Novel 3-Oxa Lipoxin A4 Analogues with Enhanced Chemical and Metabolic Stability Have Anti-inflammatory Activity in Vivo

Article abstract of DOI:10.1021/jm030569l

Lipoxin A₄ (LXA₄) is a structurally and functionally distinct natural product called an eicosanoid, which displays immunomodulatory and anti-inflammatory activity but is rapidly metabolized to inactive catabolites in vivo. A previous

Full text of DOI:10.1021/jm030569l

J . Med. Chem. 2004, 47, 2157-2165
2157
Novel 3-Oxa Lip oxin A4 An a logu es w ith En h a n ced Ch em ica l a n d Meta bolic
Sta bility Ha ve An ti-in fla m m a tor y Activity in Vivo
William J . Guilford,*,† J ohn G. Bauman,† Werner Skuballa,‡ Shawn Bauer,† Guo Ping Wei,† David Davey,†
Caralee Schaefer,§ Cornell Mallari,§ J ennifer Terkelsen,§ J ih-Lie Tseng,# J un Shen,# Babu Subramanyam,#
Arndt J . Schottelius,| and J ohn F. Parkinson
Departments of Medicinal Chemistry, Immunology, Animal Pharmacology, and Pharmacokinetics and Drug Metabolism,
Berlex Biosciences, 2600 Hilltop Drive, Richmond, California 94804, and Research Business Area Dermatology and Medicinal
Chemistry, Schering AG, Berlin, Germany
Received November 11, 2003
Lipoxin A4 (LXA4) is a structurally and functionally distinct natural product called an eicosanoid,
which displays immunomodulatory and anti-inflammatory activity but is rapidly metabolized
to inactive catabolites in vivo. A previously described analogue of LXA4, methyl (5R,6R,7E,9E,-
11Z,13E,15S)-16-(4-fluorophenoxy)-5,6,15-trihydroxy-7,9,11,13-hexadecatetraenoate (2, ATLa),
was shown to have a poor pharmacokinetic profile after both oral and intravenous administra-
tion, as well as sensitivity to acid and light. The chemical stability of the corresponding E,E,E-
trien-11-yne analogue, 3, was improved over 2 without loss of efficacy in the mouse air pouch
model of inflammation. Careful analysis of the plasma samples from the pharmacokinetic assays
for both 2 and 3 identified a previously undetected metabolite, which is consistent with
metabolism by â-oxidation. The formation of the oxidative metabolites was eliminated with
the corresponding 3-oxatetraene, 4, and the 3-oxatrien-11-yne, 5, analogues of 2. Evaluation
of 3-oxa analogues 4 and 5 in calcium ionophore-induced acute skin inflammation model
demonstrated similar topical potency and efficacy compared to 2. The 3-oxatrien-11-yne
analogue, 5, is equipotent to 2 in an animal model of inflammation but has enhanced metabolic
and chemical stability and a greatly improved pharmacokinetic profile.
In tr od u ction
Lipoxins are a structurally and functionally distinct
class of natural products called eicosanoids, which play
a key role in inflammation. Arachidonate-derived lipid
mediators such as prostaglandins or leukotrienes are
involved in the initiation and maintenance of inflam-
matory responses and are proinflammatory; however,
lipoxins are involved in the regulation and resolution
of inflammation and are anti-inflammatory.1 Since
lipoxin A4 (1, LXA4, Figure 1) was reported in 1984,2
the trihydroxytetraene structure has attracted the
attention of synthetic chemists, as seen in a number of
reported syntheses.3-6 The potent anti-inflammatory
and immunomodulatory activity of LXA4 has attracted
the attention of researchers for use as a treatment of
allergic asthma,7 T cell-mediated dermatoses, and in-
flammatory bowel diseases and as a gastroprotective
agent.8 Development of LXA4, however, has been limited
by its metabolic and chemical instability.
The metabolic instability of the natural product,
LXA4, is the result of a series of steps initiated by the
rapid oxidation of the 15(S) alcohol by prostaglandin
dehydrogenase (PGDH) to form inactive catabolites, as
15-oxo-LXA4 and 13,14-dihydro-LXA4.9 The initial de-
* To whom correspondence should be addressed. Phone: 510 669
4065. Fax: 510 669 4310. E-mail: william_guilford@berlex.com.
F igu r e 1. Lipoxin A4 and analogues.
† Department of Medicinal Chemistry, Berlex Biosciences.
‡ Medicinal Chemistry, Schering AG.
§ Department of Animal Pharmacology, Berlex Biosciences.
sign of metabolically stable LXA4 analogues focused on
identifying poor substrates for PGDH,10 which main-
tained potency in in vitro assays. The discovery that 15-
epi-LXA4, or aspirin-triggered LXA4, was equipotent in
#
Department of Pharmacokinetics and Drug Metabolism, Berlex
Biosciences.
| Research Business Area Dermatology, Schering AG.
Department of Immunology, Berlex Biosciences.
10.1021/jm030569l CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/11/2004

2158 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8
Sch em e 1. Retrosynthetic Analysis of Lipoxin Analogues
Guilford et al.
in vitro assays to LXA4 but was a poorer substrate for
PGDH provided support for the potential novel analogue
design. Several structural elements were combined in
2, methyl (5R,6R,7E,9E,11Z,13E,15S)-16-(4-fluorophen-
oxy)-5,6,15-trihydroxy-7,9,11,13-hexadecatetraenoate
(15-epi-16-(p-fluoro)phenoxy-LXA4, ATLa, ATLa2), which
has enhanced metabolic stability and was selected for
scale-up and further testing. The potent anti-inflam-
matory and immunomodulatory properties of 2 were
demonstrated in a series of in vivo models, including
an allergic airway inflammation model,11 a T-cell-
dependent skin inflammation model,12 a dextran sulfate
(DSS) induced colitis model,13 and an adaptive im-
munity model with 5-LO knockout mice.14 Taken to-
gether, these studies provide evidence that 2 can
directly or indirectly modulate T cell effector function
in the setting of Th1- and Th2-dependent inflammation
and adaptive immunity.
Although 2 has enhanced metabolic stability over
LXA4 in vivo, 2 is reported to be cleared within 15 min
after intravenous injection in the mouse.15 In addition,
2 shares the same trihydroxytetraene structure with
LXA4 and, therefore, has the same chemical stability
issues, light and acid sensitivity. LXA4 isomerizes to a
mixture of double bond isomers including the corre-
sponding E,E,E,E- or 11-trans-LXA4 in the presence of
light and decomposes to a mixture of products in the
presence of strong acid. To circumvent the metabolic and
chemical stability issues, we explored simple replace-
ments for the E,E,Z,E-tetraene unit that would main-
tain potency. Herein, we describe the results of our effort
to identify analogues of 2 that have increased metabolic
and chemical stability, improved pharmacokinetics, and
anti-inflammatory activity in vivo, which led to the
discovery of 5.
Ch em istr y
The structure of 2 and 3 can be characterized as a
linear chain of 16 carbons in which all but three carbon
atoms are either unsaturated or substituted with oxy-
gen. The 3-oxa analogues 4 and 5 differ in that all of
the carbon atoms in the chain are either unsaturated
or substituted with oxygen. As in the reported synthesis
of 2,16 the retrosynthetic analysis of the lipoxin ana-
logues is based on the Nicolaou synthesis of lipoxin A4
(1) and is shown in Scheme 1.17 This convergent
approach divides the molecule into three parts: an
enyne Wittig reagent (13),18 an aldehyde (11 or 12), and
a vinyl bromide unit (17). The stereochemistry at C5
and C6 is set with the starting sugar, 15 or 16, and the
stereochemistry at C15 is set by a chiral reduction. The
geometry of the C9-C10 double bond is established with
starting E-pentenynol, 14. The geometry of the C13-
C14 double bond is set by reduction of an acetylenic
alcohol (Scheme 2). The geometry of the C7-C8 and
C11-C12 double bonds is set during the synthesis.
Preparation of the vinyl bromide intermediate is
shown in Scheme 2. Weinreb amide 19 was prepared
from 4-fluorophenoxyacetic acid, 18, through the cor-
responding acid chloride. Treatment of amide 19 with
lithium acetylide gave ketone 20, which was reduced
with the chiral reducing agent R-Alpine-Borane.19 Al-
though the enantiomeric excess of 21 was typically in
the 60-70% range, material having >98% ee was
obtained by recrystallization of the corresponding dini-
trobenzoyl derivative (22). Hydrolysis of the dinitroben-
zoyl ester followed by bromination of the acetylene with
N-bromosuccinimide and silver nitrate gave the acetyl-
enic bromide 23.20 Reduction of the acetylenic bromide

Lipoxin A4 Analogues
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8 2159
Sch em e 2. Preparation of Vinyl Bromide 17a
a
(a) (COCl)2, DMF, ether; (b) MeON(H)Me‚HCl, sat. K2CO3(aq), EtOAc, 0 °C, 73% from 18; (c) HCCMgBr, THF, 40 min, 91%; (d) R-Alpine-
Borane, 97% (e) 3,5-dinitrobenzoyl chloride, TEA, DMAP, CH2Cl2, 0 °C, 1-2 h, followed by recrystallization, 54%; (f) K2CO3, MeOH,
THF, 3.5 h, 80%; (g) NBS, AgNO3, acetone, 99%; (h) AlCl3, LiAlH4, ether, 81%.
Sch em e 3. Synthesis of Aldehyde 12 from Rhaminosea
Introduction of the 3-oxa group into the dienyne unit,
10, was accomplished as outlined in Scheme 3. Cyclo-
hexylidene ketal, 24, prepared from rhamnose, 16,23 was
reduced with sodium borohydride to triol 25. Selective
O-alkylation of the primary hydroxyl group using phase-
transfer conditions and tert-butyl bromoacetate gave
ether 26. The desired aldehyde, 12, was prepared by
oxidative cleavage of 26 using sodium metaperiodate.
Coupling of 12 and 13 yielded dienyne, 27, as a mixture
of E,E and E,Z isomers (Scheme 4). Iodine-catalyzed
isomerization to the E,E-dienyne 27 and removal of the
trimethylsilyl protecting group gave intermediate 10.
Palladium-catalyzed Sonogashira coupling of 10 and 17
followed by hydrolysis of the cyclohexylidene ketal gave
the corresponding tert-butyl ester 28. As in the case of
esters of 3, the trienyne unit was stable to acid over a
24 h period required to cleave the ketal. Base-catalyzed
ester hydrolysis gave 5. Reduction of 28 with activated
zinc22 gave the E,E,Z,E-tetraene and subsequent ester
hydrolysis yielded 4.
a
(a) H2SO4, Cu(II)SO4, cyclohexanone, room temp, 16 h, 57%;
(b) NaBH4, MeOH, 3 h, room temp, 88%; (c) toluene, 25% NaOH(aq)
,
Bu4N, HSO4, tBu bromoacetate, 44%; (d) H2O, acetone, NaIO4,
92%.
with a lithium aluminum hydride/aluminum chloride
mixture gave the desired E-vinyl bromide 17.21
Lipoxin analogues 2 and 3 were prepared from
deoxyribose 15 via dienyne 9 using a procedure de-
scribed earlier16 (Scheme 1). Sonogashira coupling of
dienyne 9 and vinyl bromide 17 gave the complete
carbon skeleton as the trienyne 7.17 Hydrolysis of the
isopropylidene ketal under acidic conditions and methyl
ester under basic conditions gave 3. The relatively good
acid stability of the trien-11-yne unit was demonstrated
by the treatment of 7 with 1 N HCl needed to cleave
the ketal protecting group in good yield. Treatment of
deketalized 7 with activated zinc using the Boland
procedure22 gave 2.
Id en tifica tion of Meta bolite
Since the report15 that 2 is cleared from the mouse
within 15 min after intravenous injection, using a dose
of 2 µg per mouse, the pharmacokinetic parameters of
2 and related analogues were evaluated in the rat at a
dose of 5 mg/kg (about 1 mg per rat). The results are
shown in Table 1. Evaluation of the pharmacokinetic
parameters of 2 in the rat was complicated by the rapid
hydrolysis of the methyl ester in rat plasma to the
corresponding carboxylic acid, 29 (see Figure 2). The
Sch em e 4. Synthesis of 3-oxa-lipoxin Analoguesa
a
(a) BuLi, THF, 30-0 °C, 3 h; (b) I2, CH2Cl2, 1 h, 49% from 12; (c) TBAF, THF, 99%; (d) (Ph3P)4, Et2NH, CuI, THF, 17, 50%; (e) AcOH,
H2O, EtOAc, 58%; (f) Zn, AgNO3, Cu(II)(OAc)2, MeOH, H2O; (g) NaOH, H2O, MeOH.

2160 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8
Guilford et al.
F igu r e 3. Tetraene and trienyne lipoxin analogues are both
anti-inflammatory in vivo. Analogues 2 and 3 were tested for
local anti-inflammatory effects in the murine dorsal air pouch
model of TNF-induced inflammation, as described in “Experi-
mental Section” and ref 29. The figure shows the effect of 10
ng of locally delivered analogues on total leukocyte (grey bars)
and neutrophil (black bars) cell counts in the pouch exudate 4
h following TNF administration. A robust and similar inhibi-
tory effect (∼50%) on TNF-induced leukocyte trafficking into
the pouch was observed for 2 and 3. Data are the mean ( SEM
from N ) 20 mice: (/) p < 0.05 versus vehicle; (//) p < 0.05
versus TNF.
F igu r e 2. Proposed metabolites of 2.
Ta ble 1. Pharmacokinetic Data in the Rat after iv (n ) 3, 3
mg/kg) and po (n ) 3, 5 mg/kg) Dosing
iv
Cl,
po
AUCall
,
compd
T
1/2, h (mL/mg)/kg (h‚µg)/mL
T
1/2, h Cmax Tmax, h F, %
of 4 compared with 2, 29, or 3. In contrast, trienyne 5
had a significantly improved half-life and a significantly
reduced clearance after both po and iv dosing in the rat.
The similar F values in the rat for 2 and 5 hide the
improved pharmacokinetic parameters of 5 as deter-
mined by the area under the curve value, AUCall after
iv dosing. Comparison of the corresponding pharmaco-
kinetic parameters for 2 and 5 in the dog after iv dosing
(2 mg/kg) are complicated by the incomplete ester
hydrolysis of 2. Comparison of the pharmacokinetic
parameters for 5 in the dog versus the rat shows a slight
increase in the iv half-life from 2.3 to 2.7 h (2 mg/kg)
and a slight increase in the po half-life from 2.2 to 3.2
h (2 mg/kg). Similar improvements in Cmax and Tmax
values after oral dosing increased the F value for 5 in
the dog to 40.
In Vivo Mod els of In fla m m a tion . Analogue 2 has
been shown to be efficacious in the TNFR-induced
murine dorsal air pouch model, which is an acute
experimental model of inflammation.28-30 The model is
characterized by cellular infiltration into a localized
area, i.e., air pouches raised on the dorsal area of mice.
Once the air pouch is established and stabilized, an
injection of TNF-R induces a temporal infiltration of
inflammatory cells into the air pouch. The infiltrate is
predominantly composed of proinflammatory neutro-
phils but also contains eosinophils and smaller numbers
of mononuclear cells (monocytes and T- and B-lympho-
cytes). Four hours after injection of the compound and
TNF-R, the cells were harvested from the air pouch for
enumeration and differential count analysis. As seen in
Figure 3, compounds 2 and 3 showed a significant
decrease in the total number of cells and neutrophils
infiltrating into the air pouches at similar doses of 10
ng/pouch compared with vehicle-treated mice.
2
29
3
4
5
0.3
0.2
0.1
1.3
2.3
51
28
58
50
7
1.0
1.8
0.9
1.0
7.0
0.5
1.1
1.8
1.1
2.2
0.2
0.3
0.2
0.04
0.4
0.5
1.0
0.5
0.5
1.3
12
13
28
4
14
calculated phamacokinetic parameters for 2 listed in
Table 1 were obtained by dosing 2 but analyzing the
plasma samples for 29. The half-life of 2 and 29 were
about 15 min, in agreement with the data reported in
mice.15 Careful analysis of the plasma samples indicated
the presence of a metabolite, which had a molecular
weight 2 mass units lower than that of 29. Since
compound 2 was designed to minimize all the known
major metabolic routes, the identification of a new
metabolite by LC/MS-MS was unexpected. The UV
spectra of both 29 and the metabolite were identical,
which is consistent with an intact E,E,Z,E-tetraene
chromaphore9 and minimal structural modification. Two
structures were consistent with the data: 30 and 31 (see
Figure 2). Since the metabolite was stable to treatment
with sodium borodeuteride in methanol, the 5-oxo
structure, 31, could be ruled out and the structure of
the metabolite was assigned as the 2,3-dehydro ana-
logue 30. Attempts to isolate 30 from in vitro metabo-
lism studies, such as liver microsomes, for a more
complete structural determination were unsuccessful.
By analogy to lipid metabolism in the prostaglandin24,25
and leukotriene26 fields, we propose that 30 would be
formed as the first step in the â-oxidation of 2. Consis-
tent with the proposal, the corresponding 2-mass-unit-
lower metabolite was observed during a pharmacoki-
netic assay of 3 in rats but not in 3-oxa analogues 4
and 5, which were designed to eliminate this pathway.27
The phamacokinetic parameters for tetraene analogue
2 is characterized by rapid clearance and relatively short
half-lives after both oral gavage (po) and intravenous
injection (iv). The corresponding trienyne analogue, 3,
had similar properties after iv administration but a
significantly better effective half-life after oral admin-
istration. The elimination of the â-oxidation pathway
had a minimal impact on the pharmacokinetic profile
Topical efficacy for native LXA4, 1, and 2 has been
demonstrated in several models of skin inflammation.12
The calcium ionophore model was chosen to compare
3-oxa analogues 4 and 5 with 2. In this model, calcium
ionophore A-23187 with or without anti-inflammatory
agents is applied topically to the ears of a mouse, which
induces acute inflammation with edema and granulo-

Lipoxin A4 Analogues
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8 2161
F igu r e 4. 3-Oxa-lipoxin analogues are potent topical anti-inflammatory agents. Acute mouse ear skin inflammation was induced
by topical application of 0.1% w/v calcium ionophore in isopropylmyristate, as described in “Experimental Section” and ref 12.
Inflammation was measured after 24 h. Data for vehicle control (open bars), ionophore control (black bars), and treatment by
topical coapplication of 5 (gray bars), 4 (gray hashed bars), and the positive control 2 (stippled bars) are shown at doses of 100,
300, and 600 µg/cm2. Dose-dependent anti-inflammatory effects were observed on edema (ear weight, panel A), total leukocyte
infiltration (peroxidase activity, panel B), and neutrophil infiltration (elastase activity, panel C). ED50 was <300 µg/cm2 for 5 and
4 on all endpoints. Data are the mean ( SEM with N ) 8-to 10 mice for all groups, except N ) 5 for 600 µg/cm2 dose of 2: (/)
p < 0.05 versus ionophore control.
cyte infiltration that peaks at approximately 24 h. The
extent of granulocyte infiltration was determined using
the peroxidase assay. Elastase activity was used as a
measure of neutrophil infiltration. Anti-inflammatory
effects of a given compound were defined as inhibition
of edema formation (Figure 4A), peroxidase acitivity
(Figure 4B), and elastase activity (Figure 4C). The data
shown in Figure 4 clearly show that 2, 4, and 5 inhibited
edema formation and neutrophil and granulocyte infil-
tration in a dose-dependent manner. In addition, neu-
trophil and granulocyte infiltration was completely
inhibited at the highest dose tested (600 µg/cm2). The
IC50 for 4 and 5 on all three topical efficacy parameters
was <300 µg/cm2.
Con clu sion
The natural product, lipoxin A4, has immunomodu-
latory and anti-inflammatory activity but is rapidly
metabolized to inactive catabolites in vivo. Chemically
and metabolically stable LXA4 analogues of 2, which are
equipotent in animal models of inflammation, have been
reported. The synthesis of the trienyne and tetraene

2162 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8
Guilford et al.
4-(4-F lu or op h en oxy)-1-bu tyn -3-on e (20). A solution of 19
(20.00 g, 74 mmol) in THF (100 mL) was cooled in an ice bath
as a solution of ethynylmagnesium bromide (0.5 M in THF,
508 mL, 254 mmol) was added. After an additional 30 min at
0 °C, the reaction mixture was poured into a vigorously stirred
mixture of 1 M NaH2SO4 (1700 mL) and ether (1 L). The layers
were separated, and the aqueous layer was then extracted with
ether (700 mL). The combined organic phases were washed
with brine and dried over MgSO4, filtered, and concentrated
under vacuum. The residue was purified by eluting through a
plug of silica gel (10 cm × 3 cm) with 1:4 ether/petroleum ether
to afford 27.65 g (91% yield) as a low-melting solid: 1H NMR
(CDCl3) δ 3.40 (s, 1H), 4.70 (s, 2H), 6.85 (m, 2H), 7.0 (t, 2H)
ppm.
analogues is described in both the carbon and 3-oxa
series. In the carbon series, we demonstrated that the
tetraene unit is not needed for efficacy in the air pouch
model of inflammation. On the basis of the identification
of a new metabolic route for 2, the â-oxidation-resistant
3-oxa series was developed (trienyne, 5, and tetraene,
4) with anti-inflammatory activity comparable to that
of 2 in cutaneous inflammation. On the basis of the
increased acid, metabolic, and UV stability, equivalent
anti-inflammatory activity, improved pharmacokinetic
parameters, and ease of synthesis, 5 has pharmaceutical
properties suitable for clinical development. Taken
together, the results provide new insights on the in vivo
catabolism of lipoxins and structure-activity relation-
ships essential for in vivo actions. The 3-oxa-based
lipoxin A4 analogues provide superior pharmacophores
with which to explore preclinical and clinical concepts
in immunomodulation.
(3S)-4-(4-F lu or op h en oxy)-3-h yd r oxy-1-bu tyn e (21). A
solution of R-Alpine-Borane (0.5 M in THF, 930 mL, 465 mmol)
was evaporated to dryness under vacuum to give about 150 g
of a thick syrup. Neat 20 (27.6 g, 155 mmol) was added, and
the reaction mixture was stirred for 2 days. The reaction
mixture was cooled to 0 °C, and acetaldehyde (26 mL, 465
mmol) was added to quench the excess reagent. After being
stirred at ambient temperature for 2 h, the reaction mixture
was placed under vacuum and stirred first at 0 °C for 1 h,
then at 65 °C for 2 h. The reaction mixture was cooled to
ambient temperature, and ether (300 mL) was added under
nitrogen. Ethanolamine (30 mL, 465 mmol) was added drop-
wise at 0 °C, and the resulting reaction mixture was stored in
the freezer overnight. The resulting precipitate was removed
by filtration and washed with cold ether. The combined
filtrates were concentrated under vacuum. Purification by
flash chromatography on a 2.5 L column using a 10-25%
gradient of ethyl acetate in hexane gave 27 g (97%): 1H NMR
(CDCl3) δ 2.56 (s, 1H), 4.10 (m, 2H), 4.78 (m, 1H), 6.85 (m,
2H), 7.0 (m, 2H). This material was determined to be about
64% ee based on chiral HPLC of its 3,5-dinitrobenzoyl ester.
To a solution of 21 (about 490 mmol) in methylene chloride (1
L) was added, between -5 and 0 °C, 3,5-dinitrobenzoyl chloride
(125 g, 539 mmol), followed by slow addition of triethylamine
(10.8 mL, 77 mmol) and a catalytic amount of (dimethylamino)-
pyridine (20 mg). After being stirred for 40 min, the reaction
mixture was cautiously partitioned between methylene chlo-
ride and aqueous sodium bicarbonate. The aqueous layer was
extracted with dichloromethane. The combined organic layers
were washed with water and brine, dried over Na2SO4, filtered
through a pad of silica gel, and concentrated to give a tan solid.
Rapid recrystallization from a 99:1 mixture of methanol/acetic
acid (5 L) gave 101 g (54%) of the enantiomerically enriched
product as fluffy white needles. This material was determined
to have >98% ee by HPLC: 1H NMR (CDCl3) δ 2.65 (s, 1H),
4.40 (m, 2H), 6.05 (m, 1H), 6.90 (m, 2H), 7.0 (t, 2H), 9.15 (s,
2H), 9.25 (s, 1H) ppm. To a solution of 22 (10.35 g, 98% ee,
27.6 mmol) in THF (115 mL), was added methanol (115 mL)
and potassium carbonate (0.58 g). After being stirred for 3.5
h, the reaction mixture was quenched with acetic acid (2 mL).
The solvents were evaporated. The resulting slurry was
filtered, and the solid was washed with ether. The filtrate was
concentrated, and the filtration/ether wash sequence was
repeated to give 4.02 g (80%): 1H NMR (CDCl3) δ 2.56 (s, 1H),
4.10 (m, 2H), 4.78 (m, 1H), 6.85 (m, 2H), 7.0 (m, 2H).
Exp er im en ta l Section
All starting materials not described below were purchased
from commercial sources. All reagents and solvents were used
as received from commercial sources without additional pu-
rification. Elemental analyses were performed by Robertson
Microlit Laboratories (Madison, NJ ), and results were within
(0.4% of the calculated values. Proton NMR spectra were
recorded with a Varian XL-300 spectrometer at 400 MHz
proton and were consistent with the assigned structures.
Carbon-13 NMR spectra at 100 and 125 MHz were obtained
on a Varian XL-300 spectrometer and the Bruker Avance 500
spectrometer. HPLC was performed with a Rainin SD-1
Dynamax system and a C-18 reverse-phase Dynamax 60A
column using a gradient of methanol (0.1% TFA) or acetonitrile
(0.1% TFA) in water (0.1% TFA). The enantiomeric excess of
21 was determined by HPLC analysis on the corresponding
dinitrobenzoate, 22, using a Diacel Chiralpak AD (4.6 mm ×
250 mm, 60% 2-propanol/hexane, 1 mL/min), which separates
the (R) (11.5 min) and the (S) (19.3 min) enantiomers.
Id en tifica tion of th e Meta bolite of 2. Since 2 is rapidly
metabolized in plasma to the corresponding acid 29, the
standard analytical protocol for analysis of 2 in biological
samples was developed for 29. Analytical quantification was
carried out using LXA4 as an internal standard using a C18
column (Waters XTerra RP18, 5 µm, 2.1 mm × 50 mm) and a
10-80% “A” in “B” gradient over 2 min (6 min total run time).
Solvent “A” was water with 0.1% formic acid, and solvent “B”
was acetonitrile with 0.1% formic acid. The MS-MS method
monitored the m/e ) 403-111 (negative ionization conditions).
N -Me t h o x y -N -m e t h y l-2-(4-flu o r o p h e n o x y )e t h a n -
a m id e (19). Oxalyl chloride (60 mL, 686 mmol) and a catalytic
amount of dimethylformamide were added to a stirred suspen-
sion of 2-(4-fluorophenoxy)ethanoic acid (18, 97.3 g, 572 mmol)
in dichloromethane (500 mL). After 22 h, the mixture was
concentrated under vacuum to obtain 108 g of the correspond-
ing acid chloride as a yellow oil in quantitative yield: 1H NMR
(CDCl3) δ 4.90 (s, 2H), 6.84 (m, 2H), 6.99 (m, 2H) ppm. The
residue was added slowly to a stirred suspension of N,O-
dimethylhydroxylamine hydrochloride (55.80 g, 572 mmol) in
saturated potassium carbonate and ethyl acetate (375 mL). A
moderately exothermic reaction occurred. After 20 min, the
reaction mixture was partitioned between water and ether.
The ether layer was washed with 1 M hydrochloric acid and
saturated sodium chloride, dried over magnesium sulfate, and
concentrated under vacuum to give 19 as an off-white crystal-
line solid, 113.05 g (73% yield from starting acid): 1H NMR
(CDCl3 ,400 mHz) δ 3.21 (s, 3H), 3.73 (s, 3H), 4.75 (s, 2H),
6.87 (m, 2H), 6.95 (m, 2H) ppm.
(3S )-1-Br om o-4-(4-flu or op h e n oxy)-3-h yd r oxy-1-b u -
tyn e (23). A mixture of 21 (2.5 g, 14 mmol), N-bromosuccin-
imide (2.74 g, 15.4 mmol), and silver nitrate (0.12 g, 0.7 mmol)
in acetone (70 mL) was stirred at ambient temperature. The
pale solution became cloudy over 30 min. The mixture was
concentrated. Filtration through a plug of silica gel, eluting
with 20% ethyl acetate in hexane, gave 3.6 g (99%) of a pale-
yellow oil: 1H NMR (CDCl3) δ 3.95-4.15 (m, 2H), 4.75 (m, 1H),
6.86 (m, 2H), 6.97 (m, 2H) ppm.
(1E ,3S )-1-Br om o-4-(4-flu or op h e n oxy)-3-h yd r oxy-1-
bu ten e (17). Aluminum chloride (2.79 g, 21 mmol) was added
in portions to a mixture of lithium aluminum hydride (1.06 g,
28 mmol) in ether (70 mL). A solution of 23 (3.6 g) in ether
(10 mL) was added cautiously. A vigorous reaction with
evolution of hydrogen gas was observed. The mixture was

Lipoxin A4 Analogues
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8 2163
warmed to reflux in a water bath for 30 min. The reaction
mixture was cooled to 0 °C and treated sequentially with 2.8
mL of water, 2.8 mL of 15% NaOH, and 8.4 mL of water. The
resulting suspension was stirred for 10 min and filtered, and
the solids were washed with THF and ether. The filtrate was
concentrated to afford 2.94 g (81%): 1H NMR (CDCl3) δ 2.41
(t, 1H), 3.85 (dd, 1H), 3.99 (dd, 1H), 4.50 (m, 1H), 6.31 (dd,
1H), 6.52 (dd, 1H), 6.83 (m, 2H), 6.97 (t, 2H) ppm.
(m, 1H), 5.02 (m, 1H), 5.48 (dd, 1H), 5.6 (d, 1H), 6.16 (dd, 1H),
6.82 (dd, 1H) ppm. The residue was dissolved in methylene
chloride (400 mL) and was treated with iodine until a red color
persisted. After about 1 h, the reaction mixture was treated
with an aqueous solution of sodium hydrosulfite, dried, filtered,
and concentrated. Purification by chromatography on silica gel
using using a gradient of ether in hexane gave 27.8 g (49%) of
the desired product: [R]D -19.10 (10.88 mg/cc, MeOH); 1H
NMR (CDCl3) δ 0.15 (s, 9H), 1.3 (m, 2H), 1.4 (s, 9H), 1.6 (m,
8H), 3.45 (m, 2H), 3.92 (m, 2H), 4.34 (m, 1H), 4.62 (m, 1H),
5.54 (d, 1H), 5.72 (dd, 1H), 6.26 (dd, 1H), 6.56 (dd, 1H) ppm.
Anal. (C24H38SiO5‚0.4H2O) C, H.
1,1-Dim eth yleth yl [[(2S,3R)-3-[(1E,3E)-1,3-Hexa d ien -5-
yn yl]-1,4-dioxaspir o[4,5]dec-2-yl]m eth oxy]eth an oate (10).
A solution of 27 (27.8 g, 63 mmol) in THF (25 mL) was cooled
in an ice bath and treated with a solution of tetrabutylammo-
nium fluoride in THF (71 mL, 71 mmol, 1 M in THF). After
about 1 h, the reaction mixture was diluted with saturated
aqueous monobasic potassium phosphate and ether. The
aqueous layer was washed with ether (2×). The combined
organic layers were washed with saturated aqueous monobasic
potassium phosphate solution and brine solution, dried, and
concentrated to give 23.4 g (99%) of an oil: 1H NMR (CDCl3)
δ 1.3 (m, 2H), 1.4 (s, 9H), 1.6 (m, 8H), 3.02 (s, 1H), 3.5 (m,
2H), 3.96 (m, 2H), 4.38 (q, 1H), 4.66 (t, 1H), 5.54 (dd, 1H), 5.78
(dd, 1H), 6.33 (dd, 1H), 6.65 (dd, 1H) ppm.
(2R,3R)-3-(1,2-Dih yd r oxyp r op yl)-1,4-d ioxa sp ir o[4,5]-
d eca n e-2-ca r boxa ld eh yd e (24). Rhamnose hydrate (100 g,
0.55 mol) was converted into 92.3 g (0.31 mol, 57%) of
cyclohexylidene ketal as described earlier:23 [R]D +0.46 (10.49
1
mg/cc, MeOH); H NMR (CDCl3) δ 1.34 (d, 3H), 1.40 (m, 2H),
1.6 (m, 8H), 2.78 (d, 1H), 3.0 (s, 1H), 3.9 (m, 1H), 4.07 (m, 1H),
4.6 (d, 1H), 4.9 (m, 1H), 5.4 (s, 1H) ppm. Anal. (C12H20O5‚
0.5C6H10O1) C, H.
(2R,3S)-r2-(1-Hyd r oxyeth yl)-1,4-dioxa sp ir o[4,5]deca n e-
2,3-d im eth a n ol (25). A slurry of sodium borohydride (34.2
g, 0.9 mol) in methanol (400 mL) was cooled in an ice bath
and treated with a solution of 24 (92 g, 0.27 mol) in methanol
(200 mL). The reaction mixture was stirred for about 4 h before
the addition of acetic acid to adjust the pH to about 6 (about
120 mL). The reaction mixture was filtered. The filtrate was
concentrated to give a slightly yellow viscous oil. Treatment
of an ether solution with hexane gave 81.2 g (88%) of an off-
white solid after drying: 1H NMR (CD3OD) δ 1.28 (d, 3H), 1.43
(m, 2H), 1.7 (m, 8H), 3.42 (dd, 1H), 3.7 (m, 3H), 4.25 (m, 1H),
4.42 (dd, 1H) ppm.
1,1-Dim eth yleth yl [[(2S,3R)-3-[(1E,3E,7E,9S)-10-(4-F lu -
or op h e n oxy)-9-h yd r oxyl-1,3,7-d e ca t r ie n -5-yn yl]-1,4-
d ioxa sp ir o[4,5]d ec-2-yl]m eth oxy]eth a n oa te (8). A solution
of 17 (16.6 g, 63 mmol), solid tetrakistriphenylphosphine-Pd-
(0) (3.67 g, 3 mmol), and Cu(I) iodide (1.2 g, 6.3 mmol) in
diethylamine (50 mL) and THF (800 mL) was stirred and
deoxygenated by bubbling argon through the mixture for 90
min. An argon deoxygenated solution of 10 (23 g, 63 mmol) in
200 mL of THF was added dropwise. After 2 h, the reaction
mixture was diluted with hexane (about 400 mL), treated with
silica gel (about 40 g), and filtered. The solid was washed with
a 1:1 solution of ether and hexane. The filtrate was concen-
trated to give 36.8 g of an oil. Purification by chromatography
on silica gel using a 15-50% gradient of ether in hexane gave
1,1-Dim eth yleth yl [[(2S,3R)-3-(1,3-Dih yd r oxyp r op yl)-
1,4-d ioxa sp ir o[4,5]d ec-2-yl]m eth oxy]a ceta te (26). A mix-
ture of 25 (81 g, 0.32 mol) and tert-butyl bromoacetate (77 g,
0.39 mol, 1.2 equiv) in 1 L of toluene and 80 mL of aqueous
sodium hydroxide (25 wt %) was stirred as tetrabutylammo-
nium sulfate (7.8 g, 23 mmol, 0.07 equiv) was added. After
about 16 h, the reaction mixture was diluted with ethyl acetate
and saturated aqueous monobasic potassium phosphate. The
combined organic layers were dried and concentrated to give
a clear oil. Purification on silica gel using a step gradient of
ether in hexane (20-50%) gave 50.8 g (44%) of an oil: [R]D
1
+8.59 (10.30 mg/cc, MeOH); H NMR (CDCl3) δ 1.24 (d, 3H),
1
16.9 g (50%) as an oil: [R]D -21.17 (10.16 mg/cc, MeOH); H
1.35 (m, 2H), 1.47 (s, 9H), 1.6 (m, 8H), 3.6 (m, 2H), 3.8 (m,
2H), 3.95 (m, 2H), 4.32 (m, 1H), 4.4 (m, 1H) ppm. Anal.
(C18H32O7‚0.2H2O) C, H.
NMR (CDCl3) δ 1.3 (m, 2H), 1.4 (s, 9H), 1.6 (m, 8H), 2.42 (s,
1H), 3.5 (d, 2H), 3.96 (m, 4H), 4.38 (q, 1H), 4.58 (m, 1H), 4.66
(t, 1H), 5.72 (m, 1H), 5.78 (dd, 1H), 6.03 (m, 1H), 6.16 (dd,
1H), 6.33 (dd, 1H), 6.58 (dd, 1H), 6.88 (m, 4H) ppm. Anal.
(C31H39FO7) C, H, F.
1,1-Dim eth yleth yl [[(2S,3S)-3-F or m yl-1,4-d ioxa sp ir o-
[4,5]d ec-2-yl]m eth oxy]a ceta te (12). A solution of 26 (50 g,
138 mmol) in acetone (350 mL) was treated with a solution of
periodate (50 g, 235 mmol, 1.7 equiv) in water (1.2 L). After
about 4 h, solvent was removed by distillation and the residue
was extracted with ethyl acetate (3 × 500 mL). The combined
organic layers were dried and concentrated under reduced
pressure without heating to give 40 g (92%) of a clear oil: [R]D
-1.14 (10.15 mg/cc, MeOH); 1H NMR (CDCl3) δ 1.38 (m, 2H),
1.42 (s, 9H), 1.61 (m, 8H), 1.73 (m, 2H), 3.52 (dd, 1H), 3.72
(dd, 1H), 3.88 (s, 2H), 4.38 (dd, 1H), 4.52 (m, 1H), 9.62 (s, 1H)
ppm. Anal. (C16H26O6) C, H.
(5R ,6R ,7E ,9E ,13E ,15S)-16-(4-F lu or op h e n oxy)-3-oxa -
5,6,15-tr ih yd r oxy-7,9,13-h exa d eca tr ien -11-yn oic Acid (5).
A solution of 8 (1 g, 1.8 mmol) in acetic acid (50 mL) was
diluted with ethyl acetate (50 mL) and placed in a 55 °C oil
bath for 20 h. The reaction was complete by TLC analysis.
Acetic acid and ethyl acetate were removed by distillation
under high vacuum. The residue was diluted with water and
extracted with ethyl acetate (3×). The combined organic layers
were washed with water, saturated aqueous sodium carbonate,
water, and brine solution, dried, and concentrated to give 0.9
g of an oil. Chromatography on an HP-20 column, eluting with
a gradient of methanol in water, gave 0.5 g (58%) of 28 upon
concentration. The combined fractions were treated with 1 N
sodium hydroxide solution (2 mL) and concentrated. The
reaction was complete by TLC after about 1 h and the mixture
was placed on a column of CHP20P resin (75-150 µm,
Mitsubishi Chemical Corp.) and eluted using a gradient of
methanol in water to give 0.3 g (68%) of the desired product
that solidified upon standing: 1H NMR (CD3OD) δ 3.63 (m,
1H), 3.67 (m, 2H), 3.86 (dd, J ) 9.9, 6.7 Hz, 1H), 3.94 (dd, J )
9.7, 4.5 Hz, 1H), 4.11 (s, 2H), 4.15 (t, J ) 6.1, 1H), 4.50 (dd, J
) 5.4, 1.5 Hz, 1H), 5.76 (dd, J ) 15.3, 2 Hz, 1H), 5.95 (dd, J )
15, 6.6 Hz, 1H), 6.00 (dt, J ) 17, 2 Hz, 1H), 6.20 (dd, J ) 15.8,
5.5 Hz, 1H), 6.38 (dd, J ) 15.8, 10.8 Hz, 1H), 6.60 (dd, J )
15.8, 10.8 Hz, 1H), 6.95 (m, 4H) ppm; 13C NMR (100 MHz,
CD3OD) δ 69.1, 71.2, 73.2, 73.5, 73.9, 74.7, 90.7, 91.0, 112.2,
112.4, 116.7, 116.7, 116.9, 116.9, 131.9, 137.1, 142.4, 142.8,
156.5, 158.8, 174.5 ppm. Anal. (C21H23FO7‚0.2H2O) C, H, F.
1,1-Dim eth yleth yl [[(2S,3R)-3-[(1E,3E)-6-(Tr im eth ylsi-
lyl)-1,3-h e xa d ie n -5-yn yl]-1,4-d ioxa sp ir o[4,5]d e c-2-yl]-
m eth oxy]eth a n oa te (27). A slurry of phosphonium salt (13,
67.1 g, 0.14 mol) in THF (875 mL) was stirred under nitrogen,
cooled in a dry ice acetonitrile bath (-30 °C internal), and
n
treated with a solution of BuLi (66.5 mL, 0.133 mol, 2 M in
hexane) via dropwise addition. The reaction mixture was
allowed to warm to about 0 °C, then cooled back to about -30
°C. The reaction mixture was treated with a solution of 12
(40 g, 0.127 mol) in 125 mL of THF. After 1 h, the reaction
was diluted with saturated potassium phosphate (pH 5). The
aqueous layer was washed with ether (3×). The combined
organic layers were washed with water and brine, dried, and
concentrated. Impurities were precipitated from an ether
solution with hexane. The concentration of the filtrate gave
50.29 g of a 2:1 mixture of E,Z to E,E isomers by NMR
analysis. Z,E isomer: 1H NMR (CDCl3) δ 0.15 (s, 9H), 1.3 (m,
2H), 1.4 (s, 9H), 1.6 (m, 8H), 3.45 (m, 2H), 3.92 (m, 2H), 4.34

2164 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8
Guilford et al.
(3) Adams, J .; Fitzsimmons, B. J .; Rokach, J . Synthesis of Li-
poxins: Total Synthesis of Conjugated Trihydroxy Eicosatet-
raenoic Acids. Tetrahedron Lett. 1984, 25, 4713-4716.
(4) Corey, E. J .; Su, W. G. Simple Synthesis and Assignment of
Stereochemistry of Lipoxin A. Tetrahedron Lett. 1985, 26, 281-
284.
(5) Nicolaou, K. C.; Webber, S. E. A General Strategy for the Total
Synthesis of the Presumed Lipoxin Structures. J . Chem. Soc.,
Chem. Commun. 1985, 297-298.
(6) Rodriguez, A. R.; Spur, B. W. Total Synthesis of aspirin-triggered
15-epi-lipoxin A4. Tetrahedron Lett. 2001, 42, 6057-6060.
(7) Levy, B. D.; Serhan, C. N. Exploring new approaches to the
treatment of asthma: potential roles for lipoxins and aspirin-
triggered lipid mediators. Drugs Future 2003, 39, 373-384.
(8) Wallace, J . L.; Fiorucci, S. A Magic Bullet for Mucosal Protection
... and Aspirin Is the Trigger! Trends Pharmacol. Sci. 2003, 24,
323-326.
(9) Clish, C. B.; Levy, B. D.; Chiang, N.; Tai, H. H.; Serhan, C. N.
Oxidoreductases in lipoxin A4 metabolic inactivation: A novel
role for 15-oxaprostaglandin 13-reductase/leukotriene B4 12-
hydroxydehydrogenase in inflammation. J . Biol. Chem. 2000,
33, 25372-25380.
(10) Serhan, C. N.; Maddox, J . F.; Petasis, N. A.; Akritopoulou-Zanze,
I.; Papayianni, A.; Brady, H. R.; Colgan, S. P.; Madara, J . L.
Design of Lipoxin A4 Stable Analogs That Block Transmigration
and Adhesion of Human Neutrophils. Biochemistry 1995, 34,
14609-14615.
(11) Levy, B. D.; De Sanctis, G. T.; Devchand, P. R.; Kim, E.;
Ackerman, K.; Schmidt, B. A.; Szczeklil, W.; Drazen, J . M.;
Serhan, C. N. Multi-pronged inhibition of airway hyper-
responsiveness and inflammation by lipoxin A4. Nat. Med. 2002,
8, 1018-1023.
(5R,6R,7E,9E,11Z,13E,15S)-16-(4-Flu or oph en oxy)-3-oxa-
5,6,15-tr ih yd r oxy-7,9,11,13-h exa d eca tetr a en oic Acid (4).
Activated zinc was prepared from 10 g of zinc using the Boland
procedure.22 A solution of ester (0.25 g, 0.5 mmol) in methanol
(5 mL) was added to a slurry of activated zinc in a 1:1 mixture
of methanol to water (45 mL). The flask was stirred vigorously
under nitrogen for 24 h. The mixture was filtered through a
pad of non-acid-washed Celite 545, and the pad was rinsed
with methanol (3 × 25 mL). The fractions were combined,
condensed, and placed on a column of CHP20P resin. The
tetraene ester was eluted with a gradient of methanol in water
(20-100%) and treated with 1 N aqueous sodium hydroxide
(2 mL). After about 1 h, the reaction mixture was placed on
an CHP20P column and the product was eluted with a
gradient of methanol in water (20-100%) to afforded 65 mg
(30%) of the sodium salt, which solidified after concentration.
Since coupling constants could not be obtained on the sample
in methanol-d4, DMSO-d6 was used: 1H NMR (DMSO-d6) δ
3.20 (m, 1H), 3.48 (m, 2H), 3.5 (q, J ) 15.9 Hz, 1H), 3.88 (dq,
J ) 14.8, 4.8 Hz, 2H), 3.97 (m, 1H), 4.42 (m, 1H), 5.28 (d, J )
4.9 Hz, 1H), 5.82 (dd, J ) 14, 5.6 Hz, 1H), 6.00 (m, 3H), 6.32
(m, 2H), 6.55 (br s, 1H), 6.7 (dd, J ) 14, 9.9 Hz, 1H), 6.85 (dd,
J ) 15, 9.9 Hz, 1H), 6.94 (m, 2H), 7.09 (m, 2H); 13C NMR (125
mHz, CD3OD) δ 72.1, 72.2, 73.9, 74.2, 74.3, 75.3, 117.0, 117.2,
117.5, 117.5, 128.7, 129.5, 130.3, 131.6, 133.0, 133.3, 134.7,
135.4, 136.0, 157.1, 158.3, 160.2, 178.9. Anal. (C21H24FO7Na‚
0.8H2O) C, H, F, Na.
Cu ta n eou s Mod el of In fla m m a tion . The calcium iono-
phore induced model of acute inflammation was used as
described.12 Briefly, calcium ionophore A-23187 (10 µL, 0.1%
w/v), with or without the anti-inflammatory agent, was applied
topically to the dorsal surface of mouse ears, using isopropy-
lmyristate as the vehicle. The animals were euthanized after
24 h. Ear weight was used as an indicator of edema. The extent
of granulocyte and neutrophil infiltration was determined on
the basis of the measurement of peroxidase and elastase
activity, respectively, in mouse ear homogenates.
Air P ou ch Mod el of In fla m m a tion . The mice used in the
study were age-matched male Balb/c, 6-8 weeks old [pur-
chased from Sprague-Dawley] and were in groups of 20. On
day 0, all mice were anesthetized with 3% isofluorane. Dorsal
areas were shaved, cleaned with alcohol swabs, and injected
subcutaneously with sterile air (3 mL, passed through 20 µm
filter) near the tail. On day 3, all mice were injected with 3
mL of sterile air. On day 6, the air pouches of anesthetized
mice were injected sequentially with 900 µL of either 2 or 3
in 0.1% ethanol/PBS or vehicle and with 100 000 units of
recombinant mouse TNF-R (0.1 mL PBS; Boehringer Man-
nheim). Four hours after TNF injection, mice were euthanized
with carbon dioxide. Air pouches were lavaged three times
with 3 mL of sterile PBS. The cellular exudates were centri-
fuged for 15 min at 2000 rpm, and the resulting cell pellets
were resuspended in 1 mL of sterile PBS/10% BSA. The total
number of leukocytes, collected from the pouches, was counted
using a Coulter Z1 counter. Total neutrophils were enumerated
by differential cell counts on stained slides.
(12) Schottelius, A. J .; Giesen, C.; Asadullah, K.; Fierro, I. M.; Colgan,
S. P.; Bauman, J .; Guilford, W. J .; Perez, H. D.; Parkinson, J . F.
An aspirin-triggered lipoxinA4 stable analog displays a unique
topical anti-inflammatory profile. J . Immunol. 2002, 168, 7064-
7070.
(13) Gewirtz, A. T.; Collier-Hyams, L. S.; Young, A. N.; Kucharzik,
T.; Guilford, W. J .; Parkinson, J . F.; Williams, I. R.; Neish, A.
S.; Madara, J . L. Lipoxin A4 analogs attenuate induction of
intestinal epithelial proinflammatory gene expression and re-
duce the severity of dextran sodium sulfate-induced colitis. J .
Immunol. 2002, 168, 5260-5267.
(14) Aliberti, J .; Serhan, C. N.; Sher, A. Parasite-induced lipoxin A4
is an endogenous regulator of IL-12 production and immunopa-
thology in Toxoplasma gondii infection. J . Exp. Med. 2002, 196,
1253-1262.
(15) Clish, C. B.; O’Brien, J . A.; Gronert, K.; Stahl, G. L.; Petasis, N.
A.; Serhan, C. N. Local and systemic delivery of a stable aspirin-
triggered lipoxin prevents neutrophil recruitment in vivo. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 8247-8252.
(16) Phillips, E. D.; Chang, H.-F.; Holmquist, C. R.; McCauley, J . P.
Synthesis of methyl (5R,6R,7E,9E,11Z,13E,15S)-16-(4-fluorophe-
noxy)-5,6,15-trihydroxy-7,9,11,13-hexadecatetraenoate, an ana-
logue of 15R-lipoxin A4. Bioorg. Med. Chem. Lett. 2003, 13,
3223-3226.
(17) Nicolaou, K. C.; Veale, C. A.; Webber, S. E.; Katerinopoulos, H.
Stereocontrolled Total Synthesis of Lipoxins A. J . Am. Chem.
Soc. 1985, 107, 7515-7518.
(18) Robinson, R. S.; Clark, J . S.; Holmes, A. B. The Synthesis of
(+)-Laurencin. J . Am. Chem. Soc. 1993, 115, 10400-10401
(19) Gooding, O. W.; Beard, C. C.; Cooper, G. F.; J ackson, D. Y. Triply
Convergent Synthesis of 15-(Phenoxymethyl) and 4,5-Allenyl
Prostaglandins. Preparation of an Individual Isomer of Enprostil.
J . Org. Chem. 1993, 58, 3681-3686.
(20) Hofmeister, H.; Annen, K.; Laurent, H.; Wiechert, R. A Novel
Entry to 17R-Bromo- and 17R-Iodoethynyl Steroids. Angew.
Chem., Int. Ed. Engl. 1984, 23, 727-729.
Ack n ow led gm en t. The authors thank Eginhard
Matzke and Detlef Opitz (Research Business Area
Dermatology, Schering AG) and J erry Dallas (Biophys-
ics, Berlex Biosciences) for excellent technical assis-
tance. We acknowledge the support of Drs. H. Daniel
Perez, Khusru Asadullah, and Gary B. Phillips for this
work.
(21) Bohlmann, F.; Rotard, W. Synthese von (2E,4Z)-2,4,11-Dodec-
atrien-1-al, einem Abbauprodukt der Linolensaure (Synthesis
of (2E,4Z)-2,4,11-Dodecatrien-1-al, a Degradation Product of
Linolenic Acid. Liebigs Ann. Chem. 1982, 1216-1219.
(22) Boland, W.; Schroer, N.; Sieler, C.; Feigel, M. 96. Sterospecific
Syntheses and Spectroscopic Properties of Isomeric 2,4,6,8-
Undecatetraenes. New Hydrocarbons from the Marine Brown
Alga Giffordia mitchellae Helv. Chim. Acta 1987, 70, 1025-1040.
(23) Wheatley, J . R.; Beacham, A. R.; Lilley, P. M. deQ.; Watkin, D.
J .; Fleet, G. W. J . Ketals of L-Rhamnoheptonolactones: Potential
Mimics of L-Rhamnose. Tetrahedron: Asymmetry 1994, 5, 2523-
2534.
Refer en ces
(24) Ellis, C. K.; Smigel, M. D.; Oates, J . A.; Oelz, O.; Sweetman, B.
J . Metabolism of prostaglandin D2 in the monkey. J . Biol. Chem.
1979, 254, 4152-4163.
(25) Sturzebecher, S.; Haberey, M.; Muller, B.; Schillinger, E.;
Schroder, G.; Skuballa, W.; Stock, G.; Vorbruggen, H.; Witt, W.
Pharmacological profile of a novel carbacyclin derivative with
high metabolic stability and oral activity in the rat. Prostag-
landins 1986, 31, 95-109.
(1) Serhan, C. N. Lipoxins and aspirin-triggered 15-epi-lipoxin
biosynthesis: an update and role in anti-inflammation and pro-
resolution. Prostaglandins Other Lipid Mediators 2002, 68-69,
433-455.
(2) Serhan, C. N.; Hamberg, M.; Samuelsson, B. Trihydroxytet-
raenes: a novel series of compounds formed from arachidonic
acid in human leukocytes. Biochem. Biophys. Res. Comm. 1984,
118, 943-949.

Lipoxin A4 Analogues
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 8 2165
(26) Guindon, Y.; Delorme, D.; Lau, C. K.; Zamboni, R. Total
Synthesis of LTB4 and Analogues. J . Org. Chem. 1988, 53, 267-
275.
(27) Skuballa, W.; Schillinger, E.; Stuerzebecher, C. St.; Vorbrueggen,
H. Synthesis of a New Chemically and Metabolically Stable
Prostacyclin Analogue with High and Long-Lasting Oral Activ-
ity. J . Med. Chem. 1986, 29, 313-315.
(28) Takano, T.; Clish, C. B.; Gronert, K.; Petasis, N.; Serhan, C. N.
Neutrophil-Mediated Changes in Vascular Permeability Are
Inhibited by Topical Application of Aspirin-Triggered 15-Epi-
lipoxin A4 and Novel Lipoxin B4. J . Clin. Invest. 1998, 101, 819-
826.
(29) Hachicha, M.; Pouliot, M.; Petasis, N. A.; Serhan, C. N. Lipoxin
(LX) A4 and aspirin-triggered 15-epi-LXA4 inhibit tumor ne-
crosis factor 1alpha-initiated neutrophil responses and traffick-
ing: regulators of a cytokine-chemokine axis. J . Exp. Med. 1999,
189, 1923-1930.
(30) Serhan, C. N.; Takano, T.; Clish, C. B.; Gronert, K.; Petasis, N.
Aspirin-triggered 15-epi-lipoxin A4 and novel lipoxin B4 stable
analogues inhibit neutrophil-mediated changes in vascular
permeability. Adv. Exp. Med. Biol. 1999, 469, 287-293.
J M030569L