First dual M3 antagonists-PDE4 inhibitors: Synthesis and SAR of 4,6-diaminopyrimidine derivatives

Article abstract of DOI:10.1016/j.bmcl.2006.01.006

SAR around 4,6-diaminopyrimidine derivatives allowed the discovery of the first potent dual M₃ antagonists and PDE4 inhibitors. Various chemical modulations around that scaffold led to the discovery of ucb-101333-3 which is characterized by the

Full text of DOI:10.1016/j.bmcl.2006.01.006

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1834–1839
First dual M3 antagonists-PDE4 inhibitors: Synthesis
and SAR of 4,6-diaminopyrimidine derivatives
Laurent Provins,a,* Bernard Christophe,b Pierre Danhaive,a Jacques Dulieu,a
a
c
a
a
´
´
´
Veronique Durieu, Michel Gillard, Florence Lebon, Sebastien Lengele,
a
´ ´a
Luc Quere and BerendJan van Keulen
aGlobal Chemistry, UCB, R&D, Chemin du Foriest, B-1420 Braine-L’Alleud, Belgium
bGeneral Pharmacology, UCB, R&D, Chemin du Foriest, B-1420 Braine-L’Alleud, Belgium
cIn Vitro Pharmacology, UCB, R&D, Chemin du Foriest, B-1420 Braine-L’Alleud, Belgium
Received 13 December 2005; accepted 4 January 2006
Available online 24 January 2006
Abstract—SAR around 4,6-diaminopyrimidine derivatives allowed the discovery of the first potent dual M3 antagonists and PDE4
inhibitors. Various chemical modulations around that scaffold led to the discovery of ucb-101333-3 which is characterized by the
most interesting profile on both targets.
ꢀ 2006 Elsevier Ltd. All rights reserved.
COPD (chronic obstructive pulmonary disease) is a
chronic, progressive, and poorly reversible condition
characterized by impaired expiratory outflow and
abnormal inflammatory response of the lungs to nox-
ious particles and gases. Chronic bronchitis, resulting
in small airway obstruction and mucus plugging, and
emphysema, responsible for parenchyma destruction,
are the two main components of this disease state, the
top three symptoms being coughing, wheezing and,
above all, shortness of breath.1–6
primary inflammatory cells associated with asthma are
CD4+ T-cells, mast cells and eosinophils, while those
associated with COPD are macrophages, CD8+ T-cells,
and neutrophils.1,2
Therefore, there is an urgent need to develop new treat-
ment approaches.3–6
Bronchodilator drugs are the current mainstay of treat-
ment for symptoms’ relief.8,9 Anticholinergic bronchod-
ilators, particularly selective muscarinic M3 antagonists,
are currently the preferred choice for the symptomatic
management of COPD as they are more effective and
have fewer side-effects than b2-adrenoceptor agonists.8
Tiotropium bromide (Spirivaꢁ), a new anticholinergic
with a very long duration of action, has been recently
registered and proved to be suitable for once-daily
administration as a dry powder (see Fig. 1).10,11
COPD is one of the most common chronic diseases
worldwide. It affects 4–6% of people older than 45 and
is predicted to be the third leading cause of death by
2020. Smoking tobacco is by far the most important risk
factor for the development and progression of the path-
ological condition.1–6
There are currently no drug therapies capable of slowing
down the progression of disease. Patients are commonly
treated with drugs developed for asthma but it proves to
be quite ineffective probably because the pathophysiolo-
gies of asthma and COPD differ markedly.2,7 The
However, although bronchodilators are quite effective to
improve symptoms, they do not address the underlying
chronic inflammation or the changes in airway structure.
Contrary to asthma, standard treatment with corticoste-
roids as anti-inflammatory agents has demonstrated
limited efficacy. However, among the new anti-inflam-
matory agents currently being developed, PDE4 inhibi-
tors proved to be very effective in attenuating
the responses of various inflammatory cells through
their ability to elevate cAMP levels. They are known
Keywords: Muscarinic receptor(s); Phosphodiesterase; Pyrimidine;
COPD.
*
Corresponding author. Tel.: +32 2 3863217; fax: +32 2 3863669;
e-mail: laurent.provins@ucb-group.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.01.006

L. Provins et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1834–1839
1835
Compounds bearing a nitro substituent in R4 have been
prepared by nitration of the corresponding 4,6-dihydr-
oxypyrimidine I-a using glacial acetic acid and nitric
acid.20 Bromination of compound IV-a with N-bromo-
succinimide afforded IV-b.21
F
F
Br
N+
O
O
O
S
HO
Cl
N
O
O
S
O
N
Cl
Roflumilast
(Daxas®)
Tiotropium Bromide
(Spiriva®)
The reduction of IV-c with sodium dithionite in the
presence of ammonia afforded IV-d.22
Figure 1. Selected M3 antagonist and PDE4 inhibitor targeting COPD.
The synthesis of compounds VII listed in Table 4 is
outlined in Scheme 2.
to modulate activity, migration, and apoptosis of
neutrophils by inhibiting the production and release of
chemokines, superoxide free radicals, leukotrienes, and
proteolytic and toxic granular enzymes.12–14
Condensation of amidines with 2-{3-[(tert-butoxy-
hydroxy-methyl)-amino]-alkyl}-diethylmalonates in the
presence of base afforded the corresponding 4,6-dihydr-
oxypyrimidines. Their reaction with phosphorus
oxychloride in the presence of N,N-diethylaniline affor-
ded the deprotected dichloropyrimidines, which were
then heated at high temperature to give the cyclized
compounds VI. Subsequent substitution with second-
ary amines afforded the desired compounds VII.
The tendency of first generation PDE4 inhibitors (e.g.,
rolipram) to cause side effects such as nausea, emesis,
and headache has limited their beneficial effect in clinical
studies. Nevertheless, newꢁgeneration compounds13,14
such as Roflumilast (Daxas )15,16 are currently in Phase
III for the treatment of COPD and have been shown to
significantly improve lung function and symptoms of
patients (see Fig. 1).
A series of 4,6-diaminopyrimidines were synthesized and
evaluated for their ability to inhibit phosphodiesterase
type 4 (PDE4), prepared from U937 cells,23 and to bind
to the M3 receptor.24
Thus, the combination of selective muscarinic M3 antag-
onism with selective PDE4 inhibition may lead to a new
class of drugs combining both bronchodilating and anti-
inflammatory properties.
The first modification envisaged was the replacement of
the thiomorpholine moiety, which was suspected to be
rapidly oxidized. A variety of compounds bearing var-
ious secondary and tertiary amines have been prepared
and tested. As mentioned in Table 1, the replacement
of thiomorpholine by amines bearing hydrophilic
In this paper, we present the discovery of a family of
dual M3 antagonists and PDE4 inhibitors as potential
new drugs for COPD treatment. We focused on the
structural optimization of cyclopropyl-(2-cyclopropyl-
5-methyl-6-thiomorpholin-4-yl-pyrimidin-4-yl)-amine 1
(Fig. 2), one of the first chemical leads discovered by
screening from our in-house library. Full details of
SAR studies are given.17
groups
(thiomorpholine-1,1-dioxide,
morpholine,
4-hydroxypiperidine, and piperidin-4-one) resulted in
a dramatic loss of M3 affinity, PDE4 inhibiting activity
is also decreasing but to a lesser extent.
Various physicochemical parameters were measured for
compounds listed in Table 1 and were correlated to M3
affinity. A positive correlation was found between the
experimental lipophilicity parameter K0IAMex25 and
the M3 affinity for the tertiary amine substituents which
adequately fill the space of a hypothetical hydrophobic
pocket within the receptor. The introduction of larger
lipophilic groups has however an unfavorable impact
on affinity. Equation 1, obtained by multiple regression
analysis (MRA) employing diverse common physico-
chemical descriptors (p, rm, F, R, MR, and Verloop. . .),
indeed shows a dependency of M3 affinity towards both
lipophilicity (K0IAMex) and steric bulk (Verloop
parameters L, B3, and B4).
The general method of synthesis of compounds listed in
Tables 1–4 is outlined in Scheme 1.
Briefly, condensation of 2-substituted dialkylmalonates
with various amidines in the presence of sodium ethox-
18
ide affords 4,6-dihydroxypyrimidines I
which were
then chlorinated with phosphorus oxychloride in the
presence of N,N-diethylaniline.19 The resulting
4,6-dichloropyrimidines II are subsequently substituted
with various primary or secondary amines to give the
desired compounds IV.
pKi ¼ 0:83 log K0IAMex ꢀ 0:61L þ 0:73B3 ꢀ 1:02B4
R2
R3
N
N
þ 6:84
HN
R4
n ¼ 18 r2 ¼ 0:81.
N
N
R5
R1
N
R6
N
N
In most cases, no dramatic change of PDE4 inhibiting
activity was observed following those substitutions but
it nevertheless appears that the optimal ring size is 6
(piperidines) or 7 (azepane), pyrrolidine or azocane
being less potent.
S
1
Figure 2.

1836
L. Provins et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1834–1839
Table 1. Modulation of NR5R6
Table 1 (continued)
Compound NR5R6=
PDE4 M3Ki
IC50
log K0IAMex
Compound NR5R6=
PDE4 M3Ki
IC50
log K0IAMex
(nM)
(nM)
(lM)
(lM)
1
2
3
4
5
0.32
0.79
4.0
79
2.14
1.45
1.20
1.35
2.50
21
22
23
24
0.5
6.3 3.11
N
N
N
N
N
S
N
HN
#
O
4.0
200
251
316
2.36
2.70
2.97
501
3200
13
HN
N
OH
O
0.40
0.50
3.2
F
0.50
25
2.0
63
2.38
HN
6
7
0.63
0.63
16
50
2.80
3.21
N
N
F
S
HN
HN
26
27
6.3
2.5
126
79
2.21
2.57
S
8
9
0.79
0.32
0.79
2.0
32
25
2.37
3.02
3.66
3.89
N
N
N
Results are expressed as IC50 (lM), Ki (nM) or # (when <50% inhi-
bition of radioligand specific binding by 10 lM of compounds).
Table 2. Modulations of R1
10
11
158
790
Compound
R1=
PDE4 IC50 (lM)
M3Ki (nM)
*
17
0.63
3.2
N
N
CH2Ph
*
12
13
0.63
2.0
501
100
1.57
2.94
28
2.0
1.26
2.5
2.0
1.0
3.2
40
79
OH
*
N
N
CF3
29
F
*
14
0.40
50
2.50
30
25
F
*
15
16
0.40
1.6
79
2.70
2.36
N
N
31
126
501
32
100
*
32
*
*
17
18
0.63
1.6
3.2 2.89
N
N
3326
(-)
16
3.05
3426
2.5
398
(+)
19
20
2.5
2.0
316
398
2.71
2.98
HN
HN
Cycloalkylamino substituents as well as several benzyl-
amines have also been synthesized and tested
(Table 1). Lower M3 affinities are obtained in most cases
except for 25 and 27.

L. Provins et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1834–1839
1837
Table 3. Modulations of R4
After this first series of substitutions, the compound 17,
bearing an azepanyl moiety, has been found to be char-
acterized by the most interesting balance between M3
affinity (3.2 nM), and PDE4 inhibiting activity
(630 nM) and has therefore been selected for further
optimization.
Compound R4=
NR5R6= PDE4
IC50 (lM) (nM)
M3Ki
pKa
1
17
35
36
37
38
39
40
41
42
CH3
CH3
H
0.32
0.63
0.40
0.63
0.63
1.26
0.79
2.5
79
6.99
N
N
N
N
N
N
N
S
All tentatives to replace the cyclopropyl ring (R1) result-
ed in a loss of M3 affinity as well as PDE4 inhibiting
activity, as reported in Table 2.
3.2 8.13
63
8.03
6.44
6.21
4.64
7.15
6.47
5.27
6.78
These results clearly show the importance of steric fac-
tors in that area. Increased ring size as well as larger
alkyl or benzyl groups is detrimental to both activities.
F
251
316
2512
79
The nature of the substituent in 5-position (R4) is also
highly important for the basicity and therefore for the
affinity for the M3-R as exemplified in Table 3. It has
however nearly no impact on PDE4 inhibiting activity.
Br
NO2
NH2
OCH3
F
For example, electronegative substituents such as halo-
gens (pKa ꢁ 6.5)27 or nitro (pKa ꢁ 4.6) led to a 25- to
800-fold decrease in M3 affinity. The affinity is main-
tained if the methyl group is replaced by hydrogen or
a small alkyl group (ethyl).
398
#
N
N
N
S
S
S
M3 affinity is however four times lower in the particular
case of the methoxy group (40) compared to the ethyl
group (42). For compound 40, the counter-intuitive
decrease in basicity is explained by an intramolecular
H-bond between the oxygen and the cyclopropylamino
moiety. This assumption has been confirmed by a
X-ray structure of the compound 40.28
0.25
1.6
CH2CH3
100
PDE4 inhibiting activity proved to be much more sensi-
tive toward the nature of the substituent in 4-position
(NR2R3) as shown in Table 4.
Table 4. Modulations of NR2R3
Compound NR2R3=
R4= PDE4 IC50 (lM) M3Ki (nM)
For example, removal, substitution in various positions
or replacement of the cyclopropane by larger rings led to
an important loss of activity, therefore highlighting the
importance of steric factors in that area. Affinities for
the M3-R are, in most cases, insensitive to these
modifications.
17
Me
Me
Me
0.63
2.0
3.2
7.9
HN
43
HN
44
6.3
10
HN
Bicyclic derivatives 46 and 47 in which R3 and R4
are linked together to form 5,6,7,8-tetrahydropyri-
do[2,3-d]pyrimidine or 6,7,8,9-tetrahydro-5H-pyrimi-
do[4,5-b]azepine derivatives have also been synthesized
according to Scheme 2. These are characterized by an
interesting profile on both targets but suffer from lower
selectivities than open derivatives toward other
receptors.
45
Me
7.9
7.9
HN
(+,-)
HN
46
0.5
1.3
50
47
10
HN
*
*
Compound 17 (ucb-101333-3) was therefore selected as
the most promising candidate to be further studied in
several assays. Its in vitro profile as well as several phys-
icochemical data is detailed in Table 5. It is character-
ized by a high selectivity for PDE4 versus other PDE
enzymes and towards the high affinity rolipram binding
site (HARBS). The anti-inflammatory activity of 17 has
also been confirmed in several cell-based assays (TNF-a
However, these examples emphasize the importance of
tertiary amines for PDE4 inhibiting activity. For exam-
ple, N-methyl-benzylamine and 1,2,3,4-tetrahydroiso-
quinoline substituents are about 10-fold more potent
than benzylamine itself (compare compounds 22, 23,
and 24 in Table 1).
or
elastase
release
from
polymorphonuclear

1838
L. Provins et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1834–1839
Cl
OH
R4
R4
Cl
NH
NH2
b
a
N
N
R1
R1
N
O
O
R1
N
OH
EtO
OEt
R4
I
II
4
4
R4 = H, Me, Et, F, Cl, OMe
R
I-a
I-b
= H
e
R
O2
R2
R3
R2
N
R3
N
N
R4
Cl
R4
c
d
N
R5
R2 R3
R5 R6
R1
N
R1
N
N
R6
N
H
N
H
III
IV
4
4
4
R
=NO2
R
= H
IV-c
IV-a
IV-b
g
f
4
R
=NH2 IV-d
R
= Br
Scheme 1. Synthesis of compounds IV. Reagents and conditions: (a) NaOEt, EtOH, 60 ꢂC; (b) POCl3, N,N-diethylaniline, 100 ꢂC; (c) neat, 45 ꢂC;
(d) neat, 100–120 ꢂC; (e) HOAc, HNO3, 0 ꢂC then 30 ꢂC; (f) NBS, CHCl3, 60 ꢂC; (g) Na2S2O4, NH4OH, dioxane/H2O, rt.
Table 5. ucb-101333-30s in vitro profile
OH
Ucb-101333-3 (17)
NH
a
( )n
OH
NH-Boc
N
Muscarinic profile
pKi M3
pKi M1
R1
NH2
O
O
R1
N
8.5
7.7
7.2
8.0
7.2
7.9
7.0
EtO
OEt
( )n
pKi M2
pKi M4
V
n = 1, 2
NH-Boc
pKi M5
pA2 M3
pA2 M2
d
( )n
HN
( )n
HN
e
d
b, c
N
N
Physicochemical profile
Log D (pH 7.4)
Log K0IAMex
R5
R5 R6
R1
N
N
R6
R1
N
Cl
2.90
2.75
N
H
S (pH 1, 4, 7.4)
7, 4, 0.01 mg/ml
VII
VI
PDE profile
Scheme 2. Synthesis of compounds VII. Reagents and conditions: (a)
NaOEt, EtOH, 60 ꢂC; (b) POCl3, N,N-diethylaniline, 100 ꢂC; (c)
1-methoxy-2-propanol, 140 ꢂC (d) neat, 100–120 ꢂC.
pIC50 PDE4 (U937 cells)
pKi HARBS
6.1
5.3
PDE1 (bovine brain)
PDE2 (human platelets)
PDE3 (human platelets)
PDE5 (human platelets)
PDE6 (bovine retina)
PDE7 (HuT-78 cells)
Inhibition of LPS-induced TNF-a
2% inhibitiona
7% inhibitiona
4% inhibitiona
6% inhibitiona
26% inhibitionb
31% inhibitionc
5.9
neutrophils). An interesting selectivity for the M3 recep-
tor toward the other muscarinic receptor subtypes (M2
in particular) is also noticed and has been confirmed
using the isolated guinea pig trachea or left atrium both
stimulated by carbachol as functionally integrated M3
and M2 assays, respectively. ucb-101333-3 has also been
screened against more than 40 different targets and no
significant interactions were observed at 10 lM.
release from human PMN (pIC50
)
Inhibition of fMLP-induced
elastase release from human PMN (pIC50
pD2 PDE4f
6.1
6.1
)
a Inhibition at 10 lM.
b Inhibition at 1 lM.
c Inhibition at 100 lM.
However, before pursuing the complete characteriza-
tion of this compound, a salt selection was performed
in order to avoid potential formulation issues. A nice,
nonhygroscopic powder has been obtained by making
a salt with fumaric acid. The stoichiometry (sesqui-
fumarate) as well as the protonated center on the aro-
matic nitrogen in 3-position have been unambiguously
assigned through X-ray diffraction. The crystallo-
graphic data are available in Cambridge Structural
Database.29
d On guinea pig trachea contracted with carbachol.
e On guinea pig paced left atrium stimulated with carbachol.
f On guinea pig trachea precontracted with LTD4.30
As a conclusion, we have identified, for the first time, a
new family of potent dual M3 antagonists and PDE4
inhibitors. Chemical modulations and SAR study

L. Provins et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1834–1839
1839
17. Provins, L.; Christophe, B.; Danhaive, P.; Dulieu, J.;
around the 4,6-diaminopyrimidine scaffold allowed us to
identify ucb-101333-3 which has been selected for exten-
sive DMPK and pharmacological evaluation on the ba-
sis of its excellent affinity/activity and selectivity profiles
as well as adequate physicochemical properties. Those
results as well as its characterization in relevant animal
models of COPD will be reported in due course.
´
Durieu, V.; Gillard, M.; Lebon, F.; Lengele, S.; Quere, L.;
van Keulen, B. J. Sci. Pharm. 2005, 73, S180.
´ ´
18. Gershon, H.; Braun, R.; Scala, A.; Rodin, R. J. Med.
Chem. 1964, 7, 808.
19. Biagi, G.; Giorgi, I.; Livi, O.; Scartoni, V.; Lucacchini, A.
Farmaco 1997, 52, 61.
20. Beck, J. P.; Arvanitis, A. G.; Curry, M. A.; Rescinito, J.
T.; Fitzgerald, L. W.; Gilligan, P. J.; Zaczek, R.; Trainor,
G. L. Bioorg. Med. Chem. Lett. 1999, 9, 967.
Acknowledgments
21. Chen, C.; Dagnino, R.; De Souza, E. B.; Grigoriadis, D.
E.; Huang, C. Q. J. Med. Chem. 1996, 39, 4358.
22. Chorvat, R. J.; Bakthavatchalam, R.; Beck, J. P.;
Gilligan, P. J.; Wilde, R. G., et al. J. Med. Chem.
1999, 42, 833.
´
The authors thank Geraldine Longfils for compounds’
crystallization, Bernadette Norberg (FUNDP, Namur)
for X-ray resolution, Liliane Ellens for technical support
23. Inhibition of PDE4 enzyme activity was measured in the
supernatant from U937 cells prestimulated with salbuta-
mol and rolipram to upregulate PDE4 expression. Enzyme
activity was measured using [3H]cAMP as a substrate.
24. Affinity for M3 muscarinic receptor was determined by
competition experiments with [3H]N-methylscopolamine
performed in CHO cell membranes expressing recombi-
nant human receptors.
´
in physicochemistry, and Veronique Pinilla and her
team for chiral chromatography.
References and notes
1. Barnes, P. J. Pharmacol. Rev. 2004, 56, 515.
2. Rand Sutherland, E.; Martin, R. J. J. Allergy Clin.
Immunol. 2003, 112, 819.
25. Capacity factors (k0IAM) were established on
IAM.PC.DD Drug-Discovery HPLC column
a
2
3. Buhl, R.; Farmer, S. G. Proc. Am. Thorac. Soc. 2005, 2,
83.
30 · 4.6 mm (Regis Tech Inc., Morton Grove, IL, USA).
The mobile phase consisted of different mixtures of
phosphate-buffered saline (pH 7.4) and acetonitrile as
co-solvent. The published data are the extrapolated values
at 0% in CH3CN. See also: Ong, S.; Liu, H.; Pidgeon, C.
J. Chromatogr. A 1996, 728, 113.
4. Barnes, P. J.; Hansel, T. T. Lancet 2004, 364, 985.
5. Donnelly, L. E.; Rogers, D. F. Drugs 2003, 63, 1973.
6. Krishna, G.; Sankaranarayanan, V.; Chitkara, R. K.
Expert Opin. Invest Drugs 2004, 13, 255.
7. Barnes, P. J. Chest 2000, 117, 10S.
8. Donohue, J. F. Chest 2004, 126, 125S.
9. Rennard, S. I. Lung Biol. Health Dis. 2000, 145, 159.
10. Panning, C. A.; DeBisschop, M. Pharmacotherapy 2003,
23, 183.
11. Hvizdos, K. M.; Goa, K. L. Drugs 2002, 62, 1195.
12. Houslay, M. D.; Schafer, P.; Zhang, K. Y. J. Drug
Discovery Today 2005, 10, 1503, and references cited
therein.
13. Odingo, J. O. Expert Opin. Ther. Patents 2005, 15, 773.
14. Lipworth, B. J. Lancet 2005, 365, 167.
15. Hatzelmann, A.; Schudt, C. J. Pharm. Exp. Ther. 2001,
297, 267.
26. Separation of the diastereoisomers has been performed via
chiral chromatography (Daicel Chiralpak AD column).
27. All titrations were performed using a GlpKaTM (Sirius,
Forest Row, UK) and done in aqueous solution or
methanol–water mixture with 0.15 M KCL at 25 ꢂC using
standardized 0.5 M HCL titrant.
28. CCDC 290973 contains the supplementary crystallograph-
ic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre
via http://www.ccdc.cam.ac.uk/data_request/cif.
29. CCDC 290974 contains the supplementary crystallograph-
ic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre
via http://www.ccdc.cam.ac.uk/data_request/cif.
30. Christophe, B.; Lelubre, F.; Provins, L.; Peck, M.;
Massingham, R. The Pharmacologist 2002, 44, A227.
16. Bundschuh, D. S.; Eltze, M.; Barsig, J.; Wollin, L.;
Hatzelmann, A.; Beume, R. J. Pharm. Exp. Ther. 2001,
297, 280.