Development of potent non-carbohydrate imidazole-based small molecule selectin inhibitors with antiinflammatory activity

Article abstract of DOI:10.1021/jm000508c

A novel series of non-carbohydrate imidazole-based selectin inhibitors has been discovered via high-throughput screening using a P-selectin ELISA-based assay system. The initial lead 1 had an IC₅₀ of 17 μM in the P-selectin ELISA; this potency was significantly improved via an extensive SAR exploration. One of the current lead compounds (29) has an IC₅₀ of 300 nM in a P-selectin ELISA; it also has good activity in P- and E-selectin cell adhesion assays and shows efficacy in vivo. These compounds represent a novel series of sLe^X mimetics with antiinflammatory activity. Their unique profile supports our interest in their further evaluation as drug candidates for the treatment of inflammation. Herein we describe the syntheses, optimization, and SAR of this series of novel potent selectin antagonists.

Full text of DOI:10.1021/jm000508c

2094
J . Med. Chem. 2001, 44, 2094-2107
Develop m en t of P oten t Non -Ca r boh yd r a te Im id a zole-Ba sed Sm a ll Molecu le
Selectin In h ibitor s w ith An tiin fla m m a tor y Activity
Deborah H. Slee,* Suzanne J . Romano, J inghua Yu, Truc N. Nguyen, J udy K. J ohn, Neil K. Raheja,
Frank U. Axe, Todd K. J ones, and William C. Ripka
Ontogen Corporation, 2325 Camino Vida Roble, Carlsbad, California
Received November 27, 2000
A novel series of non-carbohydrate imidazole-based selectin inhibitors has been discovered via
high-throughput screening using a P-selectin ELISA-based assay system. The initial lead 1
had an IC50 of 17 µM in the P-selectin ELISA; this potency was significantly improved via an
extensive SAR exploration. One of the current lead compounds (29) has an IC50 of 300 nM in
a P-selectin ELISA; it also has good activity in P- and E-selectin cell adhesion assays and
shows efficacy in vivo. These compounds represent a novel series of sLeX mimetics with
antiinflammatory activity. Their unique profile supports our interest in their further evaluation
as drug candidates for the treatment of inflammation. Herein we describe the syntheses,
optimization, and SAR of this series of novel potent selectin antagonists.
In tr od u ction
The inflammatory response is mediated by a series
of chemoattractants, cytokines, and cell adhesion mol-
ecules (CAMs) which work in a programmed, sequential
manner to direct immune cells to the inflamed tissue.
This has been termed “the leukocyte-endothelial cas-
cade”.1,2 Three known families of CAMs participate in
this cascade: the selectins, the integrins, and the
immunoglobulin superfamily. The first step, rolling of
leukocytes and lymphocytes along the blood vessel wall,
is mediated by the selectins.1-5 A number of natural
ligands for the selectins have been identified. The
specific carbohydrate moieties necessary for selectin
binding, or the common motifs recognized by all three
selectins, are the sialylated and fucosylated tetra-
saccharide sialyl Lewisx (sLex) and a related structure,
sialyl Lewisa (sLea).6-8 The selectins have been impli-
cated in a number of inflammatory disease states, such
as asthma,9-11 atopic dermatitis, contact dermatitis,
(i.e., cutaneous inflammation),12-15 and rheumatoid
arthritis16-18 among others. The aforementioned studies
F igu r e 1.
have clearly demonstrated that selectin-mediated rolling
of immune cells along vascular endothelium is a key
early event in the pathogenesis of acute inflammation,
making it a good target for therapeutic intervention. A
selectin inhibitor would be expected to attenuate in-
flammatory reactions and alleviate the symptoms of
inappropriate immune responses by reducing cellular
infiltration.
Many reports describing sugar-based selectin inhib-
itors,14,19-25 in addition to several reports of peptidic
inhibitors of the selectin/ligand interaction, have been
published.26,27 Typically these classes of inhibitors suffer
from a lack of potency, poor pharmacokinetic properties,
and are often difficult to synthesize. More importantly,
there have been few reports28,29 of non-carbohydrate,
non-peptidyl antagonists of the selectin/ligand interac-
tion; and in these cases only ELISA data have been
reported. It has been our approach to develop a program
for discovering potent small molecule inhibitors of
selectin-ligand binding, utilizing cell-based assays, with
the aim of developing a novel class of antiinflammatory
therapeutics.
The original lead imidazole-based compound 1 (IC50
) 17 ( 26 µM, Figure 1, Table 3) was discovered via
high-throughput screening of our compound library
against P-selectin using an ELISA-based assay system.
This ELISA assay measures the inhibition of binding
of a recombinant selectin protein to a synthetic sLex
analogue.30 Recent in vivo studies with knockout mice
and selectin antibodies suggest that it may be necessary
to block both P- and E-selectin in order to obtain a
potent antiinflammatory response.31,32
Herein we report the syntheses and optimization of
a series of novel potent imidazole-based P- and E-
* To whom correspondence should be addressed. Tel: 760-930-0100.
Fax: 760-930-0955. E-mail: debbie.slee@ontogen.com.
10.1021/jm000508c CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/17/2001

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2095
Sch em e 1a
a
Reagents and conditions: (i) KtBuO, DMSO/THF, 2 h, 53%; (ii) mCPBA, DCM, 40 °C, 91%; (iii) formic acid, 0 °C, THF, 5 h, 64%; (iv)
TEMPO, DCM, 0.7 M NaBr, buffered bleach, quantitative; (v) Pd(OAc)2, TEA, P(o-tolyl)3, 90% crude.
Sch em e 2
selectin antagonists, derived from lead compound 1,
which have been readily prepared from commercially
available starting materials.
either the tert-butyl or methyl ester of aldehyde 3
provided alkene 5 as a 1:1 mixture of geometrical
isomers. Epoxide formation employing mCPBA was
followed by ring opening with formic acid and subse-
quent hydrolysis of the resulting formic acid ester in
situ to provide diol 7. TEMPO oxidation of 7 furnished
the dione intermediate 8. This dione was further func-
tionalized via a Heck reaction with acrylamide 9e,
synthesized via reaction of 3-phenyl-1-propylamine with
acryloyl chloride (Scheme 5), to give the fully derivatized
dione 10. The dione 10 was condensed with NH4OAc
and the appropriate aldehyde 11 in acetic acid to give
imidazole 12 (Scheme 2). Final deprotection gave the
final desired imidazoles 13 and 14.
Ch em istr y
The imidazole-based selectin inhibitors were prepared
using the general procedures shown in Schemes 1, 2,
and 4. These schemes are illustrative of the approach
used for the synthesis of imidazoles 1, 13, 14, and 27-
31 (Table 3). The routes differ in the methodology used
for the preparation of the required intermediate diones.
The imidazole ring synthesis were carried out using
similar methodology.33 In the first example, shown in
Schemes 1 and 2, the synthesis of imidazole 12 is
outlined (12a f 13; 12b f 14).
Preparation of the aldehydes 3 and 21, shown in
Scheme 3, began with the commercially available mono-
diethyl acetal of terephthalaldehyde 15, which, upon
Scheme 1 illustrates the synthesis of the desired
intermediate dione 10. A Wittig olefination with 4 and

2096 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13
Slee et al.
Sch em e 3a
a
Reagents and conditions: (i) dioxane, hydroxylamine‚HCl, TEA, 72 h, 95%; (ii) bleach, THF, ∼18 h; (iii) Amberlite IR120, THF/H2O,
5 h, 92%.
Sch em e 4a
a
Reagents and conditions: (i) Pd(OAc)2, TEA, P(o-tolyl)3, 40%; (ii) NH4OAc, AcOH, R2CHO ) 3, 21, or paraformaldehyde, 100 °C,
36-92%; (iii) TFA:DCM 1:1 (for yields, see Table 1).
condensation with hydroxylamine, furnished oxime 16.
1,3-Dipolar cyclization of 16 with either tert-butyl
acrylate 17 or ethyl propiolate 19 in the presence of
bleach, followed by saponification, provided the requisite
aldehydes 3 and 21 in excellent overall yield.34
removed from diones 25a -d by column chromatog-
raphy. The diones can then be reacted with NH4OAc
and the appropriate aldehyde to give the desired imid-
azole of general formula 26. Deprotection of the tert-
butyl protecting group of the intermediate of general
formula 26 gives the desired imidazoles 1 and 27-31.
The yields for the last two steps in Scheme 4 are shown
in Table 1.
Scheme 4 shows the general synthetic route employed
for the synthesis of diones 25a -d from 4,4′-dibromo-
benzil, via sequential Heck reactions with tert-butyl
protected acrylic acid 17, and the appropriate acryl-
amide 9. The di-tert-butyl ester side product 24 is
carried through the reaction sequence and is easily
Scheme 5 illustrates the general approach used for
the synthesis of the acrylamides 9a -e used in Schemes
1 and 4. Acryloyl chloride was reacted with the ap-

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2097
Ta ble 1. Final Compound Yields for Scheme 4
Sch em e 5
(2) affinity data that have been obtained via chemical
modification of the sLex ligand36-46 or genetic modifica-
tion of the E-35,47,48 or P-selectin49-52 proteins, (3) the
force field derived model of sLex,53-56 and (4) a number
of studies utilizing NMR.57-60
In addition, the structure of complexes of sugar
ligands with carbohydrate recognition domains of two
closely related members of the C-type animal lectin
family, rat serum and liver mannose-binding proteins
(MBPs), have been resolved by X-ray crystallogra-
phy.61,62 These studies revealed that terminal mannose,
N-acetylglucosamine, or fucose residues interact directly
with the protein and indirectly through a bound cal-
cium. Additional mutagenesis studies by the same group
analyzed the mechanism of oligosaccharide binding to
the selectin cell adhesion molecules by transferring
regions of the carbohydrate recognition domains of E-
and P-selectin to the corresponding sites of the homolo-
gous rat serum mannose binding protein.63 One of the
main findings of these studies was that affinity for the
sLex ligand seemed to be dependent on the presence of
a number of lysine residues and that the sialic acid
portion of the ligand interacts near the sequence Lys-
Lys-Lys which corresponds to residues 111-113 of
propriate amine 32 to give the acrylamide 9a -e quan-
titatively. Acrylamides 9a -e were then used directly
without purification.
Mod elin g
The crystal structure of E-selectin is available from
the Protein Data Bank (PDB). A number of published
studies have established the key structural features of
sLex necessary for binding to the selectin proteins. These
include (1) the X-ray crystal structure of E-selectin,35

2098 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13
Slee et al.
F igu r e 3. Additional substitution at R2 allows reduction in
size of the R5 substituent.
F igu r e 2. Criteria believed to be essential for a potent selectin
inhibitor.
Ta ble 2. Alternative Hydrophobic Groups at R5 as
Replacements for a Straight Chain Alkyl Moiety Based on
Calculated Π Values (Substituent log P Values)
E-selectin. Several independent studies also suggest the
importance of Lys 111 and Lys 113 for ligand bind-
ing.35,47,49-51
The human P-selectin lectin domain shares a high
sequence homology of >80% (with no insertions or
deletions) with the human E-selectin lectin domain
which is available as a high-resolution structure from
the PDB as mentioned previously.35 Our model of
P-selectin was obtained from the X-ray structure of
E-selectin by simple substitution of the appropriate
amino acids. Recently, Camphausen has reported the
structure of both E- and P-selectin with a bound sialyl
Lewisx ligand. In this study, the structure of P-selectin
was indeed found to adopt a conformation similar to that
of E-selectin, and the sLex binding site was highly
conserved as predicted in our model.
Complexes were generated manually by docking
inhibitors to the binding epitope of P-selectin, such that
the proposed calcium binding carboxylate group bound
the calcium ion, and key amino acid residues involved
in sLex binding were blocked. These initial structures
were energy minimized, and the lowest energy structure
was selected. The final structure was subjected to a
Monte Carlo docking protocol as described in ref 26, to
determine whether there were any other lower energy
docked complexes.65 The docked complexes of the inhibi-
tors with P-selectin were energy minimized with the
Discover program66 using the consistent valence force
field (CVFF) of Hagler.67
Resu lts a n d Discu ssion
The original imidazole-based lead compound 1 (IC50
) 17 ( 26 µM (n ) 40), P-selectin ELISA, 98 µM
P-selectin cell-cell assay), was optimized via a combi-
nation of solid support71 and solution phase chemistry.
The synthesis of focused libraries on solid support was
carried out using our OntoBLOCK system.72
Overall, three minimum criteria appeared to be
essential for activity: (I) a calcium binding moiety, (II)
a rigid template/core, and (III) a hydrophobic moiety
(Figure 2).
Several different groups were tested as linkers for the
carboxylic acid at R4. We found that 4,5-dihydro-
isoxazole-5-carboxylic acid, phenoxy acetic acid (not
shown), and cinnamic acid, which are thought to act as
calcium-binding moieties at R4, gave equivalent activi-
ties. However, incorporation of the 4,5-dihydro-isox-
azole-5-carboxylic acid moiety increased the aqueous
solubility of this series. The tert-butyl ester precursors
Biology
High-throughput screening utilized an ELISA-based
assay system developed at Nippon Organon K. K.
(formerly Kanebo Ltd., New Drug Discovery Laborato-
ries) in Osaka, J apan.30 For lead optimization, we have
developed a cell-selectin protein and a cell-cell assay
system. Inhibitors were screened against both E- and
P-selectin (see Table 3).
The animal models used to evaluate these selectin
inhibitors in vivo include a mouse thioglycollate induced
peritonitis model of inflammation68 and a model of
selectin-mediated rolling in vivo using an intravital
microscopy system.69 In this model, selectin-mediated
rolling in cytokine-stimulated dorsal skin of anesthe-
tized mice is measured.70

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2099
Ta ble 3. Biological Assay Dataa
a
NT ) not tested. NA ) not active.
that we tested (e.g., 26) were found to be inactive,
providing evidence that the carboxylic acid moiety is
essential for activity.
smaller propyl phenyl group as illustrated by imidazole
14 (Figure 3). Activity was lost, however, when the R5
side chain was reduced to a benzyl group. Many differ-
ent amines were used to couple to the R5 side chain
through the acrylamide moiety. Coupling of amines that
were more hydrophilic in character, such as tryptamine
or N-(3-aminopropyl)morpholine for example, resulted
in loss of potency. It was found that overall, the most
potent derivatives in the cell-based assay systems
contain the C-16 alkyl chain at R5. Interestingly, reduc-
tion of the double bonds in either of the R4 and/or R5
side chains when R5 was dodecylalkyl or less, resulted
We were seeking to replace the long fatty side chain
of 1; however, preliminary SAR information indicated
that the general hydrophobic character of a hexadecyl
chain was required for potency. From our combinatorial
libraries we found that removal of the hydrophobic
moiety at R5 resulted in loss of activity. If a substituted
phenyl group was introduced at R2, good activity was
maintained in vitro in the ELISA and P-selectin cell
assays when the alkyl chain (R5) was replaced with the

2100 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13
Slee et al.
acid functionality at R4 binds to calcium, filling out the
coordination sphere in a manner similar to the way the
fucose moiety is believed to bind in the natural ligand
sLex. This places the imidazole ring in close proximity
to LYS 111 and LYS 113 where it may interact via
hydrogen bonding, again analogous to the natural ligand
sLeX 2. The amide group at R5 can interact with amino
acid side chains of SER 97 and SER 99 as well as TYR
94. The hydrophobic chains at R5 may possibly provide
a hydrophobic driving force to aid in keeping the
inhibitor bound. The rigid aromatic rings at R2 are
hydrophobic in nature and tend to interact with the
hydrophobic side chains of residues such as LYS 112.
Many commercially available aldehydes were em-
ployed to synthesize a series of imidazoles substituted
at the R2 position. Most of these substitutions were well
tolerated when the R4 and R5 substituents met the basic
requirements of a calcium binding moiety and a suitable
hydrophobic group, respectively. Close examination of
the proposed mode of binding suggested that a suitably
positioned carboxylic acid moiety might pick up a key
ionic interaction with LYS 112. Indeed the most potent
analogues were found when a phenyl ring, para-
substituted with a carboxylic acid-containing moiety,
was added at the R2 position. One of the most potent
analogues is the diacid 29, with an IC50 of 0.29 µM in
our P-selectin ELISA assay (14 µM P-selectin cell-cell
assay) (Table 3). We postulate that this imidazole
derivative 29 binds in a similar way as compound 14
with the additional acid moiety picking up an ionic
interaction with the terminal amino moiety of LYS-112
in P- or E-selectin as illustrated in Figure 5 (cross-eyed
stereoview).
F igu r e 4. Schematic representation of 14 showing the
proposed interactions with the P-selectin binding site.
in complete loss of potency in the cell-based assays,
presumably due to the loss of structural rigidity. Thus
implying that the direction of the hydrophobic group is
important.
To find an alternative side chain with similar hydro-
phobic character, groups with similar calculated Π
values (substituent log P values) were evaluated. Theo-
retical Π values for R5 were calculated for the corre-
sponding amine using the MOE software package.73 The
computed values for NC12H25 (4.31) and NC16H33 (6.07)
were used as reference (NC12H25 was the smallest
substituent that gave acceptable potency). Suitable side
chains with similar log P values included a phenyl-
heptyl moiety (compound 27, Π value of 4.01) and a bis-
hexyl moiety (compound 28, Π value of 4.13), as shown
in Table 2. Compound 27 has an IC50 of 44 µM which is
comparable to compound 28 (IC50 of 21 µM) and com-
pound 1 (IC50 of 40 µM) in the P-selectin cell assay
(Table 3).
Overall, when we compare the proposed binding mode
of our inhibitors to that of the known natural ligand
sLeX 2 (Figure 1), we postulate that the two carboxylic
acid moieties act as replacements for the critical fucose
and sialic acid groups of sLeX, with a triaryl imidazole
unit replacing the lactosamine core. This proposed
binding mode is similar to that discussed by Kogan et
al.11 for the mannose-containing inhibitor TBC-1269. If
the mode of binding is indeed similar, our series of
compounds suggest that the proposed calcium binding-
mannose functionality of TBC-1269 can be replaced by
a simple carboxylic acid moiety resulting in a completely
novel sLex mimetic that contains no sugar moieties.
Our proposed model of how the imidazole-based
inhibitors bind to P-selectin is illustrated in Figure 4.
The inhibitors were modeled on the assumption that the
F igu r e 5. Stereoview of compound 29 bound to P-selectin illustrating the proposed binding mode.

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2101
To confirm that our inhibitors were effective under
conditions of shear flow (i.e., under physiological condi-
tions) and that the antiinflammatory effect we had
observed in the thioglycollate model was via inhibition
of selectin-mediated rolling, we tested compound 29 for
its ability to inhibit selectin-mediated rolling in vivo
using an intravital microscopy system.69 This compound
was found to inhibit cytokine-stimulated rolling in the
dorsal skin of anesthetized mice at concentrations of
30-50 mg/kg iv.70 A combination of P- and E-selectin
monoclonal antibodies also blocked rolling in this sys-
tem, demonstrating that the rolling was indeed medi-
ated by selectins (data not shown).
Su m m a r y
A novel series of potent small molecule inhibitors that
not only have excellent in vitro profiles but also have
activity in vivo has been developed. A combination of
parallel synthesis and structure-based design allowed
rapid optimization, resulting in compounds with in vivo
activity. We found what we believe to be the three
minimal criteria that are essential for activity: (1) a
calcium binding moiety, (2) a rigid template/core, and
(3) a hydrophobic moiety (Figure 2). Of those that were
tested, the carboxylic acid moiety was found to be
optimal as the calcium binding moiety. If a substituted
phenyl group was introduced at R2, the alkyl chain (R5)
could be replaced with a smaller propyl phenyl moiety.
However, the optimal group at the R5 position was found
to be a C-16 alkyl chain. A number of templates could
be used as the rigid template, but the imidazole core
was found to be optimal in terms of aqueous solubility.
We found that many different substituted phenyl groups
were tolerated at the R2 position, but increased potency
was observed when this ring was para-substituted with
a carboxylic acid-containing functionality such as in
compound 29.
In general, the cell-selectin protein and cell-cell
assay systems proved to be essential for lead optimiza-
tion and seemed to give more predictive results in terms
of identifying compounds with in vivo activity. These
findings are similar to that of other groups.75,76 Several
members of this series were found to be active against
both P- and E-selectin.
One of the more potent imidazole derivatives, com-
pound 29, identified as having good activity in all of the
in vitro assays, has been shown to reduce inflammation
in vivo and has also been demonstrated to inhibit
selectin-mediated rolling in vivo. We believe that this
is the first report of such a series of non-carbohydrate
inhibitors with in vivo efficacy.
F igu r e 6. Treatment with compound 29 at 10 and 50 mg/kg
reduces neutrophil influx in an in vivo model of inflammation.
Balb/c mice were dosed with compound or vehicle (buffered
saline) by intravenous injection immediately before intra-
peritoneal challenge with thioglycollate. Total cells and neu-
trophils in the peritoneum were quantified 4 h later (n ) 4-5
mice per group). *p < 0.01 by unpaired, two-tailed t-test.
It was found that the chiral center of the 4,5-dihydro-
isoxazole-5-carboxylic acid moiety at R2 of compound 29
could be removed to give the isoxazole 30 which has an
IC50 of 0.13 µM in our P-selectin ELISA assay and 8
µM in our P-selectin cell-cell assay; this is approxi-
mately 2-fold more potent than the equivalent isoxazo-
line derivative 29.
When the alkyl chain at R5 is shortened to a dodecyl
chain (compound 31, Table 3), activity is maintained in
the P-selectin ELISA assay system, or when changed
to a propyl phenyl moiety (compound 13, Table 3),
activity is maintained in the P-selectin/cell assay sys-
tem; however, in both cases, potency is lost in the cell-
cell assay.
A number of templates were explored as alternatives
to the imidazole core. It was found that either one of
the nitrogens in the imidazole core could be replaced
with sulfur to give a thiazole moiety, suggesting that
the imidazole was functioning primarily as a rigid
scaffold. This also suggests that any interaction between
the selectin protein and the imidazole core is not
important for binding. A simple appropriately substi-
tuted phenyl ring also gave active compounds. However,
both the thiazole and phenyl derivatives were distinctly
less aqueous soluble than the imidazole derivatives.
Compounds 29 and 30 have excellent activity in our
in vitro P- and E-selectin cell-based assays (Table 3).
Compound 29, with potent cell-cell activity against
both P- and E-selectin, was used for in vivo testing. The
aqueous solubility of 29 is good as the bis-sodium salt
(15 mg/mL), allowing us to use this form of the com-
pound for our in vivo studies. Compound 29 was tested
in the mouse thioglycollate-induced peritonitis model68
and found to reduce the inflammatory response in a
dose-dependent manner, causing a 30-50% reduction
in cell infiltration at doses of 10-50 mg/kg iv (statisti-
cally significant results, Figure 6).70 This compound also
showed efficacy in a mouse model of IgE-dependent
delayed-type hypersensitivity.74
Exp er im en ta l Section
P -Selectin ELISA Assa y. An ELISA-type assay was used
to screen for inhibitors of selectin-ligand interactions. A
P-selectin-IgG chimera, constructed as described by Foxall,77
and sialyl Lewisx pentaceramide were obtained from Kanebo,
Ltd. (Osaka).78
Assays were performed essentially as described by Ohmoto
et al.30 Polystyrene microtiter plates (Falcon Pro-Bind) were
coated with the sialyl LewisX analogue at 40-100 pmol/well.
Coated wells were blocked with 5% bovine serum albumin
(BSA) in 50 mM imidazole buffer, pH 7.2, for 1 h at room
temperature.

2102 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13
Slee et al.
Compounds were diluted from DMSO stock solutions in
assay buffer (50 mM imidazole buffer, pH 7.2, containing 1%
BSA and 1 mM CaCl2). Compounds were always run in
duplicate or triplicate. A complex consisting of P-selectin IgG
chimera, biotinylated goat F(ab′)2 anti-human IgG, and strepta-
vidin-alkaline phosphatase conjugate was made in assay
buffer. Selectin chimera was omitted from the complex for
negative control (“background”) wells. The complex and the
test compounds (or vehicle controls) were combined in wells
of a polypropylene microtiter plate and incubated for 30 min
at room temperature. The complex-compound mixture was
then added to the blocked, sialyl LewisX-ceramide coated plate
and allowed to incubate for 45 min at 37 °C. After washing
three to four times with 50 mM imidazole, the bound complex
was detected using the colorimetric phosphatase substrate,
p-nitrophenyl phosphate, at 1 mg/mL in 1 M diethanolamine
containing 0.01% MgCl2. After developing for 1-2 h at room
temperature, the absorbance at 405 nm was measured in a
Molecular Devices microplate reader. Percent inhibition was
calculated by comparing the test compound result with the
vehicle control after subtracting the background from each.
IC50 values were calculated by in-house data analysis software
(OntoASSAY; Ontogen, Corp.) using standard algorithms.
in cold RPMI to 2× the final test concentration were added.
Triplicate test samples were incubated with the activated
HUVECs for 20 min at 4 °C. It was found that 4 °C was the
optimal temperature for detecting primarily selectin-mediated
adhesion (unpublished observations).
HL60 cells were labeled by incubating with the membrane-
permeable fluorescent dye, BCECF-AM, at 5 µg/mL for 30 min
at 37 °C. Cells were then washed and resuspended in RPMI
medium at 2 × 106 cells/mL and added to the compound-
containing wells in a volume of 50 µL. After incubating for 30
min at 4 °C, unbound cells were removed by aspiration and
washing two times with PBS+. Bound cells were quantified
by measuring fluorescence as described for the cell-selectin
assay. Percent inhibition was calculated as described above.
IC50 values were calculated by Probit analysis using Microsoft
Excel.
P -Selectin Cell-Cell Ad h esion Assa y. The procedure for
the P-selectin cell-cell adhesion assay was similar to that of
the E-selectin cell-cell assay described above, except that
Chinese hamster ovary cells transfected with human P-selectin
(CHO-Psel) served as the source of native P-selectin. These
cells, which were kindly provided by Dr. G. Kansas (North-
western Medical School, Chicago, IL), constitutively express
P-selectin on their surface. The CHO-Psel cells were propa-
gated in Hamm’s F12 medium containing 10% FBS. The day
before the assay, the cells were trypsinized and plated in 96-
well tissue culture dishes. As described above for the E-selectin
cell-cell assay, BCECF-labeled HL60 cells were allowed to
adhere to the monolayers at 4 °C in the presence of test
compounds, and bound cells were detected using a fluorescence
plate reader. Calculations were done as described above.
Cell-Selectin Ad h esion Assa ys. The ability of compounds
to inhibit the adhesion of HL60 cells to purified selectin
proteins was measured using a “cell-selectin” assay. Recom-
binant soluble P- and E-selectin proteins purchased from R&D
Systems (Minneapolis, MN) were diluted to 2.5 µg/mL in
Dulbecco’s PBS containing calcium and magnesium (PBS+).
Falcon Pro-Bind microtiter plate wells were incubated with
50 µL of the P- or E-selectin protein solution for 1 h at 37 °C
or overnight at 4 °C. The selectin protein was omitted from
negative control (“background”) wells. Coated wells were then
washed three times with PBS+ and then blocked with 1% BSA
in PBS+ for 1 h at room temperature. After blocking, the plates
were washed three times with PBS+. Compounds were diluted
to 2× final test concentration in PBS+ and added to the
blocked, selectin-coated wells in a volume of 50 µL. Samples
were always run in duplicate or triplicate. Compounds and
vehicle controls were preincubated in the wells for ∼20 min
at room temperature.
HL60 cells obtained from the ATCC (Manassas, VA) were
cultivated in RPMI medium containing 10% heat-inactivated
fetal bovine serum (FBS). For the assay, cells were harvested
by centrifugation, washed once with PBS+, and resuspended
in PBS+ at a concentration of 2 × 106 cells/mL. Cells were
added directly to the compound-containing wells in a volume
of 50 µL per well, bringing the compound to its final test
concentration in a total volume of 100 µL. Cells and compound
were incubated on the selectin-coated wells for 45 min at 37
°C. Unbound cells were removed using a vacuum manifold and
a single wash with 200 µL of PBS+ (added slowly using a
manual multichannel pipettor). Retained cells were labeled
directly on the plate by adding 5 µg/mL of the membrane-
permeable fluorescent dye, calcein-AM, and incubating for 30
min at 37 °C. The signal was quantified in a Wallac Victor
fluorescent microplate reader using 485 nm excitation and 535
nm emission. Percent inhibition and IC50 values were calcu-
lated as described above for the ELISA assay.
Ch em istr y. Nuclear magnetic resonance spectra (1H NMR)
were measured on either a Varian 300 MHz or a Varian 400
MHz. Chemical shifts (δ) are reported in parts per million
(ppm) downfield from tetramethylsilane (TMS). Data are
reported as follows: chemical shift, multiplicity (br ) broad,
s ) singlet, d ) doublet, t ) triplet, q ) quadruplet, m )
multiplet), integration, peak assignment, and coupling con-
stant (hertz). Elemental analyses (C, H, N) were performed
by Quantitative Technology Inc., Whitehouse, NJ , and results
were within 0.4% of calculated values except where noted.
High-resolution MALDI-FTMS (HRMS) were recorded by the
Scripps Research Institute Mass Spectrometry Facility, San
Diego, CA. Mass spectra were measured using Atmospheric
Pressure Chemical Formation (APcI) looking at positive and
negative modes on a Micromass LCZ (3 keV with a probe
temperature of 400 °C and a source block at 120 °C). LC
spectra for LC/MS were measured using an eluant of CH3CN
(0.1% CF3CO2H)/H2O (0.1% CF3CO2H) (V:V) on a Hewlett-
Packard HP1100 HPLC, in the range 200-300 nm with a
diode array detector (DAD); 5 µL per injection (Gilson 215
Autosampler) at an average concentration of 1 mg/mL; gradi-
ent: 10-100% CH3CN in 5 min, 100% CH3CN for 1 min, 100-
10% CH3CN in 0.2 min, 10% CH3CN for 1.4 min; LC element
split 1:4 directly into ion source (500 µL/min). Analytical HPLC
was performed on two different reverse phase systems with a
flow rate of 1.5 mL/min: HPLC system 1 - Hewlett-Packard
HP1100 HPLC with Zorbax 150 × 4.6 mm C-8 column (5 µm
particle sizes) and 200-300 nm DAD; 10-100% CH3CN
(solvent A) and 0.1% TFA in water (solvent B) were used as
mobile phase with a run time of 15 min. HPLC system 2 -
Thermo Separation Products HPLC with Kromsil 150 × 4.6
mm C-18 column (5 µm particle size) and 220 and 254 nm
variable UV detector; 40-100% CH3CN (solvent A) and 0.1%
TFA in water (solvent B) were used as the mobile phase with
a run time of 15 min. The purity of the final compounds is
above 95% except where noted. The chromatography columns
used for LC in LC/MS and HPLC were 50 × 4.6 mm C-8 with
5 µm particle sizes and Zorbax 150 × 4.6 mm C-8 with 5 µm
particle sizes, respectively. The same gradient was used in
HPLC as in LC for LC/MS. Reactions in solution phase were
monitored by thin-layer chromatography (TLC) using Merck
silica gel 60F-254-coated plates (0.25 mm thickness). Flash
E-Selectin Cell-Cell Ad h esion Assa ys. Cell-cell adhe-
sion assays were used to determine if compounds could inhibit
native selectin-native ligand interactions. Human umbilical
vein endothelial cells (HUVEC; Clonetics, Inc., San Diego, CA)
were used as a source of native E-selectin. Frozen stocks of
HUVECs were expanded and propagated in EGM medium
(Clonetics) for a maximum of four to five passages. The day
before the assay was to be run, HUVEC cells were trypsinized
and replated in Nunc 96-well tissue culture plates which were
precoated with 2% gelatin. On the day of the assay, the
HUVECs were activated with 10 ng/mL of IL-1â (R&D
Systems, Minneapolis, MN). It was found that this treatment
significantly upregulated expression of the E-selectin protein
(data not shown). After
a 4 h activation at 37 °C, the
stimulation medium was removed, and the compounds diluted

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2103
chromatography was performed using E. Merck silica gel 60
(230-400 mesh ASTM).
tate, 3:1) to give the desired cis and trans alkenes 5 as a pale
yellow oil (9.4 g, 53% yield). 1H NMR (400 MHz, CDCl3) cis
isomer: 7.64 (d, 2H, J ) 7.7), 7.55 (d, 2H, J ) 8.2), 7.48 (d,
2H, J ) 7.7), 7.40 (d, 2H, J ) 8.5), 7.10 (s, 2H), 5.08 (t, 1H, J
) 9.6), 3.6 (d, 2H, J ) 9.6), 1.5 (s, 9H); trans isomer: 7.55 (d,
2H, J ) 8.2), 7.35 (d, 2H, J ) 8.5), 7.28 (d, 2H, J ) 8.0), 7.10
(d, 2H, J ) 8.2), 6.63 (d, 1H, J ) 12.0), 6.57 (d, 1H, J ) 12.0),
5.08 (t, 1H, J ) 9.6), 3.60 (d, 2H, J ) 9.9), 1.50 (s, 9H).
Preparation of epoxide 6: The alkene 5 (9.4 g, 22 mmol)
was dissolved in DCM (100 mL), and then mCPBA (5.0 g, 22
mmol, purity 57-86%) in DCM (100 mL) was added. The
reaction was stirred at 40 °C for 10 h and then treated with
10% sodium sulfite until testing with starch paper was
negative. The reaction mixture was then extracted with DCM.
The combined organic extracts were washed with saturated
sodium bicarbonate and brine and dried over anhydrous
sodium sulfate. The product was concentrated to dryness. The
product was purified via flash chromatography eluting with
hexane:ethyl acetate (8:1 then 6:1). The desired epoxide 6 was
obtained as a pale yellow foam (8.9 g, 91% yield). 1H NMR
(400 MHz, CDCl3) mixture of isomers: 7.47 (m, 2H), 7.27 (m,
2H), 7.17 (m, 2H), 7.01 (m, 2H), 4.97 (m, 1H), 4.35 (m, 2H),
3.49 (m, 2H), 1.48 (s, 9H).
Opening of epoxide 6 to give diol 7: The epoxide 6 (11 g)
was dissolved in THF (15 mL). The solution was cooled in an
ice bath, and formic acid (30 mL) was added slowly followed
by water (0.5 mL). The reaction was stirred at 0 °C for 5 h.
Upon completion, the reaction was concentrated in vacuo. The
residue was dissolved in THF (40 mL) and treated with 1 N
NaOH (aqueous) until a color change was observed (yellow to
brown). The reaction was monitored carefully by TLC. Upon
completion, the product was extracted into ethyl acetate (2 ×
200 mL), dried (MgSO4), and concentrated in vacuo. The
product was purified by column chromatography, eluting with
30% EtOAc in hexane, to give the desired diol 7 (7.2 g, 64%).
1H NMR (400 MHz, CDCl3) 1:1 mixture of isomers: 7.56 (d,
2H, J ) 8.0), 7.52 (d, 2H, J ) 8.0), 7.38 (d, 2H, J ) 8.0), 7.35
(d, 2H, J ) 8.0), 7.19 (d, 2H, J ) 8.0), 7.12 (d, 2H, J ) 8.0),
7.03 (d, 2H, J ) 8.0), 6.96 (d, 2H, J ) 8.0), 5.1-4.95 (m, 1H),
4.90-4.80 (m, 3H), 4.70-4.55 (m, 2H), 3.60-3.49 (m, 4H), 1.50
(s, 18H).
Gen er a l Meth od s. Gen er a l Meth od 1: Syn th esis of
Ald eh yd es 3a , 3b, a n d 21 (Sch em e 3). General procedure
for synthesis of oxime 16: The aldehyde 15 (10 g, 48 mmol)
was dissolved in dioxane (40 mL). Triethylamine (20 mL) was
added, followed by hydroxylamine hydrochloride (4.0 g, 58
mmol). The reaction mixture was sonicated for 3 h and then
stirred at room temperature for 3 days. The progress of the
reaction was monitored by 1H NMR. The reaction was worked
up by concentration in vacuo to about 50% of the original
volume. Water (60 mL) was added, and the reaction mixture
was extracted with diethyl ether (3 × 40 mL). The combined
organic extracts were then dried (MgSO4) and concentrated
in vacuo. The oxime 16 was obtained, and the crude material
was used in the next reaction (10 g, crude yield, 95%). 1H NMR
(400 MHz, CDCl3): 8.0 (s, 1H), 7.4 (d, 2H, J ) 8), 7.3 (d, 2H,
J ) 8), 5.4 (s, 1H), 3.5 (m, 4H), 1.1 (m, 6H).
The oxime 16 (25 g, 112 mmol) was dissolved in THF (200
mL). tert-Butyl acrylate 17 (29 g, 223 mmol) was added and
the reaction mixture cooled to 0 °C. Bleach (5.25% sodium
hypochlorite aqueous, 400 mL) was added, and the reaction
mixture was allowed to warm to room temperature. When all
of the starting material had been consumed, the reaction was
worked up via addition of ethyl acetate (200 mL), followed by
washing with 10% Na2S2O3 (2 × 50 mL) and brine (1 × 50
mL), dried (Na2SO4), and concentrated in vacuo. The tert-
butyl acrylate was removed by coevaporation with toluene
(monitored by NMR) to give compound 18. The acetal protect-
ing group of 18 was removed by dissolving the isoxazoline
acetal 18 in THF/water (300 mL/50 mL) followed by addition
of acidic Amberlite IR-120 ion-exchange resin (2.0 g). The
reaction mixture was stirred at room temperature for 5 h. The
resin was then removed via filtration and the product extracted
with DCM. The combined organic extracts were dried (MgSO4)
and concentrated in vacuo. The aldehyde 3a was obtained as
a pale yellow crystalline solid, which was recrystallized from
DCM/hexane (22 g, 92% yield). 1H NMR (400 MHz, CDCl3):
10.01 (s, 1H), 7.95 (d, 2H, J ) 8.1), 7.85 (d, 2H, J ) 8.1), 5.1
(t, 1H, J ) 9.6), 3.61 (d, 2H, J ) 9.6), 1.5 (s, 9H).
The aldehyde input 3b was synthesized according to general
method 1, using methyl acrylate in place of tert-butyl acrylate
to give aldehyde 3b (R ) Me) (92%). 1H NMR (300 MHz,
CDCl3): 10.06 (s, 1H), 7.94 (d, 2H, J ) 8.1), 7.86 (d, 2H, J )
8.4), 5.30-5.24 (m, 1H), 3.85 (s, 3H), 3.72-3.67 (m, 2H).
The aldehyde input 21 was synthesized according to general
method 1, using ethyl propiolate 19 in place of tert-butyl
acrylate to give aldehyde 21 as a white solid (1.8 g, 38%). TLC
Rf ) 0.5 (1:1 hexane/ethyl acetate); 1H NMR (300 MHz,
CDCl3): 10.10 (s, 1H), 8.03 (s, 4H), 7.33 (s, 1H), 4.49 (q, 2H, J
) 7.2), 1.46 (t, 3H, J ) 7.2).
Gen er a l Meth od 2: Syn th esis of Wittig Rea gen t 4
(Sch em e 1). 4-Bromobenzyl bromide (10 g, 40 mmol) was
added to triphenyl phosphine (11 g, 42 mmol), in o-xylene (50
mL). The mixture was heated to 150 °C overnight. The Wittig
reagent 4 crystallizes out of solution and is collected by
filtration as a white crystalline solid, which is washed with
hexane and dried. The yield is quantitative. 1H NMR (300
MHz, CDCl3): 7.82-7.71 (m, 9H), 7.65-7.56 (m, 6H), 7.18 (d,
2H, J ) 8.7), 7.06 (d, 2H, J ) 8.6), 5.57 (d, 2H, J ) 15.0).
Gen er a l Meth od 3: Syn th esis of Dion e 8 via a Wittig
r ea ction (Sch em e 1). Wittig reaction to give alkene 5: To
the Wittig reagent 4 (22 g, 43 mmol) in dry DMSO (65 mL)
was added potassium tert-butoxide (5.1 g, 43 mmol), and the
mixture was stirred at room temperature. After 30 min, the
aldehyde 3a (11.4 g, 41 mmol) was added in dry THF (150
mL). The reaction was stirred for 1 h at room temperature
and then quenched by pouring into ice water (100 mL). This
mixture was then extracted with DCM (3 × 100 mL). The
combined DCM extracts were washed with water (1 × 50 mL),
saturated sodium bicarbonate (1 × 50 mL), and brine (1 × 50
mL). The mixture was dried over anhydrous sodium sulfate
and concentrated to dryness. The crude product was purified
by silica gel chromatography (eluting with hexane:ethyl ace-
Oxidation of diol 7 to give dione 8a : The diol 7 (1.0 g,
2.2 mmol) was dissolved in dichloromethane (12 mL). To
this mixture were added 0.7 M NaBr (1.5 mL, 1.0 mmol) and
TEMPO (4.0 mg, 0.025 mmol), and the reaction mixture cooled
to 0 °C. A freshly prepared buffered bleach solution (270 mg,
NaHCO3 dissolved in 16 mL of bleach (5.25% aqueous sodium
hypochlorite) was added dropwise to the reaction mixture. The
reaction mixture was then stirred for a further 15 min before
work up. The reaction was quenched with 10% Na2S2O3
aqueous (30 mL) and extracted with ethyl acetate (3 × 60 mL).
The combined organic layers were then washed with water (1
× 30 mL) and brine (1 × 40 mL), dried (MgSO4), and
concentrated in vacuo, to afford the dione 8a (0.8 g, quantita-
1
tive) as a pale yellow solid. H NMR (400 MHz, CDCl3): 8.01
(d, 2H, J ) 8.5), 7.86 (d, 2H, J ) 8.2), 7.83 (d,1H, J ) 8.0),
7.69 (d, 2H, J ) 8.2), 5.12 (t, 1H, J ) 9.3), 3.62 (d, 2H, J )
9.4), 1.51 (s, 9H).
Aldehyde 3b was used to synthesize the desired dione 8b
as the methyl ester using the methodology outlined for the
synthesis of dione 8a above, and crude was used in the
following reaction.
Heck Rea ction on Dion e 8 To Give Dion e 10 (Sch em e
1). The dione 8a or 8b (1 equiv) was dissolved in DMF (to make
0.14 M solution), followed by addition of Pd(OAc)2 (0.02 equiv),
TEA (3 equiv), P(o-tolyl)3 (0.09 equiv), and acrylamide (1.20
equiv). The reaction mixture was heated to 100 °C for 2 h.
The reaction was then quenched via addition of water and
extracted with methylene chloride. The combined organic
layers were washed with 1 N HCl (aqueous) and water, dried
(Na2SO4), and concentrated in vacuo, to give the desired
derivatized diones 10a and 10b, respectively. The resulting
dione was used for subsequent reactions without further
purification.

2104 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13
Slee et al.
t
Data for 10a (quantitative yield, R1 ) Bu): 1H NMR (300
and extracted with ethyl acetate. The ethyl acetate solution
was washed with water and brine, dried over MgSO4, and
concentrated to dryness. The final acid was recrystallized using
isopropyl alcohol and ethyl acetate.
Gen er a l Meth od 8: Hyd r olyses of a ter t-Bu tyl Ester .
The tert-butyl ester was dissolved in 50% TFA dichloro-
methane solution with ice bath. The reaction mixture was
stirred at 0 °C for ∼1 h. The reaction mixture was then
concentrated in vacuo. The product was precipitated with a
mixture of acetonitrile (few drops) and ether, collected via
filtration, and can be recrystallized from methanol/ethyl
acetate.
MHz, CDCl3): 8.02 (d, 2H, J ) 8.7), 7.97 (d, 2H, J ) 8.1), 7.82
(d, 2H, J ) 9), 7.64-7.59 (m, 3H), 7.32-7.19 (m, 5H), 6.46 (d,
1H, J ) 15.6), 5.77 (t, 1H, J ) 6), 5.15-5.09 (m, 1H), 3.62-
3.58 (m, 2H), 3.48-3.41 (m, 2H), 2.71 (t, 2H, J ) 7.8), 1.98-
1.88 (m, 2H), 1.51 (s, 9H).
Data for 10b (quantitative yield, R1 ) Me): 1H NMR (300
MHz, CDCl3): 8.01 (d, 2H, J ) 6.6), 7.96 (d, 2H, J ) 8.1), 7.81
(d, 2H, J ) 9.0), 7.65-7.52 (m, 3H), 7.35-7.14 (m, 5H), 6.44
(d, 1H, J ) 15.6), 5.79 (t, 1H, J ) 5.0), 5.26 (dd, 1H, J ) 10.5,
7.2), 3.83 (s, 3H), 3.72-3.60 (m, 2H), 3.43 (q, 2H, J ) 6.3),
2.70 (t, 2H, J ) 7.5), 2.00-1.85 (m, 2H).
Gen er a l Meth od 4: Syn th esis of Acr yla m id e 9 (Sch em e
5). Acrylamides 9a -e were prepared by adding acryloyl
chloride (1 equiv) to a cooled solution (0 °C) of the desired
amine 32 (1.0 equiv) in dichloromethane (0.5 M) with tri-
ethylamine (1.0 equiv) as base. These acrylamides were used
directly, without purification in the following Heck reaction.
Gen er a l Meth od 5: Syn th esis of Dion e 25 (Sch em e 4).
To 4,4′-dibromobenzil 22 (5.0 g, 14 mmol, 1.0 equiv) in DMF
(28 mL, 0.5 M) were added Pd(OAc)2 (94 mg, 0.4 mmol, 0.03
equiv), P(o-tolyl)3 (511 mg, 1.7 mmol, 0.12 equiv), TEA (3.9
mL, 28 mmol, 2.0 equiv), and tert-butyl acrylate (2.9 mL, 20
mmol, 1.45 equiv). The reaction was stirred at 100 °C for 1 h.
Afterward, the acrylamide 9a -d (7.5 mmol, 0.55 equiv) was
added and the mixture stirred an additional hour. Upon
completion, the mixture was diluted with ethyl acetate. The
mixture was extracted with ethyl acetate (2 × 200 mL) and
washed with water (2 × 100 mL). The aqueous layer was back
extracted with an additional aliquot of ethyl acetate (250 mL).
The combined organic phase was washed with water (1 × 50
mL), dried (MgSO4), and concentrated in vacuo to obtain a
brown oil. The oil was purified by flash column chromatogra-
phy, eluting with a hexane/ethyl acetate/dichloromethane
mixture, to afford the desired product 25a -d as a light brown
solid (∼39%).
3-[4-(2-[4-((E)-2-Ca r boxy-vin yl)-p h en yl]-5-{4-[(E)-2-(3-
p h en yl-p r op ylca r ba m oyl)-vin yl]-p h en yl}-1H-im id a zol-4-
yl)-p h en yl]-4,5-d ih yd r o-isoxa zole-5-ca r boxylic Acid (13).
Compound 13 was synthesized according to general method 7
from dione 10a (0.73 g, 1.29 mmol) in acetic acid (6 mL),
4-formylcinnamic acid tert-butyl ester (0.36 g, 1.55 mmol), and
NH4OAc (3.0 g, 38.7 mmol). The resulting imidazole was
purified by flash column chromatography eluting with DCM/
MeOH (95:5) to give the protected imidazole 3-[4-(2-[4-((E)-2-
tert-butoxycarbonyl-vinyl)-phenyl]-5-{4-[(E)-2-(3-phenyl-pro-
pylcarbamoyl)-vinyl]-phenyl}-1H-imidazol-4-yl)-phenyl]-4,5-
dihydro-isoxazole-5-carboxylic acid tert-butyl ester 12a as a
yellow solid (0.42 g, 42%). 1H NMR (300 MHz, DMSO-d6): 8.15
(br s, 2H), 7.65-7.40 (br m, 12H), 7.30-7.10 (br m, 5H), 6.40-
6.20 (br m, 3H), 5.04 (t, 1H, J ) 7.5), 3.55 (d, 2H, J ) 7.5),
3.34 (br s, 2H), 2.58 (br s, 2H), 1.84 (br s, 2H), 1.58 (s, 9H),
1.53 (s, 9H).
The bis-tert-butyl ester of 12a was hydrolyzed according to
general method 8 to give, after recrystallization, the desired
imidazole 3-[4-(2-[4-((E)-2-carboxy-vinyl)-phenyl]-5-{4-[(E)-2-
(3-phenyl-propylcarbamoyl)-vinyl]-phenyl}-1H-imidazol-4-yl)-
phenyl]-4,5-dihydro-isoxazole-5-carboxylic acid 13 as a yellow
solid (0.13 g, 36%). TLC: Rf ) 0.38 (chloroform/isopropyl
1
alcohol/formic acid, 80:20:1). H NMR (300 MHz, DMSO-d6):
Data for 25a (R5′ ) H, R5′′ ) C12 25): 1H NMR (400 MHz,
H
8.20-8.14 (m, 3H), 7.86 (d, 2H, J ) 7.8), 7.88 (d, 2H, J ) 8.1),
7.69-7.57 (m, 7H), 7.45 (d, 1H, J ) 15.6), 7.32-7.18 (m, 5H),
6.67 (d, 1H, J ) 15.9), 6.63 (d, 1H, J ) 15.9), 5.22-5.16 (m,
1H), 3.81-3.51 (m, 2H), 3.23-3.17 (m, 2H), 2.65-2.60 (m, 2H),
1.82-1.72 (m, 2H). HPLC-1: tR ) 4.79 min. HPLC-2: tR ) 3.80
min. HRMS: m/z 667.2540 [M + H]+, expected 667.2551 [M
+ H]+.
CDCl3): 7.98 (d, 2H, J ) 6.3), 7.97 (d, 2H, J ) 6.0), 7.61-7.58
(m, 6H), 6.52 (d, 1H, J ) 11.7), 6.49 (d, 1H, J ) 12.0), 5.77 (t,
1H, J ) 4.2), 3.39 (q, 2H, J ) 5.4), 1.70 (brs, 2H), 1.59-1.54
(m, 9H), 1.33-1.30 (m, 18H), 0.88 (t, 3H, J ) 5.1).
Data for 25b (R5′ ) H, R5′′ ) C16H33): 1H NMR (300 MHz,
CDCl3): 7.98 (d, 4H, J ) 7.2), 7.69-57 (m, 6H), 6.51 (d, 1H, J
) 15.9), 6.49 (d, 1H, J ) 15.9), 5.77 (brs, 1H), 3.39 (q, 2H, J )
5.4), 1.82 (brs, 2H), 1.52 (s, 9H), 1.29 (s, 26H), 0.90 (t, 3H, J )
5.1).
3-{4-[5-{4-[(E)-2-(3-P h en yl-p r op ylca r b a m oyl)-vin yl]-
p h en yl}-2-(4-p yr r olid in -1-yl-p h en yl)-1H -im id a zol-4-yl]-
ph en yl}-4,5-dih ydr o-isoxazole-5-car boxylic Acid (14). Com-
pound 14 was synthesized according to general method 6 from
dione 10b (0.2 g, 0.35 mmol) in acetic acid (2.0 mL), with
4-pyrrolidin-1-yl-benzaldehyde (0.1 g, 0.38 mmol) and
NH4OAc (0.8 g, 11 mmol), to give 3-{4-[5-{4-[(E)-2-(3-phenyl-
propylcarbamoyl)-vinyl]-phenyl}-2-(4-pyrrolidin-1-yl-phenyl)-
1H-imidazol-4-yl]-phenyl}-4,5-dihydro-isoxazole-5-carboxylic acid
methyl ester. The methyl ester 12b was hydrolyzed according
to general method 7 to give, after recrystallization, the desired
imidazole 3-{4-[5-{4-[(E)-2-(3-phenyl-propylcarbamoyl)-vinyl]-
phenyl}-2-(4-pyrrolidin-1-yl-phenyl)-1H-imidazol-4-yl]-phenyl}-
4,5-dihydro-isoxazole-5-carboxylic acid 14 as a yellow solid
(0.08 g, 33%). TLC: Rf ) 0.25 (chloroform/isopropyl alcohol/
formic acid, 80:20:1). 1H NMR (300 MHz, DMSO): 8.16 (t, 1H,
J ) 5.4), 7.90 (d, 2H, J ) 9.0), 7.70-7.56 (m, 8H), 7.43 (d, 1H,
J ) 15.9), 7.32-7.18 (m, 5H), 6.67-6.61 (m, 3H), 5.17 (dd, 1H,
J ) 11.4, J ) 6.9), 3.79-3.55 (m, 2H), 3.35 (br s, 4H), 3.23-
3.16 (m, 2H), 2.62 (t, 2H, J ) 7.8), 1.98 (br s, 4H), 1.81-1.72
(m, 2H). MS (ESI): m/z 666.7 (100, [M + H]), calcd C41H40N5O4
([M + H]) 666.3. HPLC-1: tR ) 5.64 min. HPLC-2: tR ) 5.7
min. HRMS: m/z 666.3095 [M + H]+, expected 666.3075 [M
+ H]+.
Data for 25c (R5′ ) H, R5′′ ) PhC7H15): 1H NMR (300 MHz,
CDCl3): 8.44 (s, 1H), 7.97 (d, 2H, J ) 8.7), 7.90 (d, 2H, J )
8.7), 7.71 (d, 1H, J ) 15.9), 7.66-7.48 (m, 7H), 7.12 (d, 2H, J
) 8.7), 6.77 (d, 1H, J ) 15.9), 6.50 (d, 1H, J ) 15.9), 2.50 (m,
2H), 1.50 (br s, 11H), 1.30 (br s,4H), 0.88 (t, 3H, J ) 6.9).
Data for 25d (R5′ ) R5′′ ) C6H13): 1H NMR (300 MHz,
CDCl3): 7.98 (d, 4H, J ) 8.1), 7.70 (d, 1H, J ) 15.3), 7.63 (d,
4H, J ) 8.4), 7.59 (d, 1H, J ) 16.0), 6.95 (d, 1H, J ) 15.3),
6.48 (d, 1H, J ) 16.2), 3.40 (br q, 4H, J ) 8.1), 1.68-1.50 (m,
4H), 1.4-1.24 (m, 12H), 0.96-0.82 (m, 6H).
Gen er a l Meth od 6: Syn th esis of Im id a zole (Sch em es
2 a n d 4). Acetic acid (20 mL) was added to a mixture of the
dione 10 or 25 (1.0 equiv), aldehyde (1.5 equiv), and NH4OAc
(30 equiv) and heated to 100 °C for ∼2 h. The reaction has to
be monitored carefully if tert-butyl groups are present, as these
will be removed with prolonged heating. The reaction mixture
was extracted with ethyl acetate, washed with water, and then
back extracted with ethyl acetate. The organic layers were
combined, dried (MgSO4), and concentrated in vacuo. The
imidazole was purified by flash column chromatography,
eluting with hexane/ethyl acetate (3:1). The compound fluo-
resces as a yellow spot on TLC under a long-wave UV lamp.
The desired imidazole is obtained as a yellow or white solid.
Gen er a l Meth od 7: Hyd r olyses of a Meth yl or Eth yl
Ester (Also Rem oves ter t-Bu tyl Ester s). A mixture of
appropriate ester (1.0 equiv), 1 N LiOH (15 equiv), and 1,4-
dioxane (0.3 M of ethyl ester) was stirred at room temperature
overnight. The reaction mixture was acidified with 1 N HCl
(E)-3-(4-{5-[4-((E)-2-Hexa d ecylca r ba m oyl-vin yl)-p h en -
yl]-1H-im id a zol-4-yl}-p h en yl)-a cr ylic Acid (1). Imidazole
1 was synthesized from dione 25b (1.5 g, 2.38 mmol, 1.0 equiv)
in acetic acid (14 mL) and DMSO (4 mL), with hexamethyl-
enetetramine (1.67 g, 11.9 mmol, 5 equiv) and NH4OAc (5.50
g, 71.4 mmol, 30 equiv). The resulting protected imidazole was
purified by flash column chromatography eluting with a

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2105
gradient of 2-8% methanol in DCM to give (E)-3-(4-{5-[4-((E)-
2-hexadecylcarbamoyl-vinyl)-phenyl]-1H-imidazol-4-yl}-phen-
yl)-acrylic acid tert-butyl ester 26 as a yellow solid (1.4 g, 92%).
1H NMR (400 MHz, CDCl3): 7.59 (s, 1H), 7.50-7.22 (m, 10H),
6.64 (br, s, 1H), 6.38 (d, 1H, J ) 15.2), 6.28 (d, 1H, J ) 16.0),
3.36-3.29 (m, 2H), 1.52-1.46 (m, 2H), 1.51 (s, 9H), 1.23 (br,
s, 26H), 0.86 (t, 3H, J ) 6.6).
The tert-butyl esters were removed according to general
method 8 to give, after recrystallization from methanol/ethyl
acetate, the desired imidazole (E)-3-(4-{5-[4-((E)-2-hexadecyl-
carbamoyl-vinyl)-phenyl]-1H-imidazol-4-yl}-phenyl)-acrylic acid
1 as a pale yellow solid (0.77 g, 60%). TLC: Rf ) 0.13
(chloroform/isopropyl alcohol/formic acid, 80:20:1). 1H NMR
(300 MHz, DMSO-d6): 8.49 (s, 1H), 8.11 (t, 1H, J ) 5.6), 7.73
(d, 2H, J ) 8.1), 7.60 (d, 2H, J ) 8.4), 7.59 (d, 1H, J ) 15.9),
7.52 (d, 2H, J ) 8.4), 7.51 (d, 2H, J ) 8.1), 7.42 (d, 1H, J )
15.6), 6.62 (d, 1H, J ) 15.9), 6.56 (d, 1H, J ) 16.2), 3.20-3.13
(m, 2H), 1.50-1.40 (m, 2H), 1.22 (br, s, 26H), 0.84 (t, 3H, J )
6.3). HPLC-1: tR ) 7.54 min. HPLC-2: tR ) 13.2 min. MS
(ESI): m/z 584.8 (100, [M + H]+), calcd C37H49N3O3 ([M + H]+)
584.8. Anal. (C37H49N3O3) C, H, N.
3-(4-{4-[4-((E)-2-Ca r b oxy-vin yl)-p h en yl]-5-[4-(2-h exa -
decylcar bam oyl-vin yl)-ph en yl]-1H-im idazol-2-yl}-ph en yl)-
4,5-d ih yd r o-isoxa zole-5-ca r boxylic Acid (29). Compound
29 was synthesized according to general method 6 from dione
25b (0.5 g, 0.79 mmol) in acetic acid (5.5 mL), with 4-formyl-
phenyl-4,5-dihydro-isoxazole-5-carboxylic acid tert-butyl ester
3a (0.26 g, 0.95 mmol) and NH4OAc (1.8 g, 23.8 mmol). The
resulting imidazole was purified by flash column chromatog-
raphy, eluting with hexane/ethyl acetate (3:1), to give 3-(4-{4-
[4-((E)-2-tert-butoxycarbonyl-vinyl)-phenyl]-5-[4-(2-hexadecyl-
carbamoyl-vinyl)-phenyl]-1H-imidazol-2-yl}-phenyl)-4,5-dihydro-
isoxazole-5-carboxylic acid tert-butyl ester 26 as a yellow solid
(0.5 g, 72%). 1H NMR (300 MHz, CDCl3): 8.01 (br, m, 2H),
7.70-7.20 (br, m, 10H), 6.40-6.10 (br, m, 3H), 5.10 (t, 1H, J
) 9.3), 3.60 (d, 2H, J ) 9.3), 3.30 (br, s, 2H), 1.58 (s, 9H), 1.56
(s, 9H), 1.57 (br, s, 2H), 1.30 (br, s, 26H), 0.85 (t, 3H, J ) 7.5).
imidazol-2-yl}-phenyl)-isoxazole-5-carboxylic acid 30 (217 mg,
75%) as a yellow solid. TLC: Rf ) 0.3 (20% IPA/DCM, 1%
1
formic acid). H NMR (300 MHz, DMSO-d6): 8.31 (d, 2H, J )
8.1), 8.19 (d, 2H, J ) 8.4), 8.12 (t, 1H, J ) 5.1), 7.92 (s, 1H),
7.78 (d, 2H, J ) 8.1), 7.66-7.59 (m, 7H), 7.44 (d, 1H, J ) 15.9),
6.67 (d, 1H, J ) 15.6), 6.59 (d, 1H, J ) 15.9), 3.16 (q, 2H, J )
6.0), 1.45 (t, 2H, J ) 6.0), 1.23 (s, 26H), 0.84 (t, 3H, J ) 5.7).
LC/MS: LC retention time ) 4.18 min. MS (APcI): m/z 771.0
(100, [M + H]+), calcd C47H55N4O6 ([M + H]+) 771.4. Anal.
(C47H54N4O6‚H2O) C, H, N.
3-(4-{4-[4-((E )-2-Ca r b oxy-vin yl)-p h e n yl]-5-[4-(2-d o-
d ecylca r ba m oyl-vin yl)-p h en yl]-1H-im id a zol-2-yl}-p h en -
yl)-isoxa zole-5-ca r boxylic Acid (31). Imidazole 31 was
synthesized according to general method 6 from dione 25a (300
mg, 0.52 mmol) in acetic acid (2 mL), with 3-(4-formyl-phenyl)-
isoxazole-5-carboxylic acid ethyl ester 21 (synthesized accord-
ing to general method 1 using the appropriate alkyne) (192
mg, 0.78 mmol) and NH4OAc (1.2 g, 15.6 mmol), which gave,
after purification via column chromatography eluting with
DCM:methanol (95:5), 3-(4-{4-[4-((E)-2-tert-butoxycarbonyl-
vinyl)-phenyl]-5-[4-(2-dodecylcarbamoyl-vinyl)-phenyl]-1H-
imidazol-2-yl}-phenyl)-isoxazole-5-carboxylic acid ethyl ester
26 (200 mg, 48%).
The ethyl and tert-butyl esters were hydrolyzed according
to general method 7 to give, after recrystallization from
methanol/ethyl acetate, the desired imidazole 3-(4-{4-[4-((E)-
2-carboxy-vinyl)-phenyl]-5-[4-(2-dodecylcarbamoyl-vinyl)-phen-
yl]-1H-imidazol-2-yl}-phenyl)-isoxazole-5-carboxylic acid 31 (35
mg, 55%) as a yellow solid. TLC: Rf ) 0.3 (20% IPA/DCM, 1%
1
formic acid). H NMR (300 MHz, DMSO-d6): 8.24 (d, 2H, J )
8.4), 8.14 (t, 1H, J ) 4.5), 8.08 (d, 2H, J ) 8.4), 7.83 (s, 1H),
7.72 (d, 2H, J ) 8.1), 7.63-7.59 (m, 7H), 7.42 (d, 1H, J ) 15.6),
6.64 (d, 1H, J ) 15.6), 6.54 (d, 1H, J ) 15.9), 3.15 (t, 2H, J )
4.5), 1.44 (t, 2H, J ) 5.7), 1.23 (s, 18H), 0.83 (t, 3H, J ) 6.3).
LC/MS: LC retention time ) 3.72 min. MS (APcI): 715.1 (100,
[M + H]+), calcd C43H47N4O6 ([M + H]+) 715.9. Anal.
(C43H46N4O6‚2H2O) C, H, N.
The tert-butyl esters were removed according to general
method 8 to give, after recrystallization from methanol/
ethyl acetate, the desired imidazole 3-(4-{4-[4-((E)-2-carboxy-
vinyl)-phenyl]-5-[4-(2-hexadecylcarbamoyl-vinyl)-phenyl]-1H-
imidazol-2-yl}-phenyl)-4,5-dihydro-isoxazole-5-carboxylic acid
29 as a pale yellow solid (0.3 g, 69%). TLC: Rf ) 0.19
(chloroform/isopropyl alcohol/formic acid, 80:20:1). 1H NMR
(300 MHz, DMSO-d6): 8.19 (d, 2H, J ) 8.7), 8.10 (t, 1H, J )
5.4), 7.86 (d, 2H, J ) 8.1), 7.75 (d, 2H, J ) 8.4), 7.64-7.58 (m,
7H), 7.43 (d, 1H, J ) 15.6), 6.65 (d, 1H, J ) 15.9), 6.57 (d, 1H,
J ) 15.9), 5.25-5.19 (m, 1H), 3.84-3.61 (m, 2H), 3.19-3.13
(m, 2H), 1.50-1.40 (br, m, 2H), 1.22 (br, s, 25H), 0.84 (t, 3H,
J ) 6.60). MS (ESI): m/z 773.8 (30, [M + H]+), calcd for
(E)-3-[4-(5-{4-[(E)-2-(4-Heptyl-ph en ylcar bam oyl)-vin yl]-
ph en yl}-1H-im idazol-4-yl)-ph en yl]-acr ylic Acid (27). Imid-
azole 27 was synthesized according to general method 6 from
dione 25c (300 mg, 0.52 mmol) in acetic acid (6 mL), with
hexamethylene tetramine (360 mg, 2.58 mmol) and NH4OAc
(1.19 g, 15.5 mmol), to give, after purification via column
chromatography eluting with DCM:methanol (95:5), (E)-3-
[4-(5-{4-[(E)-2-(4-heptyl-phenylcarbamoyl)-vinyl]-phenyl}-1H-
imidazol-4-yl)-phenyl]-acrylic acid tert-butyl ester 26 (110 mg,
36%). 1H NMR (300 MHz, CDCl3): 8.96 (br, s, 1H), 7.77 (br, s,
1H), 7.61 (d, 2H, J ) 7.8), 7.61-7.20 (m, 10H), 7.09 (d, 2H, J
) 8.0), 6.67 (d, 1H, J ) 15.3), 6.27 (d, 1H, J ) 15.9), 2.54 (br,
t, 2H, J ) 7.2), 1.60-1.48 (m, 2H), 1.51 (s, 9H), 1.34-1.20 (m,
8H), 0.87 (t, 3H, J ) 6.6).
C
47H56N4O6 [M + H]+ 773.4. Anal. (C47H56N4O6‚H2O) C, H, N.
3-(4-{4-[4-((E)-2-Ca r b oxy-vin yl)-p h en yl]-5-[4-(2-h exa -
The tert-butyl ester was removed according to general
method 8 to give, after recrystallization from methanol/ethyl
acetate, the desired imidazole (E)-3-[4-(5-{4-[(E)-2-(4-heptyl-
phenylcarbamoyl)-vinyl]-phenyl}-1H-imidazol-4-yl)-phenyl]-
acrylic acid 27 (31 mg, 28%) as a yellow solid. TLC: Rf ) 0.21
(chloroform/isopropyl alcohol/formic acid, 80:20:1). 1H NMR
(300 MHz, DMSO-d6): 7.86 (s, 1H), 7.68 (br, d, 4H, J ) 6.3),
7.61-7.44 (m, 6H), 7.11 (d, 1H, J ) 15.0), 6.52 (d, 1H, J )
15.9), 3.45 (t, 4H, J ) 7.2), 1.51 (br, m, 4H), 1.27 (br, s, 12H),
d ecylca r ba m oyl-vin yl)-p h en yl]-1H-im id a zol-2-yl}-p h en -
yl)-isoxa zole-5-ca r boxylic Acid (30). Imidazole 30 was
synthesized according to general method 6 from dione 25b (500
mg, 0.79 mmol) in acetic acid (4 mL), with 3-(4-formyl-phenyl)-
isoxazole-5-carboxylic acid ethyl ester 21 (synthesized accord-
ing to general method 1 using the appropriate alkyne) (290
mg, 1.2 mmol) and NH4OAc (1.8 g, 23.7 mmol), which gave,
after purification via column chromatography eluting with
hexane/ethyl acetate/dichloromethane (2:1:1), 3-(4-{4-[4-((E)-
2-tert-butoxycarbonyl-vinyl)-phenyl]-5-[4-(2-hexadecylcarbam-
oyl-vinyl)-phenyl]-1H-imidazol-2-yl}-phenyl)-isoxazole-5-car-
boxylic acid ethyl ester 26 (377 mg, 56%). TLC: Rf ) 0.3 (1:1
0.86 (t, 6H, J ) 7.5). HPLC-1: tR ) 5.92 min. HPLC-2: tR
)
7.15 min. HRMS: m/z 528.3209 ([M + H]+), expected 528.3221.
(E)-3-(4-{5-[4-((E)-2-Dih exylca r ba m oyl-vin yl)-p h en yl]-
1H-im id a zol-4-yl}-p h en yl)-a cr ylic Acid (28). Imidazole 28
was synthesized according to general method 6 from dione 25d
(410 mg, 0.71 mmol) in acetic acid (5 mL), with hexamethyl-
enetetramine (1.05 g, 21.4 mmol) and NH4OAc (1.97 g, 26
mmol), to give, after purification via column chromatography
eluting with DCM:methanol (95:5), (E)-3-(4-{5-[4-((E)-2-di-
hexylcarbamoyl-vinyl)-phenyl]-1H-imidazol-4-yl}-phenyl)-
acrylic acid tert-butyl ester 26 (280 mg, 68%). 1H NMR (400
MHz, CDCl3): 8.06 (s, 1H), 7.56-7.50 (m, 6H), 7.45 (d, 2H, J
) 8.0), 7.42 (d, 2H, J ) 8.0), 6.82 (d, 1H, J ) 16.0), 6.34 (d,
1
H/EtOAc). H NMR (300 MHz, CDCl3): 8.44 (d, 2H, J ) 8.1),
8.11 (d, 2H, J ) 8.4), 7.80-7.70 (m, 5H), 7.63 (d, 2H, J ) 8.4),
7.53 (d, 3H, J ) 7.8), 7.48 (s, 1H), 6.50 (d, 1H, J ) 15.9), 6.48
(d, 1H, J ) 15.9), 6.02 (brs, 1H), 4.67 (q, 2H, J ) 7.2), 3.52 (q,
2H, J ) 6.6), 1.86 (brs, 2H), 1.73 (s, 9H), 1.64 (t, 3H, J ) 7.2),
1.44-1.41 (m, 26H), 1.07 (t, 3H, J ) 6.6).
The tert-butyl and ethyl esters were removed according to
general method 7 to give, after recrystallization from methanol/
ethyl acetate, the desired imidazole 3-(4-{4-[4-((E)-2-carboxy-
vinyl)-phenyl]-5-[4-(2-hexadecylcarbamoyl-vinyl)-phenyl]-1H-

2106 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13
Slee et al.
(14) Todderud, G.; Nair X.; Lee D.; Alford J .; Davern L.; Stanley P.;
Bachand C.; Lapointe P.; Marinier A.; Martel A.; Menard M. W.
J .; Bajorath J .; Hollenbaugh D.; Aruffo A.; Tramposch K. M.
BMS-190394, a Selectin Inhibitor, Prevents Rat Cutaneous
Inflammatory Reactions. J . Pharmacol. Exp. Ther. 1997, 282,
1298-1304.
(15) Ohnishi, M.; Koike, H.; Kawamura, N.; Tojo, S. J .; Hayashi, M.;
and Morooka, S. Role of P-selectin in the Early Stage of the
Arthus Reaction. Immunopharmacology 1996, 34, 161-170.
(16) Veale, D. J .; Maple, C. Cell Adhesion Molecules in Rheumatoid
Arthritis. Drugs Aging 1996, 9, 87-92.
(17) Walter, U.; Issekutz, A. C. The Role of E- and P-selectin in
Neutrophil and Monocyte Migration in Adjuvant-Induced Ar-
thritis in the Rat. Eur. J . Immunol. 1997, 27, 1498-1505.
(18) Ertenli, I.; Kiraz, S.; Arici, M.; Haznedaroglu, I. C.; Calguneri,
M.; Celik, I.; Kirazli, S. P-Selectin as a Circulating Molecular
Marker in Rheumatoid Arthritis with Thrombocytosis. J . Rheu-
matol. 1998, 25, 1054-1058.
1H, J ) 16), 3.34-3.36 (m, 4H), 1.66-1.56 (m, 4H), 1.62 (s,
9H), 1.32 (br, s, 12H), 0.87 (t, 6H, J ) 6.8).
The t-butyl ester was removed according to general method
8 to give, after recrystallization from methanol/ethyl acetate,
the desired imidazole (E)-3-(4-{5-[4-((E)-2-dihexylcarbamoyl-
vinyl)-phenyl]-1H-imidazol-4-yl}-phenyl)-acrylic acid 28, (50
mg, 18%) as a yellow solid. TLC: Rf ) 0.15 (chloroform/
isopropyl alcohol/formic acid, 80:20:1). 1H NMR (400 MHz,
DMSO-d6): 12.40 (br, s, 1H), 10.12 (s, 1H), 7.95 (s, 1H), 7.69
(d, 2H, J ) 8.0), 7.63-7.51 (m, 10H), 7.14 (d, 2H, J ) 8.4),
6.82 (d, 1H, J ) 16.0), 6.53 (d, 1H, J ) 16.0), 2.52 (t, 2H, J )
8.0), 1.58-1.50 (br, m, 2H), 1.30-1.22 (br, m, 8H), 0.85 (t, 3H,
J ) 6.8). HPLC-1: tR ) 6.16 min. HPLC-2: tR ) 8.38 min.
HRMS m/z 534.274 ([M + H]+), expected 534.2751.
(19) Tsukida, T.; Moriyama, H.; Kurokawa, K.; Achiha, T.; Inoue, Y.;
Kondo, H. Studies on selectin blockers. 7. Structure-activity
relationships of sialyl Lewis X mimetics based on modified Ser-
Glu dipeptides. J . Med. Chem. 1998, 41, 4279-4287.
(20) Tsujishita, H.; Kiramatsu, Y.; Kondo, N.; Ohmoto, H.; Kondo,
H.; Kiso, M.; Hasegawa, A. Selectin-ligand Interactions Revealed
by Molecular Dynamics Simulation in Solution. J . Med. Chem.
1997, 40, 362-369.
(21) Kogan, T.; Dupre, B.; Scott, I. L.; Bui, H.; Wheeler, K. L.; Keller,
K. M.; Kassir, J . M. Di- and Trivalent Small Molecule Selectin
Inhibitors. (Texas Biotechnology Corporation, WO9701335, 1997).
(22) Renkonen, O.; et al. Synthesis of a New Nanomolar Saccharide
Inhibitor of Lymphocyte Adhesion: Different Polylactosamine
Backbones Present Multiple Sialyl Lewis x Determinants to
L-Selectin in High-Affinity Mode. Glycobiology 1997, 7, 453-
461.
(23) Ikeda, T.; Kajimoto, T.; Kondo, H.; Wong, C.-H. Design and
Synthesis of an alpha-Mannosyl Terpenoid as Selective Inhibitor
of P-Selectin. Bioorg. Med. Chem. Lett. 1997, 7, 2485-2490.
(24) Marron, T. G.; Woltering, T. J .; Weitz-Schmidt, G.; Wong, C.-H.
C-Mannose Derivatives as Potent Mimics of Sialyl Lewis X.
Tetrahedron Lett. 1996, 37, 9037-9040.
(25) Marinier, A.; et al. Sulfated Galactocerebrosides as Potential
Antiinflammatory Agents. J . Med. Chem. 1997, 40, 3234-3247.
(26) Briggs, J . B.; Oda, Y.; Gilbert, J . H.; Schaefer, M. E.; Macher,
B. A. Peptides Inhibit Selectin-Mediated Cell Adhesion In Vitro,
and Neutrophil Influx into Inflammatory Sites In Vivo. Glyco-
biology 1995, 5, 583-588.
(27) Martens, C. L. et al. Peptides Which Bind to E-Selectin and Block
Neutrophil Adhesion. J . Biol. Chem. 1995, 270, 21129-21136.
(28) Kuribayashi, T.; Ohkawa, N.; Satoh, S. Aryl C-Glycosides:
Physiologically Stable Glycomimetics of Sialyl Lewis X. Bioorg.
Med. Chem. Lett. 1998, 8, 3307-3310.
(29) Hiramatsu, Y.; Tsukida, T.; Nakai, Y.; Inoue, Y.; Kondo, H. Study
on Selectin Blocker 8. Lead Discovery of a Non-Sugar Antagonist
Using a 3D-Pharmacophore Model. J . Med. Chem. 2000, 43,
1476-1483.
(30) Ohmoto, H.; Nakamura, K.; Inoue, T.; Kondo, N.; Inoue, Y.;
Yoshino, K.; Kondo, H.; Ishida, H.; Kiso, M.; Hasegawa, A.
Studies on Selectin Blocker 1. Structure-Activity Relationships
of Sialyl Lewis X Analogues. J . Med. Chem. 1996, 39, 1339-
1343.
(31) Frenette, P.; Wagner, D. Insights into Selectin Function from
Knockout Mice. Thromb. Haemostasis 1997, 78, 60-64.
(32) Hickey, M. J .; Kanwar, S.; McCafferty, D.-M.; Granger, D. N.;
Eppihimer, M. J .; Kubes, P. Varying Roles of E-Selectin and
P-Selectin in Different Microvascular Beds in Response to
Antigen. J . Immunol. 1999, 162, 1137-1143.
Ack n ow led gm en t. We would like to graciously
acknowledge the very generous collegial interaction we
had with numerous Nippon Organon K. K. scientists
and research managers (formerly part of Kanebo, Ltd.)
with whom we collaborated during the first few years
of this research program. That includes Drs. Ikeda,
Inoue, Kondo, Ohashi, Ohmoto, Saito, Yoshino, and
Hirota. The contributions of Drs. Coats and Pallai
(current Ontogen staff) and Drs. Mjalli and Moran
(former Ontogen staff) are also sincerely acknowledged.
Dr. Geoff Kansas (Northwestern Medical School, Chi-
cago, IL) kindly provided the P-selectin transfected CHO
cells.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 1, 13, 14, and 27-31. This material is available
free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Albelda, S. M.; Smith, C. W.; Ward, P. A. Adhesion Molecules
and Inflammatory Injury. FASEB J . 1994, 8, 504-512.
(2) Tedder, T. F.; Steeber, D. A.; Chen, A.; Engel, P. The Selectins:
Vascular Adhesion Molecules. FASEB J . 1995, 9, 866-873.
(3) Lasky, L. A. Selectins: Interpreters of Cell-Specific Carbohydrate
Information During Inflammation. Science 1992, 258, 964-969.
(4) McEver, R. Selectins. Curr. Opin. Immunol. 1994, 6, 75-84.
(5) McEver, R. P.; Moore, K. L.; Cummings, R. D. Leukocyte
Trafficking Mediated by Selectin-Carbohydrate Interactions. J .
Biol. Chem. 1995, 270, 11025-11028.
(6) Lasky, L. A.; Presta, L. G.; Erbe, D. V. Structure-Function
Aspects of Selectin-Carbohydrate Interactions. In Cellular
Ahesion: Molecular Definition to Therapeutic Potential; Metcalf,
B. W., Dalton, B. J ., Poste, G., Eds.; Plenum Press: New York,
1994; pp 37-53.
(7) Rosen, S. D.; Bertozzi, C. R. The Selectins and Their Ligands.
Curr. Opin. Cell Biol. 1994, 6, 663-673.
(8) Bertozzi, C. Cracking the Carbohydrate Code for Selectin
Recognition. Chem. Biol. 1995, 2, 703-708.
(9) Gundel, R. H.; Wegner, C. D.; Torcellini, C. A.; Clarke, C. C.;
Haynes, N.; Rothlein, R.; Smith, C. W.; Letts, L. G. Endothelial
Leukocyte Adhesion Molecule-1 Mediates Antigen-Induced Acute
Airway Inflammation and Late-phase Obstruction in Monkeys.
J . Clin. Invest. 1991, 88, 1407-1411.
(10) De Sanctis, G. T.; Wolyniec, W. W.; Green, F. H. Y.; Qin, S.; J iao,
A.; Finn, P. W.; Noonan, T.; J oetham, A. A.; Gelfland, E.;
Doerschuk, C. M.; Drazen, J . M. Reduction of Allergic Airway
Responses in P-Selectin-Deficient Mice. J . Appl. Physiol. 1997,
83, 681-687.
(11) Kogan, T. P.; Dupre, B.; Bui, H.; McAbee, K. L.; Kassir, J . M.;
Scott, I. L.; Hu, X.; Vanderslice, P.; Beck, P. J .; Dixon, R. A. F.
Novel Synthetic Inhibitors of Selectin-Mediated Cell Adhesion:
Synthesis of 1,6-Bis[3-(3-carboxymethylphenyl)-4-(2-R-D-manno-
pyranosyloxy)phenyl]hexane (TBC1269). J . Med. Chem. 1998,
41, 1099-1111.
(12) Teixeira, M. M.; Hellewell, P. G. Contribution of Endothelial
Selectins and R4 Integrins to Eosinophil Trafficking in Allergic
and Nonallergic Inflammatory Reactions to Skin. J . Immunol.
1998, 161, 2516-2523.
(13) Staite, N. D.; J usten, J . M.; Sly, L. M.; Beaudet, A. L.; Bullard,
D. C. Inhibition of Delayed-Type Contact Hypersensitivity in
Mice Deficient in Both E-Selectin and P-Selectin. Blood 1996,
88, 2973-2979.
(33) Brackeen, M.; Stafford, J .; Feldman, P.; Karanewsky, D. An
efficient and mild synthesis of highly substituted imidazoles.
Tetrahedron Lett. 1994, 35, 1635-1638.
(34) Cheng, J .; Mjalli, A. Solid-Phase Synthesis of ∆-Isoxazolines.
Tetrahedron Lett. 1998, 39, 939-942.
(35) Graves, B. J .; Crowther, R. L.; Chandran, C.; Rumberger, J . M.;
Li, S.; Huang, K. S.; Presky, D. H.; Familletti, P. C.; Wolitzky,
B. A.; Burns, D. K. Insight into E-selectin/Ligand Interaction
from the Crystal Structure and Mutagenesis of the Lec/EGF
Domains. Nature 1994, 367, 532-538.
(36) Huang, H.; Wong, C.-H. Synthesis of Biologically Active Sialyl
Lewis X Mimetics. J . Org. Chem. 1995, 60, 3100-3106.
(37) Uchiyama, T.; Vassilev, V.; Kajimoto, T.; Wong, W.; Huang, H.;
Lin, C.-C.; Wong, C.-H. Design and Synthesis of Sialyl Lewis X
Mimetics. J . Am. Chem. Soc. 1995, 117, 5395-5396.
(38) Ramphal, J .; Zheng, Z.; Perez, C.; Walker, L.; DeFrees, S.; Gaeta,
F. Structure-Activity Relationships of Sialyl Lewis X-Contain-
ing Oligosaccharides. 1. Effect of Modifications of the Fucose
Moiety. J . Med. Chem. 1994, 37, 3459-3463.
(39) Nelson, R.; Dolich, S.; Aruffo, A.; Cecconi, O.; Belivacqua, M.
Higher-Affinity Oligosaccharide Ligands for E-Selectin. J . Clin.
Invest. 1993, 91, 1157-1166.

Non-Carbohydrate Imidazole-Based Selectin Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 13 2107
(40) Musser, J .; Rao, N.; Nashed, M.; Dasgupta, F.; Abbas, S.;
Nematalla, A.; Date, V.; Foxall, C.; Asa, D.; J ames, P.; Tyrell,
D.; Brandley, B. Structure-Activity Studies Based on the Sialyl
Lewis X Epitope. In Pharmacochemistry Library: Trends in
Drug Research, Vol. 20; Claasen, V., Ed.; Elsevier: New York,
1993; pp 33-40.
(41) Manning, D.; Bertozzi, C.; Pohl, N.; Rosen, S.; Kiessling, L.
Selectin-Saccharide Interactions: Revealing Structure-Function
Relationships with Chemical Synthesis. J . Org. Chem. 1995, 60,
6254-6255.
(42) Thoma, G.; Schwarzenbach, F.; Duthaler, R. Synthesis of a Sialyl
Lewis x Mimic with Fixed Carboxylic Acid Group: Chemical
Approach toward the Elucidation of the Bioactive Conformation
of Sialyl Lewis x. J . Org. Chem. 1996, 61, 514-624.
(43) Stahl, W.; Sprengard, U.; Kretzschmar, G.; Kunz, H. Synthesis
of Deoxy-Sialyl Lewis X Analogs, Potential Selectin Antagonists.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2096-2098.
(44) Ramphal, J .; Hiroshige, M.; Lou, B.; Gaudino, J .; Hayashi, M.;
Chen, S.; Chiang, L.; Gaeta, F.; DeFrees, S. Ligand Interactions
with E-selectin. Identification of a New Binding Site for Recogni-
tion of N-Acyl Aromatic Glucosamine Substituents of Sialyl
Lewis X. J . Med. Chem. 1996, 39, 1357-1360.
(45) Bamford, M.; Bird, M.; Gore, P.; Holmes, D.; Priest, R.; Prodger,
J .; Saez, V. Synthesis and Biological Activity of Conformationally
Constrained Sialyl Lewis X Analogues with Reduced Carbohy-
drate Character. Bioorg. Med. Chem. Lett. 1996, 6, 239-244.
(46) Sprengard, U.; Kunz, H.; Huls, C.; Schmidt, W.; Seiffge, D.;
Kretzschmar, G. Synthesis and Biological Activity of Novel
Sialyl-Lewis X Conjugates. Bioorg. Med. Chem. Lett. 1996, 6,
509-514.
(47) Erbe, D.; Wolitzky, B.; Presta, L.; Norton, C.; Ramos, R.; Burns,
D.; Rumberger, J .; Rao, B.; Foxall, C.; et al. Identification of an
E-selectin Region Critical for Carbohydrate Recognition and Cell
Adhesion. J . Cell Biol. 1992, 119, 215-227.
(48) Kogan, T.; Revelle, B.; Tapp, S.; Scott, D.; Beck, P. A Single
Amino Acid Residue Can Determine the Ligand Specificity of
E-Selectin. J . Biol. Chem. 1995, 270, 14047-14055.
(59) Hensley, P.; McDevitt, P.; Brooks, I.; Trill, J .; Feild, J .; McNulty,
D.; Connor, J .; Griswold, D.; Kumar, N.; Kopple, K.; et al. The
Soluble Form of E-Selectin is an Asymmetric Monomer. Expres-
sion, Purification, and Characterization of the Recombinant
Protein. J . Biol. Chem. 1994, 269, 23949-23958.
(60) Poppe, L.; Brown, G.; Philo, J .; Nikrad, P.; Shah, B. Conforma-
tion of sLeX Tetrasaccharide, Free in Solution and Bound to E-,
P-, and L-Selectin. J . Am. Chem. Soc. 1997, 119, 1727.
(61) Ng, K.; Drickamer, K.; Weis, W. Structural Analysis of Mono-
saccharide Recognition by Rat Liver Mannose-Binding Protein.
J . Biol. Chem. 1996, 271, 663-674.
(62) Weis, W.; Drickamer, K.; Hendrickson, W. Structure of a C-Type
Mannose-Binding Protein Complexed with an Oligosaccharide.
Nature 1992, 360, 127-134.
(63) Torgersen, D.; Mullin, N.; Drickamer, K. Mechanism of Ligand
Binding to E- and P-selectin Analyzed Using Selectin/Mannose-
binding Protein Chimeras. J . Biol. Chem. 1998, 273, 6254-6261.
(64) Immediately before submission of this manuscript, the X-ray
crystal structure of P- and E-selectin bound to sLex was
published: Somers, W.; Tang, J .; Shaw, G.; Camphausen, R.
Insights into the Molecular Basis of Leukocyte Tethering and
Rolling Revealed by Structures of P- and E-Selectin Bound to
SLeX and PSGL-1. Cell 2000, 103, 467-479.
(65) Luty, B. A.; Wasserman, Z. R.; Stouten, P. F. W.; Hodge, C. N.;
Zacharias, M.; McCammon, J . A. A Molecular Mechanics Grid
Method for Evaluation of Ligand-Receptor Interactions. J .
Comput. Chem. 1995, 16, 454.
(66) Discover. 98 edn (Molecular Simulation Inc., San Diego, CA).
(67) Dauber-Osguthorpe, P.; Roberts, R. A.; Osguthorpe, D. J .; Wolff,
J .; Genst, M.; Hagler, A. T. Structure and Energetics of Ligand
Binding to Proteins: E. coli Dihydrofolate Reductase-Trimetho-
prim, A Drug-Receptor System. Proteins 1988, 4, 31.
(68) Labow, M.; et al. Characterization of E-Selectin-Deficient mice:
Demonstration of Overlapping Function of the Endothelial
Selectins. Immunity 1994, 1, 709-720.
(69) Sriramarao, P.; Anderson, W.; Wolitzky, B.; Broide, D. Mouse
Bone Marrow-Derived Mast Cells Roll on P-Selectin Under
Conditions of Flow In Vivo. Lab. Invest. 1996, 74, 634-643.
(70) Romano, S. J .; Slee, D. H.; J ohn, J . K.; Yu, J .; Nguyen, T. N.;
Raheja, N. K.; Newman, M. J .; Ripka, W. C. OC 229-648, A
Novel, Non-Carbohydrate Small Molecule Selectin Inhibitor with
Antiinflammatory Activity. Manuscript in preparation.
(71) Sarshar, S.; Siev, D.; Mjalli, A. Imidazole Libraries on Solid
Support. Tetrahedron Lett. 1996, 37, 835-838.
(72) Cargill, J .; Maiefski, R. Automated Combinatorial Chemistry on
Solid Phase. Lab. Rob. Autom. 1996, 3, 139-148.
(73) Molecular Operating Environment (MOE). 1.0 edn (Chemical
Computing Group, Montreal, Quebec).
(49) Erbe, D. V.; Watson, S. R.; Presta, L. G.; Wolitzky, B. A.; Foxall,
C.; Brandley, B. K.; Lasky, L. A. P- and E-Selectin Use Common
Sites for Carbohydrate Ligand Recognition and Cell Adhesion.
J . Cell Biol. 1993, 120, 1227-1235.
(50) Hollenbaugh, D.; Bajorath, J .; Stenkamp, R.; Aruffo, A. Interac-
tion of P-Selectin (CD62) and Its Cellular Ligand: Analysis of
Critical Residues. Biochemistry 1993, 32, 2960-2966.
(51) Bajorath, J .; Hollenbaugh, D.; King, G.; W Harte, J .; Eustice,
D.; Darveau, R.; Aruffo, A. CD62/P-Selectin Binding Sites for
Myeloid Cells and Sulfatides are Overlapping. Biochemistry
1994, 33, 1332-1339.
(52) Revelle, B.; Scott, D.; Beck, P. Single Amino Acid Residues in
the E- and P-Selectin Epidermal Growth Factor Domains Can
Determine Carbohydrate Binding Specificity. J . Biol. Chem.
1996, 271, 16160-16170.
(53) Ichikawa, Y.; Lin, Y.; Dumas, D.; Shen, G.; Garcia-J unceda, E.;
Williams, M.; Bayer, R.; Ketcham, C.; Walker, L.; et al. Chemi-
cal-Enzymic Synthesis and Conformational Analysis of Sialyl
Lewis X and Derivatives. J . Am. Chem. Soc. 1992, 114, 9283-
9298.
(74) (a) Wada, Y.; Saito, T.; Matsuda, N.; Ohmoto, H.; Yoshino, K.;
Ohashi, M.; Kondo, H.; Ishida, H.; Kiso, M.; Hasegawa, A.
Studies on Selectin Blockers. 2. Novel Selectin Blocker as
Potential Therapeutics for Inflammatory Disorders. J . Med.
Chem. 1996, 39, 2055-2059. (b) Compound 29 showed efficacy
at iv doses of 10-30 mg/kg in an IgE-dependent ear inflamma-
tion model developed and run by scientists at Nippon Organon
K. K. (formerly Kanebo, Ltd., New Drug Discovery); the data
for these experiments may be reported in a joint Ontogen-
Nippon Organon manuscript at a later date.
(75) J enison, R. D.; Walker, D. W.; Bargatze, R. F.; Parma, D.
Oligonucleotide Inhibitors of P-Selectin-Dependent Neutrophil-
Platelet Adhesion. Antisense Nucleic Acid Drug Dev. 1998, 8,
265-279.
(76) Sanders, W.; Gordon, E.; Dwir, O.; Beck, P.; Alon, R.; Kiessling,
L. Inhibition of L-Selectin-Mediated Leukocyte Rolling by Syn-
thetic Glycoprotein Mimics. J . Biol. Chem. 1999, 274, 5271-
5278.
(77) Foxall, C.; Watson, S. R.; Dowbenko, D.; Fennie, C.; Lasky, L.
A.; Kiso, M.; Hasegawa, A.; Asa, D.; Brandley, B. K. The Three
Members of the Selectin Receptor Family Recognize A Common
Carbohydrate Epitope, The Sialyl Lewis(x) Oligosaccharide. J .
Cell Biol. 1992, 117, 895-902.
(54) Lin, Y.; Hummel, C.; Huang, D.; Ichikawa, Y.; Nicolaou, K.;
Wong, C. Conformational Studies of Sialyl Lewis X in Aqueous
Solution. J . Am. Chem. Soc. 1992, 114, 5452-5454.
(55) Ball, G.; O’Niell, R.; Schultz, J .; Lowe, J .; Weston, B.; Nagy, J .;
Brown, E.; Hobbs, C.; Bednarski, M. Synthesis and Structural
Analysis Using 2-D NMR of Sialyl Lewis X (SLex) and Lewis X
(Lex) Oligosaccharides: Ligands Related to E-Selectin [ELAM-
1] Binding. J . Am. Chem. Soc. 1992, 114, 5449-5451.
(56) Rao, B.; Anderson, M.; Musser, J .; Gilbert, J .; Schaefer, M.;
Foxall, C.; Brandley, B. Sialyl Lewis X Mimics Derived from a
Pharmacophore Search are Selectin Inhibitors with Antiinflam-
matory Activity. J . Biol. Chem. 1994, 269, 19663-19666.
(57) Scheffler, L.; B Ernst, B.; Katopodis, A.; Magnani, J .; Wang, W.;
Weisemann, R.; Peters, T. Determination of the Bioactive
Conformation of the Carbohydrate Ligand in the E-Selectin/
Sialyl Lewis Complex. Angew. Chem., Int. Ed. Engl. 1995, 34,
1841-1844.
(78) Kiyoi, T.; Inoue, Y.; Ohmoto, H.; Yoshida, M.; Kiso, M.; Kondo,
K. Synthesis of Sialyl Lewis X Pentasaccharide Analogue for
High-throughput Screening of Selectin Blockers. Bioorg. Med.
Chem. 1998, 6, 587-593.
(58) Cooke, R.; Hale, R.; Lister, S.; Shah, G.; Weir, M. The Conforma-
tion of the Sialyl Lewis X Ligand Changes upon Binding to
E-Selectin. Biochemistry 1994, 33, 10591-10596.
J M000508C