Design and synthesis of 2-oxo-imidazolidine-4-carboxylic acid hydroxyamides as potent matrix metalloproteinase-13 inhibitors

Article abstract of DOI:10.1016/S0960-894X(01)00187-1

A novel series of imidazolidinone-based matrix metalloproteinase (MMP) inhibitors was discovered by structural modification of pyrrolidinone 1a. Potent inhibition of MMP-13 was exhibited by the analogues having 4-(4-fluorophenoxy)phenyl (4a, IC₅₀

Full text of DOI:10.1016/S0960-894X(01)00187-1

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1211–1213
Design and Synthesis of 2-Oxo-imidazolidine-4-carboxylic Acid
Hydroxyamides as Potent Matrix Metalloproteinase-13Inhibitors
Ralph P. Robinson,* Ellen R. Laird, Kathleen M. Donahue, Lori L. Lopresti-Morrow,
Peter G. Mitchell, Matthew R. Reese, Lisa M. Reeves, Amber I. Rouch, Ethan J. Stam
and Sue A. Yocum
Pfizer Global Research & Development, Groton Laboratories, Eastern Point Road, Groton, CT 06340, USA
Received 5 January 2001; accepted 9 March 2001
Abstract—A novel series of imidazolidinone-based matrix metalloproteinase (MMP) inhibitors was discovered by structural mod-
ification of pyrrolidinone 1a. Potent inhibition of MMP-13 was exhibited by the analogues having 4-(4-fluorophenoxy)phenyl (4a,
IC50=3 nM) and 4-(naphth-2-yloxy)phenyl (4h, I C =4 nM) as P10 groups. # 2001 Elsevier Science Ltd. All rights reserved.
50
Inhibitors of matrix metalloproteinases (MMPs) have
great potential for the treatment of a variety of diseases
including arthritis and cancer.1 We recently described
the structure-based design and synthesis of 1a (Table 1),
a potent and novel MMP-13 inhibitor.2 This compound
utilizes a pyrrolidinone ring as a scaffold for correct
orientation of groups that bind to the enzyme: a
hydroxamic acid for chelation of the active site zinc
atom, a 4-(4-fluorophenoxy)phenyl P10 group for inter-
action with the hydrophobic S10 pocket and a cis amide
group to establish hydrogen bonds with main chain
atoms flanking the active site. Here, we describe the
discovery of a new series of MMP inhibitors based on
1a that incorporates an imidazolidinone ring as an
alternative scaffold.
Although 1a is stable in neutral or mildly basic media,
epimerization of the relatively acidic C4 position under
moderately basic conditions (e.g., alkoxide) limits the
scope of chemistry that can be employed for the synth-
esis of analogues. For example, formation of appreci-
able amounts of the trans diastereomer3 occurs during
alkylation of the OH group of intermediate 2 using a
variety of bases. An intriguing possibility for avoiding
epimerization at C4 was to replace this center with a
nitrogen atom giving the corresponding imidazolidinone
(e.g., 3). However, based on modeling studies using a
homology model of MMP-13,2 it was clear that P10
groups would no longer be oriented correctly for occu-
0
pancy of the S1 pocket due to the planar geometry of
the cyclic urea function. We therefore considered com-
pounds such as 4a in which a methylene spacer was
introduced to allow greater flexibility of the P10 side
chain. Indeed, modeling of 4a in the active site of
MMP-13 showed that this molecule could attain a low
Table 1.a
Compound
MMP IC50 (nM)
3
1
2
9
13
1a
4a
4h
1560 (120)
5450b
12,700b
39 (17)
28 (9)
59 (37)
700 (170)
1380 (330)
940 (310)
82 (16)
62 (13)
56 (24)
7 (1)
3 (1)
4 (2)c
aValues are ꢀSD of 3 determinations unless otherwise noted.
bn=2.
cn=4.
*Corresponding author. Fax; +1-860-715-2469;
e-mail: ralph_p_robinson@groton.pfizer.com
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00187-1

1212
R. P. Robinson et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1211–1213
energy conformation positioning the 4-fluorophenoxy-
phenyl group deep within the S10 pocket while main-
taining important contacts made by the hydroxamic
acid and cis amide functions.
Upon testing, we were pleased that 4a exhibited potent
activity against MMP-13 (IC50=3 nM, Table 1). The
compound also displayed moderate potency against
MMP-2 and MMP-9 but was significantly less active
against MMP-1 and MMP-3. Overall, the activity pro-
file of 4a against the various MMPs closely resembled
that of 1a. As in the pyrrolidinone series, the weak
activity against MMP-1 was considered advantageous
in light of side effects attributed to MMP-1 inhibition in
vivo.6
The synthesis of 4a and analogues (4b–h, Table 2) was
carried out in five steps from the readily available 2-
oxo-imidazolidine-4-carboxylic acid derivative 54
(Scheme 1). Alkylation of 5 with benzyl bromide deri-
vatives5 took place smoothly, albeit slowly, to provide 6
(ꢁ75 %). Sequential removal of the CBZ and tert-butyl
ester protecting groups afforded acid 7 (ꢁ60% for the
two steps). Finally, formation of the (2-trimethylsilyl-
.
ethoxy)amide and deprotection using BF3 OEt2 intro-
duced the hydroxamic acid function to provide 4a and
analogues (ꢁ25% for the two steps). In terms of
synthesis, the imidazolidinones have clear advantages
over the pyrrolidinone series: easily obtained starting
materials, fewer steps and reduced potential for loss of
stereochemical integrity. As a result, the synthesis
offered the possibility for rapidly exploring a range of
different P10 groups to optimize activity against MMP-13
and related metalloenzymes (e.g., MMP-2 and MMP-9).
Scheme 1.
Table 2.a
Compound
R
MMP-1 IC50 (nM)
MMP-13 IC50 (nM)
Imidazolidinoneb
Pyrrolidinone
Imidazolidinone
Pyrrolidinone
a
b
c
d
e
f
5450
1560 (120)
3 (1)
7 (1)
388
8040 (2410)c
6430 (2340)
>30,000
281 (60)
1480 (570)d
182 (20)
25,900
331 (65)
1590 (380)
123 (41)
>30,000
>30,000
>30,000
3450 (210)
946 (340)
4760 (580)
>30,000
>30,000
214 (15)
g
h
8020
825 (64)
4 (2)c
12,700
aValues are ꢀSD of 3 determinations unless otherwise noted.
bn=2.
cn=4.
dn=5.

R. P. Robinson et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1211–1213
1213
Analogues 4b–h were prepared to probe the MMP-1
and MMP-13 SAR around P10. In the case of 4a–f,
comparison to the P10 SAR in the pyrrolidinone series
(1a–f) could be made and, with the exception of 4c, all
exhibited modestly increased MMP-13 inhibition rela-
tive to the corresponding pyrrolidinones. Most striking
is the 3-(4-fluorophenoxy)phenyl analogue 4f which is
over 20-fold more potent against MMP-13 than 1f. We
believe the overall improved MMP-13 potency of the
series is a consequence of the greater flexibility per-
mitted by the methylene spacer linking the imidazolidi-
none scaffold and the P10 group. This flexibility allows
productive binding conformations to be more easily
attained. Like the pyrrolidinones, the imidazolidinones
are selective for MMP-13 over MMP-1 although in the
case of 4b, the selectivity is very slight (ꢁ1.4). This
results from relatively potent inhibition of MMP-1 and
is consistent with the general preference of MMP-1 for
small hydrophobic P10 groups.
during the synthesis of analogues of 1a. As in the pyrro-
lidinone series, the most potent inhibition of MMP-13 was
exhibited by the analogue having a 4-(4-fluorophenoxy)-
phenyl P10 group (i.e., 4a). Potent MMP-13 inhibition was
also shown by the P10 4-(naphth-2-yloxy)phenyl analo-
gue 4h. The lack of a relatively acidic chiral center sus-
ceptible to epimerization and the comparative ease of
synthesis establish the imidazolidones as an attractive
new series for the continued pursuit of MMP inhibitors
as drugs.
Acknowledgements
The authors thank Drs. Lawrence A. Reiter and Martin
Wythes for helpful comments and review of the manu-
script.
References and Notes
Despite the overall improved MMP-13 potency of the
imidazolidione series relative to the pyrrolidinones, 4b–g
are nonetheless relatively weak inhibitors of MMP-13 in
comparison to 4a and 1a. In sharp contrast to 4b–g, the
4-(naphth-2-yloxy)phenyl analogue 4h displays potent
MMP-13 inhibition (IC50=4 nM). This compound was
tested against MMP-2, MMP-3, and MMP-9 giving an
inhibition profile against the various MMPs similar to that
of 4a (Table 1). Modeling of 4h in the active site of MMP-
13 shows a high degree of complementarity between the 4-
(naphth-2-yloxy)phenyl group and the S10 pocket.
1. (a) Recent reviews on MMP inhibitors: Michaelides, M. R.;
Curtin, M. L. Curr. Pharm. Des. 1999, 5, 787. (b) Whittaker,
M.; Floyd, C. D.; Brown, P.; Gearing, A. J. H. Chem. Rev.
1999, 2735. (c) Beckett, R. P.; Whittaker, M. Expert Opin.
Ther. Patents 1998, 8, 259.
2. Robinson, R. P.; Laird, E. R.; Blake, J. F.; Bordner, J.;
Donahue, K. M.; Lopresti-Morrow, L. L.; Mitchell, P. G.;
Reese, M. R.; Reeves, L. M.; Stam, E. J.; Yocum, S. A. J.
Med. Chem. 2000, 43, 2293.
3. The trans diastereomer of 1a was much less active than 1a
as an inhibitor of MMP-13.2
4. The S-enantiomer of 5 is known: Shiba, T.; Koda, A.;
Kusumoto, S.; Kaneko, T. Bull. Chem. Soc. Jpn. 1968, 41, 2748.
5. Hayashi, K.; Nunami, K.; Kato, J.; Yoneda, N.; Kubo, M.;
Ochiai, T.; Ishida, R. J. Med. Chem. 1989, 32, 289.
6. Rothenberg, M. L.; Nelson, A. R.; Hande, K.R Oncologist
1998, 3, 271.
In summary, we have discovered a novel series of imi-
dazolidinone-based MMP inhibitors by structural mod-
ification of pyrrolidinone 1a, a potent inhibitor of
MMP-13. The imidazolidinone scaffold was explored as
a strategy for avoiding the epimerization at C4 observed