Discovery of novel hydroxamates as highly potent tumor necrosis factor-α converting enzyme inhibitors: Part I - Discovery of two binding modes

Article abstract of DOI:10.1021/jm070376o

Through a de novo design approach, hydroxamates derived from trans-cyclopropyl dicarboxylate were examined as potential TNF-α converting enzyme (TACE) inhibitors. Two distinctive series of inhibitors (A and B) were identified and shown to have different structure-activity relationship trends and selectivity profiles against other matrix metalloproteases despite their close structural similarities. X-ray crystallography of the inhibitors binding to the TACE enzyme demonstrates that each series derives its activity from the opposite enantiomer of the cyclopropyl scaffolds, which display almost superimposable hydroxamate groups that coordinate to the zinc at the catalytic site. Mode A inhibitors occupy the S1′-S3′ binding pockets, whereas mode B resides in the nonprime binding sites.

Full text of DOI:10.1021/jm070376o

J. Med. Chem. 2008, 51, 725–736
725
Articles
Discovery of Novel Hydroxamates as Highly Potent Tumor Necrosis Factor-r Converting
Enzyme Inhibitors: Part IsDiscovery of Two Binding Modes†
Zhaoning Zhu,*,‡ Robert Mazzola,‡ Lisa Sinning,‡ Brian McKittrick,‡ Xiaoda Niu,§ Daniel Lundell,§ Jing Sun,§ Peter Orth,‡
Zhuyan Guo,‡ Vincent Madison,‡ Richard Ingram,‡ and Brian M. Beyer‡
Departments of Medicinal Chemistry and Inflammation and Infectious Diseases, Schering-Plough Research Institute, 2015 Galloping Hill Road,
Kenilworth, New Jersey 07033
ReceiVed March 29, 2007
Through a de novo design approach, hydroxamates derived from trans-cyclopropyl dicarboxylate were
examined as potential TNF-R converting enzyme (TACE) inhibitors. Two distinctive series of inhibitors (A
and B) were identified and shown to have different structure-activity relationship trends and selectivity
profiles against other matrix metalloproteases despite their close structural similarities. X-ray crystallography
of the inhibitors binding to the TACE enzyme demonstrates that each series derives its activity from the
opposite enantiomer of the cyclopropyl scaffolds, which display almost superimposable hydroxamate groups
that coordinate to the zinc at the catalytic site. Mode A inhibitors occupy the S1′-S3′ binding pockets,
whereas mode B resides in the nonprime binding sites.
Introduction
protein to its mature soluble form in vitro. It is also involved in
the processing of several cell-surface proteins including Hb-
EGF,7 ephrin-A2,8 APP,9 and Notch.10 ADAM10 deficiency in
mice is lethal at the embryonic stage. Despite the fact that a
few TACE inhibitors have advanced to human clinic trial stages,
one of which was 3 (DPC-333, Figure 1),11 so far, no TACE
inhibitor drug has reached the market. This has been parti-
ally attributed to the general lack of selectivity for these early
hydroxamate TACE inhibitors. Thus, identification of a highly
selective TACE inhibitor has been a major goal for many
academic and pharmaceutical industry research laboratories.
Overexpression of tumor necrosis factor R (TNF-R), a 17
kDa pro-inflammatory cytokine, has been directly associated
with inflammation diseases such as rheumatoid arthritis and
Crohn’s disease.1 Reduction of TNF-R levels through a scav-
enging mechanism using either antibodies or TNF-R receptor
fusion proteins has been highly effective in controlling the
symptoms of these diseases in various clinical trials.2 TNF-R
is released from the membrane-bound pro-TNF-Rsa 26 kDa
precursor protein3sthrough a proteolytic cleavage process
mainly involving the TNF-R converting enzyme or TACE
(ADAM17), which is a member of the zinc-metalloprotease
family known as ADAM (a disintegrin and a metalloprotease).4
Inhibition of TACE activity to control the level of TNF-R release
has long been viewed as a promising way of treating related
inflammatory diseases.5
Many known small molecule TACE inhibitors are related to
the first generation succinate-based hydroxamate peptidic matrix
metalloprotease (MMP) inhibitors.6 Most of these contain three
important structural motifs: a zinc binding elementscommonly
a hydroxamate; a hydrogen bond acceptorsusually an amide
or sulfonamidesplaced 3-4 bonds away; and a hydrophobic
group attached on the backbone (2, Figure 1). The backbones
of these inhibitors are usually rigidified through cyclization,
which might enhance the binding affinity and selectivity and
improve their pharmacokinetic profiles. The selectivity of TACE
inhibitors against other MMPs and ADAMs has been considered
to be important for drug development but difficult to achieve.
For example, ADAM10 has the highest aminoacid sequence
homology as compared with TACE among all known MMPs.
ADAM10 also cleaves the membrane-bound TNF-R precursor
Docking studies of 3 with the published TACE X-ray crystal
structure12 suggested that the hydroxamate coordinates to the
zinc atom in a bidentate fashion with the lactam carbonyl group
interacting with the hydrogen bond donors of Leu-348 and Gly-
349 in TACE.13 The hydrophobic group extends into the S1′–S3′
binding pockets. In our effort to search for novel TACE
inhibitors, we examined the possibility of arranging the three
binding elements in a general structure 1 where a hydroxamate
derived from the trans-cyclopropyl dicarboxylate would offer
a good opportunity to interact with the zinc metal and the second
carbonyl group would be in a position to pickup the H-bonding
interactions with Leu-348 and Gly-349. A hydrophobic group
could be attached to the cyclopropyl ring to reach into the prime
site binding pockets even though the available protein crystal
structure, which was obtained with a peptidic hydroxamate
inhibitor without a S3′ group, did not seem to tolerate the
hydrophobic group in compound 3 without significant protein
rearrangement. Realizing that the only TACE protein crystal
structure available to us at the time would not help us rank order
different designs of the hydrophobic group, we decided to
prepare all four scaffolds (4-7) in Figure 2 to examine their
potential TACE inhibitory activities. These four scaffolds differ
by the spacers through which the benzyl group is attached to
the cyclopropyl ring as well as the substitution pattern on the
†
Dedicated to Professor Howard E. Zimmerman on his 82nd birthday.
* To whom correspondence should be addressed. Tel: 908-740-5651.
E-mail: Zhaoning.Zhu@spcorp.com.
‡
Department of Medicinal Chemistry.
Department of Inflammation and Infectious Diseases.
§
10.1021/jm070376o CCC: $40.75
2008 American Chemical Society
Published on Web 02/05/2008

726 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
Zhu et al.
Figure 1. TACE inhibitor pharmacophore model.
Figure 2. Scaffold design.
Scheme 1. Synthetic Scheme for Cyclopropyl Hydroxamatesa
a Reagents: (a) (1) n-BuLi, -78 °C, 1 h; (2) CuCN, 0 °C, 30 min; (3) methyl 2-(bromomethyl)acrylate, THF, -78 °C, 30 min; (4) -78 to -10 °C. (b)
DBU, t-butyloxymethyl-tetrahydro-thiophenium bromide, CH3CN. (c) 30% TFA, CH2Cl2. (d) O-Tritylhydroxyl amine, HOBt, EDCI, NMM, CH2Cl2. (e)
(CH3CH2)3SiH, 30% TFA, CH2Cl2.
phenyl or benzyl group, which would offer slightly different
geometries to orient the hydrophobic group for optimal binding
in S3′.
and 1-tert-butoxycarbonylmethyl-tetrahydrothiophenium bro-
mide (Scheme 1). Cuprates generated from meta/para-benzyl-
oxyphenyl bromides 8/9 reacted with methyl R-bromomethyl-
acrylate to give the meta/para-benzyloxybenzylacrylates 10/11,
respectively.14 Acrylates 18 and 19 were obtained using a
literature procedure.15 All four acrylates (10, 11, 18, and 19)
readily gave the desired trans-dicarboxylate cyclopropanation
Chemistry
These four cyclopropyl scaffolds were obtained via a cyclo-
propanation reaction involving appropriate precursor acrylates

Hydroxamates as TNF-R ConVerting Enzyme Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 727
Scheme 2. Chiral Resolution of Intermediate and Solution Phase Chemistry for SAR Developmenta
a Reagents: (a) (1) H2, 10% Pd/C, CH3OH; (2) OD Chiral Column, 5% IPA/hexane. (b) 4-Chloromethyl-2-methylquinoline, K2CO3, TBAB, CH2Cl2. (c)
30% TFA, CH2Cl2. (d) O-Tritylhydroxyl amine, HOAt, EDCI, DIPEA, CH2Cl2. (e) LiOH, 1:1 THF:H2O. (f) NH4Cl, HOBt, EDCI, DIPEA, DMF. (g)
(CH3CH2)3SiH, 30% TFA, CH2Cl2.
products (12, 13, 20, and 21) with exclusive trans-selectivity.16
After removal of the tert-butyl esters, the resulting acids were
coupled with O-trityl hydroxylamine using a standard peptide
coupling condition followed by deprotection of the trityl group
to give 6, 7, 4, and 5, which were found to have TACE Ki values
of 4, 0.8, 0.5, and 20 µM as racemic mixtures, respectively.
We concentrated our effort on two scaffolds, 4 and 7, for
subsequent follow-up, primarily focused on two sites for SAR
developments: (i) replacement of the benzyl group and (ii) ester
modifications. At first, the benzyl groups in core intermediates
13 and 20 were removed through hydrogenation and the
corresponding racemic phenols were resolved using an OD
chiral column eluting with 5% isopropanol in hexane (Scheme
2). It was subsequently found that the TACE binding affinity
was exclusively associated with the enantiomer that eluted
second for the phenylcyclopropyl series (22) and the enantiomer
that eluted first for the benzylcyclopropyl series (30). Alkylation
of the appropriate enantiomer of 22 led to compound 23
followed by deprotection of the tert-butyl ester to give an acid.
The resulting acid was coupled with O-trityl hydroxylamine to
give protected hydroxamate 25. Subsequently, compound 25
was converted to 26 through deprotection of the trityl group.
Additionally, hydrolysis of 25 followed by amide formation and
hydroxamate deprotection led to 29. Similarly, compound 20
was converted to 34 using the same procedure.
To rapidly explore phenol SAR, a solid-phase synthesis for
ethers from a resin-bound phenol was developed.17 This required
a resin linker for the hydroxamate that would not leave an
alkylatable site, such as -OH or -NH-, on the linked
hydroxamate group to carry out the subsequent solid-phase
O-alkylation or O-arylation on the phenol.18 It seemed that an
N-linked O-trimethylsilylethylhydroxamine (37) would offer a
good opportunity to achieve this. Thus, O-trimethylsilylethyl-
hydroxamine was loaded onto aldehyde resin 35 through oxime
reduction using either borane-pyridine or cyanoborohydride
under acidic conditions. The resin-bound hydroxylamine was
acylated with acid chloride 40, generated from chiral intermedi-
ate 22, and the loading level was found to be 0.40 mmol/g,
determined after cleavage to afford compound 42.
After removal of the acetyl group from the resin bound 41
with 20% piperidine in DMF, the resulting resin-bound phenol
43 was alkylated or arylated19 using either alcohols through
Mitsunobu condition or aryl boronic acid through copper acetate-
mediated O-arylation chemistry to give 44 or 45, respectively,

728 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
Zhu et al.
Scheme 3. Solid Phase Synthesis for SAR Development on Phenolsa
a Reagents: (a) (1) 10:20:70 CH3CO2H:CH3OH:THF; (2) BH3-pyridine, CH2Cl2. (b) Acetic anhydride, DMAP, CH2Cl2. (c) 30% TFA in DCM. (d) Oxalyl
chloride, CH2Cl2 with 1 drop of DMF. (e) DIPEA, DCM. (f) 50% TFA, CH2Cl2. (g) 20% piperidine in DMF. (h) ADDP, 3,4-dichlorobenzyl alcohol, PBu3,
THF. (i) Cu(OAc)2, DIPEA, phenyl boronic acid, DCM. (j) EDCI, NMM, HOAt, CH2Cl2. (k) 1 M Bu4NHOH, THF, 60 °C. (l) 2-Propanol, EDCI, DMAP,
CH2Cl2. (m) Benzyl amine, EDCI, HOBt, NMM, NMP. (a) Superimposition of X-ray crystal structures of compound 34 and 26 in the TACE enzyme
(V353G). (b) Key interaction of 34 (green) with TACE enzyme (V353G). (c) Key interactions of 26 (yellow) with TACE enzyme (V353G).
after TFA cleavage/deprotection followed by HPLC purification.
Similarly, compound 24 was loaded on to commercially
available hydroxylamine resin. The ester group was hydrolyzed,
and the resulting acid was converted to ester or amide through
a routine solid-phase transformation to give 48 and 49 after TFA
cleavage and purification.
individual enantiomers of 22 and 30, respectively (entry 1 in
Table 1 and entry 2 in Table 2).20
Results and Discussion
The fact that two out of the first four prototype racemic
hydroxamates (Figure 2) had submicromolar TACE Ki values
offered much needed proof of principle for these novel hydrox-
amates as potential TACE inhibitors. Because of the close
structural similarities between 4 and 7, it was speculated that
the geometric requirement of the TACE S1′/S3′ binding pocket
can be met with either the meta-substituted phenyl or a para-
substituted benzyl group directly attached to the cyclopropyl
ring. However, some SAR trends differentiated these two series
from the very beginning.
The TACE activities were measured by an internally quenched
peptide substrate with its sequence derived from the pro-TNFR
cleavage site using a catalytic domain of recombinant human
TACE enzyme. The Ki values are listed in Table 1 for
hydroxamates derived from 22. The eudismic ratios for inhibitors
derived from 22 and 30 are high with over 300-fold for
compound 4 and 39-fold for compound 77, prepared from the

Hydroxamates as TNF-R ConVerting Enzyme Inhibitors
Table 1. SAR for Phenylcyclopropyl Scaffold
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 729
a Racemates. b Each Ki value is an average of three determinations, and the standard errors for all Ki determinations are less than 10%.

730 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
Table 2. SAR for Benzylcyclopropyl Scaffold
Zhu et al.
a Racemates. b Each Ki value is an average of three determinations, and the standard errors for all Ki determinations are less than 10%.
Table 3. Selectivity of TACE Inhibitorsa
Because the carboxyl group was designed to interact with
compd MMP1 MMP2 MMP3 MMP7 MMP14 BMP1 ADAM10
Leu-348 and Gly-349 through hydrogen-bonding interactions,
an amide group should give a higher binding affinity than the
corresponding ester. While the amides had comparable or lower
Ki values as compared to corresponding esters for series 51
(Table 2, 76 vs 77, 34 vs 78, and 87 vs 88), it was interesting
to find that the carbomethoxyl group had a lower TACE Ki than
the primary amide in series 50 (Table 1, 4 vs 52 and 26 vs 29).
Additionally, tertiary-amides for this scaffold (50) were not
tolerated (Table 1, 71, 73 and 75), while the tertiary-amide was
among the inhibitors with the highest binding affinities for series
51 (Table 2, 80), suggesting that the carboxyl groups in scaffolds
50 and 51 did not have exactly the same interactions with the
TACE enzyme.
Furthermore, the phenol substitution in the series represented
by 50 tolerated a wide variety of groups such as substituted
benzyls (Table 1, 44, 52, and 54–58) with 3,4-dichlorobenzyl
as the most potent (∼11 nM) and (napththyl-2-yl)methyl as the
least active in this group, although still comparable with the
O-Me ether (∼3 µM; Table 1, 44, 53, and 59). However,
the naphthylethyl ethers recovered much binding affinity (Table
1, 61 and 62), and even the phenethyl group was comparable
with the benzyl group (Table 1, 60 and 4). Some O-aryl
substitutions were somewhat less potent (Table 1, 45 and
63–65), but the ꢀ-naphthyl group (63) retained good affinity
with a TACE Ki of 72 nM. Most aliphatic groups were less
tolerated (Table 1, 66 and 67). On the other hand, the SAR on
the phenol group for scaffold 51 was far more restrictive; all
substituted benzyl groups were 20- to over 100-fold less potent
than the (2-methylquinolin-4-yl)methyl group (Table 2, 81 and
82 vs 34). Ethylene-tethered naphthyl groups were far less active
in contrast with SAR with scaffold 50 and so were phenyl and
cyclohexylmethyl groups (Table 2, 83–86). While these diver-
gent SAR trends for the groups initially designed to reach the
26
34
44
48
56
78
>8300 3400 1300 >8300 7700
1400
2200
400
800
700
4
>7000
10
>12500 1100
100
600
200 1800
5400 2400
12400 3000
>600 4000
40 1000
6800 2800
7400 1400 1500
4
8
1800 >600
9000 300
200
20
9000
12000
a Selectivity is expressed as Ki/TACE Ki. Each Ki value is an average of
three determinations, and the standard errors for all Ki determinations are
less than 10%.
same S1′-S3′ binding pocket could have resulted from the
differences between the geometries of these two scaffolds, these
differences were big enough to raise the question as to whether
these classes of inhibitors did actually bind in the same pockets
of the TACE enzyme.
Third, the selectivity profiles for TACE inhibitors derived
from both scaffolds 50 and 51 were very different against other
metalloproteases (Table 3). Most notably, inhibitors from 50
had high selectivity against MMP1, -2, -3, and -14 and BMP1
but not ADAM10 (Table 3, 26, 44, 48, and 56). On the other
hand, the selectivity of inhibitors from 51 was universally high
against ADAM10 while the selectivity against MMP3 and -7
varied greatly depending on the phenol-O substitutions (34 and
78). These two series of MMP/ADAM inhibitors (50/51) seemed
to have very different presentations of pharmacophores to the
class of Zn-metalloproteases, in sharp contrast to the close
similarities between the chemical structures of the two scaffolds.
Lastly, under identical chiral resolution conditions, com-
pound 22 is the second enantiomer coming out of the OD
column while compound 30 is the first. Even though this
alone may be circumstantial, it does raise a possibility that
the two classes of TACE inhibitors are from different
enantiomer series. This was subsequently confirmed with X-ray

Hydroxamates as TNF-R ConVerting Enzyme Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 731
Figure 3. (a) Superimposition of X-ray crystal structures of compounds 34 and 26 in the TACE enzyme (V353G). (b) Key interaction of 34
(green) with TACE enzyme (V353G). (c) Key interactions of 26 (yellow) with TACE enzyme (V353G).
crystallography of 34 (Table 2) and 26 (Table 1) binding to a
TACE mutant enzyme (V353G)21 with the superimposed view
presented in Figure 3a. Both inhibitors used the hydroxamate
group to bind to the zinc metal and formed hydrogen-bonding
interactions with Glu406 with nearly identical geometry. Com-
pound 34 had a 1S,2S configuration, and the benzyl group
that attached to the cyclopropyl ring extended into the S1′
binding pocket. The phenyl group formed a ring-stacking
interaction with the imidazole group of the side chain of
His405, and it was also packed against the hydrophobic side
chain of Ala439 and Leu348 (Figure 3b). The quinoline group
occupied the extended S3′ binding pocket defined by Glu398,
Leu401, Ala439, Ser441, and Asn447, just as the original
design. We termed this binding mode A. The mode A
inhibitors bind to the enzyme, in a fashion similar to cases
reported by Letavic and co-workers,22 with a conformational
change of the enzyme loop region (Ala-439 to Gly-442)
caused by accommodation of the quinolinyl group; for
example, Val-440 shifts up 5.5 Å in comparison with the
enzyme with the peptidic inhibitor. This flexible loop was

732 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
Zhu et al.
less ordered in comparison to the crystal structure of TACE
with a peptidic inhibitor as reported by Maskos and co-
workers7 due to lack of the hydrogen-bonding interactions
of the peptidic inhibitor to both sides of the binding groove.
The carbomethoxyl group indeed interacted with Leu-348 and
Gly-349 through H-bonding interactions, which supported
the SAR trend that some of the amides were among the most
potent TACE inhibitors in this series. It was obvious that
the S3′ binding pocket was deeply buried, and it correlated
well with the finding that the structural tolerance for the
phenol derivatives was low. The quinoline nitrogen was
within hydrogen-bonding distance with the backbone N-H
of Ser441 (3.3 Å). Also, the trajectory from the S1′ pocket
extending to S3′ fit well with a -OCH2- linkage and that
the biaryl ether or arylethyl ethers would not appropriately
place the hydrophobic terminal aryl group into the S3′ binding
pocket.
inhibitors bind to the TACE enzyme in two different modes as
established by X-ray crystallography and exhibited different
selectivity profile against other metalloproteases. Having both
ADAM10 selective (mode A) and nonselective (mode B) series
provided an opportunity to further evaluate the biochemical
pathways and mechanism of action of these TNF-R inhibitors.
Experimental Section
Molecular Modeling. The X-ray structure of a TACE-inhibitor
complex (pdb code 1BKC)7 was adapted for modeling before in-
house crystal structures were available. One of the four molecules
in the asymmetric unit was used (chain A) with a peptide inhibitor
(1BKC), which showed three key features of binding: (i) zinc ligand,
(ii) hydrogen bonding to the backbone NHs of Leu-348 and Gly-
349, and (iii) hydrophobic interactions with residues in the S1′
pocket. While the coordinates of the hydroxamate moiety were
obtained from this X-ray structure as well as the X-ray structures
of other nonpeptidic MMP and TACE inhibitors in various binding
modes, several possible scaffolds were evaluated using docking
softwares such as GLIDE (Schrodinger) and GOLD (Cambridge
Crystallographic Data Center). To reduce the size of the system
and therefore speed up the computation, residues that were more
than 12 Å away from the original bound inhibitor were deleted.
The Sybyl (Tripos) molecular modeling software package was used
for molecular manipulations such as adding hydrogen atoms and
assigning atomic partial charges, as well as to perform dynamics
simulations and minimizations. A distance constraint of 2.0–2.2 Å
from the zinc was imposed on the two hydroxamate oxygens to
maintain the distance close to that observed in the crystal structures.
The hydroxamate was deprotonated, and one of the carboxylate
oxygens (OE1) was protonated following the finding from an early
publication.24
X-ray Crystallography. Prior to crystallization, TACE (V353G)
was incubated with a 2-fold excess of TAPI-1 (see ref 15) and
transferred into 25 mM TrisHCl at pH 8. Crystals were obtained
using the hanging drop method and PEG 6000, 10% 2-propanol,
and 100 mM sodium citrate, pH 5.6, as the precipitant. Crystals
were soaked in the precipitant solution with the respective inhibitors
for 3 days and frozen in liquid nitrogen with 15% glycerol as a
cryoprotectant. Diffraction data were collected at 100 K using the
IMCA-CAT beam line BM17. Data processing was performed using
the HKL2000 and CCP4 packages. The crystals were orthorhombic
with space group P212121 and unit cell dimensions of a ) 73, b )
75, and c ) 103 Å with a resolution of 1.7 Å. There were two
molecules in the asymmetric unit, and the crystal structure was
determined by molecular replacement using the program AMORE25
and the protein data bank entry 1BKC as a model. The structure
refinement was carried out using the CNX package.26,27
TACE and Metalloprotease Assays. The catalytic domain of
recombinant human TNF-converting enzyme [TACEcat, residues
215-477 with two mutations (S266A and N452Q) and a 6xHis
tail] was purified from the baculovirus/Hi5 cells expression system
using affinity chromatography. MMPs (MMP1, -2, -3, and -7) and
ecto domain of ADAM10 (a disintegrin containing metalloprotease)
as well as MMP substrates 1 and 2 were purchased from R&D
Systems (Minneapolis, MN). TACE, ADAM10, and MMP activities
were determined by a kinetic assay measuring the rate of increase
in fluorescent intensity generated by the cleavage of an internally
quenched peptide substrate. The substrate for TACE was an
internally quenched peptide (MCA-Pro-Leu-Ala-Gln-Ala-Val-Arg-
Ser-Ser-Ser-Dpa-Arg-NH2) [MCA, (7-methoxycoumarin-4-yl)acetyl;
Dpa, N-3-(2,4-dinitrophenyl)-L-2,3-diaminopropionyl], with its
sequence derived from the pro-TNFR cleavage site. Fifty microliters
of assay mixture contained 20 mM HEPES, pH 7.3, 5 mM CaCl2,
100 µM ZnCl2, 2% DMSO, 0.04% methylcellulose, 30 µM peptide
substrate, 0.1 nM TACEcat, and varying concentrations of a
compound. The reaction was started by the addition of the substrate.
The fluorescent intensity (excitation at 320 nm, emission at 405
nm) was measured every 45 s for 30 min using a fluorospectrometer
(GEMINI XS, Molecular Devices). ADAM10 was assayed in the
Compound 26, on the other hand, extended in the opposite
direction using the cyclopropyl scaffold with 1R,2R configu-
ration, which we termed mode B. In this binding mode, the
protein did not undergo significant conformational change at
the S3′ site. The carbomethoxyl group resided in the S1 binding
pocket that was defined by side chains of Val314, Lys315, and
Leu350. The pocket was half solvent-exposed and was hydro-
phobic in nature. As the result, hydrophobic groups such as
esters were preferred at this site and over primary amides.
Tertiary amides were not tolerated in this series due to the
intramolecular conformational change of the phenyl induced by
the tertiary-amide nitrogen. The m-phenyl group that attached
to the cyclopropyl ring made an edge to face interaction with
the imidazole group of His415 and was surrounded by the side
chain of Ser355. However, because the backbone conformation
of Ser355 and Pro356 in the mutant TACE enzyme was quite
different in comparison with the WT due to the single residue
mutation (V353G), it was not clear how much these changes
had affected the binding affinity with the mutant TACE enzyme.
The terminal quinolinyl group extended into the S3 binding
pocket and made van der Waals contact with the hydrophobic
part of the side chains of Lys315 and Tyr352. This binding
pocket was also half-exposed to solvent. This explained why
this site tolerated a variety of groups on the phenol oxygen,
including substituted benzyl, aryl, and arylethyl groups (Figure
3c).
The discovery of two distinctive binding modes warranted
major revisions of the pharmacophore models for the design of
inhibitors for TACE and other metalloproteases. In addition to
the aforementioned pharmacophore model that accounted for
the large majority of TACE inhibitors, which utilized the prime
side binding pocket, this work further demonstrated that an
alternative binding mode that placed the inhibitor into the
nonprime site to obviate a hydrogen-bonding interaction with
the backbone could provide potent TACE inhibitors. A search
in the patent literature revealed additional examples of this type
of TACE/MMP3 inhibitor.23 It was intriguing to explore the
combination of both types of binding modes for the benefit of
binding affinity as well as selectivity profile against other
metalloproteases. Given the structural diversity of known TACE
inhibitors in the literature, it was highly possible that some of
these known TACE inhibitors had already taken advantage of
both prime and nonprime site binding pockets and might possess
different selectivity profiles as a result.
In summary, the development of SAR for a series of de novo-
designed cyclopropyl dicarboxylate-related hydroxamates led
to the discovery of two series of potent TACE inhibitors. The

Hydroxamates as TNF-R ConVerting Enzyme Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 733
same way as TACE. Substrates 1 and 2 were used in assays for
MMP-1, -2, -3, and -7 according to the manufacturer’s protocol.
Values of Ki were calculated by the PRISM program based on one-
site competitive inhibition mode. Each Ki value was an average of
three determinations, and the standard errors for all Ki determina-
tions were less than 10%.
Synthesis. All reagents were obtained from commercial sources
unless noted otherwise. All NMR data were collected on 400 MHz
NMR spectrometers unless otherwise indicated. The chemical shifts
were expressed as ppm downfield from tetramethylsilane or as
relative ppm from designated reference peaks. Preparative reverse-
phase HPLC with a C-18 column using a gradient of 5–95% MeCN/
water with 0.1% TFA and a mass-triggered fraction collection
system was used to purify the hydroxamates. The purity of final
compounds was analyzed on an ES-LCMS analytical reverse-phase
HPLC system equipped with a 4.5 mm × 100 mm, 5 µm, C18
column and a mass detector and a UV detector monitored at both
219 and 254 nM. All runs were carried out at room temperature
with a 10 min gradient of 5–95% MeCN/water with 0.1% TFA at
a flow rate of 1 mL/min. See the Supporting Information for
compound purity analysis data for selected compounds.
(tert)-Butyl 2-(4-Benzyloxyphenyl)-2-trans-carboethoxylcyclo-
propane Carboxylate (21). A mixture of 19 (750 mg, 2.70 mmol),
tert-butyloxycarbonylmethyl-tetrahydrothiophenium bromide (822
mg, 2.9 mmol), and DBU (0.61 mL, 608 mg, 4.5 mmol) in
acetonitrile was stirred at room temperature for 2 days. The reaction
was poured into a saturated aqueous NH4Cl and CH2Cl2 mixture.
The aqueous layer was extracted with CH2Cl2 (3×). The organic
extracts were combined, dried over sodium sulfate, filtered, and
concentrated. The crude material was chromatographed with
10–20% EtOAc/hexane to give 950 mg of 21 (68%). 1H NMR (400
MHz, CDCl3): δ 7.44–7.30 (m, 5H, ArH), 7.23–7.19 (m, 2H, ArH),
6.93–6.89 (m, 2H, ArH), 5.05 (s, 2H, OCH2Ar), 4.19–4.04 (m, 2H,
CO2CH2CH3), 2.62 (dd, 1H, J ) 6.6 Hz, 8.1 Hz, cyclopropyl-CH),
1.93 (dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.80 (dd, 1H,
J ) 4.4 Hz, 8.1 Hz, cyclopropyl-CH), 1.20–1.17 (m, 12 H,
CO2C(CH3)3, CO2CH2CH3).
2-(4-Benzyloxyphenyl)-2-trans-carboethoxylcyclopropane Hy-
droxyamic Acid (5). A solution of 21 in 30% trifluoroacetic acid
in CH2Cl2 was stirred at room temperature. After 4 h, the reaction
mixture was concentrated. The residue was adjusted to pH ∼ 9.5
with a 1:1 ratio of a saturated NaHCO3:Na2CO3 solution. The
aqueous solution was washed with ether (3×). After acidification
to pH ∼ 2, the aqueous layer was extracted with ethyl acetate. The
combined organic layers were dried over sodium sulfate, filtered,
and concentrated to afford the corresponding acid, which was used
without purification in the next step.
A mixture of the above acid (410 mg, 1.16 mmol), O-
tritylhydroxylamine (430 mg, 1.56 mmol), HOBt (237 mg, 1.55
mmol), and EDCI (280 mg, 1.47 mmol) in CH2Cl2 was stirred
overnight. The reaction mixture was diluted with CH2Cl2 and
washed with a saturated aqueous solution of NaHCO3 (3×) and
water (3×) and then dried over sodium sulfate. The solvent was
removed, and the crude material was purified by flash chromatog-
raphy eluting with 0-100% ethyl acetate in hexane to yield the
O-trityl hydroxamate.
To a solution of the above hydroxamate (50 mg, 0.08 mmol) in
CH2Cl2 (2 mL) were added triethylsilane (20 mg, 0.17 mmol) and
trifluoroacetic acid (88 mg, 0.77 mmol). The reaction mixture was
concentrated. The crude material was purified by reverse-phase
HPLC (C-18 column) eluting with 5-95% acetonitrile in water to
afford 5. 1H NMR (400 MHz, CD3OD): δ 7.43–7.28 (m, 5H, ArH),
7.20–7.16 (m, J ) 8.8 Hz, 2H, ArH), 6.92–6.88 (m, 2H, ArH),
5.04 (s, 2H, ArCH2O), 4.12–4.03 (m, 2H, CO2CH2CH3), 2.54 (m,
1H, cyclopropyl-CH), 1.98 (dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclo-
propyl-CH), 1.77 (dd, 1H, J ) 4.4 Hz, 8.8 Hz, cyclopropyl-CH),
1.14 (m, 3 H, CO2CH2CH3). LC-MS (ESI) m/z: 356.1 (M + 1)+.
2-(3-Benzyloxyphenyl)-2-trans-carbomethoxylcyclopropane Hy-
droxyamic Acid (4). Compound 4 was prepared following a
procedure similar to the synthesis of 5. 1H NMR (400 MHz,
CD3CN): δ 7.47–7.33 (m, 5H, ArH), 7.19 (dd, 1H, J ) 7.3 Hz, 8.8
Hz, ArH), 6.90–6.84 (m, 3H, ArH), 5.08–5.02 (m, 2H, ArCH2O),
3.56 (s, 3H, CO2CH3), 2.49 (dd, 1H, J ) 6.6 Hz, 8.1 Hz,
cyclopropyl-CH), 1.95–1.92 (m, cyclopropyl-CH), 1.72 (dd, 1H, J
) 4.4 Hz, 8.8 Hz, cyclopropyl-CH). LC-MS (ESI) m/z: 342.1 (M
+ 1)+.
2-(3-Benzyloxybenzyl)-2-trans-carbomethoxylcyclopropane Hy-
1
droxyamic Acid (6). H NMR (400 MHz, CD3CN): δ 7.46–7.31
(m, 5H, ArH), 7.19–7.5 (m, 1H, ArH), 6.87–6.83 (br. s, 1H, ArH),
6.82–6.79 (m, 2H, ArH), 5.07 (s, 2H, OCH2Ar), 3.56 (s, 3H,
CO2CH3), 3.19–3.00 (m, 2H, ArCH2), cyclopropyl-CH), 1.52 (dd,
1H, J ) 3.6 Hz, 8.8 Hz, cyclopropyl-CH), 1.48–1.45 (m, 1H,
cyclopropyl-CH). LC-MS obs: 356.1 (M + 1).
2-(4-Benzyloxybenzyl)-2-trans-carbomethoxylcyclopropane Hy-
1
droxyamic Acid (7). H NMR (400 MHz, CD3OD): δ 7.43–7.28
(m, 5H, ArH), 7.15–7.13 (m, 2H, ArH), 6.88–6.86 (m, 2H, ArH),
5.03 (s, 2H, ArCH2O), 3.60 (s, 3H, CO2Me), 3.18 (d, J ) 14.6 Hz,
1H, ArCH2), 2.97 (d, J ) 14.6 Hz, 1H, ArCH2), 2.24 (br s, 1H,
cyclopropyl-CH), 1.53–1.51 (m, 2H, cyclopropyl-CH). LC-MS
(ESI) m/z: 356.1 (M + 1)+.
1
Methyl 2-(3-Benzyloxybenzyl) Acrylate (10). Yield, 38%. H
NMR (400 MHz, CDCl3): δ 7.46–7.43 (m, 2H, ArH), 7.41–7.37
(m, 2H, ArH), 7.35–7.31 (m, 1H, ArH), 7.25–7.21 (m, 1H, ArH),
6.87–6.81 (m, 3H, ArH), 6.25 (br s, 1H, vinyl-CH), 5.49–5.48 (m,
1H, vinyl-CH), 5.05 (s, 2H, OCH2Ar), 3.74 (s, 3H, CO2CH3), 3.62
(br s, 2H, CH2ArO).
Methyl 2-(4-Benzyloxybenzyl) Acrylate (11). A solution of 1.5
M tert-butyl lithium in hexane (53 mL, 80 mmol) was added to a
solution of 9 (10.5g, 37 mmol) in anhydrous THF (100 mL) at
-78 °C over 5 min. After the solution was stirred at -78 °C for
1 h, it was added into a mixture of CuCN (3.58 g, 40 mmol) in
THF (20 mL) at 0 °C. The solution was stirred for 30 min and
then cooled to -78 °C and added to a solution of methyl
2-(bromomethyl)acrylate (5.2 g, 29 mmol) in THF (20 mL) at -78
°C. The reaction was stirred for 30 min at -78 °C and then warmed
to -10 °C for 10 min before it was poured into a mixture of
saturated NH4Cl in ice. The mixture was extracted with CH2Cl2,
and the residue was chromatographed with 10% EtOAc/hexane to
1
give 6.0 g of the desired product 11 (53%). H NMR (400 MHz,
CDCl3): δ 7.45–7.43 (m, 2H, ArH), 7.41–7.37 (m, 2H, ArH),
7.35–7.31 (m, 1H, ArH), 7.14–7.11 (m, 2H, ArH), 6.93–6.91 (m,
2H, ArH), 6.22 (m, 1H, vinyl-CH) 5.46 (m, 1H, vinyl-CH), 5.05
(s, 2H, OCH2Ar), 3.74 (s, 3H, CO2CH3), 3.58 (s, 2H, CH2ArO).
tert-Butyl 2-(3-Benzyloxybenzyl)-2-trans-carbomethoxylcy-
1
clopropane Carboxylate (12). Yield, 270 mg;+ 21%. H NMR
(400 MHz, CDCl3): δ 7.45–7.30 (m, 5H, ArH), 7.19–7.15 (m, 1H,
ArH), 6.90–6.87 (m, 1H, ArH), 6.85–6.83 (m, 1H, ArH), 6.81–6.78
(m, 1H, ArH), 5.05 (s, 2H, OCH2Ar), 3.62 (s, 3H, CO2CH3),
3.24–3.15 (m, 2H, ArCH2), 2.46 (dd, 1H, J ) 6.6 Hz, 8.8 Hz,
cyclopropyl-CH), 1.59 (dd, 1H, J ) 4.4 Hz, 8.8 Hz, cyclopropyl-
CH), 1.46 (dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.37 [s,
9 H, CO2C(CH3)3].
tert-Butyl 2-(4-Benzyloxybenzyl)-2-trans-carbomethoxyl Cy-
1
clopropane Carboxylate (13). Yield, 39%. H NMR (400 MHz,
CDCl3): δ 7.44–7.41 (m, 2H, ArH), 7.39–7.35 (m, 2H, ArH),
7.33–7.29 (m, 1H, ArH), 7.17–7.13 (m, 2H, ArH), 6.89–6.85 (m,
2H, ArH), 5.03 (s, 2H, OCH2Ar), 3.63 (s, 3H, CO2CH3), 3.13 (s,
2H, ArCH2), 2.42 (dd, 1H, J ) 6.6 Hz, 8.8 Hz, cyclopropyl-CH),
1.57 (dd, 1H, J ) 4.4 Hz, 8.1 Hz, cyclopropyl-CH), 1.46 (dd, 1H,
J ) 4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.36 [s, 9 H, CO2C(CH3)3].
tert-Butyl 2-(3-Benzyloxyphenyl)-2-trans-carbomethoxylcy-
clopropane Carboxylate (20). Compound 20 was prepared by the
procedure described for the synthesis of 21 (73%). 1H NMR (400
MHz, CDCl3): δ 7.44–7.30 (m, 5H, ArH), 7.24–7.20 (m, 1H, ArH),
6.95–6.88 (m, 3H, ArH), 5.03 (s, 2H, OCH2Ar), 3.64 (s, 3H,
CO2CH3), 2.64 (dd, 1H, J ) 6.6 Hz, 8.8 Hz, cyclopropyl-CH), 1.94
(dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.80 (dd, 1H, J )
4.4 Hz, 8.1 Hz, cyclopropyl-CH), 1.20 [s, 9 H, CO2C(CH3)3].
tert-Butyl 2-(3-Hydroxyphenyl)-2-trans-carbomethoxylcyclo-
propane Carboxylate (22). To a solution of 13 (2.0 g, 7 mmol) in
100 mL of methanol was added 10% Pd/C (200 mg). The solution
was stirred under H2 atmosphere until the starting material

734 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
Zhu et al.
disappeared. The solution was filtered, and the solvent was
evaporated. The racemic mixture (1.0 g) was resolved with an OD
chiral column and eluted with 5% IPA/hexane (120 mL/min). The
second enantiomer that eluted at 28.17 min was determined to be
the active isomer. 1H NMR (400 MHz, CDCl3): δ 7.17–7.14 (m, 1
H, ArH), 6.86–6.84 (m, 1H, ArH), 6.76–6.72 (m, 2H, ArH), 3.65
(s, 3H, CO2CH3), 2.63 (dd, 1H, J ) 6.6 Hz, 8.1 Hz, cyclopropyl-
CH), 1.93 (dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.79
(dd, 1H, J ) 4.4 Hz, 8.1 Hz, cyclopropyl-CH), 1.20 [s, 9H,
quinolinyl-CH3), 2.56–2.53 (m, 1H, cyclopropyl-CH), 1.90–1.88
(m, 1H, cyclopropyl-CH), 1.72–1.69 (m, 1H, cyclopropyl-CH). LC-
MS (ESI) m/z: 392.1 (M + 1)+.
2-[3-(2-Methylquinolin-4-yl)methoxyphenyl]-2-trans-car-
bomethoxylcyclopropane Hydroxyamic Acid (34). 1H NMR (400
MHz, CDCl3): δ 8.43 (d, 1H, J ) 8.8 Hz, quinolinyl-CH), 8.17–8.11
(m, 3H, quinolinyl-CH), 7.97–7.93 (m, 1H, quinolinyl-CH), 7.25
(d, 2H, J ) 8.8 Hz, ArH), 7.06 (d, 2H, J ) 8.8 Hz, ArH), 5.83 (s,
2H, quinolinyl-CH2O), 3.62 (s, 3H, CO2CH3), 3.2–3.3 (m, 2H,
ArCH2), 2.99 (s, 3H, quinolinyl-CH3), 2.28–2.24 (m, 1H, cyclo-
propyl-CH), 1.57–1.54 (m, 1H, cyclopropyl-CH). LC-MS (ESI) m/z:
421.1 (M + 1)+.
(1R,2R)-tert-Butyl 2-Acetoxylphenyl-2-trans-carbomethoxyl-
cyclopropane Carboxylate (38). To a solution of 22 (516 mg, 1.8
mmol) in CH2Cl2 were added acetic anhydride (0.5 mL, 5.3 mmol)
and 4-dimethylaminopyridine (28 mg, 0.23 mmol). After 30 min,
the reaction was concentrated. The residue was dissolved in CH2Cl2
and washed with 0.1 N HCl (3×), a saturated solution of NaHCO3
(3×), and H2O (3×). The organic solution was dried over MgSO4,
filtered, and concentrated to afford 38. 1H NMR (400 MHz, CDCl3):
δ 7.32–7.28 (m, 1 H, ArH), 7.16–7.14 (m, 1H, ArH), 7.05–7.03
(m, 2H, ArH), 3.64 (s, 3H, CO2CH3), 2.63 (dd, 1H, J ) 6.6 Hz,
8.8 Hz, cyclopropyl-CH), 2.27 (s, 3H, COCH3), 1.94 (dd, 1H, J )
4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.81 (dd, 1H, J ) 4.4 Hz, 8.8
Hz, cyclopropyl-CH), 1.17 [s, 9H, CO2C(CH3)].
1
CO2C(CH3)]. H NMR (400 MHz, CDCl3): δ 7.18–7.14 (m, 1 H,
ArH), 6.87–6.85 (m, 1H, ArH), 6.76–6.72 (m, 2H, ArH), 3.65 (s,
3H, CO2CH3), 2.63 (dd, 1H, J ) 6.6 Hz, 8.8 Hz, cyclopropyl-
CH), 1.93 (dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclopropyl-CH), 1.79
(dd, 1H, J ) 4.4 Hz, 8.1 Hz, cyclopropyl-CH), 1.20 [s, 9H,
CO2C(CH3)]. The enantiomeric excess for both enantiomers was
determined using an analytical OD column to be >99%.
tert-Butyl 2-[3-(2-Methylquinolin-4-yl)methoxyphenyl]-2-
trans-carbomethoxylcyclopropane Carboxylate (23). 4-Chloro-
methy-2-methylquinoline (56 mg, 0.25 mmol), K2CO3 (100 mg,
0.72 mmol), tetrabutyl ammonium bromide (20 mg, 0.062 mmol),
and the second enantiomer of 22 (48 mg, 0.16 mmol) in CH2Cl2
were stirred overnight. The reaction solution was washed with water
(3×). The aqueous phase was extracted with CH2Cl2 (2×). The
organic extracts were combined, dried over sodium sulfate, filtered,
1
and concentrated to afford 23. H NMR (400 MHz, CD3OD): δ
8.02 (d, 1H, J ) 7.3 Hz, quinolinyl-CH), 7.97 (d, 1H, J ) 8.8 Hz,
quinolinyl-CH), 7.74–7.70 (m, 1H, quinolinyl-CH), 7.57–7.53 (m,
1H, quinolinyl-CH), 7.50 (s, 1H, quinolinyl-CH), 7.26–7.22 (m,
1H, ArH), 7.01–6.98 (m, 2H, ArH), 6.93–6.91 (m, 1H, ArH), 5.49
(s, 2H, quinolinyl-CH2O), 3.62 (s, 3H, CO2CH3) 2.68 (s, 3H,
quinolinyl-CH3), 2.60 (dd, 1H, J ) 6.6 Hz, 8.1 Hz, cyclopropyl-
CH), 1.94 (dd, 1H, J ) 5.1 Hz, 6.6 Hz, cyclopropyl-CH), 1.76
(dd, 1H, J ) 5.1 Hz, 8.8 Hz, cyclopropyl-CH), 1.14 [s, 9H,
COC(CH3)3].
2-[3-(2-Methylquinolin-4-yl)methoxylphenyl]-2-trans-car-
bomethoxylcyclopropane Hydroxyamic Acid (26). Compound 26
(30 mg, 0.046 mmol) was prepared from 23 using method similar
to compound 7. 1H NMR (400 MHz, CD3OD): δ 8.44 (d, 1H, J )
8.1 Hz, quinolinyl-CH), 8.20–8.12 (m, 2H, quinolinyl-CH), 8.07
(s, 1H, quinolinyl-CH), 7.98–7.94 (m, 1H, quinolinyl-CH), 7.30–7.25
(m, 1H, ArH), 7.07–7.04 (m, 2H, ArH), 6.98 (d, 1H, J ) 8.1 Hz,
ArH), 5.91–5.82 (m, 2H, quinolinyl-CH2O), 3.63 (s, 3H, CO2CH3),
3.00 (s, 3H, quinolinyl-CH3), 2.57 (dd, 1H, J ) 7.3 Hz, 8.8 Hz,
cyclopropyl-CH), 2.01 (dd, 1H, J ) 4.4 Hz, 6.6 Hz, cyclopropyl-
CH), 1.82 (dd, 1H, J ) 4.4 Hz, 8.8 Hz, cyclopropyl-CH). LC-MS
(ESI) m/z: 407.1 (M + 1)+.
(1R,2R)-2-Acetoxylphenyl-2-trans-carbomethoxylcyclopro-
pane Carboxylic Acid (39). A solution of 38 (1.69 g, 5 mmol) in
30% trifluoroacetic acid in CH2Cl2 (30 mL) was stirred at room
temperature. After 2 h, the reaction was concentrated to yield 1.40 g
of 39 (100%). 1H NMR (400 MHz, CDCl3): δ 7.34–7.30 (m, 1 H,
ArH), 7.16–7.13 (m, 1H, ArH), 7.04–7.01 (m, 2H, ArH), 3.66 (s,
3H, CO2CH3), 2.74 (dd, 1H, J ) 6.6 Hz, 8.8 Hz, cyclopropyl-
CH), 2.28 (s, 3H, COCH3), 1.97 (dd, 1H, J ) 4.4 Hz, 6.6 Hz,
cyclopropyl-CH), 1.94 (dd, 1H, J ) 4.4 Hz, 8.1 Hz, cyclopropyl-
CH).
(1R,2R)-2-Acetoxylphenyl-2-transcarbomethoxylcyclopro-
pane Carboxylic Acid Chloride (40). Oxalyl chloride (4.4 mL,
50 mmol) was added to 39 (1.40 g, 5 mmol) in 7 mL of CH2Cl2.
A drop of DMF was added, and the reaction was stirred at room
temperature. After 3 h, the reaction was concentrated to afford 40.
1H NMR (400 MHz, CDCl3): δ 7.377.33 (m, 1 H, ArH), 7.17–7.15
(m, 1H, ArH), 7.11–7.08 (m, 1H, ArH), 7.07–7.06 (m, 1H, ArH),
3.70 (s, 3H, CO2CH3), 3.29 (dd, 1H, J ) 6.6 Hz, 8.1 Hz,
cyclopropyl-CH), 2.29 (s, 3H, COCH3), 2.12 (dd, 1H, J ) 5.1 Hz,
6.6 Hz, cyclopropyl-CH), 2.04 (dd, 1H, J ) 5.1 Hz, 8.1 Hz,
cyclopropyl-CH).
O-Trityl 2-[3-(2-Methylquinolin-4-yl)methoxylphenyl]-2-trans-
carboxylcyclopropane Hydroxyamic Acid (27). A solution of 1
N LiOH monohydrate in 1:1 THF:H2O (2 mL, 2.0 mmol) was added
to 25 (30 mg, 0.046 mmol). The reaction was concentrated, and
the residue was partitioned between CH2Cl2 and a saturated aqueous
solution of NH4Cl. The organic layer was washed with a saturated
solution of NH4Cl (3×) and water (3×) and dried over sodium
sulfate. The solvent was removed to afford 25 mg of product (27)
(86%).
2-[3-(2-Methylquinolin-4-yl)methoxylphenyl]-2-trans-carbox-
amidocyclopropane Hydroxyamic Acid (29). To a solution of 27
(30 mg, 0.047 mmol) in DMF, ammonium chloride (60 mg, 1.11
mmol), HOBt (60 mg, 0.39 mmol), diisopropylethylamine (0.45
mL, 2.6 mmol), and EDCI (80 mg, 0.42 mmol) were added. The
reaction was stirred at room temperature overnight. The reaction
mixture was diluted with CH2Cl2 and washed with H2O (3×). The
organic layer was dried over sodium sulfate, filtered, and concen-
trated. The crude material was purified via flash chromatography
to afford 28.
Resin Loading of 2-(4-Formyl-3-methoxyphenoxy)ethyl Poly-
styrene Resin (37). O-(2-Trimethylsilylethyl)hydroxyl amine hy-
drochloride (1.66 g, 9.8 mmol) was added to 2-(4-formyl-3-
methyoxyphenoxy)ethyl polystyrene resin (1.00 mmol/g, 8.3 g, 8.3
mmol, Novabiochem) suspended in a 10:20:70 solvent mixture of
HOAc:CH3OH:THF. The reaction was agitated at room temperature
for 16 days. The resin was washed with DMF, THF, and CH2Cl2
sequentially (5×).
A solution of 8 M borane-pyridine complex (15 mL, 120 mmol)
was added to the resin (7.55 g) suspended in CH2Cl2. The mix-
ture was cooled to 0 °C, and dichloroacetic acid (14 mL, 169 mmol)
was added. The reaction was agitated at room temperature. After 7
days, the reaction was quenched with methanol. The resin was
washed with acetic acid for 30 min (2×) followed by 2 N NH3 in
MeOH (5×) and DMF (5×). The resin was then rinsed with MeOH,
THF, and CH2Cl2 sequentially (5×) and dried in vacuo to give 37.
(1R,2R)-O-Trimethylsilylethyl 2-(3-Acetoxylphenyl)-2-trans-
carbomethoxylcyclopropane Hydroxamic Acid on Resin (41).
Diisopropylethylamine (2.6 mL, 15 mmol) was added to resin 37
(9.77 g) suspended in 40 mL of CH2Cl2. Compound 40 (2.24 g,
7.6 mmol) was added dropwise to the suspension. The reaction
was agitated for 3 days. The resin was washed with MeOH, THF,
and CH2Cl2 sequentially (5×) and dried in vacuo to afford 10.82
g of 41. A portion of resin 41 (202 mg) was agitated with 50%
TFA in CH2Cl2 overnight. The filtrate was collected and concen-
Compound 29 was prepared from 28 following the procedure
1
described for the synthesis of 7. H NMR (400 MHz, CD3OD): δ
8.46 (d, 1H, J ) 8.8 Hz, quinolinyl-CH), 8.21–8.13 (m, 2H,
quinolinyl-CH), 8.10 (s, 1H, quinolinyl-CH), 8.00–7.96 (m, 1H,
quinolinyl-CH), 7.57 (s, 1H,), 7.38–7.34 (m, 1H, ArH), 7.15–7.05
(m, 3H, ArH), 5.95–5.86 (m, 2H, quinolinyl-CH2O), 3.02 (s, 3H,

Hydroxamates as TNF-R ConVerting Enzyme Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4 735
trated to afford 22 mg of 42 (51%). The loading level of the final
resin was determined to be 0.40 mmol/g. 1H NMR of 42 (400 MHz,
CDCl3): δ 7.31–7.26 (m, ArH), 7.10–7.09 (d, 1H, J ) 7.3 Hz, ArH),
6.99–6.98 (m, 2H, ArH), 3.65 (s, 3H, CO2CH3), 2.51–2.48 (m, 1H,
cyclopropyl-CH), 2.29 (s, 3H, COCH3), 2.05–2.00 (m, 1H, cyclo-
propyl-CH), 1.85–1.82 (m, 1H, cyclopropyl-CH). LC-MS (ESI) m/z:
294.1 (M + 1)+.
(1R,2R)-O-Trimethylsilylethyl 2-(3-Phenoxy)-2-trans-car-
bomethoxylcyclopropane Hydroxamic Acid on Resin (43). The
resin was agitated with 20% piperidine in DMF overnight. The resin
was washed with MeOH, THF, and CH2Cl2 sequentially (5×) to
give resin bound 43.
in CH2Cl2 (2 mL). The resin was agitated for 10 min before the
addition of 2-propanol (0.015 mL, 0.20 mmol). This mixture was
agitated for 16 h at room temperature. The liquid was drained, and
the resin was washed with an alternating cycle of CH2Cl2 (3×),
THF (3×), and MeOH (3×). The resin was treated with 50% TFA/
CH2Cl2 (2 mL) and agitated for 1.5 h. The liquid was drained, and
the resin was washed with CH2Cl2 (2×). The residue was purified
by reverse phase HPLC to provide 48 (0.002 g). LC-MS (ESI) m/z:
372.1 (M + 1)+.
(1R,2R)-2-trans-(N-Benzylaminocarbonyl-2-[3-(2-methylquin-
olin-4-yl)methoxylphenyl]cyclopropane Hydroxyamic Acid (49).
A mixture of the carboxylic acid resin 47 (0.040 g, 0.9 mmol/g),
EDCI (0.035 g, 0.18 mmol), HOBt (0.023 g, 0.15 mmol), and NMM
(0.020 mL, 0.18 mmol) in NMP (2 mL) was agitated for 20 min
before the addition of benzyl amine (0.020 mL, 0.18 mmol). This
mixture was agitated for 18 h at room temperature. The liquid was
drained, and the resin was washed with an alternating cycle of
CH2Cl2 (3×), THF (3×), and MeOH (3×). The resin was treated
with 50% TFA/CH2Cl2 (2 mL) and agitated for 1 h. The liquid
was drained, and the resin was washed with CH2Cl2 (2×).
Concentration of the liquid afforded compound 49 after purification
(21 mg). LC-MS (ESI) m/z: 482.3 (M + 1)+.
(1R,2R)-2-[3-(3,4-Dichlorobenzyloxy)phenyl]-2-trans-car-
bomethoxylcyclopropane Hydroxamic Acid (44). 1,1′-(Azodi-
carbonyl)dipiperidine (43 mg, 0.17 mmol,) and 3,4-dichlorobenzyl
alcohol (35 mg, 0.20 mmol) were added to S6.3 (100 mg). The
resin was suspended in 3 mL of anhydrous THF under nitrogen,
and tributylphosphine (0.06 mL, 0.24 mmol) was added. The final
reaction mixture was agitated and heated at 70 °C overnight. The
resin was rinsed with anhydrous THF, and the above procedure
was repeated. After 2 days, the resin was washed with MeOH, THF,
and CH2Cl2 sequentially (5×). The resin was cleaved with 75%
TFA in CH2Cl2 overnight. The filtrate was concentrated. The crude
material was purified via preparative HPLC eluting with 5–95%
Acknowledgment. We thank Emily Luk for LC-MS analysis
and Dr. T. M. Chan and Rebecca Osterman for help in NMR
structural determination.
1
of acetonitrile in water to give desired product 44. H NMR (400
MHz, CD3OD): δ 7.62 (d, 1H, J ) 2.2 Hz, ArH), 7.53 (d, 1H, J )
8.1 Hz, ArH), 7.38 (dd, 1H, J ) 1.5, 8.1 Hz ArH), 7.21–7.17 (m,
1H, ArH), 6.91–6.87 (m, 3H, ArH), 5.04 (s, 2H, ArCH2), 3.61 (s,
3H, CO2CH3), 2.56 (dd, 1H, J ) 6.6 Hz, 8.8 Hz, cyclopropyl-
CH), 2.00 (dd, 1H, J ) 4.4 Hz, 6.6 Hz cyclopropyl-CH), 1.78 (dd,
1H, J ) 4.4 Hz, 8.8 Hz cyclopropyl-CH). LC-MS (ESI) m/z: 410.1
(M + 1)+.
Supporting Information Available: Table of compounds and
LC-MS % purity. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1R,2R)-2-(3-Phenoxylphenyl)-2-trans-carbomethoxylcyclo-
propane Hydroxamic Acid (45). A mixture of resin 43, 5 µm, 4
Å molecular sieves (100 mg), copper acetate (10 mg, 0.05 mmol),
phenyl boronic acid (24 mg, 0.020 mmol), and diisopropylethyl-
amine (0.05 mL, 0.28 mmol) in 2 mL of CH2Cl2 was agitated at rt
overnight. The resin was washed with THF (3×) and CH2Cl2 (3×).
The above procedure was repeated. The resin was washed with
0.1 N EDTA in 2:1 DMF:H2O for 30 min (2×) followed by a 1:1
DMF:H2O solution with a final wash of MeOH, THF, and CH2Cl2
sequentially (5×). The resin was cleaved with 75% TFA in CH2Cl2
overnight. After removal of organic solvent, the residue was purified
with a C-18 reverse phase column eluting with 5–95% acetonitrile
in water to give desired product 45. 1H NMR (400 MHz, CD3OD):
δ 7.36–7.32 (m, 2H, ArH), 7.27–7.23 (m, 1H, ArH), 7.11–7.07
(m, 1H, ArH), 7.03–6.99 (m, 3H, ArH), 6.93–6.92 (m, 1H, ArH),
6.87–6.85 (m, 1H, ArH), 3.63 (s, 3H, CO2CH3), 2.55 (dd, 1H, J )
6.6 Hz, 8.8 Hz cyclopropyl-CH), 2.00 (dd, 1H, J ) 4.4 Hz, 6.6 Hz
cyclopropyl-CH), 1.77 (dd, 1H, J ) 4.4 Hz, 8.8 Hz cyclopropyl-
CH). LC-MS (ESI) m/z: 328.2 (M + 1)+.
Loading of Wang Hydroxylamine Resin (46). A solution of
acid 24 (0.536 mg, 1.06 mmol), Wang hydroxylamine resin (0.931
g, 1.04 mmol/g), EDCI (0.330 g, 1.72 mmol), NMM (1.10 mL,
10.00 mmol), and HOAt (0.150 g, 1.10 mmol) in CH2Cl2 (12 mL)
was agitated for 14 h at room temperature. The liquid was drained,
and the resin was washed with CH2Cl2 (3×), THF (3×), and MeOH
(3×) in an alternating sequence. The resin was dried under high
vacuum to yield 46 (1.32 g, 0.90 mmol/g).
(1R,2R)-2-trans-Carboxyl-2-[3-(2-methylquinolin-4-yl)methox-
ylphenyl]cyclopropane Hydroxyamic Acid (47) on Resin. A
mixture of resin 46 (1.3 g, 0.90 mmol/g) and 1 M Bu4NOH in
THF (15 mL) was agitated at 60 °C for 6 h. The liquid was drained,
and the resin was washed with 1% AcOH in DMF (2 × 30 min)
followed by an alternating cycle of MeOH (3×), THF (3×), and
CH2Cl2 (3×). The resulting resin was dried under high vacuum
for 4 h.
(1R,2R)-2-trans-Carboisopropoxyl-2-[3-(2-methylquinolin-4-
yl)methoxylphenyl]cyclopropane Hydroxyamic Acid (48). EDCI
(0.035 g, 0.18 mmol) and DMAP (0.015 g, 0.12 mmol) were added
to a mixture of the carboxylic acid resin 47 (0.040 g, 0.9 mmol/g)
(1) (a) Cerretti, D. P. Characterization of the tumour necrosis factor alpha
converting enzyme, TACE/ADAM17. Biochem. Soc. Trans. 1999, 27,
219–223. (b) Vassalli, P. The pathophysiology of tumor necrosis
factors. Annu. ReV. Immunol. 1992, 10, 411–452.
(2) (a) Van Assche, G.; Rutgeerts, P. Anti-TNF agents in Crohn’s disease.
Exp. Opin. InVest. Drugs 2000, 9, 103–111. (b) Nelson, F. C.; Zask,
A. The therapeutic potential of small molecule TACE inhibitors. Exp.
Opin. InVest. Drugs 1999, 8, 383–392. (c) Doggrell, S. A. TACE
inhibition a new approach to treating inflammation. Exp. Opin. InVest.
Drugs 2002, 11, 1003–1006.
(3) Kriegler, M.; Perez, C.; DeFay, K.; Albert, I.; Lu, S. D. A novel form
of TNF/cachectin is a cell surface cytotoxic transmembrane protein:
ramifications for the complex physiology of TNF. Cell 1988, 53, 45–
53.
(4) (a) Black, R. A.; Rauch, C. T.; Kozlosky, C. J.; Peschon, J. J.; Slack,
J. L.; Wolfson, M. F.; Castner, B. J.; Stocking, K. L.; Reddy, P.;
Srinivasan, S.; Nelson, N.; Boiani, N.; Schooley, K. A.; Gerhart, M.;
Davis, R.; Fitzner, J. N.; Johnson, R. S.; Paxton, R. J.; March, C. J.;
Cerretti, D. P. A metalloproteinase disintegrin that releases tumour-
necrosis factor-R from cells. Nature 1997, 385, 729–733. (b) Moss,
M. L.; Jin, S.-L.; Milla, M. E.; Bickett, D. M.; Burkhart, W.; Carter,
H. L.; Chen, W. J.; Clay, W. C.; Didsbury, J. R.; Hassler, D.; Hoffman,
C. R.; Kost, T. A.; Lambert, M. H.; Leesnitzer, M. A.; McCauley, P.;
McGeehan, G.; Mitchell, J.; Moyer, M.; Pahel, G.; Rocque, W.;
Overton, L. K.; Schoenen, F.; Seaton, T.; Su, J. L.; Becherer, J. D.
Cloning of a disintegrin metalloproteinase that processes precursor
tumour-necrosis factor-R. Nature 1997, 385, 733–736.
(5) (a) Newton, R. C.; Solomon, K. A.; Covington, M. B.; Decicco, C. P.;
Haley, P. J.; Friedman, S. M.; Vaddi, K. Biology of TACE inhibition.
Ann. Rheum. Dis. 2001, 60, 25–32. (b) Moss, M. L.; White, J. M.;
Lambert, M. H.; Andrews, R. C. TACE and other ADAM proteases
as targets for drug discovery. Drug DiscoVery Today 2001, 6, 417–
426.
(6) (a) Skotnicki, J. S.; Zask, A.; Nelson, F. C.; Albright, J. D.; Levin,
J. I. Design and synthetic considerations of matrix metalloproteinase
inhibitors. Ann. N. Y. Acad. Sci. 1999, 878, 61–72. (b) De, B.; Natchus,
M. G.; Cheng, M.; Pikul, S.; Almstead, N. G.; Taiwo, Y. O.; Snider,
C. E.; Chen, L.; Barnett, B.; Gu, F.; Dowty, M. The next generation
of MMP inhibitors: Design and synthesis. Ann. N. Y. Acad. Sci. 1999,
878, 40–60. (c) Nelson, F. C.; Zask, A. The therapeutic potential of
small molecule TACE inhibitors. Exp. Opin. InVest. Drugs 1999, 8,
383–392.
(7) Yan, Y.; Shirakabe, K.; Werb, Z. The metalloprotease Kuzbanian
(ADAM10) mediates the transactivation of EGF receptor by G protein-
coupled receptors. J. Cell Biol. 2002, 158, 221–226.

736 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 4
Zhu et al.
(8) Hattori, M.; Osterfield, M.; Flanagan, J. G. Regulated cleavage of a
contact-mediated axon repellent. Science 2000, 289, 1360–1365.
(9) Lammich, S.; Kojro, E.; Postina, R.; Gilbert, S.; Pfeiffer, R.;
Jasionowski, M.; Haass, C.; Fahrenholz, F. Constitutive and regulated
alpha-secretase cleavage of Alzheimer’s amyloid precursor protein by
a disintegrin metalloprotease. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
3922–3927.
(10) Hartmann, D.; De Strooper, B.; Serneels, L.; Craessaerts, K.; Herreman,
A.; Annaert, W.; Umans, L.; Lubke, T.; Lena, I. A.; Von Figura, K.;
Saftig, P. The disintegrin/metalloprotease ADAM 10 is essential for
Notch signaling but not for alpha-secretase activity in fibroblasts. Hum.
Mol. Genet. 2002, 11, 2615–2624.
(11) (a) Grootveld, M.; McDermott, M. F. BMS-561392 (Bristol-Myers
Squibb). Curr. Opin. InVest. Drugs 2003, 4, 598–602. (b) Liu, R.;
Magolda, R.; Newton, R.; Duan, J.; Vaddi, K.; Madskuie, T.; Qian,
M.; Collins, R.; Taylor, T.; Giannaris, J. Pharmacological profile of
DPC 333, a selective TACE inhibitor. Inflammation Res. 2002, 51,
A27. (c) Zhang, Y.; Xu, J.; Udata, C.; McDevitt, J.; Cummons, T.;
Sun, L.; Zhu, Y.; Li, G.; Rao, V.; Wang, Q; Gibbons, J. Indentification
and characterization of Aprarastat (TMI-005), a potent, dual TACE/
MMP inhibitor for the treatment of RA. Ann. Rheum. Dis. 2005, 64,
462–463. (d) Fleischmann, R.; Kivitz, A. J.; Franklin, C.; Pavlik, G. S.;
Sridharan, S.; Kirsch, T. M. A randomized, double blind, placebo
controlled, sequential dose study of the safety of Apratastat (TMI-
005), a novel dual inhibitor of TNF-alpha converting enzyme/
metalloproteinase, in patients with rheumatoid arthritis on a background
of methotrexate. Ann. Rheum. Dis. 2005, 64, 132.
(12) Maskos, K.; Fernandez-Catalan, C.; Humber, R.; Bourenkov, G. P.;
Bartunik, H.; Eliestad, G. A.; Reddy, P.; Wolfson, M. F.; Rauch, C. T.;
Castner, B. J.; Davis, R.; Clarke, H. R. G.; Petersen, M.; Fitzner, J. N.;
Cerretti, D. P.; March, C. J.; Paxton, R. J.; Black, R. A.; Bode, W.
Crystal structure of the catalytic domain of human tumor necrosis
factor-alpha-converting enzyme. Proc. Natl. Acad. Sci. U.S.A. 1998,
95, 3408–3412.
(13) A cyclopropyl dicarboxylate scaffold had been previously explored
as potential MMP inhibitors as part of an approach for rigidification
of peptide backbones. It was concluded that these scaffolds were “poor
mimics of the bound conformation” of the peptidic substrates. Reichelt,
A.; Gaul, C.; Frey, R. R.; Kennedy, A.; Martin, S. F. J. Org. Chem.
2002, 67 (12), 4062–4075.
(14) Majid, T. N.; Knochel, P. A new preparation of highly functionalized
aromatic and heteroaromatic zinc and copper organometallics. Tetra-
hedron Lett. 1990, 31, 4413–4416.
(15) Serelis, A. K.; Simpson, G. W. Stereoselectivity in the thermal
cycloaddition reactions of tetrafluoroethylene to derivatives of R-(4-
ethoxyphenyl)acrylic acid. Tetrahedron Lett. 1997, 38, 4277–4280.
(16) Von der Saal, W.; Reinhardt, R.; Seidenspinner, H.-M.; Satwitz, J.;
Quast, H. Cyclopropanediamines. 3. Pure diastereomers of 1,2-
cyclopropanedicarboxylic acids and derivatives. Liebigs Ann. Chem.
1989, 703–712.
(17) Zhu, Z.; Mazzola, R.; Guo, Z.; Lavey, B.; Sinning, L.; Kozlowski, J.;
McKittrick, B.; Shih, N.-Y. Compounds for the treatment of inflam-
matory disorders. U.S. Patent 6,838,466, 2005.
hydes and hydroxamates from a common backbone amide-linked
(BAL) intermediate. J. Pept. Res. 2005, 66 (6), 324–332. (d) Khan,
S. I.; Grinstaff, M. W. A facile and convenient solid-phase procedure
for synthesizing nucleoside hydroxamic acids. Tetrahedron Lett. 1998,
39 (44), 8031–8034.
(19) (a) Chan, D. M.; Monaco, K. L.; Wang, R.-P.; Winters, M. New N-
and O-arylation with phenylboronic acids and cupric acetate. Tetra-
hedron Lett. 1998, 39 (19), 2933–2936. (b) Evans, D. A.; Katz, J. L.;
West, T. R. Synthesis of diaryl ethers through the copper-promoted
arylation of phenols with arylboronic acids. An expedient synthesis
of thyroxine. Tetrahedron Lett. 1998, 39 (19), 2937–2940. (c) Combs,
A. P.; Saubern, S.; Rafalski, M.; Lam, P. Y. S. Solid supported aryl/
heteroaryl C-N cross-coupling reactions. Tetrahedron Lett. 1999, 40
(9), 1623–1626.
(20) Detailed SAR work for scaffold 30 will be the subject of another paper,
and only limited data are listed in Table 2 for discussion.
(21) (a) Ingram, R. N.; Orth, P.; Strickland, C. L.; Le, H. V.; Madison, V.;
Beyer, B. M. Stabilization of the autoproteolysis of TNF-alpha
converting enzyme (TACE) results in a novel crystal form suitable
for structure-based drug design studies. Protein Eng., Protein Des.
Sel. 2006, 19 (4), 155–161. (b) Beyer, B. M.; Ingram, R. N.; Orth, P.;
Strickland, C. Crystal structure of human TNF-R convertase mutant
complexed with inhibitor for use in structure-based rational drug
design. U.S. Patent US7138264, 2006.
(22) Letavic, M. A.; Axt, M. Z.; Barberia, J. T.; Carty, T. J.; Danley, D. E.;
Geoghegan, K. F.; Halim, N. S.; Hoth, L. R.; Kamath, A. V.; Laird,
E. R.; Lopresti-Morrow, L. L.; McClure, K. F.; Mitchell, P. G.;
Natarajan, V.; Noe, M. C.; Pandit, J.; Reeves, L.; Schulte, G. K.; Snow,
S. L.; Sweeney, F. J.; Tan, D. H.; Yu, C. H. Synthesis and biological
activity of selective pipecolic acid-based TNF-R converting enzyme
(TACE) inhibitors. Bioorg. Med. Chem. Lett. 2002, 12, 1387–1390.
(23) (a) Hirata, T.; Itoh, K.; Misumi, K.; Kuramoto, Y.; Inokuma, K.;
Amano, H.; Aoki, S.; Yoshimatsu, T. Discovery of potent, highly
selective, and orally active propenohydroxamate TNF-R converting
enzyme (TACE) inhibitors. Abstracts of Papers, 222nd ACS National
Meeting, Chicago, IL, Aug 26–30, 2001; American Chemical Society:
Washington, DC, 2001; MEDI-262. (b) Hirata, T.; Misumi, K.; Ito,
K.; Inokuma, K.; Katayama, K. Preparation of propenohydroxamic
acid derivatives as TACE inhibitors for treatment of sepsis, infectious
and autoimmune diseases, etc. International patent WO2002018326,
2002. (c) Hirata, A.; Nishimura, H.; Katayama, K.; Tamura, K.;
Amano, H.; Sugimoto, K. Preparation of 3-phenyl- and 3-pyridylpro-
penohydroxamic acid derivatives as new MMP (MMP-3) inhibitors.
Japan Patent JP2004277311, 2004.
(24) Cross, J. B.; Duca, J. S.; Kaminski, J. J.; Madison, V. S. The active
site of a zinc-dependent metalloproteinase influences the computed
pK(a) of ligands coordinated to the catalytic zinc ion. J. Am. Chem.
Soc. 2002, 124, 11004–11007.
(25) Navaza, J. AMoRe: An automated package for molecular replacement.
Acta Crystallogr., Sect. A: Found. Crystallogr. 1994, 50, 157–163.
(26) Bruenger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros,
P.; Grosse-Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges, M.;
Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Crystallography & NMR System: A new software suite for macro-
molecular structure determination. Acta. Crystallogr., Sect. D: Biol.
Crystallogr. 1998, 54, 905–921.
(18) For other related linkers, see (a) Stanger, K. J.; Krchnak, V. Polymer-
supported N-derivatized, O-linked hydroxylamine for concurrent solid-
phase synthesis of diverse N-alkyl and N-H hydroxamates. J. Comb.
Chem. 2006, 8 (3), 435–439. (b) Krchnak, V. Solid-phase synthesis
of biologically interesting compounds containing hydroxamic acid
moiety. Med. Chem 2006, 6 (1), 27–36. (c) Gazal, S.; Masterson, L. R.;
Barany, G. Facile solid-phase synthesis of C-terminal peptide alde-
(27) Bruenger, A. T. Free R value: A novel statistical quantity for assessing
the accuracy of crystal structures. Nature 1992, 355, 472–475.
JM070376O