Oxyguanidines: Application to non-peptidic phenyl-based thrombin inhibitors

Article abstract of DOI:10.1016/S0960-894X(03)00125-2

Although thrombin has been extensively researched with many examples of potent and selective inhibitors, the key characteristics of oral bioavailability and long half-life have been elusive. We report here a novel series non-peptidic phenyl-based, highly potent, highly selective and orally bioavailable thrombin inhibitors using oxyguanidines as guanidine-mimetics.

Full text of DOI:10.1016/S0960-894X(03)00125-2

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1495–1498
Oxyguanidines: Application to Non-peptidic Phenyl-Based
Thrombin Inhibitors
Bruce Tomczuk,* Tianbao Lu, Richard M. Soll,y Cynthia Fedde, Aihua Wang,
Larry Murphy, Carl Crysler, Malini Dasgupta, Stephen Eisennagel, John Spurlino
and Roger Bone
3-Dimensional Pharmaceuticals, Inc., 665 Stockton Drive, Exton, PA 19341 USA
Received 23 October 2002; accepted 13 January 2003
Abstract—Although thrombin has been extensively researched with many examples of potent and selective inhibitors, the key
characteristics of oral bioavailability and long half-life have been elusive. We report here a novel series non-peptidic phenyl-based,
highly potent, highly selective and orally bioavailable thrombin inhibitors using oxyguanidines as guanidine-mimetics.
# 2003 Elsevier Science Ltd. All rights reserved.
A key strategy in the development of novel anti-
thrombotic therapy has been directed towards the dis-
covery of small molecule inhibitors of the coagulation
cascade. One molecular target is thrombin (Factor
IIa), which is a serine protease which plays a pivotal
role in both fibrin generation penultimate to clot for-
mation and platelet activation.1 Although this target
has been extensively researched with many examples of
potent and selective inhibitors, the key characteristics
of oral bioavailability and long half-life have been
elusive. A first generation (intravenous only) active site
thrombin inhibitor, argatroban, has recently been
approved in the U.S. for HITT and HITTS.2 It is
likely that a second generation compound, such as the
double-prodrug ximelagatran (1) by AstraZeneca or a
compound from the pyridinone (2a) or pyrazinone
series (2b) by Merck, will have improved oral and
pharmacokinetic properties.3ꢀ5 We have developed a
series of non-peptidic phenyl-based thrombin inhibitors
which have addressed these elusive properties. These
inhibitors incorporate a novel oxyguanidine moiety,
which is a guanidine-mimetic of the RGD-type throm-
bin inhibitors.6 The oxyguanidine possesses a greatly
reduced pKa of approximately 7.0–7.5 compared to a
guanidine of 13–14.7 In the GI tract, a basic com-
pound of reduced pKa would result in a larger per-
centage of uncharged molecules with greater
propensity to cross cell membranes and enter the sys-
temic circulation. This positive distinguishing char-
acteristic will be exemplified by in vitro permeability
and pharmacokinetic studies.
a
These compounds were synthesized starting with the
monosulfonated orcinol 3,8 which undergoes two
sequential Mitsunobu reactions, first with a diol and the
second with N-hydroxyphthalimide.
The phthalimide 4 is deprotected with aqueous methyl-
amine to reveal the alkoxyamine 5 (Scheme 1). Sub-
sequent reaction with bis-Boc guanidinopyrazole gives a
protected alkoxy guanidine 6. Deprotection with TFA
provides the final products 7–29 as TFA salts. These
phenyl arylsulfonates have excellent hydrolytic stability
in buffers from pH 2 to 10 at 37 ꢁC for at least 24 h.
Tables 1 and 2 indicate the structure–activity relation-
ships in the aryl- and the heteroaryl-sulfonate function-
alities, respectively. The ortho-substituted derivatives
(7–12) were generally potent with the exception being
the anilino derivative 10. Focus was placed on the sul-
fones (11,12) due to their increased aqueous solubility
characteristic. The 3-methylsulfone 13 was an order of
magnitude less potent than 11. A combination of
methylsulfone groups at the ortho- and para-positions
(14) caused a loss of two orders of magnitude in
*Corresponding author. Tel.: +1-610-458-6058; fax: +1-610-458-
8258; e-mail: bruce.tomczuk@3dp.com
yPresent address: Ontogen Corporation, 6451 El Camino Real,
Carlsbad, CA 92009, USA.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00125-2

1496
B. Tomczuk et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1495–1498
potency. In the heteroarylsulfonates shown in Table 2,
the 8-substituted quinoline 15 appears to be consistently
more potent than the 4-substituted isoquinoline 16.
Reduction of 15 to the tetrahydroquinoline, which is
similar to that found in argatroban, actually reduced
potency. Interestingly, the tetrahydrobenzothiophene-
1,1-dioxide 18, which is a constrained ortho-sulfone
derivative, is less potent than 11 or 12.
poration of fluorine was detrimental with a single fluor-
ination (21) losing one order of magnitude and gem-
difluorination (22) another 3-fold loss. The rigidification
of the side chain with an sp2 center as in olefin 23 or the
addition of a hydrophilic hydrogen-bond donor group
as in 20 were not beneficial. However, the length of the
side-chain is critical as the one-carbon abrogation (24)
lost over 800-fold in potency.
Focusing on the ortho-methylsulfones, we varied the
oxyguanidine side-chain as shown in Table 3. These
variations centered on the 2-position of the 1,3-pro-
panedioxy side-chain. The only variation that did not
reduce potency was the cyclopropyl 19. The incor-
The central phenyl scaffold variations at the 3-position,
shown in Table 4, indicate that a small substituent is
preferential. Thus, even chloro 27 that is slightly larger
is less potent than 11. The larger groups such as meth-
oxy 25 and ethyl 26 were also less potent. This struc-
Scheme 1. Synthesis of Alkoxyguanidines.

B. Tomczuk et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1495–1498
1497
Table 1. Inhibition of a-human thrombin by substituted
arylsulfonates
Table 3. Inhibition of a-human thrombin by 2-methylsulfonyl-aryl-
sulfonates with side-chain variations
Compd
Thrombin inhibition
Ki, mMa
Compd
Thrombin inhibition
Ki, mMa
R
Q
7
8
9
10
11
12
13
14
2-Cl
2-CN
2-OCH3
0.0021 (ꢃ0.0001)
0.0099 (ꢃ0.0005)
0.0067 (ꢃ0.0001)
0.0866 (ꢃ0.0027)
0.0119 (ꢃ0.0003)
0.0048 (ꢃ0.0004)
0.204 (ꢃ0.009)
19
0.0077 (ꢃ0.0002)
20
21
22
23
24
CHOH
CHF
CF2
C=CH2
bond
0.137 (ꢃ0.005)
0.114 (ꢃ0.005)
0.364 (ꢃ0.031)
0.0324 (ꢃ0.0028)
2.62 (ꢃ0.17)
2-NH2
2-SO2CH3
2-SO2C6H5
3-SO2CH3
2,4-(SO2CH3)2
0.749 (ꢃ0.040)
aValues are means of three experiments, standard deviation is given in
parentheses.
aValues are means of three experiments, standard deviation is given in
parentheses.
Table 2. Inhibition of a-human thrombin by heteroarylsulfonates
Table 4. Inhibition of a-human thrombin by 3-substituted central
phenyl scaffold
Compd
Thrombin inhibition
Ki, mMa
Compd
Thrombin inhibition
Ki, mMa
15
16
0.0067 (ꢃ0.0005)
0.0132(ꢃ0.0008)
X
Q
25
26
27
–OCH3
–CH2CH3
–Cl
–CH2–
–CH2–
–CH2–
0.0346 (ꢃ0.0011)
0.0297 (ꢃ0.0008)
0.0394 (ꢃ0.0024)
28
29
–CH2OH
–F
0.403 (ꢃ0.010)
17
18
0.031(ꢃ0.0003)
0.102(ꢃ0.002)
0.0493 (ꢃ0.0001)
aValues are means of three experiments, standard deviation is given in
parentheses.
aValues are means of three experiments, standard deviation is given in
parentheses.
dino, and aminopyridine moieties. This is an important
characteristic because it would result in a higher propor-
tion of neutral species of the compound at the pH of the
lower intestine where absorption is likely to occur. This
higher absorption potential was realized when the per-
meability of compound 11 was quantitated in the
human Caco-2 monolayer assay.11 The apparent per-
meability coefficient (Papp) from the apical to baso-
lateral side was determined to be 2.4ꢂ10ꢀ6 cm/s. This
result would be considered in the medium range from
our laboratory.
ture–activity is consistent with that found in our pre-
vious studies with phenyl templated thrombin inhibi-
tors.9 The incorporation of a hydrophilic hydrogen-
bonding functionality as found in 28 resulted in a 800-
fold loss in potency.
These oxyguanidine phenyl-based thrombin inhibitors
retain the excellent selectivity against other serine pro-
teases that were observed in other previous work with
other phenyl-based thrombin inhibitors.10 A representa-
tive compilation of activities against other serine pro-
teases is shown in Table 5 for compound 11. The
oxyguanidine unit is a guanidine mimetic that possesses a
pKa of 7.05. This is the lowest pKa reported for any gua-
nidine mimetic that would include various amino, ami-
The in vitro metabolic potential of 11 in liver microsomes
from several species was studied as shown in Table 6.12
The disappearance of parent compound after incu-
bation with liver microsomes and NADPH was deter-
mined and the rate was calculated. It appears that
compound 11 has a low metabolic liability in dog
microsomes. These were distinct from the results with

1498
B. Tomczuk et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1495–1498
Table 5. In vitro inhibition of serine proteases by compound 11 (Ki (nM))
Thrombin
11.9
Fxa
Plasmin
Urokinase
67,000
Trypsin
Chymotrypsin
Elastase
>22,000
>22,000
>22,000
>22,000
>22,000
Table 6. Characterization of compound 11
pKa
Caco-2
Papp (ꢂ10ꢀ6 cm/s)
2.4
Microsomal stability (rate in pg/min/mg protein)
Dog
Rat
Mouse
Human
7.05
21.1
Propranolol 36.2
69.5
92.0
30.8
81.8
50.4
18.9
Wenxi Pan for the synthesis of compound 27 and Nor-
man Huebert for the pK re-analysis.
References and Notes
1. Chi, L.; Rogers, K. L.; Upritchard, A. C. G.; Gallagher, K.
Exp. Opin. Invest. Drugs 1997, 6, 1591.
2. Jeske, W.; Walenga, J.; Lewis, B. E.; Fareed, J. Exp. Opin.
Invest. Drugs 1999, 8, 625.
3. Gustafsson, D.; Nystrom, J.; Carlsson, S.; Bredberg, U.;
Eriksson, U.; Gyzander, E.; Elg, M.; Antonsson, T.; Hoff-
mann, K.; Ungell, A.; Sorensen, H.; Nagard, S.; Abra-
hamsson, A.; Bylund, R. Thrombosis Research 2001, 101, 171.
4. Sanderson, P. E. J.; Cutrona, K. J.; Dorsey, B. D.; Dyer,
D. L.; McDonough, C. M.; Naylor-Olsen, A. M.; Chen, I. W.;
Chen, Z.; Cook, J. J.; Gardell, S. J.; Krueger, J. A.; Lewis,
S. D.; Lin, J. H.; Lucas, B. J.; Lyle, E. A.; Lynch, J. J.; Stra-
nieri, M. T.; Vastag, K.; Shafer, J. A.; Vacca, J. P. Bioorg.
Med. Chem. Lett. 1998, 8, 817.
5. Sanderson, P. E. J.; Lyle, T. A.; Cutrona, K. J.; Dyer, D. L.;
Dorsey, B. D.; McDonough, C. M.; Naylor-Olsen, A. M.;
Chen, I. W.; Chen, Z.; Cook, J. J.; Cooper, C. M.; Gardell,
S. J.; Hare, T. R.; Krueger, J. A.; Lewis, S. D.; Lin, J. H.;
Lucas, B. J.; Lyle, E. A.; Lynch, J. J.; Stranieri, M. T.; Vastag,
K.; Yan, Y.; Shafer, J. A.; Vacca, J. P. J. Med. Chem. 1998,
41, 4466.
6. Von der Saal, W.; Heck, R.; Leinert, H.; Poll, T.; Stegme-
ier, K.; Michel, H., 1994, WO 942046.
7. Charton, M. J. Org. Chem. 1965, 30, 969.
8. Soll, R. M.; Lu, T.; Tomczuk, B.; Illig, C. R.; Fedde, C.;
Eisennagel, S.; Bone, R.; Murphy, L.; Spurlino, J.; Salemme,
F. R. Bioorg. Med. Chem. Lett. 2000, 10, 1.
9. Lu, T.; Soll, R. M.; Illig, C. R.; Bone, R.; Murphy, L.;
Spurlino, J.; Salemme, F. R.; Tomczuk, B. E. Bioorg. Med.
Chem. Lett. 2000, 10, 79.
Figure 1. In vivo pharmacokinetics in dog (n=3) after 6 mg/kg iv over
15 min and 30 mg/kg po of compound 11.
rat and human microsomes that have higher rates of
metabolism.
Confident that the series was permeable in a human cell
line and stable in microsomal metabolism in the dog,
compound 11 was administered as a solution at a dose
of 6 mg/kg intravenously and 30 mg/kg by oral gavage
to beagle dogs.13 The pharmacokinetic analysis indi-
cated that the terminal half-life after iv administration
was 7.3 h (Fig. 1). The calculated clearance rate was 6.9
mL/min/kg that was 1/5 the hepatic blood flow in dog.
This rate would be considered medium. The calculated
volumn of distribution was 4.4 L/kg that would be
considered high. After oral administration, the Cmax
reached 7.0 mM at 0.75 h. The dose-corrected oral
bioavailability (F) was 95%. These pharmacokinetic
parameters are among the best reported orally bioa-
vailable thrombin inhibitors.14
10. Lu, T.; Tomczuk, B.; Illig, C. R.; Bone, R.; Murphy, L.;
Spurlino, J.; Salemme, F. R.; Soll, R. M. Bioorg. Med. Chem.
Lett. 1998, 8, 1595.
11. Yamashita, S.; Konishi, K.; Yamazaki, Y.; Taki, Y.;
Sakane, T.; Sezaki, H.; Furuyama, Y. J. Pharm. Sci. 2002, 91,
669.
12. Obach, R. S.; Baxter, J. G.; Liston, T. E.; Silber, M.;
Jones, B. C.; MacIntire, F.; Rance, D. J.; Wastall, P. J. Phar-
macol. Exp. Ther. 1997, 283, 46.
13. The pharmacokinetic analysis was performed using the
Kinetica software package (Innaphase, Philadelphia, PA) in a
non-compartmental model.
We have discovered a novel series of phenyl-based
thrombin inhibitors using oxyguanidines as guanidine
mimetics. We believe that the key to the superior oral
pharmacokinetics is the use of the hydrophilic, but non-
basic oxyguanidine moiety.
Acknowledgements
We wish to thank Mike Kolpak for NMR spectra,
Harvey Fries for the Caco-2 permeability data, and
14. Sanderson, P. E. J. Annu. Rep. Med. Chem. 2001, 36, 79.