Bicyclic N-hydroxyurea inhibitors of 5-lipoxygenase: Pharmacodynamic, pharmacokinetic, and in vitro metabolic studies characterizing N-hydroxy-N-(2,3-dihydro-6-(phenylmethoxy)-3-benzofuranyl)urea

Article abstract of DOI:10.1021/jm960271d

A series of N-hydroxyurea derivatives have been prepared and examined as inhibitors of 5-lipoxygenase. Oral activity was established by examining the inhibition of LTB₄ biosynthesis in an ex vivo assay in the mouse. The pharmacodynamic performance in the mouse of selected compounds was assessed using an ex vivo LTB₄ assay and an adoptive peritoneal anaphylaxis assay at extended pretreat times. Compounds with an extended duration of action were reexamined as the individual enantiomers in the ex vivo assay, and the (S) enantiomer of N-hydroxy-N-[2,3-dihydro-6-(phenylmethoxy)-3-benzofuranyl]urea, (+)-1a (SB 202235), was selected as the compound with the best overall profile. Higher plasma concentrations and longer plasma half-lives were found for (+)-1a relative to its enantiomer in the mouse, monkey, and dog. In vitro metabolic studies in mouse liver microsomes established enantiospecific glucuronidation as a likely mechanism for the observed differences between the enantiomers of 1a. Enantioselective glucuronidation favoring (-)-1a was also found in human liver microsomes.

Full text of DOI:10.1021/jm960271d

J . Med. Chem. 1996, 39, 5035-5046
5035
Articles
Bicyclic N-Hyd r oxyu r ea In h ibitor s of 5-Lip oxygen a se: P h a r m a cod yn a m ic,
P h a r m a cok in etic, a n d in Vitr o Meta bolic Stu d ies Ch a r a cter izin g
N-Hyd r oxy-N-(2,3-d ih yd r o-6-(p h en ylm eth oxy)-3-ben zofu r a n yl)u r ea †
J erry L. Adams,*,‡ Ravi S. Garigipati,‡ Margaret Sorenson,‡ Stanley J . Schmidt,‡,§ William R. Brian,|,
J ohn F. Newton,|, Kathy A. Tyrrell,| Eric Garver,| Lee A. Yodis,| Marie Chabot-Fletcher,∇ Maritsa Tzimas,∇
Edward F. Webb,∇ J ohn J . Breton,∇ and Don E. Griswold*,∇
SmithKline Beecham Pharmaceuticals, P.O. Box 1539, King of Prussia, Pennsylvania 19406
Received April 8, 1996X
A series of N-hydroxyurea derivatives have been prepared and examined as inhibitors of
5-lipoxygenase. Oral activity was established by examining the inhibition of LTB4 biosynthesis
in an ex vivo assay in the mouse. The pharmacodynamic performance in the mouse of selected
compounds was assessed using an ex vivo LTB4 assay and an adoptive peritoneal anaphylaxis
assay at extended pretreat times. Compounds with an extended duration of action were re-
examined as the individual enantiomers in the ex vivo assay, and the (S) enantiomer of
N-hydroxy-N-[2,3-dihydro-6-(phenylmethoxy)-3-benzofuranyl]urea, (+)-1a (SB 202235), was
selected as the compound with the best overall profile. Higher plasma concentrations and
longer plasma half-lives were found for (+)-1a relative to its enantiomer in the mouse, monkey,
and dog. In vitro metabolic studies in mouse liver microsomes established enantiospecific
glucuronidation as a likely mechanism for the observed differences between the enantiomers
of 1a . Enantioselective glucuronidation favoring (-)-1a was also found in human liver
microsomes.
In tr od u ction
biochemical pharmacology of inflammatory bowel dis-
ease (IBD) suggests an important role for 5-LO products
in the pathophysiology of this group of diseases. As is
the case with asthma, supporting evidence for the role
of leukotrienes has been obtained in the clinic. The
efficacy of sulphasalazine and the active principle,
5-aminosalicylic acid, has been suggested to be due to
inhibition of leukotriene biosynthesis.4 Perhaps most
cogent are the recent findings that the 5-LO inhibitor,
zileuton, reduced colonic levels of LTB4 in ulcerative
colitis patients. Since these results were encouraging,
a preliminary clinical trial was conducted with zileuton
in which significant symptomatic improvement was
noted.5 Lastly, leukotrienes are believed to be contrib-
uting mediators to the pathophysiology of rheumatoid
arthritis, and clinical results with zileuton indicate
significant improvement in some disease parameters.6
These results suggest that blockade of the leukotriene
cascade will be of therapeutic benefit in the treatment
of a number of inflammatory diseases for which the
current therapeutic modalities have proven inadequate.
Recognition that the reduction of multiple lipid media-
tors via the inhibition of 5-LO theoretically offers a
greater therapeutic benefit than a selective leukotriene
antagonist has made the goal of developing effective
5-LO inhibitors an important therapeutic target. Within
the field of 5-lipoxygenase inhibition, multiple strategies
have been developed, and compounds derived from these
efforts are currently being tested in the clinic.7
The 5-lipoxygenase enzyme (5-LO) transforms arachi-
donic acid into leukotriene A4 (LTA4) which can be
further metabolized to the peptidoleukotrienes (LTC4
and LTD4), potent vasoactive and spasmogenic media-
tors capable of inducing increased vascular permeability
and bronchoconstriction, as well as leukotriene B4
(LTB4) and 5-hydroxyeicosatetraenoic acid (HETE),
which function as chemotactic and chemokinetic media-
tors playing a central role in inflammatory cell infiltra-
tion. Current evidence suggests that several inflam-
matory diseases are associated with the presence and
actions of leukotrienes. First, bronchoalveolar lavages
and sputum from patients with asthma are replete with
lipid-derived mediators including peptidoleukotrienes
and LTB4.1 Elevated levels of these mediators have
been associated with the key features of the disease
including, increased vascular permeability, airway ob-
struction, mucus hypersecretion, and leukocytic infiltra-
tion of pulmonary tissue.2 The improvement of acute
airway function in asthmatic patients upon administra-
tion of either a peptidoleukotriene antagonist or a 5-LO
inhibitor is strong evidence supporting the importance
of peptidoleukotrienes in this disease.3 Likewise, the
† Preliminary communication given at 204th ACS Washington
meeting, August 1992, Division of Medicinal Chemistry, abstract 21.
* To whom correspondence should be addressed.
‡ Department of Medicinal Chemistry.
§ Present address: Nycomed Inc., 466 Devon Park Drive, Wayne,
PA 19087.
| Department of Drug Metabolism.
We were particularly intrigued by the observation
that lipophilic N-hydroxyureas which are orally active
inhibitors of leukotriene biosynthesis can be prepared.8
The mechanistic rationale for these compounds first
Present address: Sanofi-Winthrop Pharmaceuticals, 9 Great Val-
ley Parkway, Malvern, PA 19355.
∇ Department of Pharmacology.
X Abstract published in Advance ACS Abstracts, November 15, 1996.
S0022-2623(96)00271-3 CCC: $12.00 © 1996 American Chemical Society

5036 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26
Adams et al.
F igu r e 1. Bicyclic and tricyclic templates.
Ta ble 1. Yield and Analytical Data
Sch em e 1
yield
no.
R1
R2
â-OCH2Ph
â-OCH2(4-MeOPh)
â-OPh
γ-OPh
formula
mp (°C) (%)a,b
1a
1b
1c
1d
2a CH2
2b CH2
2c CH2
2d CH2
O
O
O
O
C16H16N2O4
C17H18N2O5
15H14N2O4
15H14N2O4
175-6
181-2
168-9
170-1
167
166-7
164-5
161-2
160-2
163-4
168-9
187-8
165-6
160-1
175-6
168-9
45
13
12
16
48
10
26
27
26
27
33
35
22
24
44
46
21
30
28
C
C
â-OCH2Ph
â-OCH2(4-MePh)
â-OPh
C17H18N2O3
C18H20N2O4
C16N16N2O3
C16H16N2O3
C18H20N2O3
C12H16N2O3
C11H14N2O2
C18H20N2O3
C19H22N2O4
C18H20N2O3
C17H18N2O2
C17H18N2O3
viz. oximation, reduction, and hydroxyurea formation
(Scheme 1).11 The 6-alkoxybenzofuranones 8a (X ) O)
were efficiently synthesized from resorcinol in three
steps12 (Scheme 2). The 5-alkoxyindanones (8b, X )
CH2) and 6-alkoxytetralones (8c, X ) (CH2)2) were
prepared from the commercially available 5-methoxy-
indanone and 6-methoxytetralone, respectively, by de-
methylation and realkylation with the appropriate alkyl
halide. 6-Phenoxybenzofuranone (10) was prepared by
an intramolecular Friedel-Crafts reaction of (3-
phenoxyphenoxy)acetyl chloride 9 (Scheme 3). 5-Phe-
noxybenzofuranone (12) was prepared by an extension
of a procedure published by J ung et al.13 starting from
the methoxymethyl ether of 4-phenoxyphenol which was
converted to the benzoyl chloride 11 by the directed
metalation strategy.14 The benzoyl chloride 11 was
transformed to the corresponding diazo ketone, which
was treated with glacial acetic acid to afford 12. The
hydroxyureas 1a , 2a , and 3a were resolved by the
previously published procedure from our laboratories.15
The preparation of 5- and 7-phenoxyindanones (14 and
15) was accomplished by an intramolecular Friedel-
Crafts cyclization of 3-(3-phenoxyphenyl)propionic acid
(13).
Ullman coupling was used to synthesize 5-phenoxy-
tetralone (17) from the corresponding phenol (Scheme
4). 6-Phenyltetralone (16) was prepared by a Pd-
catalyzed coupling of 6-hydroxytetralone triflate and
phenylzinc chloride. Pyrrolo[1,2-a]indole skeleton 19
was prepared from the readily available ethyl indole-
2-carboxylates 18 by standard procedures16 involving
condensation with ethyl acrylate and decarbethoxylation
in refluxing acetic acid. The furanones required for the
synthesis of hydroxyureas 5 and 6 were prepared from
R- and â-naphthol, respectively, e.g., the naphthols were
reacted with chloroacetic acid to afford the naphthoxy-
acetic acids 20 and 22 which underwent a clean intra-
molecular Friedel-Crafts reaction to afford 21 and 23.
The hydroxyurea 7 was prepared from the commercially
available 1,2,3,4-tetrahydrophenanthren-1-one.
δ-OPh
3a (CH2)2 â-OCH2Ph
3b (CH2)2 â-OMe
3c (CH2)2
H
3d (CH2)2 R-OCH2Ph
3e (CH2)2 â-OCH2(4-MePh)
3f (CH2)2 γ-OCH2Ph
3g (CH2)2 â-Ph
3h (CH2)2 R-OPh
3i (CH2)2 â-OCH2(4-ClPh)
3j (CH2)2 â-(CH2)2Ph
C18H19N2O3Cl 166-7
C19H22N2O2
162-3
169
3k (CH2)2 â-OCH2-(2-naphthyl) C22H22N2O3
4a
4b
4c
4d
4e Me
4f
H
H
H
H
OCH2Ph
H
Cl
C18H19N3O3
12H13N3O3
12H12N3O2Cl 186-9
14H17N3O2
16H21N3O2
18H17N3O3
19H19N3O3
13H12N2O3
13H12N2O3
15H16N2O2
199-200 40
C
C
C
C
C
C
C
C
C
190-2
30
45
7
15
35
22
18
12
60
Et
209-10
175-7
186-8
181-2
195-6
191-2
192-3
i-Pr
OPh
OPh
H
4g Me
5
6
7
a
Elemental analyses were within (0.4% of the calculated value
b
except where stated otherwise. Conversion of the ketone to the
hydroxyurea (three steps, Scheme 1).
proposed by Corey was bidentate binding of the hydrox-
amate to the ferric ion of 5-LO.9 Recent studies on the
mechanism of the hydroxamic acid/hydroxyurea class
suggest that reduction of ferric to ferrous enzyme, and
not the formation of a stable enzyme inhibitor complex,
may be the actual mechanism of inhibition.10 As
mentioned above, initial clinical studies with the N-
alkyl-N-hydroxyurea, zileuton, suggest that it will have
therapeutic utility in the treatment of asthma and
perhaps IBD and rheumatoid arthritis. However, the
pharmacodynamics of zileuton in humans may require
dosing four times daily for maximum efficacy.7a Hence,
there exists a need to develop 5-lipoxygenase inhibitors
which will have a more extended duration of action in
humans. This paper outlines our efforts to identify an
N-hydroxyurea with the potential for an extended
duration of action in humans.
Ch em istr y
Biologica l Resu lts a n d Discu ssion
Hydroxyureas (Figure 1, 1-7, and Table 1) were
prepared from the corresponding ketones in three steps,
In order to probe the structural requirements for
binding, the templates chosen for investigation were

Bicyclic N-Hydroxyurea Inhibitors of 5-Lipoxygenase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26 5037
Sch em e 2
Sch em e 3
conformationally restricted analogs of the benzylic hy-
droxamic acids and hydroxyureas that had been previ-
ously identified by the work of Summers and co-workers
at Abbott as orally active 5-LO inhibitors.11,17 In
addition to providing some knowledge of the binding,
we thought that conformational constriction might
afford enhanced potency relative to the known com-
pounds in which the N-hydroxyurea was not constrained
in a ring. The investigation was further restricted to
the synthesis of N-hydroxyureas as workers at both
Abbott and Wellcome had demonstrated that the N-
hydroxyurea group afforded superior pharmacokinetic
and hence pharmacological behavior relative to the
acetohydroxaminc acids.8 Because lipophilicity was
known to be very important for good 5-LO inhibition,18
the N-hydroxyureas chosen for the initial study were
prepared with substituents which would afford com-
pounds of similar lipophilicity. This was achieved by
preparing the benzyloxy analogs 1a -4a and the parent
heterocycles for the tricyclic analogs 5-7 (Figure 1,
Table 1).
The IC50 for the inhibition of 5-lipoxygenase was
determined for the semipurified enzyme from a rat
basophilic leukemia cell line (RBL-1) using a continuous
assay based upon oxygen consumption. Previous stud-
ies in our labs employing the oxygen electrode assay had
demonstrated that the rat and recombinant human
enzyme afford similar IC50 values for a number of
structurally diverse 5-LO inhibitors (ref 19 and unpub-
lished observations). Moreover, we have shown that the
oxygen consumption assay affords results comparable
to those obtained by monitoring 5-HPETE production
in a 5-LO-containing supernatant prepared from human
monocytes.19 The benzyloxy analogs 1a -4a , as ex-
pected on the basis of the findings of Summers18 and
their similarity to Abbott-63162 and zileuton (Scheme
1), demonstrated in vitro IC50 values of ∼1-2 µM. The
unsubstituted tricyclic compounds 5-7 were slightly
less active, a result that for compounds 6 and 7 could
be related to reduced lipophilicity (ClogP, Table 2). An
additional assay which was examined early on as an
indicator of the potential efficacy of these compounds
in humans was inhibition of leukotriene biosynthesis
in calcium ionophore-stimulated human whole blood.
This assay has the additional complexities of protein
binding and partitioning that are encountered in the
in vivo setting and has been used as a criteria to predict
the potential utility in humans.20 When examined in
human whole blood, compounds 1a -3a performed well,
displaying IC50 values similar to those obtained on the
isolated enzyme, whereas the activity of 4a was notably
attenuated. The remaining tricyclic compounds, 5-7,
which were weaker inhibitors of the isolated enzyme,
had IC50 values in excess of 10 µM in human whole
blood. These studies established the 5-LO inhibitory
activity of all seven cyclic templates to which the
hydroxyurea moiety was attached. Subsequent studies
focused on the templates represented by compounds
1a -4a .
The importance of optimizing lipophilicity (log P) for
in vitro 5-LO inhibitor activity is illustrated in Table 3
for a series of tetraline derivatives. The less lipophilic
methoxy and unsubstituted compounds (3b and 3c)
were weak inhibitors. For the compounds having the
more lipophilic substituents (3a and 3d -k ), similar IC50
values were obtained regardless of the nature of this
substituent, suggesting that lipophilicity was the main
factor governing affinity at this site remote from the
hydroyurea. The inhibition of LTB4 production in an
ex vivo assay in the mouse was used to examine the
potential of these compounds to inhibit leukotriene
biosynthesis in vivo following oral administration. The

5038 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26
Adams et al.
Sch em e 4
Ta ble 2. In Vitro Evaluation of Hydroxyurea Templates
IC50 (µM)
which had modest activity in the 5-LO assay was active
in the ex vivo assay. Alkyl substitution of the pyrrolo-
indole decreased oral activity for both the 9-unsubsti-
tuted and 7-phenoxy compounds (compare 4b to 4d and
4f to 4g). Finally, analogues of the indan (2) and
dihydrobenzofuran (1) possessing the preferred benzyl-
oxy and phenoxy substituents performed well in the
mouse ex vivo assay for oral activity (Table 5).
compound
5-LO
human whole blooda
calcd log Pb
1a
2a
3a
4a
5
6
7
2.3
0.78
1.8
0.65
4.2
8.0
4.3 (3.2-6.8)
0.64 (0.41-0.80)
1.2 (0.76-1.8)
4.5 (4.0-4.9)
29% at 10 µM
27% at 10 µM
31% at 10 µM
2.4 (1.2-5.5)
3.6
4.7
5.3
4.6
4.7
2.9
2.9
3.6
Having established oral activity for a number of
compounds, we next examined the pharmacological
duration of action. Summers17 had reported data in rats
that demonstrates that a hydroxamic acid inhibitor of
5-LO containing a benzyl ether linkage (Abbott-63162)
was metabolically inactivated by oxidative cleavage of
this linkage. A similar experience was also reported in
the rat for BW-A137C.21 In this latter example, sub-
stitution with an aryl ether improved duration of
action.22 Because these examples suggested that oxida-
tive metabolism would limit the duration of action of
compounds containing the benzyl ether linkage, repre-
sentative examples of benzyl and phenyl ethers were
chosen for pharmacodynamic comparison.
5.6
3.2
zileuton
a
Numbers in parentheses are 95% confidence limits calculated
by linear regression analysis of percent inhibition at the highest
b
concentration tested. Numbers calculated using the Pomona
ClogP program and do not include hydroxyurea fragment.
optimum oral activity was achieved with benzyloxy and
phenoxy substituents (3a , 3e, 3h , and 3i). A further
increase in lipophilicity (3j and 3k ) slightly enhanced
in vitro activity, but was deleterious to ex vivo activity.
Our findings on the optimization of aryl substituent
group closely parallel the observations made by Sum-
mers in the corresponding A-63162 series.17 These
hydroxyureas are poorly water soluble (the measured
aqueous solubility for 3a was 0.67 µg/mL). One possible
explanation for the reduced in vivo activity of 3j and
3k may be that the increased lipophilicity further
decreases aqueous solubility, resulting in solubility-
limited absorption. In order to minimize any variation
due to solubility-limited absorption, all compounds were
dosed in solution (DMA/sesame oil) and not as suspen-
sions. Oral dosing with solutions does not ensure that
the compounds will remain in solution, and it is possible
that 3j and 3k precipitated from solution in the stom-
ach. Interestingly, the 5- and 6-benzyloxy-substituted
tetralins, but not the 7-benzyloxy-substituted tetralin
(3d and 3a , but not 3f) were orally active. Because all
three compounds were equally effective as 5-LO inhibi-
tors, this result also suggests differences in absorption
and metabolism.
Duration of action was examined using the mouse ex
vivo assay, but now extending pretreatment time from
0.5 to 6.0 h. To make this comparison more representa-
tive of the individual pharmacodynamic performance of
each compound, animals were dosed at 1.5 times the
ED50 determined with the initial protocol (0.5 h pre-
treatment). Time courses were performed at these doses
to determine the maximal inhibition achievable with the
compound. Test compounds were then examined to
determine those which best retained their maximal
potency in the ex vivo assay at the 6 h time point. To
our surprise, the duration of pharmacological action in
mice for the benzyl esters, 1a and 3a , was superior to
that of the phenyl ethers, 1d and 2c (Table 6). Ad-
ditional compounds of interest having the benzyl ether
linkage were evaluated in the adoptive peritoneal ana-
phylaxis (APA) reaction in the mouse using a pretreat-
ment time of 7 h. The APA model, which is comple-
mentary to the ex vivo assay, utilizes antigen-specific
IgE to stimulate leukotriene biosynthesis in vivo.23
While not statistically significant at a p < 0.05 due to
abnormally high variability in the vehicle control group,
The relationship between lipophilicity and in vitro/
ex vivo activity seen in the pyrroloindole series is similar
to that seen for the tetralins (Table 4). As was found
for the tetralins, compounds with benzyloxy (4a ) and
phenoxy (4f) substituents demonstrated good ex vivo
activity. In addition the unsubstituted compound (4b)

Bicyclic N-Hydroxyurea Inhibitors of 5-Lipoxygenase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26 5039
Ta ble 3. Tetralins: Correlation of Lipophilicity with in Vitro and ex Vivo 5-LO Inhibition
5-LO
IC50 (µM)
b
compound
position of X
X
calcd log Pa
LTB4 mouse ex vivo % of control [ED50]
3a
3b
3c
3d
3e
3f
3g
3h
6
6
OCH2Ph
OMe
H
OCH2Ph
OCH2(4-OMePh)
OCH2Ph
5.3
3.6
3.6
5.3
5.3
5.3
5.5
5.7
6.0
6.2
6.4
3.6
1.8
48
29
-67***[9.9 (7.9-14.4)]
NS
-30*
-44***
-81***[4.8 (4.0-5.8)]
NS
-52***
-57***
-70*** [4.5 (1.8-8.6)]
NS
NS
5
6
7
6
5
6
6
6
0.91
0.5
1.0
3.0
0.84
2.2
0.6
0.5
3.2
Ph
OPh
3i
3j
OCH2(4-ClPh)
(CH2)2Ph
OCH2(2-naphthyl)
3k
zileuton
-78***[1.0 (0.05-1.8]
a
b
Numbers calculated using the Pomona ClogP version 3.4 program and do not include hydroxyurea fragment. Compounds screened
at a dose of 10 mg/kg, po; ED50 values (mg/kg) along with 95% confidence limits (shown in parentheses) were determined for compounds
demonstrating 60% or greater inhibition at the screen dose and were calculated by linear regression analysis; statistical significance,
which was judged by Student’s t test versus vehicle control, is given as *p < 0.05; **p > 0.01; ***p < 0.001 or NS for no stastistically
significant inhibition at the screen dose.
compounds.21 However, in the ex vivo LTB4 assay of
Ta ble 4. Pyrrolo[1,2-a]indole Inhibitors of 5-LO
the individual enantiomers, notable differences were
seen. The separation of activity was most clearly seen
for the (+) and (-) enantiomers of 1a which produced
ED50 values of 0.9 and 22 mg/kg, respectively. On the
basis of these findings for the individual enantiomers
and the duration of action studies, (+)-1a (SB 202235)
was chosen as the best overall compound. In order to
determine if the profile of (+)-1a met our criteria of the
potential for improved pharmacodynamic performance
in humans, more detailed studies focusing on the
pharmacokinetics and metabolism of (+)-1a were initi-
ated.
calcd
5-LO
LTB4 mouse ex vivo
b
compound R1
R2
OCH2Ph 4.6
log Pa IC50 (µM) % of control [ED50
]
4a
4b
4c
4d
4e
4f
H
H
H
H
0.65
10
1.3
1.7
1.7
-68***[1.8 (0.9-55)]
H
C;
Et
2.8
3.6
4.0
5.0
5.0
5.7
-74***c
not tested
NS
Me i-pr
OPh
Me OPh
NS
-58***
-26***
H
0.52
0.45
4g
P h a r m a cok in etic Ch a r a cter iza tion of th e En a n -
tiom er s of 1a . Assuming that 1a has no pharmaco-
logically active metabolites, the inhibition results from
the ex vivo LTB4 assay should correlate with the
pharmacokinetics of the compound. To test this the oral
pharmacokinetics of (+)-1a and (-)-1a (20 mg/kg) were
characterized following oral administration of a suspen-
sion in 0.5% tragacanth to mice. The maximal plasma
concentrations and AUCs achieved for (+)-1a were 2.6
µg/mL and 21 µg‚h/mL, and for (-)-1a these values were
1.0 µg/mL and 7.5 µg‚h/mL (Figure 2). The approxi-
mately 60% lower Cmax and AUC for (-)-1a relative to
(+)-1a following oral administration explains, at least
in part, the differences in the oral pharmacology (ex vivo
LTB4) of the two compounds in the mouse. Further-
more, the difference in oral pharmacokinetics appears
to be due to enantioselective differences in clearance as
intravenous studies indicated that (+)-1a was cleared
from plasma much slower than (-)-1a (data not shown).
The oral pharmacokinetics of (+)-1a compares favorably
with published data on zileuton in the mouse.8a Similar
AUCs and half-lives were observed for both compounds.
a,b
Same as Table 3. c ED50 not determined because of poor IC50
for 5-LO inhibition.
2a showed a clear inhibitory trend and approached
statistical significant inhibition at p < 0.10. This trend
was not seen with any of the compounds tested in the
APA model, other than 1a , which strongly inhibited the
response (55%, p < 0.01). In addition to confirming the
duration of action demonstrated for 1a in the ex vivo
LTB4 assay, the APA model identified the (benzyloxy)-
dihydroindan (2a ) as a compound with an extended
duration of action (Table 7). In summary these two in
vivo models identified three compounds (1a , 2a , and 3a )
with an improved pharmacodynamic profile relative to
the standard, zileuton.
For these initial studies all of the compounds had
been examined as their racemates. As there are many
potential advantages to developing homochiral thera-
peutic agents, we chose to examine the importance of
chirality on both the in vitro and in vivo potency of the
three lead compounds before proceeding further. No
greater than a 2-fold difference was observed in the
oxygen electrode assay IC50 values of the individual
enantiomers of the three enantiomeric pairs (Table 8).
These findings are in accordance with previous observa-
tions demonstrating that the chirality of the asymmetric
center adjacent to the hydroxyurea does not substan-
tially influence the 5-LO inhibitory activity of these
Differences favoring (+)-1a were also seen in the dog
following iv administration of the enantiomers (Table
9). As the clinical data for zileuton24 suggested that the
pharmacokinetics in the dog was the least predictive of
the four species examined preclinically,8a further phar-

5040 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26
Adams et al.
Ta ble 5. Benzyl and Phenyl Ethers of Indanyl and Dihydrobenzofuranyl 5-LO inhibitors
5-LO
IC50 (µM)
LTB4 mouse ex vivo
% of control[ED50]
a
compound
position of subst
R
Z
1a
1b
1c
1d
2a
2b
2c
2d
1
1
1
2
1
1
1
3
CH2Ph
CH2(4-MeOPh)
Ph
Ph
CH2Ph
CH2(4-MeOPh)
Ph
Ph
O
O
O
O
CH2
CH2
CH2
CH2
2.3
2.7
3.3
3.5
0.78
1.4
0.85
5.6
-100***[3.9 (3.1-4.9)]
-73***[9.8 (5.6-90)]
-56***
-64***[1.9 (1.0-3.8)]
-85***[2.6 (1.803.4)]
NS
-88***[6.4 (4.7-9.0)]
NT
a
Same as Table 3.
Ta ble 6. Duration of Action Comparison in the Mouse Using
the ex Vivo LTB4 Assay
maximum
time of
dose
inhibition
maximum
% inhibition of
LTB4 inhibition LTB4 of 6 ha
a
compound (mg/kg) of LTB4
1a
3a
1d
2c
6
15
3
10
2
91***
61***
67***
88***
70***
1
2
0.5
1
1
87***
67***
0 NS
57***
40***
zileuton
a
***p ) <0.001.
Ta ble 7. Duration of Action Comparison in the Mouse
Adoptive Peritoneal Anaphylaxis Assay
5-LO
IC50 (µM)
ex vivo
screen ED50
APA % change
from controlsb
a
compound
F igu r e 2. Mean plasma concentrations of (+)- and (-)-1a in
1a
2a
4a
3e
2.3
3.9
4.0
1.8
4.8
4.5
1.0
-55**
-31c
-8 NS
-13 NS
0 NS
male CD-1 mice following oral administration (20 mg/kg).
0.78
0.65
0.5
2.2
3.2
Ta ble 9. Plasma Clearance Values and Half-Lives following
Intravenous Administrationa
3i
zileuton
-10 NS
(+)-1a
CLp
(-)-1a
CLp
zileutonb
a
b
t1/2
t1/2
t1/2
Data from Tables 3-5. Following a 7 h treatment with the
compounds dosed at 1.5 times the ex vivo ED50 or vehicle (DMA/
oil), sensitized 1BD8 cells were injected ip. Five minutes later
ARS-BSA was injected, and following an additional 5 min the
animals were sacrificed and the LTB4 levels of the peritoneal
determined according to published procedures.22 c Approached
statistical significance at a p < 0.10 (t ) 1.3353, df ) 15).
mouse
dog
monkey
8.6
3.5
6.1
1.0
2.9
6.0
IDc
10.9
10.3
ID
1.1
1.1
0.9
7.5
0.3
a
Abbreviations and units of pharmacokinetic parameters are
as follows: CLp ((mL/min)/kg), apparent plasma clearance; t1/2 (h),
apparent terminal half-life. Compounds were administered in-
a
travenously as a solution. Obtained from ref 8a. b ID, insufficient
Ta ble 8. Selection of Preferred Enantiomer
data to calculate parameter.
5-LO
IC50 (µM)
LTB4 mouse ex vivo,
compound
ED50 (mg/kg)a
following administration of an aqueous solution in 30%
(w/v) (2-hydroxypropyl)-â-cyclodextrin. Maximal plasma
concentrations (1.3 µg/mL) were reached approximately
2 h after drug administration, and (+)-1a disappeared
from plasma with an apparent terminal half-life of
approximately 10 h. The longer half-life following oral
administration for (+)-1a when compared to the half-
life measured after iv administration suggested that the
apparent terminal half-life is a reflection of prolonged
absorption rather than elimination. The mean AUC0-t
for (+)-1a was approximately 13 µg‚h/mL. Comparison
of the oral pharmacokinetic data of (+)-1a (10 mg/kg)
with published data on zileuton (20 mg/kg) indicates
that systemic exposure of (+)-1a , based upon AUCs of
parent drug, was almost 3-fold greater than that of
zileuton even though a higher dose of zileuton was
administered.8a The mean bioavailability was good, but
somewhat variable (>50% in two monkeys and ap-
proximately 20% in a third).
(+)-1a
(-)-1a
(+)-2a
(-)-2a
(+)-3a
(-)-3a
1.9
2.0
0.88
0.62
0.76
1.3
0.9 (0.5-1.3)
22 (10.3-170)
1.9 (1.6-2.1)
0.9 (0.57-1.2)
28 % inhibition at 10 mg/kg
6.6 (4.7-9.7)
a
95% confidence limits (shown in parentheses) were calculated
by linear regression analysis.
macokinetic characterization of (+)- and (-)-1a in the
dog was not pursued.
Unlike mouse and dog, the differences in plasma
clearance between (+)-1a and (-)-1a in the monkey
were small (6.1 and 10.3 (mL/min)/kg, respectively).
However, a longer half-life was observed for (+)-1a (6
h) when compared with that of (-)-1a (1 h) following iv
administration presumably due to a larger volume of
distribution for (+)-1a . The oral pharmacokinetics of
(+)-1a , the more active enantiomer, were characterized

Bicyclic N-Hydroxyurea Inhibitors of 5-Lipoxygenase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26 5041
glucuronidation forming a small amount of the 14C-
labeled (+)-1a glucuronide at a rate of 2.5 (nmol/mg of
protein)/h. However, the (-) enantiomer of 1a formed
the glucuronide at a much higher rate of 60.3 (nmol/
mg of protein)/h. The rate with (-)-1a was approxi-
mately 10-fold higher than was observed in mouse
microsomes. On the basis of the in vitro data, enantio-
selective differences in metabolic clearances of (+)- and
(-)-1a in humans may be similar to or even larger than
in the mouse. The metabolism of zileuton was also
examined in mouse and human liver microsomes. An
apparent glucuronide of zileuton was formed in mouse
microsomes, at a rate of 0.7 (nmol/mg of protein)/h. This
glucuronide was also formed in human liver micro-
somes, at a rate of 6.8 (nmol/mg of protein)/h.
Ta ble 10. Glucuronidation Rates for (+)- and (-)-1 and
Zileuton in Mouse and Human Liver Microsomes
test
system
rate of formation
compound
((nmol/mg of protein)/h)a
mouse
human
(+)-1a
(-)-1a
zileuton
(+)-1a
(-)-1a
zileuton
none detected
5.9
0.7
2.5
60.3
6.8 (0.7)b
a
Values are the mean of duplicate determinations. b Along with
the major glucuronide evidence for formation of a second
glucuronide at a reduced rate was seen.
a
Pharmacokinetic studies in the rat following the oral
administration of (+)- and (-)-1a (20 mg/kg as a
suspension in 0.5% tragacanth) were also performed.
Low (generally <200 ng/mL) and erratic plasma con-
centrations were noted up to 24 h after dosing, which
made it impossible to calculate any relevant pharma-
cokinetic parameters. Similarly, both (+)- and (-)-1a
rapidly disappeared from plasma following iv adminis-
tration, and again no relevant pharmacokinetics could
be calculated. The lack of significant plasma concentra-
tions for the enantiomers of 1a is consistent with the
performance in the rat ex vivo LTB4 assay in which poor
and erratic inhibition was observed. Of the four species
in which pharmacokinetics were examined, the rapid
clearance of 1a proved unique to the rat.
In Vitr o Meta bolism . The selection of (+)-1a as the
preferred enantiomer was initially based upon the
results of the ex vivo LTB4 assay in the mouse. The
subsequent pharmacokinetic data from three species
provided further support for this choice (Table 9).
Nevertheless, without any understanding of the origin
of this difference, one could not predict with confidence
which would be the preferred enantiomer in humans.
Experiments in mouse hepatic microsomes were con-
ducted to investigate if the difference in clearances of
(+)- and (-)-1a in the mouse were the result of enantio-
selective metabolism. Originally, cytochrome P-450-
catalyzed oxidation of (+)- and (-)-1a was investigated;
however, no evidence was found for the involvement of
P-450 enzymes in the metabolism of these enantiomers.
Glucuronidation was examined next as this mechanism
had been reported as an important pathway of metabo-
lism for zileuton.25 Incubating mouse microsomes with
(-)-1a and [14C]UDP-glucuronic acid (UDPGA) resulted
in the formation of a labeled metabolite that was
identified as a glucuronide of (-)-1a using quadrupole
MS-MS. This metabolite was produced at an apparent
rate of 5.9 (nmol/mg of protein)/h (Table 10). In
contrast, there was very little, if any, 14C-labeled glu-
curonide of (+)-1a formed under identical incubation
conditions. Appropriate controls were run to verify that
the formation of the glucuronide metabolite was not an
artifact of the in vitro incubations, i.e., the metabolite
was not produced in the absence of (-)-1a or mi-
crosomes. These studies suggest that the difference in
clearances of (+)- and (-)-1a in the mouse could be at
least partially due to metabolism via hepatic stereo-
selective glucuronidation.
Con clu sion s
The 5-LO inhibitory properties of seven cyclic tem-
plates possessing the hydroxyurea moiety have been
examined. Although not rigorously evaluated, it ap-
pears that attaching the hydroxyurea moiety to a ring
did not enhance 5-LO inhibition relative to less confor-
mationally restricted compounds, such as zileuton and
A-63162. Moreover, the enzyme showed very little
discrimination between the individual enantiomers in
the three cases examined (1a -3a ). These results are
consistent with a redox mechanism of inhibition, as this
would require only transient association of the inhibitor
with the enzyme.10 A nonredox mechanism involving
formation of a stable complex, presumably with the
active-site iron, would be expected to be more sensitive
to the immediate environment of the hydroxyurea. The
lag phase seen in the kinetics of inhibition using the
oxygen electrode assay is further support for the in-
volvement a redox active species in the mechanism of
inhibition.19
The protocol for selection of a compound of potential
therapeutic interest began with the determination of
IC50 for inhibition of the semipurified 5-LO enzyme from
RBL-1 cells in combination with the inhibition of
ionophore-induced LTB4 generation in whole human
blood. Analogues of those bicyclic systems (1a -4a )
having good activity in both assays were examined for
their ability to block LTB4 production in a murine ex
vivo assay, and the best compounds from this assay were
then chosen based upon duration of action studies.
From an evaluation of individual enantiomers of the
most promising racemic compounds, (+)-1a (SB 202235)
was selected, and further characterization was under-
taken to evaluate the compound’s suitability as a
potential clinical development candidate. Pharmaco-
kinetic studies in three species demonstrated enantio-
specific differences between (+)-1a and (-)-1a . In all
three species, (+)-1a was eliminated more slowly and
was clearly superior to (-)-1a . In the mouse a good
correspondence between and pharmacokinetics and
pharmacodynamics (ex vivo assay) was established. The
demonstration of a much higher rate of glucuronidation
for (-)-1a when compared to that for (+)-1a in mouse
hepatic microsomes supports enantiospecific glucuron-
idation as the probable mechanism to explain the
enantioselective clearance. Enantioselective glucuron-
idation favoring (-)-1a was also found in human liver
microsomes. These data clearly pointed to the selection
of (+)-1a over (-)-1a as the preferred enantiomer for
clinical development.26
Since the ultimate goal of this work was to select the
best compound for use in humans, the glucuronidation
of (+)- and (-)-1a was also examined in human liver
microsomes. As was found with the mouse microsomes,
the (+)-enantiomer of 1a was a poor substrate for

5042 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26
Adams et al.
(dd, 1H), 6.55 (d, 1H), 5.17 (s, 2H), 5.07 (s, 2H), 4.07 (br s,
1H). The oxime (118 g, 0.45 mol) in 1:1 MeOH/CH2Cl2 (2.3 L)
was reduced to the hydroxylamine (100 g, 86%), which was
reacted with trimethylsilyl isocyanate (77 mL, 0.57 mol) to
afford 1a as a white powder (62 g, 53%). Mp: 174.5-175.5
°C; 1H NMR (CDCl3, MeOH-d4): δ 7.45-7.25 (m, 5H), 7.14 (d,
1H), 6.51 (dd, 1H), 6.41 (d, 1H), 5.87 (t, 1H), 5.03 (s, 2H), 4.53
(d, 1H). IR (cm-1): 3460, 3320, 3180, 2880, 1650-1620.
(S)-(+)- a n d (R)-(-)-N-Hyd r oxy-N-[2,3-d ih yd r o-6-(ben -
zyloxy)ben zofu r a n -3-yl]u r ea [(+)-1a a n d (-)-1a ]. To a
solution of (()-N-hydroxy-N-[2,3-dihydro-6-(benzyloxy)benzo-
furan-3-yl]urea (37.87 g, 0.12 mol) in CH2Cl2 (750 mL) at 0 °C
was added triethylamine (32 mL, 0.23 mol). After 10 min of
stirring, (R)-4-benzyl-2-oxazolidinone 3-carbamoyl chloride
(43.80 g, 0.18 mol) was added, and the mixture was stirred at
0 °C for 1 h. Water was added, and the organic phase was
washed with brine. The solvent was evaporated, and the
residue was purified by flash chromatography (1% MeOH/CH2-
Cl2) to give the desired diasteriomers (47.04 g, 76% yield) that
were separated by preparative HPLC (Chiralcel OJ , 4.6 mm
× 250 mm, 1:1 EtOH/hexane, 1.0 mL/min, UV 210 nm), with
diasteriomer I having a retention time of 9.72 min and
diasteriomer II having a retention time of 13.68 min.
To a solution of diasteriomer I from the above procedure
(15.3 g, 29.6 mmol) in THF/H2O (4:1, 300 mL) at 0 °C was
added dropwise hydrogen peroxide (12.5 mL, 116.1 mmol,
30%). Lithium hydroxide (0.716 g, 45.5 mmol) dissolved in
H2O (15 mL) was added, and the mixture was stirred at 0 °C
for 1 h. The reaction was quenched with saturated aqueous
NaHSO3, and the mixture was diluted with EtOAc. The
organic phase was washed with brine, and the solvent was
evaporated. The residue was triturated with Et2O and then
recrystallized from EtOH to give (+)-1a (3.27 g, 35% yield).
Mp: 189-190 °C. [R]D: +10.1° in methanol. Following the
same procedure the diastereomer II yielded (-)-1a . Mp: 188-
189 °C. [R]D: -10.7° in methanol.
N-Hyd r oxy-N-[2,3-d ih yd r o-5-(b en zyloxy)-1H-in d en -1-
yl]u r ea (2a ). Using the above general procedure 8b was
converted to hydroxyurea 2a . Mp: 167 °C. 1H NMR (250
MHz, DMSO): δ 8.89 (s, 1H), 7.40 (m, 4H), 7.31 (m, 1H), 7.05
(d, 1H), 6.84 (s, 1H), 6.78 (dd, 1H), 6.36 (s, 2H), 5.58 (m, 1H),
5.07 (s, 1H), 2.88 (m, 1H), 2.71 (m, 1H), 2.17 (m, 1H), 2.09 (m,
1H).
Pharmacodynamic studies in the mouse demonstrated
that compounds containing a benzyl ether linkage were
equivalent, if not superior, to compounds possessing the
phenyl ether linkage. The superior in vivo activity of
the benzyl ethers in the mouse was unexpected in light
of the reported metabolic lability of related benzyl ether
hydroxyureas in the rat. In agreement with these
reports, the benzyl ether 1a was rapidly cleared follow-
ing iv administration in the rat. However, 1a was not
rapidly inactivated in the two additional species which
were examined (dog and monkey). These data suggest
that oxidative cleavage of the benzylic ether is the rate-
limiting metabolic pathway in the rat for clearance of
1a , but that oxidation of the benzylic linkage may be
less important in other species. In support of this
proposal, glucuronidation and not cytochrome P-450
oxidation was the predominant metabolic pathway
demonstrated in both mouse and human liver micro-
somes.
Glucuronidation is the major clearance pathway of
zileuton in humans, accounting for as much as 70% of
the administered dose. If the relative rates of glucuron-
idation demonstrated for (+)-1a and zileuton in vitro
are reflective of the relative ease of glucuronidation in
humans, then (+)-1a (SB 202235) should demonstrate
an improved duration of action relative to that of
zileuton.27,28
Exp er im en ta l Section
Ch em istr y: Gen er a l P r oced u r es. Melting points were
determined with a Thomas-Hoover melting point apparatus
and are uncorrected. Infrared spectra were recorded on a
Perkin-Elmer Model 783 or 683 spectrophotometer. 1H NMR
spectra were recorded on a Bruker Model AM 250. Chemical
shifts are reported as parts per million on the δ scale relative
to tetramethylsilane as internal standard. Data are presented
as follows: chemical shift (multiplicity, integrated intensity,
coupling constants, assignment). Unless otherwise noted, the
NMR solvent was CDCl3. Chromatography refers to purifica-
tion by flash chromatography on E. Merck silica gel 60 (230-
400 mesh). When air or moisture sensitive reagents were
used, reactions were run under argon, and all aqueous
workups culminated in washing the organic layer with water
and brine, drying over magnesium sulfate, filtering, and
concentrating at reduced pressure.
(+)- a n d (-)-N-Hyd r oxy-N-[2,3-d ih yd r o-5-(ben zyloxy)-
1H-in d en -1-yl]u r ea [(+)-2a a n d (-)-2a ]. The resolution was
performed using the above general procedure described above
for 1a . Data for (+)-2a . Mp: 167-168 °C. [R]D: +36.02° in
methanol. Data for (-)-2a . Mp: 165.6-166.5 °C. [R]D:
-39.2° in methanol.
N-Hydr oxy-N-[1,2,3,4-tetr ah ydr o-6-(ben zyloxy)-1-n aph -
th a len yl]u r ea (3a ). Using the above general procedure, 8c
was converted to hydroxyurea 3a . Mp: 160-162 °C. 1H NMR
(250 MHz, MeOD): δ 7.37 (m, 5H), 7.20 (d, 1H), 6.80 (dd, 1H),
6.71 (d, 1H), 5.43 (t, 1H), 5.05 (s, 2H), 2.90-2.63 (m, 2H), 2.00
(m, 3H), 1.78 (m, 1H).
(+)- a n d (-)-N-Hyd r oxy-N-[1,2,3,4-tetr a h yd r o-6-(ben -
zyloxy)-1-n a p h th a len yl]u r ea [(+)-3a a n d (-)-3a ]. The
resolution was performed using the above general procedure
described above for 1a . Data for (+)-3a . Mp: 154.5-155.5
°C. [R]D: +29.8° in methanol. Data for (-)-3a . Mp: 146-
147 °C. [R]D ) -15.3° in methanol.
6-(Ben zyloxy)-3-oxo-2,3-d ih yd r oben zofu r a n (8a ). The
requisite 2,3-dihydro-6-hydroxy-3-oxobenzofuran was prepared
using a modification of the procedure of Shriner.12 To a
solution of 2,3-dihydro-6-hydroxy-3-oxobenzofuran (363 g, 2.42
mol) in DMF (4 L) was added anhydrous potassium carbonate
(668 g, 4.84 mol). After 5 min of stirring at room temperature,
benzyl bromide (582 g, 3.40 mol) was added to the mixture
dropwise over 15 min. The resulting mixture was stirred at
room temperature for 18 h, at which time the potassium
carbonate was removed by filtration and washed with DMF.
The combined organic material was poured into cold H2O (12
L) and stirred. The solid was collected by filtration, washed
with H2O (4 L), and dried to provide the benzyl ether (554 g,
100%).
Gen er a l P r oced u r e for Con ver sion of Cycloa lk a n on es
to Hyd r oxyu r ea s (Sch em e 1). To a solution of the cyclo-
alkanone in anhydrous pyridine (0.5-1.0 M solution) was
added hydroxylamine hydrochloride (2.0 equiv), and the
mixture was stirred for at 50 °C for 1-2 h. The solution was
cooled to room temperature and poured into water. The thick
solid, which forms slowly, is filtered, washed with water, and
dried to afford the required oxime usually as a mixture of E
and Z isomers. The oximino compound obtained above was
dissolved in a suitable solvent (such as EtOH, CH2Cl2/EtOH
or CH2Cl2/MeOH, 0.1-0.5 M solution), and BH3-pyridine (3-7
equiv) was added. Hydrogen chloride was added as an
aqueous or ethanolic solution (6 N), and the resulting solution
was stirred at room temperature until the starting oxime was
completely consumed. The product was isolated either by
extraction or as the hydrochloride salt. The hydroxylamine
was dissolved in THF (0.1-0.4 M solution), and trimethylsilyl
isocyanate (2 equiv) was added. The reaction was stirred at
55 °C until the starting hydroxylamine was consumed. The
reaction mixture was cooled, and the product was isolated by
crystallization or extraction. A final recrystallization from a
suitable solvent mixture afforded pure hydroxyurea.
N-Hyd r oxy-N-[2,3-d ih yd r o-6-(b en zyloxy)b en zofu r a n -
3-yl]u r ea (1a ). Using the above general procedure, 8a (116
g, 0.5 mol) was converted to the oxime as a yellow solid (118
g, 92%). 1H NMR (CDCl3): δ 7.45 (d, 1H), 7.40 (m, 5H), 6.64

Bicyclic N-Hydroxyurea Inhibitors of 5-Lipoxygenase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26 5043
5-(Ben zyloxy)-1-in d a n on e (8b). To a solution of 5-hy-
droxy-1-indanone29 (8.02 g, 54.2 mmol) in dry DMF (120 mL)
under an argon atmosphere was added slowly sodium hydride
(1.64 g of 80% suspension in mineral oil, 59.6 mmol). After
the evolution of hydrogen ceased, benzyl chloride (7.12 mL,
60.0 mmol) was added, and the resulting mixture was stirred
for 15 min. The reaction mixture was concentrated under
reduced pressure, and the residue was partitioned between
EtOAc and 1:1 saturated aqueous NaCl/3 N HCl. The organic
extract was wsahed with saturated aqueous NaCl and dried
(MgSO4). The solvent was removed in vacuo, and the residue
was used without further purification.
6-(Ben zyloxy)-1-tetr a lon e (8c). To a solution of 6-hy-
droxy-1-tetralone29 (9.2 g, 56.8 mmol) in DMF (150 mL) was
added potassium hydride (2.49 g, 62 mmol). When the
evolution of hydrogen subsided, benzyl bromide (10.6 g, 62
mmol) was added. After 2 h of stirring at room temperature,
the reaction mixture was concentrated under reduced pressure.
The residue was dissolved in EtOAc and washed successively
with 3 N HCl, H2O, and saturated aqueous NaCl. Removal
of the solvent in vacuo and purification by flash chromatog-
raphy eluting with a gradient of 0-100% CH2Cl2/hexanes
yielded the desired product (9.7 g, 73%). The infrared spec-
trum of the product indicated a conjugated ketone at 1665-
1685 cm-1. The NMR spectrum indicated the presence of the
benzyl methylene at δ 5.0 and aromatic benzyl protons at δ
7.4.
(3-P h en oxyp h en oxy)a cetic a cid (9). To a solution of
3-phenoxyphenol (4.48 g, 24.1 mmol) in acetone (20 mL) was
added potassium carbonate (4.00 g, 29.0 mmol) and ethyl
bromoacetate (4.84 g, 29.0 mmol). After overnight reflux, the
mixture was cooled and then filtered. The solvent was
evaporated, and the residue was dissolved in EtOAc and
washed with H2O and brine. The solvent was evaporated,
yielding a yellow oil (5.14 g) that was used without further
purification. 1H NMR (250 MHz, CDCl3): δ 7.31 (m, 2H), 7.20
(t, 1H), 7.10 (m, 1H), 7.00 (m, 2H), 6.60 (m, 3H), 4.54 (s, 2H),
4.19 (q, 2H), 1.22 (t, 3H). To a solution of this compound in
1:4 H2O/THF (10 mL) was added lithium hydroxide (0.188 g,
7.83 mmol). After 1 h of stirring the mixture was neutralized
with 3 N HCl and diluted with H2O. The product was
extracted from the aqueous phase with EtOAc. The organic
phase was washed with brine and evaporated, yielding a red
oil (0.739) that was used without further purification. 1H NMR
(250 MHz, CDCl3): δ 7.30 (m, 2H), 7.21 (t, 1H), 7.10 (m, 1H),
7.00 (m, 2H), 6.60 (m, 3H), 4.61 (s, 2H).
(6-P h en oxy)ben zofu r a n -3-on e (10). To a solution of (3-
phenoxyphenoxy)acetic acid (0.739 g, 2.61 mmol) in benzene
(3 mL) was added oxalyl chloride (3 mL, 45.8 mmol). The
mixture was refluxed 0.5 h. The solution was concentrated,
and the residue was dissolved in benzene (10 mL). Aluminum
chloride was added (0.46 g, 3.48 mmol), and the mixture was
stirred at room temperature for 1 h. The mixture was poured
over ice, and the organic layer was separated and washed with
brine. The solvent was evaporated, and the residue was
purified with flash chromatography (1:1 hexane/CH2Cl2) to give
the desired product (0.12 g, 20% yield). 1H NMR (250 MHz,
CDCl3): δ 7.60 (d, 1H), 7.42 (m, 2H), 7.24 (m, 2H), 7.09 (m,
1H), 6.70 (dd, 1H), 6.50 (d, 1H), 4.60 (s, 2H).
was stirred until the gas evolution ceased. After evaporation
of the solvent, the residue was dissolved in glacial acetic acid
(20 mL) and stirred for 30 min at room temperature. The
solvent was removed, and the crude product was purified by
flash chromatography on silica gel (1:10 hexane/EtOAc) to
afford the benzofuranone (0.83 g, 50%). 1H NMR (CDCl3): δ
7.40-6.96 (m, 8H), 4.65 (s, 2H).
3-(3-P h en oxyp h en yl)p r op a n oic Acid (13). To a suspen-
sion of sodium hydride (1.36 g of 60% dispersion in mineral
oil, 33.9 mmol) in DMF (30 mL) under an argon atmosphere
at 0 °C was added, over 10 min, a solution of diethyl malonate
(5.42 mL, 35.7 mmol) in DMF (20 mL). The reaction mixture
was stirred at 0 °C for 30 min, at which time a solution of
3-phenoxybenzyl bromide30 (4.7 g, 17.9 mmol) in DMF (25 mL)
was added. The reaction mixture was allowed to warm to room
temperature and stirred for 1 h. The mixture was partitioned
between Et2O and aqueous NH4Cl, and the organic extract was
washed successively with H2O and saturated aqueous NaCl
and dried (MgSO4). The solvent was removed in vacuo, and
the residue was purified by flash chromatography, eluting with
10% EtOAc/hexanes to afford diethyl (3-phenoxybenzyl)-
malonate as a colorless oil (4.17 g, 68%).
To a solution of diethyl (3-phenoxybenzyl)malonate (4.17 g,
12.2 mmol) in EtOH (100 mL) was added 1 M NaOH (61 mL,
60.9 mmol). The mixture was heated at reflux overnight and
then allowed to cool, and the pH was adjusted to pH 4 by the
addition of 1 M HCl. The mixture was diluted with CH2Cl2,
and the organic layer was washed successively with H2O and
saturated aqueous NaCl and dried (MgSO4). The solvent was
removed in vacuo, and the residue was dissolved in acetic acid
and heated at reflux overnight. The reaction mixture was
allowed to cool and was partitioned between CH2Cl2 and H2O.
The aqueous phase was extracted with CH2Cl2 (2 × 100 mL),
and the combined organic extracts were washed successively
with H2O and saturated aqueous NaCl and dried (MgSO4). The
solvent was removed in vacuo to afford 13 as a pale yellow oil
(2.67 g, 90%).
5-P h en oxy-1-in d a n on e a n d 7-P h en oxy-1-in d a n on e (14
a n d 15). To a round-bottom flask fitted with a mechanical
stirrer was added polyphosphoric acid (600 g). To this was
added, over 1 h, 3-(3-phenoxyphenyl)propanoic acid (49 g, 0.202
mol), maintaining the temperature at 75 °C. Upon completion
of the addition, the mixture was stirred for 1 h at 80 °C, then
cooled to 0 °C, and diluted with ice cold H2O. Ether was added,
and the mixture was stirred for 30 min. The layers were
separated, and the aqueous phase was extracted with Et2O (2
× 200 mL). The combined organic extracts were washed
successively with H2O, saturated K2CO3, H2O, and saturated
aqueous NaCl and dried (K2CO3). The solvent was removed
in vacuo, and the residue was purified in portions by flash
chromatography, eluting with 10% EtOAc/hexanes. The major
component was recrystallized from cyclohexane to afford
5-phenoxy-1-indanone 14 (4.63 g, 10%). Mp: 67-69 °C. 1H
NMR (CDCl3): δ 7.72 (d, 1H), 7.42 (t, 2H), 7.22 (m, 1H), 7.10
(d, 1H), 6.98 (d, 1H), 6.93 (s, 1H), 3.06 (dd, 2H), 2.71 (dd, 2H).
A minor component was also isolated and recrystallized from
cyclohexane to afford 7-phenoxy-1-indanone (15) (548 mg, 1%).
Mp: 102-103 °C.
6-P h en yl-1-tetr a lon e (16). To a solution of zinc chloride
(5.1 mL of 1.0 M solution, 5.1 mmol) in dry THF (20 mL) was
added phenyllithium (2.6 mL of 2.0 M solution, 5.1 mmol). The
mixture was stirred at room temperature for 30 min and added
to a solution containing 6-(1-tetralonyl (trifluoromethyl)-
sulfonate (1.09 g, 3.7 mmol), palladium acetate (7.6 mg, 0.03
mmol), and 1,3-bis(diphenylphosphino)propane (14 mg, 0.03
mmol) in dry THF (50 mL). The mixture was stirred at room
temperature for 1 h and then partitioned between EtOAc and
3 N HCl. The organic extract was washed with saturated
aqueous NaCl and dried (MgSO4). The solvent was removed
in vacuo, and the residue was purified by flash chromatogra-
phy, eluting with 10% EtOAc/hexanes to provide the desired
product which was recrystallized from hexanes (0.42 g, 51%).
1H NMR (CDCl3): δ 8.10 (d, 1H), 7.66-7.35 (m, 7H), 3.03 (t,
2H), 2.70 (t, 2H), 2.20 (m, 2H).
5-P h en oxyben zofu r a n on e (12). To a solution of meth-
oxymethyl 4-phenoxyphenyl ether (2.9 g, 12.6 mmol) in
anhydrous THF (20 mL) was added n-BuLi (6.0 mL, 2.5 M in
hexane) at -78 °C over 10 min. The mixture was stirred at
-78 °C for 2 h, and dry CO2 was bubbled through the reaction
mixture for 10 min. The reaction mixture was allowed to
warm to room temperature over 2 h and was partitioned
between saturated NH4Cl and EtOAc. The organic extract was
dried and concentrated to afford the product (2.4 g, 70%), which
was used without any purification.
A solution of the benzoic acid (2.0 g, 7.3 mmol) in oxalyl
chloride (15 mL) was refluxed for 40 min. The excess oxalyl
chloride was removed in vacuo to afford the required benzoyl
chloride which was used in the next step without any further
purification. IR (neat): 3010, 2970, 1775, 1470, 1310. Freshly
prepared ethereal diazomethane solution was added to a
solution of the benzoyl chloride in CH2Cl2, and the mixture
5-P h en oxy-1-tetr a lon e (17). To a solution containing
5-hydroxy-1-tetralone (970 mg, 6.0 mmol) and iodobenzene (4.0

5044 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26
Adams et al.
mL, 35.7 mmol) in DMF (12 mL) was added slowly with cooling
sodium hydride (150 mg, 6.25 mmol). The mixture was heated
until dissolution occurred and then allowed to cool. To the
mixture was added slowly with cooling cuprous chloride (600
mg, 6.1 mmol), followed by tris[2-(2-methoxyethoxy)ethyl]-
amine (0.68 mL, 2.1 mmol). The mixture was heated at 145-
150 °C overnight and then allowed to cool. The reaction
mixture was partitioned between 3 N HCl and EtOAc was
filtered. The organic extract was washed successively with
H2O and saturated aqueous NaCl and dried (MgSO4). The
solvent was removed in vacuo, and the residue was purified
by flash chromatography, eluting with a solvent gradient of
0-5% EtOAc/hexanes to afford the title compound (300 mg,
21%). 1H NMR (CDCl3): δ 7.89 (dd, 1H), 7.40-7.25 (m, 3H),
7.10 (m, 2H), 6.94 (m, 2H), 2.93 (apparent t, 2H), 2.67 (dd,
2H), 2.14 (apparent quintet, 2H).
was separated, washed with brine, and concentrated, yielding
the desired product (0.24 g, 93% yield). 1H NMR (250 MHz,
CDCl3): δ 8.74 (d, 1H), 8.06 (d, 1H), 7.82 (d, 1H), 7.66 (ddd,
1H), 7.50 (ddd, 1H), 7.25 (d, 1H), 4.54 (s, 2H).
Biology. The IC50 values for inhibition of 5-lipoxygenase
by test compounds were determined using a continuous oxygen
consumption assay of Breton et al. using a semipurified
cytosolic preparation of 5-LO dervied for RBL-1 cells.19 The
IC50s for inhibition of leukotriene formation in A23187-
stimulated human whole blood were determined using a LTB4
specific radioimmunoassay as previously reported by Chabot-
Fletcher.27
(A) An im a ls. Male CD-1 mice were obtained from Charles
River Breeding Laboratories (Kingston, NY). Within a single
experiment, mice were age-matched. The ex vivo experiments
involved the use of five animals per treatment and control
groups.
Gen er a l P r oced u r e for th e Con ver sion of In d oles to
19, Illu str a ted for 2,3-Dih yd r o-1H-p yr r olo[1,2-a ]in d ol-1-
on e (19a ). To a suspension of sodium hydride (2.65 g of 60%
dispersion in mineral oil, 66 mmol) was added ethyl 2-indole-
carboxylate (10.0 g, 53 mmol). The mixture was stirred for
15 min, at which time ethyl acrylate (6.3mL, 58 mmol) was
added. Heating was continued for 1 h, at which time ad-
ditional sodium hydride (1.5g of 60% dispersion in mineral oil,
37.5 mmol) and ethyl acrylate (1.0 mL, 9.2 mmol) were added.
The reaction mixture was heated at reflux for an additional 4
h and allowed to cool to room temperature. Ethanol was added
to the reaction mixture, followed by saturated aqueous NH4Cl.
The pH was adjusted to 4 with 5% HCl, and the mixture was
extracted with CH2Cl2. The organic extract was washed with
saturated aqueous NaCl. The solvent was removed in vacuo,
and the residue was dissolved in acetic acid (750 mL) contain-
ing H2O (40 mL). The reaction was heated at reflux for 14 h
and then allowed to cool to room temperature. The reaction
mixture was concentrated under reduced pressure. The
residue was dissolved in CH2Cl2 and washed with aqueous
NaHCO3 and brine. The solvent was removed in vacuo, and
the residue was purified by flash chromatography, eluting with
a solvent gradient of 10-100% EtOAc/hexane to provide 2,3-
dihydro-1H-pyrrolo[1,2-a]indol-1-one (8.00 g, 89%). 1H NMR
(CDCl3): δ 7.74 (apparent dd, 1H), 7.41-7.30 (m, 2H), 7.19
(apparent dt, 1H), 6.97 (s, 1H), 4.34 (t, J ) 8 Hz, 2H), 3.14 (t,
J ) 8 Hz, 2H).
1-Na p h th oxyeth a n oic Acid (20). Chloroacetic acid (2.3
g, 24.7 mmol) and 1-naphthol (3.1g, 21.4 mmol) were added
to a solution of potassium hydroxide (3.0 g, 52.1 mmol) in H2O
(35 mL). After 3 h of refluxing and then cooling, the solution
was extracted with EtOAc to remove the remaining 1-naph-
thol. The remaining aqueous phase was made acidic with 3
N HCl and was extracted with EtOAc. The organic phase was
washed with brine, and the solvent was removed in vacuo,
affording the title compound (2.8 g, 64%). 1H NMR
(CDCl3/MeOD): δ 8.35 (m, 1H), 7.77 (m, 1H), 7.47 (m, 3H),
7.33 (t, 1H), 6.77 (d, 1H), 4.71 (s, 2H), 4.62 (s, 1H).
Na p h th o[1,2-a ]-1H-fu r a n -3-on e (21). To a solution of
1-naphthoxyethanoic acid (2.26 g, 11.2 mmol) in benzene (20
mL) was added oxalyl chloride (5 mL, 57.3 mmol), and the
solution was refluxed 30 min. The mixture was cooled and
concentrated. Additional benzene (20 mL) was added followed
by aluminum chloride (1.8 g, 13.9 mmol). The mixture was
stirred at room temperature 1 h and then poured over ice. The
organic phase was separated and washed with brine, and the
solvent was removed in vacuo to afford the title compound
(1.74 g, 84%). 1H NMR (CDCl3): δ 8.23 (d, 1H), 7.89 (d, 1H),
7.72 (t, 1H), 7.60 (m, 2H), 7.46 (d, 1H), 4.83 (s, 2H).
2-Na p h th oxyeth a n oic Acid (22). This compound was
prepared using the procedure described above for 1-naph-
thoxyethanoic acid (20). 1H NMR (CD3OD): δ 8.4 (m, 1H),
7.7 (m, 1H), 7.41 (m, 2H), 7.34 (m, 2H), 6.70 (d, 1H), 4.65 (s,
2H).
(B) Ex Vivo Mou se Wh ole Blood LTB4 Assa y. The
5-lipoxygenase product LTB4 was extracted from mouse whole
blood following A23187 stimulation. Aliquots of pooled hep-
arinized mouse blood (1 mL each aliquot) from male CD1 mice
were placed into 4 mL polypropylene tubes. The samples were
preincubated for 5 min at 37 °C. A23187 (60 µM) was added
to stimulate eicosanoid production. All samples were incu-
bated for 30 min at 37 °C. The blood samples were centrifuged
at 2500g for 15 min, and the plasma was recovered for
extraction. One volume of chilled acetonitrile was added to
all at 4 °C. The supernatants were recovered after centrifuga-
tion at 2500g for 15 min and diluted with 1% formic acid to
achieve a final concentration of 20% acetonitrile. These
supernatants were then loaded onto the extraction cartridges
that have been conditioned according to the manufacturer’s
instructions (solid phase extraction columns, J . T. Baker, C18
3 mL size). The samples were washed with 3 mL of 1% formic
acid, air-dried, and then washed with 3 mL of petroleum ether.
After air-drying, the samples were eluted with methyl formate.
The eluents were concentrated under vacuum. The concen-
trates were resuspended in 1 mL of enzyme immunoassay
(EIA) buffer. Leukotriene B4 was quantitated by EIA (Cayman
Chemical Co., Ann Harbor, MI).
P h a r m a cok in etics. In mice (N ) 3/time point), (+)- and
(-)-1a (in ethanol/propylene glycol, 30:70, v/v) were adminis-
tered intravenously (0.5 mg/kg) by lateral tail vein injection
over a 5-10 s period or orally (20 mg/kg) by gavage (as a
suspension in 0.5% tragacanth). At various times after drug
administration, mice were anesthetized with ether and ter-
minal blood samples were taken via the vena cava or by
cardiac puncture and immediately chilled.
Pharmacokinetic experiments with (+)- and (-)-1a (in 7.5%
(w/v) aqueous (2-hydroxypropyl)-â-cyclodextrin, Molecusol,
Pharmtec, Inc., Alachua, FL) in male cynomolgus monkeys and
beagle dogs (N ) 3) were performed using intravenous (1-1.4
mg/kg) infusion (15 min) via a catheter placed in the saphenous
(monkeys) or cephalic (dogs) vein. In monkeys, (+)-1a was
also administered orally (approximately 10 mg/kg) by naso-
gastric intubation (in 30% (w/v) aqueous Molecusol). At
various times after drug administration, blood samples were
taken from the opposite cepahilic vein (dogs) or from a vascular
access port with the catheter tip in the vena cava (monkeys)
and immediately chilled. In all cases, plasma was separated
by centrifugation at 4 °C and stored at -20 °C until HPLC
analysis.
For dogs and monkeys, noncompartmental pharmacokinetic
parameters for (+)- and (-)-1a were first estimated for
individual animals and then averaged to obtain mean esti-
mates. Because destructive sampling was used for mice, a
mean plasma concentration was calculated for each time point
prior to pharmacokinetic analysis. The area under the plasma
concentration time curve was determined using linear/log-
linear trapezoidal rule. The half-life of (+)- and (-)-1a was
obtained from the slope of the log transformed plasma con-
centration time profile; the appropriate number of points used
to calculate the half-life in an individual animal was deter-
mined by visual inspection of the data. Maximal (Cmax) plasma
concentrations and the time Cmax was achieved (Tmax) were
determined from actual data. The plasma clearance (CLp) was
Na p h th o[2,1-b]fu r -1-on e (23). Oxalyl chloride (5 mL, 56
mmol) was added to a solution of 2-naphthoxyacetic acid (0.29
g, 1.4 mmol) in toluene (5 mL). The solution was refluxed for
1 h, cooled, and concentrated. Aluminum chloride (0.37 g, 2.8
mmol) in toluene (5 mL) was added, and the reaction was
stirred for 2 h before being poured over ice. The organic phase

Bicyclic N-Hydroxyurea Inhibitors of 5-Lipoxygenase
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26 5045
(5) (a) Laursen, L. S.; Naesdal, J .; Bukhave, K.; Lauristen, K.; Rask-
Madsen, J . Selective 5-lipoxygenase inhibition in ulcerative
colitis. Lancet 1990, 335, 683-685. (b) Collawn, C.; Rubin, P.;
Perez, N.; Bobadilla, J .; Cabrera, G.; Reyes, E.; Borovoy, J .;
Kershenobich, D. Phase II study of the safety and efficacy of a
5-lipoxygenase inhibitor in patients with ulcerative colitis. Am.
J . Gastroenterol. 1992, 87, 342-346.
(6) Weinblatt, M. E.; Kremer, J . M.; Coblyn, J . S.; Helfott, S.; Maier,
A. L.; Petrillo, G.; Hensen, B.; Rubin, P.; Sperling, R. Zileuton,
A 5-lipoxygenase inhibitor in rhematoid arthritis. J . Rheumatol.
1992, 19, 1537-1541.
(7) (a) Isreal, E.; Rubin, P.; Kemp, J . P.; Grossman, J .; Pierson, M.
D.; Siegel S. C.; Tinkelman, D.; Murray, J . J .; Busse, W.; Segal,
A. T.; Fish, J .; Kaiser, H. B.; Ledford, D.; Wenzel, S.; Rosenthal,
R.; Cohn, J .; Lanni, C.; Pearlman, H.; Karahalios, P.; Drazen,
J . M. The effect of inhibition of 5-lipoxygenase by zileuton in
mild-to-moderate asthma. Ann. Intern. Med. 1993, 119 (11),
1059-1066. (b) Friedman, B. S.; Bel, E. H.; Buntinx, A.; Tanaka,
W.; Han, Y. H. R.; Shingo, S.; Spector, R.; Sterk, P. Oral
leukotriene inhibitor (MK-886) blocks allergen-induced airway
responses. Am. Rev. Respir. Dis. 1993, 147 (4), 839-844. (c)
Crawley, G. C.; Bird, T. G. C.; Bruneau, P.; Dowell, R. I.;
Edwards, P. N.; Foster, S. J .; Girodeau, J . M.; Mcmillan, R. M.;
Walker, E. R. H.; Waterson, D. Structure and activity relation-
ships leading to the discovery of ICI-D2138, a selective, potent
and orally active inhibitor of 5-lipoxygenase. J . Lipid Mediators
1993, 6, 249-257.
(8) (a) Carter, G. W.; Young, P. R.; Albert, D. H.; Bouska, J .; Dyer,
R.; Bell, R. L.; Summers, J . B.; Brooks, D. W. 5-lipoxygenase
inhibitory activity of zileuton. J . Pharmacol. Exp. Ther. 1991,
256, 929-937. (b) Garland, L. G.; Salmon, J . A. Hydroxamic
acids and hydroxyureas as inhibitors of arachidonate 5-lipoxy-
genase. Drugs Future 1991, 16 (6), 547-558.
(9) Corey, E. J .; Cashman, J . R.; Kantner, S. S. Rationally designed,
potent competitive inhibitors of leukotriene biosynthesis. J . Am.
Chem. Soc. 1984, 106, 1503-1504.
(10) (a) Falgueyret, J .-P.; Hutchinson, J . H.; Riendeau, D. Criteria
for the identification of non-redox inhibitors of 5-lipoxygenase.
Biochem. Pharmacol. 1993, 45, 978-981. (b) Nelson, M. J .; Batt,
D. G.; Thompson, J . S.; Wright, S. W. Reduction of the active-
site iron by potent inhibitors of lipoxygenases J . Biol. Chem.
1991, 266, 8225-8229.
(11) Summers, J . B., J r. Lipoxygenase inhibiting compounds. Eur.
Pat. Appl. 279281, 1987.
(12) Shriner, R. L.; Grosser, F. Coumaran derivatives IX. Synthesis
of 3,4,6,3′,4′-pentahydroxy-2-benzyl coumarin. J . Am. Chem.
Soc. 1942, 64, 382-384.
(13) J ung, M. E.; Abrecht, S. Improved synthesis of 3-substituted
methoxybenzofurans, useful intermediates for preparation of
morphine analogues. J . Org. Chem. 1988, 53, 4233-425.
(14) Townsend, C. A.; Davis, S. G.; Christensen, S. B.; Link, J . C.;
Lewis, C. P. Methoxymethyl-directed aryl metalation, a total
synthesis of (() averufin. J . Am. Chem. Soc. 1982, 103, 6885-
6888.
(15) Garigipati, R. S.; Sorenson, M. E.; Erhard, K. F.; Adams, J . L.
Resolution of hydroxyureas. Tetrahedron Lett. 1993, 34, 5537-
5540.
(16) Casner, M. L.; Remers, W. A.; Bradner, W. T. Synthesis and
biological activity of 6-substituted mitosene analogues of the
mitomycins. J . Med. Chem. 1985, 28, 921-926.
(17) Summers, J . B.; Gunn, B. P.; Martin, J . G.; Martin, M. B.;
Mazdiyasni, H.; Stewart, A. O.; Young, P. R.; Bouska, J . B.;
Goetze, A. M.; Dyer, R. D.; Brooks, D. W.; Carter, G. W.
Structure-activity Analysis of a class of orally active hydroxamic
acid inhibitors of leukotriene biosynthesis. J . Med. Chem. 1988,
31, 1960-1964.
(18) Summers, J . B.; Kim, K. H.; Mazaditasni, H.; Holms, J . H.;
Ratajeczyk, J . D.; Stewart, A. O.; Dyer, R. D.; Carter, G. W.
Hydroxamic acid inhibitors of 5-lipoxygenase: quantitative
structure-activity relationships. J . Med. Chem. 1990, 33, 992-
998.
(19) Breton, J .; Keller, P.; Chabot-Fletcher, M.; Hillegass, L.; DeWolf,
W., J r.; Griswold, D. Use of a continuous assay of oxygen
consumption to evaluate the pharmacology of 5-lipoxygenase
inhibitors. Prostaglandins, Leukotrienes Essent. Fatty Acids
1993, 49, 929-937.
obtained by dividing the dose administered by the AUC. The
bioavailability (F) of (+)-1a was calculated by dividing the
dose-normalized AUC obtained after iv administration by the
AUC obtained following oral administration.
HP LC Assa y. Plasma concentrations of (+)- and (-)-1a
were quantified by HPLC from standard curves dissolved in
plasma. The assay had a limit of quantitation of 0.05 µg/mL
and was linear up to 15 µg/mL for each enantiomer. Samples
and standards (50 µL) were analyzed after addition of the
internal standard (3a , 10 µL of a 50 µg/mL solution in
acetonitrile/water, 1:1) and 2.5 mL of 30% methanol in 1%
aqueous acetic acid using the following procedure. Following
mixing, samples were passed through conditioned (methanol
followed by 1% aqueous acetic acid) Baker 3 mL LD C18-bonded
phase extraction columns, washed with 2 mL of 1% aqueous
acetic acid, and eluted with methanol (2 × 0.5 mL). The
extracts were evaporated under nitrogen, redissolved in 100
µL of mobile phase, and analyzed by HPLC at 215 nm using
a Waters NOVAPAK C18 column (5 × 100 mm). Chromatog-
raphy was carried out isocratically at a flow rate of 2 mL/min
with a mobile phase of 34% acetonitrile in 10 mM aqueous
ammonium formate.
In Vitr o Meta bolism . Sou r ces of Ma ter ia ls. [14C]UDP-
glucuronic acid was purchased from DuPont NEN (Boston,
MA). All other materials were of the highest quality from
commercial suppliers.
Tissu e Sou r ce a n d P r ep a r a tion . Livers were obtained
from male CD-1 mice [(ICR)Br OUTBRED] greater than 10
weeks of age. Human livers were obtained from organ donors
through the Association of Human Tissue Users (Tucson, AZ).
Microsomes were prepared from fresh mouse liver and frozen
human liver using a standard method [Guengerich, In Prin-
ciples and Methods of Toxicology; Hayes, A. W., Ed.; Raven
Press, New York, 1989; pp 777-814.
En zym e Assa y Con d ition s. Glucuronidation reaction
mixtures contained 2.5 mM substrate [(+)- and (-)-1a , zileu-
ton], 15 mM [14C]UDP-glucuronic acid (0.3 mCi/mmol), 0.25-
1.0 mg/mL microsomal protein, 10 mM MgCl2, 2% bovine
serum albumin, and 100 mM sodium phosphate buffer (pH
7.4) in a final volume of 250 mL. Reaction mixtures were
incubated at 37 °C for 1-4 h, and reactions were terminated
by adding acetonitrile to a final concentration of 20% (v/v).
Samples were centrifuged at 14 000 rpm for 10 min, and
supernatant was analyzed by HPLC.
HP LC Con dition s. Samples were analyzed using a NOVA-
PAK C18 column (4 m, 5 × 100 mm) in a radial compression
module (Millipore, Milford, MA). The column was equilibrated
with water/acetonitrile/trifluoroacetic acid (84.9:15.0:0.1, v/v)
and was eluted by linearly increasing the concentration of
acetonitrile to 50% over 15 min, with a flow rate of 2 mL/min.
Eluent from the HPLC was monitored at 215 nm and with a
Radiomatic Flo-One\ Beta radioflow detector (Packard, Meri-
den, CT).
Refer en ces
(1) (a) Wardlaw, A. J .; Hay, H.; Cromwell, O.; Collins, J . V.; Kay,
A. B. Leukotrienes, LTC4 and LTB4, in bronchoalveolar lavage
in bronchial asthma and other respiratory diseases. J . Allergy
Clin. Immunol. 1989, 84, 19-26. (b) Lewis, R. A.; Austen, K.
F.; Soberman, R. J . Leukotrienes and other products of the
5-lipoxygenase pathway. N. Engl. J . Med. 1990, 323, 645-655.
(2) Henderson, W. R., J r. Lipid-derived and other chemical media-
tors of inflammation in the lung. J . Allergy Clin. Immunol.
1987, 79, 543-553.
(3) (a) Torphy, T. J .; Faiferman, I.; Gleason, J . G.; Hall, R. F.; Lewis,
M. A.; Broom, C.; Helfrich, H. M.; Newton, J . F.; Hay, D. W. P.
The preclinical and clinical pharmacology of SK&F 104353, a
potent and selective peptidoleukotriene receptor antagonist.
Ann. N.Y. Acad. Sci. 1991, 629, 157-167. (b) Nakagawa, T.;
Mizushima, Y.; Ishii, A; Nambu, F.; Motishi, M.; Yui, Y.; Shida,
T.; Miyamoto, T. Effect of a leukotriene antagonist on experi-
mental and clinical bronchial asthma. Adv. Prostaglandin,
Thromboxane Leukotriene Res. 1990, 21, 465-468. (c) Busse,
W. W.; Gaddy, J . N. The role of leukotriene antagonists and
inhibitors in the treatment of airway diseases. Am. Rev. Respir.
Dis. 1991, 143, S103-S107.
(20) Gresele, P.; Arnout, J .; Deckmyn, H.; Vermylen, J . L-653,343,
A novel dual cyclolipoxygenase inhibitor, inhibits LTB4-produc-
tion by stimulated human polymorphonuclear cells but not
stimulated human whole blood. Biochem. Pharmacol. 1987, 36,
3529-3531.
(21) Salmon, J . A.; J ackson, W. P.; Garland, L. G. Development and
in vivo evaluation of 5-lipoxygenase inhibitors: potential drugs
for asthma and inflammation. In Therapeutic Approaches to
Inflammatory Diseases; Lewis, A. J ., Doherty, N. S., Ackerman,
N. R., Eds.; Elsevier Science Publishing Co.: New York, 1989;
pp 137-146.
(4) J ewell, D. P.; Snook, J . A. Immunology of ulcerative colitis and
Crohn’s disease. In Inflammatory Bowel Diseases; Allan, R. N.,
Keighley, M. R. B., Alexander-Williams, J ., Hawkins, C., Eds.;
1990, pp 127-146.

5046 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 26
Adams et al.
(22) Tateson, J . E.; Randall, R. W.; Reynolds, C. H.; J ackson, W. P.;
Bhattacherjee, P.; Salmon, J . A.; Garland, L. G. Selective
inhibition of arachidonate 5-lipoxygenase by novel acetohydrox-
amic acids: biochemical assessment in vitro and ex vivo. Br. J .
Pharmacol. 1988, 94, 528-539.
(23) Griswold, D. E.; Hillegass, L. M.; White, J . R. Pharmacological
evaluation of leukotriene biosynthesis and inflammation in an
adoptive model of peritoneal anaphylaxis in the mouse. Drug
Dev. Res. 1993, 30, 83-90.
(24) Braeckman, R. A.; Granneman, G. R.; Rubin, P.; Kesterson, J .
W. Pharmacokinetics and metabolism of the new 5-lipoxygenase
inhibitor A-64077 after single oral administration in man. J .
Clin. Pharmacol. 1989, 29, A22.
of the potency and duration of action of N-hydroxyurea 5-lipoxy-
genase inhibitors. J . Pharmacol. Exp. Ther. 1995, 272, 724-
731.
(27) The effects of SB 202235 in models of asthma and inflammation
have recently been published. Chabot-Fletcher, M. C.; Under-
wood, D. C.; Breton, J . J .; Adams, J . L.; Kageysobotka, A.;
Griswold, D. E.; Marshall, L. A.; Sarau, H. M.; Winkler, J . D.;
Hay, D. W. P. Pharmacological characterization of SB-202235,
a
potent and selective 5-lipoxygenase inhibitor - Effects in
Models of Allergic Asthma. J . Pharmacol. Exp. Ther. 1995, 273,
1147-1155.
(28) A manuscript further detailing the chemistry and evaluation of
additional analogues is currently in preparation and will address
the clinical results with SB 202235.
(29) Miyake, A.; Katsumi, I.; Norio, T.; Masao, T.; Hirata, M.; Oka,
Y. Synthesis of 2-(N-Substituted amino)-6-hydroxy-1,2,3,4-tetra-
hydronaphthalen-1-ol derivatives. Chem. Pharm. Bull. 1983, 31,
2329-2348.
(25) Sweeny, D. J .; Bouska, J .; Machinst, J .; Bell, R.; Caret, G.; Cepa,
S.; Nellans, H. N. Glucuronidation of zileuton (A-64077) by
human hepatic microsomes. Drug. Metab. Dispos. 1992, 20,
328-329.
(26) A similar selection process to chose the prefered enantiomer of
a racemic hydroxyurea has been described in the following: Bell,
R. L.; Bouska, J . B.; Malo, P. E.; Lanni, C.; Harris, R. R.; Otis,
E. R.; Stewart, A. O.; Brooks, D. W.; Carter, G. W. Optimization
(30) Kim, K.; Kang, S.; Kim, I. K.; Hong, J . H. A Short synthesis of
ethoproxyfen (MIT-500), a non-ester pyrethroid insecticide. Bull.
Korean Chem. Soc. 1987, 8, 128.
J M960271D