Second-generation lymphocyte function-associated antigen-1 inhibitors: 1H-imidazo[1,2-α]imidazol-2-one derivatives

Article abstract of DOI:10.1021/jm049657b

A novel class of lymphocyte function-associated antigen-1 (LFA-1) inhibitors is described. Discovered during the process to improve the physicochemical and metabolic properties of BIRT377 (1, Figure 1), a previously reported hydantoin-based LFA-1 inhibito

Full text of DOI:10.1021/jm049657b

5356
J . Med. Chem. 2004, 47, 5356-5366
Secon d -Gen er a tion Lym p h ocyte F u n ction -Associa ted An tigen -1 In h ibitor s:
1H-Im id a zo[1,2-r]im id a zol-2-on e Der iva tives
J iang-Ping Wu,* J onathan Emeigh, Donghong A. Gao, Daniel R. Goldberg, Daniel Kuzmich, Clara Miao,
Ian Potocki, Kevin C. Qian, Ronald J . Sorcek, Deborah D. J eanfavre, Kei Kishimoto, Elizabeth A. Mainolfi,
Gerald Nabozny, J r., Charline Peng, Patricia Reilly, Robert Rothlein, Rosemarie H. Sellati,
J oseph R. Woska, J r., Shirlynn Chen, J ocelyn A. Gunn, Drane O’Brien, Stephen H. Norris, and Terence A. Kelly
Research and Development, Boehringer Ingelheim Pharmaceuticals, 900 Ridgebury Road, Ridgefield, Connecticut 06877
Received May 6, 2004
A novel class of lymphocyte function-associated antigen-1 (LFA-1) inhibitors is described.
Discovered during the process to improve the physicochemical and metabolic properties of
BIRT377 (1, Figure 1), a previously reported hydantoin-based LFA-1 inhibitor, these compounds
are characterized by an imidazole-based 5,5-bicyclic scaffold, the 1,3,3-trisubstituted 1H-
imidazo[1,2-R]imidazol-2-one (i.e. structure 3). The structure-activity relationship (SAR) shows
that electron-withdrawing groups at C5 on the imidazole ring benefit potency and that oxygen-
containing functional groups attached to a C5-sulfonyl or sulfonamide group further improve
potency. This latter gain in potency is attributed to the interaction(s) of the functionalized
sulfonyl/sulfonamide groups with the protein, likely polar-polar in nature, as suggested by
SAR data. X-ray studies revealed that these bicyclic inhibitors bind to the I-domain of LFA-1
in a pattern similar to that of compound 1.
In tr od u ction
Lymphocyte function-associated antigen-1 (LFA-1) is
a cellular adhesion molecule involved in many funda-
mental immunological processes, such as leukocyte
trafficking, antigen presentation, B-cell activation, and
activation of cytotoxic T lymphocytes.1 LFA-1 inhibitors
have attracted considerable attention in the pharma-
ceutical industry as potential therapeutic agents for
immunological diseases.2 An anti-LFA-1 antibody has
been approved for treatment of psoriasis.2c,e
We previously reported the discovery of BIRT377 (1,
Figure 1), a hydantoin-based reversible and specific
small molecule inhibitor of LFA-1.3 Shown to be an
allosteric modulator that binds to the I-domain of
LFA-1, compound 1 demonstrated good molecular and
cellular potency and was orally active in a mouse model
that measures inhibition of Staphylococcal enterotoxin
superantigen (SEB) induced IL-2 production. However,
this compound was poorly soluble in aqueous solution
and was rapidly metabolized in a human liver mi-
crosome assay (Figure 1). Metabolite identification
studies revealed that N4-demethylation was the major
metabolic pathway.3,4 Developing second-generation
LFA-1 inhibitors with improved solubility and metabolic
stability therefore became the primary objective of our
strucure-activity relationship (SAR) study.
F igu r e 1. The profiles of compound 1.
F igu r e 2.
Previous SAR studies on compound 1 showed that
substitutions at N4 and C5 were well-tolerated.4a This
prompted us to investigate bicyclic structures such as
those depicted in Figure 2 (Structure A). Such struc-
tures would provide additional sites to incorporate
solublizing groups4b and eliminate the possibility of
N4-demethylation. The nonhydantoin nature of struc-
ture A was also particularly attractive in that it would
lead to a structurally novel class of LFA-1 inhibitors.
To test if a second fused ring could be tolerated, a
series of bicyclic analogues were synthesized. As shown
in Figure 3, all four of the initially prepared compounds
were active in the LFA-1/ICAM-1 binding assay,5 indi-
cating that the bicyclic structures were indeed accept-
able. While compounds 4 and 5 were found to be un-
stable in aqueous solution,6 compounds 2 and 3 were
stable and therefore warranted further investigation.
* Corresponding author. Department of Medicinal Chemistry, Boe-
hringer Ingelheim Pharmaceuticals, 900 Ridgebury Road, Ridgefield,
CT 06877. E-mail jwu@rdg.boehringer-ingelheim.com.
10.1021/jm049657b CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/29/2004

Lymphocyte LFA-1 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5357
F igu r e 3. Initial series of bicyclic analogues. Kds are the dissociation constants obtained from the LFA-1/ICAM-1 binding assay.5
Sch em e 1. Synthesis of the Bicyclic Compound 3
Sch em e 2. Regioselective Iodination of Compound 3
In this and subsequent papers, we detail the SAR
studies centered around structure 3 that led to the
development of a novel class of potent LFA-1 inhibitors.7
generated in situ from triphenylphosphine, and azide
9, to produce the guanidine derivative 10. Compound
10 was subsequently cyclized to product 3 by treatment
with trifluoroacetic acid (Scheme 1).
Substitution of the imidazole ring in compound 3 is
summarized in Schemes 2-4. Compound 3 was treated
with 1 equiv of N-iodosuccinimide (NIS) to give the
C5-iodide 11. With 2 equiv of NIS, the compound was
converted to the diiodide 12, which was subsequently
treated with cyclopentylmagnesium bromide followed by
quenching with aqueous ammonium chloride to gener-
ate the C6-iodide 13 (Scheme 2).
Ch em istr y
The syntheses of the compounds discussed in this
paper are summarized in Schemes1-4.8,9 Shown here
as racemic syntheses, these schemes were also adapted
for the preparation of the single enantiomers, using (R)-
or (S)-enantiomers of amino ester 610 as the starting
material.
Reaction of amino ester 6 with 3,5-dichlorophenyl
isothiocyanate afforded the thiohydantoin 7. Compound
7 was then reacted with aza-phosphonium ylide 8,
The two isomers of iodide were further functionalized
via a magnesium-iodine exchange reaction.11 Treat-

5358 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Sch em e 3. Functionalization of the Imidazole Ring
Wu et al.
Sch em e 4. Synthesis of the C5 Sulfones and Sulfonamides
Ta ble 1. Dissociation Constants of Compounds 14a -ea
a
Compounds are racemic.
ment of iodides 11 or 13 with cyclopentylmagnesium
bromide generated the metalated intermediates 11a or
13a , which were trapped in situ with a range of
electrophilic reagents to provide compounds 14a -e and
15a -g, as shown in Tables 1 and 2.
(Tables 3 and 4). Alternatively, the magnesium sulfinate
16 was treated with N-chlorosuccinimide to generate
in situ the sulfonyl chloride intermediate, which was
reacted with amines to produce sulfonamides 18a -f
(Table 5).
The magnesium-iodide exchange reaction was modi-
fied for the synthesis of the sulfones and sulfon-
amides listed in Tables 3 and 4. As shown in Scheme
4, the metalated intermediate 11a generated from
iodide 11 was reacted with SO2 to give the magnesium
sulfinate salt 16.12 This sulfinate salt was in turn re-
acted with alkyl bromide/iodide to give sulfones 17a -g
Resu lts a n d Discu ssion
Ster eoselective In h ibition . It was previously es-
tablished that the stereochemistry at C3 in the hydan-
toin-based LFA-1 inhibitors was important for com-
pound potency. For example, compound 1, the (R)-
enantiomer, was shown to be ∼35 times more potent

Lymphocyte LFA-1 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5359
Ta ble 2. Dissociation Constants of Compounds 15a -ga
a
Compounds are racemic unless otherwise noted.
Ta ble 3. Dissociation Constant for 17a -c
Ta ble 4. SAR at the New Binding Site
than its (S)-enantiomer.3a,4a The existence of the same
C3 chiral center in the bicyclic series led us to examine
the stereochemical influence early on in the exploration
of the SAR. The two enantiomers of compound 3, (R)-3
and (S)-3, were prepared and tested in the binding
assay. While (R)-3 produced a Kd of 172 nM, (S)-3 was
substantially less active (Kd > 7 µM). Thus, the stereo-
selective inhibition observed in the hydantoin series was
retained in the bicyclic series (Figure 4).
P r elim in a r y P r ofile of (R)-3. With molecular po-
tency in the submicromolar range, compound (R)-3 was
deemed as a favorable SAR starting point and fur-
ther profiled. In the J Y cell aggregation assay,13 known
to be LFA-1 dependent, the compound generated an
IC50 of 1.2 µM. However, in an experiment that mea-
sures SEB-induced IL-2 production in mice (oral dosing),
the compound showed no inhibition at the dose of 100
mg/kg. Subsequent analysis revealed an extremely low
plasma concentration that, together with a fast degra-
dation rate (114 pmol/min/mg protein) in human liver
microsomes, suggested metabolic instability as the cause
of in vivo inefficacy. Thus, our effort focused on improv-
ing metabolic stability as well as increasing potency
(Figure 5).
Although no metabolite information was available at
the time, the imidazole moiety in (R)-3 was considered
a potential metabolic liability, due to its electron-rich
nature and known sensitivity toward oxidation. It was
reasoned that attaching functional groups to the imi-
dazole ring would alter the electronic and steric envi-
ronment and modulate the metabolic properties of the
molecule. A search of substitutions capable of stabilizing
the molecule as well as increasing compound potency
was thus started.

5360 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Ta ble 5. Dissociation Constants for Compounds 18a -f
Wu et al.
experiment, a result attributed to its insufficient po-
tency in serum rich media. Additionally, the compound
was shown to be poorly soluble, with a solubility of ∼0.1
µg/mL at pH 7.4. Thus, improving solubility and further
increasing potency became our next objectives.
A New Bin d in g In ter a ction . Our strategy for
increasing the solubility of (R)-15g was to attach polar
groups to the sulfonyl moiety. For this purpose, a series
of sulfones containing ω-hydroxy groups was prepared,
as shown in Table 3. The 2-hydroxy ethyl sulfone
derivative 17a (n ) 2) produced a Kd of 43 nM,
comparable to that of (R)-15g. When the OH group was
further extended out, as in compounds 17b and 17c (n
) 3 and 4), a marked increase in potency, to 13 and 11
nM, respectively, was observed. It appeared that the OH
groups in 17b and 17c, located three and four carbons
from the sulfur atom, picked up a new interaction.
F igu r e 4. Stereoisomers of Compound 3.
Str u ctu r e-Activity Rela tion sh ip s. The initial se-
ries of compounds substituted at C6 or C5 are shown in
Tables 1 and 2. As revealed by Table 1, substitutions
at C6 were generally not well tolerated. With the
exception of the CN group, which had little influence
on potency, all other substitutions reduced potency. At
C5 (Table 2), the electronic characters of the substituting
groups proved important. While electron-rich groups
(i.e., Me, OMe, and I) did not improve potency, the
electron-withdrawing groups (i.e., CN, CO2CH3, and
SO2CH3) did increase the potency. The most effective
substitution was the C5-methanesulfonyl group, the
introduction of which resulted in a compound of 74 nM
(15g).
The (R)-enantiomer of compound 15g, (R)-15g, was
next synthesized and profiled. As shown in Figure 6,
(R)-15g was 50 nM in the binding assay and 0.26 µM
in J Y cell aggregation assay, a considerable improve-
ment compared to the unsubstituted compound (R)-3
(Figure 1). More importantly, (R)-15g was found to be
stable in a human liver microsome assay. In the mouse
SEB/IL-2 experiment, the plasma concentration of (R)-
15g 4 h after dosing was determined to be 2800 and
14 400 ng/mL at doses of 30 and 100 mg/kg, respectively,
a striking improvement compared to that of (R)-3. Thus,
introducing the methanesulfonyl group at C5 had the
double benefit of increasing compound potency as well
as improving metabolic stability.
Subsequent studies were designed to elucidate the
nature of the newly discovered interaction(s). Replacing
the OH groups in 17b and 17c with OAc or OMe
produced compounds of comparable potency (17d -f),
excluding the possibility that the OH groups in 17b and
17c were H-bond donors.14 Substituting the OH group
in 17b with amino groups resulted in loss of potency
(17g and 17h ),15 indicating that oxygen atom(s) were
required. Placing an acetyl group on the amino group
in 17g restored potency (17i, Kd ) 34 nM), further
demonstrating the essential role of oxygen atom(s) in
the interaction. Although the exact nature of the
interaction could not be deduced, it likely involves a
polar-polar interaction(s) (Table 4).16
Further SAR studies focused on the directional pref-
erences of the interaction. A series of cyclic analogues
were synthesized and tested, as summarized in Table
5. The piperidine sulfonamide 18a has a Kd of 50 nM,
similar to the methyl sulfone derivative (R)-15g. When
a hydroxy or carbonyl group was placed at the 4-position
of the piperidine ring, a significant increase in potency
was observed (18d or 18e). Placing an OH at the
3-position had no beneficial effect (18c). The morpholine
sulfonamide 18b, with an ethereal oxygen at the 4-posi-
tion, also showed a moderate gain in potency. An amino
group, as in 18f, showed no beneficial effect. Thus, a
Despite of the good exposure level in mice, compound
(R)-15g did not show in vivo efficacy in the mouse

Lymphocyte LFA-1 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5361
F igu r e 5. Profiles of R-3.
F igu r e 6. Profiles of compound (R)-15g.
Ta ble 6. Cellular Activities for Selected Compounds
proper orientation, in addition to the presence of oxygen
atom, is required for an effective interaction.
account for the lower potency in the whole blood assay.
Studies on the SAR of protein binding will be reported
shortly.
P r otein Bin d in g Stu d ies. With compounds of low
nanomolar potency identified, we proceeded to the
cellular level of compound evaluation. In the J Y cell
aggregation13 assay, compounds with low nanomolar Kds
consistently generated IC50s less than 1 µM. However,
in a human whole blood assay,5 which measures the
SEB-induced IL-2 production, the IC50s were well above
the micromolar level (Table 6).
The discrepancies between the two cellular assays
were attributed to protein binding. In a modified binding
assay,5,17 which included 1% of human plasma proteins
during incubation, a dramatic reduction in binding
potency was observed (Table 6, column 4). The presence
of plasma proteins apparently reduced the free com-
pound concentration and resulted in the lower observed
potency. The same protein binding effect was likely to
X-r a y Cr ysta llogr a p h y Stu d ies. The bicyclic class
of LFA-1 inhibitors had been assumed to bind to the
I-domain of LFA-1, similar to compound 1.3b This was
confirmed when the cocrystal structure of compound
17d and I-domain was obtained, as presented in Fig-
ure 7. Similar to compound 1,3b the N-1 dichlorophenyl
moiety of 17d was located in a hydrophobic pocket
between R-helices 1 and 7 and â-strands 1, 3, and 4.
The bromophenyl group at C3 was oriented toward
the two R-helices 1 and 7. However, the X-ray structure
did not reveal the detailed binding pattern of the
C5-acetylpropylsulfonyl group, which was apparently
responsible for the potency increase in 17d . In Figure
7, the C5 acetylpropylsulfonyl was shown to extend
toward solvent, not in apparent contact with any amino

5362 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Wu et al.
mixture was extracted with CH2Cl2. The organic layer was
washed with water, dried, and concentrated to give azide 9 in
quantitative yield: 1H NMR (CDCl3, 400 MHz) 5.10 (1H, t, J
) 3.5 Hz), 4.15-4.08 (2H, m), 4.05-3.92 (2H, m), 3.35 (2H, d,
J ) 3.5 Hz) ppm.
5-(4-Br om oben zyl)-3-(3,5-d ich lor op h en yl)-2-[(E)-[1,3]-
d ioxola n -2-ylm e t h ylim in o]-5-m e t h ylim id a zolid in -4-
on e (10). To a solution of PPh3 (2.36 g, 9.0 mmol) in toluene
(20 mL) was added azide 9 (1.2 g, 9.0 mmol). After stirring at
room temperature overnight, thiohydantoin 7 (2 g, 4.5 mmol)
was added. The mixture was sealed under N2 in a pressure
tube and heated at 130-140 °C for 3-4 days. The mixture
was then cooled to room temperature, concentrated, and
purified by silica gel chromatography to give the product 10
(1.96 g, yield 85%): 1H NMR (CDCl3, 400 MHz) δ 7.40-7.35
(3H, m), 7.05 (2H, d, J ) 8 Hz), 6.55 (2H, s), 5.05-4.92 (1H,
m), 4.00-3.90 (4H, m), 3.82 (1H, m), 3.75-3.65 (1H, m), 3.55-
3.45 (1H, m), 3.02 (2H, abq, J ) 14, 22 Hz), 2.28 (1H, s), 1.58
(3H, s) ppm.
3-(4-Br om ob en zyl)-1-(3,5-d ich lor op h en yl)-3-m et h yl-
1H-im id a zo[1,2-r]im id a zol-2-on e (3). To a solution of 10 in
CH2Cl2 was added trifluoroacetic acid (TFA, 5-6 equiv). The
mixture was heated under N2 at 90 °C overnight. The mixture
was then cooled to room temperature, diluted with EtOAc,
washed with saturated NaHCO3, dried (Na2SO4), and concen-
trated. The residue was purified by silica gel chromatography
to give the title compound 3: yield >85%; mp 36.0-37.5 °C;
1H NMR (CDCl3, 400 MHz) δ 7.76 (2H, s), 7.35 (2H, d, J ) 8.3
Hz), 7.00 (1H, s), 6.92 (1H, s), 6.81 (2H, d, J ) 8.3 Hz), 3.32
(1H, d, J ) 14.0 Hz), 3.18 (1H, d, J ) 13.9 Hz), 1.80 (3H, s)
ppm; 13C NMR (CDCl3, 100 M Hz) δ 174.3, 145.3, 134.8, 134.2,
131.5, 131.2, 130.6, 128.5, 126.6, 121.6, 119.8, 110.6, 65.5, 43.6,
22.7 ppm; mass spectrum (EI) m/z 449 (M+). Anal. Calcd for
F igu r e 7. X-ray cocrystal structure of 17d and the I-domain.
acid residues. One possibility is that the acetylpropyl-
sulfonyl residue interacts with portions of LFA-1 outside
of the I-domain.
Su m m a r y
We have identified a novel class of potent small
molecule inhibitors of LFA-1. These compounds are
characterized by the 1,3,3-trisubstituted 1H-imidazo-
[1,2-R]imidazol-2-one scaffold. Substitution at C5 with
electron-withdrawing groups benefits potency. Certain
functionalized sulfonyl or sulfonamide groups at this
position further enhance potency by engaging in a new
interaction not known previously. SAR data suggests
that the new interaction most likely involves a polar-
polar interaction(s). X-ray studies reveal that these
compounds bind to the I-domain in a pattern similar to
that of compound 1.
C
19H14BrCl2N3O: C, 50.58; H, 3.13; N, 9.31. Found: C, 50.64;
H, 3.17; N, 9.12.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-5-iod o-3-
m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (11). To a solu-
tion of compound 3 (1.82 g, 4.04 mmol), in CH2Cl2 (20 mL),
cooled to 0 °C was added in small portions N-iodosuccinimide
(1.43 g, 6.04 mmol). Pyridinium p-toluenesulfonate (100 mg,
0.40 mmol) was added and the mixture was stirred at 0 °C for
3 h, during which time additional N-iodosuccinimide (400 mg,
1.68 mmol) was added to complete the reaction. The mixture
was diluted with CH2Cl2, washed with 10% Na2SO3 solution,
dried, and concentrated. The residue was purified by silica gel
chromatography to give the title compound 11 (1.8 g): yield
80%; 1H NMR (CDCl3, 400 MHz) δ 7.61 (2H, s), 7.39 (2H, d, J
) 8.3 Hz), 7.30-7.32 (m, 1H), 7.01 (1H, s), 6.83 (2H, d, J )
8.3 Hz), 3.29 (1H, d, J ) 13.9 Hz), 3.16 (1H, d, J ) 14.0 Hz),
1.78 (3H, s) ppm; 13C NMR (CDCl3, 100 MHz) δ 174.3, 146.4,
135.5, 134.3, 131.9, 131.7, 131.2, 127.6, 122.4, 120.7, 116.8,
80.3, 66.8, 44.0, 23.2. ppm; mass spectrum (CI) m/z 576 (M +
H)+.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-5,6-d iiod o-3-
m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (12). Diiodide 12
was synthesized from 3 using 2.2 mol equiv of N-iodosuccin-
imide by the same procedure as described above: yield 60%;
1H NMR (CDCl3, 400 MHz) 7.44 (2H, s), 7.35-7.25 (3H, m),
6.82 (2H, d, J ) 8 Hz), 3.58 (1H, d, J ) 14 Hz), 3.26 (1H, d, J
) 14 Hz), 1.92 (3H, s) ppm.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-6-iod o-3-
m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (13/14b). To a
solution of compound 12 (92 mg, 0.131 mmol) in THF (2 mL)
at -30 °C was added cyclopentylmagnesium bromide (2.0 M
in ether, 0.131 mL, 0.262 mmol) under nitrogen. The solu-
tion was stirred at -30 °C for 1.5 h before saturated NH4Cl
solution was added. The mixture was warmed to room tem-
perature and extracted with EtOAc. The organic layer was
dried with Na2SO4 and concentrated. The residue was purified
by silica gel chromatography to give the iodide 13 (45 mg, yield
60%): 1H NMR (CDCl3, 400 MHz) δ 7.62 (2H, s), 7.40 (2H, d,
J ) 8 Hz), 7.30 (1H, m), 7.04 (1H, s), 6.84 (2H, d, J ) 8 Hz),
3.30 (1H, d, J ) 14 Hz), 3.18 (1H, d, J ) 14 Hz), 1.84 (3H, s)
ppm.
Exp er im en ta l Section
All reactions were carried out in oven or flame-dried
glassware under nitrogen atmosphere in commercially sup-
plied (EMS or Aldrich) dry solvents. Thin-layer chromatogra-
phy was performed on Analtech or E-Merck silica gel TLC
plates. Developed plates were visualized using 254 nm UV
illumination or by PMA stain. Flash chromatography was
performed using 230-400 mesh Merck silica gel.
1H and 13C NMR spectra were recorded on Bruker Ultra-
Shield-400 MHz spectrometers in solvents as noted. IR spectra
were recorded on a Nicolet Impact-410 instrument. Mass
spectra were obtained on Micromass Platform LCZ mass
spectrometers using either electrospray positive/negative ion-
izations or chemical ionization. Elemental analysis was per-
formed by Quantitative Technologies, Whitehouse, NJ .
Syn th esis of Com p ou n d 3 (Sch em e 1). 5-(4-Br om oben -
zyl)-3-(3,5-d ich lor op h en yl)-5-m eth yl-2-th ioxoim id a zoli-
d in -4-on e (7). A solution of amino-ester 6 (3.40 g, 12.5 mmoL)
and 3,5-dichlorophenylisothiocyanate (2.60 g, 12.7 mmoL) in
1,4-dioxane (15 mL) was heated at 90 °C overnight. The
mixture was concentrated to give a thick oil, which was
crystallized from EtOAc-hexane to give 5.0 g of thiohydantoin
derivative 7: yield 90%; 1H NMR (CDCl3, 400 MHz) δ 7.56
(2H, d, J ) 8 Hz), 7.48 (1H, s), 7.15 (2H, d, J ) 8 Hz), 6.78
(2H, s), 3.22 (1H, d, J ) 14 Hz), 3.05 (1H, d, J ) 14 Hz), 1.70
(3H, s) ppm.
2-(Azid om eth yl)[1,3]d ioxola n e (9). A mixture of 10.0 g
(0.060 moL) of 2-(bromomethyl)-1,3-dioxolane and 20 g (0.3
moL) of sodium azide in 100 mL of DMSO was heated at 100
°C for 30 min. The mixture was then cooled to room temper-
ature and poured into 200 mL of an ice-water mixture. The

Lymphocyte LFA-1 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5363
Syn th esis of Com pou n ds 14a-14e an d 15a-15 g (Tables
compound 15b (22 mg, 26% two steps). M.p. 170.1-171.3 °C;
1H NMR (CDCl3, 400 MHz) δ 7.72 (2H, s), 7.30 (3H, m), 6.82
(2H, d, J ) 8 Hz), 6.10 (1H, s), 4.00 (3H, s), 3.35 (1H, d, J )
14 Hz), 3.28 (1H, d, J ) 14 Hz), 1.82 (3H, s) ppm; mass
spectrum (CI) m/z 480 (M + H)+.
1
a n d 2). 6-Br om o-3-(4-b r om ob en zyl)-1-(3,5-d ich lor o-
p h en yl)-3-m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (14a ).
Compound 14a was prepared from compound 3 and N-
bromosuccinimide via the same procedure as for the prepara-
tion of iodide 13: 1H NMR (CDCl3, 400 MHz) δ 7.62 (2H, s),
7.40 (2H, d, J ) 8 Hz), 7.30 (1H, m), 6.90 (1H, s), 6.85 (2H, d,
J ) 8 Hz), 3.30 (1H, d, J ) 14 Hz), 3.15 (1H, d, J ) 14 Hz),
1.81 (3H, s) ppm; mass spectrum (ES+) m/z 528 (M + H)+.
Syn th esis of 14b: See Syn th esis of Com p ou n d 13.
Gen er a l P r oced u r e I. To a solution of the iodide 13 or 11 (1
equiv) in THF at -35 °C was added cyclopentylmagnesium
bromide (2 M in ether, 3 equiv) under nitrogen. The solution
was stirred at -35 °C for 1 h before an electrophilic reagent
(5-8 equiv) was added. The mixture was stirred at -35 °C
for 30 min and then at room temperature for 1 h. Saturated
NaHCO3 solution was added at 0 °C. The mixture was
extracted with EtOAc, and the organic layer was dried with
Na2SO4 and concentrated. The residue was purified by silica
gel chromatography to give the products 14a -e or 15a -g.
5-(4-Br om oben zyl)-7-(3,5-d ich lor op h en yl)-5-m eth yl-6-
oxo-6,7-d ih yd r o-5H -im id a zo[1,2-r]im id a zole-2-ca r b on i-
tr ile (14c). Compound 14c was prepared from iodide 13
according to general procedure I, using tosyl cyanide as the
electrophile: yield 38%; 1H NMR (CDCl3, 400 MHz) δ 7.66 (2H,
s), 7.46 (1H, s), 7.41 (2H, d, J ) 8.3 Hz), 7.54 (1H, m), 6.81
(2H, d, J ) 8.3 Hz), 3.36 (1H, d, J ) 14.1 Hz), 3.20 (1H, d, J
) 14.1 Hz), 1.85 (3H, s) ppm; mass spectrum (CI) m/z 475 (M
+ H)+.
5-(4-Br om oben zyl)-7-(3,5-d ich lor op h en yl)-5-m eth yl-6-
oxo-6,7-d ih yd r o-5H-im id a zo[1,2-r]im id a zole-2-ca r boxy-
lic Acid Meth yl Ester (14d ). Compound 14d was prepared
from iodide 13 according to general procedure I, using methyl
chloroformate as the electrophile: yield 16%; 1H NMR (CDCl3,
400 MHz) δ 7.69 (1H, s), 7.51 (2H, s), 7.38-7.34 (3H, m), 6.80
(2H, d, J ) 8.3 Hz), 3.95 (3H, s), 3.37 (1H, d, J ) 14.0 Hz),
3.20 (1H, d, J ) 14.0 Hz), 1.85 (3H, s) ppm; mass spectrum
(CI) m/z 508 (M + H)+.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-6-m eth a n e-
su lfon yl-3-m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (14e).
Compound 14e was prepared from iodide 13 according to
general procedure I, using methylsulfonyl chloride as the
electrophile: yield 22%; 1H NMR (CDCl3, 400 MHz) δ 7.66 (2H,
s), 7.62 (1H, s), 7.40-7.36 (3H, m), 6.84 (2H, d, J ) 8.2 Hz),
3.38 (1H, d, J ) 14.1 Hz), 3.22 (1H, d, J ) 14.0 Hz), 3.21 (3H,
s), 1.86 (3H, s) ppm.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-3,5-d im eth -
yl-1H-im id a zo[1,2-r]im id a zol-2-on e (15a ). To a suspension
of anhydrous LiCl (10.0 mg, 0.236 mmol) and CuCN (10.5 mg,
0.117 mmol) in THF (0.2 mL), cooled at -20 °C, was added
CH3MgBr (1.4 M in THF, 0.21 mL, 0.294 mmol) under N2. The
solution was stirred at -20 °C for 15 min. A solution of
compound 11 (34 mg, 0.059 mmol) in THF (0.5 mL) was added.
The reaction mixture was stirred at -20 °C for 2 h and then
room temperature overnight before being quenched with
saturated NH4Cl at 0 °C. The mixture was extracted with
EtOAc, dried with Na2SO4, and concentrated. The residue was
purified with preparative thin-layer chromatography (prep-
TLC) to give 2 mg (yield 6%) of 15a : 1H NMR (CDCl3, 400 MHz)
δ 7.65 (2H, s), 7.32-7.28 (3H, m), 6.74 (2H, d, J ) 8 Hz), 6.65
(1H, s), 3.33 (1H, d, J ) 14.0 Hz), 3.27 (1H, d, J ) 14.0 Hz),
2.47 (3H, s), 1.98 (3H, s) ppm; mass spectrum (EI) m/z 463 (M
+ H)+.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-5-m eth oxy-
3-m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (15b). To a so-
lution of compound 3 (80 mg) in methanol (2 mL) was added
toluenesulfonic acid (2 mg). The mixture was stirred at room
temperature overnight. The mixture was concentrated. The
residue was redissolved in HOAc (1 mL). To this mixture was
added NaOAc (anhydrous, 50 mg). The mixture was heated
at 100 °C for 4 h. The mixture was concentrated, and the
residue was taken up in EtOAc (5 mL). The mixture was
filtered through a short silica gel plug, concentrated and
purified by Prep-TLC using 33% ethyl acetate-hexane to give
5-Br om o-3-(4-b r om oben zyl)-1-(3,5-d ich lor op h en yl)-3-
m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (15c). Compound
15c was synthesized from compound (R)-3 and N-bromosuc-
cinimide according to the same procedure as the synthesis of
1
iodide 11: yield 85%; H NMR (CDCl3, 400 MHz) δ 7.60 (2H,
s), 7.35-7.26 (3H, m), 6.89 (1H, s), 6.85 (2H, d, J ) 8 Hz),
3.54 (1H, d, J ) 14 Hz), 3.32 (1H, d, J ) 14 Hz), 1.98 (3H, s)
ppm; mass spectrum (CI) m/z 528 (M + H)+.
(R)-3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-5-iod o-3-
m eth yl-1H-im id a zo[1,2-r]im id a zol-2-on e (15d ) ((R)-En a n -
tiom er of Iod id e 11). See preparation of compound 11.
5-(4-Br om oben zyl)-7-(3,5-d ich lor op h en yl)-5-m eth yl-6-
oxo-6,7-d ih yd r o-5H -im id a zo[1,2-r]im id a zole-3-ca r b on i-
tr ile (15e). Compound 15e was prepared from iodide 11
according to general procedure I, using tosylcyanide as the
electrophile: yield 35%; 1H NMR (CDCl3, 400 MHz) δ 7.56 (3H,
m), 7.37-7.30 (3H, m), 6.84 (2H, d, J ) 8.3 Hz), 3.48 (1H, d,
J ) 14.3 Hz), 3.41 (1H, d, J ) 14.2 Hz), 1.99 (3H, s) ppm; mass
spectrum (CI) m/z 475 (M + H)+.
5-(4-Br om oben zyl)-7-(3,5-d ich lor op h en yl)-5-m eth yl-6-
oxo-6,7-d ih yd r o-5H-im id a zo[1,2-r]im id a zole-3-ca r boxy-
lic Acid Meth yl Ester (15f). Compound 15f was prepared
from iodide 11 according to general procedure I, using methyl
chloroformate as the electrophile: yield 38%; 1H NMR (CDCl3,
400 MHz) 7.61 (1H, s), 7.57 (2H, m), 7.35 (1H, s), 7.29 (2H, d,
J ) 8.0 Hz), 6.75 (2H, d, J ) 8.0 Hz), 4.01 (3H, s), 3.86 (1H, d,
J ) 16.0 Hz), 3.34 (1H, d, J ) 16.0 Hz), 1.99 (3H, s) ppm; mass
spectrum (CI) m/z 508 (M + H)+.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-5-m eth ylsu l-
fon yl-3-m et h yl-1H -im id a zo[1,2-r]im id a zol-2-on e (15g).
Compound 15g was prepared from iodide 11 according to
general procedure I, using methylsulfonyl chloride as the
1
electrophile: yield 66%; mp 92-93 °C; H NMR (CDCl3, 400
MHz) δ 7.51(1H, s), 7.38 (2H, s), 7.33 (1H, s), 7.26 (2H, d,
J ) 8.2 Hz), 6.80 (2H, d, J ) 8.3 Hz, 3.86 (1H, d, J ) 13.9
Hz), 3.25 (3H, s), 3.24 (1H, d, J ) 14 Hz), 2.03 (3H, s) ppm;
mass spectrum (ES+) m/z 528 (M + H)+. Anal. Calcd for
C20H16BrCl2N3O3S1: C, 45.39; H, 3.05; N, 7.94. Found, C, 45.64;
H, 3.13; N, 7.95.
Alter n a tive Syn th esis of Com p ou n d 15g. Compound
15g was also prepared from the magnesium sulfinate 16, ac-
cording to general procedure II (see below), using iodo-
methane as the alkylating reagent: yield 83%.

5364 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Wu et al.
Br om om a gn esiu m 5-(4-b r om ob en zyl)-7-(3,5-d ich lo-
r op h en yl)-5-m eth yl-6-oxo-6,7-d ih yd r o-5H-im id a zo[1,2-r]-
im id a zole-3-su lfin a te (16). A solution of compound 11 (2.5
g, 4.33 mmol) in 25 mL of THF was treated with cyclopentyl-
magnesium bromide (2.6 mL, 2 M, 5.2 mmol) at -40 °C under
argon. The mixture was stirred at -40 °C for 40 min and then
SO2 was bubbled in over 1 min. The mixture was stirred at
-40 °C for 15 min and then room temperature for 1 h before
being concentrated. The residue was twice redissolved in fresh
dry THF (25 mL each time) and concentrated to give 3.8 g of
the solid magnesium salt 16. The product thus obtained was
assumed to be 70% pure based on theoretical yield and used
in later reactions without further purification.
Syn th esis of Com p ou n d s 17a -i (Ta bles 3 a n d 4)
Gen er a l P r oced u r e II. The magnesium salt (R)-16 (1 equiv)
was dissolved in dry DMF. To this solution were added
powdered potassium carbonate (3 equiv) and an alkyl bromide/
iodide. The mixture was heated at 70-80 °C for 2-16 h until
the reaction was complete. The mixture was then cooled to
room temperature, diluted with water, and extracted with
EtOAc. The organic layer was dried over MgSO4 and concen-
trated. The residue was purified by silica gel chromatography
to give sulfones 17a -g.
prepared from (R)-16 and 3-bromopropyl acetate according to
general procedure II: yield 43%; H NMR (CDCl3, 400 MHz)
1
δ 7.53 (1H, s), 7.41 (2H, s), 7.38 (1H, m), 7.31 (2H, d, J ) 7.2
Hz), 6.84 (2H, d, J ) 8.3 Hz), 4.27-4.21 (2H, m), 3.90 (1H, d,
J ) 13.9 Hz), 3.43-3.28 (2H, m), 3.27 (1H, d, J ) 14.0 Hz),
2.34-2.25 (1H, m), 2.22-2.13 (1H, m), 2.10 (3H, s), 2.05 (3H,
s) 2.00-1.75 (3H, m) ppm.
3-(4-Br om ob en zyl)-1-(3,5-d ich lor op h en yl)-5-(3-m et h -
oxyp r op yl-1-su lfon yl)-3-m eth yl-1H-im id a zo[1,2-r]im id a -
zol-2-on e (17e). To a solution of 20 mg (35 µmol) of 17b in 2
mL of ether was added 0.1 mL (1.6 mmol) of MeI, 16 mg (70
mmol) of Ag2O, and 19 mg of Ag2CO3. The mixture was stirred
for 3 days. The mixture was diluted with CH2Cl2 and filtered.
The filtrate was concentrated and purified by silica gel Prep-
TLC with 20% acetone-hexane to give 8 mg of 17e: yield 39%;
1H NMR (CDCl3, 400 MHz) δ 7.52 (1H, s), 7.41 (2H, s), 7.36
(1H, s), 7.31-7.27 (2H, m), 6.85 (2H, d, J ) 8 Hz), 3.90 (1H, d,
J ) 14 Hz), 3.50 (2H, t, J ) 6.0 Hz), 3.40-3.28 (5H, m), 3.25
(1H, d, J ) 14 Hz), 2.25-2.10 (1H, m), 2.10-2.02 (1H, m), 2.02
(3H, s) ppm; mass spectrum (ES+) m/z 586 (M + H)+.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-5-(2-h y-
d r oxyeth a n esu lfon yl)-3-m eth yl-1H-im id a zo[1,2-r]im id a -
zol-2-on e (17a ). The sulfinate salt (R)-16 was reacted with
2-bromoethyl acetate according to general procedure II. Only
the deacetylated product 17a was isolated: yield 7.5%; mp
1
139.2-140.0; H NMR (CDCl3, 400 MHz) δ 7.57 (1H, s), 7.41
(2H, s), 7.38 (1H, s), 7.31 (2H, d, J ) 8.1 Hz), 6.85 (2H, d, J )
8.2 Hz), 4.21-4.16 (2H, m), 3.89 (1H, d, J ) 13.8 Hz), 3.58-
3.47 (2H, m), 3.29 (1H, d, J ) 13.9 Hz), 2.43 (1H, t, J ) 5.2
Hz), 2.06 (3H, s) ppm; mass spectrum (ES+) m/z 558 (M + H)+.
Anal. Calcd for C21H18BrCl2N3O4S1‚0.5H2O: C, 44.39; H, 3.37;
N, 7.39. Found: C, 44.47; H, 3.41; N, 7.39.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-5-(3-h y-
d r oxyp r op yl-1-su lfon yl)-3-m eth yl-1H-im id a zo[1,2-r]im i-
d a zol-2-on e (17b). A solution 17d (44 mg) in methanol was
treated with 30% HBr/HOAc. The mixture was stirred at room
temperature overnight and concentrated. The residue was
diluted with CH2Cl2, washed with aq NaHCO3, and concen-
trated. Purification by prep-TLC gave 13 mg of 17b: yield 31%;
mp 112.6-113.6 °C; 1H NMR (CDCl3, 400 MHz) δ 7.53
(1H, s), 7.42 (2H, s), 7.37 (1H, s), 7.31 (2H, d, J ) 8 Hz), 6.86
(2H, d, J ) 8.3 Hz), 3.92 (1H, d, J ) 13.8 Hz), 3.87-3.84 (2H,
m), 3.53-3.40 (2H, m), 3.28 (1H, d, J ) 13.8 Hz), 2.22-2.20
(2H, m), 2.06 (3H, s) ppm; mass spectrum (ES+) m/z 573.9 (M
+ H)+.
Acetic a cid 4-[5-(4-Br om oben zyl)-7-(3,5-d ich lor op h e-
n yl)-5-m eth yl-6-oxo-6,7-d ih yd r o-5H-im id a zo[1,2-r]im id a -
zolyl-3-su lfon yl]bu tyl Ester (17f). Compound 17f was pre-
pared from (R)-16 and 4-bromobutyl acetate according to
1
general procedure II: yield 30%; H NMR (CDCl3, 400 MHz)
δ 7.51 (1H, s), 7.42 (2H, s), 7.37 (1H, m), 7.31 (2H, d, J ) 8.3
Hz), 6.85 (2H, d, J ) 8.3 Hz), 4.20-4.10 (2H, m), 3.90 (1H, d,
J ) 13.9 Hz), 3.30-3.20 (3H, m), 2.07 (3H, s), 2.10-2.00
(4H, m), 2.00-1.75 (3H, m) ppm; mass spectrum (ES+) m/z
628 (M + H)+.
Compounds 17g-i were prepared according to the following
scheme.
Met h a n esu lfon ic Acid 3-[5-(4-Br om ob en zyl)-7-(3,5-
d ich lor op h en yl)-5-m eth yl-6-oxo-6,7-d ih yd r o-5H-im id a zo-
[1,2-r]im id a zolyl-3-su lfon yl]p r op yl Ester (17b-OMs). To
a solution of 37 mg (64.5 µmol) of 17b in 2 mL of CH2Cl2 at
room temperature was added 20 µL (0.14 mmol) of triethyl-
amine and 4 µL of methylsulfonyl chloride. The mixture was
stirred at room temperature for 3 h. The reaction was
quenched with saturated NH4Cl. The mixture was extracted
with CH2Cl2, and the organic layer was dried and concen-
trated. The residue was purified by Prep-TLC with 5% MeOH-
CH2Cl2 to give 30 mg of the mesylate 17b-OMs.
5-(3-Am in op r op yl-1-su lfon yl)-3-(4-br om oben zyl)-1-(3,5-
d ich lor op h en yl)-3-m eth yl-1H-im id a zo[1,2-r]im id a zol-2-
on e (17g). A solution of 53 mg (81 µmol) of 17b-OMs in 2 mL
of 2 M NH3-EtOH was heated in a sealed tube at 80 °C
overnight. The mixture was then cooled to room temperature
and concentrated. The residue was purified by Prep-TLC using
5% MeOH-CH2Cl2 with 1% concentrated NH3‚H2O to give 31
mg of 17g: yield 66.5%; mp 75 °C; 1H NMR (CDCl3, 400 MHz)
δ 7.52 (1H, s), 7.42 (2H, m), 7.37 (1H, m), 7.30 (2H, d, J ) 8.3
Hz), 6.85 (2H, d, J ) 8.3 Hz), 3.91 (1H, d, J ) 13.9 Hz), 3.50-
3.32 (2H, m), 3.27 (1H, d, J ) 13.9 Hz), 2.91 (2H, t, J ) 6.5
Hz), 2.00-1.98 (4H, m), 1.98-1.87 (1H, m), ppm; mass
spectrum (ES+) m/z 571 (M + H)+.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-5-(4-h y-
d r oxyb u t yl-1-su lfon yl)-3-m et h yl-1H -im id a zo[1,2-r]im i-
d a zol-2-on e (17c). The acetate 17f was hydrolyzed to 17c
using 30% HBr/HOAc in methanol in a similar procedure for
1
the synthesis of 17b: yield 84%; H NMR (CDCl3, 400 MHz)
δ 7.52 (1H, s), 7.42 (2H, s), 7.37 (1H, m), 7.32-7.29 (2H, m),
6.85 (2H, d, J ) 8.3 Hz), 3.91 (1H, d, J ) 13.9 Hz), 3.75 (2H,
t, J ) 6.0 Hz), 3.38-3.28 (2H, m), 3.27 (1H, d, J ) 13.9 Hz),
2.05 (3H, s), 2.00-1.85 (2H, m), 1.85-1.70 (2H, m) ppm; mass
spectrum (ES+) m/z 586 (M + H)+.
Acetic Acid 3-[5-(4-Br om oben zyl)-7-(3,5-d ich lor op h e-
n yl)-5-m eth yl-6-oxo-6,7-d ih yd r o-5H-im id a zo[1,2-r]im id a -
zolyl-3-su lfon yl]p r op yl Ester (17d ). Compound 17d was
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-3-m eth yl-5-
(3-m eth yla m in op r op yl-1-su lfon yl)-1H-im id a zo[1,2-r]im i-
d a zol-2-on e (17h ). A solution of 20 mg (31 µmol) of 17b-OMs
in 2 mL of 2 M MeNH2-MeOH was heated in a sealed tube at

Lymphocyte LFA-1 Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22 5365
d, J ) 8.3 Hz), 6.87 (2H, d, J ) 8.3 Hz), 3.89 (1H, d, J ) 13.6
Hz), 3.33-3.15 (5H, m), 2.00 (3H, s), 1.78-1.68 (4H, m), 1.62-
1.50 (2H, m) ppm; mass spectrum (ES+) m/z 597 (M + H)+.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-3-m eth yl-5-
(m or p h olin yl-4-su lfon yl)-1H -im id a zo[1,2-r]im id a zol-2-
on e (18b). Compound 18b was prepared according to general
procedure III, using morpholine as the amine: yield 21%; mp
76.1-77.4 °C; 1H NMR (CDCl3, 400 MHz) δ 7.43 (2H, s), 7.40
(1H, s), 7.37 (1H, m), 7.31 (2H, d, J ) 7 Hz), 6.86 (2H, d, J )
8.4 Hz), 3.88-3.85 (5H, m), 3.37-3.25 (5H, m), 2.02 (3H, s)
ppm; mass spectrum (ES+) m/z 599 (M + H)+.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-5-(3-h y-
d r oxyp ip er id in yl-1-su lfon yl)-3-m eth yl-1H-im id a zo[1,2-r]-
im id a zol-2-on e (18c). Compound 18c was prepared according
to general procedure III, using 3-hydroxypiperidine as the
amine: yield 70%; 1H NMR (CDCl3, 400 MHz) δ 7.43-7.30
(6H, m), 6.87 (2H, d, J ) 8.2 Hz), 4.00 (1H, br s), 3.86 (1H, d,
J ) 13.8 Hz), 3.60-3.51 (1H, m), 3.40-3.30 (1H, m), 3.26 (1H,
d, J ) 13.9 Hz), 3.24-3.06 (2H, m), 2.09 (3H, s), 2.09-1.62
(4H, m) ppm; mass spectrum (ES+) m/z 613 (M + H)+.
3-(4-Br om ob e n zyl)-1-(3,5-d ich lor op h e n yl)-5-(4-h y-
d r oxyp ip er id in yl-1-su lfon yl)-3-m eth yl-1H-im id a zo[1,2-r]-
im id a zol-2-on e (18d ). Compound 18d was prepared accord-
ing to general procedure III, using 4-hydroxypiperidine as the
amine: yield 41%; mp 67.2-68.5 °C; 1H NMR (CDCl3, 400
MHz) δ 7.43 (2H, s), 7.40 (1H, s), 7.36 (1H, m), 7.30 (2H, d, J
) 8 Hz), 6.87 (2H, d, J ) 8.4 Hz), 4.17 (1H, br s), 3.87 (1H, d,
J ) 13.8 Hz), 3.57-3.50 (3H, m), 3.27-3.24 (3H, m), 2.08-
2.00 (2H, m), 2.02 (3H, s), 1.82-1.79 (2H, m) ppm; mass
spectrum (ES+) m/z 613 (M + H)+.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-3-m eth yl-5-
(4-oxop ip er id in yl-1-su lfon yl)-1H-im id a zo[1,2-r]im id a zol-
2-on e (18e). Compound 18e was prepared according to general
procedure III, using 4-piperidone as the amine: yield 60%; 1H
NMR (CDCl3, 400 MHz) δ 7.35 (1H, s), 7.32 (2H, s), 7.30 (1H,
m), 7.25 (2H, d, J ) 8 Hz), 6.82 (2H, d, J ) 8 Hz), 3.88 (1H, d,
J ) 14 Hz), 3.74-3.60 (4H, m), 3.22 (1H, d, J ) 14 Hz), 2.65
(4H, m), 1.98 (3H, s) ppm.
70 °C overnight. The mixture was then cooled to room
temperature and concentrated. The residue was purified by
Prep-TLC using 5% MeOH-CH2Cl2 with 1% concentrated
NH3‚H2O to give 3 mg of 17h : yield 17%; 1H NMR (CDCl3,
400 MHz) δ 7.48 (1H, s), 7.37 (2H, s), 7.33 (1H, s), 7.26 (2H,
m), 6.90 (2H, m), 3.85 (1H, d, J ) 14.9 Hz), 3.40-3.27 (2H,
m), 3.23 (1H, d, J ) 14.0 Hz), 2.85-2.78 (3H, m), 2.44 (3H, s),
2.45-2.00 (4H, m), 2.01 (3H, s) ppm.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-3-m eth yl-5-
(p ip er a zin yl-1-su lfon yl)-1H -im id a zo[1,2-r]im id a zol-2-
on e (18f). Compound 18f was obtained according to general
1
procedure III, using piperazine as the amine: yield 90%; H
NMR (CDCl3, 400 MHz) δ 7.42 (2H, m), 7.38 (1H, s), 7.36 (1H,
m), 7.30 (2H, d, J ) 8.3 Hz), 6.86 (2H, d, J ) 8.3 Hz), 3.87
(1H, d, J ) 14.0 Hz), 3.32-3.24 (5H, m), 3.08-3.03 (4H, m),
1.98 (3H, s) ppm; mass spectrum (CI) m/z 598 (M + H)+.
N-{3-[5-(4-Br om oben zyl)-7-(3,5-dich lor oph en yl)-5-m eth -
yl-6-oxo-6,7-d ih yd r o-5H -im id a zo[1,2-r]im id a zolyl-3-
su lfon yl]p r op yl}a ceta m id e (17i). To a solution of 17g (23
mg, 40 µmol) in CH2Cl2 (2.5 mL) was added triethylamine (7
µL, 48 µmol) followed by acetyl chloride (3 µL, 48 µmol). The
mixture was stirred at room temperature for 2 h. The reaction
was then quenched by saturated ammonium chloride. The
mixture was extracted with CH2Cl2 (3×), dried, and concen-
trated. The residue was purified by prep-TLC using 5%
MeOH-CH2Cl2 to give compound 17i (18 mg, yield 73%): 1H
NMR (CDCl3, 400 MHz) δ 7.52 (1H, s), 7.42 (2H, s), 7.36 (1H,
s), 7.30 (2H, d, J ) 8 Hz), 6.84 (2H, d, J ) 8 Hz), 5.73 (1H, br
s), 3.89 (1H, d, J ) 14 Hz), 3.50-3.20 (5H, m), 2.20-2.05 (2H,
m), 2.05 (3H, s) ppm; mass spectrum (ES+) m/z 615 (M + H)+.
Syn th esis of Su lfon a m id es 18a -f (Ta bles 5). Gen er a l
P r oced u r e III. To a stirred solution of N-chlorosuccinimide
(1.5 equiv) in THF (10 mL) was added (R)-16 (1.0 equiv) in
portions at 0 °C. After the addition, the cooling bath was
removed and the mixture was stirred at room temperature for
5 min. An amine R′R′′NH (2-5 equiv) was added next. The
reaction mixture was stirred for 1 h and quenched with
saturated ammonium chloride at 0 °C. The mixture was
extracted with EtOAc, washed successively with brine, dried
over MgSO4, and concentrated. The residue was purified by
silica gel chromatography to afford the sulfonamides 18a -f.
3-(4-Br om oben zyl)-1-(3,5-d ich lor op h en yl)-3-m eth yl-5-
(p ip e r id in yl-1-su lfon yl)-1H -im id a zo[1,2-r]im id a zol-2-
on e (18a ). Compound 18a was prepared according to general
Ack n ow led gm en t. The authors thank Dr. Xi-Ping
Su and Denis Byrne from the Chemical Development
Department for providing large quantities of intermedi-
ate 11. Analytical support from Scott Campbell and Dr.
Scott Leonard, suggestion by Dr. Larry Nammy to use
SO2 for introducing the sulfonyl moiety, and manuscript
proofreading by Dr. Matthew Netherton are specially
acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Synthetic schemes,
experimental procedures, and spectral data for compounds 4
and 5. This material is available free of charge via the Internet
at http://pubs.acs.org.
Refer en ces
(1) (a) Springer, T. A. Adhesion receptors of the immune system.
Nature 1990, 346, 425-434. (b) Gahmberg, C. G. Leukocyte
adhesion: CD11/CD18 integrins and intercellular adhesion
molecules. Curr. Opin. Cell Biol. 1997, 9, 643-650. (c) Vazeux,
R.; Hoffman, P.; Tomfta, J .; Dkkinson, E.; J asman, R.; St. J ohn,
T. and Gallatin, W. Cloning and characterization of a new
intercellular adhesion molecule ICAM-R. Nature 1992, 360,
485-488.
(2) (a) Fischer, A.; Friedrich, W.; Fasth, A.; Blanche, S.; Le Deist,
F.; Girault, D.; Veber, F.; Vossen, J .; Lopez, M.; Griscelli, C.
Reduction of graft failure by a monoclonal antibody (anti-LFA-1
CD11a) after HLA nonidentical bone marrow transplantation
in children with immunodeficiencies, osteopetrosis, and Fanco-
1
procedure III, using piperidine as the amine: yield 46%; H
NMR (CDCl3, 400 MHz) δ 7.44 (2H, s), 7.36 (2H, s), 7.30 (2H,

5366 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 22
Wu et al.
ni’s anemia: A European Group for Immunodeficiency/European
Group for Bone Marrow Transplantation report. Blood 1991, 77,
249-256. (b) Hourmant, M.; Le Mauff, B.; Le Meur, Y.; Dantal,
J .; Cantarovich, D.; Giral, M.; Caudrelier, P.; Albericci, G.;
Soulillou, J . Administration of an anti-CD11a monoclonal an-
tibody in recipients of kidney transplantation: A pilot study.
Transplantation 1994, 58, 377-380. (c) Gottlieb, A.; Krueger,
J .; Bright, R.; Ling, M.; Lebwohl, M.; Kang, S.; Feldman, S.;
Spellman, M.; Wittkowski, K.; Ochs, H. D.; J ardieu, P.; Bauer,
R.; White, M.; Dedrick, R.; Garavoy, M. Effects of administration
of a single dose of a humanized monoclonal antibody to CD11a
on the immunobiology and clinical activity of psoriasis. J . Am.
Acad. Dermatol. 2000, 42, 428-435. (d) Sanfilippo, P. J .; J etter,
M. C.; Cordova, R.; Noe, R. A.; Chourmouzis, E.; Lau, C. Y. and
Wang, E. Novel thiazole based heterocycles as inhibitors of LFA-
1/ICAM-1 mediated cell adhesion. J . Med. Chem. 1995, 38,
1057-1059. (e) Efalizumab (Raptiva), a humanized anti-LFA-1
monoclonal antibody, was approved for the treatment of chronic
moderate-to-severe psoriasis in October 2003.
(9) (a) The chemical synthesis of compounds 4 and 5 are provided
as Supporting Information. (b) Compound 15b was synthesized
as follows:
(3) (a) Kelly, T.; J eanfavre, D.; McNeil, D.; Woska, J .; Reilly, P.;
Mainolfi, E.; Kishimoto, K.; Nabozny, G.; Zinter, R.; Bormann,
B.-J .; Rothlein, R. Cutting Edge: A Small Molecule Antagonist
of LFA-1-Mediated Cell Adhesion. J . Immunol. 1999, 163, 5173-
5177. (b) Last-Barney, K.; Davidson, W.; Cardozo, M.; Frye, L.;
Grygon, C.; Hopkins, J .; J eanfavre, D.; Pav, S.; Qian, C.-g.;
Stevenson, J .; Tong, L.; Zindell, R.; Kelly, T. Binding site
elucidation of hydantoin-based antagonists of LFA-1 using
multidisciplinary technologies: Evidence for the allosteric in-
hibition of a protein-protein interaction. J . Am. Chem. Soc.
2001, 123, 5643-5650.
(4) (a) Kelly, T.; et al. To be published. (b) As will be discussed in
ref 4a, the 3,5-dichlorophenyl substituent at N1 is critical for
the activity. This precludes the option to improve the solubility
by replacing the N1-dichlorophenyl group. Alternative ap-
proaches such as incorporating solubilizing groups therefore are
especially relevant in the current work.
(5) The LFA-1/ICAM-1 binding assay, J Y cell aggregation assay,
and human whole blood SEB/IL-2 assay were performed as
previously described; see ref 3a.
(10) The synthesis of the pure enantiomers of amino acid 6 has been
previously reported. Yi, N. K. Self-regeneration of stereo-
centers: A practical enantiospecific synthesis of LFA-1 antago-
nist BIRT377. Org. Lett. 2000, 2, 2781-2783.
(11) Boymond, L.; Rottlander, M.; Cahiez, G.; Knochel, P. Preparation
of highly functionalized Grignard reagents by an iodine-
magnesium exchange reaction and its application in solid-phase
synthesis. Angew. Chem., Int. Ed. 1998, 37, 1701-1703.
(12) The sulfinate 16 can be prepared in large quantity (procedure
provided in Experimental Section) and stored in refrigerator for
months without noticeable degradation.
(6) Compounds 4 and 5 were hydrolyzed to the corresponding ring-
opened acids after 24 h at pH 7.4.
(13) Rothlein, R.; Springer, T. A. The requirement for lymphocyte
function-associated antigen 1 in homotypic leukocyte adhesion
stimulated by phorbol ester. J . Exp. Med. 1986, 163, 1132-1149.
(14) The initial impression from Table 3 was that the OH groups in
17b or 17c acted as H-bond donors in a H-bonding interaction.
This was proved incorrect by subsequent SAR data.
(15) The fact that the amino derivatives 17g and 17h were less potent
than the methanesulfonyl derivative (R)-15g (50 nM) seemed
to suggest the presence of positive charges on the protein near
the sulfonyl binding site.
(16) (a) A H-bonding interaction where an oxygen atom functioning
as a H-bond acceptor cannot be ruled out. (b) There were
suggestions that the apparent increase in potency could be the
entropic effect due to solvent exclusion. However, the fact that
sulfonyl groups similar in size had different effects (compare
17g,h vs 17b,d ,e,i; 18a vs 18b,d ,e) and that the interaction has
orientational preference (Table 5) argue against this suggestion.
(17) The protein shift assay is a modified binding assay in which 14%
of human serum, corresponding to 1% of plasma protein, was
added during incubation. The experiment details are otherwise
identical to that described in ref 3a.
(7) The SAR of the 6,5-bicyclic structure 2 will be the topic of a
separate publication.
(8) The details of the chemical synthesis outlined herein will be
published in separate papers. Wu, J .-P.; et al. Manuscripts in
preparation.
J M049657B