Novel fluorinated derivatives of the broad-spectrum MMP inhibitors N-hydroxy-2(R)-[[(4-methoxyphenyl)sulfonyl](benzyl)- and (3-picolyl)-amino]-3- methyl-butanamide as potential tools for the molecular imaging of activated MMPs with PET

Article abstract of DOI:10.1021/jm0708533

An approach to the in vivo imaging of locally upregulated and activated matrix metalloproteinases (MMPs) found in many pathological processes is offered by positron emission tomography (PET). Hence, appropriate PET radioligands for MMP imaging are required. Here, we describe the syntheses of novel fluorinated MMP inhibitors (MMPIs) based on lead structures of the broad-spectrum inhibitors N-hydroxy-2(R)-[[(4-methoxyphenyl)sulfonyl](benzyl)-amino]-3-methyl-butanamide (CGS 25966) and N-hydroxy-2(R)-[[(4-methoxyphenyl)sulfonyl](3-picolyl)-amino]-3- methyl-butanamide (CGS 27023A). Additionally, tailor-made precursor compounds for radiolabeling with the positron-emitter ¹⁸F were synthesized. All prepared hydroxamate target compounds showed high in vitro MMP inhibition potencies for MMP-2, MMP-8, MMP-9, and MMP-13. As a consequence, the promising fluorinated hydroxamic acid derivative 1f was resynthesized in its ¹⁸F-labeled version via two different procedures yielding the potential PET radioligand [¹⁸F]If. As expected, the biodistribution behavior of this novel compound and that of the more hydrophilic variant [ ¹⁸F]1j, also developed by our group, indicates that there was no tissue specific accumulation in wild-type (WT) mice.

Full text of DOI:10.1021/jm0708533

5752
J. Med. Chem. 2007, 50, 5752-5764
Novel Fluorinated Derivatives of the Broad-Spectrum MMP Inhibitors
N-Hydroxy-2(R)-[[(4-methoxyphenyl)sulfonyl](benzyl)- and
(3-picolyl)-amino]-3-methyl-butanamide as Potential Tools for the Molecular Imaging of
Activated MMPs with PET
Stefan Wagner,*,†,‡,# Hans-Jo¨rg Breyholz,†,‡,# Marilyn P. Law,‡,# Andreas Faust,‡,# Carsten Ho¨ltke,‡,# Sandra Schro¨er,‡,#
,#
Gu¨nter Haufe,§,# Bodo Levkau,|,# Otmar Schober,‡,# Michael Scha¨fers,‡, and Klaus Kopka‡,#
Department of Nuclear Medicine, UniVersity Hospital of the Westfa¨lische Wilhelms-UniVersita¨t Mu¨nster, Germany, Institute of Organic
Chemistry, UniVersity of Mu¨nster, Mu¨nster, Germany, Institute of Pathophysiology, Center of Internal Medicine, UniVersity Hospital, Essen,
Germany, Interdisciplinary Center of Clinical Research (IZKF), Mu¨nster, Germany, and European Institute of Molecular Imaging (EIMI),
Mu¨nster, Germany
ReceiVed July 13, 2007
An approach to the in ViVo imaging of locally upregulated and activated matrix metalloproteinases (MMPs)
found in many pathological processes is offered by positron emission tomography (PET). Hence, appropriate
PET radioligands for MMP imaging are required. Here, we describe the syntheses of novel fluorinated
MMP inhibitors (MMPIs) based on lead structures of the broad-spectrum inhibitors N-hydroxy-2(R)-[[(4-
methoxyphenyl)sulfonyl](benzyl)-amino]-3-methyl-butanamide (CGS 25966) and N-hydroxy-2(R)-[[(4-
methoxyphenyl)sulfonyl](3-picolyl)-amino]-3-methyl-butanamide (CGS 27023A). Additionally, tailor-made
precursor compounds for radiolabeling with the positron-emitter 18F were synthesized. All prepared
hydroxamate target compounds showed high in Vitro MMP inhibition potencies for MMP-2, MMP-8, MMP-
9, and MMP-13. As a consequence, the promising fluorinated hydroxamic acid derivative 1f was resynthesized
in its 18F-labeled version via two different procedures yielding the potential PET radioligand [18F]1f. As
expected, the biodistribution behavior of this novel compound and that of the more hydrophilic variant
[18F]1j, also developed by our group, indicates that there was no tissue specific accumulation in wild-type
(WT) mice.
Introduction
found in numerous disease processes, including inflammation,
tumor cell metastasis, and atherosclerosis.8 Increases in MMP
activity in inflammatory, malignant, and degenerative diseases
have been attributed to increased cytokine and growth-factor-
stimulated gene transcription, enhanced zymogen activation, and
an imbalance in the MMP-to-TIMP ratio.9 The impact of MMPs,
especially in tumor progression and metastasis, has led to the
development of a number of different synthetic matrix metal-
loproteinase inhibitors (MMPIs) that block the activated enzyme
in pathological conditions by chelating the active site zinc(II)
ion of the MMP with a metal complexing moiety (e.g.,
carboxylic or hydroxamic acid).10 Synthetic MMPIs available
for the treatment of degenerative diseases are hydroxamate based
pseudopeptides (e.g., batimastat or marimastat)11,12 and non-
peptidyl sulfonamide-based hydroxamic acid inhibitors (e.g., AG
3340).13 Beside this therapeutical application, MMPIs are
potentially useful for the noninvasive in ViVo imaging of
activated MMPs with techniques used in diagnostic nuclear
medicine (e.g., single photon emission computed tomography
(SPECT) and positron emission tomography (PET)). Therefore,
radiolabeled analogues of MMPIs can serve as radiotracers for
the detection of unbalanced MMP levels. Such improved
diagnosis of diseases associated with dysregulated MMP activity
would be of great clinical interest and would strongly influence
therapy and diagnosis algorithms.
Matrix metalloproteinases (MMPsa) are a family of zinc- and
calcium-dependent endopeptidases. The family of MMPs is able
to degrade all protein components of the extracellular matrix
(ECM) with overlapping substrate specificities and is subdivided
into several subclasses.1-4
MMPs are secreted as inactive proenzymes or zymogens.
Proteolytic cleavage of the prodomain results in the release of
the cysteine residue from the zinc ion and in the subsequent
activation of the enzyme. This mechanism is termed “cysteine
switch”.5 MMP activity is tightly controlled by several endog-
enous inhibitors, for example, the four tissue inhibitors of
metalloproteinases (TIMPs).6,7
MMPs and TIMPs are involved in the normal physiology of
connective tissue during development, morphogenesis, and
wound healing. Elevated MMP expression and activity has been
* To whom correspondence should be addressed. Address: Department
of Nuclear Medicine, University Hospital of the Westfa¨lische Wilhelms-
Universita¨t, Albert-Schweitzer-Str. 33, D-48149 Mu¨nster, Germany.
Phone: +49-251-83-47362. Fax: +49-251-83-47363. E-mail: stwagner@uni-
muenster.de.
† These authors contributed equally to this work.
‡ Department of Nuclear Medicine, University Hospital, Mu¨nster.
§ Institute of Organic Chemistry, University of Mu¨nster.
| University Hospital, Essen.
Interdisciplinary Center of Clinical Research (IZKF).
# European Institute of Molecular Imaging (EIMI).
One approach for the imaging of activated MMPs is based
on the application of labeled peptides or proteins acting as
MMPIs.14-19
a Abbreviations: EDC, N-ethyl-N′-(3-dimethylaminopropyl)-carbodiimide
hydrochloride; EM, exact mass; EOS, end of synthesis; ESI, electron spray
ionization; HOBt, 1-hydroxy-benzotriazole; K 222, Kryptofix 2.2.2; MMP,
matrix metalloproteinase; MMPI, matrix metalloproteinase inhibitor; NMM,
4-methyl-morpholine; PET, positron emission tomography; p.i., post injec-
tion; SD, standard deviation; SPECT, single photon emission computed
tomography; WT, wild-type.
Another approach is based on the application of radiolabeled
small molecule nonpeptidyl MMPIs. Most developments of
radiolabeled MMPIs are based on the synthesis and in Vitro as
10.1021/jm0708533 CCC: $37.00 © 2007 American Chemical Society
Published on Web 10/23/2007

Broad-Spectrum MMPIs for MMP Imaging with PET
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5753
sor compounds of the 18F-labeled analogues were synthesized.
These hydroxamic acids 1a-f, 1h, and 1j-n were tested in
Vitro in MMP-2, -8, -9, and -13 fluorogenic assays and compared
with selected carboxylic acid analogues 5a-c, 5f, and 5h-l
occurring in our synthesis sequences, because compounds
containing a carboxylic acid moiety that functions as the zinc
complexing group and a suitable substitution pattern can show
MMPI potency as well.1
Additionally, the optimized radiosynthesis of [18F]1f is shown.
Compound 1f was identified as one of the most potent MMPIs
of our new library (Table 1). Futhermore, the biodistribution
of [18F]1f was investigated in wild-type (WT) mice side-by-
side with that of the more hydrophilic variant [18F]1j.
Figure 1. Lead structures of CGS 25966 and CGS 27023A with a
selection of radiolabeled derivatives.
Results
Chemistry. From the (radio)chemical point of view, the
introduction of fluorine into the CGS lead structure is only
meaningful for distinct molecule positions because the radio-
chemical synthesis of the corresponding 18F-labeled isotopomers
from potential precursor compounds most often is realized with
the labeling species [18F]F- that is frequently used in nucleo-
philic substitutions. Therefore, the nonradioactive CGS deriva-
tives with a radiochemically meaningful fluorine substitution
pattern and the precursor compounds for nucleophilic 18F-
fluorination were initially synthesized to investigate their in Vitro
MMP inhibition potencies in enzyme assays and to select
suitable candidates for resynthesis of the radioactive 18F-labeled
isotopomers. The synthesized fluorinated CGS 25966 and CGS
27023A derivatives and their potential precursor compounds
can be classified in three groups.
The first group is represented by the corresponding core-
fluorinated analogues 1a and 1c characterized by a fluorine
substitution at the phenyl ring in para-position to the electron
withdrawing sulfonamide group (Scheme 1). The synthesis of
the nonradioactive CGS derivatives 1a and 1c, based on the
modification of D-valine, is depicted in Scheme 1. The
sulfonamide 3a was derived from the commercially available
tert-butylester of D-valine hydrochloride 2 and 4-fluorobenze-
nesulfonyl chloride in the presence of pyridine (yield: 94%).
The phenyl or picolyl group was introduced to give the
intermediates 4a and 4c, respectively (yields: 72-73%),
followed by the acidic removal of the ester protective group
(yields: 49-79%). Conversion of the resulting carboxylic acids
5a and 5c into the corresponding hydroxamic acid esters 6a
and 6d was achieved by reaction with O-tert-butylhydroxyl-
amine hydrochloride, N-ethyl-N′-(3-dimethylaminopropyl)-car-
bodiimide hydrochloride (EDC), 1-hydroxy-benzotriazole (HOBt),
and 4-methyl-morpholine (NMM) (yields: 78-87%). Depro-
tection of the hydroxamic acid esters 6a and 6d was, once again,
performed by treatment with hydrochloric acid gas, yielding
the target compounds 1a and 1c (yields: 31-82%).
well as in ViVo evaluation of such nonpeptidyl tracers, and they
were reviewed recently.20 In this second group, the N-sulfonyl
amino acid hydroxamates CGS 25966 and CGS 27023A
represent prominent lead structures for potential radiolabeled
tracers (Figure 1). The lead compounds CGS 25966 and CGS
27023A are broad-spectrum inhibitors that inhibit MMP-1
(interstitial collagenase, Ki ) 43 and 33 nM, respectively),
MMP-2 (gelatinase A, Ki ) 11 and 20 nM, respectively),
MMP-3 (stromelysin-1, Ki ) 34 and 43 nM, respectively), and
MMP-9 (gelatinase B, Ki ) 27 and 8 nM, respectively) by
chelating the zinc ion of the enzyme active site with the
hydroxamic acid moiety.21,22
Lately, our group synthesized a SPECT-compatible 123I-
labeled derivate of CGS 27023A ([123I]I-HO-CGS 27023A) for
the assessment of MMPs in Vitro and in ViVo (Figure 1). This
compound was successfully used to image specifically in ViVo
activated MMPs in vascular lesions developing after carotid
artery ligation in apolipoprotein E-deficient (ApoE-/-) mice by
means of planar scintigraphy. This was the first published in
ViVo experiment demonstrating that upregulated MMP levels
can be detected with appropriate radiotracers and may provide
a new diagnostic parameter in nuclear medicine.23,24 In the next
step, our group and others focused on the development of PET-
compatible CGS 25966 and CGS 27023A derivatives labeled
with the positron emitters 11C and 18F. To date, several CGS
25966 and CGS 27023A derivatives, mainly labeled with 11C,
have been synthesized as potential PET tracers for activated
MMPs in breast cancer.25-28 However, only [11C]CGS 25966
(Figure 1) has been studied in two animal models of breast
cancer. Due to unpromising results arising from these studies,
[11C]CGS 25966 was excluded from further in ViVo evalua-
tions.29 Furthermore, only two 18F-labeled derivatives of CGS
25966 and CGS 27023A have been mentioned in the literature.
In 2001, compound [18F]1c (Figure 1) was introduced without
giving detailed information about the synthesis. To date, in Vitro
and in ViVo evaluations of [18F]1c have not been published.25,26
Recently, our group succeeded in the synthesis of another 18F-
labeled CGS 27023A analogue with a 2-fluoroethoxy subunit
[18F]1j (Figure 1) via a three-step radiosynthesis. The nonra-
dioactive counterpart 1j was found to be a potent inhibitor of
MMP-2 and MMP-9 (IC50 ) 3 nM (MMP-2), IC50 ) 7 nM
(MMP-9)).30
Potential precursor compounds of the 18F-labeled analogues
[18F]1a and [18F]1c possess a suitable leaving group for a
nucleophilic substitution with [18F]F- (e.g., -NO2, -NMe3+).
Therefore, the syntheses of the corresponding trimethylammo-
nium- and nitro-CGS precursors were achieved. Additionally,
the dimethylamino derivatives 1b and 1d were synthesized for
comparative investigations in the enzyme assays.
The work presented here focuses on the systematic investiga-
tion of further new fluorinated derivatives of CGS 25966 and
CGS 27023A, that should also be achievable as corresponding
18F-labeled isotopomers via fast radiosyntheses starting with the
most prominent 18F-labeling species [18F]fluoride ([18F]F-).
Therefore, the nonradioactive counterparts and potential precur-
Suitable precursors for the substitution of the trimethylam-
monium group with [18F]F- should still feature the O-tert-butyl-
protective group, because the acidic deprotection of this moiety
with hydrochloric acid would lead to an undesired cleavage of
the trimethylammonium group, making a subsequent 18F-
fluorination impossible. In detail, the hydroxamic acid esters

5754 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Scheme 1. Syntheses of MMPIs 1a-1na
Wagner et al.
a Reagents: (a) 4-R1-PhSO2Cl, pyridine; (b) benzylbromide, 3-picolyl chloride, 3-bromomethyl-2-fluoropyridine, 2-bromo-3-bromomethylpyridine,
5-bromomethyl-2-fluoropyridine or 2-bromo-5-bromomethylpyridine, K2CO3, DMF; (c) HCl, dichloroethane; (d) t-Bu-ONH2, EDC, NMM, HOBt, CH2Cl2;
(e) NaI, 2-butanone; (f) MeOTf, CH2Cl2, CH3CN; (g) MeOTf, solvent or MeI, CH2Cl2; (h) NaI, acetone; (i) AgOTs, CH3CN.
6b and 6e, synthesized in analogy to the above-mentioned
sequence (Scheme 1, overall yields: 31-46%), were quaternized
with methyltriflate or methyliodide. While the quaternization
of 6b yielded the desired trimethylammonium salt 6c (yield:
28%) (Scheme 1), the N-methylation of 6e under various
reaction conditions always led to the undesired methylpyri-
dinium compound 6f (yield: 94%) (Scheme 1).
However, the synthesis of the nitro-substituted CGS 27023A
derivative 1e, that also represents a potential precursor of
[18F]1c, was successful via the general synthesis route with a
high overall yield of 53%.
that displaces the methoxy moiety of the lead compounds. The
CGS 25966 analogue 1f was prepared in a five-step procedure
with 5% overall chemical yield (Scheme 1). The CGS 27023A
analogue 1j and its 18F-radiolabeled isotopomer [18F]1j were
prepared by our group as described previously.30 A potential
precursor for a two-step radiosynthesis of [18F]1f was synthe-
sized via the chloro/bromo compound mixture 3e, that was
formed in the first synthesis step from the reaction of
D-valine-tert-butylester hydrochloride 2 and 4-(2-bromoethoxy)-
benzenesulfonyl chloride. Obviously, the unintended synthesis
of the chloro dervivative in the mixture 3e arises from the
nucleophilic substitution of the bromine in the bromoethoxy
core with chloride that descended from compound 2 and/or
The second group of fluorinated CGS 27023A and CGS
25966 analogues is characterized by a 2-fluoroethoxy subunit

Broad-Spectrum MMPIs for MMP Imaging with PET
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5755
Scheme 2. Synthesis of Radiolabeled MMPI [18F]1fa
a Reagents: (a) [18F]K(K 222)F, K2CO3, CH3CN; (b) TFA.
the benzenesulfonyl chloride derivative. Nevertheless, the
mixture 3e is suitable for the synthesis of the desired corre-
sponding hydroxamic acid ester mixture 6i. Mixture 6i was
iodinated to provide 6j, which was subsequently tosylated to
yield precursor 6k (Scheme 1, yield: 98%). Alternatively, a
potential precursor for a one-step radiosynthesis of [18F]1f was
prepared from the chloro/bromo compound mixture 1g. Sub-
sequent iodination and tosylation yielded the potential tosyl
precursor 1i.
The third group of target compounds is exclusively composed
of CGS 27023A derivatives and based on the possibility to
introduce [18F]F- in an electron-deficient aromatic subunit with
a nucleofuge group (e.g., a bromo-substituted picolyl subunit).31
Here, the 2- and 6-fluoro-3-picolyl CGS 27023A derivatives
1k and 1m (overall yield: 3-9%) and the potential bromo
precursors 1l and 1n (overall yield: 8-9%) of the 18F-labeled
isotopomers were prepared according to Scheme 1. For that
purpose, the pyridine derivatives used in step two (3-bromom-
ethyl-2-fluoropyridine, 2-bromo-3-bromomethylpyridine, 5-bro-
momethyl-2-fluoropyridine, and 2-bromo-5-bromomethylpyri-
dine) were synthesized according to ref 32.
Radiochemistry. Due to the high in Vitro MMPI potency of
compound 1f (see the section on enzyme assays), the radio-
synthesis of its 18F-labeled analogue [18F]1f was developed and
optimized (Scheme 2). The radiosyntheses of the remaining
ligands [18F]1a, [18F]1c, [18F]1k, and [18F]1m are in develop-
ment.
Two different routes for the preparation of [18F]1f were
evaluated (a two-step and a one-step procedure). In the two-
step procedure, the tosylate precursor 6k was converted into
the 18F-substituted ester intermediate [18F]6h using [18F]K-
(Kryptofix 2.2.2)F ([18F]K(K 222)F) and potassium carbonate
in acetonitrile. Compound [18F]6h was produced with a radio-
chemical yield of 39.5 ( 5.1% (noted as mean ( SD, decay-
corrected, n ) 5). The hydrolysis of [18F]6h in trifluoroacetic
acid (TFA) resulted in the target compound [18F]1f with a
radiochemical yield of 31.4 ( 6.5% (decay-corrected, n ) 5).
The overall radiochemical yield of [18F]1f for both steps was
12.4 ( 3.0% (decay-corrected). The synthesis was achieved with
a radiochemical purity of 97.3 ( 1.6% in 113 ( 6 min from
end of radionuclide production and a calculated specific
radioactivity of 44 ( 19 GBq/µmol (32-72 GBq/µmol) at the
end of synthesis (EOS) (n ) 5).
Additionally, radiosynthesis of the ligand [18F]1f was suc-
cessfully reduced to a one-step procedure using the tosylate
precursor 1i. Direct nucleophilic substitution of hydroxamic acid
1i with [18F]fluoride provided [18F]1f with a radiochemical yield
of 45.6 ( 5.6% (decay-corrected), a radiochemical purity of
95.5 ( 3.1% in 115 ( 10 min from end of radionuclide
production, and a calculated specific radioactivity of 60 ( 25
GBq/µmol (42-97 GBq/µmol) at the EOS (n ) 5). Altogether,
both procedures provided the chemically pure radioligand
[18F]1f. Besides the injection peak the UV-trace of the HPLC
quality control of the purified [18F]1f fraction showed only an
additional peak for the carrier, from which the specific
radioactivity was calculated.
Enzyme Assays and log D Values. The MMP inhibition
potencies of the hydroxamic acids CGS 27023A, 1a-1f, 1h,
and 1j-1n and of the carboxylic acids 5a-5c, 5f, and 5h-5l
were measured in fluorogenic in Vitro inhibition assays for
MMP-2, MMP-8, MMP-9, and MMP-13. The resulting IC50-
values for the investigated MMPs were calculated from a
nonlinear regression fit of the concentration-dependent reaction
rates and were compared to those of the parent compound CGS
27023A. The results are shown in Table 1.
The table also contains the Ki-values of CGS 25966 taken
from ref 22 and the calculated log P and log D values (clog P
and clog D) of the synthesized hydroxamic and carboxylic acids
to indicate the changes of lipophilicities caused by the structural
modifications of the CGS lead compounds. Additionally, the
logD values of the radioligands [18F]1f and [18F]1j were
determined experimentally (Table 1 footnotes d and e). Com-
parison of calculated and experimental log D values showed
that these data differ in 1-2 units (clog D of 1f: 4.03 versus
log D of [18F]1f: 2.02 and clog D of 1j: 2.53 versus log D of
[18F]1j: 1.34).
Our results verify that CGS 27023A is a potent broad-
spectrum inhibitor of the tested MMPs with IC50-values in the

5756 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Wagner et al.
Table 1. MMP Inhibition Potencies of the CGS 25966 and CGS 27023A Derivatives for MMP-2, -8, -9, and -13 Expressed in IC50-Values
IC50 [nM]a
cpd
MMP-2
MMP-8
MMP-9
MMP-13
34 ( 12
clog Pb
clog Db
CGS 25966
CGS 27023A
11c
13c
27c
3.81
2.32
3.60
4.99
3.72
5.11
2.11
3.50
2.23
2.00
4.04
5.42
4.92
6.31
2.54
2.43
3.81
3.12
4.51
2.43
3.81
3.12
4.51
3.80
2.30
3.60
1.41
3.72
1.56
2.10
-0.10
2.22
1.98
4.03d
1.85
4.91
2.74
2.53e
2.41
0.21
3.11
0.90
2.41
0.21
3.11
0.91
3 ( 1
2 ( 0.4
12 ( 9
6 ( 3
1a
5a
1b
5b
1c
5c
1d
1e
1f
239 ( 154
4350 ( 950
335 ( 45
(41 ( 36) × 103
120 ( 76
>10 × 104
992 ( 105
756 ( 215
4 ( 3
281 ( 101
821 ( 253
614 ( 239
(15 ( 1) × 103
414 ( 122
>10 × 104
1330 ( 1100
5150 ( 1840
50 ( 27
384 ( 147
9140 ( 3280
10 ( 1
754 ( 65
105 ( 34
(14 ( 11) × 103
228 ( 56
(71 ( 6) × 103
1200 ( 89
14 ( 2
(14 ( 4) × 103
527 ( 90
>10 × 104
1340 ( 506
10 ( 0.5
2 ( 1
2460 ( 464
3 ( 1
11 ( 0.3
5f
5590 ( 638
22 ( 9
1300 ( 114
69 ( 29
>10 × 104
1h
5h
1j
1k
5i
1l
5j
1m
5k
1n
5l
10 ( 5
207 ( 28
3 ( 1
n.d.f
1350 ( 554
7 ( 1
n.d.
2 ( 1
0.8 ( 0.03
19 ( 9
18 ( 8
51 ( 24
50 ( 23
(20 ( 11) × 103
(78 ( 11) × 103
(10 ( 5) × 103
103 ( 6
(41 ( 7) × 103
59 ( 20
30 ( 14
75 ( 7
9110 ( 1170
8 ( 1
7480 ( 2050
0.9 ( 0.3
(22 ( 2) × 103
20 ( 4
(35 ( 1) × 103
(34 ( 7) × 103
0.9 ( 0.2
0.5 ( 0.1
(35 ( 1) × 103
9820 ( 2060
23 ( 8
(30 ( 5) × 103
2 ( 1
24 ( 4
6120 ( 3400
6760 ( 2440
6930 ( 4110
3260 ( 213
a Values are the mean ( SD of three assays. b log P and log D values were calculated with ACD/Chemsketch Labs 6.00 (log D ) log P at physiological
pH). c Ki-values from ref 22, where SDs are not denoted. d Experimental determination of log D for compound [18F]1f: 2.02 ( 0.03 (noted as mean ( SD,
n ) 4). e Experimental determination of log D for compound [18F]1j: 1.34 ( 0.04 (noted as mean ( SD, n ) 4). f n.d.: not determined.
low nanomolar range (IC50 ) 2-34 nM). Compared to this lead
structure, the core-fluorinated compound 1a is less potent and
the IC50-values range between 12 and 384 nM. Interestingly,
this compound shows a moderate selectivity (20-32 fold) for
collagenase 2 (MMP-8, IC50 ) 12 nM). To a minor extent, this
tendency also occurred for the corresponding carboxylic acid
5a (MMP-8, IC50 ) 754 nM). Here, a low selectivity for MMP-8
and MMP-9 (∼5-13 fold) is observed in comparison to MMP-2
and MMP-13. The carboxylic acid derivative 5a shows inhibi-
tion potentcies that are decreased by a factor of 3-63 compared
to the hydroxamic acid analogue 1a. This result is not surprising,
because the intrinsic binding affinity of bidentate hydroxamates
for zinc (of enzyme active sites) is much higher than that of
monodentate carboxylates.1 A different result is pointed out by
the compound pair of the dimethylamino-substituted hydrox-
amate 1b and carboxylate 5b. Compound 1b is also a potent
inhibitor (IC50 ) 10-614 nM) but possesses a 11-61-fold
selectivity for collagenase 3 (MMP-13). In contrast, carboxylic
acid 5b shows nonselective binding and only low inhibition
potencies with IC50-values in the micromolar range (14-41 µM).
The core-fluorinated CGS 27023A derivative 1c exhibits similar
binding potencies for MMP-2, -9 and, -13 (IC50 ) 120-527
nM) compared to the CGS 25966 counterpart 1a (IC50 ) 239-
384 nM), but the binding selectivity of derivative 1a for MMP-8
was not found for 1c. The corresponding carboxylate 5c is an
ineffective inhibitor of the tested MMPs (IC50 ) 71 to >100
µM). Furthermore, the dimethylamino hydroxamate 1d (a CGS
27023A derivative) possesses IC50-values of ∼1 µM (992-1340
nM) and is, in contrast to the CGS 25966 analogue 1b, not
specific for MMP-13. The nitro-substituted hydroxamate 1e was
identified as collagenase-selective MMPI (54-515 fold). In
comparison to the lead structure CGS 27023A, compound 1e
remains very potent for MMP-8 and -13 (IC50 ) 10-14 nM)
but its MMP inhibition potency for gelatinases A and B (MMP-2
and -9) is remarkably reduced (IC50 ) 756-5150 nM).
introduction of a slightly more sterically demanding group, 1f
conserves the broad-spectrum MMPI potency of the lead
structures with low nanomolar IC50-values (2-50 nM). Again,
the corresponding carboxylate 5f clearly exhibits a decreased
MMP inhibition potency (26 to >9000-fold) compared to the
active hydroxamate analogue 1f. The halogeno analogue 1h of
compound 1f with the sterically demanding iodine again leads
to broad-spectrum inhibition potencies which are not remarkably
influenced by the introduction of the bigger iodine atom (IC50
) 3-69 nM). The corresponding carboxylic acid 5h remains a
moderate inhibitor for MMP-2 (IC50 ) 207 nM). The CGS
27023A derivative 1j is a very potent inhibitor of the investi-
gated MMPs. This compound is even more potent against MMP-
13 compared to the lead structure CGS 27023A (43-fold).
Compared to the CGS 25966 counterpart 1f, compound 1j shows
an increased effective inhibition of MMP-9 and MMP-13 (IC50
) 50 versus 7 and 11 versus 0.8 nM, respectively).
The compound groups 1k-1n and 5i-5l are composed of
the CGS 27023A derivatives and the corresponding carboxylates
that possess a halogen substituent (fluorine or bromine) at the
picolyl subunit. In comparison to the lead structure CGS
27023A, the fluorine substitution in 1k causes a slightly
decreased but effective inhibition potency with IC50-values in
the low nanomolar range (IC50 ) 18-51 nM). As expected,
the corresponding carboxylate moiety in compound 5i leads to
a dramatic loss of inhibition potency (IC50 ) 10-78 µM). This
behavior is also shown by the carboxylic acids 5j-5l that are
g100-fold less potent than the related hydroxamic acids 1l-
1n. A comparison of the bromo-substituted derivative 1l with
the corresponding fluoro compound 1k indicates that the
introduction of a sterically more demanding bromo atom leads
to a slight decrease in the inhibitory effectivity for all tested
MMPs (IC50 ) 18-51 versus 30-103 nM). Additionally, the
hydroxamate with a 6-fluoro substituent in the picolyl core 1m
possesses superior inhibition potential compared to the above-
Such selectivity is not shown by the CGS 25966 type
hydroxamate 1f with a 2-fluoroethoxy subunit that displaces
the methoxy moiety of the lead compound. In spite of the
mentioned analogue with fluorine at the 2-position 1k (IC50 )
0.5-8 versus 18-51 nM). Compound 1m is the most potent
inhibitor of the here presented compound series with even

Broad-Spectrum MMPIs for MMP Imaging with PET
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5757
Figure 2. Radioactivity in tissues of mice at 20 min after i.v. injection
of [18F]1f (solid bar) or [18F]1j (slanted stripe bar). Averages for two
mice with ranges are shown. The upper panel illustrates uptake in tissues
involved with metabolism and excretion (uptake indices 1-15), and
the lower panel shows possible target tissues (uptake indices 0.1-1.5).
Indices for blood are shown in both panels to assist comparison.
improved IC50-values compared to the lead structure CGS
27023A (IC50 ) 0.5-8 versus 2-34 nM). The bromo analogue
1n remains a potent broad-spectrum MMPI with slightly
decreased IC50-values for MMP-8, -9, and -13 in comparison
to the fluorinated derivative 1m (IC50 ) 20 versus 0.9 nM
(MMP-8), 23 versus 0.5 nM (MMP-9), and 24 versus 0.9 nM
(MMP-13)). In contrast to the fluoro/bromo compound pair 1k
and 1l, the bromo analogue 1n is a more effective inhibitor for
MMP-2 than the fluoro derivative 1m (IC50 ) 2 versus 8 nM
(MMP-2)). The comparison of the compound series 1k-1n
shows the general tendency that the halogen substitution at the
6-position results in more effective MMP inhibitors than the
substitution at the 2-position of the picolyl-ring.
Biodistribution and Metabolism. Figure 2 shows radioactiv-
ity in tissue samples of mice at 20 min after intravenous (i.v.)
injection of [18F]1f or [18F]1j. Averages for two mice with
ranges are shown. The upper panel illustrates uptake in tissues
involved with metabolic and excretory processes (intestinal tract,
liver, and kidney), and the lower panel shows possible target
tissues (lung, heart, muscle, and brain). The high uptake in the
duodenum may be related to the high levels of radioactivity
(uptake index: >30) seen in the bile (data not shown). Results
for [18F]1f and [18F]1j were comparable.
Figure 3. Radioactivity in tissues of mice after i.v. injection of
[18F]1f or [18F]1j. Each symbol indicates an uptake index for one mouse.
Compound [18F]1f (b) or [18F]1j (2) (13-38 MBq‚kg-1, 0.98-4.5
nmol‚kg-1) was given alone or 10 min after unlabeled 1f (O) or 1j
(4), respectively (g25 µmol‚kg-1). As there was no difference
between blood activities after injection of [18F]1f or [18F]1j, a single
fit to all data is shown. For other tissues, the solid lines are fitted to all
data for [18F]1f radioactivity and the broken lines are fitted to all data
for [18F]1j.
very high levels of radioactivity were detected in the bile for
both compounds. Uptake indices ranged from 13 at 2 min after
injection to 23-89 between 45 and 60 min.
Figure 3 illustrates time activity curves for selected tissues.
Exponential fits are shown to assist comparison of the data for
various tissues. The first sampling time was 2 min after i.v.
injection, at which time radioactivity in the blood was low
(uptake index: 1.5) and little further change was observed during
the next hour. Uptake in the myocardium was comparable to
that in the blood for both [18F]1f or [18F]1j, with there being
no evidence for accumulation of radioactivity. Radioactivity was
cleared from the tissues which showed high levels at early times
after injection. The results for [18F]1j in the kidney, however,
were very variable; three mice showed uptake indices of ∼7 at
20, 45, and 60 min, while another three mice showed values of
e1 at the same times, comparable with those for [18F]1f.
Predosing with unlabeled 1f or 1j had no demonstratable effect.
Radioactivity was detected in urine at 2 min after injection, and
although there was an increase with time, only moderate levels
were observed (uptake indices: 8-12 at 60 min). In contrast,
Additionally, plasma samples of WT mice 2.5 and 20 min
postinjection (p.i.) of [18F]1f or [18F]1j were analyzed by high-
performance liquid chromatorgraphy (HPLC). After 2.5 min,
86% of the remaining plasma radioactivity originated from
[18F]1f and 14% originated from a polar metabolite detected in
the γ-trace (trace of radioactivity) of the HPLC chromatogram.
After 20 min, the plasma radioactivity was divided into three
fractions. In detail, 37% of [18F]1f was left, the percentage of
the polar metabolite increased to 21%, and in addition a second
metabolite with a similar retention time to the parent compound
[18F]1f occurred (42%). The picolyl derivative [18F]1j showed
a different behavior. After 2.5 min, the ratio of intact [18F]1j to
a polar metabolite in plasma was 6/4. The amount of the polar
metabolite increased to 47%, and that of [18F]1j decreased to
53% of radioactivity remaining in the plasma 20 min p.i. To
get an idea about the extent of tracer defluorination and
generation of [18F]F- that accumulates in ViVo in the skeleton,

5758 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Wagner et al.
the uptake indices of the right tibiae in comparison to those of
the heart were determined for the above-mentioned WT mice
20 min after injection of [18F]1f or [18F]1j. After injection of
[18F]1f, an uptake index for the tibiae of 0.37 ( 0.10 (n ) 4)
was observed. In each case, the percentage of injected dose of
the tibiae was even lower than that of the heart (0.72 ( 0.11,
n ) 4). Similar results were provided by the study with [18F]-
1j. The uptake index was 0.17 ( 0.02% for the heart and 0.12
( 0.02% for the right tibiae (n ) 4).
in a commerially available remotely controlled PET tracer
synthesizer. Both procedures were achieved in 113-115 min,
which complies with ∼1 half-life of 18F (109.7 min). Compared
to the two-step procedure, the one-step approach took the same
amount of time because the cartridge purification of the one-
step procedure is more time-consuming than that of the two-
step synthesis. However, being aware of the time-consuming
and challenging preparation of the CGS 27013A-derived MMPI
[18F]1j (three-step synthesis: radiochemical yield, 13.4 ( 5.3%
(decay-corrected); synthesis time, 211 ( 30 min after end of
radionuclide production (n ) 8)30), the radiosynthesis of the
CGS 25966 analogue [18F]1f presented here was further
developed, optimized, and finally simplified, resulting in a faster
one-step radiosynthesis characterized by higher radiochemical
yields.
Furthermore, the experimental log D value determination of
the radioligands [18F]1f and [18F]1j (Table 1 footnotes d and e)
in comparison to the log D value calculation of 1f and 1j pointed
out that experimental measurements and theoretical calculations
exhibit a distinct difference of 1-2 units (clog D of 1f: 4.03
versus log D of [18F]1f: 2.02 and clog D of 1j: 2.53 versus
log D of [18F]1j: 1.34). The solubility of the radioligands in
the buffer was 10-100-fold increased in comparison to the
calculated solubility. Maybe the polar character of a hydrox-
amate moiety is not considered adequately by the calculation
software as in the case for the corresponding carboxylic acids.
Nevertheless, the expected tendency of a higher lipophilicity
of the phenyl compound 1f compared to the picolyl derivative
1j was reflected by the calculations.
The prepared hydroxamic acids and selected carboxylic acids
were evaluated in MMP inhibition assays using MMP-2, -8,
-9, and -13. For the first time, in this study, the aromatic
substituents of the lead structures CGS 25966 and CGS 27023A
were varied, resulting in new and valuable results.
First, the hydroxamates of the series are more potent MMP
inhibitors than the corresponding carboxylates. In most cases,
the MMP inhibition potencies of the carboxylates are g10-fold
decreased. This result was also found for other MMPI classes,
and it is caused by the stronger intrinsic binding affinity of
hydroxamates for the active site zinc ion in comparison to that
of carboxylates.1
Second, the CGS 27023A lead structure tolerates a fluoro or
bromo substitution at the picolyl ring. While compounds 1k
and 1l substituted in 2-position are comparable to CGS 27023A
(IC50 ) 18-103 versus 2-34 nM), the 6-substituted derivatives
1m and 1n are even more potent MMP inhibitors for some of
the investigated MMPs (IC50 ) 0.5-24 nM). Furthermore, a
fluoro substitution is favored over a bromo substitution. Obvi-
ously, a linear substitution pattern at the picolyl ring with a
little halogen substituent (1m) leads to more attractive and/or
less repulsive interactions in the S2′ enzyme pocket (the enzyme
pocket that is commonly occupied by the picolyl ring; see, e.g.,
ref 1) than a nonlinear substitution with a sterically more
demanding halogen (1l).
Third, the replacement of the methoxy moiety of the
phenylsulfonamide substructure (this residue commonly oc-
cupies the S1′ enzyme pocket; see, e.g., ref 1) leads to selective
MMPIs in three cases, including one core-fluorinated derivative.
These results are not expected for derivatives of lead structures
that are classical examples of nonselective broad-spectrum
MMPIs such as CGS 25966 and CGS 27023A. From our results,
the collagenase-selective inhibitors 1a (MMP-8 selective), 1b
(MMP-13 selective), and 1e (MMP-8 and MMP-13 selective)
were identified. This unexpected selectivity might be caused
Discussion
Most of the 123I-, 11C-, and 18F-labeled MMPI radioligands
synthesized so far are based on nonpeptidyl MMPIs with
carboxylate or hydroxamate moieties as zinc binding function-
alities. The radioligand [123I]I-HO-CGS 27023A (Figure 1)
which was evaluated by our group represents the first designed
radiotracer with moderate lipophilicity (calculated log D ) 1.60
(pH 7.4)) that specifically accumulates in vascular MMP-9-rich
lesions of apolipoprotein-deficient mice (ApoE-/-).24 The
specific accumulation can be observed despite lower potencies
(experimental data of ref 23: IC50 ) 153 nM (MMP-9)) in
comparison with the parent compound CGS 27023A (Table 1:
IC50 ) 6 nM (MMP-9)). The aim of our present study was the
development of an improved CGS 25966/CGS 27023A-based
radioligand that possesses high inhibition potencies for gelati-
nases and collagenases and, at the same time, can be labeled
with the positron emitter 18F using a straightforward radiosyn-
thesis. Subsequently, nonradioactive CGS derivatives with a
fluorine substituent were synthesized (Scheme 1). The fluorine
substitution pattern of these compounds was chosen in a manner
that a radiochemical resynthesis by nucleophilic 18F-fluorination
should be possible. Thus, six new fluorinated CGS 25966 and
CGS 27023A derivatives (1a, 1c, 1f, 1j, 1k, and 1m) were
prepared in five steps with overall chemical yields ranging
between 3 and 34%. Inhibition potencies of compounds were
investigated in fluorogenic MMP inhibition assays for MMP-
2, -8, -9, and -13. Additionally, nine potential precursors for
syntheses of the corresponding 18F-labeled radioligands were
prepared, that serve for a one-step (one-step precursor) or two-
step radiosynthesis (two-step precursor) (overall chemical yields
in 4-7 steps: 8-58%). In detail, trimethylammonium com-
pound 6c was prepared as a two-step precursor for the
radiosynthesis of the potential PET MMPI [18F]1a, nitro
compounds 6g and 1e were obtained as two- or one-step
precursors for 18F-fluorinations giving [18F]1c, tosylates 6k and
1i were synthesized as two- or one-step precursors for the CGS
25966 derivative [18F]1f, and bromo compounds 6m, 1l, 6o,
and 1n were prepared as two- or one-step precursors for the
synthesis of the CGS 27023A analogues [18F]1k and [18F]1m,
respectively. Syntheses of the precursor, nonradioactive refer-
ence 1j and radioligand [18F]1j were already described by our
group elsewhere.23,30 As a result of our in Vitro evaluation (see
below for discussion), we have chosen the potent inhibitor 1f
for synthesis of its 18F-labeled counterpart [18F]1f. Two alterna-
tive approaches for the radiosynthesis of [18F]1f were success-
fully developed. In a two-step procedure, target compound
[18F]1f was prepared from the hydroxamic acid ester 6k with a
moderate radiochemical yield of 12.4 ( 3.0% (decay-corrected)
(Scheme 2). An improvement was achieved by realization of a
one-step procedure starting from the tosylate 1i which provided
[18F]1f with a higher radiochemical yield of 45.6 ( 5.6%
(decay-corrected), circumventing the subsequent hydrolysis of
the ester function of compound 6k. The one-step procedure
offers the improvement of radiation protection by a feasible
setup of a fully automated radiosynthesis of compound [18F]1f

Broad-Spectrum MMPIs for MMP Imaging with PET
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5759
by electronic and/or sterical reasons. To clarify this observation,
a systematic investigation with appropriate substituted deriva-
tives of the lead compounds has to be carried out, but this is
out of the focus of our work.
these radioligands. The compounds were tested in in Vitro MMP
inhibition assays for their MMP-2, -8, -9, and -13 inhibition
potencies, resulting in the identification of a collagenase-
selective fluorinated inhibitor 1a (MMP-8 selective). With
exception of compound 1a, it was shown that all synthesized
fluorinated hydroxamates are still potent broad-spectrum MMPIs
(IC50 ) 0.5-527 nM). Due to their excellent in Vitro inhibition
potential, especially the 2-fluoroethoxy compounds 1f and 1j
and the 2- and 6-fluoropicolyl derivatives 1k and 1m are suitable
candidates for the setup of radiolabeling procedures. As a
consequence, we successfully established the 18F-radiosyntheses
of [18F]1f and [18F]1j. As expected, biodistribution and me-
tabolism studies of both PET-compatible radioligands indicate
that there was no tissue specific accumulation in WT mice. Low
uptake in normal subjects may be an advantage in the study of
activated and dysregulated MMPs.
Fourth, the synthesized fluorinated compounds (1a, 1c, 1f,
1j, 1k, and 1m) remain potent MMPIs on the basis of the CGS
25966 or CGS 27023A lead structure (IC50 ) 0.5-527 nM).
In this context, they can be classified in an order of increasing
MMP inhibition potencies: core-fluorinated analogues 1a and
1c < 2-fluoroethoxy compounds 1f and 1j ≈ 2- and 6-fluo-
ropicolyl CGS 27023A derivatives 1k and 1m. As a conse-
quence, the potent MMPI 1f was chosen for synthesis of its
18F-labeled counterpart. Additionally, our results suggest com-
pound [18F]1a as a suitable candidate for the selective imaging
of collagenase 2 (MMP-8).
Ex ViVo dissection of WT mice indicates that there was no
tissue retention of radioactivity after i.v. injection of either
[18F]1f or [18F]1j (Figure 3). Levels of radioactivity in the
gastrointestinal tract, liver, and kidney were initially high but
decreased with time after injection. Radioactivity levels in the
heart, lung, and muscle were comparable with those in the blood,
while brain uptake was negligible (Figure 2). Predosing with
unlabeled 1f or 1j had no demonstratable effect on biodistrubtion
and clearance, suggesting there was no specific binding to
MMPs in any of the tissues studied. Indeed, low uptake in
normal subjects may be an advantage in the study of inflam-
mation, tumor cell metastasis, and atherosclerosis.
The examination of metabolites in the plasma of WT mice
showed only one polar metabolite and the corresponding parent
compound [18F]1f 2.5 min p.i. The polar metabolite, which is
presumably not free [18F]F- (see next paragraph), likely occurs
by phase I metabolism via reaction of [18F]1f with oxygenases
(e.g., resulting in O-dealkylation of the parent compound).33 An
additional metabolite of [18F]1f appeared 20 min p.i. in plasma
that is slightly more hydrophilic compared with [18F]1f. This
metabolite is probably generated by phase II metabolism and
conjugation reaction by transferases (e.g., glucuronidation or
glycine conjugation).33 However, the parent compound [18F]1f
is still available 20 min p.i., assuring the putative in ViVo
detection of activated MMPs in diseases associated with MMP
upregulation. Similar to radioligand [18F]1f, a polar metabolite
appeared in the plasma of WT mice 2.5 min p.i. of compound
[18F]1j, which as the parent compound is also detectable. In
contrast to [18F]1f, a slightly more hydrophilic metabolite of
[18F]1j did not occur 20 min p.i., but the parent compound
[18F]1j was still observable, indicating its MMP detection
potential 20 min p.i.
The comparison of the uptake index of the right tibiae to
that of the heart indicated that the in ViVo metabolic formation
of [18F]F- from [18F]1f or [18F]1j was low because the bone
accumulation of radioactivity was negligible for both radioli-
gands. Furthermore, the images of initial small animal PET
experiments with these radioligands gave no evidence of
radioactivity accumulation in the skeleton coming along with
in ViVo formation of [18F]F- (data not shown).
In summary, the here introduced class of hydroxamate-based
MMPI radioligands (e.g., [18F]1f and [18F]1j) could evolve into
an innovative tool to noninvasively image activated MMPs in
ViVo in diseases such as inflammation, cancer, or atherosclerosis.
Experimental Section
General Methods. All chemicals, reagents, and solvents for the
syntheses were analytical grade and purchased from commercial
sources. The melting points (uncorrected) were determined on a
Stuart Scientific SMP3 capillary melting point apparatus. 1H NMR
and 13C NMR spectra were recorded on Bruker ARX 300 and AMX
400 spectrometers, respectively. Chemical shifts are reported in parts
per million (ppm, δ) relative to tetramethylsilane. Mass spectrometry
was performed using a Varian MAT 212 (EI ) 70 eV) spectrometer
and a Bruker MALDI-TOF-MS Reflex IV (matrix: DHB)
instrument. Exact mass analyses were conducted on a Bruker
MicroTof apparatus. Elemental analyses were realized by using a
Vario EL III analyzer. Radiosyntheses were partly carried out using
an automated PET tracer synthesizer (TRACERLab FxFDG Syn-
thesizer; GE Functional Imaging GmbH). The recorded data were
processed by the TRACERLab Fx software (GE Functional Imaging
GmbH). Separation and purification of the radiosynthesized com-
pounds were performed by gradient radio-HPLC using a Knauer
K-500 and a Latek P 402 pump, a Knauer K-2000 UV-detector (λ
) 254 nm), a Crismatec NaI(Tl) Scintibloc 51 SP51 γ-detector,
and a Nucleosil 100-10 C18 column (250 × 8 mm2). Sample
injection was carried out using a Rheodyne injector block (type
7125 incl. 1000 µL loop). The recorded data were processed by
the NINA version 4.9 software (GE Functional Imaging GmbH).
The radiochemical purities and the specific activities were acquired
with a radio-HPLC system composed of a Syknm S1021 pump, a
Knauer K-2501 UV-detector (λ ) 254 nm), a Crismatec NaI(Tl)
Scintibloc 51 SP51 γ-detector, a Nucleosil 100-3 C18 column (200
× 3 mm2), a VICI injector block (type C1 incl. 20 µL loop), and
the NINA version 4.8, revision 4 software (GE Functional Imaging
GmbH). Unless specified, the coupling constant J represents H,H-
couplings.
General Procedure for the Synthesis of the (R)-2-(N-Benzyl-/
picolyl-arylsulfonamido)-methylbutanoic Acids (5a-5g and 5i-
5l). A solution of the corresponding ester 4a-4k in dichloroethane
was cooled to 0 to -10 °C. Hydrochloric acid was bubbled through
the solution for 2-4 h. The reaction vessel was sealed, and the
mixture was warmed to RT and stirred overnight. Removal of the
solvent in Vacuo as well as subsequent treatment with diethyl ether
(or diisopropyl ether) provided the colorless solid products (5b-
5e). Silica gel column chromatography using cyclohexane/ethyl
acetate mixtures (2:1 (5g and 5l); 3:2 (5a, 5f, 5i, 5j, 5k)) gave the
pure products 5a, 5i, 5j, and 5l as colorless solids, except for 5f
and mixture 5g and 5k which were obtained as viscous yellow oils.
Significance
The aim of the present study was the development of
radioligands for the molecular imaging of activated MMPs in
ViVo by means of PET. Therefore, the hydroxamate-based MMPI
lead structures CGS 25966 and CGS 27023A were systemati-
cally modified by introduction of fluorine, resulting in six
nonradioactive reference compounds of corresponding potential
18F-labeled PET tracers and nine radiolabeling precursors of
(R)-2-(N-Benzyl-4-fluorophenylsulfonamido)-3-methylbutano-
ic Acid (5a). Yield: 49%. mp: 135-136 °C. 1H NMR (300 MHz,
CDCl3) δ 7.72 (dd, 3J ) 9.1 Hz, 4JH,F ) 5.1 Hz, 2H, HAryl), 7.32-

5760 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Wagner et al.
3
3
2
7.21 (m, 5H, HAryl), 7.05 (dd, J ) JH,F ) 9 Hz, 2H, HAryl), 4.56
67.80 (| JC,F| ) 20.2 Hz), 66.54, 49.38, 28.72, 20.02. 19F NMR
(282 MHz, CDCl3) δ -223.80. MS (ESI-EM) m/e: 432.1235 (M
+ Na)+ calcd for C20H24FNO5SNa 432.1251.
2
3
(AB system, J ) 15.6 Hz, 2H, CH2), 4.20 (d, J ) 10.5 Hz, 1H,
N-CH), 2.07-1.95 (m, 1H, CH(CH3)2), 0.91 (d, 3J ) 6.6 Hz, 3H,
CH(CH3)2), 0.82 (d, 3J ) 6.6 Hz, 3H, CH(CH3)2). 13C NMR (75.5
(R)-2-(N-Benzyl-4-(2-bromoethoxy)phenylsulfonamido)-3-
methylbutanoic Acid and (R)-2-(N-Benzyl-4-(2-chloroethoxy)-
phenylsulfonamido)-3-methylbutanoic Acid (5g) (Mixture). H
1
MHz, CDCl3) δ 176.10, 165.17 (d, | JC,F| ) 264.1 Hz), 136.52,
2
1
130.49, 130.37, 129.17, 128.57, 127.95, 116.11 (d, | JC,F| ) 25.5
Hz), 66.55, 49.34, 28.52, 19.81, 19.75. 19F NMR (282 MHz, CDCl3)
δ -105.36. MS (ESI-EM) m/e: 388.0973 (M + Na)+ calcd for
C18H20FNO4SNa 388.0989. Anal. (C18H20FNO4S) C, H, N.
(R)-2-(N-Benzyl-4-(dimethylamino)phenylsulfonamido)-3-
methylbutanoic Acid Hydrochloride (5b). Yield: 86%. mp:
166-168 °C. 1H NMR (300 MHz, DMSO-d6) δ 7.54 (d, 3J ) 9.1
Hz, 2H, HAryl), 7.38-7.36 (m, 2H, HAryl), 7.30-7.21 (m, 3H, HAryl),
3
NMR (300 MHz, DMSO-d6) δ 7.69 (d, J ) 8.8 Hz, 2H, HAryl),
7.38-7.36 (m, 2H, HAryl), 7.31-7.23 (m, 3H, HAryl), 7.06 (d, 3J )
8.8 Hz, 2H, HAryl), 4.55 (AB system, 2J ) 16.3 Hz, 2H, CH2), 4.43-
3.81 (m, 5H, O-CH2CH2X and N-CH), 1.87-1.75 (m, 1H,
3
3
CH(CH3)2), 0.78 (d, J ) 6.9 Hz, 3H, CH(CH3)2), 0.64 (d, J )
6.9 Hz, 3H, CH(CH3)2). 13C NMR (75.5 MHz, DMSO-d6) δ 171.64,
161.50, 138.28, 132.02, 129.89, 128.70, 128.37, 127.52, 115.03,
68.69, 66.44, 48.66, 43.23, 41.88, 31.46, 28.41, 19.86, 19.68.
(R)-2-(N-((2-Fluoropyridin-3-yl)methyl)-4-methoxyphenylsul-
fonamido)-3-methylbutanoic Acid (5i). Yield: 72%. mp: 129-
131 °C. 1H NMR (300 MHz, CDCl3) δ 8.20-8.06 (m, 2H, HAryl),
3
2
6.77 (d, J ) 9.1 Hz, 2H, HAryl), 4.51 (AB system, J ) 16.4 Hz,
2H, CH2), 3.91 (d, 3J ) 10.1 Hz, 1H, N-CH), 2.98 (s, 6H,
3
N(CH3)2), 1.84-1.72 (m, 1H, CH(CH3)2), 0.76 (d, J ) 6.5 Hz,
3H, CH(CH3)2), 0.63 (d, J ) 6.5 Hz, 3H, CH(CH3)2). 13C NMR
3
3
(75.5 MHz, DMSO-d6) δ 171.45, 152.28, 138.37, 128.74, 128.26,
127.93, 126.99, 125.38, 111.28, 65.94, 48.07, 39.90, 28.15, 19.59,
19.33. MS (MALDI-TOF) m/e: 413 [M + Na]+, 391 [M + H]+.
Anal. (C20H26N2O4S‚1.0HCl) C, H, N.
7.72 (d, J ) 9.1 Hz, 2H, HAryl), 7.22-7.17 (m, 1H, HAryl), 6.91
3
2
(d, J ) 9.1 Hz, 2H, HAryl), 4.65 (AB system, J ) 17.2 Hz, 2H,
3
CH2), 4.17 (d, J ) 10.1 Hz, 1H, N-CH), 3.83 (s, 3H, OCH3),
3
2.05-1.93 (m, 1H, CH(CH3)2), 0.94 (d, J ) 6.5 Hz, 3H, CH-
3
(R)-2-(4-Fluorophenylsulfonamido-N-(pyridin-3-ylmethyl))-
(CH3)2), 0.82 (d, J ) 6.5 Hz, 3H, CH(CH3)2). 13C NMR (75.5
1
1
3-methylbutanoic Acid (5c). Yield: 79%. mp: 199-200 °C. H
MHz, CDCl3) δ 174.72, 163.31, 160.74 (d, | JC,F| ) 234.4 Hz),
2
NMR (300 MHz, DMSO-d6) δ 8.88-8.79 (m, 2H, HAryl), 8.51 (d,
146.25, 142.31, 130.75, 129.83, 121.68, 118.97 (d, | JC,F| ) 27.1
3J ) 8.6 Hz, 1H, HAryl), 8.00 (dd, J ) 8.2 Hz, J ) 5.6 Hz, 1H,
Hz), 114.21, 66.06, 55.75, 41.97, 28.55, 19.95, 19.52. 19F NMR
(282 MHz, CDCl3) δ -72.96. MS (ESI-EM) m/e: 419.1045 (M +
Na)+ calcd for C18H21FN2O5SNa 419.1047. Anal. (C18H21FN2O5S)
C, H, N.
3
3
H
Aryl), 7.92 (dd, 3J ) 9.1 Hz, 4JH,F ) 5.2 Hz, 2H, HAryl), 7.42 (dd,
3 2
3J ) JH,F ) 9.0 Hz, 2H, HAryl), 4.82 (AB system, J ) 17.4 Hz,
3
2H, CH2), 3.99 (d, J ) 9.9 Hz, 1H, N-CH), 2.07-1.93 (m, 1H,
3
3
CH(CH3)2), 0.86 (d, J ) 6.7 Hz, 3H, CH(CH3)2), 0.66 (d, J )
(R)-2-(N-((2-Bromopyridin-3-yl)methyl)-4-methoxyphenylsul-
fonamido)-3-methylbutanoic Acid Hydrochloride (5j). Yield:
6.7 Hz, 3H, CH(CH3)2). 13C NMR (75.5 MHz, DMSO-d6) δ 171.54,
1
1
165.11 (d, | JC,F| ) 253.3 Hz), 144.76, 141.84, 141.37, 138.65,
65%. mp: 169-171 °C. H NMR (300 MHz, CDCl3) δ 11.34 (s,
2
3
3
135.21, 131.02, 126.77, 116.82 (d, | JC,F| ) 23.0 Hz), 66.35, 45.88,
2H), 8.23 (d, J ) 4.9 Hz, 1H, HAryl), 8.13 (d, J ) 7.7 Hz, 1H,
28.31, 19.85, 19.48. 19F NMR (282 MHz, DMSO-d6) δ -100.58.
MS (MALDI-TOF) m/e: 367 [M + H]+. Anal. (C17H19FN2O4S)
C, H, N.
3
3
3
H
Aryl), 7.64 (d, J ) 9.0 Hz, 2H, HAryl), 7.34 (dd, J ) 7.7 Hz, J
3
) 4.9 Hz, 1H, HAryl), 6.85 (d, J ) 8.9 Hz, 2H, HAryl), 4.63 (AB
system, 2J ) 18.7 Hz, 2H, CH2), 4.01 (d, 3J ) 9.2 Hz, 1H, N-CH),
3.75 (s, 3H, OCH3), 1.80-1.73 (m, 1H, CH(CH3)2), 0.81 (d, 3J )
(R)-2-(4-(Dimethylamino)phenylsulfonamido-N-(pyridin-3-yl-
methyl))-3-methylbutanoic Acid Hydrochloride (5d). Yield:
99%. mp: 146-152 °C. 1H NMR (300 MHz, DMSO-d6) δ 9.30-
6.8 Hz, 3H, CH(CH3)2), 0.70 (d, 3J ) 6.8 Hz, 3H, CH(CH3)2). 13
C
NMR (75.5 MHz, CDCl3) δ 171.81, 163.06, 146.40, 140.59, 138.69,
136.62, 129.99, 129.72, 123.16, 114.07, 65.82, 55.50, 47.78, 28.77,
19.97, 19.20. MS (ESI-EM) m/e: 479.0233 (M + Na)+ calcd for
C18H21BrN2O5SNa 479.0247. Anal. (C18H21BrN2O5S‚1.4HCl) C, H,
N.
3
3
9.23 (m, 2H, HAryl), 8.99 (d, J ) 8.3 Hz, 1H, HAryl), 8.48 (dd, J
) 8.2 Hz, 3J ) 5.7 Hz, 1H, HAryl), 7.80 (d, 3J ) 9.2 Hz, 2H, HAryl),
2
2
6.98 (d, J ) 9.2 Hz, 2H, HAryl), 5.02 (AB system, J ) 17.5 Hz,
2H, CH2), 4.20 (d, 3J ) 9.9 Hz, 1H, N-CH), 3.25 (s, 6H, N(CH3)2),
3
2.27-2.15 (m, 1H, CH(CH3)2), 1.10 (d, J ) 6.6 Hz, 3H, CH-
(R)-2-(N-((6-Fluoropyridin-3-yl)methyl)-4-methoxyphenyl-
sulfonamido)-3-methylbutanoic Acid Hydrochloride (5k).
(CH3)2), 0.93 (d, J ) 6.6 Hz, 3H, CH(CH3)2). 13C NMR (75.5
3
1
MHz, DMSO-d6) δ 171.94, 153.22, 145.44, 140.42, 139.94, 139.65,
129.26, 126.84, 123.52, 111.31, 65.77, 55.03, 45.20, 28.13, 19.81,
19.34. MS (MALDI-TOF) m/e: 414 [M - HCl + Na]+, 392 [M
- HCl + H]+. Anal. (C19H25N3O4S‚2.3HCl) C, H, N.
Yield: 76%. H NMR (300 MHz, CDCl3) δ 8.72 (s, br, COOH),
8.12 (s, 1H, HAryl), 8.01-7.59 (m, 1H, HAryl), 7.66 (d, 3J ) 8.8 Hz,
3
2H, HAryl), 6.88-6.84 (m, 1H, HAryl), 6.87 (d, J ) 8.8 Hz, 2H,
2
3
H
Aryl), 4.60 (AB system, J ) 15.9 Hz, 2H, CH2), 4.21 (d, J )
10.6 Hz, 1H, N-CH), 3.82 (s, 3H, OCH3), 2.00-1.93 (m, 1H,
(R)-2-(4-Nitrophenylsulfonamido)-N-(pyridin-3-ylmethyl))-3-
methylbutanoic Acid Hydrochloride (5e). Yield: 89%. mp: 200-
201 °C. 1H NMR (300 MHz, DMSO-d6) δ 8.83 (s, 2H, HAryl), 8.48
3
3
CH(CH3)2), 0.94 (d, J ) 6.9 Hz, 3H, CH(CH3)2), 0.82 (d, J )
6.9 Hz, 3H, CH(CH3)2). 13C NMR (75.5 MHz, CDCl3) δ 173.74,
1
3
3
163.13, 162.96 (d, | JC,F| ) 242.1 Hz), 147.27, 142.97, 131.56,
(d, J ) 8.3 Hz, 1H, HAryl), 8.38 (d, J ) 8.9 Hz, 2H, HAryl), 8.11
2
2
3
3
131.25, 129.67, 114.14, 109.39 (d, | JC,F| ) 36.3 Hz), 66.18, 55.74,
(d, J ) 8.9 Hz, 2H, HAryl), 7.98 (dd, J ) 8.1 Hz, J ) 5.6 Hz,
45.33, 28.65, 19.68, 19.56. 19F NMR (282 MHz, CDCl3) δ -73.53.
MS (ESI-EM) m/e: 419.1040 (M + Na)+ calcd for C18H21FN2O5-
SNa 419.1047. Anal. (C18H21FN2O5S‚0.1HCl) C, H, N.
3
1H, HAryl), 4.84 (s, 2H, CH2), 4.05 (d, J ) 9.9 Hz, 1H, N-CH),
3
2.19-1.97 (m, 1H, CH(CH3)2), 0.88 (d, J ) 6.6 Hz, 3H, CH-
(CH3)2), 0.67 (d, J ) 6.6 Hz, 3H, CH(CH3)2). 13C NMR (75.5
3
MHz, DMSO-d6) δ 171.23, 150.43, 144.90, 144.12, 141.89, 141.52,
138.14, 129.52, 126.88, 124.85, 66.63, 46.19, 28.30, 19.85, 19.49.
MS (ESI-EM) m/e: 392.0938 (M - H)- calcd for C17H18N3O6S
392.0922.
(R)-2-(N-((6-Bromopyridin-3-yl)methyl)-4-methoxyphenyl-
sulfonamido)-3-methylbutanoic Acid (5l). Yield: 71%. mp: 73-
1
74 °C. H NMR (300 MHz, CDCl3) δ 9.84 (s, 2H), 8.27 (s, 1H,
H
Aryl), 7.79-7.75 (m, 1H, HAryl), 7.60 (d, 3J ) 9.0 Hz, 2H, HAryl),
3 3
7.41 (d, J ) 8.1 Hz, 1H, HAryl), 6.84 (d, J ) 9.3 Hz, 2H, HAryl),
4.59 (AB system, 2J ) 16.1 Hz, 2H, CH2), 4.27 (d, 3J ) 10.6 Hz,
1H, N-CH), 3.82 (s, 3H, OCH3), 2.03-1.92 (m, 1H, CH(CH3)2),
(R)-2-(N-Benzyl-4-(2-fluoroethoxy)phenylsulfonamido)-3-
methylbutanoic Acid (5f). Yield: 52%. 1H NMR (300 MHz,
CDCl3) δ 8.91 (s, br, 1H, COOH), 7.69 (d, 3J ) 8.8 Hz, 2H, HAryl),
7.36-7.33 (m, 2H, HAryl), 7.27-7.23 (m, 3H, HAryl), 6.90 (d, 3J )
3
3
0.97 (d, J ) 6.4 Hz, 3H, CH(CH3)2), 0.87 (d, J ) 6.4 Hz, 3H,
CH(CH3)2). 13C NMR (75.5 MHz, CDCl3) δ 173.78, 163.01, 149.81,
140.59, 139.92, 133.06, 131.55, 129.50, 127.93, 114.17, 65.95,
55.67, 45.38, 28.69, 19.67, 19.09. MS (ESI-EM) m/e: 479.0239
(M + Na)+ calcd for C18H21BrN2O5SNa 479.0247. Anal. (C18H21-
BrN2O5S) C, H, N.
2
3
8.8 Hz, 2H, HAryl), 4.74 (dt, JH,F ) 47.2 Hz, J ) 7.8 Hz, 2H,
2
O-CH2CH2F), 4.55 (AB system, J ) 15.9 Hz, 2H, CH2), 4.22
3
3
3
(dt, JH,F ) 28.2 Hz, J ) 7.8 Hz, 2H, O-CH2CH2F), 4.09 (d, J
) 10.3 Hz, 1H, N-CH), 2.04-1.92 (m, 1H, CH(CH3)2), 0.89 (d,
3J ) 6.5 Hz, 3H, CH(CH3)2), 0.77 (d, 3J ) 6.9 Hz, 3H, CH(CH3)2).
13C NMR (75.5 MHz, CDCl3) δ 175.75, 162.05, 137.22, 132.65,
(R)-2-(N-Benzyl-4-(2-iodoethoxy)phenylsulfonamido)-3-methyl-
butanoic Acid (5h). An amount of 1.60 g (10.6 mmol) of anhydrous
1
130.16, 129.29, 128.63, 128.01, 114.85, 81.95 (| JC,F| ) 170.6 Hz),

Broad-Spectrum MMPIs for MMP Imaging with PET
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5761
sodium iodide (dried at 100 °C in Vacuo for 3 h) and 600 mg of
compound mixture 5g in 40 mL of 2-butanone were heated to reflux
for 6 days. After cooling to ambient temperature, the solvent was
evaporated and the residue was subjected to silica gel column
chromatography (cyclohexane/ethyl acetate 3:2), affording 370 mg
(0.72 mmol) of pure compound 5h as a yellow oil, which solidified
m/e: 404 [M + Na]+, 382 [M + H]+. Anal. (C17H20FN3O4S‚
1.3HCl) C, H, N.
(R)-2-(4-(Dimethylamino)phenylsulfonamido-N-(pyridin-3-yl-
methyl))-N-hydroxy-3-methylbutanamide (1d). Yield: 87%.
1
3
mp: 147-151 °C. H NMR (300 MHz, DMSO-d6) δ 8.77 (d, J
) 5.7 Hz, 1H, HAryl), 8.74 (s, 1H, HAryl), 8.46 (d, 3J ) 8.2 Hz, 1H,
Aryl), 7.98 (dd, J ) 7.9 Hz, J ) 5.7 Hz, 1H, HAryl), 7.51 (d, J
) 8.9 Hz, 2H, HAryl), 6.69 (d, J ) 8.9 Hz, 2H, HAryl), 4.78 (AB
system, 2J ) 17.3 Hz, 2H, CH2), 3.83 (d, 3J ) 10.7 Hz, 1H,
N-CH), 2.98 (s, 6H, N(CH3)2), 2.02-1.90 (m, 1H, CH(CH3)2),
1
3
3
3
on standing. mp: 57-59 °C. H NMR (400 MHz, CDCl3) δ 7.68
H
3
3
(d, J ) 8.7 Hz, 2H, HAryl), 7.36-7.32 (m, 2H, HAryl), 7.27-7.23
(m, 3H, HAryl), 6.86 (d, 3J ) 8.7 Hz, 2H, HAryl), 4.55 (AB system,
3
2J ) 15.8 Hz, 2H, CH2), 4.25 (t, J ) 6.9 Hz, 2H, O-CH2CH2I),
3
3
3
3
4.16 (d, J ) 10.7 Hz, 1H, N-CH), 3.39 (t, J ) 6.9 Hz, 2H,
0.77 (d, J ) 6.3 Hz, 3H, CH(CH3)2), 0.59 (d, J ) 6.3 Hz, 3H,
CH(CH3)2). 13C NMR (75.5 MHz, DMSO-d6) δ 166.43, 152.76,
145.76, 140.49, 139.70, 139.67, 128.98, 126.82, 124.96, 111.38,
63.42, 52.44, 44.68, 28.51, 19.45, 19.25. MS (ESI-EM) m/e:
429.1553 (M + Na)+ calcd for C19H26N4O4SNa 429.1567.
(R)-2-(4-Nitrophenylsulfonamido-N-(pyridin-3-ylmethyl))-N-
hydroxy-3-methylbutanamide (1e). Yield: 90%. mp: 125-126
3
O-CH2CH2I), 2.04-1.93 (m, 1H, CH(CH3)2), 0.89 (d, J ) 6.5
Hz, 3H, CH(CH3)2), 0.78 (d, J ) 6.9 Hz, 3H, CH(CH3)2). 13C
3
NMR (75.5 MHz, CDCl3) δ 174.96, 161.25, 136.86, 132.54, 129.92,
129.02, 128.31, 127.73, 114.59, 68.86, 66.24, 49.08, 28.44, 19.77,
19.72, 0.21. MS (ESI-EM) m/e: 516.0362 (M - H)+ calcd for
C20H23INO5S 516.0347.
1
General Procedure for the Synthesis of the (R)-2-(N-Benzyl-/
picolyl-4-arylsulfonamido)-N-hydroxy-3-methylbutanamides (1a-
1g and 1k-1n). A solution of the corresponding (R)-2-(N-benzyl-/
picolyl-4-arylsulfonamido)-N-tert-butyloxy-3-methylbutanamides 6a,
6b, 6d, 6e, 6g-6i, and 6l-6o in dichloroethane was cooled to
-10 °C. Hydrochloric acid was bubbled through the solution for
2-4 h. The reaction vessel was sealed, and the mixture was warmed
to ambient temperature and stirred overnight. Removal of the
solvent to 1/3 of its original volume in Vacuo and subsequent
treatment with diisopropyl ether provided the colorless solid
products. Compounds 1a, 1f, and 1k-1n were purified by silica
gel column chromatography eluting with cyclohexane/ethyl acetate
mixtures (3:2) (1a and 1f), (1:1) (1m), (1:2) (1k, 1l, and 1n),
yielding the colorless solid products.
°C. H NMR (300 MHz, DMSO-d6) δ 11.37 (s, 1H, OH), 8.73-
8.68 (m, 2H, HAryl), 8.31 (d, 3J ) 9.0 Hz, 1H, HAryl), 8.24 (d, 3J )
8.3 Hz, 2H, HAryl), 8.02 (d, 3J ) 9.0 Hz, 1H, HAryl), 7.77 (dd, 3J )
3
2
8.0 Hz, J ) 5.5 Hz, 2H, HAryl), 4.93 (AB system, J ) 16.3 Hz,
2H, CH2), 3.97 (d, 3J ) 10.6 Hz, 1H, N-CH), 2.10-1.99 (m, 1H,
3
3
CH(CH3)2), 0.83 (d, J ) 6.5 Hz, 3H, CH(CH3)2), 0.72 (d, J )
6.5 Hz, 3H, CH(CH3)2). 13C NMR (75.5 MHz, DMSO-d6) δ 174.26,
165.27, 149.69, 144.80, 143.60, 136.40, 128.54, 125.60, 124.35,
115.46, 63.29, 48.50, 27.78, 19.19, 19.75. MS (EI): m/e (intensity
%): 408 (M+, 6), 348 (4), 250 (6), 235 (16), 207 (11), 125 (100),
82 (59), 55 (30). Anal. (C17H20N4O6S) C, H, N.
(R)-2-(N-Benzyl-4-(2-fluoroethoxy)phenylsulfonamido)-N-hy-
droxy-3-methylbutanamide (1f). Yield: 27%. mp: 144-146 °C.
1H NMR (400 MHz, CDCl3) δ 7.60 (m, 2H, HAryl), 7.37-7.35 (m,
2H, HAryl), 7.27-7.24 (m, 3H, HAryl), 6.89 (m, 2H, HAryl), 4.76 (d,
(R)-2-(N-Benzyl-4-fluorophenylsulfonamido)-N-hydroxy-3-
methylbutanamide (1a). A solid-phase synthesis of the (S)-
enantiomer is described in ref 34. Yield: 31%. mp: 166-167 °C.
1H NMR (300 MHz, CDCl3) δ 10.60 (s, 1H, OH), 8.36 (s, 1H,
NH), 7.60 (dd, 3J ) 8.1 Hz, 4JH,F ) 5.1 Hz, 2H, HAryl), 7.34-7.17
2
2JH,F ) 49.1 Hz, 2H, O-CH2CH2F), 4.60 (AB system, J ) 15.6
3
Hz, 2H, CH2), 4.23 (d, JH,F ) 27.3 Hz, 2H, O-CH2CH2F), 3.79
3
(m, 1H, N-CH), 2.23-2.14 (m, 1H, CH-CH(CH3)2), 0.81 (d, J
3
) 6.5 Hz, 3H, CH(CH3)2), 0.55 (d, J ) 6.9 Hz, 3H, CH(CH3)2).
3
3
(m, 5H, HAryl), 6.97 (dd, J ) JH,F ) 9 Hz, 2H, HAryl), 4.71 (s,
13C NMR (100 MHz, CDCl3) δ 167.15, 161.69, 136.56, 132.33,
2H, CH2), 3.97 (d, 3J ) 10.8 Hz, 1H, N-CH), 2.21-2.09 (m, 1H,
129.35, 129.03, 128.10, 127.48, 114.63, 82.20 (| JC,F| ) 172.7 Hz),
1
3
3
2
CH(CH3)2), 0.85 (d, J ) 6.5 Hz, 3H, CH(CH3)2), 0.77 (d, J )
67.33 (| JC,F| ) 20.4 Hz), 63.72, 48.57, 27.20, 19.57, 19.08. 19F
6.5 Hz, 3H, CH(CH3)2). 13C NMR (75.5 MHz, CDCl3) δ 166.50,
NMR (282 MHz, CDCl3) δ -224.01. MS (ESI-EM) m/e: 447.1363
(M + Na)+ calcd for C20H25FN2O5SNa 447.1360. Anal. (C20H25-
FN2O5S) C, H, N.
1
164.65 (d, | JC,F| ) 254.0 Hz), 136.84, 136.63, 129.89, 129.37,
2
128.03, 127.38, 115.82 (d, | JC,F| ) 22.7 Hz), 64.01, 48.60, 28.05,
19.61, 19.61. 19F NMR (282 MHz, CDCl3) δ -106.20. MS (ESI-
EM) m/e: 403.1094 (M + Na)+ calcd for C18H21FN2O4SNa
403.1098. Anal. (C18H21FN2O4S) C, H, N.
(R)-2-(N-Benzyl-4-(2-bromoethoxy)phenylsulfonamido)-N-hy-
droxy-3-methylbutanamide and (R)-2-(N-Benzyl-4-(2-chloro-
ethoxy)phenylsulfonamido)-N-hydroxy-3-methylbutanamide (1g)
1
(Mixture). H NMR (300 MHz, DMSO-d6) δ 10.79 (s, 1H, OH),
(R)-2-(N-Benzyl-4-(dimethylamino)phenylsulfonamido)-N-hy-
droxy-3-methylbutanamide Hydrochloride (1b). Yield: 45%. mp
8.93 (s, 1H, NH), 7.61 (d, 3J ) 8.8 Hz, 2H, HAryl), 7.35-7.32 (m,
3
1
2H, HAryl), 7.25-7.19 (m, 3H, HAryl), 6.97 (d, J ) 8.8 Hz, 2H,
98-101 °C (decomposition). H NMR (300 MHz, DMSO-d6) δ
2
3
H
Aryl), 4.66 (AB system, J ) 15.7 Hz, 2H, CH2), 4.40-3.80 (m,
10.77 (s, 1H, OH), 8.13 (s, 1H, NH), 7.50 (d, J ) 8.2 Hz, 2H,
Aryl), 7.37-7.20 (m, 5H, HAryl), 6.67 (d, 3J ) 8.2 Hz, 2H, HAryl),
5H, O-CH2CH2X and N-CH), 1.93-1.80 (m, 1H, CH(CH3)2),
H
3
3
4.60 (AB system, 2J ) 15.7 Hz, 2H, CH2), 3.79 (d, 3J ) 10.7 Hz,
1H, N-CH), 2.96 (s, 6H, N(CH3)2), 1.85-1.70 (m, 1H, CH(CH3)2),
0.73 (d, J ) 6.5 Hz, 3H, CH(CH3)2), 0.67 (d, J ) 6.5 Hz, 3H,
CH(CH3)2). 13C NMR (75.5 MHz, DMSO-d6) δ 165.75, 160.81,
137.94, 132.49, 129.17, 128.76, 127.87, 127.05, 114.65, 68.14,
63.17, 47.59, 42.94, 28.11, 19.50, 19.09.
3
3
0.71 (d, J ) 6.0 Hz, 3H, CH(CH3)2), 0.62 (d, J ) 6.0 Hz, 3H,
CH(CH3)2). 13C NMR (75.5 MHz, DMSO-d6) δ 175.50, 166.26,
152.58, 138.89, 129.05, 129.95, 128.15, 127.27, 111.44, 63.37,
47.68, 28.53, 19.87, 19.48. (13C NMR signal for N(CH3)2 is not
detectable.) MS (ESI-EM) m/e: 428.1614 (M + Na)+ calcd for
C20H27N3O4SNa 428.1616. Anal. (C20H27N3O4S‚1.1HCl) C, H, N.
(R)-2-(4-Fluorophenylsulfonamido-N-(pyridin-3-ylmethyl))-
N-hydroxy-3-methylbutanamide Hydrochloride (1c). Yield: 82%.
(R)-2-(N-((2-Fluoropyridin-3-yl)methyl)-4-methoxyphenyl-
sulfonamido)-N-hydroxy-3-methylbutanamide (1k). Yield: 28%.
1
mp: 167-169 °C. H NMR (300 MHz, DMSO-d6) δ 10.83 (s,
1H, OH), 8.95 (s, 1H, NH), 8.07 (s, 1H, HAryl), 7.90-7.84 (m, 1H,
H
Aryl), 7.68 (d, 3J ) 9.2 Hz, 2H, HAryl), 7.25-7.20 (m, 1H, HAryl),
3 2
7.00 (d, J ) 9.2 Hz, 2H, HAryl), 4.72 (AB system, J ) 17.1 Hz,
2H, CH2), 3.80 (d, 3J ) 12.1 Hz, 1H, N-CH), 3.82 (s, 3H, OCH3),
1
mp: 144-145 °C. H NMR (300 MHz, DMSO-d6) δ 10.95 (s,
3
1H, OH), 8.77-8.76 (m, 2H, HAryl), 8.40 (d, 3J ) 8.2 Hz, 1H, HAryl),
1.98-1.84 (m, 1H, CH(CH3)2), 0.77 (d, J ) 6.5 Hz, 3H, CH-
(CH3)2), 0.67 (d, J ) 6.5 Hz, 3H, CH(CH3)2). 13C NMR (75.5
3
7.92 (dd, 3J ) 8.1 Hz, 3J ) 5.8 Hz, 1H, HAryl), 7.88 (dd, 3J ) 8.1
1
3
3
3
MHz, DMSO-d6) δ 165.53, 162.47, 160.17 (d, | JC,F| ) 237.0 Hz),
Hz, J ) 5.8 Hz, 2H, HAryl), 7.34 (dd, J ) JH,F ) 9 Hz, 2H,
2
2
3
145.94, 141.34, 130.98, 129.19, 121.69, 120.28 (d, | JC,F| ) 29.3
HAryl), 4.85 (AB system, J ) 16.4 Hz, 2H, CH2), 3.82 (d, J )
Hz), 114.17, 62.65, 55.63, 40.47, 27.80, 19.32, 18.90. 19F NMR
(282 MHz, CDCl3) δ -73.30. MS (MALDI-TOF) m/e: 434 [M +
Na]+, 412 [M + H]+. Anal. (C18H22FN3O5S) C, H, N.
3
10.8 Hz, 1H, N-CH), 2.05-1.92 (m, 1H, CH(CH3)2), 0.80 (d, J
3
) 6.5 Hz, 3H, CH(CH3)2), 0.62 (d, J ) 6.5 Hz, 3H, CH(CH3)2).
13C NMR (75.5 MHz, DMSO-d6) δ 174.93, 164.89 (d, | JC,F| )
1
254.6 Hz), 145.00, 142.08, 141.25, 138.53, 136.11, 130.56, 130.43,
(R)-2-(N-((2-Bromopyridin-3-yl)methyl)-4-methoxyphenylsul-
fonamido)-N-hydroxy-3-methylbutanamide (1l). Yield: 52%.
mp: 174 °C. 1H NMR (300 MHz, DMSO-d6) δ 10.81 (s, 1H, OH),
2
116.78 (d, | JC,F| ) 23.0 Hz), 64.68, 46.05, 27.59, 19.90, 19.59.
19F NMR (282 MHz, DMSO-d6) δ -101.05. MS (MALDI-TOF)

5762 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Wagner et al.
3
3
3
8.90 (s, 1H, NH), 8.16 (d, J ) 4.5 Hz, 1H, HAryl), 7.68 (d, J )
3H, CH(CH3)2), 0.49 (d, J ) 6.5 Hz, 3H, CH(CH3)2). 13C NMR
(75.5 MHz, CDCl3) δ 167.17, 161.36, 145.26, 136.54, 132.61,
132.48, 129.98, 129.41, 129.12, 128.28, 128.03, 127.70, 114.65,
67.62, 65.76, 63.62, 48.63, 27.10, 21.72, 19.77, 19.16. MS (ESI-
EM) m/e: 599.1487 (M + Na)+ calcd for C27H32N2O8S2Na
599.1492. Anal. (C27H32N2O8S2) C, H, N.
3
8.9 Hz, 2H, HAryl), 7.71-7.64 (m, 1H, HAryl), 7.28 (dd, J ) 7.7
Hz, 3J ) 4.5 Hz, 1H, HAryl), 7.00 (d, 3J ) 8.9 Hz, 2H, HAryl), 4.71
2
(AB system, J ) 17.7 Hz, 2H, CH2), 3.78 (s, 3H, OCH3), 3.78-
3
3.75 (m, 1H, N-CH), 1.87-1.75 (m, 1H, CH(CH3)2), 0.71 (d, J
3
) 6.4 Hz, 3H, CH(CH3)2), 0.62 (d, J ) 6.6 Hz, 3H, CH(CH3)2).
13C NMR (75.5 MHz, DMSO-d6) δ 165.38, 162.69, 148.37, 140.89,
138.06, 134.80, 130.61, 129.49, 122.93, 114.32, 62.57, 55.59, 46.80,
27.88, 19.46, 18.92. MS (ESI-EM) m/e: 496.0332 (M + Na)+ calcd
for C18H22BrN3O5SNa 496.0337.
Radiochemistry. Production of [18F]Fluoride and Synthesis of
[18F]K(K 222)F. No-carrier-added aqueous [18F]fluoride was
produced on a RDS 111e cyclotron (CTI-Siemens) by irradiation
of a 1.2 mL water target using 10 MeV proton beams on 97.0%
enriched [18O]water by the 18O(p,n)18F nuclear reaction. Typical
[18F]fluoride ion batches were 9.4 GBq and 4.5 GBq for the two-
step and one-step procedures at end of radionuclide production for
currents of 32 µA and irradiation times of 12 and 4 min,
respectively. To recover the [18O]water, the batch of aqueous
[18F]fluoride was passed through an anion exchange resin (Sep-
Pak Light Waters Accell Plus QMA cartridge, preconditioned with
5 mL of 1 M K2CO3 and 10 mL of water). [18F]Fluoride was eluted
from the resin with a mixture of 40 µL of 1 M K2CO3, 200 µL of
water for injection, and 800 µL of DNA-grade CH3CN containing
18 mg of Kryptofix 2.2.2 (K 222). Subsequently, the aqueous
[18F]K(K 222)F solution was carefully evaporated to dryness in
Vacuo.
(R)-2-(N-((6-Fluoropyridin-3-yl)methyl)-4-methoxyphenyl-
sulfonamido)-N-hydroxy-3-methylbutanamide (1m). Yield: 42%.
mp: 156 °C. 1H NMR (300 MHz, DMSO-d6) δ 10.94 (s, 1H, OH),
9.03 (s, 1H, NH), 8.09 (s, 1H, HAryl), 7.85-7.79 (m, 1H, HAryl),
3
7.55 (d, J ) 8.8 Hz, 2H, HAryl), 7.01-6.95 (m, 1H, HAryl), 6.91
3
2
(d, J ) 8.8 Hz, 2H, HAryl), 4.67 (AB system, J ) 15.8 Hz, 2H,
3
CH2), 3.86 (d, J ) 10.6 Hz, 1H, N-CH), 3.80 (s, 3H, OCH3),
3
2.01-1.89 (m, 1H, CH(CH3)2), 0.88 (d, J ) 6.6 Hz, 3H, CH-
(CH3)2), 0.71 (d, J ) 6.6 Hz, 3H, CH(CH3)2). 13C NMR (75.5
3
1
MHz, DMSO-d6) δ 165.88, 162.23, 162.13 (d, | JC,F| ) 235.6 Hz),
2
146.25, 147.58, 142.39, 131.93, 128.88, 114.02, 108.63 (d, | JC,F
|
) 38.0 Hz), 62.96, 55.62, 44.18, 27.87, 19.21, 18.93. 19F NMR
(282 MHz, CDCl3) δ -71.15. MS (ESI-EM) m/e: 434.1154 (M +
Na)+ calcd for C18H22FN3O5SNa 434.1156. Anal. (C18H22FN3O5S)
C, H, N.
(R)-2-(N-Benzyl-4-(2-[18F]fluoroethoxy)phenylsulfonamido)-
N-hydroxy-3-methylbutanamide ([18F]1f) (Two-Step Procedure).
An amount of 5.7 mg (9.0 µmol) of compound 6k and the carefully
dried [18F]K(K 222)F residue were heated at 84 °C in 1 mL of
DNA-grade acetonitrile for 4 min. The mixture was cooled to
40 °C, diluted with 10 mL of water for injection, and passed through
a Waters Sep-Pak C18 cartridge. The cartridge was washed with
10 mL of water for injection and eluted with 4 mL of acetonitrile.
Subsequently, the eluate was evaporated to dryness. Compound
[18F]6h was obtained in radiochemical yields of 39.5 ( 5.1%
(decay-corrected, n ) 5). An amount of 1 mL of trifluoroacetic
acid (TFA) taken from a freshly opened ampule was then added.
The reaction mixture was vortexed and refluxed for 20 min. TFA
was evaporated, 350 µL of acetonitrile followed by 650 µL of water
was added, and the resulting solution was purified using a gradient
radio-HPLC procedure (conditions: λ ) 254 nm; flow ) 4 mL/
min; eluents ) A: CH3CN/H2O/TFA, 950/50/1 and B: CH3CN/
H2O/TFA, 50/950/1; time program ) eluent B from 70% to 30%
in 30 min, halt 5 min, from 30% to 70% in 5 min). The product
fraction of compound [18F]1f (retention time tR ) 15.3 min) was
evaporated to dryness in Vacuo and redissolved with 1 mL of H2O/
EtOH (9:1). The second step provided the target compound
[18F]1f in radiochemical yields of 31.4 ( 6.5% (decay-corrected,
n ) 5). Product compound [18F]1f was obtained in an overall
radiochemical yield (two steps including synthesis of [18F]K(K
222)F) of 12.4 ( 3.0% (decay-corrected) and a radiochemical purity
of 97.3 ( 1.6% after 113 ( 6 min from end of radionuclide
production. The determined specific radioactivity was 44 ( 19 GBq/
µmol (32-72 GBq/µmol) at the EOS (n ) 5). The specific activity
of product [18F]1f was estimated by comparing the peak area of
purified [18F]1f in the UV channel (retention time tR ) 11.8 min)
with a standard curve of known concentrations of reference
compound 1f measured with a RP-HPLC system (conditions: λ )
254 nm; flow ) 0.3 mL/min; eluent ) CH3CN/H2O/TFA, 450/
550/1). The chemical identity of [18F]1f was proved by coinjection
and coelution of [18F]1f and the nonradioactive counterpart 1f on
the same HPLC system.
(R)-2-(N-((6-Bromopyridin-3-yl)methyl)-4-methoxyphenyl-
sulfonamido)-N-hydroxy-3-methylbutanamide (1n). Yield: 47%.
1
mp: 168-172 °C. H NMR (300 MHz, DMSO-d6) δ 10.92 (s,
1H, OH), 9.03 (s, 1H, NH), 8.22 (s, 1H, HAryl), 7.60-7.54 (m, 1H,
3
3
H
Aryl), 7.55 (d, J ) 9.0 Hz, 2H, HAryl), 7.43 (d, J ) 8.2 Hz, 1H,
3 2
H
Aryl), 7.68 (d, J ) 9.0 Hz, 2H, HAryl), 4.63 (AB system, J )
3
16.0 Hz, 2H, CH2), 3.85 (d, J ) 10.8 Hz, 1H, N-CH), 3.81 (s,
3H, OCH3), 2.00-1.88 (m, 1H, CH(CH3)2), 0.78 (d, 3J ) 6.5 Hz,
3H, CH(CH3)2), 0.70 (d, J ) 6.5 Hz, 3H, CH(CH3)2). 13C NMR
3
(75.5 MHz, DMSO-d6) δ 165.86, 162.29, 150.42, 139.80, 139.63,
133.38, 131.79, 128.92, 127.16, 114.05, 62.93, 55.63, 44.36, 27.86,
19.23, 18.92. MS (ESI-EM) m/e: 494.0361 (M + Na)+ calcd for
C18H22BrN3O5SNa 494.0356. Anal. (C18H22BrN3O5S) C, H, N.
(R)-2-(N-Benzyl-4-(2-iodoethoxy)phenylsulfonamido)-N-
hydroxy-3-methylbutanamide (1h). An amout of 794 mg (5.58
mmol) of anhydrous sodium iodide (dried at 100 °C in Vacuo for
3 h) and 1.20 g of compound mixture 1g in 30 mL of 2-butanone
were heated to reflux overnight. After cooling to room temperature,
the solvent was evaporated and the residue was subjected to silica
gel column chromatography (cyclohexane/ethyl acetate 2:1), af-
fording 550 mg (1.03 mmol) of compound 1h as a colorless foam.
1
mp: 140-142 °C. H NMR (300 MHz, CDCl3) δ 9.19 (s, 1H,
3
OH), 7.58 (d, J ) 9.0 Hz, 2H, HAryl), 7.36-7.32 (m, 2H, HAryl),
3
7.27-7.23 (m, 3H, HAryl), 6.87 (d, J ) 9.0 Hz, 2H, HAryl), 4.57
2
3
(AB system, J ) 16.1 Hz, 2H, CH2), 4.27 (t, J ) 6.9 Hz, 2H,
3
3
O-CH2CH2I), 3.74 (d, J ) 10.7 Hz, 1H, N-CH), 3.42 (t, J )
6.9 Hz, 2H, O-CH2CH2I), 2.25-2.12 (m, 1H, CH(CH3)2), 0.80
3
3
(d, J ) 6.4 Hz, 3H, CH(CH3)2), 0.52 (d, J ) 6.4 Hz, 3H, CH-
(CH3)2). 13C NMR (75.5 MHz, CDCl3) δ 167.32, 161.25, 136.55,
132.43, 129.53, 129.18, 128.30, 127.70, 114.83, 68.81, 63.72, 48.67,
27.19, 19.80, 19.21, 0.08. MS (ESI-EM) m/e: 555.0404 (M + Na)+
calcd for C20H25IN2O5SNa 555.0421.
(R)-2-(N-Benzyl-4-(2-(tosyloxy)ethoxy)phenylsulfonamido)-N-
hydroxy-3-methylbutanamide (1i). An amount of 2.80 g (10.0
mmol) of silver tosylate and 550 mg of compound 1h in 60 mL of
acetonitrile were heated to reflux overnight. After cooling to ambient
temperature, the solvent was evaporated and the residue was
subjected to silica gel column chromatography (cyclohexane/ethyl
acetate 1:1), affording 120 mg (0.208 mmol) of compound 1i as a
(R)-2-(N-Benzyl-4-(2-[18F]fluoroethoxy)phenylsulfonamido)-
N-hydroxy-3-methylbutanamide ([18F]1f) (One-Step Procedure).
An amount of 5.8 mg (10.0 µmol) of compound 1i and the carefully
dried [18F]K(K 222)F residue were heated at 84 °C in 1 mL of
DNA-grade CH3CN for 20 min. The mixture was cooled to 40 °C,
diluted with 10 mL of water for injection, and passed through a
Waters Sep-Pak Light C18 cartridge. The cartridge was washed
with 10 mL of water for injection and eluted with 0.4 mL of warm
dimethylformamide (DMF) (warmed to 120 °C before elution). The
eluate was diluted with 0.5 mL of water for injection and purified
by HPLC (conditions: see two-step procedure). The product fraction
1
colorless foam. Yield: 21%. mp: 101-104 °C. H NMR (300
MHz, CDCl3) δ 9.13 (s, 1H, NH), 7.84 (d, 3J ) 8.3 Hz, 2H, HAryl),
3
7.55 (d, J ) 8.9 Hz, 2H, HAryl), 7.37-7.25 (m, 7H, HAryl), 6.78
3
2
(d, J ) 8.9 Hz, 2H, HAryl), 4.57 (AB system, J ) 15.3 Hz, 2H,
CH2), 4.39-4.36 (m, 2H, O-CH2CH2OTs), 4.20-4.17 (m, 2H,
O-CH2CH2OTs), 3.70 (d, 3J ) 10.8 Hz, 1H, N-CH), 2.46 (s, 3H,
C6H4-CH3), 2.25-2.13 (m, 1H, CH(CH3)2), 0.80 (d, 3J ) 6.5 Hz,

Broad-Spectrum MMPIs for MMP Imaging with PET
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23 5763
of compound [18F]1f was evaporated to dryness in Vacuo and
redissolved with 1 mL of H2O/EtOH (9:1). The radiosynthesis
provided [18F]1f with radiochemical yields of 45.6 ( 5.6% (decay-
corrected, n ) 5), a radiochemical purity of 95.5 ( 3.1% in 115 (
10 min from end of radionuclide production, and a calculated
specific radioactivity of 60 ( 25 GBq/µmol (42-97 GBq/µmol)
at the EOS (n ) 5).
Measurement of log D Values. The determination of the log D
values of [18F]1f and [18F]1j was performed in 1-octanol and 0.02
M phosphate buffer at pH 7.4 as described previously.35,36 The
layers were counted in an automated gamma counter (Wallac
Wizard 3′′, Perkin-Elmer Life Sciences, Boston, MA), back
correcting for decay. The measurement was repeated four times.
The log D value was determined by calculating the ratio of 1-octanol
to that of buffer.
Metabolism. Adult C57BL/6 mice (male and female, 25-30 g)
were prepared in accordance to the section on biodistribution (see
above). Compounds [18F]1f or [18F]1j each were injected as a bolus
(100 µL) via the tail vein. Aliquots of each injectant were diluted
in saline and measured to determine the radioactivity injected into
each animal. The animals were sacrificed at 2.5 or 20 min after
injection of the radiotracer by intravenous injection of sodium
pentobarbitone (Euthatal) at 200 mg‚(kg body weight)-1. Blood
was taken immediately by cardiac puncture and centrifuged at
10 000g for 4 min to separate the plasma. Plasma from 2-3 (2.5
min) or 4 (20 min) mice was pooled for analysis by HPLC. Ice-
cold acetonitrile (0.7 mL) was added to the pooled plasma (0.7-
1.0 mL), and the precipitated proteins were removed by centrifu-
gation (10 000g for 4 min). A sample of the resulting supernatant
(900 µL) was spiked with nonradioactive 1f or 1j (10 µg) and
analyzed by HPLC (conditions for [18F]1f: λ ) 254 nm; flow )
4 mL/min; eluents ) A: CH3CN/H2O/TFA, 950/50/1 and B: CH3-
CN/H2O/TFA, 50/950/1; time program ) eluent B from 70% to
30% in 30 min, halt 5 min, from 30% to 70% in 5 min, retention
time tR(polar metabolite) ) 2.2 min, tR(metabolite) ) 11.5 min,
tR([18F]1f) ) 13.3 min; conditions for [18F]1j: λ ) 254 nm; flow
) 8 mL/min; eluents ) A: CH3CN/H2O/TFA, 950/50/1 and B:
CH3CN/H2O/TFA, 50/950/1; time program ) eluent B from 92%
to 30% in 45 min, halt 5 min, from 30% to 92% in 5 min, retention
time tR(polar metabolite) ) 1.2 min, tR([18F]1j) ) 9.1 min).
At 20 min, the hearts and the right tibiae were removed and
transferred to weighed vials for reweighing and measurement of
radioactivity using an automated gamma counter. Radioactivity was
expressed as uptake index (as defined above).
cpm/mL octanol
cpm/mL buffer
log D ) log
(
)
In Vitro Enzyme Inhibition Assays. The synthetic broad-
spectrum fluorogenic substrate (7-methoxycoumarin-4-yl) acetyl
pro-Leu-Gly-Leu-(3-(2,4-dinitrophenyl)-L-2,3-diamino-propionyl)-
Ala-Arg-NH2 (R & D Systems, Minneapolis, MN) was used to
assay activated MMP-2, MMP-8, MMP-9, and MMP-13 as
described previously.37 The inhibitions of human active MMP-2,
MMP-8, MMP-9, and MMP-13 by CGS 27023A, its nonradioactive
analogues 1a-1f, 1h, and 1j-1n, and the carboxylic acids 5a-
5c, 5f, 5h, and 5i-5l were assayed by preincubating MMP-2 (2
nM), MMP-8 (2 nM), MMP-9 (2 nM), or MMP-13 (2 nM) and
inhibitor compounds at varying concentrations (10 pM-1 mM) in
50 mM Tris‚HCl, pH 7.5, containing 0.2 M NaCl, 5 mM CaCl2,
20 µM ZnSO4, and 0.05% Brij 35 at 37 °C for 30 min. An aliquot
of substrate (10 µL of a 50 µM solution) was then added to 90 µL
of the preincubated MMP/inhibitor mixture, and the fluorescence
was determined at 37 °C by following product release with time.
The fluorescence changes were monitored using a Fusion Universal
Microplate analyzer (Packard Bioscience, MA) with excitation and
emission wavelengths set to 330 and 390 nm, respectively. Reaction
rates were measured from the initial 10 min of the reaction profile
where product release was linear with time and plotted as a function
of inhibitor dose. From the resulting inhibition curves, the IC50-
values for each inhibitor were calculated by nonlinear regression
analysis, performed using the Grace 5.1.8 software (Linux).
Acknowledgment. This work was supported by grants
from the Deutsche Forschungsgemeinschaft (DFG),
Sonderforschungsbereich 656, Mu¨nster, Germany (Project SFB
656 A2, A3 and B1). We gratefully acknowledge Sven Fatum
for technical assistance and the BASF AG, Ludwigshafen,
Germany, for kindly providing O-tert-butylhydroxylamine
hydrochloride.
Supporting Information Available: Elemental analyses and
ESI-EM data of target compounds and routine experimental and
spectroscopic data of intermediates. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Skiles, J. W.; Gonnella, N. C.; Jeng, A. Y. The design, structure,
and clinical update of small molecular weight matrix metallopro-
teinase inhibitors. Curr. Med. Chem. 2004, 11, 2911-2977.
(2) Tayebjee, M. H.; Lip, G. Y. H.; MacFadyen, R. J. Matrix metallo-
proteinases in coronary artery disease: Clinical and therapeutic
implications and pathological significance. Curr. Med. Chem. 2005,
12, 917-925.
(3) Beaudeux, J. L.; Giral, P.; Bruckert, E.; Foglietti, M. J.; Chapman,
M. J. Matrix metalloproteinases, inflammation and atherosclerosis:
Therapeutic perspectives. Clin. Chem. Lab. Med. 2004, 42, 121-
131.
(4) Whittaker, M.; Ayscough, A. Matrix metalloproteinases and their
inhibitors-current status and future challenges. Celltransmissions
2001, 17, 3-14.
(5) Borkakoti, N. Structural studies of matrix metalloproteinases. J. Mol.
Med. 2000, 78, 261-268.
Biodistribution. Animals were handled and maintained accord-
ing to approved protocols of the animal welfare committees of the
University of Mu¨nster, Germany.
Adult C57/BL6 mice (male and female, 21-30 g) were
anesthetized by isoflurane/N2O/O2, and one lateral tail vein was
canulated using a 27 G VenofixA needle with 15 cm polyethylene
catheter tubing (od 1 mm, id 0.38 mm). Animals were allowed to
recover from the anesthesia for ∼1 h, and during the studies they
were conscious but under light restraint. Compounds [18F]1f or
[18F]1j (13-32 MBq‚kg-1, 1-4 nmol‚kg-1) each were injected as
a bolus (50-100 µL) via the tail vein. In some animals, unlabeled
ligand (1f or 1j) at >25 µmol‚kg-1 was injected 10 min before the
radioligand. Aliquots of each injectant were diluted in saline and
measured to determine the radioactivity injected into each animal.
At selected times after injection of the radioligand, the animals were
sacrificed by intravenous injection of sodium pentobarbitone
(Euthatal) at 200 mg‚(kg body weight)-1. Blood samples were taken
by cardiac puncture, samples of urine and bile were taken by
puncture of the urinary and gall bladders, respectively, and selected
tissues were rapidly removed. Tissue samples were blotted dry with
filter paper and transferred to weighed vials for reweighing and
measurement of radioactivity using an automated gamma counter.
Radioactivity was expressed as an uptake index, defined as
(6) Stamenkovic, I. Extracellular matrix remodelling: The role of matrix
metalloproteinases. J. Pathol. 2003, 200, 448-464.
(7) Rhee, J. S.; Coussens, L. M. RECKing MMP function: Implications
for cancer development. Trends Cell Biol. 2002, 12, 209-211.
(8) Galis, Z. S.; Khatri, J. J. Matrix metalloproteinases in vascular
remodeling and atherogenesis: The good, the bad, and the ugly. Circ.
Res. 2002, 90, 251-62.
(9) George, S. J. Therapeutic potential of matrix metalloproteinase
inhibitors in atherosclerosis. Expert Opin. InVest. Drugs 2000, 9,
993-1007.
(10) Brinckerhoff, C. E.; Matrisian, L. M. Matrix metalloproteinases: A
tail of a frog that became a prince. Nat. ReV. Mol. Cell Biol. 2002,
3, 207-214.
(11) Rasmussen, H. S.; McCann, P. P. Matrix metalloproteinase inhibition
as a novel anticancer strategy: A review with special focus on
batimastat and marimastat. Pharmacol. Ther. 1997, 75, 69-75.
tissue radioactivity (cpm)/tissue wet weight (g)
uptake index )
radioactivity injected (cpm)/body weight (g)

5764 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 23
Wagner et al.
(12) Steward, W. P. Marimastat (BB2516): Current status of development.
Cancer Chemother. Pharmacol. 1999, 43, S56-S60.
(26) Fei, X.; Zheng, Q. H.; Hutchins, G. D.; Liu, X.; Stone, K. L.; Carlson,
K. A.; Mock, B. H.; Winkle, W. L.; Glick-Wilson, B. E.; Miller, K.
D.; Fife, R. S.; Sledge, G. W.; Sun, H. B.; Carr, R. E. Synthesis of
MMP inhibitor radiotracers [11C]methyl-CGS 27023A and its analogs,
new potential PET breast cancer imaging agents. J. Labelled Compd.
Radiopharm. 2002, 45, 449-470.
(27) Zheng, Q. H.; Fei, X.; Liu, X.; Wang, Q. J.; Sun, H. B.; Mock, B.
H.; Stone, K. L.; Martinez, T. D.; Miller, K. D.; Sledge, G. W.;
Hutchins, G. D. Synthesis and preliminary biological evaluation of
MMP inhibitor radiotracers [11C]methyl-halo-CGS 27023A analogs,
new potential PET breast cancer imaging agents. Nucl. Med. Biol.
2002, 29, 761-770.
(28) Fei, X.; Zheng, Q. H.; Liu, X.; Wang, J. Q.; Stone, K. L.; Miller, K.
D.; Sledge, G. W.; Hutchins, G. D. Synthesis of MMP inhibitor
radiotracer [11C]CGS 25966, a new potential PET tumor imaging
agent. J. Labelled Compd. Radiopharm. 2003, 46, 343-351.
(29) Zheng, Q. H.; Fei, X.; Liu, X.; Wang, J. Q.; Stone, K. L.; Martinez,
T. D.; Gay, D. J.; Baity, W. L.; Miller, K. D.; Sledge, G. W.;
Hutchins, G. D. Comparative studies of potential cancer biomarkers
carbon-11 labeled MMP inhibitors (S)-2-(4′-[11C]methoxybiphenyl-
4-sulfonylamino)-3-methylbutyric acid and N-hydroxy-(R)-2-[[(4′-
[11C]methoxyphenyl)sulfonyl]benzylamino]-3-methylbutanamide. Nucl.
Med. Biol. 2004, 31, 77-85.
(30) Breyholz, H.-J.; Wagner, S.; Levkau, B.; Schober, O.; Scha¨fers, M.;
Kopka, K. A 18F-radiolabelled analogue of CGS 27023A as a
potential agent for assessment of matrix-metalloproteinase activity
in ViVo. Q. J. Nucl. Med. Mol. Imaging 2007, 51, 24-31.
(31) Lasne, M.-C.; Perrio, C.; Rouden, J.; Barre´, L.; Roeda, D.; Dolle´,
F.; Crouzel, C. Chemistry of â+-emitting compounds based on
fluorine-18. Top. Curr. Chem. 2002, 222, 201-258.
(32) Smith, C.; Porter, B.; Walsh, R.; Majid, T.; McCarthy, C.; Harris,
D; Astles, P.; McLay, I.; Morley, A.; Bridge, A.; Van Sickle, A.;
Halley, F.; Roach, A.: Foster, M. (Rhone-Poulenc Rorer Ltd.).
Substituted phenyl compounds with a substituent having a 1,3-
benzodioxole ring. U.S. Patent 6034893, April 11, 2000.
(33) Mutschler, E. Arzneimittelwirkungen; wissenschaftliche Verlagsge-
sellschaft mbH: Stuttgart, 1996; pp 20-34.
(34) Caldarelli, M.; Habermann, J.; Ley, S. V. Synthesis of an array of
potential matrix metalloproteinase inhibitors using a sequence of
polymer-supported reagents. Bioorg. Med. Chem. Lett. 1999, 9,
2049-2052.
(35) Wilson, A. A.; Jin, L.; Garcia, A.; DaSilva, J. N.; Houle, S. An
admonition when measureing the lipophilicity of radiotracers using
counting techniques. Appl. Radiat. Isot. 2001, 54, 203-208.
(36) Wilson, A. A.; Houle, S. Radiosynthesis of carbon-11 labelled
N-Methyl-2-(arylthio)benzylamines: Potential radiotracers for sero-
tonin reuptake receptors. J. Labelled Compd. Radiopharm. 1999, 42,
1277-1288.
(37) Huang, W.; Meng, Q.; Suzuki, K.; Nagase, H.; Brew, K. Mutational
study of the amino-terminal domain of human tissue inhibitor of
metalloproteinasees 1 (TIMP-1) locates an inhibitory region for matrix
metalloproteinases. J. Biol. Chem. 1997, 272, 22086-22091.
(13) Price, A.; Shi, Q.; Morris, D.; Wilcox, M. E.; Brasher, P. M.;
Rewcastle, N. B.; Shalinsky, D.; Zou, H.; Appelt, K.; Johnston, R.
N.; Yong, V. W.; Edwards, D.; Forsyth, P. Marked inhibition of tumor
growth in a malignant glioma tumor model by a novel synthetic
matrix metalloproteinase inhibitor AG3340. Clin. Cancer Res. 1999,
5, 845-854.
(14) Giersing, B. K.; Rae, M. T.; CarballidoBrea, M.; Williamson, R. A.;
Blower, P. J. Synthesis and characterization of 111In-DTPA-N-TIMP-
2: A radiopharmaceutical for imaging matrix metalloproteinase
expression. Bioconjugate Chem. 2001, 12, 964-971.
(15) Oltenfreiter, R.; Burvenich, I.; Staelens, L.; Lejeune, A.; Frankenne,
F.; Foidart, J. M.; Slegers, G. Synthesis, quality control and in ViVo
evaluation of [123I]rhTIMP-2, a potential tumour-imaging agent. J.
Labelled Compd. Radiopharm. 2005, 48, 387-396.
(16) Li, W. P.; Lewis, J. S.; Lan, S.; Anderson, C. J. Cu-64-DOTA
CTTHWGFTLC, a selective gelatinase inhibitor for tumor imaging.
J. Nucl. Med. 2002, 43, S277-S278.
(17) Kulasegaram, R.; Giersing, B.; Page, C. J.; Blower, P. J.; Williamson,
R. A.; Peters, B. S.; O’Doherty, M. J. In ViVo evaluation of 111In-
DTPA-N-TIMP-2 in Kaposi sarcoma associated with HIV infection.
Eur. J. Nucl. Med. 2001, 28, 756-761.
(18) Li, W. P.; Lewis, J. S.; Eiblmaier, M.; Lan, S.; Anderson, C. J. In
Vitro and in ViVo evaluation of a radiolabelled gelatinase inhibitor
for microPET imaging of metastatic breast cancer. Mol. Imaging Biol.
2002, 4, S23.
(19) Spargue, J. E.; Li, W. P.; Liang, K.; Achilefu, S.; Anderson, C. J. In
Vitro and in ViVo investigation of matrix metalloproteinase expression
in metastatic tumor models. Nucl. Med. Biol. 2006, 33, 227-237.
(20) Wagner, S.; Breyholz, H.-J.; Faust, A.; Ho¨ltke, C.; Levkau, B.;
Schober, O.; Scha¨fers, M.; Kopka, K. Molecular imaging of matrix
metalloproteinases in ViVo using small molecule inhibitors for SPECT
and PET. Curr. Med. Chem. 2006, 13, 2819-2838.
(21) MacPherson, L. J.; Bayburt, E. K.; Capparelli, M. P.; Carroll, B. J.;
Goldstein, R.; Justice, M. R.; Zhu, L.; Hu, S. I.; Melton, R. A.; Fryer,
L.; Goldberg, R. L.; Doughty, J. R.; Spirito, S.; Blancuzzi, V.; Wilson,
D.; O’ Byrne, E. M.; Ganu, V.; Parker, D. T. Discovery of CGS
27023A, a non-peptidic, potent, and orally active stromelysin inhibitor
that blocks cartilage degradation in rabbits. J. Med. Chem. 1997, 40,
2525-2532.
(22) Scozzafava, A.; Supuran, C. T. Carbonic anhydrase and matrix
metalloproteinase inhibitors: sulfonylated amino acid hydroxamates
with MMP inhibitory properties act as efficient inhibitors of CA
isozymes I, II, and IV, and N-hydroxysulfonamides inhibit both these
zinc enzymes. J. Med. Chem. 2000, 43, 3677-3687.
(23) Kopka, K.; Breyholz, H.-J.; Wagner, S.; Law, M. P.; Riemann, B.;
Schro¨er, S.; Trub, M.; Guilbert, B.; Levkau, B.; Schober, O.; Scha¨fers,
M. Synthesis and preliminary biological evaluation of new radioio-
dinated MMP inhibitors for imaging MMP activity in ViVo. Nucl.
Med. Biol. 2004, 31, 257-267.
(24) Scha¨fers, M.; Riemann, B.; Kopka, K.; Breyholz, H.-J.; Wagner, S.;
Scha¨fers, K. P.; Law, M. P.; Schober, O.; Levkau, B. Scintigraphic
imaging of matrix metalloproteinase activity in the arterial wall in
ViVo. Circulation 2004, 109, 2554-2559.
(25) Zheng, Q. H.; Hutchins, G. D.; Mock, B. H.; Winkle, W. L. MMP
inhibitor radiotracer [11C]Methyl-CGS 27023A: A new PET breast
cancer imaging agent. J. Labelled Compd. Radiopharm. 2001, 44,
S104-S106.
JM0708533