Orally bioavailable dual MMP-1/MMP-14 sparing, MMP-13 selective α-sulfone hydroxamates

Article abstract of DOI:10.1016/j.bmcl.2010.04.130

A series of phenyl piperidine α-sulfone hydroxamate derivatives has been prepared utilizing a combination of solution-phase and resin-bound library technologies to afford compounds that are potent and highly selective for MMP-13, are dual-sparing of MMP-1

Full text of DOI:10.1016/j.bmcl.2010.04.130

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3557–3560
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Orally bioavailable dual MMP-1/MMP-14 sparing, MMP-13 selective
hydroxamates
a-sulfone
Stephen A. Kolodziej, Susan L. Hockerman, Terri L. Boehm, Jeffery N. Carroll, Gary A. DeCrescenzo,
Joseph J. McDonald, Debbie A. Mischke, Grace E. Munie, Theresa R. Fletcher, Joseph G. Rico,
*
Nathan W. Stehle, Craig Swearingen, Daniel P. Becker
Departments of Medicinal Chemistry and Pharmacology, Pfizer Research & Development, 700 Chesterfield Village Parkway, St. Louis, MO 63198, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of phenyl piperidine a-sulfone hydroxamate derivatives has been prepared utilizing a combina-
Received 11 March 2010
Revised 23 April 2010
Accepted 27 April 2010
Available online 23 May 2010
tion of solution-phase and resin-bound library technologies to afford compounds that are potent and
highly selective for MMP-13, are dual-sparing of MMP-1 and MMP-14 (MT1-MMP) and exhibit oral bio-
availability in rats.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Matrix metalloproteinase inhibitor
MMP inhibitor
MMP-13 selective
Hydroxamate
MMP-1/14 dual-sparing
Metalloenzyme
The matrix metalloproteinases (MMPs) are a family of zinc-
dependent enzymes that degrade all components of the extracellu-
lar matrix.1,2 There are at least 24 isozymes in the MMP family, and
they are roughly classified on the basis of their substrate specific-
ity: collagenases (MMP-1, -8, -13 and -18), gelatinases (MMP-2,
and-9), stromelysins (MMP-3, -10 and -11) and membrane-type
MMPs (MMP-14, -15, -16, -17, and -24) and others (MMP-7, -11,
-12, -19, -20, -21, -22, and -23). Clinical experience with the pan
MMP inhibitor Marimastat has revealed a constellation of adverse
effects collectively referred to as musculoskeletal syndrome
(MSS).3 We have hypothesized that this is predominantly a result
of inhibiting both MMP-1 and MMP-14, and that MMP inhibitors
sparing these two isozymes should be devoid of MSS.4 Toward
obtaining efficacy in mitigating damage suffered in osteoarthritic
patients, it has been reported that MMP-13 mRNA levels are in-
creased in osteoarthritic cartilage.5 Thus, we envisioned the devel-
opment of a selective inhibitor of MMP-13 sparing both MMP-1
and MMP-14 as a safe means of treating osteoarthritis (OA) which
we refer to as the dual-sparing hypothesis.
potent and selective MMP-13 inhibitors. Figure 2 shows a sequence
comparison for a set of MMP family members focused on the S01 loop.
There are differences in both amino acid identity and in length of the
loop, as MMP-1, -2 and -9 are two residues shorter than MMP-13, -8
and -14. Interaction of the amide N-substituents of 1 deep in the S10
pocket was expected to affect isozyme selectivity across the MMP
family. In an effort to further investigate SAR in this region, we de-
signed templates 2 and 3, where the distal phenyl group resides in
approximately the same region as the amide N-substituents of 1.
To explore the effect of substituents on the distal aryl rings of 2
and 3, we undertook a parallel synthetic approach with the goal of
optimizing MMP-13 potency and selectivity.
A solid-phase parallel synthesis approach was used to create a
small library of N-arylpiperazine a-sulfone hydroxamic acid deriv-
atives (2) from commercially available N-aryl piperazines 5. Nucle-
ophilic aromatic substitution of the previously reported polymer-
bound aryl fluoride 44 with N-arylpiperazines was found to require
a 10-fold excess of 5 and presence of 2 equiv of cesium carbonate
in N-methylpyrrolidinone (NMP) at 100 °C over night to achieve
good conversions. Acidic deprotection with trifluoroacetic acid
In the preceding paper,4 we demonstrated that the N-substituted
phenyl isonipecotamide hydroxamic acid template 1 (Fig. 1) yielded
afforded
Alternatively, a solution phase approach was developed to syn-
thesize analogs with a basic amine in the -heterocycle (8). THP-
a-sulfone hydroxamic acids 2 in good yields (Scheme 1).
a
* Corresponding author at present address: Department of Chemistry, Loyola
University, 6525 North Sheridan Road, Chicago, IL 60626, USA. Tel.: +1 773 508
3089; fax: +1 773 508 3086.
protected hydroxymate 7 was obtained by reaction of carboxylic
acid 6 with THP-protected hydroxylamine using the water-soluble
carbodiimide EDC.6 Subsequent nucleophilic aromatic substitution
E-mail address: dbecke3@luc.edu (D.P. Becker).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.130

3558
S. A. Kolodziej et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3557–3560
Figure 1. 4-Substituted piperidine/piperazine sulfone hydroxamic acids.
Figure 2. S01 loop sequence variation across selected MMP family members.
with the requisite N-arylpiperazines followed by acidic deprotec-
tion afforded -sulfone hydroxamic acids 6 in high yields.
a
For the synthesis of 4-arylpiperidine phenyl sulfones 3, the re-
quired 4-aryl piperidines 11 were purchased commercially or pre-
pared either via the Grignard route (Scheme 3) or via Suzuki
coupling (Scheme 4). For the Grignard route, similar to the method
of Burns,7 the appropriate aryl Grignard was added to N-benzyl
piperidine-4-one followed by dehydration with TFA in methylene
chloride. Reduction then gave 4-arylpiperidines 11. Alternatively,
enol triflate 12 was reacted with the requisite arylboronic acid with
a catalytic amount of tetrakistriphenylphosphine palladium to af-
ford N-Boc-4-aryl tetrahydropyridine 13, following the procedure
of Wustrow and Wise8 (Scheme 2). Catalytic hydrogenation of 13
followed by removal of the N-Boc protecting group afforded 4-aryl-
piperidines 11.
Scheme 2. Solution phase synthesis of N-arylpiperazine sulfone hydroxamic acids 8.
Scheme 3. Syntheses of 4-arylpiperidines via the Grignard route.
Assembly of the N-arylpiperidine sulfone hydroxamic acids 3
was accomplished by a solution phase approach (Scheme 5) similar
to that described above. Carboxylic acid 146 was coupled with
THP-protected hydroxylamine, then aromatic nucleophilic substi-
tution with 11 followed by acidic deprotection afforded
hydroxamic acids 3 in good yields.
a
-sulfone
The MMP inhibitory potency values for N-aryl piperazine
a-sul-
fone hydroxamic acids (2 and 8) are summarized in Table 1. All com-
pounds were determined to have no measurable potency at MMP-1
(IC50 >10,000 nM), hence selectivity for MMP-13 over MMP-1 is
>1000ꢀ in most cases. The N-aryl piperazines (2a–2j and 8a–8b)
exhibited mostly single-digit nanomolar potency for MMP-13, but
most were also potent for MMP-2, resulting in only very modest
selectivity for MMP-13 over MMP-2 with a number of compounds
which were nearly equipotent. N-Phenyl piperazine 2a was very po-
tent for MMP-13 (IC50 = 1.7 nM) with a 14-fold selectivity versus
Scheme 4. Syntheses of 4-arylpiperidines via Suzuki coupling.
MMP-2, and nearly 6000-fold selectivity versus MMP-14, whereas
the corresponding a-piperidine 8a was nearly equipotent at MMP-
13 and MMP-2 (IC50 = 3.3 and 5.4 nM, respectively.) The reduction
in selectivity due to the change from X = O to X = N-cyclopropyl
was unexpected given the continuity between
a-tetrahydropyran
Scheme 1. Solid phase synthesis of N-arylpiperazine sulfone hydroxamates 2.

S. A. Kolodziej et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3557–3560
3559
3c was found to have an increased potency relative to parent com-
pound 3a at both MMP-13 and -2. The substantial effect of ortho
substitution on selectivity prompted further evaluation of addi-
tional ortho-substituted analogs (3d–i). Generally, MMP-13 poten-
cies were similar and reduced compared to 3a, but IC50’s for MMP-
2 (and thus the MMP-2/13 selectivity ratio) corresponded approx-
imately to the size of the substituent, with methoxy being optimal:
H < Cl, OH < CH3, CF3 < OMe, OEt, 4-F–C6H4. The effect of an addi-
tional substituent was explored in an attempt to increase potency
for MMP-13 while maintaining micromolar affinity for MMP-2. The
1-naphthyl derivative 3j was slightly more potent than the ortho-
methoxy analog 3b at MMP-13, but potency at MMP-2 increased
7-fold. Other disubstituted analogs (3k–n) showed a similar trend,
except for 3n with a 2-methoxy and a 5-isopropyl substitution
where MMP-13 potency dropped fourfold. Presumably the de-
creased affinity for MMP-13 was due to steric reasons. Comparison
of the MMP-2/13 selectivity for ortho-methoxy substituted N-aryl
piperazine 2e (2.8-fold) with that of the 4-arylpiperidine analog
3b (211-fold) is noteworthy. Presumably, 3b adopts a conforma-
tion where the aryl group is orthogonal to the piperidine ring, evi-
denced by the substantial effect of ortho-substitution on
selectivity. The energetic penalty for an N-aryl piperazine to adopt
such a conformation would be high, which is likely responsible for
the reduced potency of 2e at MMP-13 and the increased potency at
MMP-2 relative to 3b.
Scheme 5. Solution phase synthesis of 4-arylpiperidine sulfone hydroxamates.
and
a
-piperidine analogs in our earlier MMP-1 sparing series,6
although this single pair does not necessarily constitute a trend.
The ortho-fluorinated derivative 2b exhibited similar potency to
the N-phenyl parent 2a, whereas the bulkier ortho-methyl and
chloro derivatives 2c–2d dropped five-fold in potency for MMP-
13, and the ortho-methoxyderivative 2e dropped 76-fold in potency.
The meta-derivatives 2f and 2g suffered a similar loss in potency for
MMP-13. On the other hand, para-substituted derivatives main-
tained high potency and selectivity for MMP-13, in particular
para-methoxy derivative 2h with an IC50 = 0.5 nM for MMP-13, a
40-fold selectivity versus MMP-2, and the highest selectivity
observed versus MMP-14 (>20,000). para-Methyl and para-trifluo-
romethyl derivatives 2i and 8b exhibited good potency for MMP-
13 (1.9 and 2.4 nM, respectively) and selectivity versus MMP-14
Based on the superior MMP-13 potency and dual MMP-1 and -
14 sparing profiles of the p-substituted N-aryl piperazines, addi-
tional analogs were prepared for more thorough enzyme and PK
evaluation (Table 3). Potent MMP-13 inhibition was observed for
compounds 2l, 2m and 8c with IC50’s of 0.6, 1.0, 0.5 nM, respec-
tively. Selectivity versus other MMP family members was generally
ꢁ100-fold except for MMP-2 (4–20-fold) and MMP-3 (58–500-
fold). Rat PK for these three compounds showed low to moderate
values for half-life and bioavailability. Aryl piperazine 8c had an
acceptable BA of 20.7%, but a very short t1/2 of only 0.24 h. Aryl
piperidine 2l exhibited a modest bioavailability of 16%, but a much
improved half-life of 2.59 h, which we attribute to the trifluoro-
methylphenyl moiety in P01, which has enhanced the PK of other
series as well. 4-Chlorophenyl piperidine 2m possessed a longer
half-life but a disappointing BA of only 7.4%. Included for compar-
(both >4000ꢀ), noting that 8b is an
a-piperidine. The more sterically
demanding 2,4-dimethylphenylpiperazine 2j suffered a drop in
potency at MMP-13 (IC50 = 28.6 nM), although selectivity against
MMP-2 was the highest of all N-arylpiperazines at approximately
50ꢀ.
Table 2 shows MMP inhibitory potencies of 4-aryl piperidine a-
sulfone hydroxamates (3). 4-Phenylpiperidine 3a was 3ꢀ less po-
tent at MMP-13 than N-phenylpiperazine 2a but its potency for
MMP-2 increased to 4.4 nM, making 3a equipotent for MMP-13
and MMP-2. A substantial boost in MMP-13 selectivity was
achieved by the presence of an ortho-methoxy substituent (3b). Po-
tency of 3b for MMP-13 dropped threefold from the parent com-
pound (3a), while MMP-2 potency dropped 840-fold, generating
a selectivity ratio of 211X. On the other hand, para-chloro analog
ison is broader-spectrum, MMP-1 sparing
pound SC-276 lacks selectivity among MMPs, only significantly
a
-sulfone SC-276.6 Com-
Table 1
MMP inhibitory potency of N-aryl piperazine
a
-sulfones 2 (X = O) and 8 (X = N-cPr)
Compd
X
R
IC50 (nM) at MMP-X
Ratio MMP 2/13
2
9
13
14
2a
2b
2c
2d
2e
2f
2g
2h
2i
O
O
O
O
O
O
O
O
O
O
N-cPr
N-cPr
H
2-F
2-Me
2-Cl
2-MeO
3-MeO
3-CF3
4-MeO
4-Me
2,4-diMe
H
23.5
21.1
88.0
40
370
48.5
330
18.4
35.0
1400
5.4
450
356
2860
2090
—
163
—
722
682
—
1.7
2.7
>10,000
>10,000
>10,000
>10,000
—
2860
—
>10,000
8220
—
13.8
7.8
11.4
4.4
2.8
6.1
16.5
29
18.4
49.0
1.6
7.7
9.0
130
8.0
20
0.63
1.9
28.6
3.3
2.4
2j
8a
8b
—
—
4-CF3
32.6
>10,000
>10,000
13.6

3560
S. A. Kolodziej et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3557–3560
Table 2
MMP inhibitory potency of N-aryl piperidine
a-sulfones (3)
Compd
R1
R2
IC50 (nM) at MMP-X
13
6.0
Ratio MMP 2/13
2
9
14
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
H
MeO
H
Cl
Me
CF3
EtO
OH
H
H
4-Cl
H
H
H
H
H
H
4.4
3700
2.0
82
700
900
3100
70
5400
500
505
1750
800
265
6700
350
—
—
—
>10,000
>10,000
>10,000
—
—
—
0.7
211
2.9
17.5
0.7
42.5
32.7
28.8
35.0
11.3
33.7
11.4
7
1.9
21.4
31.3
88.6
6.2
160
43.9
72.1
159
70.2
143
—
—
>10,000
—
2501
>10,000
>10,000
—
>10,000
—
>10,000
>10,000
>10,000
—
4-F–C6H4
2,3-(CH@CH) (naphthyl)
Me
MeO
MeO
MeO
4-MeO
4-diMeO
5-diMeO
5-iPr
11
11.4
70
>10,000
—
—
Table 3
MMP inhibitory potency of N-aryl piperazine
a
-tetrahydropyranyl sulfone hydroxamates 8c and 2 compared with SC-276
Compd
X
R1
IC50 at MMP-X (nM)
cLog P
Cmax
C6h
t1/2 (h)
BA (%)
1
2
3
7
8
9
13
14
2l
2m
O
O
cPr-N
CF3
Cl
OCH3
>10000
>10000
>10000
8660
12.1
3.7
5.3
300
130
24.5
13.0
>10000
—
—
670
1000
59.2
1.8
5000
270
729
0.6
1.0
0.42
0.40
>10000
5000
>10000
19
1.76
1.46
1.32
2.04
4210
49
751
10
36
19
2.59
1.91
0.24
1.1
16.3
7.4
20.7
28
8c
SC-276a
0.33
>10000
1.5
13630
281
a
Data for SC-276 from Ref. 6, except for MMP-7 and MMP-14 were not previously reported.
sparing MMP-1, yet this compound has the very high exposure in
the rat that is compelling for development, consistent with its po-
tent and efficacious antitumor activity.6
weightspecies, while applying our learnings about P01 manipulations
toward optimizing MMP-13 selectivity.
In summary, the related series of compounds described herein
have demonstrated single-digit to sub-nanomolar potency for
MMP-13 combined with exceptional selectivity versus MMP-1 and
MMP-14 of typically >100ꢀ up to 20,000ꢀ. Selectivities versus other
MMPs when tested varied for MMP-3 (40–500ꢀ), MMP-8 (140–
2500ꢀ) and for MMP-9 (20 to >4000ꢀ). Selectivity for MMP-13 over
MMP-2 ranged from equipotency to 200ꢀ, thus selectivities were
References and notes
1. Brinckerhoff, C. E.; Matrisian, L. M. Nat. Rev. Mol. Cell Biol. 2002, 3, 207.
2. Nagase, H.; Woessner, J. F., Jr. J. Biol. Chem. 1999, 274, 21491.
3. Peterson, J. T. Heart Fail. Rev. 2004, 9, 63.
4. Kolodziej, S. A.; Hockerman, S. L.; DeCrescenzo, G. A.; McDonald, J. J.; Mischke, D.
A.; Munie, G. E.; Fletcher, T. R.; Stehle, N.; Swearingen, C.; Becker, D. P. Bioorg.
Med. Chem. Lett. 2010, 20, 3561.
5. Freemont, A. J.; Byers, R. J.; Taiwo, Y. O.; Hoyland, J. A. Ann. Rheum. Dis. 1999, 58,
357.
6. Becker, D. P.; Villamil, C. I.; Barta, T. E.; Bedell, L. J.; Boehm, T. L.; DeCrescenzo, G.
A.; Freskos, J. N.; Getman, D. P.; Hockerman, S.; Heintz, R.; Howard, S. C.; Li, M.
H.; McDonald, J. J.; Carron, C. P.; Funckes-Shippy, C. L.; Mehta, P. P.; Munie, G. E.;
Swearingen, C. A. J. Med. Chem. 2005, 48, 6713.
7. Burns, D. M.; He, C.; Li, Y.; Scherle, P.; Liu, X.; Marando, C. A.; Covington, M. B.; Yang,
G.; Pan, M.; Turner, S.; Fridman, J. S.; Hollis, G.; Vaddi, K.; Yeleswaram, S.; Newton,
R.; Friedman, S.; Metcalf, B.; Yao, W. Bioorg. Med. Chem. Lett. 2008, 18, 560.
8. Wustrow, D. J.; Wise, L. D. Synthesis 1991, 993.
9. Veber, D. F.; Johnson, S. R.; Cheng, H.; Smith, B. R.; Ward, K. W.; Kopple, K. D. J.
Med. Chem. 2002, 45, 2615.
somewhat lower relative to the related isonipecotate
a-sulfone
hydroxamate series.4 Rat PK for selected members of these series
demonstrated bioavailabilities of up to 20.7% (8c) and a half lives
of up to 2.6 h (2l) yet individual compounds lacked a compelling
complete package to initiate development. Undoubtedly the high
molecular weights of these analogs (551 a.u. for 2m) plays a role gi-
ven the recommendations of Lipinski although the limited number
of rotatable bonds9 favors the limited bioavailability that was ob-
served. We therefore turned our attention to lower molecular