Synthesis, SAR exploration, and X-ray crystal structures of factor XIa inhibitors containing an α-ketothiazole arginine

Article abstract of DOI:10.1016/j.bmcl.2006.02.052

Using an α-ketothiazole arginine moiety as a key recognition element, a series of small peptidomimetic molecules was designed and synthesized, and their co-crystal structures with factor XIa were studied in an effort to develop smaller, less peptidic inhibitors as antithrombotic agents.

Full text of DOI:10.1016/j.bmcl.2006.02.052

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3049–3054
Synthesis, SAR exploration, and X-ray crystal structures of factor
XIa inhibitors containing an a-ketothiazole arginine
Hongfeng Deng,a,* Thomas D. Bannister,a Lei Jin,b Robert E. Babine,b Jesse Quinn,a
Pamela Nagafuji,a Cassandra A. Celatka,a Jian Lin,a Tsvetelina I. Lazarova,a
Michael J. Rynkiewicz,b Frank Bibbins,c Pramod Pandey,d Joan Gorga,c
Harold V. Meyers,a Sherin S. Abdel-Meguida,b,c,d and James E. Stricklerc
aDepartment of Medicinal Chemistry, Daiichi Asubio Medical Research Laboratories, LLC (DAIAMED),
One Kendall Square, Bldg 700, Cambridge, MA 02139, USA
bDepartment of Structural Biology and Computational Chemistry, Daiichi Asubio Medical Research Laboratories,
LLC (DAIAMED), One Kendall Square, Bldg 700, Cambridge, MA 02139, USA
cDepartment of Biochemistry, Daiichi Asubio Medical Research Laboratories, LLC (DAIAMED),
One Kendall Square, Bldg 700, Cambridge, MA 02139, USA
dDepartment of Biology, Daiichi Asubio Medical Research Laboratories, LLC (DAIAMED),
One Kendall Square, Bldg 700, Cambridge, MA 02139, USA
Received 15 December 2005; revised 17 February 2006; accepted 20 February 2006
Available online 9 March 2006
Abstract—Using an a-ketothiazole arginine moiety as a key recognition element, a series of small peptidomimetic molecules was
designed and synthesized, and their co-crystal structures with factor XIa were studied in an effort to develop smaller, less peptidic
inhibitors as antithrombotic agents.
Ó 2006 Elsevier Ltd. All rights reserved.
Thromboembolic diseases are the leading causes of mor-
bidity and mortality in the developed countries.1 Con-
ventional antithrombotic therapy includes intravenous
administration of heparin followed by oral treatment
with warfarin.2 Since both agents are indirect and non-
specific inhibitors of the coagulation serine proteases,
this treatment carries high risk of excessive bleeding
and requires routine monitoring. To overcome these
limitations, new antithrombotic agents have been de-
signed to target specific enzymes in the coagulation
pathway, such as thrombin and factor Xa (FXa).
Although these newer anticoagulants offer some advan-
tages over heparin therapy, some undesirable features
such as bleeding and a narrow therapeutic window have
been demonstrated and their safety profiles have yet to
be established in clinical trials.3
Coagulation factor XIa (FXIa) is a trypsin-like serine
protease that plays a major role in amplification of the
thrombotic response and in maintaining clot integrity.4
The inhibition of FXIa in vivo may result in the forma-
tion of smaller thrombi without affecting the initiation
of clot formation. The lack of an effect upon initiation
may result in a lesser bleeding liability for FXIa inhibi-
tors than anticoagulants that do suppress clot initiation.
Supporting evidence for this hypothesis includes the
observation that administration of an anti-human FXIa
antibody significantly reduces intraluminal thrombus
growth in a baboon thrombosis model without prevent-
ing thrombus initiation.5 The anti-FXIa antibody
prolongs the partial thromboplastin time (a measure
of the coagulation function involving FXIa) without
affecting prothrombin time (a measure of the coagula-
tion function involving FVIIa and tissue factor). Most
significantly, the anti-FXIa antibody does not prolong
bleeding time. Also supporting the FXIa lowered bleed-
ing risk hypothesis is the finding that individuals with a
natural FXIa deficiency do not show severe bleeding
problems.6
Keywords: Factor XIa inhibitor; a-Ketothiazole arginine; Anticoagu-
lant; X-ray crystal structure.
*
Corresponding author. Tel.: +1 781 795 4302; fax: +1 781 795
4496; e-mail: hongfeng.deng@praecis.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.02.052

3050
H. Deng et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3049–3054
H2N
NH
NH
H2N
NH
NH
H2N
NH
NH
X = NH or CH2
S
O
S
O
S
H
N
H
N
X
R
N
N
R
N
H
R
N
H
N
N
H
O
O
O
O
O
I
II
III
Figure 1.
O
O
O
H
N
BocHN
OMe
O
H2N
S
R
N
S
N
S
a, b
c, d
N
O
HN
HN
O
O
O
HN
HN
S
HN
N
H
HN
N
H
NH2 TFA
OMe
OMe
3 (a~h)
1
2
Scheme 1. Reagents and conditions: (a) thiazole (4.6 equiv), n-BuLi (4.16 equiv), TMEDA (4.6 equiv), THF, À78 °C to À41 °C then À78 °C, 75%;
(b) 25% TFA–DCM, rt; (c) RCOOH, EDC (1.24 equiv), HOBt (1.24 equiv), DIEA (4.0 equiv), THF, rt; or RCOCl (RSO2Cl), TEA, DCM; (d) 95%
TFA, 2.5% thioanisole, 2.5% H2O, 30–45% for three steps.
O
O
H
N
H
N
a, b, c
NHFmoc
O
O
R
O
4
H
N
H2N
NH
NH
Mtr
NH
NH
d, e
f
S
O
S
O
H
N
H
N
H
N
H
N
N
H
N
N
H
N
R
R
O
O
O
O
5
6 (a~h)
Scheme 2. Reagents and conditions: (a) 20% piperidine–DMF, 2 h; (b) 4-nitrophenyl chloroformate, DIEA, THF–DCM (1/1), 1 h; (c) RNH2, TEA,
DMF, 2 h; (d) 50% TFA–DCM, 30 min; 28–40% for four steps; (e) 2, EDC, HOBt, DIEA, DCM–THF (1/1), 1 h; (f) 95% TFA, 2.5% H2O, 2.5%
thioanisole, 3 h, 30–45% for two steps.
O
O
a
b, c, d
O
O
O
O
HN
HN
NH2
7
Cl
Cl
Cl
Cl
e, f
O
O
H
N
H
N
S
OH
N
H
N
O
O
O
8
9
Scheme 3. Reagents and conditions: (a) LDA (1.2 equiv), THF, À78 °C to 0 °C; tert-butyl 2-bromoacetate, THF, 0 °C to rt, 48%; (b) 50% TFA–
DCM, 30%; (c) 3,4-dichlorobenzylamine, EDC, HOBt, DIEA, THF, 99%; (d) H2, Pd–C (5%), MeOH, 70%; (e) compd 2, EDC, HOBt, DIEA, THF;
(f) 95% TFA, 2.5% H2O, 2.5% thioanisole, DCM, 3 h, 36% for two steps.

H. Deng et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3049–3054
3051
We recently reported a series of specific tripeptidomi-
metic FXIa inhibitors.7 In vivo pharmacological studies
revealed that these inhibitors effectively reduce the size
of clot without altering bleeding times and other obser-
vable side effects. In this paper, we report the design and
synthesis of smaller, less peptidic FXIa inhibitors, repre-
sented by the generic structures I, II, and III (Fig. 1).
pyroglutamyl-prolyl-arginyl (Glu-Pro-Arg). The assay
was carried out in a buffer containing 20 mM Tris–
HCl, pH 7.4, 150 mM NaCl, and 0.02% Tween 20.
Enzyme and substrate concentrations were 0.25 nM
and 50 lM, respectively. The assay was incubated at
30 °C for 15–30 min and then released AMC was mea-
sured (emission 460 nm, excitation 355 nm) on a Wallac
1420 Multilabel Counter. The data were fit to a one-site
dose–response curve using non-linear least squares to
determine the IC50 value.
The synthesis of type I compounds is shown in Scheme 1.
The Mtr-protected a-ketothiazole arginine (2) was pre-
pared from Weinreb amide (1) using a modified proce-
dure in which lithium thiazole, generated in situ, was
added by cannula to pre-cooled Weinreb amide at
À78 °C.8 After workup, the isolated product was careful-
ly treated with 25% TFA-DCM to afford the desired
amine 2. Direct coupling of 2 with selected acids, acid
chlorides, or sulfonyl chlorides, followed by Mtr-depro-
tection provided compounds 3 (a–h).
Table 1. SAR of ketothiazole arginine analogs with one amide bond
HN
HN
NH2
S
R
N
N
H
A combination of solid-phase and solution-phase syn-
thesis was used to prepare urea type II compounds
(X = NH) (Scheme 2). Pre-loaded Fmoc Wang resin
was treated with 20% piperidine in DMF followed by
para-nitrophenyl chloroformate to form the carbamate
intermediate, which was then treated with various select-
ed amines to form the N-terminal ureas (4). Cleavage
from the resin followed by standard EDC-HOBt
coupling with the aminoketone (2) yielded compounds
6 (a–h) after Mtr-deprotection.
O
Compound
R
MH+
(m/z)
IC50 (lM)
a
FXIa
123
O
3a
3b
326.3
410.2
428.3
>250
37.9
O
Cl
Cl
O
3c
The synthesis of a reverse amide type II compound
(X = CH2) is shown in Scheme 3. 3-Methyl butyric acid
benzyl ester was treated with LDA followed by tert-
butyl a-bromoacetate to give the 2-isopropyl succinic
acid derivative 7. After removal of the tert-butyl ester,
the acid was coupled with 3,4-dichlorobenzylamine
and then hydrogenolysis afforded the acid 8. Coupling
of 8 with aminoketone 2 provided compound 9 after
standard Mtr-deprotection.
O
3d
3e
396.3
396.2
2.25
O
>200
O
O
S
3f
432.2
432.2
86.0
250
O
O
S
The synthesis of type III compounds is shown in
Scheme 4. Standard EDC–HOBt coupling of 2 with
CBz-Val-OH followed by hydrogenolysis yielded 10.
EDC–HOBt coupling with various carboxylic acids
and standard Mtr-deprotection provided the N-acylat-
ed Val-Arg-ketothiazoles 11 (a–h).
3g
O
O
O
3h
422.2
5.90
F
a IC50s were derived from triplicate measurements whose standard
errors were normally <5% in a given assay. Assay-to-assay variability
was within 2-fold based on the results of a standard compound,
benzamidine.
Peptidolytic activities for FXIa were measured using a
fluorogenic reporter group, 7-methyl-4-aminocoumarin
(AMC), attached via an amide linkage to the tripeptide
HN
HN
NH2
HN
HN
NH Mtr
a, b
c
O
O
2
N
N
H
N
H2N
S
S
N
H
R
N
H
O
O
10
11 (a~h)
Scheme 4. Reagents and conditions: (a) CBz-Val-OH, EDC, HOBt, DIEA, DMF, 85%; (b) H2, Pd–C (10%), EtOAc, 95%; (c) i—RCO2H, EDC,
HOBt, DIEA, DMF, 60–80%; ii—95% TFA, 2.5% H2O, 2.5% thioanisole, DCM, 3 h, 55–70%.

3052
H. Deng et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3049–3054
Table 2. SAR of ketothiazole arginine analogs with two amide bonds
and a urea moiety at the N-terminus
Table 3. SAR of ketothiazole arginine analogs with two amide bonds
and an acyl moiety at the N-terminus
HN
HN
NH2
HN
HN
NH2
S
O
S
O
H
N
H
N
X
R
N
N
N
H
R
N
H
O
O
O
O
MH+
(m/z)
IC50 (lM)
a
Compound
R
MH+ (m/z)
IC50 (lM) FXIa
a
Compound
X
R
FXIa
6a
6b
NH
NH
H
384.3
0.502
11a
439.3
0.850
Cl
542.1
544.1
0.480
O
Cl
Cl
Cl
11b
11c
11d
11e
11f
11g
11h
507.2
509.2
509.2
493.2
475.2
475.2
474.2
0.254
0.116
0.790
0.956
0.425
0.431
1.75
6c
6d
NH
NH
488.3
1.72
1.20
OH
566.6
568.2
Br
Br
F
Cl
OH
6e
NH
566.6
568.2
12.5
OH
F
6f
6g
6h
9
NH
NH
NH
CH2
506.4
520.4
520.4
2.32
0.713
3.31
6.22
OH
OH
NH2
Cl
Cl
541.5
543.3
a IC50s were derived from triplicate measurements whose standard
errors were normally <5% in a given assay. Assay-to-assay variability
was within 2-fold based on the results of a standard compound,
benzamidine.
a IC50s were derived from triplicate measurements whose standard
errors were normally <5% in a given assay. Assay-to-assay variability
was within 2-fold based on the results of a standard compound,
benzamidine.
Table 1 shows selected SAR of type I compounds with
the ketothiazole arginine moiety directly coupled to a
variety of different non-amino acid fragments through
either an amide or a sulfonamide bond. In general,
such compounds have very weak to moderate potency.
The most potent analog in the series has R = naphtha-
lene-1-carbonyl (3d), with an IC50 (FXIa) of 2.25 lM.
When R = naphthalene-2-carbonyl (3e), the potency is
100-fold less. The same diminished affinity is seen for
sulfonamide analog 3g. It seems that the 6-fluoro-ben-
zo[1,3]-dioxane ring (3h) is a good bioisostere of a
1-naphthyl ring since 3d has about the same potency as
3h. Due to the overall low affinity for FXIa, no further
optimization effort was pursued for type I compounds.
by inhibitor 6a. 3,4-Dichlorobenzyl substitution of the
urea maintained the FXIa potency (6b). While the
chirality of R groups attached to the urea appears to
significantly affect potency of these analogs (e.g., 6d vs
6e and 6g vs 6h), different substituents on the para-posi-
tion of the phenyl ring, such as, hydrogen (6c), bromo
(6d), or fluoro (6f), cause little change in affinity. Nota-
bly, replacement of the urea NH of 6b with a methylene
moiety (9) significantly decreases potency, suggesting
that this NH is important for FXIa inhibition.
We then explored whether the external urea NH group is
necessary for inhibition. The results of this study are
summarized in Table 3. Compound 11a, with an isopentyl
group replacing the external urea NH2 of 6a, has similar
potency, an indication that this NH is dispensable. When
R is 3-chlorobenzoyl (11b), the potency increases ꢀ3-fold.
Changing the carbonyl to a hydroxyl group further
improves potency (11c). The stereochemistry of this
hydroxyl group seems to be not important, since both R
Table 2 shows the SAR of type II compounds having the
ketothiazole arginine linked to a urea derivative through
a Val residue.9 The addition of the Val residue as the P2
recognition element and the urea unit as an N-terminal
cap further increased FXIa inhibition, as demonstrated

H. Deng et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3049–3054
3053
and S are equally potent (11f and 11g). Replacing this
hydroxyl with an amino group is detrimental for affinity
(11h). The chloro group and its position on the phenyl ring
of 11c are important for the activity. Compounds where it
is switched from meta-position (11c) to the ortho-position
(11d), replacedwithafluorogroup(11e), orcompletely re-
moved (11f) all have decreased potency.
Figure 2 shows the X-ray crystal structures of 11a (PDB
code: 1ZPB), 6a (PDB code: 1ZPC), and 11c (PDB code:
2FDA) in co-complex with FXIa catalytic domain
(FXIac). All three structures show salt bridge interac-
tion between the P1 arginine and D189 in the S1 pocket,
while the keto group is covalently attached to S195 in
the enzyme active site.10 The P2 Val side chain occupies
the S2 pocket. The N-terminal carbonyl groups in 11a,
6a, and 11c form an H-bond interaction with the back-
bone NH of G216. The differences among these struc-
tures include: (1) the isopentyl group in 11a forms a
hydrophobic interaction with residues that form the S4
pocket; (2) the NH2 of the urea in 6a forms an addition-
al H-bond interaction with the backbone carbonyl
group of G216; and (3) in 11c, the above two interac-
tions, a hydrophobic S4 interaction and a hydrogen
bond with the hydroxyl group, are combined together.
Since the hydrogen bonds between the hydroxyl group
of 11c and G216 are weak, the enhanced hydrophobic
interactions of the chlorophenyl moiety in the S4 pocket
might contribute to the higher affinity of 11c.
In conclusion, three types of FXIa inhibitors contain-
ing fewer peptide bonds than those previously report-
ed are disclosed. Several compounds from these series
were found to have submicromolar potency. The
SAR for these compound series and the analysis of
their co-crystal structures provide guidance for the
design of future FXIa inhibitors as antithrombotic
agents.
References and notes
1. Qiao, J. X.; Cheng, X.; Modi, D. P.; Rossi, K. A.;
Luettgen, J. M.; Knabb, R. M.; Jadhav, P. K. Bioorg.
Med. Chem. Lett. 2005, 15, 25.
2. Hyers, T. M. Arch. Intern. Med. 2003, 163, 759.
3. Hirsh, J.; O’Donnell, M.; Weitz, J. I. Blood 2005, 105, 453.
4. Walsh, P. N. Thromb. Haemost. 2001, 86, 75.
5. Gruber, A.; Hanson, S. R. Blood 2003, 102, 953.
6. Ragni, M. V.; Sinha, D.; Seaman, F.; Lewis, J. H.; Spero,
J. A.; Walsh, P. N. Blood 1985, 65, 719.
7. Lin, J.; Deng, H.; Jin, L.; Pandey, P.; Rynkiewicz, M.;
Bibbins, F.; Cantin, S.; Quinn, J.; Magee, S.; Gorga, J.;
Celatka, C.; Nagafuji, P.; Bannister, T.; Meyers, H. V.;
Babine, R.; Hayward, N.; Abdel-Meguid, S. S.; Strickler,
J. E. Abstracts of Papers, 229th National Meeting of the
American Chemical Society, San Diego, CA, March
13–17, 2005; American Chemical Society: Washington,
DC, 2005; MEDI 247.
8. Berryman, K.; Doherty, A.; Edmunds, J.; Plummer, J.
PCT Int. Appl. WO 97/48687, 1997.
9. Compounds in Tables 2 and 3 were also assayed against
thrombin and trypsin. In general, the affinity for these
enzymes was similar to that for FXIa, that is, within 100-
fold.
Figure 2. X-ray crystal structures of 11a, 6a, and 11c with FXIac.
FXIac is in stick representation and the ligands are in ball-and-stick
representation. H-bond distances are labeled in angstrom. Substrate-
binding sites are labeled in gray letters. The crystallization and ligand-
soaking conditions have been published.11

3054
H. Deng et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3049–3054
10. Jin, L.; Pandey, P.; Babine, E. R.; Gorga, J. C.; Seidl, K.
J.; Gelfand, E.; Weaver, D. T.; Abdel-Meguid, S. S.;
Strickler, J. E. J. Biol. Chem. 2005, 280, 4704.
11. Jin, L.; Pandey, P.; Babine, E. R.; Weaver, D. T.; Abdel-
Meguid, S. S.; Strickler, J. E. Acta Crystallogr. 2005, D61,
1418.