4-hydroxymethyl isomers, the latter (21) was separated chromatographically); (c) MeSO2Cl, Et3N, CH2Cl2, then NaI/acetone (80%); (d) MeCO2But, LiHMDS,
THF, ꢀ78 °C, then addition of 22 (89%); (e) LiHMDS, THF, ꢀ78 °C (93%); (f) CF3CO2H, CH2Cl2 then PhMe/reflux (74%); (g) LiHMDS, THF, ꢀ78 °C, then BrCH2CO2Me (57%); (h)
H2N–NH2, AcOH, EtOH, H2O, reflux (77%); (i) BnBr, NaH, THF (57%).
2), were synthesised via tricyclic ketoester 24, Scheme 2. The latter
was prepared, as shown, by adaptation of a previously reported
route to a similar compound starting from pyridine-3-methanol
(19).19 The pyridazinone ring in 27 was then constructed in much
the same way as for the Series 1 analogues shown in Scheme 1.
Thus, cleavage of the ester in 24 with decarboxylation afforded ke-
tone 25. Reaction of the enolate derived from the latter with bro-
moacetate led to ketoester 26. The dihydropyridazinone ring in
27 was completed by reaction with hydrazine. Alkylation of 27
was also applied to afford the N-benzyl derivative (28).
Activity data for torsionally-constrained analogues 13–18, 27
and 28, which (for the structures possessing a stereogenic centre)
were assessed in racemic form, are summarised in Table 1. PDE5-
inhibitory activity was weak in all analogues. The Series 2 parent
structure (27) exhibited comparable PDE4-inhibitory activity to
the unconstrained parent structure (2). In contrast, the Series 1
parent structure (13) performed less well as a PDE4 inhibitor.
The oxidised, Series 1 pyridazinone analogue (17) exhibited com-
parable activity as a PDE4 inhibitor to the corresponding uncon-
strained structure (3). Introduction of the dihydropyridazinone
ring 5-methyl group (15) and N-benzylation of the (dihydro)pyrid-
azinone (14, 16, 18) markedly improved PDE4-inhibitory potency
in Series 1 analogues. However, the N-benzyl Series 2 compound
(28) also exhibited improved performance in the PDE4 assay, sug-
gesting that the unconstrained pyrazolopyridine–pyridazinones
may potentially bind the PDE4 catalytic pocket with a near copla-
nar topology to the ring systems but where the enzyme may toler-
ate either of the two possible relative orientations for the rings. In
contrast to the results with PDE4, the Series 2 torsionally-con-
strained analogues (27 and 28) performed poorly as inhibitors of
PDE3, whereas, the Series 1 analogues (13, 15, 17) showed respect-
able activity. These data suggest a PDE3-bound conformation in
which the pyridazinone ring diazine functionality preferentially
lies ‘syn’ to the pyrazolopyridine pyrido ring.
In addition to our efforts to probe the enzyme-bound topology
of the pyrazolopyridine–pyridazinone inhibitors, we also under-
took a SAR survey, focusing mainly on pyridine ring substitutions
and modifications to the pyridazinone ring lactam functionality.
The most important findings from this work are concisely summa-
rised with reference to the compounds shown in Figure 3, which
were prepared using the synthetic routes recently described by
us.16 Substitutions at the N(2)- and 70-positions were found to offer

3310
R. W. Allcock et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3307–3312
fit the PDE3 catalytic pocket in this mode, as shown for compound
15 (S-ent) docked into the PDE3B structure, Figure 4A. Signifi-
cantly, this binding mode also accounts for the stereochemical dis-
crimination between 1R and 1S exhibited by PDE3. Thus, a 5-
methyl group occupying the pseudoaxial position, and orientated
approximately orthogonal to the near coplanar core heterocycle
arrangement, can only be accommodated on one face of the ligand.
In the case of 1R and 15 (S-ent) the methyl group is predicted to fit
a small subpocket adjacent to the purine-stacking Phe residue. The
methyl group of 1S and 15 (R-ent) cannot be accommodated in
such a binding mode, thus the activity of 15 is predicted to reside
with the (S)-enantiomer, although we did not confirm this by sep-
arate evaluation of the two enantiomers for 15.
Figure 3. Representative 6-(pyrazolo[1,5-a]pyridin-3-yl)pyridazin-3(2H)-ones and
related compounds that provided informative SAR data.
In PDE4 Tyr takes the place of the distal His of PDE3 and the
conformational preferences of the purine-scanning Gln differ
somewhat between PDE3 and PDE4, thereby contributing to the
cAMP-binding specificity exhibited by PDE4 isoforms.22 The reten-
tion of PDE4-inhibitory activity in chloropyridazine 32 as well as
the insensitivity of PDE4 to the stereochemistry of 1R/1S and the
activity enhancement associated with introduction of the 70-meth-
oxy group (30 and 31) are all indicative of a ligand binding mode in
PDE4 which is fundamentally different from that which operates
for PDE3. Hydrogen-bonded engagement of the purine-scanning
glutamine is a universal feature seen in co-crystal structures of
inhibitors bound to PDE4, and the SAR data presented here suggest
that it is the pyrazolopyridine subunit (not the pyridazinone) that
fulfils this role in our compound series. This implies that the com-
pounds bind in reverse orientation in PDE4 (relative to PDE3), such
that the pyridazinone ring occupies the region proximal to the cat-
alytic metal centres in PDE4. The binding mode for 30 and 31
would then be reminiscent of that established for catechol ether
PDE4 inhibitors such as zadaverine21,23 and roflumilast,21,24 with
the pyrazolopyridine N(10) and 70-OMe groups hydrogen bonding
to the purine-scanning Gln (Fig. 4B).
Consideration of the PDE4 binding model for compounds such
as 30 (Fig. 4B) suggested that interaction of the (dihydro)pyridaz-
inone ring within the catalytic pocket may be suboptimal. This
led us to investigate the impact of repositioning its site of attach-
ment from C-30 to C-40 of the pyrazolopyridine in order to set the
ring more deeply within the catalytic pocket and improve surface
contact with the protein. We therefore prepared a test set of ‘40-
linked’ analogues (41–45) for evaluation, Scheme 3. The synthetic
strategy employed a Suzuki coupling of 3,6-dichloropyridazine
with a pyrazolopyridine-4-boronic acid derived from bromide 39,
greatest potential for enhancement of PDE4 inhibitory activity, and
a combination of 70-methoxy with N(2)-benzyl (29–31) was partic-
ularly effective (Table 1). The enhancement of PDE4-inhibitory
activity achieved with N-benzylation (29 and 31) mirrors the effect
seen in the N-benzyl torsionally-constrained analogues (14, 16,
18, 28) when compared with their respective parent structures
(13, 15, 17, 27). For PDE3 inhibition N-benzylation reduced activity
in the torsionally-constrained series (with the exception of com-
pound 16, where a slight improvement of activity was seen over
parent 15). In the unconstrained compounds (29 and 31) the
reduction in PDE3-inhibitory activity with N-benzylation was more
pronounced. This effect was also reproduced with a range of sub-
stituents (data not shown), and N(2)-alkylation consistently com-
promised inhibition of PDE3 (and PDE5) but enhanced inhibition
of PDE4. Chlorination of the pyridazinone ring (32) or transforma-
tion to the methoxypyridazine (34) also caused a collapse in PDE3
inhibition whilst allowing retention or improvement in the modest
PDE4-inhibitory activity of parent 3. The presence of the in-plane
5-methyl group in pyridazine 33 caused loss of activity across
the panel, mirroring the effect seen with pyridazinone 4. The key
finding here was that PDE3-inhibitory activity in the pyrazolopyri-
dine–pyridazinone series is critically dependent on the lactam
functionality of the (dihydro)pyridazinone ring. This is consistent
with a binding mode in which the pyridazinone engages the
purine-scanning Gln and distal His of the catalytic pocket, as seen
in the X-ray co-crystal structure of the PDE3B core catalytic do-
main in complex with another 5-methyldihydropyridazinone-con-
taining ligand.20 Such a binding mode accounts for the topological
preferences for inhibitor binding to PDE3 deduced from our tor-
sionally-constrained series. Thus, only the Series 1 structures can
Figure 4. (A) Model of 15 (S-ent) (orange stick) docked into the catalytic pocket of PDE3B (PDB: 1SO220). An alternative binding mode in which the inhibitor oxygen hydrogen
bonds to the purine-scannning Gln (rather than the distal His) may operate for some compounds alkylated on the pyridazinone N(2) (e.g., 14, 16). (B) Models of 30 (cyan stick)
and 42 (yellow stick) docked into the catalytic pocket of PDE4D (PDB: 1XOR21). Hydrophobic pyridazinone N(2)-substituents (e.g., Bn in 43, 45) favourably extend surface
contact with residues around the catalytic pocket rim.

R. W. Allcock et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3307–3312
3311
Scheme 3. Reagents and conditions: (a) NBS, MeCN, 90 °C (53%); (b) 3-methyl-1-butyne, Pd(TPP)4, CuI, THF, Pri2NH, 70 °C (sealed tube) (97%); (c) O-mesitylenesulfonyl
hydroxylamine, CH2Cl2, (d) K2CO3, DMF (39, 28% two steps from 37); (e) (1) n-BuLi, THF, ꢀ78 °C, (2) B(OMe)3, ꢀ78 °C to 20 °C, (3) SiO2 chromatography (34%); (f) 3,6-
dichloropyridazine, Pd(PPh3)4, Na2CO3 (aq), THF, reflux (72%); (g) AcOH, reflux (42, 59% from 41); (h) BnBr, NaH, THF; (i) Zn, AcOH, 110 °C, 10 min (44, 35% from 42; 45, 83%
from 43). [X-ray crystal structures of 43 and 45 are shown.]30
the latter prepared from pyridine 35. Thus, bromination of 35
with NBS in hot MeCN afforded 2,3-dibromo-6-methoxypyridine
in 53% isolated yield after an easy separation from the minor 2,5-
dibromide product by crystallization. Sonogashira coupling with
3-methyl-1-butyne efficiently produced alkynylpyridine 36. Amina-
tion of 36 with O-mesitylenesulfonyl hydroxylamine followed by
cyclisation of the resulting N-aminopyridinium salt (37) under ba-
sic conditions via the pyridine N-imine (38) produced bicycle 39 in
28% yield. Simple 1-amino-2-(alk-1-ynyl)pyridinium salts undergo
reasonably efficient cyclisation to pyrazolo[1,5-a]pyridines.25 In
the reaction leading to 39, however, competing deamination of
N-aminopyridinium salt 37 and intermediate pyridine N-imine
38 was observed, which returned alkynylpyridine 36 together with
the reduced alkenylpyridine derivative (40). Copious quantities of
N2 gas were generated during the base treatment of the salt 37
and this suggests a decomposition mechanism in which the pyri-
dine N-imine deaminates the salt to generate diimine. The produc-
tion of diimine then accounts for both the reduction of the alkyne
side chain, affording the cis-alkene (40), and for the observed evo-
lution of N2. Pyrazolopyridine 39 was boronated under standard
conditions by lithium–bromine exchange followed by reaction
with trimethylborate. The intermediate boronic acid was then used
in the Suzuki coupling with 3,6-dichloropyridazine to afford 41.
Solvolysis of 41 with hot acetic acid gave pyridazinone 42 and
thence dihydropyridazinone 44 after reduction with zinc dust in
acetic acid. Pyridazinone 42 was also N-benzylated to give 43,
which was subsequently reduced to 45.
enzyme-bound core structure arrangement of the inhibitors is
likely to be preserved in the 40-linked compound series, and indeed
this conformation is also observed in the solid state single crystal
X-ray structures of the two most potent PDE4-inhibitory com-
pounds in the series (43 and 45), Scheme 3.
The results presented here suggest that the nonselective PDE-
inhibitory activity of ibudilast likely arises from the capacity of this
small structure to fit the catalytic pockets of all the PDE families,
with a hydrogen bond established from the pyrazolopyridine
N(1) centre to the conserved purine-scanning glutamine residue
of these enzymes. The structural modifications that we have de-
scribed establish a firm basis for development of both potent
PDE4-selective and dual PDE3/4-selective inhibitors derived from
ibudilast. Compound 42, in particular, possesses well-balanced
PDE3- and PDE4-inhibitory activity with considerable selectivity
over PDE5. In contrast, N-benzyl pyridazinone 43 exhibits potent
activity against PDE4 with 2–3 magnitude orders of selectivity over
PDE5 and PDE3.
Acknowledgments
PyMOL from W.L. DeLano, DeLano Scientific, was used to create
the protein structure images in this work. We thank the EPSRC
Mass Spectrometry Centre at Swansea for mass spectra and Dr
S.J. MacKenzie, Dr S. Hastings and Dr R.A. Clayton of the Kyorin
Scotland Research Laboratory for performing PDE inhibition assays.
Pleasingly, activity data for 41–45 confirmed out hypothesis
that repositioning of the (dihydro)pyridazinone would enhance
PDE4-inhibitory potency. Thus, compounds 44 and 45 exhibited
6.5- and 14-fold enhancements of PDE4-inhibitory activity over
their isomeric 30-linked counterparts (30 and 31, respectively).
The improvement in PDE4 inhibition was particularly striking with
N-benzylpyridazinone 43 (IC50 55 nM). Significantly, pyridazinone
42 and chloropyridazine 41 possessed comparable activity as
PDE4 inhibitors, consistent with the proposed PDE4 binding mode
(Fig. 4B) in which the pyridazinone does not contribute key hydro-
gen bond sites. In contrast, the low micromolar PDE3-inhibitory
activity of 42 was completely lost in chloropyridazine 41. As with
the 30-linked pyrazolopyridine–pyridazinones, this suggests that
the preferred binding mode for 40-linked compounds involves the
pyridazinone ring in hydrogen bonded engagement of residues in
the distal region of the PDE3 catalytic pocket. The near coplanar
References and notes
1. Giovannoni, M. P.; Vergelli, C.; Graziano, A.; Dal Piaz, V. Curr. Med. Chem. 2010,
17, 2564.
2. Klocke, R. K.; Mager, G.; Kux, A.; Hopp, H. W.; Hilger, H. H. Am. Heart J. 1991,
121, 1965.
3. (a) Packer, M.; Carver, J. R.; Rodeheffer, R. J.; Ivanhoe, R. J.; DiBianco, R.;
Zeldis, S. M.; Hendrix, G. H.; Bommer, W. J.; Elkayam, U.; Kukin, M. L., et al
N. Engl. J. Med. 1991, 325, 1468; (b) Nony, P.; Boissel, J. P.; Lievre, M.;
Leizorovicz, A.; Haugh, M. C.; Fareh, S.; de Breyne, B. Eur. J. Clin. Pharmacol.
1994, 46, 191.
4. (a) Thompson, P. E.; Manganiello, V.; Degerman, E. Curr. Top. Med. Chem. 2007,
7, 421; (b) Movsesian, M.; Stehlik, J.; Vandeput, F.; Bristow, M. R. Heart Fail. Rev.
2009, 14, 255.
5. (a) Teixeira, M. M.; Gristwood, R. W.; Cooper, N.; Hellewell, P. G. Trends
Pharmacol. Sci. 1997, 18, 164; (b) Doherty, A. M. Curr. Opin. Chem. Biol. 1999, 3,
466; (c) Houslay, M. D.; Schafer, P.; Zhang, K. Y. Drug Discovery Today 2005, 10,
1503; (d) Huang, Z.; Mancini, J. A. Curr. Med. Chem. 2006, 13, 3253; (e) Dastidar,
S. G.; Rajagopal, D.; Ray, A. Curr. Opin. Investig. Drugs 2007, 8, 364.

3312
R. W. Allcock et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3307–3312
6. (a) Rose, G. M.; Hopper, A.; De Vivo, M.; Tehim, A. Curr. Pharm. Des. 2005, 11,
3329; (b) Reneerkens, O. A.; Rutten, K.; Steinbusch, H. W.; Blokland, A.;
Prickaerts, J. Psychopharmacology (Berl.) 2009, 202, 419.
7. Giembycz, M. A.; Field, S. K. Drug Des. Develop. Ther. 2010, 4, 147.
8. Chung, K. F. Eur. J. Pharmacol. 2006, 533, 110.
23. Lee, M. E.; Markowitz, J.; Lee, J. O.; Lee, H. FEBS Lett. 2002, 530, 53.
24. 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.
25. (a) Tsuchuja, T.; Sashida, H.; Konoshita, A. Chem. Pharm. Bull. 1983, 31, 4568;
(b) Tsuchuja, T.; Sashida, H. J. Chem. Soc., Chem. Commun. 1980, 1109.
26. The isolated catalytic domains of PDE3A, PDE4B and PDE5A were used in our
primary assays and the inhibitory potencies found for ibudilast differ from
those reported in earlier studies,11b,c which used different constructs. We
evaluated compound activity against the PDE core catalytic domains in our
foundational studies because sensitivity to inhibitors can be influenced by the
nature of the N-terminal regulatory regions, especially for the PDE4 enzymes.
These are encoded by four genes (PDE4A–D) that each give rise to multiple
isoforms.22a,27 The isoforms are categorised as ‘long’, ‘short’ or ‘supershort’
according to the presence or absence of regulatory modules, upstream
conserved regions 1 and 2 (UCR1, UCR2), located between a unique isoform-
specific N-terminal sequence and the core catalytic domain.22a The UCR
modules transduce the functional consequences of regulatory phosphorylation
by kinases such as PKA,22a but we have recently shown that ibdudilast is one of
a group of inhibitors that can interact with a ‘gating sequence’ in UCR2,
switching the PDE4 conformational state to one in which the inhibitor-
occupied catalytic pocket is capped by UCR2.28 Such conformational switching
can lead to profound changes in inhibitor affinities and is sensitive to both
phosphorylation and partner protein binding.29 Thus, for the archetypal PDE4
inhibitor, rolipram, conformational interchange can switch PDE4 enzymes
between low-affinity and high-affinity rolipram binding states (LARBS/HARBS)
with IC50s in the micromolar and nanomolar ranges, respectively.22a
27. Conti, M.; Richter, W.; Mehats, C.; Livera, G.; Park, J. Y.; Jin, C. J. Biol. Chem. 2003,
278, 5493.
28. (a) Terry, R.; Cheung, Y. F.; Praestegaard, M.; Baillie, G. S.; Huston, E.; Gall, I.;
Adams, D. R.; Houslay, M. D. Cell. Signal. 2003, 15, 955; (b) Christian, F.;
Anthony, D. F.; Vadrevu, S.; Riddell, T.; Day, J. P.; McLeod, R.; Adams, D. R.;
Baillie, G. S.; Houslay, M. D. Cell. Signal. 2010, 22, 1576; (c) Day, J. P.; Lindsay, B.;
Riddell, T.; Jiang, Z.; Allcock, R. W.; Abraham, A.; Sookup, S.; Christian, F.;
Bogum, J.; Martin, E. K.; Rae, R. L.; Anthony, D.; Rosair, G. M.; Houslay, D. M.;
Huston, E.; Baillie, G. S.; Klussmann, E.; Houslay, M. D.; Adams, D. R. J. Med.
Chem. 2011, doi:10.1021/jm200070e.
29. (a) Houslay, M. D.; Adams, D. R. Nat. Biotechnol. 2010, 28, 38; (b) Alvarez, R.;
Sette, C.; Yang, D.; Eglen, R. M.; Wilhelm, R.; Shelton, E. R.; Conti, M. Mol.
Pharmacol. 1995, 48, 616; (c) McPhee, I.; Yarwood, S. J.; Scotland, G.; Huston, E.;
Beard, M. B.; Ross, A. H.; Houslay, E. S.; Houslay, M. D. J. Biol. Chem. 1999, 274,
11796; (d) Saldou, N.; Obernolte, R.; Huber, A.; Baecker, P. A.; Wilhelm, R.;
Alvarez, R.; Li, B.; Xia, L.; Callan, O.; Su, C.; Jarnagin, K.; Shelton, E. R. Cell. Signal.
1998, 10, 427.
9. Torphy, T. J.; Undem, B. J.; Cieslinski, L. B.; Luttmann, M. A.; Reeves, M. L.; Hay,
D. W. J. Pharmacol. Exp. Ther. 1993, 265, 1213.
10. (a) Van der Mey, M.; Bommele, K. M.; Boss, H.; Hatzelmann, A.; Van
Slingerland, M.; Sterk, G. J.; Timmerman, H. J. Med. Chem. 2003, 46, 2008; (b)
Giembycz, M. A. Proc. Am. Thorac. Soc. 2005, 2, 326; (c) Boswell-Smith, V.;
Spina, D.; Oxford, A. W.; Comer, M. B.; Seeds, E. A.; Page, C. P. J. Pharmacol. Exp.
Ther. 2006, 318, 840; (d) Banner, K. H.; Press, N. J. Br. J. Pharmacol. 2009, 157,
892; (e) Yamazaki, T.; Anraku, T.; Matsuzawa, S. Eur. J. Pharmacol. 2011, 650,
605.
11. (a) Kishi, Y.; Ohta, S.; Kasuya, N.; Sakita, S.; Ashikaga, T.; Isobe, M. Cardiovasc.
Drug Rev. 2001, 19, 215; (b) Gibson, L. C. D.; Hastings, S. F.; McPhee, I.; Clayton,
R. A.; Darroch, C. E.; Mackenzie, A.; MacKenzie, F. L.; Nagasawa, M.; Stevens, P.
A.; MacKenzie, S. J. Eur. J. Pharmacol. 2006, 538, 39; (c) Huang, Z.; Liu, S.; Zhang,
L.; Salem, M.; Greig, G. M.; Chan, C. C.; Natsumeda, Y.; Noguchi, K. Life Sci. 2006,
78, 2663.
12. (a) Kohno, Y.; Ogata, T.; Awano, K.; Matsuzawa, K.; Tooru, T. PCT Int. Appl.
Patent WO9814448, 1998.; (b) Yoshida, N.; Aono, M.; Tsubuki, T.; Awano, K.;
Kobayashi, T. Tetrahedron: Asymmetry 2003, 14, 529; (c) Yoshida, N.; Awano, K.;
Kobayashi, T.; Fujimori, K. Synthesis 2004, 1554.
13. Sircar, I.; Duell, B. L.; Bobowski, G.; Bristol, J. A.; Evans, D. B. J. Med. Chem. 1985,
28, 1405.
14. Bristol, J. A.; Sircar, I.; Moos, W. H.; Evans, D. B.; Weishaar, R. E. J. Med. Chem.
1984, 27, 1099.
15. Schudt, C.; Winder, S.; Muller, B.; Ukena, D. Biochem. Pharmacol. 1991, 42, 153.
16. Johnston, K. A.; Allcock, R. W.; Jiang, Z.; Collier, I. D.; Blakli, H.; Rosair, G. M.;
Bailey, P. D.; Morgan, K. M.; Kohno, Y.; Adams, D. R. Org. Biomol. Chem. 2008, 6,
175.
17. Tamura, Y.; Tsujimoto, N.; Sumida, Y.; Ikeda, M. Tetrahedron 1972, 28, 21.
18. This route to the tricyclic ketone was more versatile as it could be successfully
applied with other N-aminopyridinium salts such as the 2-methoxy
substituted salt (not shown).
19. (a) Gmeiner, P.; Sommer, J. Arch. Pharm. (Weinheim, Ger.) 1994, 327, 435; (b)
Gmeiner, P.; Sommer, J.; Mierau, J.; Höfner, G. Bioorg. Med. Chem. Lett. 1993, 3,
1477.
20. Scapin, G.; Patel, S. B.; Chung, C.; Varnerin, J. P.; Edmondson, S. D.; Mastracchio,
A.; Parmee, E. R.; Singh, S. B.; Becker, J. W.; Van der Ploeg, L. H.; Tota, M. R.
Biochemistry 2004, 43, 6091.
21. Card, G. L.; England, B. P.; Suzuki, Y.; Fong, D.; Powell, B.; Lee, B.; Luu, C.;
Tabrizizad, M.; Gillette, S.; Ibrahim, P. N.; Artis, D. R.; Bollag, G.; Milburn, M. V.;
Kim, S. H.; Schlessinger, J.; Zhang, K. Y. J. Structure 2004, 12, 2233.
22. (a) Houslay, M. D.; Adams, D. R. Biochem. J. 2003, 370, 1; (b) Zhang, K. Y. J.; Card,
G. L.; Suzuki, Y.; Artis, D. R.; Fong, D.; Gillette, S.; Hsieh, D.; Neiman, J.; West, B.
L.; Zhang, C.; Milburn, M. V.; Kim, S. H.; Schlessinger, J.; Bollag, G. Mol. Cell
2004, 15, 279; (c) Ke, H.; Wang, H. Curr. Top. Med. Chem. 2007, 7, 391; (d) Wang,
H.; Robinson, H.; Ke, H. J. Mol. Biol. 2007, 371, 302.
30. The X-ray CIF files for 43 and 45 have been deposited at the Cambridge
Crystallographic Data Center with the deposition numbers CCDC 811960 and
CCDC 811961. Copies of the data can be obtained, free of charge, from CCDC, 12
Union Road, Cambridge, CB2 1EZ UK (e-mail: deposit@ccdc.cam.ac.uk;
Internet: //www.ccdc.cam.ac.uk).