Selective water-soluble gelatinase inhibitor prodrugs

Article abstract of DOI:10.1021/jm200566e

SB-3CT (1), a selective and potent thiirane-based gelatinase inhibitor, is effective in animal models of cancer metastasis and stroke; however, it is limited by poor aqueous solubility and extensive metabolism. We addressed these issues by blocking the pr

Full text of DOI:10.1021/jm200566e

ARTICLE
pubs.acs.org/jmc
Selective Water-Soluble Gelatinase Inhibitor Prodrugs
Major Gooyit,†,§ Mijoon Lee,†,§ Valerie A. Schroeder,‡ Masahiro Ikejiri,† Mark A. Suckow,‡
Shahriar Mobashery,† and Mayland Chang*,†
†Department of Chemistry and Biochemistry and ‡Freimann Life Sciences Center and Department of Biological Sciences, University of
Notre Dame, Notre Dame, Indiana 46556, United States
ABSTRACT:
SB-3CT (1), a selective and potent thiirane-based gelatinase inhibitor, is effective in animal models of cancer metastasis and stroke;
however, it is limited by poor aqueous solubility and extensive metabolism. We addressed these issues by blocking the primary site
of metabolism and capitalizing on a prodrug strategy to achieve >5000-fold increased solubility. The amide prodrugs were
quantitatively hydrolyzed in human blood to a potent gelatinase inhibitor, ND-322 (3). The arginyl amide prodrug (ND-478, 5d)
was metabolically stable in mouse, rat, and human liver microsomes. Both 5d and 3 were nonmutagenic in the Ames II mutagenicity
assay. The prodrug 5d showed moderate clearance of 0.0582 L/min/kg, remained mostly in the extracellular fluid compartment
(Vd = 0.0978 L/kg), and had a terminal half-life of >4 h. The prodrug 5d had superior pharmacokinetic properties than those of 3,
making the thiirane class of selective gelatinase inhibitors suitable for intravenous administration in the treatment of acute gelatinase-
dependent diseases.
’ INTRODUCTION
and neuronal death in a rat model of subarachnoid hemorrhage,17
a type of hemorrhagic stroke. Compound 1 significantly reduces
flow-induced vascular remodeling, an adaptive process that
allows blood vessels to normalize hemodynamic stress in response
to increased blood flow in atherosclerosis, aneurysms, and brain
arteriovenous maltransformations in mice.18 In a rat model of
spinal cord injury, treatment with 1 results in decreases in MMP-9
activity, extravasation, and apoptotic cell death.19 This novel
inhibitor has found use in many other biological systems, which
we will not outline here in the interest of brevity.
Matrix metalloproteinases 2 and 9 (MMP-2 and MMP-9) are
zinc-dependent endopeptidases that are also known as gelati-
nases. The unregulated activities of these enzymes have been
implicated in a host of human diseases, including tumor invasion
and metastasis, cardiovascular and neurological diseases, and
inflammation to name a few.1ꢀ4 Gelatinases were shown to be
essential for leukocyte penetration into the brain parenchyma in
a mouse model of experimental autoimmune encephalomyelitis.5
All of these diseases involve restructuring of the extracellular
matrix for manifestation of the disease;4,6ꢀ8 hence, selective
gelatinase inhibitors are highly sought.9,10
We have described the first highly selective mechanism-based
gelatinase inhibitor, (4-phenoxyphenylsulfonyl)methylthiirane
(SB-3CT, 1).11 This inhibitor undergoes a chemical transforma-
tion within the active site of gelatinases, a process that is likely at
the root of the selectivity that the compound exhibits in its potent
inhibition.12 Compound 1 is efficacious in animal models of
prostate cancer metastasis to the bone,13 breast cancer metastasis
to the lungs,14 and T-cell lymphoma metastasis to the liver.15
Compound 1 also blocks the MMP-9-dependent degradation of
the extracellular matrix protein laminin and thus rescues neurons
from apoptotic cell death in a mouse model of transient focal
cerebral ischemia.16 In addition, 1 prevents laminin degradation
Despite documented excellent biological activity, administra-
tion of 1 to animals is problematic due to its poor water solubility
of 2.3 μg/mL.20 Because of poor solubility, the compound is
typically dosed as a suspension intraperitoneally. Furthermore, it
Received: May 5, 2011
Published: August 25, 2011
r
2011 American Chemical Society
6676
dx.doi.org/10.1021/jm200566e J. Med. Chem. 2011, 54, 6676–6690
|

Journal of Medicinal Chemistry
ARTICLE
Table 1. Structures of Ester and Amide Prodrugs
is rapidly metabolized by hydroxylation at the para-position of
the terminal phenyl ring (2) to a more potent gelatinase inhibitor
than the parent 1 and by oxidation at the α-position to the
sulfonyl that leads to the formation of the inactive sulfinic acid.21
We addressed oxidation at the α-position to the sulfonyl by the
addition of a methyl substituent at that site; however, this
modification led to a 10-fold decrease in inhibitory potency.22
However, derivatization at the para-position of the terminal
phenyl ring with methanesulfonate did not only improve meta-
bolic stability but also enhanced potency.23
Scheme 1. Synthesis of 3a
Our interest in developing therapeutics for the treatment of
acute neurological conditions such as stroke, aneurysm, and
traumatic brain injury, for which intravenous infusion is the
preferred route of administration, led us to explore increasing the
water solubility of 1. For parenteral administration, a solubility of
10 mg/mL is usually needed to adequately formulate a drug.24
This presented a delivery challenge in the case of 1, which was
administered intraperitoneally as a suspension in studies of
animal models of disease. A prodrug strategy is useful to solve
formulation problems associated with poor aqueous solubility. A
prodrug (drug + promoiety) is a chemical entity with little or no
pharmacological activity, which undergoes transformation
(biological or otherwise) to a therapeutically active drug.25 The
majority of prodrugs for parenteral administration have mod-
ifications where a polar ionizable promoiety is introduced to
increase water solubility.25 Amino acid promoieties have been
employed to improve the aqueous solubility of drugs containing
an alcohol or amine26,27 and are rapidly and quantitatively
converted to the active drug by the action of esterases and/or
peptidases in blood. We surmise that functionalization at the
para-position of the terminal phenyl ring with water-solubiliz-
ing agents to create prodrugs might address the solubility
problem of 1, as well as block the primary site of metabolism.
We used amino acids (and a peptide) as promoieties to impart
water solubility to the gelatinase inhibitor. The p-hydroxy
derivative 2 is a good candidate for ester conjugation, as it
was previously shown to be a more potent gelatinase inhibitor
than 1.21 We also explored the use of the p-amino derivative 3,
which was found to have potent gelatinase inhibitory activity, to
synthesize aminoacidꢀamide conjugates. The resulting ester
and amide prodrug series (Table 1) are expected to undergo
hydrolysis in plasma and blood to release the active gelatinase
inhibitors 2 or 3. The syntheses and evaluation of the properties
of these prodrug series, compounds 4 and 5, are the subjects of
this article.
a Reagents and conditions: (a) (i) Cs2CO3, room temperature, 2 h; (ii)
Zn, AcOH, 0 °C to room temperature, 2 h, 79%. (b) Boc2O, Et3N,
MeOH, 60 °C, 2 h, 82%. (c) m-CPBA, CH2Cl2, room temperature, 72 h,
81%. (d) Thiourea, MeOH/CH2Cl2, room temperature, 18 h, 72%. (e)
4 N HCl in 1,4-dioxane, ethyl acetate/CH2Cl2, 0 °C to room tempera-
ture, 48 h, 82%.
’ RESULTS AND DISCUSSION
Syntheses of Prodrugs. The p-aminophenoxybenzene scaf-
fold in 3 was assembled by the methodology developed by Ikejiri
et al.,28 which is the coupling of 1-fluoro-4-nitrobenzene (6) and
4-(allylthio)phenol (7) under basic conditions, followed by
reduction of the nitro group to the amine over elemental zinc
in acetic acid (Scheme 1). The resulting amine 8 was treated with
di-tert-butyl dicarbonate to give the Boc-protected compound 9.
The Boc group was chosen as the amine-protecting group
because sulfonylmethylthiirane is relatively stable under the
acidic condition that is used for the removal of Boc at the end
of the synthesis (see transformation of 10 to 3). The rest of the
transformations, including the oxidation to sulfone, epoxide
formation using m-chloroperoxybenzoic acid (m-CPBA), and
thiirane conversion using thiourea, were performed by the
methodology developed by us earlier.29,30 Removal of the Boc
group in compound 10 was carried out in the presence of 4 N
HCl in 1,4-dioxane at room temperature for 48 h to yield 3 as the
HCl salt in 82% yield.
The syntheses of amino acid conjugates of phenol 2 and of
aniline 3 as prodrugs of gelatinase inhibitors are outlined in
Scheme 2. Compound 2, prepared according to the literature,23
was acylated with Boc-protected N-hydroxysuccinimide (NHS)
esters of amino acids (Gly, L-Lys, L-Glu, and L-Arg) or a peptide
(L-Arg-L-Arg) in the presence of 4-(dimethylamino)pyridine
(DMAP) to give the esters 11. The Boc group in compounds
6677
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Scheme 2. Syntheses of the Ester and Amide Prodrugsa
a Reagents and conditions: (a) DMAP, iPr2EtN, CH2Cl2, room temperature, 3 h, 72ꢀ81%. (b) 4 N HCl in 1,4-dioxane, ethyl acetate/CH2Cl2, 0 °C to
room temperature, 48 h, 71ꢀ79%. (c) THF, ꢀ20 °C to room temperature, 1 h, 61ꢀ79%.
Table 2. Ex Vivo Hydrolyses of Amide Prodrugs of 3
half-life at 37 °C (min)
amide
human
plasma
human
blood
aqueous solubility
(mg/mL)
prodrugs
5a
5b
5c
5d
5e
32.4 ( 2.7
25.1 ( 5.2
32.8 ( 3.6
29.5 ( 0.7
11.8 ( 0.5
27.1 ( 2.5
15.3 ( 1.9
27.9 ( 3.8
25.4 ( 0.1
5.9 ( 0.6
>10a
>10a
>10a
>10a
>10a
a Because of the limited sample supply, no further effort was made to
measure the maximum solubility.
11, in turn, was readily removed by treatment with 4 N HCl in
1,4-dioxane to result in the ester prodrugs 4. The acid treatment
for the removal of the Boc group in the presence of thiirane
worked well, as described in Scheme 1 for the synthesis of 3.
When 3 was subjected to the same acylation conditions (using
NHS esters of an amino acid), the desired amide bond was not
formed. Instead, it gave only the recovery of the two starting
materials. The Boc-protected amino acid was activated to a mixed
anhydride by the treatment of isobutyl chloroformate in the
presence of N-methylmorpholine and was allowed to react with 3
to give the desired amide linkage. The removal of the Boc group
in compounds 12 was carried out by acid treatment resulting in
the amide prodrugs 5.
Aqueous Solubility and ex Vivo Hydrolyses of Prodrugs.
The aqueous solubilities of the prodrugs were assessed by UV
spectroscopy (see Experimental Section). The solubility of
compounds 4 and 5 exceeded 10,000 μg/mL, which indicates
a significant enhancement in aqueous solubility over that of the
parent 1 (2.3 μg/mL) of over 5000-fold. Next, we explored the
stability of compounds 4 and 5 in aqueous solution. The ester
prodrugs 4 were unstable under aqueous conditions, and within
2 h, the aqueous solutions turned turbid. The precipitate was
analyzed and was confirmed as compound 2. The instability
of the ester prodrugs in aqueous solution might be due to the
promotion of a water molecule by the aminoacyl moiety itself for
hydrolysis of the ester group or by intermolecular aminolysis of
the ester moiety by the free amino group(s) in the promoiety.
Figure 1. Ex vivo hydrolysis of 5d in human blood generates 3.
Hence, the ester derivatives 4 would appear to be inherently
unstable under these conditions. However, the corresponding
amide derivatives 5 were stable in aqueous solution for over one
month, as assessed by HPLC.
We subsequently explored the properties of the prodrugs in
the presence of human plasma and human whole blood. The
ester prodrugs 4 hydrolyzed completely in human plasma within
2 min. In contrast, the amide derivatives 5 displayed increased
stability in human plasma. The half-lives of the amide prodrugs in
human plasma and whole blood are summarized in Table 2. The
amide prodrugs 5aꢀd quantitatively released the active consti-
tuent 3 in human plasma within 30 min. Whereas the half-life for
hydrolysis of the dipeptide variant 5e was shorter in the series,
the hydrolytic enzymes in plasma first gave rise to the mono-
arginine variant 5d, which experienced further hydrolysis to 3.
Hydrolyses of the amide conjugates were in general more rapid
in whole blood than in plasma, significantly so in the case of
6678
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Table 3. Kinetic Parameters for the Inhibition of Gelatinases
MMP-2
MMP-9a
ꢀ3
ꢀ3
on
amide prodrugs
10
k
on
(Mꢀ1sꢀ1
)
104 koff (sꢀ1
)
Ki (μM)
10
k
(Mꢀ1sꢀ1
)
104 koff (sꢀ1
)
Ki (μM)
5a
1.9 ( 0.6
3.4 ( 0.1
2.5 ( 0.1
1.5 ( 0.1
1.5 ( 0.1
8.9 ( 0.2
8.2 ( 0.9
2.1 ( 0.9
1.8 ( 1.6
7.6 ( 2.9
6.1 ( 2.1
2.1 ( 1.3
0.44 ( 0.05
0.21 ( 0.04
0.27 ( 0.04
0.28 ( 0.02
0.031 ( 0.006
0.044 ( 0.004
0.61 ( 0.01
6.5 ( 0.6
4.2 ( 0.7
12.0 ( 0.7
6.0 ( 0.3
8.3 ( 2.3
5.3 ( 0.7
3.1 ( 0.3
1.6 ( 0.3
4.3 ( 0.4
19.6 ( 1.0
18.7 ( 5.6
0.87 ( 0.11
5b
0.062 ( 0.025
0.069 ( 0.064
0.52 ( 0.20
5c
5d
5e
0.42 ( 0.15
3
0.024 ( 0.015
a Catalytic domain.
Table 4. Half-Life and Apparent Intrinsic Clearance in Liver
Microsomes
5d
3
half-life
(min)
CLint,app
half-life
(min)
CLint,app
species
(mL/min/kg)
(mL/min/kg)
mouse
rat
>60
<47
27.3 ( 2.1
23.4 ( 0.7
>60
102.8 ( 7.9
60.0 ( 1.8
<10
Figure 2. Reconstructed ion chromatogram following a 60-min incuba-
tion of 5d with rat S9 liver microsomes.
59.3 ( 4.5
23.7 ( 1.8
<10
human
>60
in the mouse and rat, in contrast to the case in human, were
>70% hepatic blood flow, indicating that 3 was a higher-clearance
compound in these species. In contrast, the prototype inhibitor
1 was readily metabolized and had an intrinsic clearance of
117 mL/min/kg in the rat, significantly higher than the hepatic
blood flow of 66 mL/min/kg.20 These data collectively indicate
that 5d has superior pharmacokinetic properties compared to
those of 3 and 1.
Metabolism Pathway. The metabolism of 5d was similar
across species. The major metabolic pathway of 5d was hydro-
lysis to the active constituent 3 (Figure 2). Minor oxidation at the
α position to the sulfonyl group to lead to the corresponding
sulfinic acid M1 and N-acetylation of 3 to give M3 (Figure 2)
were observed. M1 is not observed on the reconstructed ion
chromatogram as it is a minor metabolite, and it readily experi-
ences hydrolysis to give M2, which was detected. The prodrug 5d
was hydrolyzed to 3 by amidases present in microsomes, and
further oxidation gave its sulfinic acid metabolite M2. The
identities of the sulfinic acids M1 and M2 were assigned on the
basis of MS/MS spectra (Figure 3). We had previously observed
a similar fragmentation pattern for the sulfinic acid of 1.21 The
identity of M3 was confirmed by comparison of the HPLC
retention time and MS/MS spectrum (Figure 3) with those of an
authentic synthetic standard, prepared as reported earlier.28
compounds 5b and 5e. The time-dependent release of 3 from the
arginyl amide prodrug 5d is depicted in Figure 1.
Gelatinase Inhibition. The kinetics of gelatinase inhibition of
the amide prodrugs is given in Table 3. The amide prodrugs
showed slow-binding behavior in the inhibition of gelatinases,
but the activities were consistently worse for the prodrugs,
compared to that of the active constituent 3, which is released
from them. Several general observations can be made of these
compounds. First, the rate constants for the onset of inhibition
(kon) are rapid, and the rate constants for the reversal of
inhibition (koff) are slow. From the ratio of koff/kon, the dissocia-
tion constants (Ki) are evaluated. Compound 3, the species that
is released from the prodrugs, is the most effective of the
compounds listed in Table 3, inhibiting the enzymes with Ki
values in the nanomolar range.
Metabolic Stability in Mouse, Rat, and Human Liver
Microsomes. The in vitro metabolism of 5d and 3 was further
evaluated in mouse, rat, and human liver microsomal incuba-
tions. The prodrug 5d was metabolically stable, as indicated by
half-lives of 60 min or longer in mouse, rat, and human liver
microsomes (Table 4). The prodrug 5d was more metabolically
stable than 3. Compound 3 was stable in human liver micro-
somes, with t1/2 > 60 min; the rates of metabolism in mouse and
rat liver microsomes were faster, with t1/2 values of 27.3 and 23.4
min, respectively. Compound 3 was significantly more metabo-
lically stable than 1, which had a t1/2 of 12.0 min in rat liver
microsomes.20
Intrinsic clearance (CLint), a measure of how readily the liver
metabolizes a compound, was calculated from the in vitro half-
lives. The prodrug 5d had intrinsic clearances of <47, 23.7, and
<10 mL/min/kg in mouse, rat, and human, respectively, which
were <50% of the hepatic blood flow (86, 66, and 21 mL/min/kg
in mouse, rat, and human, respectively),31 indicating that the
compound would not be readily metabolized by the liver.
Intrinsic clearances for 3 were 103, 60, and <10 mL/min/kg in
mouse, rat, and human, respectively. Intrinsic clearances of 3
To gain further insight into the metabolism of 5d after the
removal of the promoiety, we investigated the metabolism of 3.
Incubation of 3 with liver microsomes showed oxidation at the
α-position to the sulfonyl to give the sulfinic acid derivative M2.
6679
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
To detect other minor metabolites of 3, the ion chromatogram at
m/z 338 (MH + 16)+ was reconstructed (Figure 4, inset). The
chromatogram displayed two peaks, labeled as M4 and M5, with
distinct fragmentation patterns. Product ion mass spectrometric
analysis of M5 showed cleavage of the sulfonyl-methylenethiir-
ane linkage giving an ion at m/z 248, followed by a subsequent
loss of sulfur to give a fragment at m/z 216, and further loss of
two oxygens to generate the fragment at m/z 184. The product
ion mass spectrum of 3 produced identical fragments at m/z 248,
216, and 184, indicating that the terminal phenyl ring and the
middle phenyl ring were intact and that oxidation of the thiirane
sulfur was the most likely transformation to account for the
increase in mass of 16 amu in M5. Loss of 16 amu from MH+ to
give a fragment at m/z 322 confirmed the assignment of M5 as
the sulfoxide of 3. We had seen this fragmentation previously for
the sulfoxide metabolite of 1.21 The MS/MS spectrum of M4,
however, indicated cleavage of the sulfonyl-methylenethiirane
linkage to generate a fragment at m/z 264, 16 amu higher than
that of 3. Loss of SO2 gave an ion at m/z 200. These data
suggested the potential oxidation of the amino group or oxida-
tion at the terminal phenyl or the middle phenyl ring. While
hydroxylation of the middle ring is possible, we had not observed
such an oxidative pathway for 1 and its derivatives previously.21,22
Since the fragmentation pattern could not discriminate between
the variant 13 and the N-hydroxylamine 14, we prepared the
authentic synthetic standards for each of compounds 13 and 14
to confirm the identity of the minor metabolite M4.
The synthesis of derivative 13 started by base-mediated
coupling of the activated fluorobenzene 15 and phenolic 16 to
yield 17, followed by the reduction of the nitro group to the
amine and subsequent N-Boc protection (Scheme 3). Basic
hydrolysis of 18 at reflux selectively deprotected the thiophenol,
which was then allowed to react with epichlorohydrin to generate
the epoxide 19. The succeeding steps involved oxidation to the
sulfone 20, conversion to the thiirane 21, and final MOM and
N-Boc removal by acid reflux to give 13. A typical method to
prepare N-hydroxylamine is the partial reduction of the nitro
compound using zinc dust in the presence of ammonium
chloride.32 The transformation of the nitro derivative 22, pre-
pared as published,28 to N-hydroxylamine 14 proceeded without
decomposition of the thiirane moiety.
HPLC analysis revealed that M4 had an identical retention
time as the synthetic 13 (Figure 5). Identification of M4 as
derivative 13 was further confirmed by MS/MS analyses of M4
and the synthetic standards (Figure 6). Compound 13 had the
identical product ion mass spectrum as metabolite M4. The
MS/MS spectrum of the synthetic N-hydroxylamine 14 showed
similar fragments (at m/z 264 and 200) as the metabolite.
However, the ions at m/z 321, due to the loss of OH from the
molecular ion, and at m/z 109, corresponding to the phenyl-
hydroxylamine fragment, established that M4 was not the
Figure 3. Product ion mass spectra of (A) metabolite M1 (MH+, m/z
406), (B) M2 (MH+, m/z 250), and (C) M3 (MH+, m/z 364).
Figure 4. Product ion mass spectra of (A) metabolite M4 and (B) M5, and (inset) a reconstructed ion chromatogram of m/z 338 (MH + 16)+, following
a 30-min incubation of 3 with rat liver microsomes.
6680
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Scheme 3. Syntheses of 13 and the N-Hydroxylamine 14 Derivativea
a Reagents and conditions: (a) Cs2CO3, DMF, room temperature, 24 h, 88%. (b) Fe, NH4Cl, MeOH/H2O, reflux, 2 h, 83%. (c) Boc2O, Et3N, MeOH,
room temperature, 24 h, 76%. (d) KOH, MeOH, reflux, 4 h; then epichlorohydrin, room temperature, 10 min, 68%. (e) m-CPBA, CH2Cl2, 0 °C to room
temperature, 10 min, 89%. (f) Thiourea, MeOH/CH2Cl2, room temperature, 24 h, 83%. (g) HCl, MeOH, reflux, 1 h, 95%. (h) Zn, NH4Cl, CH2Cl2/
H2O, room temperature, 0.5 h, 69%.
Figure 5. Identification of M4 by comparison of HPLC retention time to synthetic standards. (A) Reconstructed ion chromatogram of m/z 338
following a 30-min incubation of 3 with rat liver microsomes and (B) HPLC chromatogram of a mixture of synthetic standards 13 and 14 using UV
absorbance at 245 nm.
N-hydroxylamine derivative 14. These data conclusively identi-
fied metabolite M4 as compound 13.
derivative (compound 14) is not generated in vivo, we dosed
mice with 5d or 3 and screened plasma and urine samples for the
presence of compound 14. The HPLC and mass spectrometry
experiments, along with the direct comparison to the authentic
synthetic sample, documented the absence of 14 in vivo.
The metabolism pathway for 5d is summarized in Figure 7.
Prodrug 5d is hydrolyzed to the active gelatinase inhibitor 3 as
the major biotransformation. A minor metabolism pathway of 5d
and 3 is the oxidation at the α-methylene to give rise to the
sulfinic acids M1 and M2. In another minor pathway, the
gelatinase inhibitor 3 is oxidized at the terminal phenyl ring to
the 3-hydroxy analog 13. N-Oxidation of the aromatic amine of 3
to N-hydroxylamine 14 does not take place.
Exploration of Potential Toxicity. Aromatic amines could
be toxic due to the oxidation by cytochromes P450 to reactive
N-hydroxylamine derivatives, which can be conjugated with
DNA and proteins or oxidized to nitrosoarenes, which are
reactive entities in their own right.33 These undesired reactivities
could result in toxicity and mutagenicity. While 3 contains an
aromatic amine, we know definitively from the foregoing experi-
ments that it is not oxidized to the corresponding N-hydroxyla-
mine derivative 14 in vitro. To confirm that the N-hydroxylamine
Notwithstanding the total absence of evidence for the existence of
the N-hydroxylamine derivative 14, we also investigated the poten-
tial mutagenicity of both 5d and 3 by the Ames II mutagenicity
test.34 This assay evaluates the mutagenic potential of a compound
by measuring its ability to induce reverse mutations at selected loci
of Salmonella typhimurium mixed and tester strains in the presence
and absence of rat S9 liver activation (see Experimental Section).
The mixed strain contains an equimolar mixture of six strains, which
are individually designed to revert by only one specific base-pair
substitution. Thus, when mixed, all base-pair mutations are repre-
sented in one culture. The tester strain is used to detect frame-shift
mutations. Both 5d and 3 gave negative (nonmutagenic) responses
in the Ames II mutagenecity assay with and without S9 metabolic
activation up to concentrations of 1 mg/mL (equivalent to 1.8 and
6681
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Figure 6. Product ion mass spectra of (A) metabolite M4, (B) synthetic standard 13, and (C) synthetic standard 14.
Figure 7. Metabolism pathway of 5d. Active gelatinase inhibitors are boxed; the major pathway is represented by a thick arrow, minor pathways are
depicted by thin arrows, and minute pathways are represented by broken thin arrows. The crossed red arrow indicates the absence of 14.
2.8 mM of 5d and 3, respectively). These results substantiate the
nonmutagenic and nontoxic properties of the prodrug 5d, as well as
the active gelatinase inhibitor 3.
MMP Selectivity. The arginyl amide prodrug 5d and 3 were
evaluated for selectivity toward inhibition of MMPs. Although 5d
showed slow-binding behavior against MMP-14cat, its high Ki of
28.9 ( 2.3 μM essentially renders it inactive against the enzyme
(Table 5). Furthermore, 5d was an extremely poor linear
competitive inhibitor of MMP-1cat, MMP-3cat, and MMP-7. As
with gelatinases, 3 also showed good slow-binding inhibition of
MMP-14cat with a Ki of 0.21 ( 0.02 μM. MMP-14 is the cell
surface activator of pro-MMP-2,35 cleaving the N-terminal
6682
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Table 5. Kinetic Parameters for the Inhibition of MMPs
which was not observed in the plasma. As predicted from the
ex vivo experiments, administration of 5d to mice resulted in
hydrolysis to 3, which further metabolized to the p-N-acetyl
derivative (M3). Pharmacokinetic parameters are calculated in
Table 6.
Following intravenous administration of 5d, plasma levels of
5d were initially 198 μM and declined rapidly, with a t1/2α
(distribution half-life) of less than 1 min and a longer elimination
half-life (t1/2β) of >4 h. Systemic exposure (AUC) of 5d was 363
ꢀ3
10
k
104 koff
(sꢀ1
Ki
on
(Mꢀ1 sꢀ1
)
)
(μM)
Prodrug 5d
12.0 ( 0.1
MMP-14a
MMP-1a
MMP-3a
MMP-7
0.042 ( 0.001
28.9 ( 2.3
43% inhibition at 1 mM
40.5 ( 1.2
160 ( 14
μM min. The calculated clearance (CL) of 0.0582 L/min/kg
3
was lower than the hepatic blood flow of 0.086 L/min/kg,31
indicating moderate clearance of the prodrug from systemic
circulation. The volume of distribution (Vd) of 0.0978 L/kg
(equivalent to 2.30 mL) was significantly lower than the total
body water of 14.5 mL,31 suggesting that 5d was poorly dis-
tributed to tissues and remained primarily in the extracellular
fluid compartment. The low volume of distribution for the
prodrug 5d was attributed to the fact that this compound was
designed to be readily hydrolyzed to 3 by amidases present in
blood. Thus, 5d is not anticipated to be distributed to tissues and
is expected to remain in the extracellular fluid compartment, as
was found experimentally. This is consistent with prodrugs that
are readily hydrolyzed in blood to the active drugs37,38 and have
volumes of distribution significantly lower than total body water.
For example, fospropofol is a phosphate ester water-soluble
prodrug which is readily hydrolyzed to propofol in blood.
Fospropofol shows a Vd of 0.25 L/kg in rats and 0.11 L/kg in
humans, while propofol has a Vd of 2.3 L/kg in rats and 12.4 L/kg
Compound 3
12.7 ( 0.8
MMP-14a
MMP-1a
MMP-3a
MMP-7
6.0 ( 0.5
0.21 ( 0.02
10% inhibition at 250 μM
23.4 ( 1.6
16% inhibition at 250 μM
Compound 13 (M4)
1.4 ( 0.7
MMP-2
MMP-9a
MMP-14a
MMP-1a
MMP-3a
MMP-7
1.7 ( 0.1
1.1 ( 0.1
2.7 ( 0.3
0.078 ( 0.043
5.3 ( 1.0
0.49 ( 0.11
7.0 ( 1.6
0.26 ( 0.07
11% inhibition at 300 μM
54% inhibition at 300 μM
116 ( 13
M3
MMP-2
MMP-9
MMP-14
MMP-1
MMP-3
MMP-7
1.2 ( 0.3
1.3 ( 0.3
0.11 ( 0.04 b
0.13 ( 0.01b
0.68 ( 0.05b
5.4 ( 0.4 b
12.2 ( 0.9 b
39 ( 3 b
in humans.37,38 Systemic exposure of 3 was 138 μM min, 2.6-
3
fold lower than that of 5d. The terminal half-life of 3 was 69 min
or 3-fold shorter than that of 5d. The p-N-acetyl derivative (M3),
generated from N-acetylation of 3, had a distribution half-life of
a Catalytic domains. b Reported earlier and given here for the sake of
12 min and a systemic exposure of 4.22 μM min, which was
3
side-by-side comparison.28
significantly lower (33-fold) than the systemic exposure of 3.
This is consistent with the observation of N-acetylation as a
minor metabolism pathway. Mice that received 5d had quantifi-
able levels of both the prodrug and the released active constituent
3 up to 4 h postadministration of 5d.
prodomain of pro-MMP-2 and generating the activated inter-
mediate, which matures into the fully active form of MMP-2.36
However, 3 extremely poorly inhibited MMP-1cat, MMP-3cat, and
MMP-7.
In contrast, intravenous administration of 3 resulted in initial
As observed for the sulfinic acid and sulfoxide derivatives of
1,21 the minor sulfinic acid metabolites of 5d, M1 and M2, as well
as the sulfoxide M5, are also expected to be poor inhibitors of
gelatinases, as the critical thiirane warhead in these compounds is
missing. The N-acetyl metabolite (M3) is, however, a potent
inhibitor of gelatinases.28 Compound 13 (M4) was also tested for
inhibitory activities toward MMPs. It displayed a slow-binding
inhibition profile similar to that of 3, with inhibition constants
(Ki) of 0.078 ( 0.043, 0.49 ( 0.11, and 0.26 ( 0.07 μM for
MMP-2, MMP-9cat, and MMP-14cat, respectively, while sparing
MMP-1cat, MMP-3cat, and MMP-7 (Table 5). The concept of the
prodrug strategy is well demonstrated by 5d, which is hydrolyzed
in the bloodstream to the active 3, which, in turn, gets further
metabolized to two additional minor metabolites, M3 and M4,
both of which are potent gelatinase inhibitors in their own right.
Note that while M4 was detected as a minor metabolite in vitro, it
was not observed in vivo.
plasma levels of 36.4 μM and systemic exposure of 221 μM min.
3
The terminal half-life of 21 min was considerably shorter than
that of 3 generated from the administration of 5d. Clearance of
0.0875 L/min/kg was >70% hepatic blood flow of 0.086 L/min/
kg,31 indicating high clearance of 3. The volume of distribution
was 0.532 L/kg (equivalent to 12.5 mL) and approximated total
body water of 14.5 mL, indicating that 3 was distributed to the
tissues. Moreover, the resulting p-N-acetyl metabolite (M3)
had an initial half-life of 5.1 min, 2.4-fold shorter than that
generated after the administration of 5d. Systemic exposure of
7.71 μM min, 29-fold lower than that of 3, further indicated that
3
N-acetylation was a minor metabolism pathway and/or that the
p-N-acetyl metabolite was cleared from systemic circulation.
Comparison of the blood levels of 3 after a single bolus
intravenous administration of 5d and 3 to mice is shown graphi-
cally in Figure 8. More favorable pharmacokinetic properties were
observed following intravenous administration of 5d than 3.
In particular, the terminal half-life of 3 generated from 5d was
more than 3-fold longer than that after the administration of 3
itself. The in vivo pharmacokinetic data were consistent with the in
vitro results and indicated that 5d had superior pharmacokinetic
properties compared to those of 3. The >4 h elimination half-life
for 5d indicates that this prodrug can be administered as an
Pharmacokinetics. The pharmacokinetics of 5d and 3 were
separately evaluated in mice, following a single bolus intravenous
dose administration at 12.5 mg/kg. Concentrations of 5d, 3, and
the p-N-acetyl derivative (M3) are summarized in Table 6. We
did not determine the blood levels of the other metabolites
because their concentrations were quite low, including M4,
6683
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Table 6. Pharmacokinetic Parameters of 5d and 3 after a Single Bolus Intravenous Dose to Mice
IV dose of 5d
IV dose of 3
time (min)
5d (μM)
3 (μM)
M3 (μM)
3 (μM)
M3 (μM)
0
198
25.7
36.4
2ꢀ3
4ꢀ5
7ꢀ8
11ꢀ12
22
35.1 ( 9.65
12.3 ( 3.79
10.4 ( 2.31
4.50 ( 0.361
2.45 ( 0.664
0.787 ( 0.373
0.0354 ( 0.0184
0.231 ( 0.0197
0.176 ( 0.0913
0.149 ( 0.0243
0.0627 ( 0.0169
18.4 ( 3.56
13.4 ( 3.29
5.33 ( 0.863
3.44 ( 0.775
0.197 ( 0.0616
0.462 ( 0.313
0.192 ( 0.0121
0.150 ( 0.0272
7.32 ( 5.76
1.28 ( 0.0544
1.58 ( 0.423
0.138 ( 0.0451
30
1.11 ( 1.11
0.0958 ( 0.0830
42
0.0872 ( 0.00313
0.0385 ( 0.00582
0.0171 ( 0.00652
0.00830 ( 0.00820
0.207 ( 0.0388
0.0908 ( 0.0403
0.0248 ( 0.00652
0.0145 ( 0.00573
0.0358 ( 0.0178
0.0121 ( 0.00181
NQa
62ꢀ66
120ꢀ123
240
0.138 ( 0.0609
0.0190 ( 0.00350
NQa
0.0273 ( 0.00628
0.00277 ( 0.00480
NQa
NQq
AUC0ꢀlast (μM min)
360
137
138
4.22
NCb
221
7.71
NCb
3
AUC0ꢀ∞ (μM min)
363
221
3
CL (L/min/kg)
0.0582
0.0978
0.88
0.0875
0.532
3.0
Vd (L/kg)
t1/2α (min)
t1/2β (min)
2.9
69
12
NCb
5.1
NCb
260
21
a NQ = not quantifiable. b NC = not calculated because the low levels observed did not allow for the calculation of the terminal half-life and AUClastꢀ∞
.
thiolate that ties up the enzyme in manifestation of the biological
activity of this class of compounds in vivo.
’ CONCLUSION
Gelatinases are important targets of inhibition in many
diseases, in particular acute neurological conditions such as
stroke, aneurysm, and traumatic brain injury. For treatment of
these diseases, intravenous infusion is the preferred route
of administration, which requires a drug to possess a water
solubility of 10 mg/mL. We addressed the poor water solubility
and extensive metabolism of the prototype thiirane inhibitor 1 by
blocking the primary site of metabolism and capitalizing on a
prodrug strategy. The prodrugs increased aqueous solubility by
more than 5000-fold. The amide prodrugs were quantitatively
hydrolyzed in human blood to a potent and selective gelatinase
inhibitor. The arginyl amide prodrug 5d was metabolically stable
in mouse, rat, and human liver microsomes. The major metabo-
lism pathway of 5d was hydrolysis to the corresponding p-amino
derivative 3; minor pathways were N-acetylation, oxidation at the
terminal phenyl ring, and oxidation at the α-methylene. Both N-
acetylation and hydroxylation at the terminal phenyl ring gener-
ated active gelatinase inhibitors. N-Hydroxylation of the aromatic
amine was not observed in vitro or in vivo. Both 5d and 3 were
nonmutagenic in the Ames II mutagenicity assay. Following a
single intravenous bolus administration of 5d to mice, systemic
exposures of 5d, 3, and the p-N-acetyl derivative were 363, 138,
Figure 8. Plasma concentrationꢀtime curves of 3 after a single bolus
intravenous dose of 5d and 3 to mice.
intravenous infusion, which is the preferred route of administra-
tion in the treatment of acute gelatinase-dependent diseases, such
as stroke, aneurysm, and traumatic brain injury.
Plasma levels of 3 remained above the Ki for MMP-2 for 2 h
and were higher than the Ki for MMP-9 for 20 min. However, we
had previously demonstrated that the mechanism of inhibition of
the thiirane class of gelatinase inhibitors is a rate-limiting
gelatinase-mediated generation of a thiolate within the enzyme
active site.12 Once the thiolate is generated, it has significantly
tighter binding than the parental thiirane, as documented for the
prototype 1, which shows a 26-fold more potent inhibition of
MMP-2 compared to the thiirane (Ki of 530 pM for the thiolate
versus 14 nM for the thiirane).12 The ability of gelatinases to
perform this chemistry within the active site is the reason for the
selectivity that this class of inhibitors enjoys in the inhibition of
gelatinases. Similarly, once 3 is generated from 5d in vivo, we
anticipate that on binding to the active site of gelatinases, the
same chemistry leads to the formation of the tight-binding
and 4.22 μM min, respectively. The prodrug 5d showed a
3
moderate clearance of 0.0582 L/min/kg, remained mostly in
the extracellular fluid compartment (Vd = 0.0978 L/kg), and had
a moderate terminal half-life of >4 h. The prodrug 5d had
superior pharmacokinetic properties compared to those of 3.
The high water solubility, chemical stability, quantitative enzymatic
hydrolysis of the amide prodrugs, metabolic stability, lack of
toxicity, and favorable pharmacokinetic properties make the
6684
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
thiirane class of selective gelatinase inhibitors suitable for intrave-
nous administration in the treatment of acute gelatinase-dependent
diseases.
δ 24.5, 26.3, 28.5, 62.8, 81.0, 117.4, 120.6, 121.4, 130.9, 131.8, 136.0,
150.0, 153.0, 163.6. HRMS-FAB (m/z): [M]+, calcd for C20H23NO5S2,
421.1018; found, 421.1009.
4-[4-((Thiiran-2-yl)methylsulfonyl)phenoxy]benazamine
3
HCl Salt (3). The thiirane (10, 4.0 g, 9.5 mmol) was dissolved in a 1:1
mixture of CH2Cl2 and ethyl acetate (40 mL), and HCl (10 mL, 4 N in
dioxane) was added. The reaction mixture was stirred for 48 h and
concentrated under reduced pressure. The resulting crude compound
was triturated with diethyl ether, and the product was obtained by
filtration (2.8 g, 82%). 1H NMR (500 MHz, D2O) δ 2.03 (d, J = 4.8 Hz,
1H), 2.42 (d, J = 6.4 Hz, 1H), 2.94 (m, J = 6.4, 6.4 Hz, 1H), 3.42 (dd, J =
14.4, 6.4 Hz, 1H), 3.61 (dd, J = 14.4, 6.4 Hz, 1H), 7.11 (d, J = 8.0 Hz,
2H), 7.16 (d, J = 8.0 Hz, 2H), 7.35 (d, J = 8.0 Hz, 2H), 7.80 (d, J = 8.0 Hz,
2H); 13C NMR (126 MHz, D2O) δ 23.5, 26.0, 61.4, 118.6, 121.9, 125.2,
126.8, 130.8, 131.1, 155.2, 162.5. HRMS-FAB (m/z): [M + H]+, calcd
for C15H16NO3S2, 322.0572; found, 322.0569.
Syntheses of Ester Prodrugs (4). Synthesis of the lysine prodrug
(4b) is given here as a representative example. To a solution of 2 (0.20 g,
0.62 mmol) in CH2Cl2 (5 mL), which was prepared by a literature
method,23 was added Nα,Nε-di-Boc-L-lysine hydroxysuccinimide ester
(0.41 g, 0.93 mmol) at room temperature. DMAP (76 mg, 0.62 mmol)
and diisopropylethyl amine (108 uL, 0.62 mmol) were added, and the
reaction mixture was stirred for 3 h at room temperature. After
concentrating the solution under reduced pressure, the crude product
was purified by column chromatography on silica gel to give the desired
product (11b, 0.30 g, 74%). The ester (0.30 g, 0.46 mmol) was dissolved
in CH2Cl2 and ethyl acetate (1:1, 4 mL), and HCl (2 mL, 4 N in
dioxane) was added. The reaction mixture was stirred for 48 h and was
concentrated under reduced pressure. The resulting crude compound
was triturated with diethyl ether, and the product (4b) was obtained by
filtration (0.18 g, 75%).
Syntheses of Amide Prodrugs (5). Synthesis of the lysine
prodrug (5b) is shown as a representative example. i-Butylchlorofor-
mate (83 μL, 0.53 mmol) was added to a mixture of Nα,Nε-di-Boc-L-
lysine dicyclohexylamine salt (0.28 g, 0.64 mmol) and N-methylmor-
pholine (140 μL, 1.3 mmol) in THF (4 mL) at ꢀ15 °C. After stirring for
0.5 h at the same temperature, the suspension of 3 (0.19 g, 0.53 mmol)
and N-methylmorpholine (58 μL, 0.53 mmol) in THF (2 mL) was
added to the reaction mixture. Stirring was continued for 1 h, while the
temperature was gradually increased to room temperature. The reaction
mixture was diluted with CH2Cl2/water, and layers were separated. The
aqueous layer was washed with CH2Cl2, and the combined organic layer
was dried (MgSO4) and evaporated to dryness. The crude material was
purified by column chromatography on silica gel to afford the desired
product (12b, 0.21 g, 64%). The amide (0.20 g, 0.30 mmol) was
dissolved in CH2Cl2 and ethyl acetate (1:1, 3 mL), and HCl (2 mL,
4 N in dioxane) was added. The reaction mixture was stirred for 48 h and
concentrated under reduced pressure. The resulting crude compound
was triturated with diethyl ether, and the product (5b) was obtained by
filtration (0.11 g, 71%).
’ EXPERIMENTAL SECTION
Chemistry. Organic reagents and solvents were purchased from
either Sigma-Aldrich Chemical Co. (St. Louis, MO, USA) or Acros
Organics (Geel, Belgium), unless otherwise stated, and were used
without further purification. All reactions were performed under an
atmosphere of nitrogen, unless noted otherwise. H and 13C NMR
1
spectra were recorded on a Varian INOVA-500 (Varian Inc., Palo Alto,
CA, USA) or a Varian DirectDrive 600 spectrometer (Varian Inc., Palo
Alto, CA, USA). Thin-layer chromatography was performed with What-
man reagents 0.25 mm silica gel 60-F plates. Flash chromatography was
carried out with silica gel 60, 230ꢀ400 mesh (0.040ꢀ0.063 mm particle
size) purchased from EM Science (Gibbstown, NJ, USA). High-resolu-
tion mass spectra were obtained at the Department of Chemistry and
Biochemistry, University of Notre Dame by FAB ionization, using a JEOL
AX505HA mass spectrometer (Tokyo, Japan) or ESI ionization, using a
Bruker micrOTOF/Q2 mass spectrometer (Bruker Daltonik, Bremen,
Germany). Purity of the prepared compounds was generally >95%,
confirmed by HPLC. Detailed conditions are given in the HPLC section.
t-Butyl [4-(4-allylthiophenoxy)phenyl]carbamate (9).
Compound 8 (10.3 g, 40.0 mmol), which was prepared by a literature
method,28 was dissolved in a mixture of MeOH and triethylamine (7:1,
80 mL), and di-t-butyl dicarbonate (17.5 g, 80 mmol) was added. The
resulting solution was stirred for 2 h at 60 °C and was concentrated
under reduced pressure. The crude material was purified by column
chromatography on silica gel to give the title compound (11.7 g, 82%).
1H NMR (600 MHz, CDCl3) δ 1.53 (s, 9H), 3.48 (d, J = 6.7 Hz, 2H),
4.96ꢀ5.09 (m, 2H), 5.86 (m, J = 16.7, 10.3 Hz, 1H), 6.47 (br. s, 1H),
6.89 (d, J = 8.8 Hz, 2H), 6.96 (d, J = 8.8 Hz, 2H), 7.24ꢀ7.39 (m, 4H);
13C NMR (151 MHz, CDCl3) δ 28.5, 38.9, 80.7, 117.7, 118.6, 120.3,
120.5, 128.9, 133.3, 134.0, 134.5, 152.2, 153.1, 157.4. HRMS-FAB
(m/z): [M]+, calcd for C20H23NO3S, 357.1399; found, 357.1402.
t-Butyl [4-(4-((Thiiran-2-yl)methylsulfonyl)phenoxy)phenyl]
carbamate (10). To a solution of compound 9 (10.0 g, 28.0 mmol) in
CH2Cl2 (100 mL) was added a solution of m-CPBA (31.2 g, 140 mmol,
77%) in an iceꢀwater bath. After 72 h, the suspension was filtered, and the
filtrate was diluted with EtOAc and washed with 10% aqueous sodium
thiosulfate, followed by washes with saturated sodium bicarbonate and
brine. The organic layer was dried over MgSO4 and was concentrated. The
product was purified by silica gel chromatography to yield the oxirane (7.4 g,
81%) with recovery of some of the allylsulfone derivative (2.1 g, 19%). 1H
NMR (600 MHz, CDCl3) δ 1.50 (s, 9H), 2.45 (dd, J = 5.0, 1.2 Hz, 1H),
2.79 (dd, J = 4.4, 2.6 Hz, 1H), 3.24ꢀ3.34 (m, 3H), 6.85 (s, 1H), 6.99 (d, J =
8.8 Hz, 2H), 7.02 (d, 2H), 7.42 (d, J = 8.2 Hz, 2H), 7.84 (d, J = 8.8 Hz, 2H);
13C NMR (151 MHz, CDCl3) δ 28.4, 46.0, 46.0, 59.7, 117.2, 120.5, 121.3,
130.6, 132.2, 136.0, 149.8, 153.0, 163.4. HRMS-FAB (m/z): [M]+, calcd for
C20H23NO6S, 405.1246; found, 405.1230.
Spectral Data of Gly Ester Prodrug Compound 11a. 1H NMR (500
MHz, CDCl3) δ 1.48 (s, 9H), 2.17 (dd, J = 5.0, 1.8 Hz, 1H), 2.55 (dd, J =
6.2, 1.8 Hz, 1H), 3.04ꢀ3.10 (m, 1H), 3.20 (dd, J = 14.4, 7.8 Hz, 1H),
3.52 (dd, J = 14.2, 5.6 Hz, 1H), 4.19 (d, J = 5.8 Hz, 2H), 7.11 (d, J = 9.0
Hz, 2H), 7.12 (d, J = 8.8 Hz, 2H), 7.18 (d, J = 9.0 Hz, 2H), 7.88 (d, J = 9.0
Hz, 2H); 13C NMR (126 MHz, CDCl3) δ 24.4, 26.2, 28.5, 42.7, 62.7,
80.5, 117.9, 121.5, 123.3, 131.0, 132.4, 147.3, 152.6, 162.8, 169.3.
HRMS-ESI (m/z): [M + H]+, calcd for C22H26N2O6S2, 480.1145;
found, 480.1145. Compound 4a: 1H NMR (500 MHz, D2O) δ 2.17 (dd,
J = 5.2, 1.2 Hz, 1H), 2.56 (d, J = 6.2 Hz, 1H), 3.09 (quin, J = 6.2 Hz, 1H),
3.57 (dd, J = 14.6, 7.2 Hz, 1H), 3.74 (dd, J = 14.6, 6.4 Hz, 1H), 4.25
(s, 2H), 7.21ꢀ7.35 (m, 6H), 7.93 (d, J = 8.8 Hz, 2H). HRMS-ESI (m/z):
[M + H]+, calcd for C17H18NO5S2, 380.0621; found, 380.0597.
Thiourea (2.4 g, 31.5 mmol) was added to a solution of the oxirane
(6.0 g, 14.8 mmol), obtained above, in a 1:1 mixture of methanol and
CH2Cl2 (50 mL). The reaction mixture was stirred at room temperature
for 18 h. The solvent was removed under reduced pressure. The residue
was partitioned between CH2Cl2 and water, the organic layer was
washed with brine and water and dried (MgSO4), and the suspension
was filtered. Evaporation of solvent gave the crude product, which was
purified by column chromatography on silica gel to give thiirane 10
(4.4 g, 72%). 1H NMR (600 MHz, CDCl3) δ 1.53 (s, 9H), 2.16 (dd, J =
5.1, 1.6 Hz, 1H), 2.54 (dd, J = 6.2, 1.2 Hz, 1H), 3.06 (dq, J = 7.6, 5.7 Hz,
1H), 3.17 (dd, J = 14.1, 7.9 Hz, 1H), 3.53 (dd, J = 14.1, 5.6 Hz, 1H), 6.61
(br. s, 1H), 7.06 (d, J = 8.8 Hz, 2H), 7.03 (d, J = 8.8 Hz, 2H), 7.44 (d, J =
8.5 Hz, 2H), 7.84 (d, J = 8.8 Hz, 2H); 13C NMR (151 MHz, CDCl3)
6685
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
Spectral Data of Lys Ester Prodrug Compound 11b. 1H NMR
(500 MHz, CDCl3) δ 1.45 (s, 9H), 1.47 (s, 9H), 1.52 (m, 2H), 1.56
(m, 4H), 1.85 (m, 1H), 1.99 (m, 1H), 2.17 (dd, J = 5.2, 1.8 Hz, 1H), 2.55
(dd, J = 6.2, 1.8 Hz, 1H), 3.07 (m, 1H), 3.17 (m, 2H), 3.19 (dd, J = 14.2,
7.8 Hz, 1H), 3.53 (dd, J = 14.2, 5.6 Hz, 1H), 4.49 (m, 1H), 4.61 (m, 1H),
5.23 (d, J = 7.4 Hz, 1H), 7.10 (d, J = 9.0 Hz, 1H), 7.11 (d, J = 8.8 Hz, 2H),
7.17 (d, J = 9.0 Hz, 1H); 13C NMR (126 MHz, CDCl3) δ 22.7, 24.4,
26.3, 28.5, 28.6, 29.9, 32.1, 40.1, 53.8, 62.8, 79.5, 80.3, 81.3, 117.9, 121.6,
123.4, 131.0, 132.4, 147.6, 152.6, 162.9, 171.8. HRMS-ESI (m/z):[M+H]+,
calcd for C31H43N2O9S2, 651.2404; found, 651.2414. Compound 4b: 1H
NMR (500 MHz, 10% CDCl3 in CD3OD) δ 1.62ꢀ1.84 (m, 4H),
2.05ꢀ2.26 (m, 2H), 2.14 (dd, J = 5.0, 1.4 Hz, 1H), 2.52 (dd, J = 6.3, 1.1
Hz, 1H), 3.01 (t, J = 7.6 Hz, 2H), 3.06 (m, 1H), 3.46 (dd, J = 14.6, 7.0 Hz,
1H), 3.54 (t, J = 6.6 Hz, 1H), 4.40 (t, J = 6.5 Hz, 1H), 7.19 (d, J = 9.0 Hz,
2H), 7.23 (d, J = 9.0 Hz, 2H), 7.32 (d, J = 9.2 Hz, 2H), 7.94 (d, J = 9.0 Hz,
2H). HRMS-ESI (m/z): [M + H]+, calcd for C21H27N2O5S2, 451.1356;
found, 541.1298.
CD3OD) δ1.71ꢀ2.23 (m, 8H), 2.16 (dd, J = 5.0, 1.2 Hz, 1H), 2.54 (dd, J =
6.2, 0.8 Hz, 1H), 3.03ꢀ3.10 (m, 1H), 3.23ꢀ3.32 (m, 4H), 3.47 (dd, J =
14.4, 7.0 Hz, 1H), 3.54 (dd, J = 14.2, 6.4 Hz, 1H), 4.16 (t, J = 6.4 Hz, 1H),
4.72 (dd, J = 9.1, 4.5 Hz, 1H), 7.15ꢀ7.23 (m, 4H), 7.25 (d, J = 9.0 Hz, 2H),
7.94 (d, J = 8.8 Hz, 2H). HRMS-ESI (m/z): [M + H]+, calcd for
C27H39N8O6S2, 635.2428; found, 635.2449.
1
Spectral Data of Gly Amide Prodrug Compound 12a. H NMR
(500 MHz, CDCl3) δ 1.37 (s, 9H), 2.05 (dd, J = 5.2, 1.8 Hz, 1H), 2.44
(dd, J = 6.2, 1.8 Hz, 1H), 2.94 (m, 1H), 3.14 (dd, J = 14.4, 7.6 Hz, 1H),
3.41 (dd, J = 14.4, 6.0 Hz, 1H), 3.80 (d, J = 5.2 Hz, 2H), 3.82 (s, 1H),
6.96 (d, J = 15.2 Hz, 2H), 6.98 (d, J = 15.4 Hz, 2H), 7.52 (d, J = 8.8 Hz,
2H), 7.75 (d, J = 9.0 Hz, 2H); 13C NMR (126 MHz, CDCl3) δ 24.0,
25.9, 28.2, 44.2, 62.5, 80.4, 117.4, 121.0, 121.8, 130.7, 131.4, 150.7, 163.3,
168.3, 172.9. HRMS-ESI (m/z): [M + Na]+, calcd for C22H26N2O6S2-
1
Na, 501.1124; found, 501.1112. Compound 5a: H NMR (500 MHz,
10% CDCl3 in CD3OD) δ 2.14 (dd, J = 5.2, 1.6 Hz, 1H), 2.51 (dd, J =
6.3, 1.5 Hz, 1H), 3.04 (m, J = 7.0, 5.2 Hz, 1H), 3.42 (dd, J = 14.4, 7.0 Hz,
1H), 3.53 (dd, J = 14.4, 6.4 Hz, 1H), 4.14 (s, 2H), 7.17 (d, J = 9.0, 2.0, 1.8
Hz, 1H), 7.20 (d, J = 9.2 Hz, 2H), 7.29 (d, J = 9.2 Hz, 2H), 7.92 (d, J =
9.2 Hz, 1H). HRMS-ESI (m/z): [M + H]+, calcd for C17H19N2O4S2,
379.0781; found, 379.0805.
Spectral Data of Glu Ester Prodrug Compound 11c. 1H NMR
(500 MHz, CDCl3) δ 1.45 (s, 18H), 2.09 (m, 1H), 2.15 (dd, J = 5.2,
1.8 Hz, 1H), 2.29 (m, 1H), 2.43 (q, J = 7.4 Hz, 2H), 2.52 (dd, J = 6.2,
1.8 Hz, 1H), 3.05 (m, 1H), 3.18 (dd, J = 14.4, 7.8 Hz, 1H), 3.51 (dd, J =
14.4, 5.8 Hz, 1H), 4.52 (m, 1H), 5.26 (d, J = 8.2 Hz, 1H), 7.08 (d, J =
9.0 Hz, 2H), 7.09 (d, J = 9.0 Hz, 2H), 7.17 (d, J = 9.2 Hz, 2H), 7.86 (d, J =
9.0 Hz, 2H); 13C NMR (126 MHz, CDCl3) δ 24.4, 26.2, 27.3, 28.2, 28.4,
31.7, 53.5, 62.7, 80.4, 81.1, 117.9, 121.4, 123.3, 130.9, 132.3, 147.5, 152.5,
155.6, 162.8, 171.3, 172.1. HRMS-ESI (m/z): [M + H]+, calcd for
Spectral Data of Lys Amide Prodrug Compound 12b. 1H NMR (500
MHz, CDCl3) δ 1.41 (s, 9H), 1.43 (s, 9H), 2.02 (s, 3H), 2.13 (dd, J =
5.0, 1.6 Hz, 1H), 2.50 (dd, J = 6.2, 1.6 Hz, 1H), 3.02 (m, 1H), 3.09
(m, 2H), 3.16 (dd, J = 14.4, 7.8 Hz, 1H), 3.50 (dd, J = 14.4, 5.8 Hz, 1H),
4.28 (m, 1H), 4.75 (m, 1H), 5.53 (d, J = 7.4 Hz, 1H), 6.93 (d, J = 8.4 Hz,
2H), 7.00 (d, J = 8.6 Hz, 2H), 7.55 (d, J = 8.8 Hz, 2H), 7.82 (d, J = 9.0 Hz,
2H); 13C NMR (126 MHz, CDCl3) δ 24.4, 26.2, 28.5, 28.6, 29.8, 31.9,
39.8, 55.3, 62.7, 79.3, 80.5, 117.4, 121.1, 121.7, 130.8, 131.8, 135.6,
150.5, 156.4, 156.6, 163.3, 171.2. HRMS-ESI (m/z): [M + Na]+, calcd
for C31H43N3O8S2Na, 672.2384; found, 672.2393. Compound 5b: 1H
NMR (500 MHz, 10% CDCl3 in CD3OD) δ 1.5ꢀ1.86 (m, 4H),
2.06ꢀ2.25 (m, 2H), 2.14 (dd, J = 5.2, 1.4 Hz, 1H), 2.52 (dd, J =
6.4, 1.4 Hz, 1H), 3.00 (t, J = 7.6 Hz, 2H), 3.05 (m, 1H), 3.45 (dd, J =
14.4, 6.8 Hz, 1H), 3.53 (dd, J = 14.6, 6.6 Hz, 1H), 4.39 (t, J = 6.5 Hz, 1H),
7.18 (d, J = 8.8 Hz, 2H), 7.22 (d, J = 9.2 Hz, 2H), 7.32 (d, J = 9.0 Hz, 2H),
7.93 (d, J = 9.0 Hz, 2H). HRMS-ESI (m/z): [M + H]+, calcd for
C21H28N3O4S2, 450.1516; found, 450.1485.
1
C29H38NO9S2, 608.1982; found, 608.1981. Compound 4c: H NMR
(500 MHz, 10% CDCl3 in CD3OD) δ 2.14 (dd, J = 5.1, 1.3 Hz, 1H),
2.37 (m, 2H), 2.52 (dd, J = 6.2, 1.0 Hz, 1H), 2.68 (td, J = 7.1, 1.6 Hz, 2H),
3.05 (quin, J = 6.2 Hz, 1H), 3.44 (dd, J = 14.4, 7.0 Hz, 1H), 3.54 (dd, J =
14.4, 6.4 Hz, 1H), 4.43 (t, J = 6.8 Hz, 1H), 7.19 (d, J = 8.8 Hz, 2H), 7.22
(d, J = 9.2 Hz, 2H), 7.32 (d, J = 9.0 Hz, 2H), 7.93 (d, J = 9.0 Hz, 2H).
HRMS-ESI (m/z): [M + H]+, calcd for C20H22NO7S2, 452.0832: found,
452.0769.
Spectral Data of Arg Ester Prodrug Compound 11d. 1H NMR
(500 MHz, CDCl3) δ 1.46 (s, 9H), 1.50 (s, 9H), 1.77 (m, 1H), 1.94
(m, 1H), 2.13 (dd, J = 5.0, 1.6 Hz, 1H), 2.51 (dd, J = 6.2, 1.6 Hz, 1H),
3.03 (m, 1H), 3.17 (dd, J = 14.2, 7.8 Hz, 1H), 3.49 (dd, J = 14.2, 5.6 Hz,
1H), 3.84 (m, 1H), 3.96 (m, 1H), 4.50 (m, 1H), 5.83 (d, J = 8.2 Hz, 1H),
7.07 (d, J = 9.0 Hz, 2H), 7.08 (d, J = 8.8 Hz, 2H), 7.14 (d, J = 9.0 Hz, 1H),
7.85 (d, J = 9.0 Hz, 2H); 13C NMR (126 MHz, CDCl3) δ 24.3, 26.2,
28.1, 28.3, 28.5, 44.1, 53.5, 60.5, 62.7, 84.1, 117.8, 121.4, 123.4, 130.9,
132.3, 147.7, 152.4, 155.0, 155.6, 160.6, 162.8, 163.6, 171.2. HRMS-ESI
(m/z): [M + H]+, calcd for C36H51N4O11S2, 779.2990; found, 779.3001.
Compound 4d: 1H NMR (500 MHz, 10% CDCl3 in CD3OD) δ
1.78ꢀ2.26 (m, 4H), 2.14 (dd, J = 5.1, 1.3 Hz, 1H), 2.52 (dd, J = 6.2,
0.8 Hz, 1H), 3.04 (quin, J = 6.2 Hz, 1H), 3.27 (m, 2H), 3.43 (dd, J = 14.4,
7.0 Hz, 1H), 3.52 (dd, J = 14.8, 6.6 Hz, 1H), 4.39 (t, J = 6.5 Hz, 1H), 7.17
(d, J = 9.0 Hz, 1H), 7.21 (d, J = 9.0 Hz, 1H), 7.30 (d, J = 9.0 Hz, 2H), 7.92
(d, J = 8.8 Hz, 2H). HRMS-ESI (m/z): [M + H]+, calcd for
C21H27N4O5S2, 479.1417; found, 479.1423.
1
Spectral Data of Glu Amide Prodrug Compound 12c. H NMR
(500 MHz, CDCl3) δ 1.47 (s, 9H), 1.49 (s, 9H), 2.00 (m, 1H), 2.17
(dd, J = 5.2, 1.8 Hz, 1H), 2.41 (m, 1H), 2.56 (m, 2H), 3.07 (m, 1H), 3.18
(dd, J = 14.3, 7.9 Hz, 1H), 3.54 (dd, J = 14.1, 5.5 Hz, 1H), 4.27 (m, 1H),
7.07 (d, J = 8.6 Hz, 2H), 7.08 (d, J = 9.0 Hz, 2H), 7.61 (d, J = 9.0 Hz, 1H),
7.86 (d, J = 9.0 Hz, 1H); 13C NMR (126 MHz, CDCl3) δ 24.5, 26.3,
28.2, 28.5, 37.1, 51.5, 62.8, 82.3, 117.6, 121.3, 121.9, 130.9, 131.9, 135.2,
151.0, 163.3, 169.3, 171.7. HRMS-ESI (m/z): [M + Na]+, calcd for
C29H38N2O8S2Na, 629.1962; found, 629.1969. Compound 5c: 1H NMR
(500 MHz, 10% CDCl3 in CD3OD) δ 2.32 (m, 1H), 2.42 (m, 1H), 2.51
(dd, J = 6.2, 1.4 Hz, 1H), 2.68 (dt, J = 7.2, 6.8, 2.0 Hz, 2H), 3.04 (m, 1H),
3.44 (dd, J = 14.4, 7.0 Hz, 1H), 3.53 (dd, J = 14.4, 6.4 Hz, 1H), 4.43 (t, J =
6.8 Hz, 1H), 4.87 (s, 10H), 7.19 (d, J=9.2 Hz, 2H), 7.22 (d, J=9.2 Hz, 2H),
7.32 (d, J = 9.2 Hz, 2H). HRMS-ESI (m/z): [M + H]+, calcd for
C20H23N2O6S2, 451.0992; found, 451.1004.
Spectral Data of Arg-Arg Ester Prodrug Compound 11e. 1H NMR
(500 MHz, CDCl3) δ 1.39ꢀ1.53 (m, 45H), 1.54ꢀ2.11 (m, 8H), 2.16
(dd, J = 5.2, 1.4 Hz, 1H), 2.54 (d, J = 5.2 Hz, 1H), 3.06 (m, 1H), 3.19 (dd,
J = 14.2, 7.8 Hz, 1H), 3.47 (dd, J = 12.8, 7.0 Hz, 2H), 3.52 (dd, J = 14.4,
5.6 Hz, 1H), 3.74ꢀ3.95 (m, 2H), 4.32 (dd, J = 14.8, 6.4 Hz, 1H), 4.77
(m, 1H), 5.92 (d, J = 7.8 Hz, 1H), 7.07ꢀ7.12 (m, 4H), 7.16 (d, J = 9.0
Hz, 2H), 7.33 (d, J = 7.6 Hz, 1H), 7.87 (d, J = 8.8 Hz, 2H), 8.38 (t, J = 4.9
Hz, 1H), 9.30 (br. s, 2H), 11.50 (s, 1H); 13C NMR (126 MHz, CDCl3)
δ 24.4, 24.9, 25.6, 26.2, 28.2, 28.2, 28.5, 28.6, 29.3, 40.4, 44.1, 52.5, 54.3,
62.8, 79.5, 80.2, 83.4, 83.5, 84.2, 117.9, 121.5, 121.5, 123.4, 130.8, 131.0,
132.3, 147.5, 152.5, 153.4, 155.0, 156.4, 160.8, 160.9, 162.9, 163.7, 170.6,
172.7. HRMS-ESI (m/z): [M+ H]+, calcd for C52H79N8O16S2, 1135.5050;
1
Spectral Data of Arg Amide Prodrug Compound 12d. H NMR
(500 MHz, CDCl3) δ 1.38 (s, 9H), 1.48 (s, 9H), 1.53 (s, 9H), 1.59ꢀ1.96
(m, 4H), 2.16 (dd, J = 5.2, 1.8 Hz, 1H), 2.54 (dd, J = 6.2, 1.6 Hz, 1H),
3.06 (m, 1H), 3.17 (dd, J = 14.2, 8.0 Hz, 1H), 3.53 (dd, J = 14.2, 5.6 Hz,
1H), 3.72 (m, 1H), 4.12 (s, 1H), 4.55 (m, 1H), 5.99 (d, J = 8.2 Hz, 1H),
7.05 (d, J = 9.0 Hz, 2H), 7.07 (d, J = 8.8 Hz, 2H), 7.53 (d, J = 8.6 Hz, 2H),
7.85 (d, J = 9.0 Hz, 2H), 9.13 (br. s, 1H); 13C NMR (126 MHz, CDCl3)
δ 24.5, 24.7, 26.3, 28.2, 28.6, 29.3, 44.1, 54.0, 62.8, 84.6, 117.6, 121.2,
123.4, 130.9, 132.0, 135.1, 151.3, 155.0, 161.3, 163.2, 163.4, 171.1.
HRMS-ESI (m/z): [M + H]+, calcd for C36H52N5O10S2, 778.3150;
found, 778.3154. Compound 5d: 1H NMR (500 MHz, 10% CDCl3 in
1
found, 1135.5039. Compound 4e: H NMR (500 MHz, 10% CDCl3 in
6686
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
CD3OD) δ 1.71ꢀ2.11 (m, 4H), 2.16 (d, J = 5.2 Hz, 1H), 2.54 (dd, J =
6.2, 1.2 Hz, 1H), 3.06 (quin, J = 6.2 Hz, 1H), 3.29 (m, 1H), 3.45 (m, 2H),
3.54 (dd, J = 14.6, 6.4 Hz, 1H), 4.14 (m, 1H), 7.15 (d, J = 9.0 Hz, 2H),
7.15 (d, J = 9.0 Hz, 2H), 7.77 (d, J = 9.0 Hz, 2H), 7.92 (d, J = 8.8 Hz, 2H).
HRMS-ESI (m/z): [M + H]+, calcd for C21H28N5O4S2, 478.1577;
found, 478.1580.
EtOAc. The combined organic portions were dried over anhydrous
Na2SO4. The solvent was removed under reduced pressure to provide
the amine as a solid (1.9 g, 5.5 mmol, 83%). This product was dissolved
in MeOH (30 mL), treated with TEA (0.6 g, 5.9 mmol) and Boc2O
(1.9 g, 8.4 mmol, 97%), and stirred at room temperature for 24 h. The
solvent was removed under reduced pressure, and the product was
purified by silica gel chromatography (hexanes/EtOAc = 3/1) to give
the title compound (1.9 g, 76%). 1H NMR (500 MHz, CDCl3) δ 1.54
(s, 9H), 3.02 (br. s., 3H), 3.08 (br. s., 3H), 3.48 (s, 3H), 5.18 (s, 2H),
6.71 (dd, J = 2.6, 8.8 Hz, 1H), 6.90 (d, J = 2.6 Hz, 1H), 6.92ꢀ6.97 (m,
2H), 7.00 (br. s., 1H), 7.36ꢀ7.42 (m, 2H), 8.05 (br. s., 1 H); 13C NMR
(126 MHz, CDCl3) δ 28.4, 36.9, 56.5, 80.5, 95.3, 107.2, 113.6, 117.7,
119.3, 121.6, 125.3, 137.4, 146.5, 150.6, 152.9, 159.4, 167.4. HRMS-FAB
(m/z): [M + H]+, calcd for C22H29N2O6S, 449.1741; found, 449.1735.
t-Butyl (2-(Methoxymethoxy)-4-(4-((oxiran-2-ylmethyl)thio)
phenoxy)phenyl)carbamate (19). To a solution of compound 18
(0.65 g, 1.5 mmol) in anhydrous MeOH (10 mL) was added potassium
hydroxide (0.42 g, 7.5 mmol). The mixture was refluxed for 4 h and then
cooled to room temperature. Epichlorohydrin (0.21 g, 2.3 mmol) was
added dropwise, and the mixture was stirred at room temperature for
15 min, after which the solvent was removed under reduced pressure. The
concentrate was diluted with water and extracted with EtOAc. The
combined organic extracts were dried over anhydrous Na2SO4, and the
suspension was filtered, followed by concentration of the filtrate in vacuo.
Purification by silica gel chromatography (hexanes/EtOAc = 4/1) gave 19
in 68% yield (0.43 g). 1H NMR (500 MHz, CDCl3) δ 1.53 (s, 9H), 2.47
(dd, J = 2.6, 5.0 Hz, 1H), 2.77 (t, J = 4.4 Hz, 1H), 2.85 (dd, J = 6.1, 13.9 Hz,
1H), 3.05ꢀ3.12 (m, 1H), 3.12ꢀ3.18 (m, 1H), 5.18 (s, 2H), 3.48 (s, 3H),
6.66 (dd, J = 2.6, 8.8 Hz, 1H), 6.85ꢀ6.92 (m, 3H), 6.98 (br. s., 1H),
7.37ꢀ7.44 (m, 2 H), 8.03 (br. s., 1H); 13C NMR (126 MHz, CDCl3) δ
28.5, 38.2, 47.6, 51.3, 56.5, 80.6, 95.4, 106.8, 113.1, 118.4, 119.4, 125.1,
128.0, 131.8, 133.8, 146.6, 151.2, 152.9, 157.9. HRMS-FAB (m/z):
[M + H]+, calcd for C22H28NO4S, 434.1632; found, 434.1637.
t-Butyl (2-(Methoxymethoxy)-4-(4-((oxiran-2-ylmethyl)sul-
fonyl)phenoxy)phenyl)carbamate (20). Compound 19 (0.44 g,
1.0 mmol) was dissolved in CH2Cl2 (10 mL) and cooled in an iceꢀwater
bath. It was treated with m-CPBA (0.48 g, 2.1 mmol, 77%), and the resulting
white suspension was stirred at room temperature for 10 min. The mixture
was filtered, and the filtrate was washed with saturated Na2S2O3, followed
by saturated NaHCO3. The organic layer was separated, and the aqueous
layer was washed with CH2Cl2. The combined organic solution was dried
over anhydrous Na2SO4, the suspension was filtered, and the filtrate was
concentrated in vacuo. Purification of the product by silica gel chromatog-
raphy (hexanes/EtOAc = 2/1) gave the title compound in 89% yield
(0.42 g). 1H NMR (500 MHz, CDCl3) δ 1.54 (s, 9H), 2.48 (dd, J = 2.2,
4.8 Hz, 1H), 2.82 (t, J = 4.5 Hz, 1H), 3.19ꢀ3.28 (m, 1H), 3.28ꢀ3.38 (m,
2H), 3.49 (s, 3H), 5.20 (s, 2H), 6.73 (dd, J = 2.6, 8.8 Hz, 1H), 6.91 (d, J =
2.6 Hz, 1H), 7.03 (s, 1H), 7.04ꢀ7.09 (m, 2H), 7.83ꢀ7.89 (m, 2H), 8.13
(br. s., 1 H); 13C NMR (126 MHz, CDCl3) δ 28.5, 46.0, 46.1, 56.6, 59.8,
80.9, 95.4, 107.5, 114.2, 117.2, 119.5, 126.3, 130.6, 132.3, 146.7, 149.4,
152.9, 163.5. HRMS-FAB (m/z): [M + H]+, calcd for C22H28NO8S,
466.1530; found, 466.1543.
Spectral Data of Arg-Arg Amide Prodrug Compound 12e. 1H NMR
(500 MHz, CDCl3) δ 1.40ꢀ1.54 (5 ꢁ s, 45H), 1.46ꢀ1.90 (m, 8H), 2.16
(dd, J = 5.2, 1.8 Hz, 1H), 2.53 (dd, J = 6.2, 1.4 Hz, 1H), 3.05 (m, 1H),
3.16 (dd, J = 14.2, 7.8 Hz, 1H), 3.39 (m, 1H), 3.53 (dd, J = 14.2, 5.6 Hz,
1H), 3.55 (m, 2H), 3.82 (t, J = 6.8 Hz, 2H), 4.28 (dd, J = 13.8, 6.4 Hz,
1H), 4.59 (q, J = 7.4 Hz, 1H), 6.10 (d, J = 7.4 Hz, 1H), 7.03 (d, J = 9.0 Hz,
2H), 7.05 (d, J = 8.8 Hz, 2H), 7.22 (d, J = 6.4 Hz, 1H), 7.61 (d, J = 8.4 Hz,
2H), 7.84 (d, J = 9.0 Hz, 2H), 8.41 (t, J = 5.6 Hz, 1H), 8.84 (s, 1H), 9.26
(br. s, 1H), 9.35 (br. s, 1H), 11.47 (s, 1H); 13C NMR (126 MHz,
CDCl3) δ 22.8, 24.5, 25.4, 26.1, 26.2, 28.2, 28.2, 28.4, 28.4, 28.6, 29.2,
31.8, 34.8, 40.1, 44.1, 53.8, 55.0, 62.8, 79.6, 80.6, 83.4, 84.3, 117.5, 121.2,
122.5, 130.9, 131.8, 135.4, 150.9, 153.4, 154.9, 156.6, 160.8, 163.4, 163.4,
163.5, 169.6. HRMS-ESI (m/z): [M + H]+, calcd for C52H80N9O15S2,
1134.5210; found, 1134.5215. Compound 5e: 1H NMR (500 MHz, 10%
CDCl3 in CD3OD) δ 1.68ꢀ2.05 (m, 8H), 2.13 (dd, J = 5.1, 1.3 Hz, 1H),
2.51 (dd, J = 6.2, 1.0 Hz, 1H), 3.17 (m, 1H), 3.22ꢀ3.28 (m, 4H), 3.42
(dd, J = 14.6, 7.0 Hz, 1H), 3.50 (dd, J = 14.0, 7.0 Hz, 1H), 4.10 (t, J =
6.3 Hz, 1H), 4.52 (dd, J = 8.7, 5.1 Hz, 1H), 7.10 (t, J = 8.9 Hz, 4H), 7.66
(d, J = 9.0 Hz, 2H), 7.89 (d, J = 9.0 Hz, 2H). HRMS-ESI (m/z):
[M + H]+, calcd for C27H40N9O5S2, 634.2588; found, 634.2585.
4-Fluoro-2-(methoxymethoxy)-1-nitrobenzene (15). 5-Fluoro-
2-nitrophenol (10 g, 63 mmol, 99%) was dissolved in anhydrous DMF
(100 mL) in an oven-dried flask. The solution was stirred under an
atmosphere of nitrogen and cooled in an iceꢀwater bath. Sodium hydride
(3.0 g, 75mmol) was added with stirring, after whichchloromethyl methyl
ether (5.3 g, 66 mmol) was added dropwise. The resulting mixture was
aged at room temperature for 1 h. The reaction was quenched with
MeOH, diluted with ether, and washed with water and brine. The ether
layer was dried over anhydrous Na2SO4 and concentrated in vacuo.
Purification of the product by silica gel chromatography (hexanes/EtOAc =
1/7) gave the title compound in 88% yield. 1H NMR (500 MHz, CDCl3)
δ 3.53 (s, 3H), 5.30 (d, J = 2.0 Hz, 2H), 6.78 (dd, J = 7.3, 9.1 Hz, 1H), 7.06
(dt, J = 2.3, 10.4 Hz, 1H), 7.91 (dd, J = 6.0, 9.0 Hz, 1 H); 13C NMR
(126 MHz, CDCl3) δ 57.0, 95.6, 105.0 (d, J = 26.3 Hz), 108.8 (d,
J = 23.0 Hz), 127.8 (d, J = 10.7 Hz), 152.8 (d, J = 11.5 Hz), 165.5 (d, J =
255.9 Hz). HRMS-FAB (m/z): [M + Na]+, calcd for C8H8FNNaO4,
224.0330; found, 224.0328.
S-(4-(3-(Methoxymethoxy)-4-nitrophenoxy)phenyl) Di-
methylcarbamothioate (17). To a round-bottom flask was added
dimethylcarbamothioate 16 (2.1 g, 11 mmol), compound 15 (2.5 g,
12 mmol), DMF (50 mL), and Cs2CO3 (7.1 g, 22 mmol). The resulting
mixture was stirred for 24 h, followed by filtration through a layer of silica
gel. The solvent was evaporated, and the product was purified by silica
gel chromatography (hexanes/EtOAc = 2/1) to give 17 as a solid in 88%
(3.5 g). 1H NMR (500 MHz, CDCl3) δ 3.05 (br. s., 3H), 3.12 (br. s.,
3H), 3.53 (s, 3H), 5.28 (s, 2H), 6.61 (ddd, J = 0.8, 2.5, 9.1 Hz, 1H), 6.99 (d,
J = 2.4 Hz, 1H), 7.02ꢀ7.12 (m, 2H), 7.50ꢀ7.56 (m, 2H), 7.88ꢀ7.93 (m,
1 H); 13C NMR (126 MHz, CDCl3) δ 37.1, 37.2, 57.1, 95.5, 107.1, 110.4,
120.5, 125.1, 127.9, 135.7, 137.9, 153.0, 156.1, 162.0, 166.9. HRMS-FAB
(m/z): [M + H]+, calcd for C17H19N2O6S, 379.0958; found, 379.0945.
t-Butyl (4-(4-((Dimethylcarbamoyl)thio)phenoxy)-2-
(methoxymethoxy)phenyl)carbamate (18). A solution of 17
(2.5 g, 6.6 mmol) in MeOH/H2O (30 mL/15 mL) was treated with
elemental iron (1.9 g, 34 mmol) and NH4Cl (0.35 g, 6.5 mmol), and the
resulting mixture was heated at reflux for 2 h. The crude reaction mixture
was filtered through Celite, and the filtrate was concentrated in vacuo.
The residue was dissolved in EtOAc and was washed with water. The
organic layer was separated, and the aqueous layer was washed with
t-Butyl (2-(Methoxymethoxy)-4-(4-((thiiran-2-ylmethyl)sul-
fonyl)phenoxy)phenyl)carbamate (21). To a solution of 20
(0.40 g, 0.86 mmol) in CH2Cl2 (5 mL) was added a mixture of thiourea
(0.10 g, 1.3 mmol, 99%) in methanol (5 mL). The resulting mixture was
stirred for 24 h at room temperature, after which the solvent was
removed under reduced pressure. The residue was partitioned between
CH2Cl2 and water. The organic layer was dried over anhydrous Na2SO4
and was filtered. Evaporation of solvent and purification by silica gel
chromatography (hexanes/EtOAc = 4/1) gave 0.34 g of 21 (83%). 1H
NMR (500 MHz, CDCl3) δ 1.53 (s, 9H), 2.14 (dd, J = 1.8, 5.0 Hz, 1H),
2.52 (dd, J = 1.5, 5.9 Hz, 1H), 3.04 (dq, J = 5.6, 7.8 Hz, 1H), 3.15 (dd, J =
8.0, 14.2 Hz, 1H), 3.47 (s, 3H), 3.51 (dd, J = 5.6, 14.4 Hz, 1H), 5.19
6687
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
(s, 2H), 6.72 (dd, J = 2.6, 8.8 Hz, 1H), 6.91 (d, J = 2.6 Hz, 1H), 7.01ꢀ
7.08 (m, 3H), 7.80ꢀ7.85 (m, 2H), 8.08ꢀ8.16 (m, 1 H); 13C NMR
(126 MHz, CDCl3) δ 24.4, 26.2, 28.5, 56.6, 62.7, 80.8, 95.3, 107.3, 114.1,
117.3, 119.4, 126.2, 130.8, 131.6, 146.6, 149.3, 152.8, 163.5. HRMS-FAB
(m/z): [M + H]+, calcd for C22H28NO7S2, 482.1302; found, 482.1301.
2-Amino-5-(4-((thiiran-2-ylmethyl)sulfonyl)phenoxy)phenol
(13). To a solution of compound 21 (0.10 g, 0.21 mmol) in methanol
(5 mL) was added a few drops of concd HCl. After the mixture was stirred at
reflux for 1 h, the solvent was evaporated in vacuo. The crude product was
taken up into water and EtOAc. Layers were separated, and the aqueous layer
was washed with EtOAc. The combined organic layers were dried over
anhydrous Na2SO4, the suspension was filtered, and the filtrate was
concentrated in vacuo. Purification by silica gel chromatography (hexanes/
EtOAc = 1/1) gave 13 in 95% yield (66 mg). 1H NMR (500 MHz, CDCl3)
δ2.14 (dd, J=1.3,4.9Hz,1H),2.18(d,J= 1.0 Hz, 1H), 2.49ꢀ2.56 (m, 1H),
3.00ꢀ3.08 (m, 1H), 3.17 (dd, J = 7.8, 14.4 Hz, 1H), 3.51 (dd, J = 5.6, 14.2
Hz, 2H), 6.46ꢀ6.54 (m, 2H), 6.73ꢀ6.80 (m, 1H), 6.97ꢀ7.07 (m, 2H), 7.80
(m, 2 H); 13C NMR (126 MHz, CDCl3) δ 24.4, 26.3, 62.8, 108.5, 113.1,
117.2, 117.9, 130.7, 131.1, 131.9, 145.9, 147.3, 164.1. HRMS-FAB (m/z):
[M + H]+, calcd for C15H16NO4S2, 338.0515; found, 338.0507.
N-(4-(4-((Thiiran-2-ylmethyl)sulfonyl)phenoxy)phenyl)
hydroxylamine (14). Compound 22 (75 mg, 0.21 mmol), prepared
as described previously,28 in CH2Cl2/H2O (0.2 mL/0.1 mL) was treated
with Zn (50 mg, 0.76 mmol) and NH4Cl (25 mg, 0.47 mmol). The
resulting mixture was stirred for 30 min at room temperature. It was then
filtered over Celite, and the filtrate was concentrated in vacuo. Purifica-
tion by silica gel chromatography (CH2Cl2) gave 14 in 69% yield
(50 mg). 1H NMR (500 MHz, CDCl3) δ 2.16 (dd, J = 1.8, 5.2 Hz, 1H),
2.53 (dd, J = 1.0, 6.2 Hz, 1H), 3.01ꢀ3.09 (m, 1H), 3.17 (dd, J = 7.9, 14.3
Hz, 1H), 3.53 (dd, J = 5.8, 14.2 Hz,1H), 6.98ꢀ7.10 (m, 6H), 7.81ꢀ7.88
(m, 2 H); 13C NMR (126 MHz, CDCl3) δ 24.5, 26.3, 62.8, 116.5, 117.3,
121.6, 130.9, 131.5, 147.4, 149.4, 163.8. HRMS-FAB (m/z): [M + H]+,
calcd for C15H16NO4S2, 338.0515; found, 338.0501.
Solubility Determination and Solution Stability. The water
solubilities of the prodrugs were measured at room temperature by using
a 10 mg/mL solution of each compound in water. The molar extinction
coefficient was calculated at 245 nm. The aqueous solutions were
centrifuged, and the absorbance at 245 nm of the supernatant was
recorded. The concentration was calculated using Beer’s law. Assays
were performed in triplicate. The ester and amide prodrugs had
solubilities of over 10 mg/mL. Over a period of one month, appro-
priately diluted aliquots were analyzed by HPLC for the presence of the
prodrug and the active gelatinase inhibitors (compound 2 from the
esters or compound 3 from the amides).
Ex Vivo Hydrolyses of the Amide Prodrugs. Fresh blood was
collected in heparinized syringes from human volunteers. Blood collec-
tion from human volunteers was approved by the University of Notre
Dame Institutional Review Board. The blood was centrifuged to obtain
plasma. The stability of the prodrugs in plasma and whole blood (1 mL)
was determined by spiking the plasma or blood with 10 μL of 10 mM of
the prodrug and incubating at 37 °C. Aliquots were drawn periodically
and processed for analysis. For the plasma samples, the reaction was
immediately terminated by the addition of two volumes of acetonitrile.
For the blood samples, aliquots were first centrifuged at 4 °C, and the
resultant plasma was treated with two volumes of acetonitrile. The
precipitated protein was centrifuged (10 min, 10 500g), and the super-
natant was analyzed by reversed-phase HPLC with UV detection. First-
order rate constants for the disappearance of the prodrugs were
calculated from which the half-lives were determined.
MOCAcPLGLA2pr(Dnp)AR-NH2 was purchased from Peptides Inter-
national (Louisville, KY, USA); MOCAcRPKPVE(Nva)WRK(Dnp)-
NH2 and (Dnp)P(Cha)GC(Me)HAK(NMa)NH2 were obtained from
R&D Systems (Minneapolis, MN, USA). The Km values used for the
reaction of MMP-2, MMP-9, and MMP-14 with the fluorogenic sub-
strate MOCAcPLGLA2pr(Dnp)AR-NH2 were 18.9 ( 1.0, 5.0 ( 0.1,
and 5.6 ( 0.4 μM, respectively. Inhibitor stock solutions (10 mM) were
prepared in water (except for compound 13, which was prepared in
DMSO). The methodology for enzyme inhibition and assays was the
same as that reported previously.28 Substrate hydrolysis was measured
with a Cary Eclipse fluorescence spectrophotometer (Varian, Walnut
Creek, CA, USA).
Microsomal Incubations. Mouse, rat, and human liver micro-
somes and rat S9 were purchased from BD Biosciences (Woburn, MA,
USA). Incubations consisted of liver microsomes (1.0 mg), NADPH
(0.5 mM, final concentration), and 100 μM compound (final con-
centration) in potassium phosphate buffer (100 mM, pH 7.4) at 37 °C in
a total volume of 1 mL. Aliquots were drawn at different time points,
and the reactions were terminated by the addition of one volume of
acetonitrile containing 25 μM (S)-2-((S)-1-((4-phenoxyphenyl)sulfonyl)-
ethyl)thiirane, prepared as published,22 as an internal standard. The pre-
cipitated protein was centrifuged (10 min, 10 500g), and the supernatant
was analyzed by reversed-phase HPLC with UV detection.
Apparent Intrinsic Clearance (CLint,app) Determination in
Microsomes. Apparent intrinsic clearance was calculated by the
following equation:39
CLint , app ¼ ð0:693=in vitro t1=2
Þ
ꢁðincubation volume=mg of microsomal proteinÞ
ꢁð45 mg of microsomal protein=gram of liverÞ
ꢁð20 g of liver=kg body weightÞ
In the previous equation, the weight of the liver can be 20, 45, or 90 g of
liver/kg of body weight for human, rat, and mouse, respectively.
High-Performance Liquid Chromatography. The HPLC
system consisted of a Perkin-Elmer series 200 quaternary pump
(Perkin-Elmer Corp., Norwalk, CT, USA), a Perkin-Elmer UV detector
series 200, and a Perkin-Elmer autosampler series 200 with a thermo-
statted tray. Samples were analyzed on a YMC 5 μm basic column (4.6
mm i.d. ꢁ 150 mm, YMC Inc., Wilmington, NC, USA) connected to a
YMC 5 μm basic guard cartridge (4.0 mm i.d. ꢁ 20 mm). The mobile
phase consisted of elution at 1 mL/min with 75% A/25% B for 3 min,
followed by a 17-min linear gradient to 10% A/90% B, then 10% A/90%
B for 5 min (A = 100 mM NH4OAc at pH 4; B = acetonitrile). Effluent
was monitored by UV detection at 245 nm.
Liquid ChromatographyꢀMass Spectrometry. The LC sys-
tem consisted of a Dionex Ultimate 3000 RSLC (Dionex, Sunnyvale,
CA, USA) equipped with a Dionex Ultimate 3000 autosampler and a
Dionex Ultimate 3000 diode array detector. All mass spectrometric
experiments were performed with a Bruker MicroTOF-Q II (Billerica,
MA, USA) monitored with Hystar 3.2 software. Mass spectrometry
acquisition was performed using electrospray ionization with the
following parameters: end plate offset voltage = ꢀ0.5 kV, capillary
voltage = 4.5 kV, and nitrogen as both a nebulizer and dry gas (8.0 L/min
flow rate, 200 °C dry temperature, 4.0 bar). Samples were analyzed on an
Acclaim RSLC 120 C18 column (2.2 μm, 2.1 mm i.d. ꢁ 100 mm,
Dionex, Sunnyvale, CA, USA). The mobile phase consisted of elution at
0.5 mL/min with 85% A/15% B for 3 min, followed by a 9-min linear
gradient to 10% A/90% B, and then 10% A/90% B for 2 min (A = 0.1%
formic acid in water; B = 0.1% formic acid in acetonitrile).
Enzyme Inhibition Studies. Human recombinant active MMP-2
and MMP-7 and catalytic domains of MMP-3 and MMP-14 were
purchased from EMD Biosciences (La Jolla, CA, USA). The catalytic
domains of human recombinant MMP-1 and MMP-9 were from Biomol
International (Plymouth Meeting, PA, USA). The fluorogenic substrate
AMES II Mutagenecity Assay. The Ames II mutagenecity assay
was carried out by BioReliance (Rockville, MD, USA). Briefly, the strains
included the Ames II mixed strains (TA7001 to TA7006) and tester
strain TA98, as described.34 The strains were mixed with the test article
6688
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
(5d or 3) and appropriate metabolic activation mixture. The mixture was
incubated at 37 °C for 90 min. The mixture was then mixed with an
indicator medium, and the contents were plated in 384-well plates. After
48 h of incubation, the plates were scored for positive wells. The test
article was tested at a maximum of six dose levels (3.3, 10.0, 33.3, 100.0,
333.3, and 1000 μg/mL) along with appropriate negative and positive
controls with mixed Ames II strains and tester strain TA98, with and
without S9 activation. All dose levels of the test article, negative controls,
and positive controls were tested in triplicate. 2-Aminoanthracene
(5.0 μg/mL) and 4-nitroquinoline N-oxide (1.0 μg/mL) were used as
positive controls with S9 activation. 2-Nitrofluorene (2.0 μg/mL) was
used as a positive control without S9 activation.
Animals. Male C57Bl/6J mice (The Jackson Laboratory, Bar
Harbor, ME, USA), 6ꢀ8 weeks old, 22ꢀ25 g body weight, specific
pathogen free, were provided with Laboratory 5001 Rodent Diet (PMI,
Richmond, IN, USA) and water ad libitum. Animals were maintained in
polycarbonate shoebox cages with hardwood bedding in a room under a
12:12 h light/dark cycle and at 72 ( 2 °F. Animal studies were approved
by the University of Notre Dame Institutional Animal Care and Use
Committee.
Animal Dosing and Sample Collection. Compounds 5d and 3,
as HCl salts, were dissolved in saline to a final concentration of 2.5
mg/mL. The solution was sterilized by passage through an Acrodisc syringe
filter (Pall Life Sciences, 0.2 μm, 13 mm diameter, PTFE membrane).
Mice (n = 3 per time point per compound) received a single 120 μL dose
of 5d or 3 intravenously (equivalent to 12.5 mg/kg) by tail vein injection.
Mice used for later time points (1 h, 2 h, and 4 h) were housed in
Nalgene rat metabolic cages equipped with flooring grates and feeders
for mice (Tecniplast, West Chester, PA, USA), for easy collection of
urine. Terminal blood samples were collected at various time points by
cardiac puncture following CO2 asphixation, using heparin as an
anticoagulant. Blood was stored on ice and centrifuged to obtain plasma.
Urine samples were collected, pooled by time point (1 h, 2 h, and 4 h),
and stored on ice.
Sample Analysis. A 200-μL aliquot of plasma was mixed with
400 μL of 1 μM (S)-2-((S)-1-((4-phenoxyphenyl)sulfonyl)ethyl)thiirane
as the internal standard in acetonitrile. The sample was centrifuged at
10 000g for 20 min. The supernatant was concentrated, reconstituted to
120 μL with water, and analyzed by reversed-phase LC with MS detection at
m/z 478.1577 (for 5d), m/z 322.0566 (for 3), and m/z 364.0672 (for the
p-N-acetyl derivative, M3). Standard curves of 5d, 3, and the p-N-acetyl
derivative in blank mouse plasma were prepared by fortification of 200 μLof
plasma with each compound at concentrations of 0, 0.005, 0.01, 0.025, 0.05,
0.1, 0.25, 0.5, 1, 2.5, 5, 10, 25, and 50 μM. Quantification of 5d, 3, and the
p-N-acetyl derivative (M3) was performed using peak area ratios relative to
the internal standard and linear regression parameters calculated from the
calibration curve standards prepared in blank mouse plasma. The assay was
linear from 0.005 to 50 μM, with coefficients of determination, R2, of 0.998
for 5d, 0.999 for 3, and 0.999 for the p-N-acetyl derivative.
concentrationꢀtime data by linear regression, where the slope of the
line was the rate constant k, and t1/2 = ln 2/k. Volume of distribution
(Vd) was calculated as the dose divided by the initial concentration (C0).
Clearance (CL) was calculated from the dose divided by AUC0ꢀ∞
.
Concentrations of 3 and the p-N-acetyl derivative (M3) after intrave-
nous administration of 3 at 12.5 mg/kg were normalized to 8.13 mg/kg
(equivalent to 12.5 mg/kg intravenous administration of 5d).
’ AUTHOR INFORMATION
Corresponding Author
*Tel: +1-574-631-2965. Fax: +1-574-631-6652. E-mail: mchang@
nd.edu.
Author Contributions
§These authors contributed equally to this work.
’ ACKNOWLEDGMENT
This work was supported by grant CA122417 from the
National Institutes of Health (to M.C. and S.M.) and a grant
from the NFL Charities (to M.C.). M.G. is a Fellow of the
Chemistry-Biochemistry-Biology Interface (CBBI) Program at
the University of Notre Dame, supported by the training grant
T32GM075762. M.L. is an Innovation Fellow supported by the
Edison Foundation. The Mass Spectrometry & Proteomics
Facility of the University of Notre Dame is supported by grant
CHE-0741793 from the National Science Foundation.
’ ABBREVIATIONS USED
amu, atomic mass unit; AUC, area under the concentrationꢀ
time curve; Boc, tert-butoxycarbonyl; CL, clearance; DMAP, 4-(di-
methylamino)pyridine; DMSO, dimethyl sulfoxide; Dnp, 2,4-dini-
trophenyl; HPLC, high-performance liquid chromatography; M1-
(ꢀ5), metabolite 1(ꢀ5); m-CBPA, meta-chloroperbenzoic
acid; MMP, matrix metalloproteinase; MOCAc, (7-methoxy-
coumarin-4-yl)acetyl; MOM, methoxymethyl; MS, mass spectro-
metry; NADPH, nicotinamide adenine dinucleotide phosphate
reduced; NHS, N-hydroxysuccinimide; Nva, L-norvalyl; Vd, volume
of distribution
’ REFERENCES
(1) Moses, M. A. The regulation of neovascularization by matrix
metalloproteinases and their inhibitors. Stem Cells 1997, 15, 180–189.
(2) Nguyen, M.; Arkell, J.; Jackson, C. J. Human endothelial
gelatinases and angiogenesis. Int. J. Biochem. Cell Biol. 2001, 33, 960–970.
(3) Stamenkovic, I. Extracellular matrix remodelling: the role of
matrix metalloproteinases. J. Pathol. 2003, 200, 448–464.
(4) Sorokin, L. The impact of the extracellular matrix on inflamma-
tion. Nat. Rev. Immunol. 10, 712-23.
(5) Agrawal, S.; Anderson, P.; Durbeej, M.; van Rooijen, N.; Ivars, F.;
Opdenakker, G.; Sorokin, L. M. Dystroglycan is selectively cleaved at the
parenchymal basement membrane at sites of leukocyte extravasation in
experimental autoimmune encephalomyelitis. J. Exp. Med. 2006, 203,
1007–19.
(6) Kessenbrock, K.; Plaks, V.; Werb, Z. Matrix metalloproteinases:
regulators of the tumor microenvironment. Cell 2010, 141, 52–67.
(7) Massova, I.; Kotra, L. P.; Fridman, R.; Mobashery, S. Matrix
metalloproteinases: structures, evolution, and diversification. FASEB
J. 1998, 12, 1075–1095.
(8) Montaner, J.; Alvarez-Sabin, J.; Molina, C. A.; Angles, A.; Abilleira,
S.; Arenillas, J.; Monasterio, J. Matrix metalloproteinase expression is related
to hemorrhagic transformation after cardioembolic stroke. Stroke 2001,
32, 2762–2767.
Urine samples were centrifuged at 10 000g for 10 min, and a 50-μL
aliquot of each supernatant was analyzed by LC/MS. In addition to
analyzing for 5d, 3, and p-N-acetyl M3, the plasma and urine samples
were screened specifically for N-hydroxylamine 14 with MS detection at
m/z 338.0515.
Pharmacokinetic Parameters. Pharmacokinetic parameters
were determined by noncompartmental analyses of the concentra-
tionꢀtime data for 5 and 3. Area under the concentrationꢀtime curve
(AUC) was determined by the linear trapezoidal rule. The concentration
at time = 0 (C0) was estimated from the first sampling times by back-
extrapolation using logꢀlinear regression analysis. AUC0ꢀ∞ was calcu-
lated as AUC0ꢀt + (Clast/k), where AUC0ꢀt is the area under the mean
plasma concentrationꢀtime curve up to the last quantifiable sample, and
Clast is the concentration at the last quantifiable sampling time. Half-lives
(t1/2) were estimated from the terminal linear portion of the
6689
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690

Journal of Medicinal Chemistry
ARTICLE
(9) Fisher, J. F.; Mobashery, S. Recent advances in MMP inhibitor
design. Cancer Metastasis Rev. 2006, 25, 115–136.
(10) Hu, J.; Van den Steen, P. E.; Sang, Q. X.; Opdenakker, G. Matrix
metalloproteinase inhibitors as therapy for inflammatory and vascular
diseases. Nat. Rev. Drug Discovery 2007, 6, 480–98.
(27) Pochopin, N. L.; Charman, W. N.; Stella, V. J. Amino-acid
derivatives of dapsone as water-soluble prodrugs. Int. J. Pharm. 1995,
121, 157–167.
(28) Ikejiri, M.; Bernardo, M. M.; Bonfil, R. D.; Toth, M.; Chang,
M. L.; Fridman, R.; Mobashery, S. Potent mechanism-based inhibitors
for matrix metalloproteinases. J. Biol. Chem. 2005, 280, 33992–34002.
(29) Lee, M.; Bernardo, M. M.; Meroueh, S. O.; Brown, S.; Fridman,
R.; Mobashery, S. Synthesis of chiral 2-(4-phenoxyphenylsulfonyl-
methyl)thiiranes as selective gelatinase inhibitors. Org. Lett. 2005,
7, 4463–4465.
(30) Lim, I. T.; Brown, S.; Mobashery, S. A convenient synthesis of a
selective gelatinase inhibitor as an antimetastatic agent. J. Org. Chem.
2004, 69, 3572–3573.
(31) Davies, B.; Morris, T. Physiological-parameters in laboratory-
animals and humans. Pharm. Res. 1993, 10, 1093–1095.
(32) Tanaka, K.; Yamamoto, Y.; Kuzuya, A.; Komiyama, M. Synth-
esis of photo-responsive acridine-modified DNA and its application to
site-selective RNA scission. Nucleosides, Nucleotides Nucleic Acids 2008,
27, 1175–1185.
(11) Brown, S.; Bernardo, M. M.; Li, Z. H.; Kotra, L. P.; Tanaka, Y.;
Fridman, R.; Mobashery, S. Potent and selective mechanism-based
inhibition of gelatinases. J. Am. Chem. Soc. 2000, 122, 6799–6800.
(12) Forbes, C.; Shi, Q. C.; Fisher, J. F.; Lee, M.; Hesek, D.; Llarrull,
L. I.; Toth, M.; Gossing, M.; Fridman, R.; Mobashery, S. Active site ring-
opening of a thiirane moiety and picomolar inhibition of gelatinases.
Chem. Biol. Drug Des. 2009, 74, 527–534.
(13) Bonfil, R. D.; Sabbota, A.; Nabha, S.; Bernardo, M. M.; Dong,
Z.; Meng, H.; Yamamoto, H.; Chinni, S. R.; Lim, I. T.; Chang, M.; Filetti,
L. C.; Mobashery, S.; Cher, M. L.; Fridman, R. Inhibition of human
prostate cancer growth, osteolysis and angiogenesis in a bone metastasis
model by a novel mechanism-based selective gelatinase inhibitor. Int.
J. Cancer 2006, 118, 2721–2726.
(14) Martin, M. D.; Carter, K. J.; Jean-Philippe, S. R.; Chang, M.;
Mobashery, S.; Thiolloy, S.; Lynch, C. C.; Matrisian, L. M.; Fingleton, B.
Effect of ablation or inhibition of stromal matrix metalloproteinase-9 on
lung metastasis in a breast cancer model is dependent on genetic
background. Cancer Res. 2008, 68, 6251–6259.
(15) Kruger, A.; Arlt, M. J. E.; Gerg, M.; Kopitz, C.; Bernardo, M. M.;
Chang, M.; Mobashery, S.; Fridman, R. Antimetastatic activity of a novel
mechanism-based gelatinase inhibitor. Cancer Res. 2005, 65, 3523–3526.
(16) Gu, Z. Z.; Cui, J.; Brown, S.; Fridman, R.; Mobashery, S.; Strongin,
A. Y.; Lipton, S. A. A highly specific inhibitor of matrix metalloproteinase-9
rescues laminin from proteolysis and neurons from apoptosis in transient
focal cerebral ischemia. J. Neurosci. 2005, 25, 6401–6408.
(17) Guo, Z. D.; Sun, X. C.; He, Z. H.; Jiang, Y.; Zhang, X. D.; Zhang,
J. H. Matrix metalloproteinase-9 potentiates early brain injury after
subarachnoid hemorrhage. Neurol. Res. 2010, 32, 715–720.
(18) Ota, R.; Kurihara, C.; Tsou, T. L.; Young, W. L.; Yeghiazarians,
Y.; Chang, M.; Mobashery, S.; Sakamoto, A.; Hashimoto, T. Roles of
matrix metalloproteinases in flow-induced outward vascular remodeling.
J. Cereb. Blood Flow Metab. 2009, 29, 1547–1558.
(19) Yu, F.; Kamada, H.; Niizuma, K.; Endo, H.; Chan, P. H.
Induction of MMP-9 expression and endothelial injury by oxidative
stress after spinal cord injury. J. Neurotrauma 2008, 25, 184–195.
(20) Testero, S. A.; Lee, M.; Staran, R. T.; Espahbodi, M.; Llarrull,
L. I.; Toth, M.; Mobashery, S.; Chang, M. Sulfonate-containing thiiranes
as selective gelatinase inhibitors. ACS Med. Chem. Lett. 2011, 2, 177–181.
(21) Lee, M.; Villegas-Estrada, A.; Celenza, G.; Boggess, B.; Toth,
M.; Kreitinger, G.; Forbes, C.; Fridman, R.; Mobashery, S.; Chang, M.
Metabolism of a highly selective gelatinase inhibitor generates active
metabolite. Chem. Biol. Drug Des. 2007, 70, 371–382.
(33) Coleman, M. D. Human Drug Metabolism; John Wiley & Sons,
Ltd.: New York, 2005.
(34) Gee, P.; Maron, D. M.; Ames, B. N. Detection and classification
of mutagens: a set of base-specific salmonella tester strains. Proc. Natl.
Acad. Sci. U.S.A. 1994, 91, 11606–11610.
(35) Sato, H.; Takino, T.; Okada, Y.; Cao, J.; Shinagawa, A.;
Yamamoto, E.; Seiki, M. A matrix metalloproteinase expressed on the
surface of invasive tumor-cells. Nature 1994, 370, 61–65.
(36) Deryugina, E. I.; Ratnikov, B.; Monosov, E.; Postnova, T. I.;
DiScipio, R.; Smith, J. W.; Strongin, A. Y. MT1-MMP initiates activation
of pro-MMP-2 and integrin alpha v beta 3 promotes maturation of
MMP-2 in breast carcinoma cells. Exp. Cell Res. 2001, 263, 209–223.
(37) Schywalsky, M.; Ihmsen, H.; Tzabazis, A.; Fechner, J.; Burak,
E.; Vornov, J.; Schwilden, H. Pharmacokinetics and pharmacodynamics
of the new propofol prodrug GPI 15715 in rats. Eur. J. Anaesthesiol. 2003,
20, 182–90.
(38) Fechner, J.; Schwilden, H.; Schuttler, J. Pharmacokinetics and
pharmacodynamics of GPI 15715 or fospropofol (Aquavan injection): a
water-soluble propofol prodrug. Handb. Exp. Pharmacol. 2008, 253–66.
(39) Lu, C.; Li, P.; Gallegos, R.; Uttamsingh, V.; Xia, C. Q.; Miwa,
G. T.; Balani, S. K.; Gan, L. S. Comparison of intrinsic clearance in liver
microsomes and hepatocytes from rats and humans: evaluation of free
fraction and uptake in hepatocytes. Drug Metab. Dispos. 2006, 34,
1600–1605.
(22) Gooyit, M.; Lee, M.; Hesek, D.; Boggess, B.; Oliver, A. G.;
Fridman, R.; Mobashery, S.; Chang, M. Synthesis, kinetic characteriza-
tion and metabolism of diastereomeric 2-(1-(4-phenoxyphenylsulfo-
nyl)ethyl)thiiranes as potent gelatinase and MT1-MMP inhibitors.
Chem. Biol. Drug Des. 2009, 74, 535–546.
(23) Lee, M.; Celenza, G.; Boggess, B.; Blase, J.; Shi, Q.; Toth, M.;
Bernardo, M. M.; Wolter, W. R.; Suckow, M. A.; Hesek, D.; Noll, B. C.;
Fridman, R.; Mobashery, S.; Chang, M. A potent gelatinase inhibitor
with anti-tumor-invasive activity and its metabolic disposition. Chem.
Biol. Drug Des. 2009, 73, 189–202.
(24) Stella, V. J. Prodrug strategies for improving drug-like proper-
ties. In Optimizing the “Drug-Like” Properties of Leads in Drug Discovery,
Borchardt, R. T., Kerns, E. H., Hageman, M. J., Thakker, D. R., Stevens,
J. L., Eds.; Springer Science and Business Media: New York, 2006;
Vol. IV, pp 221ꢀ237.
(25) Stella, V. J.; Nti-Addae, K. W. Prodrug strategies to overcome
poor water solubility. Adv. Drug Delivery Rev. 2007, 59, 677–694.
(26) Nam, N. H.; Kim, Y.; You, Y. J.; Hong, D. H.; Kim, H. M.; Ahn,
B. Z. Water soluble prodrugs of the antitumor agent 3-[3-amino-4-
methoxy)phenyl]-2-(3,4,5-trimethoxyphenyl)cyclopent-2-ene-1-one.
Bioorg. Med. Chem. 2003, 11, 1021–1029.
6690
dx.doi.org/10.1021/jm200566e |J. Med. Chem. 2011, 54, 6676–6690