Novel thiol-based TACE inhibitors. Part 2: Rational design, synthesis, and SAR of thiol-containing aryl sulfones

Article abstract of DOI:10.1016/j.bmcl.2007.11.014

A series of potent thiol-containing aryl sulfone TACE inhibitors were designed and synthesized. The SAR and MMP selectivity of the series were investigated. In particular, compound 8b showed excellent in vitro potency against the isolated enzyme and good selectivity over MMP-2, -7, -8, -9, and -13. The X-ray structure of 8b in complex with TACE was also obtained.

Full text of DOI:10.1016/j.bmcl.2007.11.014

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 44–48
Novel thiol-based TACE inhibitors. Part 2: Rational
design, synthesis, and SAR of thiol-containing aryl sulfones
Upul K. Bandarage,* Tiansheng Wang, Jon H. Come, Emanuele Perola,
Yunyi Wei and B. Govinda Rao
Vertex Pharmaceuticals Incorporated, 130 Waverly Street, Cambridge, MA 02139, USA
Received 13 September 2007; revised 6 November 2007; accepted 7 November 2007
Available online 13 November 2007
Abstract—A series of potent thiol-containing aryl sulfone TACE inhibitors were designed and synthesized. The SAR and MMP
selectivity of the series were investigated. In particular, compound 8b showed excellent in vitro potency against the isolated enzyme
and good selectivity over MMP-2, -7, -8, -9, and -13. The X-ray structure of 8b in complex with TACE was also obtained.
Ó 2007 Elsevier Ltd. All rights reserved.
Rheumatoid arthritis (RA) is an autoimmune disease
which affects over 2 million people in the United States
alone.1 The disease is currently treated with non-steroi-
dal anti-inflammatory drugs (NSAIDs),1 disease-modi-
fying anti-rheumatoid drugs (DMARD),1 and,
recently, COX-2 selective inhibitors.2 Due to limited
effectiveness and/or side effects of these drug classes,
development of active small molecule TACE (TNF-a
converting enzyme) inhibitors might offer a higher ther-
apeutic value. TNF-a (tumor necrosis factor) is a cyto-
kine that is highly expressed in patients with RA.3
TACE is responsible for the release of soluble TNF-a
from its parent membrane-bound form. Recently, a
number of small molecule TACE inhibitors have been
reported to be highly potent and very selective over
other matrix metalloproteases (MMPs).4 Most TACE
and MMP inhibitors contain a hydroxamic acid moiety
as the zinc-binding group (ZBG).4 Generally, hydroxa-
mic acids exhibit poor oral absorption in vivo and sig-
nificant metabolic liabilities (rapid hydrolysis and
glucuronidation).5 Therefore, we were interested in
exploring non-hydroxamates as ZBGs for our investiga-
tion. Previous efforts utilizing a thiol as the ZBG group
offered an advantage in imparting specificity in the inhi-
bition of MMPs.6 However, the use of a thiol as ZBG in
TACE inhibitors has not been widely explored. In our
previous paper,7 we disclosed a series of thiol-containing
sulfonamides that are very potent and selective TACE
inhibitors. Compound 2 had shown excellent in vitro
potency against the isolated TACE enzyme and good
selectivity over MMP-2, -7, -9, and -137 (Fig. 1). The
TACE potency of 2 is slightly lower than that of
hydroxamic acid analog 1, while its selectivity over other
MMPs such as MMP-2, -8, and -13 is slightly higher.7
Examination of the crystal structure of the thiol-based
sulfonamide inhibitor in complex with TACE showed
that the sulfonamide nitrogen is not required for any
specific interaction and can be substituted with a meth-
ylene group. Modeling suggested that replacement of the
sulfonamide group with a sulfone would not alter the
bound conformation of the inhibitor and would main-
HS
H
N
OH
N
S
N
S
O
O
O
O
O
O
O
Keywords: TACE inhibitors; MMP selectivity; Rheumatoid arthritis;
Hydroxamates; Sulfones.
2 (Ki = 28 nM)
Figure 1. Hydroxamic acid and thiol-based TACE inhibitors.
1 (Ki = 10 nM)
*
Corresponding author. Tel.: +1 617 444 6882; fax: +1 617 444
7827; e-mail: upul_bandarage@vrtx.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.014

U. K. Bandarage et al. / Bioorg. Med. Chem. Lett. 18 (2008) 44–48
45
tain all the key interactions with the enzyme active site.
In this study, we extended our investigation to design
and synthesize thiol-containing sulfones as selective
TACE inhibitors.
ethanol gave 10 in high yield. Oxidation of 10 with m-
CPBA followed by alkylation with butynyl bromide
and removal of the THP group afforded 11. Mitsunobu
reaction of 11 gave thioacetate 12 in good yield. Hydro-
lysis of thioacetate 12 with 4 N HCl in dioxane gave de-
sired thiol 13 in moderate yield. Michael reaction of 9
with acrolein gave aldehyde 14 which was reduced with
NaBH4 in methanol to afford 15. Compound 15 was
converted to desired thiol 16 in a manner similar to com-
pound 13.
The general synthesis of sulfone thiols with butynyloxy
tail is outlined in Scheme 1. Selective S-alkylation of
the commercially available 4-hydroxythiophenol 3 with
2-chloroethanol or 3-chloropropanol and subsequent
O-alkylation with butynyl bromide provided analogs 5
in workable yields. Subsequent oxidation of 5 with ox-
one gave sulfones 6, which were then subjected to stan-
dard Mitsunobu conditions to provide thioacetates 7 in
good yield. Hydrolysis of thioacetates 7 with aqueous
sodium methoxide afforded the desired sulfone thiols 8.
The synthesis of cyclic sulfone thiol 25 (Scheme 4)
began with Michael addition of 9 to cyclohexenone to
provide compound 17 in high yield. Reduction of ketone
17 with NaBH4 followed by protection of the resulting
alcohol with TBDMSCl gave compound 18 in moderate
yield. Compound 18 was then alkylated with butynyl
bromide to afford compound 19 in good yield. Oxone
oxidation of 19 provided sulfone 20 and the subsequent
The synthesis of sulfone thiols-containing a butynylami-
no group is outlined in Schemes 2 and 3. Alkylation of
commercially available 9 with THP-protected 2-bromo-
OH
OH
(C)n
(C)n
S
SH
S
b
a
OH
OH
O
3
4a, n=1
4b, n=2
5a, n=1
5b, n=2
O
O
O
SAc
e
OH
d
SH
(C)n
(C)n
(C)n
S
S
S
O
O
O
c
O
O
O
7a, n=1
7b, n=2
8a, n=1
8b, n= 2
6a, n=1
6b, n= 2
Scheme 1. Reagents and conditions: (a) Cl(CH2)nOH, K2CO3/DMF 4a 49% 4b 68%; (b) CH3CH„CHCH2Br, K2CO3, DMF, 5a 36% 5b 41%; (c)
oxone, MeOH/H2O 6a 82% 6b 99%; (d) Ph3P, DEAD, CH3COSH, 7a 07% 7b 59%; (e) i—10% NaOCH3, MeOH; ii—10% HCl, 8a 70% 8b 71%.
O
O
OTHP
OH
S
SH
S
a
b, c, d
NH
NH
N
O
O
O
10
9
11
O
O
SAc
O
O
SH
S
S
e
f
N
HN
13
O
12
Scheme 2. Reagents and conditions: (a) THPO(CH2)2Br, K2CO3, acetone, 82%; (b) m-CPBA, CHCl3, 85%; (c) NaH, CH3CH„CHCH2Br, DMF;
(d) TsOH, MeOH/CH2Cl2, 36% over 2 steps; (e) Ph3P, DEAD, CH3COSH; (f) 4 M HCl in dioxane, 60 °C, 30 min, 28% over 2 steps.

46
U. K. Bandarage et al. / Bioorg. Med. Chem. Lett. 18 (2008) 44–48
O
CHO
O
O
S
S
OH
S
SH
O
a
b, c
d, e, f, g, h
9
NH
NH
HN
16
O
O
14
15
Scheme 3. Reagents and conditions: (a) acrolein, CHCl3; (b) NaBH4, MeOH; (c) m-CPBA, CHCl3, 96% over 3 steps; (d) t-BuMe2SiCl, Et3N; (e) NaH,
CH3CH„CHCH2Br, DMF, (f) HCl in dioxane, 0 °C, 30 min, 88% over 3 steps; (g) Ph3P, DEAD, CH3COSH; (h) 4 M HCl in dioxane, 60 °C, 30 min,
22% over 2 steps.
O
O
OTBS
a
b, c
S
O
S
O
N
H
N
H
17
18
OTBS
OTBS
d
e
O
S
S
O
O
O
N
N
19
20
OH
H
N
f
g, h
O
S
O
O
S
O
R
N
SH
25
21 R= Ac
22 R = H
Scheme 4. Reagents and conditions: (a) i—9, Et3N, MeOH, rt, 1 h; ii—recrystallization, 97%; (b) NaBH4, MeOH, 78%; (c) TBSCl, imidazole, DMF,
rt, 71%; (d) NaH, CH3CH„CHCH2Br, DMF, 3 h, 56%; (e) oxone, MeOH, 78%; (f) TBAF, THF, rt, 21 42%, 22 21%; (g) Ph3P, DEAD, CH3COSH,
75%; (h) 4 M HCl in dioxane, 63%.
fluoride-mediated removal of the TBS group resulted in
compounds 21 and 22 in a 2:1 ratio. Compound 22 was
subjected to Mitsunobu conditions followed by hydroly-
sis of the resulting thioacetate, affording the desired thiol
25 in good yield. Compounds 23 and 24 (Fig. 2) were
synthesized in a manner similar to 25 starting from
cyclohexenone or cyclopentenone and 4-hydroxythi-
ophenol. All final compounds were characterized by
HPLC and 1H NMR analysis. Purity of these com-
pounds was >98%.
In our previous publication on sulfonamides,7 we found
a three-atom spacer between the sulfonyl SO2 and the
thiol group to be optimal among the linear inhibitors.
The same spacer length in the current sulfone series pro-
duced a slightly more potent compound, 8b. We have
obtained the X-ray crystal structure of this compound
in complex with TACE,8 which is shown in Figure 3.
Compound 8b binds to the enzyme active site in the
same orientation as the sulfonamide inhibitor 2 described
in our previous publication.7 The thiol group makes a
strong interaction with the active site Zn2+ ion forming
a tetrahedral coordination around the metal ion along
with the three histidine side-chains (His-405, His-409,
and His-415). The hydrogen bond interaction between
the sulfone oxygen and the backbone NH of Gly-349 re-
mains strong. The phenyl ring of the inhibitor is in the
hydrophobic environment of the S10 pocket and is also
involved in a stacking interaction with the His-405
side-chain. The butynyloxy tail occupies the narrow
channel between the S10 and S30 pockets.
O
S
X
O
S
O
X
O
SH
SH
23
24 X = O
25 X = NH
X = O
Figure 2.

U. K. Bandarage et al. / Bioorg. Med. Chem. Lett. 18 (2008) 44–48
47
Figure 3. X-ray crystal structure of compound 8b bound to TACE.
A closer examination of the X-ray crystal structure
showed that the thiol moiety can be placed near the
Zn2+ ion using a spacer that is one bond shorter. There-
fore, compound 8a with a shorter spacer was made and
was found to be as potent as compound 8b. The replace-
ment of the butynyloxy tail of these two compounds
with butynylamino tails (compounds 13 and 16) resulted
in very potent TACE inhibitors (Table 1, entries 3 and
4). These compounds are 2-fold more potent than their
butynyloxy counterparts. However, cyclization of the
linker into five-membered and six-membered rings pro-
duced compounds with reduced potencies. Cyclopentyl-
sulfone thiol 23 is 5-fold less potent while cyclohexyl
sulfone thiol 24 is 3- to 4-fold less potent than the corre-
sponding acyclic inhibitors. Cyclohexyl sulfone thiol 24
appears to be slightly more potent than its cyclopentyl
analog 23 (Table 2, entries 1 and 2). Replacement of
the butynyloxy tail of compound 24 with butynylamino
in compound 25 showed the usual 2-fold improvement
in Ki (Table 2, entries 1 and 3). It should be noted that
cyclization of the spacer did not affect binding in the
case of sulfonamides,7 whereas we observe a significant
loss in binding in the current sulfone series. Cyclic sulf-
Table 2. In vitro potency of thiol-aryl sulfone series against TACE
O
S
X
O
n(C)
SH
Entry
Compound
n
X
TACE Ki (nM)
1
2
3
23
24
25
1
2
2
O
O
N
50
30
17
ones (23, 24, 25) have one additional chiral center com-
pared to the sulfonamide compounds and were
produced as racemic mixtures of four compounds. Mod-
eling analysis showed that some diastereomers of these
compounds do not fit the active site as well as the linear
compounds and are likely to have lower binding affini-
ties for TACE. It is possible that the potency of a partic-
ular diastereomer is similar or better than those of the
corresponding acyclic compounds (Table 3).
Table 1. In vitro potency of thiol-aryl sulfone series against TACE
In summary, we have designed and synthesized a novel
series of thiol-containing aryl sulfones as inhibitors of
TACE. Most of these compounds show very potent
inhibition in an enzyme assay using the isolated enzyme.
One of the potent TACE compounds, 8b, is far more
selective against MMP-2, -7, -8, -9, -13 than previously
reported analogs such as 1. These results also demon-
strate that in the scaffold explored here the sulfonamide
moiety can be replaced with a sulfone without loss of
TACE potency. The crystal structure of an inhibitor of
this novel class bound to TACE enhances our under-
standing of the binding interactions with the TACE ac-
X
HS
(C)n S
O
O
Entry
Compound
n
X
TACE Ki (nM)
1
2
3
4
8a
8b
13
16
1
2
1
2
O
8
10
2
O
NH
NH
4.5

48
U. K. Bandarage et al. / Bioorg. Med. Chem. Lett. 18 (2008) 44–48
Table 3. Selectivity profile of selected compounds
Compound
TACE Ki (nM)
MMP-2 Ki (nM)
MMP-7 Ki (nM)
MMP-8 Ki (nM)
MMP-9 Ki (nM)
MMP-13 Ki (nM)
1
10
28
10
27
5000
1300
3400
>6000
>2500
43
1900
2700
800
200
17
1200
1700
2
8b
6300
2. Penning, T. D.; Talley, J. T.; Bertenshaw, S. R.; Carter, J.
S.; Collins, P. W.; Docter, S. D.; Graneto, M. J.; Lee, L. F.;
Malecha, J. W.; Miyashiro, J. M.; Rogers, R. S.; Rogier, D.
J.; Yu, S. S.; Anderson, G. D.; Burton, E. G.; Cogburn, J.
N.; Gregory, S. A.; Koboldt, C. M.; Perkins, W. E.; Seibert,
K.; Veenzen, A. W.; Zhang, Y. Y.; Isakson, P. C. J. Med.
Chem. 1997, 40, 1347.
tive site and facilitates the structure-based design of
inhibitors with improved potency. Further exploration
of this novel class of TACE inhibitors may provide
new opportunities for the treatment of rheumatoid
arthritis.
3. Feldmann, M.; Brennan, F.; Elliot, M.; Katsikis, E.; Maini,
R. Circ. Shock 1994, 43, 179.
Acknowledgments
4. Levin, J. I. Curr. Top. Med. Chem. 2004, 4, 1289.
5. Duan, J. J.-W.; Lu, Z.; Wasserman, Z. R.; Liu, R.-Q.;
Covington, M. B.; Decicco, C. P. Bioorg. Med. Chem. Lett.
2005, 15, 2970.
We thank Dr. Youssef Bennani, Dr. Jeremy Green, and
David Deininger at Vertex Pharmaceuticals for useful
comments and suggestions.
6. Freskos, J. N.; Mischke, B. V.; DeCrescenzo, G. A.; Heintz,
R.; Getman, D. P.; Howard, S. C.; Kishore, N. N.;
McDonald, J. J.; Munie, G. E.; Rangwala, S. Bioorg.
Med. Chem. Lett. 1999, 9, 943.
References and notes
1. Duan, J. J.-W.; Chen, L.; Wasserman, Z. R.; Lu, Z.; Liu,
R.-Q.; Covington, M. B.; Qian, M.; Hardman, K. D.;
Magolda, R. L.; Newton, R. C.; Christ, D. D.; Wexler, R.
R.; Decicco, C. P. J. Med. Chem. 2002, 45, 4954.
7. Rao, B. G.; Bandarage, U. K.; Wang, T.; Come, J. H.;
Perola, E.; Wei, Y.; Tian, S.; Saunders, J. O. Bioorg. Med.
Chem. Lett. 2007, 17, 2250.
8. PDB deposition number: 3B92.