Azaindoles: Moderately basic P1 groups for enhancing the selectivity of thrombin inhibitors

Article abstract of DOI:10.1016/S0960-894X(03)00017-9

Starting from a 2-amino-6-methylpyridine P1 group and following a strategy of enlarging it whilst reducing its polarity, we have developed a series of potent, moderately basic azaindoles which are intrinsically much more selective for thrombin versus tryp

Full text of DOI:10.1016/S0960-894X(03)00017-9

Bioorganic & Medicinal Chemistry Letters 13 (2003) 795–798
Azaindoles: Moderately Basic P1 Groups for Enhancing the
Selectivity of Thrombin Inhibitors
Philip E. J. Sanderson,a,* Matthew G. Stanton,a Bruce D. Dorsey,a Terry A. Lyle,a
Colleen McDonough,a William M. Sanders,a Kelly L. Savage,a Adel M. Naylor-Olsen,b
Julie A. Krueger,c S. Dale Lewis,c Bobby J. Lucas,c Joseph J. Lynchd and Youwei Yane
aDepartment of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
bDepartment of Molecular Systems, Merck Research Laboratories, West Point, PA 19486, USA
cDepartment of Biological Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
dDepartment of Pharmacology, Merck Research Laboratories, West Point, PA 19486, USA
eDepartment of Structural Biology, Merck Research Laboratories, West Point, PA 19486, USA
Received 24 October 2002; revised 3 January 2003; accepted 7 January 2003
Abstract—Starting from a 2-amino-6-methylpyridine P1 group and following a strategy of enlarging it whilst reducing its polarity,
we have developed a series of potent, moderately basic azaindoles which are intrinsically much more selective for thrombin versus
trypsin. Certain pyrazinone acetamide azaindole derivatives have pharmacokinetic parameters after oral administration to dogs, or
efficacy in vitro, comparable to an optimized pyrazinone acetamide 2-amino-6-methylpyridine derivative.
# 2003 Elsevier Science Ltd. All rights reserved.
The development of an orally active thrombin inhibitor
as an anticoagulant is a well established medicinal
chemistry goal.1 Selectivity versus other serine protease
inhibitors, particularly those which share thrombin’s
preference for P1 groups containing strong bases such
as guanidines and amidines, is a critical hurdle to over-
come in the design of such a compound.1a Trypsin is
arguably the most demanding serine protease in this
regard owing to the similarity of their active sites and its
location in the gut.
affinity of an inhibitor for thrombin until the size limit
of thrombin’s ‘specificity pocket’ is reached.
Strongly basic P1 groups will tend to retard absorption
across the gut wall and may be associated with unwan-
ted side effects.4 A reduction in the basicity of the P1
group has been used in concert with the selectivity
increasingstrategy mentioned above to overcome these
two liabilities. For example, the 2-amino-6-methylpyr-
idine P1 group (conjugate acid pKa=6.9) was devel-
oped with a pyrazinone acetamide P2/P3 template to
give 1, a selective, orally active thrombin inhibitor.5,6
A focus in the design of selective thrombin inhibitors
has been on the ‘specificity pocket’ (S1). Thrombin’s
‘specificity pocket’ is identical to that of trypsin except
at residue 190 which is alanine and serine, respectively.
Consequently the pocket is slightly larger and more
lipophilic in thrombin than it is in trypsin.2 This fact has
guided the design of inhibitors selective for thrombin
and a general design strategy which involves addition of
hydrophobic surface area to, and subtraction of hydro-
philic groups from, a P1 group is likely, but not guar-
anteed,3 to have the effect of increasingthe relative
For a second generation compound we recognized the
advantages inherent in a structure like 1, with the
exception that it should have an intrinsically more
selective P1 group. To this end we chose to investigate
bicyclic 6/5 heteroaromatic P1 groups which, in accor-
dance with the design strategy outlined above, are larger
and have fewer hydrogen bond donors and acceptors
than the 2-amino-6-methylpyridine. First we looked at
*Correspondingauthor. Tel.:+1-215-6524-957; fax:+1-215-6523-971;
e-mail: phil_sanderson@merck.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00017-9

796
P. E. J. Sanderson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 795–798
the imidazopyridine derivative 2 of aminopyridine 1
(Table 1).7 The affinity for thrombin and trypsin both
dropped and the selectivity actually diminished slightly.
Deletion of the methyl group (compound 3) resulted in
a further 12- to 15-fold drop in affinities for thrombin
and trypsin. The isomeric imidazopyridine 4 was no
better. On the other hand, the affinity for thrombin of
simple 5-linked indole 5a was a significant improvement
over imidazopyridine 3. Moreover, it was completely
inactive (>100 mM) against trypsin.8,9 The orientation
of the indole appears to be important since the isomeric
6-linked indole 6 is distinctly less active than 5a.10
Interestingly, benzimidazole 7 is slightly more active
against thrombin than 5a but its selectivity is eroded.11
such that the 2-position is essentially equidistant (2.8–
2.9 A) to the two oxygens of the carboxyl side chain of
Asp-189. A water molecule forms a hydrogen bond
bridge from the indole NH to Phe-227. The indole 7-
position is 3.2 A from the side chain of Val-213. It
should be noted that this bindingconformation is quite
different from that of the 3-chlorophenyl P1 groups in
which the chlorine atom makes contact with the phenol
side chain of Tyr-228.15 The rest of the molecule binds
in the expected fashion in the P2/P3 region and is simi-
lar to the reported structure of compound 1.5a
The indole groups of compounds 5a and b are chemi-
cally labile. To stabilize the indole and to introduce a
basic group we then prepared the three possible azain-
dole isomers 8, 9, and 10.8 The synthesis of 8 and 9 fol-
lowed our usual practice of couplingthe preformed P1
amine to the preformed P2/P3 carboxylic acid under
standard, EDC conditions.5a,13 The synthesis of the 4-
azaindole P1 amine starts from commercially available
6-methyl-5-nitropyridin-2-ol 11 (Scheme 1). Conversion
to the nitrile 12 in two steps followed by a Batcho-
Leimgruber procedure with concomitant reduction of
the nitrile gave the 5-aminomethyl-4-azaindole 13. 5-
Aminomethyl-6-azaindole was prepared from commer-
cially available 4-methyl-5-nitropyridin-2-ol usingthe
same reaction sequence. The synthesis of the 7-azain-
doles starts from 7-aza-5-bromoindoline 14.16 Conver-
sion of 14 to 5-aminomethyl-7-azaindoline 15 followed
by couplingwith the P2/P3 acid and oxidation gave 10.
Improvement in the affinity of indole 5a for thrombin
was possible by introducinga 2-pyridin-2-ylethyl P3
group to give compound 5b (Ki=1.6 nM).13 Notably, 5b
is over five orders of magnitude more potent for
thrombin than trypsin. The crystal structure of 5b
bound in thrombin’s active site was determined and it is
shown in Figure 1.14 The indole binds in a conformation
Table 1. Inhibition constants for compounds 2–1012
Compd
X
Ar
Thrombin Ki (nM) Trypsin Ki (nM)
2
CH
11
6800
3
4
CH
CH
140
270
100,000
120,000
5a
5b
CH
N
18
1.6
>100,000
290,000
6
7
CH
CH
830
11
>100,000
Figure 1. Crystal structure of compound 5b bound in thrombin’s
active site.
2000
8a
8b
CH
N
11
3.2
>100,000
>1,000,000
9a
9b
CH
N
3.7
1.2
95,000
200,000
10a
10b
CH
N
18
6.6
>100,000
140,000
Scheme 1. Synthesis of 10 and amine 13:8 (a) POBr3, (CHCl2)2, reflux,
4 h; (b) Zn(CN)2, (Ph3P)4Pd, DMF, 80 ꢀC, 5 h; (c) DMF dimethyl
acetal, DMF, 90 ꢀC, 2 h; (d) 10% Pd/C, MeOH/6M HCl, 16 h; (e)
carboxylic acid,5a,13 EDC, HOBT, DMF, 4 h; (f) MnO2, DMF, 2 h.

P. E. J. Sanderson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 795–798
797
Table 2. Final assay concentration of compound required to double
the human activated partial thromboplastin time (2ÂAPTT)12 and
pharmacokinetic parameters in dogs after oral administration at 1mg/
kg
2. Chen, Z.; Li, Y.; Mulichak, A. M.; Lewis, S. D.; Shafer,
J. A. Arch. Biochem. Biophys. 1995, 322, 198.
3. Paradoxically, addition of an ortho chloro substituent to
arylamidine serine protease inhibitors was shown to increase
selectivity for Ser-190 proteases over Ala-190 proteases. This
was ascribed to the loss of a hydrogen bond from an ordered
water molecule to the amidine which could be compensated by
the side chain of Ser-190 but not by that of Ala-190: Mack-
man, R. L.; Katz, B. A.; Breitenbucher, J. G.; Hui, H. C.;
Verner, E.; Luong, C.; Liu, L.; Sprengeler, P. A. J. Med.
Chem. 2001, 44, 3856. Katz, B. A.; Sprengeler, P. A.; Luong,
C.; Verner, E.; Elrod, K.; Kirtley, M.; Janc, J.; Spencer, J. R.;
Breitenbucher, J. G.; Hui, H.; McGee, D.; Allen, D.; Martelli,
A.; Mackman, R. L. Chem. Biol. 2001, 8, 1107.
Compd 2ÂAPTT mM Dog Cmax (mM) t1=2 (h) AUC (mM h)
1
0.41
0.80
2.63
0.40
0.95
5.37
blqa
2.69
0.18
2.10
3.18
—
3.60
ndb
2.42
23.1
—
15.2
0.17
9.1
8b
9a
9b
10b
aBelow the level of quantitation.
bNot determined due to insufficient data.
4. Kikumoto, R.; Tamao, Y.; Ohkubo, K.; Tezuka, T.;
Tonomura, S.; Okamoto, S.; Hijikata, A. J. Med. Chem. 1980,
23, 1293. Kaiser, B.; Hauptmann, J.; Weiss, A.; Markwardt,
F. Biomed. Biochim. Acta 1985, 44, 1204. Reers, M.;
Koschinsky, R.; Dickneite, G.; Hoffmann, D.; Czech, J.; Stu-
ber, W. J. Enzyme Inhib. 1995, 9, 61.
5. (a) 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. (b) 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.; Stranieri, M. T.; Vastag, K.; Shafer, J. A.; Vacca, J. P.
Bioorg. Med. Chem. Lett. 1998, 8, 817.
The affinity of the azaindole derivatives for thrombin
correlates with the acidity of the conjugate acids of the
parent azaindoles.17 Thus, while 7-azaindole derivative
10a is equipotent to indole 5a, 4-azaindole 8a is slightly
more potent and 6-azaindole 9a is the most potent
(Ki=3.7 nM). Satisfyingly, despite the addition of a
polar functional group, all three azaindoles remained
highly selective. To improve the activity of the com-
pounds, the corresponding3-[(2-pyridin-2-ylethyl]ami-
no)pyrazinones 8b, 9b, and 10b were made and in each
case a roughly three-fold improvement in the Ki was
8
seen alongwith better selectivity versus trypsin.
The potent azaindoles were examined in more depth for
their efficacy in vitro (2ÂAPTT) and for their pharma-
cokinetic parameters after oral administration in dogs
(Table 2). Compound 9a was well absorbed in dogs and
was comparable to compound 1.5a However, it was
poorly efficacious in vitro. On the other hand pyridine
9b was as efficacious as 1 in vitro but was absorbed very
poorly. 7-Azaindole 10b had intermediate properties
with moderate efficacy and oral absorption, while
4-azaindole 8b was moderately efficacious but was not
absorbed in dogs.
6. It should be noted that the absolute affinity of 1 for trypsin
is adequate (Ki=1.8 mM) given its solubility (1.6 mg/mL 10
mM citrate buffer, pH 7.0, 23 ꢀC).
7. Imidazopyridines 2–4 were prepared by treatment of the
correspondingaminopyridine with bromoacetaldehyde and
potassium acetate in water/acetic acid/DMF, see: Sanderson,
P. E., Naylor-Olsen, A. M. US Patent 6,093,717, 2000.
8. For the synthesis of compounds 5 and 7–10 see: Sanderson,
P. E.; Lyle, T. A.; Dorsey, B. D., Stanton, M. G., Staas, D.,
Coburn, C., Naylor-Olsen, A. M., Morrissette, M. M., Sel-
nick, H. G., Nantermet, P. G., Williams, P. D., Stauffer, K. J.,
Burgey, C., Isaacs, R. Barrow, J. C. WO2000026211, 2000.
5-Aminomethylindole and 5-aminomethylbenzimidazole are
commercially available.
9. Similar indoles, indazoles, benzimidazoles and benzo-
triazoles havinga single basic center with p Ka greater than 6
have been claimed as antithrombotic agents: Blagg, J.; Brown,
A. D.; Gautier, E. C. L.; Smith, J. D.; McElroy, A. B. US
Patent 6,180,627, 2001.
In conclusion, we have developed a series of moderately
basic azaindole P1 groups which are intrinsically much
more selective than the aminopyridine P1 group of
compound 1. Certain of these azaindole derivatives
have pharmacokinetic parameters after oral adminis-
tration to dogs, or efficacy in vitro, comparable to
compound 1. The challenge remains to find a compound
which combines these latter properties in the same
molecule.
10. For the synthesis of compound 6 see: Sanderson, P. E.;
Lyle, T.; Dorsey, B.; Stanton, M.; Naylor-Olsen, A. M. US
Patent 6,376,499, 2002. 6-Aminomethylindole is commercially
available.
11. Amides of benzimidazolylalanine have been claimed to
inhibit thrombin induced coagulation: Heckel, A.; Sauter, R.;
Psiorz, M; Binder, K.; Mueller, T.; Zimmermann, R. US
Patent 5,391,556, 1995.
Acknowledgements
We thank Orn Almarsson for performingthe p Ka and
solubility determinations for compound 1 and Dan
McMasters for his help with the figures.
12. Lewis, S. D.; Ng, A. S.; Baldwin, J. J.; Fusetani, N.;
Naylor, A. M.; Shafer, J. A. Thromb. Res. 1993, 70, 173. All
the Ki data quoted in this report are for the human enzymes.
13. The 2-pyridin-2-ylethyl side chain was originally devel-
oped in the aminopyridine P1 series: Sanderson, P. E. J., Lyle,
T. A., Dorsey, B. D., Varsolona, R. J. US Patent 5,866,573,
1999. For its characterization in the aminopyridine series see:
Burgey, C. S.; Robinson, K. A.; Lyle, T. A.; Sanderson, P. E.
J.; Lewis, S. D.; Lucas, B. J.; Krueger, J. A.; Singh, R.; Miller-
Stein, C.; White, R. B.; Wong, B.; Lyle, E. A.; Williams, P. D.;
References and Notes
1. (a) Wiley, M. R.; Fisher, M. J. Exp. Opin. Ther. Patents
1997, 7, 1265. (b) Coburn, C. A. Exp. Opin. Ther. Patents
2001, 11, 1.

798
P. E. J. Sanderson et al. / Bioorg. Med. Chem. Lett. 13 (2003) 795–798
Coburn, C. A.; Dorsey, B. D.; Stranieri, M. T.; Holahan, M.
A.; Sitko, G. R.; Cook, J. J.; McMasters, D. R.; McDonough,
C. M.; Sanders, W. M.; Wallace, A. A.; Clayton, F. C.; Bohn,
D.; Leonard, Y. M.; Detwiler, T. J. Jr.; Lynch, J. J. Jr.; Yan,
Y.; Chen, Z.; Kuo, L.; Gardell, S. J.; Shafer, J. A.; Vacca J. P.
J. Med. Chem. In press.
14. The crystals of 5b bound to the a-thrombin-hirugen com-
plex were prepared and the diffraction data collected as pre-
viously described: Lyle, T. A.; Chen, Z.; Appleby, S. D.;
Freidinger, R. M.; Gardell, S. J.; Lewis, S. D.; Li, Y.; Lyle,
E. A.; Lynch, J. J.; Mulichak, A. M.; Ng, A. S.; Naylor-Olsen,
A. M.; Sanders, W. M. Bioorg. Med. Chem. Lett. 1997, 7, 67.
The diffraction data was collected to 2.2 A resolution with
Rsym of 0.061 and completeness of 98%. The structure was
determined usingthe difference Fourier method and was
refined to an R factor of 0.185.
15. (a) Tucker, T. J.; Brady, S. F.; Lumma, W. C.; Lewis,
S. D.; Gardell, S. J.; Naylor-Olsen, A. M.; Yan, Y.; Sisko,
J. T.; Stauffer, K. J.; Lucas, B. J.; Lynch, J. J.; Cook, J. J.;
Stranieri, M. T.; Holahan, M. A.; Lyle, E. A.; Baskin, E. P.;
Chen, I.-W.; Dancheck, K. B.; Krueger, J. A.; Cooper, C. M.;
Vacca, J. P. J. Med. Chem. 1998, 41, 3210. (b) Sanderson, P.
E. J.; Cutrona, K. J.; Dyer, D. L.; Krueger, J. A.; Kuo, L. C.;
Lewis, S. D.; Lucas, B. J.; Yan, Y. Bioorg. Med. Chem. Lett.
In press.
16. (a) Taylor, E. C.; Macor, J. E.; Pont, J. L. Tetrahedron
1987, 43, 5145. (b) Frissen, A. E.; Marcelis, A. T. M.; Van Der
Plas, H. C. Tetrahedron 1989, 45, 803.
17. 7-Azaindole: pKa=4.59; 4-azaindole: pKa=6.94; 6-azain-
dole: pKa=7.95 (all measured at 20 ꢀC in water), see:
Katritzky, A. E. Physical Methods in Heterocyclic Chemistry;
Academic Press: 1963; Vol. 1, p 96.