SAR of N-phenyl piperidine based oral integrin α5β1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2009.11.030

Recently, a new class of selective integrin α5β1inhibitors consisting of a heterocyclic based scaffold was published. Herein the SAR and pharmacokinetic profiles of N-phenyl piperidine derivatives are described.

Full text of DOI:10.1016/j.bmcl.2009.11.030

Bioorganic & Medicinal Chemistry Letters 20 (2010) 65–68
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
SAR of N-phenyl piperidine based oral integrin
a5b1 antagonists
Gunther Zischinsky, Frank Osterkamp, Doerte Vossmeyer, Grit Zahn, Dirk Scharn,
*
Ariane Zwintscher, Roland Stragies
Jerini AG, Invalidenstraße 130, Berlin D-10115, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Recently, a new class of selective integrin a5b1inhibitors consisting of a heterocyclic based scaffold was
published. Herein the SAR and pharmacokinetic profiles of N-phenyl piperidine derivatives are described.
Received 28 September 2009
Revised 6 November 2009
Accepted 10 November 2009
Available online 14 November 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Integrin a5b1
Antagonists
N-Phenyl piperidine
Stereoisomers
Pharmacokinetic profile
Integrins are a family of transmembrane adhesion receptors
that mediate cell contact to neighboring cells and proteins of the
extracellular matrix.1 They have been described to be involved in
several pathological conditions including inflammatory diseases,
autoimmune diseases, tumor progression and metastasis by regu-
lating leukocyte trafficking, cell proliferation, migration, angiogen-
esis and lymphangiogenesis.2–5 The therapeutic efficacy of integrin
Compounds 3a–i were synthesized according to the convergent
synthesis strategy shown in Schemes 1–3. The three piperdines
6a–c were obtained from alkylation of intermediate 513 and subse-
quent deprotection of the piperidine nitrogen atom (Scheme 1).
The boronic acid derivatives 8 and 10 were synthesized from the
corresponding commercially available starting materials (Scheme 2).
The key building blocks 11a–c and (S)-11a were obtained by a
copper mediated coupling reaction14 of 6a–c with either the race-
mic boronic acid 815 or the optically pure potassium trifluoro bo-
rate 10. After liberation of the primary amines, amide formation
with different benzoic acids and saponification of the methyl ester
gave the desired integrin antagonists 3a–i.
antagonists like cilengitide (avb3) and natalizumab (a4b1) has
been shown for the treatment of glioblastoma and multiple sclero-
sis, respectively.6,7
Knockout studies of the fibronectin receptor
tumor angiogenesis studies with 5b1 integrin inhibitors have
revealed a role in blood vessel formation for this integrin.8,9
Furthermore, systemic treatment with an 5b1 integrin small mol-
ecule antagonist or an 5b1 integrin antibody reduced tumor
growth in proof-of-concept studies for different tumor types.10,11
Integrin 5b1 antagonists are therefore promising drug candidates
a5b1 integrin and
a
Previously, the first piperidine based integrin
3b was reported13 and was shown to have excellent integrin
5b1 affinity (2.3 nM). Selectivity against integrin vb3 was med-
ium, whereas selectivities against integrins vb5 and IIbb3 were
a5b1 antagonist
a
a
a
a
a
a
a
excellent (Table 1). Compound 3b showed high total body clear-
for the therapy of diseases that involve angiogenic and prolifera-
ance (CL = 70 ml/min/kg) after iv administration in rat. Therefore
tive processes.
selectivity against integrin
should be improved.
avb3 and the pharmacokinetic profile
Recently, the discovery of integrin
a5b1 antagonists with low
nanomolar affinities based on heterocyclic core structures (2) de-
rived from a 3-hydroxypyrrolidine scaffold like in 112 was reported
(Fig. 1).13 Herein, the SAR and optimization of one of these new
scaffolds with an N-phenyl piperidine core is disclosed.
As previous reported,12 the electron accepting or donating nat-
ure of substituents in the 4-position of the 2-aminopyridine has a
great influence on integrin binding (Table 1). As expected, the elec-
tron donating 4-OMe derivative 3c, had the highest integrin a5b1
binding activity in the ELISA and the cellular binding assay. Other-
wise, the 4-H and 4-Me derivatives 3a and 3b showed similar IC50
values in the ELISA, whereas the cellular IC50 value is of 3c is signif-
icantly lower. However, as the cellular binding assay represents the
more functional assay it was considered to be more relevant one.12
* Corresponding author. Correspondence should be addressed to: Pericles Calias,
Shire HGT, 125 Spring Street Lexington, MA 02421, USA. Tel.: +1 781 482 0701; fax:
+1 781 482 2961.
E-mail address: pcalias@shire.com (R. Stragies).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.11.030

66
G. Zischinsky et al. / Bioorg. Med. Chem. Lett. 20 (2010) 65–68
O
R2
NH
OH
O
O
O
NH
O
NH
OH
N
O
N
H
NH
O
HO
O
N
O
N
R1
NH
NH
1
2
3
N
N
Figure 1. Generic structure 3 with an N-phenyl piperidine core was discovered during the optimization of the new integrin
3-hydroxypyrrolidine lead 1. The dotted cycle in 2 represents heterocyclic five- or six-membered ring systems.
a5b1 antagonists type 2 which were based on the
Table 1
In contrast to the increasing integrin
a5b1 binding affinity, an
IC50 values of 3a–c for integrins
binding assays
a
5b1,
avb3,
avb5,
a
IIbb3 in competitive integrin
opposing trend was found for oral bioavailability and the in vitro
permeability (Table 2). This observation can be rationalized by
the different pKa values caused by the increasing electron donating
effect of the 4-H < 4-Me < 4-OMe substituents in 3a–c. The differ-
ent fractions of unprotonated base at physiological pH may explain
the observed in vitro permeabilities as well as the oral bioavailabil-
ities of 3a–c.
R
N
N
H
N
HN
O
OH
Furthermore, the same trend was found for the systemic clear-
ance after iv administration and in the microsomal stability (MS
rat). In favor of better pharmacokinetic properties, the unsubstitut-
ed 2-aminopyridine of derivative 3a was therefore selected and
kept fixed for further optimization.
O
Compds
R
IC50a, nM
vb5
a
5b1
avb3
a
a
IIbb3
cell
3a
3b
3c
H
Me
OMe
2.1
2.3
0.92
642
84
89
15,000
10,700
3540
40,200
23,400
34,400
198
69
34
a
Values are means of three experiments.
R1
OTos
R1
a,b
It is well known that
itors can gain selectivity against
the unsubstituted
-benzoylamide to mesitoylamide.12,16–18
As previously discussed, the selectivity for integrin vb3 for the
N-phenyl piperidine scaffold had to be improved. For optimization
the 2,6-dimethyl derivative 3d was selected as starting point (Ta-
ble 3). To increase selectivity a more bulky ethyl group was incor-
porated (3e). But both derivatives showed only medium stability to
human microsomal degradation. To overcome this issue a 4-fluoro
atom was introduced (3f–g), improving the microsomal stability.
Specifically, the 2-ethyl-4-fluoro-6-methyl compound 3g showed
a
-amino acid derived integrin
a
5b1 inhib-
N
N
N
N
NHBoc
Boc
Cbz
NH
av-integrins by switching from
6a: R1 = H
4a-c
5
a
6b: R1 = Me
a
1
6c
: R = OMe
Scheme 1. Reagents and conditions: (a) KHMDS, DMSO, 0 °C; (b) Pd/C, H2, ethyl
acetate, rt.
OH
B
OH
B
HO
HO
NH2
CO2H
HO
NHBoc
CO2Me
good integrin
a5b1 affinities in both the ELISA (2.0 nM) and the
a, b
cellular assay (76 nM) in combination with excellent selectivities
and human microsomal stability.
7
9
8
Analysis of the various stereo isomers of 3g (Table 4), indicates
that stereo chemistry has a profound impact on the activity as well
as the selectivity toward other integrins.
The (S)-enantiomer exhibited the highest binding activity and
selectivity. The (R)-enantiomer was found to be less potent than
the (S)-enantiomer or the racemic mixture and the relative selec-
tivity toward other integrins dropped by two orders of magnitude.
As previously described, the 2,6-disubstituted benzamide moi-
KF3B
NHBoc
CO2Me
NHBoc
CO2Me
c, d, e
10
Scheme 2. Reagents and conditions: (a) HCl, MeOH, reflux; (b) Boc2O, NaHCO3,
tBuOH; (c) Tf2O, pyridine, DCM, ꢀ30 °C to rt; (d) PdCl2(dppf), bis(pinacolato)dibo-
rane, KOAc, DMSO, 80 °C; (e) KHF2, H2O/MeOH, ꢀ5 °C.
ety represents the key for selectivity of integrin a5b1 antagonists
toward other integrins. To elucidate this, derivative 3i with a plane
isoindolin-1-one was synthesized and the integrin activities were
compared with the more flexible benzamides 3h and 3g in the
same position (Table 5).
R1
a, b
*3HCl
N
N
c, d
The increasing activities for integrin
decrease of integrin vb3 activities, which resulted in high integrin
5b1 selectivities. The calculated torsion angle (Fig. 2) between
the amide and phenyl planes correlates positively with these activi-
ties. On the other hand, no direct correlation of the integrin IIbb3
a5b1 are associated with a
6a-c
3a-h
H
N
NH2
a
11a: R1 = H
a
CO2Me
11b: R1 = Me
e, d
11c: R1 = OMe
a
3i
0
Scheme 3. Reagents and conditions: (a) 8 or 10, Cu(OAc)2, DIPEA, DCE, 3 ÅA MS,
40 °C; (b) HCl, MeOH, reflux; (c) R2CO2H, HATU, DIPEA, DMF, rt; (d) LiOH, THF/H2O,
rt; (e) 2-(bromomethyl)benzoyl bromide, DIPEA, DCM, 0 °C to rt.
For energy minimization the MM2 force field implemented in Chem3D Pro 10.0Ò
(CambridgeSoft) was used.

G. Zischinsky et al. / Bioorg. Med. Chem. Lett. 20 (2010) 65–68
67
Table 2
In vitro and in vivo pharmacokinetic properties of 3a–c
Compds
pKa
Papp, 10ꢀ6 cm/s
Efflux ratio
CLa, ml/min/kg
t1/2a, min
Fa, %
MS ratb, %
3a
3b
3c
7.0
7.5
7.7
4.18
1.42
0.96
0.9
2.0
2.3
16
70
110
127
66
25
24
16
6
85
77
62
a
The compounds were administered into two groups of male rats (4 animals in each group) at 1 mg/kg iv (group 1) and 10 mg/kg po (group 2). Eight blood samples were
taken from each animal. The aliquots of the plasma samples were transferred, precipitated with methanol containing a structural analogue of the analyte as internal standard
and analyzed by HPLC–MS/MS for any taken time point to determine time/concentration values. The pharmacokinetic parameters were calculated by using Winnonlin
version 5.1.
b
% Remaining after 1 h. The stability of the compounds at 1
lM in microsomal preparations was determined as described in Ref. 20. Microsomal preparations from
different species were obtained from Tebu-bio, Offenbach, Germany.
Table 3
Table 5
IC50 values of 3d–g for integrins
a
5b1,
avb3,
a
vb5,
aIIbb3 in competitive integrin
IC50 values of (S)-3g and 3h–i for integrins
a5b1,
avb3,
avb5,
a
IIbb3 in competitive
binding assays
integrin binding assays
N
N
H
R
N
H
N
N
N
R
HN
O
OH
OH
O
O
a
hMSb, %
a
Compds
R
IC50 (nM)
Torsionb
(°)
Compds
R
IC50 (nM)
a
5b1
a
vb3
a
vb5
aIIbb3
a5b1
a
vb3
avb5
aIIbb3
F
3d
3e
1.6
2.2
239
648
1870
77,700
87,200
59
52
(S)-3g
0.90 614
12,400
121,000 57
14,800 28
HN
O
11,500
8290
F
F
3h
3i
70
4.3
32.1
3f
2.7
2.0
402
105,000
71,700
69
HN
N
O
O
234
13.4
3320
76,700
0
3g
1050
11,700
105
a
Values are means of three experiments.
Torsion angles between amid and phenyl plane (Fig. 2).
b
a
Values are means of three experiments.
% Remaining after 1 h, The stability of the compounds at 1 lM in microsomal
b
preparations was determined as described in Ref. 20. Microsomal preparations from
different species were obtained from Tebu-bio, Offenbach, Germany.
Table 4
IC50 values of the stereoisomers of 3g for integrins
a5b1,
a
vb3,
a
vb5,
a
IIbb3 in
competitive integrin binding assays
Compds
IC50a, nM
vb5
a5b1
a
vb3
a
aIIbb3
Cell
3g
(S)-3g
(R)-3g
2.0
0.90
351
10,450
614
16,600
11,700
12,400
15,600
71,700
121,000
159,000
76
21
6912
Figure 2. Torsion angles of N-methylamide and phenyl planes. (a) 3i = 0°; (b)
3h = 28°; (c) 3g = 57°.
a
Values are means of three experiments.
sponded to 37% of dog hepatic blood flow.19 The mean value for
and avb5 binding and the torsion angles was found. It could be dis-
elimination half-life (t ) was 215 min while the bioavailability
½
cussed that the missing NH-donor in 3i is responsible for the low
integrin 5b1 activity, but we assume that the twisted binding con-
formation of the 2,6-substituted benzamide of 3g is directly corre-
lated with the high activity on integrin 5b1 and the selectivity
toward integrin vb3.
after oral administration of an aqueous solution was found to be
11%. The observed kinetics of (S)-3g represents a solid foundation
for further investigations.
In summary, the SAR and pharmacokinetic properties of N-phe-
nyl piperidine scaffolds as integrin antagonists are described. Mod-
ulating pharmacokinetic properties with the substitution of the 2-
aminopyridine and selectivities by benzamide substitution pattern
a
a
a
Compound (S)-3g was further characterized in male beagle
dogs. Following an iv bolus injection of (S)-3g at 10 mg/kg, the sys-
temic clearance was found to be 11 mL/min/kg, which corre-

68
G. Zischinsky et al. / Bioorg. Med. Chem. Lett. 20 (2010) 65–68
6. Reardon, D. A.; Nabors, L. B.; Stupp, R.; Mikkelsen, T. Expert Opin. Invest. Drugs
2008, 17, 1225.
7. O’Connor, P. Exp. Opin. Biol. Ther. 2007, 7, 123.
led to compound (S)-3g with overall acceptable properties for fur-
ther development.
8. Yang, T. J.; Rayburn, H.; Hynes, R. O. Development 1993, 119, 1093.
9. Kim, S.; Bell, K.; Mousa, S. A.; Varner, J. A. Am. J. Pathol. 2000, 156, 1345.
10. Christner-Albrecht, C.; Vossmeyer, D.; Stragies, R.; Zischinsky, G.; Celik, I.;
Dietz, C.; Soulaiman, M.; Zahn, G. AACR Meeting Abstracts, 2007, p B266.
11. Bhaskar, V.; Zhang, D.; Fox, M.; Seto, P.; Wong, M. H. L.; Wales, P. E.; Powers, D.;
Chao, D. T.; DuBridge, R. B.; Ramakrishnan, V. J. Translat. Med. 2007, 5, 61.
12. Stragies, R.; Osterkamp, F.; Zischinsky, G.; Vossmeyer, D.; Kalkhof, H.; Reimer,
U.; Zahn, G. J. Med. Chem. 2007, 50, 3786.
Acknowledgments
We thank Pericles Calias and Brigitte Hoch for proofreading the
manuscript and for their consent to handle any future correspon-
dence. We also thank Edith Weigt, Corinna Mewes, Seike Gericke,
Kerstin Mißbach, Frank Polster, Daniel Ohlendorf, and Thorsten
Lanz for technical assistance.
13. Stragies, R., et al. Bioorg. Med. Chem. Lett., in press. doi:10.1016/
j.bmcl.2009.10.073.
14. Zhang, H.; Cai, Q.; Ma, D. J. Org. Chem. 2005, 70, 5164.
15. Kusturin, C. L.; Liebeskind, L. S.; Neumann, W. L. Org. Lett. 2002, 4, 983.
16. Heckmann, D.; Meyer, A.; Marinelli, L.; Zahn, G.; Stragies, R.; Kessler, H. Angew.
Chem. 2007, 119, 3641.
17. Heckmann, D.; Meyer, A.; Laufer, B.; Zahn, G.; Stragies, R.; Kessler, H. Chem. Bio.
Chem. 2008, 9, 1.
References and notes
1. Hynes, R. O. Cell 1992, 69, 11.
2. Hynes, R. O. Cell 2002, 110, 673.
3. Avraamides, C. J.; Garmy-Susini, B.; Varner, J. A. Nat. Rev. Cancer 2008, 8, 604.
4. Umeda, N.; Kachi, S.; Akiyama, H.; Zahn, G.; Vossmeyer, D.; Stragies, R.;
Campochiaro, P. A. Mol. Pharmacol. 2006, 69, 1820.
5. Dietrich, T.; Onderka, J.; Bock, F.; Kruse, F. E.; Vossmeyer, D.; Stragies, R.; Zahn,
G.; Cursiefen, C. Am. J. Pathol. 2007, 171, 361.
18. Heckmann, D.; Laufer, B.; Marinelli, L.; Limongelli, V.; Novellino, E.; Zahn, G.;
Stragies, R.; Kessler, H. Angew. Chem. 2009, 121, 4501.
19. Davis, B.; Morris, T. Pharm. Res. 1993, 10, 1093.
20. Kunz, W.; Gieschen, H. Drug. Metab. Dispos. 1998, 26, 1120.