Solid-phase synthesis of an arylsulfone hydroxamate library

Article abstract of DOI:10.1016/S0960-894X(00)00285-7

Synthesis of an arylsulfone hydroxamate lead optimization library is presented. Biological activity of representative examples is given to demonstrate the value of this approach for lead optimization. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00285-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1637±1640
Solid-Phase Synthesis of an Arylsulfone Hydroxamate Library
Joseph M. Salvino,* Rose Mathew, Terence Kiesow, Ramesh Narensingh,
Helen J. Mason, Amy Dodd, Robert Groneberg, Christopher J. Burns,
Gerald McGeehan, Jane Kline, Edward Orton, Sheng-Yuh Tang,
Mathew Morrisette and Richard Labaudininiere
Rhone Poulenc Rorer, Lead Discovery and Medicinal Chemistry Departments, 500 Arcola Road, Collegeville, PA 19426, USA
Received 12 November 1999; accepted 8 May 2000
AbstractÐSynthesis of an arylsulfone hydroxamate lead optimization library is presented. Biological activity of representative
examples is given to demonstrate the value of this approach for lead optimization. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
Chemistry
The solid-phase synthesis and biological activity of an
arylsulfone hydroxamate library is described. The goal
of this work was to develop a solid-phase route to the
arylsulfone hydroxamate chemical series1 to facilitate
optimization of this lead series. The preliminary SAR on a
b-arylsulfone hydroxamate sca€old capable of inhibiting
either MMPs or PDE4 has recently been disclosed. These
targets are of considerable interest as anti-in¯ammatory
agents.1,2 As part of this study a general synthetic route
for chemical libraries of carboxylic and hydroxamic
acids was designed and carried to fruition. In this report
we wish to disclose this solid-phase synthesis and the
resulting SAR required for potent, selective MMP or
PDE4 inhibitors.
A ®ve-step solution route to the arylsulfone inhibitors
(compound 8; Scheme 2), had been developed prior to
starting this project.1 However, this route involved
chromatographic puri®cation after each synthetic step.
A robust solid-phase route would facilitate puri®cation
and thus make the entire process more ecient. A solid-
phase synthesis of the carboxylic acid sulfone was
developed (Scheme 1).3 This synthesis was ¯exible
enough to allow for the isolation and screening of all the
carboxylic acid intermediates. A resin bound hydroxyl-
amine was employed to transform the carboxylic acid into
the hydroxamic acid4 (Scheme 2). Typically, the reaction
products displayed >80% purity as determined by
Scheme 1. Reagents. (a) Phosphonoacetic acid (3 equiv), 2,6-dichlorobenzoylchloride (3.2 equiv), anhydrous pyridine (6.4 equiv), DMF, 25ꢀC, 8 h; (b)
Lithium bis(trimethylsilyl)amide (5 equiv), THF, 0 ꢀC to 25ꢀC, 60 min then ®lter under argon, add R1CHO (4 equiv), 60% cyclohexane in THF 25 ꢀC 24±
48 h; (c) HSR2 (5 equiv), THF, nBuLi (0.1 equiv), 25ꢀC, 12 h; (d) mCPBA (5 equiv), dioxane, 25ꢀC, 12h; (e) 30% TFA in CH2Cl2 (excess), 25ꢀC, 1 h.
*Corresponding author. Adolor Corp., 371 Phoenixville Pike, Malvern,
PA 19355, USA. Tel.: +1-610-287-6953; fax: +1-610-889-2203; e-mail:
Jsalvino@aol.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00285-7

1638
J. M. Salvino et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1637±1640
Scheme 2. Reagents. (a) EDC (3 equiv), DMF, Hydroxylamine resin, DMF, 25 ꢀC 4 h; (b) 30% TFA in CH2Cl2 (excess), 25 ꢀC, 1 h.
Table 1. Structure±activity relationships for PDE4 activitya
Compound
R1
R2
PDE4 IC50
nMb
9
10d
-(CH2)4-Ph
-(CH2)4-Ph
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
6.9
1.3
11
12
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
11
6.4
13c,d
14
15
16c
17
18
19
20
21
22
23
24
25
26
27
28
29
-(CH2)3-Ph
-(CH2)2-Ph
-(CH2)2-Ph
-Ph
-C6H4-4-OBn
-C6H3-4-OMe-3-OMe
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
-C6H3-4-OMe-3-OMe
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
14
31
3030
580
21
89
1500
63
64
32
260
49
9.5
55
-C6H4-3-OBn
-C6H4-2-OBn
-(CH2)2-C6H4-4-OBn
-(CH2)2-C6H4-4-OBn
-(CH2)2-C6H4-4-OPh
-C6H4-3-OPh
-C6H4-3-O-C6H4-4-OMe
-(CH2)5CH3
-(CH2)2CH3
-(CH2)2CH3
-CH2CH(CH3)2
-CH2CH(CH3)2
5250
102
>10,000
30
31
32
33
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OMe
16
564
36
295
34
35
36
37
38
39
40
-(CH2)3-Phthalimide
-(CH2)3-CON(CH3)Ph
-(CH2)2-CON(CH3)Ph
-(CH2)2-OCON(CH3)Ph
-(CH2)2-N(CH3)COPh
-CH2-N(CH3)COPh
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
-C6H3-4-OMe-3-OcC5H9
32
118
50
140
1040
974
460
-(CH2)2-N(CH3)CO2Bn
aMMP IC50s are all >10,000 nM, see ref 7.
bSee ref 6.
cSee ref 8; Compound 13 inhibits TNF a IC50=9 nM; Compound 16 inhibits TNF a IC50=450 nM.
dPuri®ed analogues.

J. M. Salvino et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1637±1640
Table 2. Structure±activity relationships for MMP activity
1639
Compound
R1
TNF a
PDE4
IC50 nMa,b
MMP-1,
Ki (nM)c
MMP-2,
Ki (nM)c
MMP-3,
Ki (nM)c
IC50 nMd
41
42
43
44
45
46
47e
48
49
50
51
52
53
54
55
56
57
58
-(CH2)4-Ph
-(CH2)3-Ph
-CH2-O-CH2-Ph
-(CH2)2-Ph
-(CH2)2-Cyclohexyl
-CH-(CH3)Ph
-Ph
-C6H4-4-CO2Me
-C6H4-4-OBn
-C6H4-4-OPh
>10,000
5000
NA
NA
NA
NA
NA
NA
4600
NA
520
280
NA
3600
NA
NA
NA
NA
NA
NA
5000
2580
>10,000
2160
400
1710
>10,000
1900
>10,000
>10,000
>10,000
>10,000
1710
20
44
38
43
36
8
270
700
1390
1430
350
590
300
>10,000
>10,000
>10,000
>10,000
>10,000
>10,000
1200
730
>10,000
4400
>10,000
4600
1400
>10,000
8000
>10,000
3300
4420
>10,000
>10,000
>10,000
>10,000
86
88
>10,000
>10,000
>10,000
>10,000
8
-C6H4-O-isoPropyl
-C6H4-4-C6H4
-cyclo-C6H11
-cyclo-C6H10-3-CO2Et
-(CH2)5CH3
-CH(Et)(CH2)3CH3
-CH2CH(CH3)CH2C(CH3)3
-(CH2)3CH3
1010
9590
357
1720
0.8
31
4
27
30
60
734
184
676
4170
404
aNA=data not available.
bSee ref 6.
cSee ref 7.
dSee ref 8.
ePuri®ed analogues.
HPLC/MS using UV220 detection. This purity was su-
cient for biological evaluation and generation of struc-
ture±activity relationships. To con®rm that the observed
activity was derived from the expected compound, several
analogues in the library were puri®ed and re-screened.
Potent, selective MMP-2 inhibitors were found in the
library. Due to the focus towards PDE4 inhibitors in
this study only three R2 groups were examined. Only the
R2=4-methoxy-phenyl analogues showed activity for
the MMPs from this set of analogues. Variation in the
R1 position revealed interesting results. For example,
alkyl groups were well tolerated, a cyclohexyl is better than
a phenyl moiety (47 vs 53). Note the 10-fold increase
in potency in the 3-carbethoxy-cyclohexyl analogue, (53
versus 54), resulting in a sub-nanomolar inhibitor
against MMP-2 for this series.
Biological Results
Presented in Tables 1 and 2 are representative biological
data to highlight this chemical library as a tool for gen-
erating SAR data rapidly. Carboxylic acid intermediates
were screened but showed minimal activity at 10 mM
concentration and therefore are not presented in the
tables. The purpose of this study was to develop PDE4
SAR for the arylsulfonehydroxamate series. Screening
against the various MMPs was performed mainly as a
selectivity screen. Screening results indicate that a 3,4-dia-
lkoxy-substituted moiety in R2 such as 3,4-dimethoxy-
phenyl group provided inhibitors of PDE4 e€ective at
nanomolar concentrations while suppressing activity
against MMPs (Table 1). The use of the 3-O-cyclopentyl-
4-O-methyl-benzene moiety for R2 in some cases enhanced
PDE4 activity in this series5 (compounds 14±15, 26±27,
28±29, 30±31 and 32±33). A variety of R1 groups were
tolerated and reveal interesting SAR. The length of the
alkyl sidechain is important for activity (compounds 25
and 26). Importantly polar R1 substituents were toler-
ated with only a 10 to 20 fold loss in activity (compounds
30 and 40). Polar R1 substituents may enhance the water
solubility of the analogues, thus giving a handle to bal-
ance the pharmokinetic pro®le within the series.
Conclusion
In summary, a novel seven-step, solid-phase synthesis of
potent MMP and PDE4 inhibitors has been accom-
plished. Both carboxylic and hydroxamic acids are
available for screening from this synthetic route. This
type of lead optimization library demonstrates the utility
and the potential to accelerate the optimization process.
Biological data could be generated directly from library
compounds without any special puri®cation. Selective
compounds were shown to speci®cally inhibit several
related enzymes and structure±activity relationships
were easily extracted from the library.
Acknowledgements
The authors would like to thank Dr. Richard Harrison,
Dr. Joanne Uhl, Dr. John Souness, and Ms. Allison

1640
J. M. Salvino et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1637±1640
Capolino for help in setting up the biological assays and
helpful discussions.
tion. N-(3,5-Dichloro-N-4-pyridyl-3-(cyclopentyloxy)-4-methoxy-
benzamide (RP73401) inhibits TNF-a release IC50=5 nM
and was used as a control. (a) In vitro inhibitory e€ects on
TNF-a release by human monocytes. The e€ects of com-
pounds on TNF-a production by human peripheral blood
monocytes (PBMs) are examined as follows: blood is drawn
from normal donors, mixed with dextran, and the erythrocytes
allowed to sediment for 35 min at 37 ꢀC. Leukocytes are frac-
tionated by centrifugation through a discontinuous (18, 20,
and 22%) metrizamide gradient. The mononuclear cell frac-
tion comprising 30±40% PBMs is suspended in Hank's
balanced salt solution and stored at 4 ꢀC until use. Cells from
the PBM-rich metrizamide fraction are spun down (200 g for
10 min at 20 ꢀC), resuspended at 106 PBM s/mL of medium;
RPMI 1640 containing 1% v/v FCS, 50 U/mL penicillin and
50 mg/mL streptomycin (Gibco, UK), then plated out in 96-
well plates at 2Â105 cells/well. The medium (200 mL) is chan-
ged to remove any non-adherent cells and the remaining,
adherent PBMs left in the incubator overnight (18 h). One h
prior to challenge, the medium is changed to that containing
compound for test or drug vehicle. Control treatments and
compounds for test are assayed in quadruplicate wells. Com-
References and Notes
1. Groneberg, R. D.; Burns, C. J.; Morrisette, M. M.; Ullrich,
J. W.; Morris, R. L.; Darnbrough, S.; Djuric, S.; Condon, S.;
McGeehan, G. M.; Labaudiniere, R.; Neuenschwander, K.;
Scotese, A. C.; Kline, J. J. Med. Chem. 1999, 42, 541.
2. Burns, C. J.; Groneberg, R. D.; Salvino, J.; McGeehan, G.;
Condon, S. M.; Morris, R.; Morrissette, M.; Mathew, R.;
Darnbrough, S.; Neuenschwander, K.; Scotese, A.; Djuric, S.;
Ullrich, J.; Labaudiniere, R. Angew. Chem., Int. Ed. Engl.
1998, 37, 2848.
3. Salvino, J.; Kiesow, T. J.; Darnbrough, S.; Labaudiniere,
R. J. Combi. Chem. 1999, 1, 134.
4. Salvino, J.; Mervic, M.; Mason, H. J.; Kiesow, T.; Teager,
D.; Airey, J.; Labaudiniere, R. F. J. Org. Chem. 1999, 64,
1823.
5. The 3-O-cyclopentyl-4-O-methyl benzene moiety is a com-
mon pharmaaphcore for the rolipram class of PDE4 inhibi-
tors. (a) Activity for RP73401 (compd 15j in the reference)
inhibits PDE4 with an IC50=1 nM; Ashton, M.; Cook, D. C.;
Fenton, G.; Karlsson, J.-A.; Palfreyman, M. N.; Raeburn, D.;
Ratcli€e, A. J.; Souness, J. E.; Thurairatnam, S.; Vicker, N. J.
Med. Chem. 1994, 37, 1696. (b) Kleinman, E. F.; Campbell,
E.; Giordano, L. A.; Cohan, V. L.; Jenkinson, T. H.; Cheng, J.
B.; Shirley, J. T.; Pettipher, E. R.; Salter, E. D.; Hibbs, T. A.;
DiCapua, F. M.; Bordner, J. J. Med. Chem. 1998, 41, 266. (c)
Newton, C.; Decicco, C. P. J. Med. Chem. 1999, 42, 2295. (d)
Christensen, S.; Torphy, T. Annu. Rep. Med. Chem. 1994, 29,
185.
6. PDE4 inhibitory activity was measured as the mean of
three determinations against guinea pig macrophage homo-
genate PDE4 according to the methods of Thompson, W. J.;
Terasaki, W. L.; Epstein, P. M.; Strada, S. J. Adv. Cyclic Nucl.
Res. 1979, 10, 69.
7. Inhibitory activity on recombinant MMP-1, MMP-2, and
MMP-3 (biogenesis) was measured according to the methods
of Knight, C. G.; Willenbrock, F.; Murphy, G. FEBS Lett.
1992, 296, 263.
11
pounds are tested within the concentration range of 3Â10
6
M to 3Â10 M. Medium (50 mL) with or without 10 ng/mL
LPS (E. Coli, 055 B5 from Sigma, UK) is then added. The
incubation is then continued for a further 4 h. Cell super-
natants are removed for storage at 20 ꢀC. TNF-a levels in
cell supernatants are quanti®ed using a standard sandwich
ELISA technique. ELISA plates (Costar, UK) are coated
overnight at 4 ꢀC with 3 mg/mL polyclonal goat anti-human
TNF-a antibody (British Biotechnology, UK) in pH 9.9
bicarbonate bu€er. Rabbit polyclonal anti-human TNF-a
antiserum (Janssen Biochimicha, Belgium) at 1/500 dilution is
used as the second antibody and polyclonal goat anti-rabbit
IgG horseradish peroxidase (Calbiochem, USA) at 1/8000
dilution is used as the detection antibody. Color development
is measured by absorbance at 450 nm using a Titek plate
reader. TNF-a levels are calculated by interpolation from a
standard curve using recombinant human TNF-a (British
Biotechnology UK)(0.125-8 ng/mL). Data (log-concd vs log-
resp) are ®tted by linear regression (P>0.99) using a Multicalc
(Wallac Pharmacia, UK) software program. Basal TNF-a
levels are less than 100 pg/mL whilst LPS (lipopoly-sacchar-
ide) stimulation of the PBMs increases TNF-a levels to 3±10
ng/mL.
8. All compounds shown to inhibit TNF-a release are inactive
against TACE, and presumably inhibit TNF-a by PDE4 inhibi-