Design and synthesis of xanthine analogues as potent and selective PDE5 inhibitors

Article abstract of DOI:10.1016/S0960-894X(02)00646-7

We have discovered potent and selective xanthine PDE5 inhibitors. Compound 25 (PDE5 IC₅₀=0.6 nM, PDE6/PDE5=101) demonstrated similar functional efficacy and PK profile to Sildenafil (PDE5 IC₅₀=3.5 nM, PDE6/PDE5=7).

Full text of DOI:10.1016/S0960-894X(02)00646-7

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3149–3152
Design and Synthesis of Xanthine Analogues as Potent and
Selective PDE5 Inhibitors
Yuguang Wang,* Samuel Chackalamannil, Zhiyong Hu, Craig D. Boyle,
Claire M. Lankin, Yan Xia, Ruo Xu, Theodros Asberom, Dmitri Pissarnitski,
Andrew W. Stamford, William J. Greenlee, Jeffrey Skell, Stanley Kurowski,
Subbarao Vemulapalli, Jairam Palamanda, Madhu Chintala, Ping Wu,
Joyce Myersand Peng Wang
Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
Received 11 June 2002; accepted 25 July 2002
Abstract—We have discovered potent and selective xanthine PDE5 inhibitors. Compound 25 (PDE5 IC50=0.6 nM, PDE6/
PDE5=101) demonstrated similar functional efficacy and PK profile to Sildenafil (PDE5 IC50=3.5 nM, PDE6/PDE5=7).
# 2002 Elsevier Science Ltd. All rights reserved.
The development of phosphodiesterase type 5 (PDE5)
inhibitorsfor the treatment of male erectile dysfunction
(ED) hasattracted great attention since the commercial
introduction of Sildenafil (1, Viagra, Fig. 1).1 PDE5, a
cGMP hydrolytic enzyme, breaksdown cGMP in the
corpora cavernosa smooth muscle cells. Inhibition of
PDE5 has been demonstrated to increase the levels of
cGMP induced by nitric oxide (NO) from neuronal and
endothelial sources during sexual stimulation. The
increased cGMP levels, in turn, enhance the cavernosal
smooth muscle relaxation, arterial inflow and venous
congestion to achieve penile erection.2 Despite its com-
mercial success, Sildenafil has showed clinically sig-
nificant side effects, some of which could be related to
its low selectivity versus other PDEs such as PDE6. To
discover potent and more selective PDE5 inhibitors for
the treatment of male ED, we embarked on the design
and synthesis of xanthine analogues based on com-
pound 2 which wasidentified asa lead tsructure from
high throughput screening. Herein, we would like to report
the successful outcome of our efforts directed at discover-
ing potent and selective xanthine PDE5 inhibitors.
inhibitory activity and selectivity for this series of com-
pounds, the structure–activity relationship (SAR) stud-
iesof compound 2 were focused on chemical
modification at the C8, N7, N1 and N3 positions.
For the C8 modification, a variety of amino alcohols
that have differencesin their chirality, tseric hindrance
and aromatic character were tried. To further identify
the SAR role of the hydroxy group of the amino alco-
hols, two cycloamines were also included. These ana-
3
logueswere prepared ahsoswn in Scheme 1.
Commercially available compound 3 wasreacted with
benzyl bromide to form compound 4. The bromide
atom of compound 4 wasdpislaced with different
aminesto give the final targetsshown in Table 1.
Ashsown in Figure 1, xanthine analogue 2 showed
weak PDE5 inhibitory activity. To improve the PDE5
*Corresponding author. Tel.: +1-908-740-3508; fax: +1-908-740-
7152; e-mail: yuguang.wang@spcorp.com
Figure 1. Structuresof PDE5 inhibitors.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00646-7

3150
Y. Wang et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3149–3152
Scheme 1. C8 modification: (a) K2CO3, BnBr, DMF (90%); (b) NH2R, iPrNEt, NMP, sealed tube, 160 ꢀC, overnight, (30–75%).
Scheme 2. N1 and N3 Modification: (a) K2CO3, R2I or R2 Br, DMF (70–90%); (b) Pd(OH)2, HCONH4, MeOH (60–80%); (c) Br2, NaOAc, HOAc
(70–85%); (d) K2CO3, BnBr, DMF (80–90%); (e) NHR, iPrNEt, NMP, sealed tube, 160 ꢀC, overnight (75–95%).
Table 1. PDE5 inhibitory results of selected analogues of C8 modifi-
cation
The PDE5 inhibitory activity of the C8 modified
analogueswere determined according to the reported
procedures.4
Change of the chiral amino alcohol (2) to the achiral
amino alcohol (5) led to the decrease of the PDE5
inhibitory activity. Introduction of other chiral amino
alcohols( 6–10) also significantly reduced the PDE5
inhibitory activity. However, removal of the hydroxy
group increased the PDE5 inhibitory activity from 17%
(2) to >50% (12 and 13). In addition, these C8
replacements( 12 and 13) resulted in achiral compounds
that are synthetically easier to access.
Compd
R
(% Inhibition @ 100 nM)
17
2
5
10
Next, with the cyclohexyl and cyclopentyl aminesasthe
optimal C8 groups, we carried out N1 and N3 modifi-
cations. These compounds were prepared as shown in
Scheme 2.3
6
7
2
11
Alkylation of the N3 position of commercially available
14 afforded 15 that washydrogenated and brominated
to form 16. N7 benzylation of 16 produced 17. The
bromide atom of 17 wasdisplaced by the cyclohexyl or
cyclopentyl aminesto form the target compounds.
8
0
4
5
9
10
The results of PDE5 inhibitory activity of a selected
number of N1 and N3 modified analoguesare shown in
Table 2. The ethyl group at the N1 and N3 positions (18
and 21) showed the highest inhibitory activity.
11
12
13
0
62
54
Finally, we modified the N7 position, maintaining ethyl
groups at the N1 and N3 positions, while substituting
cyclohexyl amine and cyclopentyl amine at the C8 posi-
tion. The target compoundswere prepared asdescribed

Y. Wang et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3149–3152
3151
Table 2. PDE5 inhibitory results of selected analogues of N1 and N3
modification
in Scheme 1 and Scheme 2. The complete PDE5 inhibi-
tion and selectivity profile for the selected analogues
with C8 cyclohexyl amine isshown in Table 3.
Since rapid clearance of PDE5 inhibitorsisa desirable
property asillutsrated by Sildenafil, a metabolically
labile methoxy group wasintroduced to the para or the
meta position of the N7 benzyl ring. Compounds 24, 25,
Compd
Structure
R1, R2
(% Inhibition @ 100 nM)
and 29 aswell asthe halogen substituted analogues
31–
35 showed dramatic improvement of PDE5 inhibition
and selectivity. It is especially worth noting that the
halogen atoms increased PDE5 selectivity versus other
PDE isozymes significantly. These analogues demon-
strated superior PDE5 inhibition and selectivity over
Sildenafil (PDE5 IC50=3.5 nM, PDE6/PDE5=7). This
SAR trend wasalso observed for the analogueswith C8
cyclopentyl amine (Table 4).
12
18
19
A
A
A
R1R2=Me
R1R2=Et
R1=Me
62
91
49
R2=I-Bu
R1=Et
20
A
56
R2=I-Bu
R1=R2=Me
R1=R2=Et
R1=Me
R2=I-Bu
R1=Et
R2=I-Bu
13
21
22
B
B
B
54
93
52
Compound 25,5 the first compound to achieve 100-fold
PDE6/5 ratio, was tested in the functional efficacy assay
and the pharmacokinetic assay. In the electrical fre-
quency stimulated (EFS)-induced relaxation of rabbit
corpus cavernosum test,6 compound 25 demonstrated
efficacy similar to that of Sildenafil (Fig. 2).
23
B
63
Table 3. PDE5 inhibition and selectivity profile (C8 cyclohexyl ana-
logues)
Compd
R1
R2
PDE5 IC50
(nM)
PDE6/
PDE5
PDE 1–4/
PDE5
21
24
25
26
27
28
29
30
31
32
33
34
35
H
H
H
OMe
OH
Me
Me
H
OMe
OH
H
H
OMe
OH
12
2.6
0.6
100
1.2
19
4.5
3.7
0.58
2.0
0.30
0.33
1.7
35
51
101
na
36
58
133
14
345
250
350
364
219
naa
na
>3000
na
na
na
na
na
>9500
–OCH2O—
Cl
Cl
F
F
Br
OMe
OH
OMe
OH
OMe
>28,000
>12,000
>10,000
Figure 2. EFS-induced relaxation of rabbit corpusCavernosum.
ana: Not available.
Table 4. PDE5 binding and selectivity profile (C8 cyclopentyl
analogues)
Compd R1 R2 PDE5 IC50 (nM) PDE6/PDE5 PDE 1–4/PDE5
18
36
37
38
39
40
41
H
H
H
H
OMe
OH
10
2.1
1.5
1.1
1.8
0.69
0.82
21
33
46
118
261
89
naa
na
na
na
na
na
>4,400
Cl OMe
Cl OH
F
F
OMe
OH
182
Figure 3. Rapid rat pharmacokinetics.

3152
Y. Wang et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3149–3152
The oral PK profile of compound 25 wasalso examined
in a rapid rat PK assay.7 The result is shown in Figure
3. Compound 25 wasrapidly absorbed and cleared.
Egan, R. W.; Billah, M. M. Life Sci. 2001, 68, 1997). Recom-
binant human PDE1A3 was assayed at the final concentration
of cAMP of 10 mM. Recombinant human PDE1A3 was
assayed at the final concentration of cAMP of 10 mM.
Recombinant human PDE4B2 was assayed at the final con-
centration of cAMP of 1 mM. Human PDE6 isolated from
retina was assayed at the final concentration of cGMP of
0.5mM. All the PDE assays were performed in the presence of
2% DMSO and 0.1% BSA. All the assays were done in
duplicate, and data shown are means from 2–4 experiments
(SEM <15%).
5. Analytical data for compound 25: 1H NMR (400 Hz,
CDCl3): d 0.87 (m, 1H), 1.02–1.34 (m, 10H), 1.57 (m, 4H),
1.90 (d, J=10.4 Hz, 2H), 3.71 (m, 1H), 3.94 (d, J=7.6 Hz,
1H), 4.07 (m, 4H), 5.21 (s, 2H), 6.73 (d, J=8.4 Hz, 2H), 7.03
(d, J=8.4 Hz, 2H). EI–MS: m/z 412 (M+1). HRMS calcd for
C22H30N5O3: 412.2349, found: 412.2332.
6. The corpus cavernosum from anesthetized male New Zeal-
and rabbitswasmounted in organ bath chamberscontaining
physiological salt solution. The optimum length-tension curves
were obtained by gradual stretching and contracting the tis-
sues with phenylephrine 10 mM. The tissues were equilibrated
in a solution containing atropine (1 mM), guanethideine (5
mM) and indomethacin (5 mM). After the tissues were equili-
brated, they were subjected to electrical field stimulation (10
V, 0.5 ms pulse duration) at 0.5–16 Hz. The tissues were then
incubated with vehicle, or test compound (30 nM) for 30 min
contracted with phenylephrine. The electrical stimulation was
repeated. Relaxation wascalculated asa percentage of phen-
ylephrine-induced contraction. For details, see: Vemulapalli,
S.; Kurowski, S. Fundamental & Clinical Pharmacol. 2001, 15,
1.
In summary, we have discovered potent and selective
PDE5 inhibitors. Compounds such as 25 have demon-
strated good efficacy in the EFS induced relaxation
assay. Compound 25 also showed a desirable oral PK
profile with rapid onset and fast clearance. Further
development of thisseriesof compoundswill be repor-
ted in the future.
Acknowledgements
We thank DrsJohn Clader, John Hunter, John Piwinski
and Catherine Strader for helpful discussions and their
advice.
References and Notes
1. Bi, Y.; Stoy, P.; Adam, L.; He, B.; Krupinski, J.; Norman-
din, D.; Pongrac, R.; Seliger, L.; Watson, A.; Macor, J. E.
Bioorg. Med. Chem. Lett. 2001, 11, 2461 and referencescited
there .
2. (a) Corbin, J. D.; Francis, S. H. J. Biol. Chem. 1999, 274,
13729. (b) Gibson, A. Eur. J. Pharmacol. 2000, 411, 1. (c)
Langtry, H. D.; Markham, A. Drugs 1999, 57, 967.
3. All of the target compoundswere purified by preparative
TLC and showed satisfactory results in the analysis by NMR,
MS and HRMS.
7. Compound 25 wasformulated in PG/TW80 due to itslow
water solubility. For the reference of the rapid rat assay, see: Cox,
K. A.; Dunn-Meynell, K.; Korfmacher, W. A.; Broske, L.;
Nomeir, A. A.; Lin, C. C.; Cayen, M. N.; Barr, W. H. A. Drug
Discov. Today 1999, 4, 232.
4. Human PDEs 5 and 2 isolated from corpus cavernosum
and PDE3 isolated from platelets were assayed as described
previously (Wang, P.; Wu, P.; Myers, J. G.; Stamford, A.;