Selective, orally active MMP inhibitors with an aryl backbone

Article abstract of DOI:10.1016/S0960-894X(01)00487-5

This letter describes SAR exploration and rat PK optimization of a series of novel, MMP-1 sparing aryl hydroxamate sulfonamides with activity against MMP-2 and MMP-13.

Full text of DOI:10.1016/S0960-894X(01)00487-5

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2481–2483
Selective, Orally Active MMP Inhibitors with an Aryl Backbone
Thomas E. Barta,a,* Daniel P. Becker,a Louis J. Bedell,a Gary A. De Crescenzo,b
Joseph J. McDonald,b Pramod Mehta,b Grace E. Munieb and Clara I. Villamila
aPharmacia, Department of Medicinal Chemistry, 4901 Searle Parkway, Skokie, IL 60077, USA
bPharmacia, Department of Medicinal Chemistry, 700 North Chesterfield Parkway, St. Louis, MO 63198, USA
Received 9 May 2001; accepted 2 July 2001
Abstract—This letter describes SAR exploration and rat PK optimization of a series of novel, MMP-1 sparing aryl hydroxamate
sulfonamides with activity against MMP-2 and MMP-13. # 2001 Elsevier Science Ltd. All rights reserved.
There are over 20 known human matrix metalloprotei-
nases (MMPs). MMPs play a crucial role in develop-
ment and in tissue growth and repair.1 In certain
pathologies, including cancer metastasis,2 arthritis,3 and
congestive heart failure,4 elevated MMP levels may
exacerbate the disease. Thus, there has been consider-
able effort in the drug industry (Fig. 1) directed towards
identifying small molecule inhibitors and several
MMPIs are in clinical trials. Prolonged administration
of broad spectrum inhibitors, such as marimistat, has
been associated with a progressive, albeit reversible,
musculoskeletal syndrome (MSS) involving pain and
stiffening in the joints.5 For this reason, researchers are
interested in identifying selectivity profiles that avoid
MSS. Our paradigm has been to design compounds that
inhibit those MMPs implicated in disease pathology
while sparing the ubiquitous MMP-1.6
In our previous paper,7,8 we described our early work in
a series of aryl-backbone, MMP-1 sparing hydrox-
amates.9 Our aryl analogues were dosed in rat and
found to be orally bioavailable (e.g., 4a, Fig. 2: Cmax
1230 ng/mL @ 20 MPK; t1/2=0.8 h). Subsequent analo-
gues were designed to improve both exposure and half-
life in the rat.
Specifically, it occurred to us that we might improve the
Cmax values the compounds could attain by synthesizing
analogues with water-solubilizing groups built into the
structure. Additionally, we felt we might be able to
extend the half-life of compounds by increasing steric
hindrance around the potentially metabolizable hydrox-
amate moiety. X-ray crystallography (of 4n in MMP-8)7
revealed that substituents on the W, X, and Y positions
are solvent exposed rather than pointing towards the
enzyme backbone. Thus, replacing the hydrogens in
these positions should not deleteriously impact
enzyme potency. The W position looked particularly
attractive as a substitution point that could improve
t1/2 because of its proximity to the hydroxamate
(Table 1).
Figure 1. Some reported MMPIs.
*Corresponding author. Tel.: +1-847-982-4639; fax: +1-847-982-
4714; e-mail: barta@netmug.org
Figure 2. Aryl-backbone MMPIs.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00487-5

2482
T. E. Barta et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2481–2483
Table 1. SAR of aryl-linked compounds
b
c
Compd
W
X
Y
Ar
MMP-1
(nM)a
MMP-2
(nM)a
MMP-13
(nM)a
Cmax
(mg/mL)
t1/2
(h)
%BA
1
2
3
23.5
2.9
800
0.3
0.75
0.4
1.2
2.2
2.4
0.8
330
12
1.8
1.1
0.3
1.0
3.0
1.1
1.4
3.3
1.9
1.0
0.5
2.0
0.6
27
2.06
2.40
1.37
1.23
0.67
22.5
5.16
—
—
—
—
—
9.20
3.06
—
3.20
—
1.31
2.67
1.6
0.9
1.5
0.8
—
1.9
1.5
—
—
—
—
—
—
—
—
—
—
—
—
28
9
22
23
39
32
31
—
—
—
—
—
—
—
—
—
—
—
—
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
p-CF3-Ph
p-OCF3-Ph
p-CF3-Ph
p-OCF3-Ph
o-OMe-Ph
m-OMe-Ph
p-OMe-Ph
Ph
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
>104
33
MeO–
MeO–
MeO–
MeO–
MeO–
MeO–
MeO–
MeO–
–O-CH2-O–
–O-CH2-O–
–O-CH2CH2-O–
F–
MeO–
MeO–
MeO–
MeO–
MeO–
MeO–
MeO–
MeO–
2.7
3.7
2500
250
14
6.7
1.5
13
18
19
7.7
12
d
p-Cl-Ph
Piperonyl
p-CF3-Ph
Piperonyl
p-CF3-Ph
p-CF3-Ph
p-CF3-Ph
p-CF3-Ph
H
H
H
Cl–
CH3-O-(CH2)2-O–
14.8
4.3
4q
4r
H
H
H
p-CF3-Ph
p-CF3-Ph
>104
>104
1.1
21
2.2
—
—
—
—
—
—
MeO–
MeO–
110
aSee ref 12 for assay conditions.
bOral dose of 20 mpk, in rats. See ref 13 for assay conditions.
civ dose of 20 mpk, in rats. See ref 13 for assay conditions.
dNot performed.
With these goals in mind, we selected 3,4-dimethoxy-
benzenesulfonyl chloride as one of our starting materials
(Scheme 1). The chloride was reacted with piperidine 5,
whose synthesis we described previously.7 We hoped
that ring deprotonation of the resulting sulfonamide
would occur preferentially in the 2 position, since that
position is doubly activated both by an oxygen from the
sulfonamide and by the 3-methoxy. This is, in fact, what
we observed. Metalation was typically clean and
facile;10 on quenching with carbon dioxide, good yields
of acids were obtained.
When we attempted to perform EDC coupling on the
carboxylic acid 7c, as we had done in our earlier work,7
we knew immediately that we had achieved our goal of
increasing hindrance at the carbonyl. The HOBT ester
formed in the coupling reaction proved resistant to dis-
placement by THPONH2 under the reaction conditions.
The ester had to be isolated and heated in the presence
of excess THPONH2 in order to obtain the THP
hydroxamate, and the conversion proceeded in low yield.
Fortunately, coupling could be easily achieved by con-
verting the acids into the corresponding, more reactive
acid chlorides. The O-tetrahydropyranyl hydroxamates
thus obtained were deprotected to the corresponding
hydroxamic acids in good yield using HCl.
Scheme 1. (a) TEA, CH3CN, DMAP, 82%; (b) t-BuLi (2 equiv), 0 ꢀC;
then CO2; then HCl (aq), 57%; (c) oxalyl chloride, CH2Cl2;
THPONH2, pyridine, CH2Cl2; (d) AcCl, MeOH, 51% from acid.
Analogues with more elaborate aryl substituents, such
as 4p and 4q, could be made by similar methods, except

T. E. Barta et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2481–2483
2483
References and Notes
1. Greenwald, R. A.; Zucker, S.; Golub, L. M., Eds. Inhibition
of Matrix Metalloproteinases: Therapeutic Applications; New
York Academy of Sciences: New York, 1999.
2. Lauer-Fields, J. L.; Fields, G. B. Exp. Opin. Ther. Patents
2000, 10, 1873.
3. Shaw, T.; Nixon, J. S.; Bottomley, K. M. Exp. Opin. Invest.
Drugs 2000, 9, 1469.
4. Spinale, F. G.; Coker, M. L.; Heung, L. J.; Bond, B. R.;
Gunasinghe, R.; Etoh, T.; Goldberg, A. T.; Zellner, J. L.;
Crumbley, A. J. Circulation 2000, 102, 1944.
5. Heath, E. I.; Grochow, L. B. Drugs 2000, 59, 1043.
6. The hypothesis that sparing MMP-1 is necessary and suffi-
cient to prevent the musculoskeletal stiffening has been called
into question in view of recent clinical results, where it is
observed that certain inhibitors that inhibit MMP-1 never-
theless do not appear to show the side effect. See ref 2.
7. Barta, T. E.; Becker, D. P.; Bedell, L. J.; De Crescenzo,
G. A.; McDonald, J. J.; Munie, G. E.; Rao, S.; Shieh, H.;
Stegeman, R.; Stevens, A. M.; Villamil, C. I. Bioorg. Med.
Chem. Lett. 2000, 10, 2815.
Scheme 2. (a) TEA, CH3CN, DMAP, 84%; (b) MeO(CH2)2OH,
NMP, 115 ꢀC, 20 h, 71%.
8. Barta, T. E.; Becker, D. P.; Bedell, L. J.; Freskos, J. N.;
McDonald, J. J.; Mischke, B. V.; Rao, S. N. WO 9838859,
1998; Chem. Abstr. 1998, 129, 260344.
9. Recently Wyeth-Ayerst reported a series of very different
aryl-backboned MMPIs based on an anthranilate linker. See:
Levin, J. I.; Du, M. T.; DiJoseph, J. F.; Killar, L. M.; Sung,
A.; Walter, T.; Sharr, M. A.; Roth, C. E.; Moy, F. J.; Powers,
R.; Jin, G.; Cowling, R.; Skotnicki, J. S. Bioorg. Med. Chem.
Lett. 2001, 11, 235.
10. For the halogen examples, 6n and 6o, the metalation was
conducted at ꢁ40 ꢀC to minimize the potential for competing
benzyne formation.
that we had to first construct the metalation substrates
by performing a fluoride displacement on the inter-
mediate 6n, as depicted in Scheme 2.
In addition to exploring modifications on the left-hand
aryl groupof the molecule, one of our goals was to
further investigate the right-hand aryl ether moieties
which fit into the P10 pocket in the enzyme. A variety of
aryl groups we studied had good enzyme potency, as we
see from examination of the data for 4g, 4h, 4j, and 4l.
Aryls with ortho or meta substitution, such as 4e and 4f,
were less potent, probably because of steric constraints
in the tight P10 pocket.
11. We were asked by the referees whether the alkoxy groups
we introduced in fact increased aqueous solubility (e.g., 4a and
4b vs 4c and 4d, respectively). We have no data on this, since
we do not routinely measure this physical parameter, but we
did obtain calculated log P values 4a and 4c using several
methods and found the results to be within 0.5 log units. This
leads us to believe that our improved PK may be more the
result of enhanced half-life than enhanced solubility.
12. Assays were conducted at six dilutions with an n at least
equal to 2. Inhibitors were tested against purified hMMP-13,
hMMP-1 and hMMP-2 using an enzyme assay based on clea-
vage of the fluorogenic peptide substrate MCA-Pro-Leu-Gly-
Leu-Dpa-Ala-Arg-NH2. This is similar to conditions described
by Knight, C. G.; Willenbrock, F.; Murphy, G. FEBS Lett.
1992, 296, 263, except that 0.02% final concentration of 2-
mercaptoethanol was used in the MMP-13 and MMP-1
assays.
13. One groupof four rats receives compound via the oral
route at a dosing volume of 2 mL/kg (10 mg/mL, dissolved in
0.5% methylcellulose, 0.1% Tween 20), while another group
of 4 rats receives compound via intravenous cannula at a dos-
ing volume of 2 mL/kg (10 mg/mL, dissolved in 10% EtOH,
50% PEG 400, 40% saline). The blood samples are collected
from the arterial cannula at 3 (iv only), 15, 30, 60, 120, 240,
and 360 min. After each sample, the cannulae are flushed with
PBS containing 10 U/mL heparin. Plasma samples are extrac-
ted with acetonitrile, evaporated under nitrogen, reconstituted
in DMSO, and then analyzed for inhibitor using MMP-13
enzyme and thiopeptolide as a substrate.
Introduction of hindering groups on the ‘backbone’ aryl
ring bearing the hydroxamate (4a and 4b vs 4c and 4d,
respectively) resulted in an unexpected increase in the
MMP-13 potency, which we cannot easily explain from
our understanding of the binding of our solved crystal
structure, 4n, in MMP-8.
As hoped, the alkoxy substituents on the left-hand aryl
ring served to enhance the PK characteristics of the aryl
series.11 The Cmax values were substantially improved
with respect to the unsubstituted analogues. In parti-
cular, 4c and 4d are seen to have markedly improved
Cmax and half-life values.
In summary, we have prepared a series of aryl-linked8
hydroxamate inhibitors that are potent for MMP-2 and
MMP-13 and which spare MMP-1. Analogue 4c exhi-
bits enhanced PK properties in rat and was selected for
testing in efficacy models. Further details will be repor-
ted in due course.