Identification of metabolites of the tryptase inhibitor CRA-9249: Observation of a metabolite derived from an unexpected hydroxylation pathway

Article abstract of DOI:10.1016/j.bmcl.2006.05.003

The metabolites of the tryptase inhibitor CRA-9249 were identified after exposure to liver microsomes. CRA-9249 was found to be degraded rapidly in liver microsomes from rabbit, dog, cynomolgus monkey, and human, and less rapidly in microsomes from rat. T

Full text of DOI:10.1016/j.bmcl.2006.05.003

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4053–4058
Identification of metabolites of the tryptase inhibitor CRA-9249:
Observation of a metabolite derived from an
unexpected hydroxylation pathway
Walter Yu,a Jeffrey M. Dener,c,* Daniel A. Dickman,c Paul Grothaus,b Yun Ling,a
Liang Liu,a Chris Havel,a Kimberly Malesky,b Tania Mahajan,b Colin O’Brian,c
Emma J. Shelton,b David Sperandio,b Zhiwei Tong,b Robert Yeeb and Joyce J. Mordentia
aDepartment of Drug Metabolism and Pharmacokinetics, Celera Genomics, 180 Kimball Way, South San Francisco, CA 94080, USA
bDepartment of Medicinal Chemistry, Celera Genomics, 180 Kimball Way, South San Francisco, CA 94080, USA
cDepartment of Process Chemistry, Celera Genomics, 180 Kimball Way, South San Francisco, CA 94080, USA
Received 24 February 2006; revised 28 April 2006; accepted 1 May 2006
Available online 18 May 2006
Abstract—The metabolites of the tryptase inhibitor CRA-9249 were identified after exposure to liver microsomes. CRA-9249 was
found to be degraded rapidly in liver microsomes from rabbit, dog, cynomolgus monkey, and human, and less rapidly in micro-
somes from rat. The key metabolites included cleavage of an aryl ether, in addition to an unexpected hydroxylation of the amide
side chain adjacent to the amide nitrogen. The chemical structures of both metabolites were confirmed by synthesis and comparison
to material isolated from the liver microsomes. Several suspected hydroxylated metabolites were also synthesized and analyzed as
part of the structure identification process.
Ó 2006 Elsevier Ltd. All rights reserved.
Metabolite identification is a critical step in the develop-
ment of new chemical entities (NCEs).1 Understanding
the metabolic fate of potential therapeutic agents early
in the drug discovery process helps to identify those
compounds that may be rapidly metabolized and excret-
ed before they can exert their pharmacological benefit
in vivo.2 This understanding is particularly important
before the development of the active pharmaceutical
ingredient (API) and the initiation of the more expensive
and time consuming preclinical studies, such as GLP
toxicology studies. Increasingly, an integrated effort
between medicinal chemistry, drug metabolism, and
toxicology departments to understand the metabolic
liabilities and toxicological profile of the lead candidate
will provide direction to identify and develop a clinical
candidate with an improved metabolic profile. Typical-
ly, this approach incorporates an iterative design effort
through the use of in vitro predictors of metabolic liabil-
ities. In fact, several recent reports have highlighted the
successful application of iteration in identifying metabo-
lites that have a better pharmacological profile than the
lead drug candidate. The best example of a marketed
product to come from a metabolite-driven drug develop-
ment program is the cholesterol absorption inhibitor,
ZetiaÒ (Ezetimibe; SCH 58235).3 While the identifica-
tion of an active metabolite may be beneficial to a drug
development program, some active metabolites can in-
duce adverse effects in vivo,4 and appropriate testing is
required to assess this possibility.
Celera has developed a novel series of bis(benzimid-
azole)methane-containing tryptase inhibitors as poten-
tial agents for allergic and inflammatory diseases, in
particular asthma.5 One such derivative is the bis(benz-
imidazole) difluoromethane inhibitor CRA-9249 (1)
(Scheme 1), an orally bioavailable inhibitor of mast cell
tryptase. During our preclinical development program
for CRA-9249 for the treatment of asthma, we studied
its metabolism in vitro to identify its metabolic profile
and stability, and to predict its metabolic fate in vivo.
Keywords: Metabolism; Metabolite identification; Synthesis; Active
metabolite.
Exposure of CRA-9249 to an aqueous suspension of
homogenized liver microsomes from four species
[mouse, dog, cynomolgus (cyno) monkey, and human]
*
Corresponding author. Tel.: +1 650 866 6695; fax: +1 650 866
6655; e-mail: jdener@yahoo.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.003

4054
W. Yu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4053–4058
H
F
F
N
F
CH3
N
N
a
N
F
F
H
N
O
O
1
CRA-9249
H
N
H
H
N
F
F
F
F
F
F
F
N
F
F
HO
CH3
CH3
CH3
N
N
N
N
N
N
HO
N
N
N
F
F
F
F
F
H
N
H
H
N
N
O
O
OH
O
O
O
2
3
4
OH
F
H
N
H
N
H
N
F
F
F
F
F
F
F
F
F
HO
CH3
CH3
CH3
N
N
N
N
N
HO
N
N
N
HO
N
F
F
F
F
F
H
H
N
H
N
N
O
O
O
O
O
O
7
5
6
Scheme 1. Metabolism of the tryptase inhibitor CRA-9249 (1). Reagents and conditions: (a) 20 lM CRA-9249, liver microsomes (1.5 mg protein/
mL) from five species, 2 mM NAPDH, 10 mM MgCl2, 37 °C, up to 3 h.
resulted in rapid degradation of the parent and
formation of several metabolites. In these species, less
than 10% of the parent drug was detected after a 3-h
exposure period. However, metabolism of 1 in rat liver
microsomes was much slower, with 56% of the parent
remaining after exposure over the same 3-h period.
Employing LC/MS/MS analysis, a total of five Phase 1
metabolites were identified from these in vitro liver
microsome incubation experiments. Furthermore, the
relative rate of disappearance of the parent 1 was species
dependent, according to the order of cyno monkey >
rabbit > dog > human ꢀ rat.
O-dealkylation product 4 and aromatic hydroxylation
derivatives 6 and 7. Finally, a large-scale incubation of
CRA-9249 in cyno monkey liver microsomes was per-
formed to assist in the confirmation of the identity of these
metabolites, allowing for the generation of proton-, car-
bon-, and fluorine-NMR data from the metabolite mix-
ture. Synthesis of the O-dearylated species (4) was
accomplished in one step by the reaction of the acid 86
with PyBroP and 500 mole percent of ethanolamine in
DMF as shown in Scheme 2. O-dearylated metabolite 4
was obtained after aqueous work-up and crystallization.
The synthesis of proposed 4- and 6-hydroxybenzimidaz-
ole derivatives 6 and 7 proved to be more challenging.
The synthesis of 6 started from commercially available
2,4,5-trifluoroaniline (9) (Scheme 3), which was acetylat-
ed and nitrated according to a literature procedure to
give nitro derivative 11.7 Heating 11 with aqueous sodi-
um hydroxide resulted in displacement of the fluorine
adjacent to the nitro group, as well as deacetylation of
the aniline. The crude product was not isolated but
Monohydroxylation and O-dearylation were the two pre-
dominant metabolites produced from 1 after exposure to
liver microsomes from all five species, although the rate of
formation of the monohydroxylated metabolite was dif-
ferent depending on the species. To better understand
the fate of 1, we initiated a program to synthesize several
of the proposed structures to confirm the structural iden-
tity of each of the major metabolites. In addition, we eval-
uated the in vitro potency (Ki) of these synthetic samples
in Celera’s zinc-dependent tryptase assay to determine if
any of these could be active metabolites of the parent
compound CRA-9249. As shown in Scheme 1, these
metabolites included monohydroxylation, dihydroxyla-
tion, and dearylation. Based on these LC/MS/MS
studies, generic structures 2 and 3 were considered the
most plausible structures for the monohydroxylated
metabolite. The hydroxyethylamide 4 was proposed as
the dearylated metabolite. Finally, generic structure 5
was proposed for the minor dihydroxylated metabolite.
Our initial target structures for synthesis included the
H
F
F
N
F
CH3
a
N
N
4
N
F
OH
O
8
Scheme 2. Reagents and conditions: (a) PyBroP, 2-aminoethanol
(500 mole %), DMF then crystallization, 31%.

W. Yu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4053–4058
4055
F
F
F
F
F
F
NO2
NH2
F
F
NO2
NH2
a
b
F
F
F
NO2
Ac
b
a
HO
Ac
NH2
N
H
F
N
F
H
F
16
17
10
11
9
H
N
F
c
F
NH2
NH2
EtO
CO2Et
HCl
+
OH
OH
CO2Et
N
NH
HO
HO
F
NO2
NH2
F
NH2
NH2
d
c
F
F
18
19
20
F
F
H
N
H
F
H
12
13
CH3
N
N
HO
d
e
CH3
O
N
7
F
F
HN
O
e-g
H
N
H
HO2C
CO2H
N
+
6
O
F
F
H2N
F
O
21
14
15
Scheme 4. Reagents and conditions: (a) aq NaOH, DMPU, 120 °C,
86%; (b), H2, 10% Pd–C, EtOH; (c) 19, EtOH, reflux, 43% from 17; (d)
15, DMPU, 185 °C, 73%; (e) (PhSO2)2N–F, EtOH, reflux, 3%.
Scheme 3. Reagents and conditions: (a) Et3N, THF, RT to À20 °C
then AcCl, <20 °C, 90%; (b) 25% AcOH in H2SO4 then HNO3/H2SO4/
AcOH (4:1:1), 50%; (c) aq NaOH, 85 °C to reflux; (d) H2, 10% Pd–C,
EtOH, 48% from 11; (e) PyBroP, N-methylmorpholine (NMM), DMF,
À10 °C; (f) PyBrOP, DMF, À10 °C, then 13, NMM. À10 °C; (g)
AcOH, 85 °C, 9% from 15.
However, a clue to the structural identity of the proba-
ble mono-hydroxylated species arose from identification
of a secondary metabolite, the primary amide 22, which
was identified by LC–MS analysis of the metabolite
broth when it was reanalyzed about one month after
incubation (Scheme 5). Subsequently, the presence of
22 in the microsome incubation broth was confirmed
by HPLC analysis by co-injection of a mixture of an
authentic sample of 22 with the metabolite broth. From
this observation, hydroxylation at the a-methylene car-
bon of the amide nitrogen of 1 (compound 23) was pro-
posed as a potentially unstable metabolite, which could
undergo hydrolysis to give amide 22 and the 2-(4-fluoro-
phenoxy)acetaldehyde (24) upon standing in the aque-
ous medium from the liver microsome incubations.
converted directly to 13 by catalytic reduction of the ni-
tro group. The formation of the bis(benzimidazole)diflu-
oromethane fragment was accomplished in a one-pot
reaction consisting of four chemical transformations.8
2,2-Difluoromalonic acid (14) was treated with PyBroP
and N-methylmorpholine (NMM) in DMF, then by
4-amino-3-(methylamino)benzamide 15 was added.
Another equivalent of PyBroP was added, followed by
diamine 13 and another portion of NMM. Finally, the
mixture was worked up and the crude product was heat-
ed in acetic acid to give the desired putative metabolite 6
after chromatographic purification.
Synthesis of putative metabolite 7 was accomplished as
described in Scheme 4. Commercially available 2,3,4-tri-
fluoro-6-nitroaniline (16) was reacted with aqueous
sodium hydroxide to introduce the required hydroxyl
functionality at C-3. The nitro group in 17 was reduced
to the 1,2-diaminobenzene 18, which was converted to
the corresponding 4,6-difluoro-5-hydroxybenzimidazole
20 by heating with imidate 19 in ethanol. Condensation
of the ester in 20 with diamine 15 in DMPU at 185 °C
afforded the bis(benzimidazole)methane derivative 21.
Fluorination of the bridgehead methylene of 21 was
accomplished with N-fluorobenzenesulfonylimide to
provide the desired 6-hydroxybenzimidazole derivative
7.9
N-dealkylation of secondary and tertiary amides via met-
abolic pathways is well documented in the literature and
presumably occurs via hydrolysis of the unstable a-
hydroxyamide; however, literature reports of the isola-
tion and characterization of these a-hydroxyamide
metabolites are limited due to this propensity to undergo
hydrolysis to the amide and aldehyde.11 An early example
of a-hydroxylation of an amide was reported for an anti-
convulsant oxadiazole, containing an N-methyl-N-acet-
amide moiety, which underwent hydroxylation at the N-
methyl group, as well as hydrolysis to the secondary
amide.12 Furthermore, N,N-dimethylbenzamides have
been shown to provide N-(hydroxymethyl)-N-methylben-
zamides in liver microsomes.13 A more relevantexample is
the metabolism of the anti-diabetic agent glyburide (25)
(Scheme 6), which contains some structural similarity to
1.14 Glyburide also undergoes a-hydroxylation to the
hydroxyamide 26, in addition to hydroxylation adjacent
to the benzene ring (27) and hydroxylation at various
positions on the cyclohexyl ring.
Analysis of DMSO–water solutions obtained from the
liver microsome incubation of 1 confirmed the presence
of compound 4 by HP LC–UV and LC–MS analysis
and comparison of these data with the authentic sample
of 4 prepared as described above.10 However, the presence
of 6 or 7 could not be confirmed in the metabolite mixture.
Analysis of the metabolite broth from the large-scale
microsome incubation study by 19F NMR did not provide
a match to the fluorine NMR of synthetic compound 7.
In order to provide experimental evidence to support
the a-hydroxylation of 1, the synthesis of proposed

4056
W. Yu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4053–4058
F
F
F
F
N
N
F
F
CH3
CH3
F
N
H
N
H
N
N
a
b
N
N
H
F
F
+
F
1
O
H
O
N
NH2
O
O
OH
O
24
22
23
Scheme 5. Proposed pathway for the formation and degradation of the a-hydroxyamide metabolite 23. (a) 20 lM CRA-9249, liver microsomes
(1.5 mg protein/mL) from five species, 2 mM NAPDH, 10 mM MgCl2, 37 °C, up to 3 h; (b) DMSO–water, ꢁ1 month.
aldehyde in refluxing acetonitrile afforded the desired a-
O
O
O
hydroxybenzamide 23 in low yield but in sufficient quan-
tity for isolation and characterization of a small sample.
The corresponding dehydration product, the N-acyli-
mine 29, was also produced in trace amounts in the reac-
tion.16 Compound 23 was unstable on standing and was
determined by LC–MS analysis to degrade back to 22
and 24. The synthetic material corresponding to pro-
posed structure 23 was shown to be identical to the ma-
jor metabolite of CRA-9249 (1) upon analysis of this
mixture based on HPLC retention time, mass spectral
fragmentation, and parent ions. Furthermore, proton
NMR of the synthetic material provided a match of
the major signals for the proton NMR spectrum that
were obtained from the large-scale metabolite incuba-
tion study.17–19
S
ref. 14
O
N
H
N
H
Cl
N
H
OCH3
25
O
O
O
O
O
S
S
O
OH
N
H
N
H
Cl
Cl
N
H
OCH3
26
+
O
O
N
H
N
H
N
H
OCH3
OH
Finally, we determined the equilibrium constants for
binding of synthetic metabolite derivatives 4, 6, 7, and
23 to tryptase. Apparent inhibition constants (Ki) for
these compounds were determined using Celera’s zinc-
dependent tryptase assay as described previously.5 Data
for these metabolites, as well as the parent compound
CRA-9249 (1), are shown in Table 1.
27
Scheme 6. Metabolites of glyburide (25).
metabolite 23 was performed according to Scheme 7, the
key step of which is essentially the reverse of the metab-
olite degradation process. Reaction of acid 8 with
ammonia and PyBrOP afforded the primary amide com-
ponent. Synthesis of the aldehyde fragment 24 was
accomplished by alkylation of 4-fluorophenol (28) with
bromoacetaldehyde dimethoxy acetal to give the pro-
tected aldehyde, which was hydrolyzed directly to the
aldehyde 24 with a catalytic quantity of sulfuric acid
in aqueous acetic acid.15 Reaction of amide 22 with this
As shown in Table 1, the O-dearylated metabolite 4 and
the synthetic aryl monohydroxy metabolite 7 possessed
low micromolar activity against tryptase and were 4-
and 23-fold less active than the parent in the enzyme as-
say; interestingly, however, both the synthetic aryl
monohydroxylated derivative 6 and the major metabo-
lite, a-hydroxyamide 23, were 8.8 and 12.5 times more
potent than the parent compound 1.
F
Although the increased potency of 6 could be a result of
participation by the aryl hydroxy group in the zinc-med-
iated binding of the compound at the active site of tryp-
tase,5 the increased potency of a-hydroxyamide 23 is not
so clear. However, one possible explanation is that 23 is
converted to N-acylimine 29 which then irreversibly
binds to the enzyme.20 In any event, due to the rapid
a-c
d
24
HO
28
F
F
N
F
CH3
N
H
N
N
Table 1. In vitro inhibition of mast cell tryptase by compound 1 and
synthetic samples of proposed metabolites
F
23
+
F
N
O
Compound
Ki (lM)
O
1
4
0.3
7.0
29
6
0.034
1.4
0.024
Scheme 7. Reagents and conditions: (a) 1 M t-BuOK in THF, DMSO;
(b) BrCH2CH(OCH3)2; (c) H2SO4 (cat), acetone–water (10:1), 33%
from 28; (d) 22, CH3CN, 90 °C, 3% after preparative HPLC.
7
23

W. Yu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4053–4058
4057
9. Sperandio, D.; Gangloff, A. R.; Litvak, J.; Goldsmith, R.;
Hataye, J. M.; Wang, V. R.; Shelton, E. J.; Elrod, K.;
Janc, J. W.; Clark, J. M.; Rice, K.; Weinheimer, S.;
Yeung, K.-S.; Meanwell, N. A.; Hernandez, D.; Staab, A.
J.; Venables, B. L.; Spencer, J. S. Bioorg. Med. Chem. Lett.
2002, 12, 3129.
metabolism of 1 in human liver microsomes and the
instability of the major metabolite 23, compound 1
was not considered as a viable candidate for continued
development.21
10. Positive and negative ion mass spectral data for metabolite
are as follows: m/z = 422.2 [M+H]+; m/z = 420.3
Acknowledgments
4
[MÀH]À. The synthetic sample of 4 provided the following
data: m/z = 422.1 [M+H]+; m/z = 420.2 [MÀH]À.
11. For representative examples of metabolism-mediated N-
dealkylation of amides in vitro and in vivo, please see the
following references: (a) Sugnaux, F. R.; Benakis, A. Eur.
J. Drug Metab. Pharmacokinet. 1978, 3, 235; (b) Yoshida,
K.; Manbu, K.; Arakawa, S.; Miyazaki, H.; Hashimoto,
M. Biomed. Mass Spectrom. 1979, 6, 253; (c) Schoenig, G.
P., ; Hartnagel, R. E., Jr.; Osimitz, T. G.; Llanso, S. Drug.
Metab. Dispos. 1996, 24, 156; (d) Constantino, L.; Iley, J.
Xenobiotica 1999, 29, 409.
We thank Drs. Mike Green and Mike Venuti for their
support, guidance, and commitment during the execu-
tion of this work.
References and notes
1. (1) Yan, Z.; Caldwell, G. W. Curr. Topics Med. Chem.
2001, 1, 403; (2) Elkins, S.; Ring, B. J.; Grace, J.;
McRobie-Belle, D. J.; Wrighton, S. A. J. Pharmacol.
Toxicol. Methods 2000, 44, 313.
12. Allen, J. G.; Blackburn, M. J.; Caldwell, S. M. Xenobiotica
1971, 1, 3–12.
2. Fura, A.; Shu, Y.-Z.; Zhu, M.; Hanson, R. L.; Roongta,
V.; Humphreys, W. G. J. Med. Chem. 2004, 47, 4339.
3. (a) Rosenblum, S. B.; Huynh, T.; Alfonso, A.; Davis, H.
R., Jr.; Yumibe, N.; Clader, J. W.; Burnett, D. A. J. Med.
Chem. 1998, 41, 973; (b) Clader, J. W. J. Med. Chem. 2001,
47, 1.
4. (a) Piromohamed, M.; Kitteringham, N. R.; Park, B. K.
Drug Safety 1994, 11, 114; (b) Williams, D. P.; Naisbitt,
D. J. Curr. Opin. Drug Disc. Dev. 2002, 5, 104; (c) Nassar,
A.-E. F.; DeMaio, W.; Davis, M.; Talaat, R. E. Curr.
Drug Disc. 2004, 20.
5. (a) Janc, J. W.; Clark, J. M.; Warne, R. L.; Katz, B. A.;
Moore, W. R. Biochemistry 2000, 39, 4792; (b) Katz, B.
A.; Clark, J. M.; Finer-Moore, J. S.; Jenkins, T. E.;
Johnson, C. R.; Ross, M. J.; Luong, C.; Moore, W. R.;
Stroud, R. M. Nature 1998, 391, 608; (c) Church, T. J.;
Cutshall, N. S.; Gangloff, A. R.; Jenkins, T. E.; Linsell, M.
S.; Litvak, J.; Rice, K. D.; Spencer, J. R.; Wang, V. R.
PCT Int. Appl. WO 9845275, 1998.
13. Ross, D.; Farmer, P. B.; Geschner, A.; Hickman, J. A.;
Threadgill, M. D. Biochem. Pharmacol. 1983, 32, 1773.
14. Tiller, P. R.; Land, A. P.; Jardine, I.; Murphy, D. M.;
Sozio, R.; Ayrton, A.; Schaefer, W. H. J. Chromatogr. A
1998, 794, 15.
15. Smith, R. L.; Bicking, J. B.; Gould, N. P.; Lee, T.-J.;
Robb, C. M.; Keuhl, F. A., Jr.; Mandel, L. R.; Cragoe, E.
J., Jr. J. Med. Chem 1977, 20, 540.
16. The presence of imine 29 in this reaction was proposed
based on LC–MS data from the crude product obtained
from the synthesis of 23. Only a few reports of N-
acylimines derived from aliphatic aldehydes are represent-
ed in the literature since these are known to isomerize to
the corresponding enamide: Gizecki, P.; Dhal, R.; Pou-
lard, C.; Gosselin, P.; Dujardin, G. J. Org. Chem. 2003,
68, 4338. Imine 29 may also be produced during the
metabolism of 1 in liver microsomes, but it is expected to
readily hydrolyze to 22 and 24 under the aqueous
conditions employed (see also Ref. 19).
6. Dener, J. M.; O’Bryan, C.; Yee, R.; Shelton, E. J.;
Sperandio, D.; Mahajan, T.; Palmer, J.; Spencer, J. R.;
Tong. Z. Tetrahedron Letters, accepted for publication.;
7. Finger, G. C.; Reed, F. H.; Burness, D. M.; Fort, D. M.;
Blough, R. R. J. Am. Chem. Soc. 1951, 73, 145.
8. Mittendorf, J.; Henning, R.; Raddatz, S.; Schlemmer,
K.-H.; Hiraoka, M.; Kadono, H.; Mogi, M.; Moriwaki,
T.; Murata, T.; Sakakibara, S.; Shimada,.M.; Yoshida,
N.; Yoshino. T. PCT Int. Appl. WO 0020401, 2000. The
sequence presumably goes through the monoamide and
bis(amide) intermediates A and B. Yields of this process
tended to be low due to the formation of symmetrical
bis(amide) derivatives from two molecules of 14 with
reaction 15.
17. The analytical data for the synthetic sample of metabolite
1
23 are as follows: H NMR (DMSO-d6; 500 MHz) d 8.97
(d, 1H, CONH), 8.29 (s, 1H, ArH), 8.18 (s, 1H, NH) 7.88
(dd, 1H, ArH), 7.79 (d, 1H, ArH), 7.08–7.15 (m, 2H,
ArH), 6.97–7.03 (m, 2H, ArH), 6.83 (br s, 2H, ArH), 6.32
(d, 1H, OH), 5.75–5.81 (m, 1H, OCHN), 4.07 (dt, 2H,
NCH2), 3.87 (s, 3H, CH3); LC–MS: m/z 532.2 [M+H]+,
377.9 [MÀ154]+. The mass-to-charge ratio of 377.9
corresponds to the ion for the protonated ([M+H]+)
primary amide 22. For comparison, compound 1 provided
the following analytical data: 1H NMR (DMSO-d6;
500 MHz) d 8.82 (t, 1H, CONH), 8.31 (s, 1H, ArH),
7.85 (dd, 1H), 7.78 (d, 1H, ArH), 7.33 (d, 1H, ArH), 7.23
(br t, 1H, ArH), 7.14–7.08 (m, 2H, ArH), 7.01–6.96 (m,
2H, ArH), 4.12 (t, 2H, OCH2), 4.02 (s, 3H, CH3), 3.67 (q,
2H, NCH2); LC–MS: m/z 516.1 [M+H]+, 538.0 [M+Na]+.
18. The metabolite broth provided a proton NMR spectrum
that contained the following signals in common with the
proton spectrum for synthetic 23: 1H NMR (DMSO-d6;
500 MHz) d 9.01 (d, CONH), 8.36 (s, ArH), 7.89 (d ArH),
7.80 (d, ArH), 7.08–7.14 (m, ArH), 6.96–7.03 (m, ArH),
6.83 (br s, 1H, ArH), 6.32 (d, OH), 5.74–5.82 (m, OCHN),
4.07 (dt, NCH2), 4.03 (s, CH3).
CH3
N
O
F
F
O
HO2C
N
14
+
15
H
O
F
H2N
A
HO
F
CH3
N
O
F
F
H
F
N
O
N
H
O
O
19. The proton NMR spectrum of the metabolite changes
upon storage for about one month. For example, the
sharp doublet at 9.01 ppm disappears and two new signals
are observed at 10.2 and 9.65 ppm. The latter signal is
consistent with the chemical shift for the aldehyde proton
in 28, as determined by the chemical shift for the aldehyde
F
NH2
H2N
B
6

4058
W. Yu et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4053–4058
proton observed in the proton NMR spectrum from an
authentic sample of the aldehyde.
potency relative to 1; however, this option was not
considered for two reasons. First of all, this class of
inhibitors did not possess sufficient potency at pH 7.4
in physiological concentrations of free zinc. More
importantly, we were unable to establish evidence for
the formation of stable, inhibitory ternary complexes
20. An alternative explanation is that benzamide 22 is
responsible for this activity; however, the Ki for this
compound was determined to be 2.7 lM, about 100-fold
less active than the synthetic sample of 23 and 3-fold less
active than the parent compound 1.
of zinc, tryptase, and
1 in the human mast cell
21. Hydroxybenzimidazole
candidate for development due to its increased
6
could have been
a
viable
granules (Amos Baruch and James Clark, unpublished
work).