Discovery of 4-[(2S)-2-{[4-(4-Chlorophenoxy)phenoxy]methyl}-1-pyrrolidinyl] butanoic acid (DG-051) as a novel leukotriene A4 hydrolase inhibitor of leukotriene B4 biosynthesis

Article abstract of DOI:10.1021/jm900838g

Both in-house human genetic and literature data have converged on the identification of leukotriene 4 hydrolase (LTA₄H) as a key target for the treatment of cardiovascular disease. We combined fragmentbased crystallography screening with an iterative medicinal chemistry effort to optimize inhibitors of LTA₄H. Ligand efficiency was followed throughout our structure-activity studies. As applied within the context of LTA₄H inhibitor design, the chemistry team was able to design a potent compound 20 (DG-051) (K_d=26 nM) with high aqueous solubility (>30 mg/mL) and high oral bioavailability (>80% across species) that is currently undergoing clinical evaluation for the treatment of myocardial infarction and stroke. The structural biology-chemistry interaction described in this paper provides a sound alternative to conventional screening techniques. This is the first example of a gene-to-clinic paradigm enabled by a fragment-based drug discovery effort.

Full text of DOI:10.1021/jm900838g

J. Med. Chem. 2010, 53, 573–585 573
DOI: 10.1021/jm900838g
Discovery of 4-[(2S)-2-{[4-(4-Chlorophenoxy)phenoxy]methyl}-1-pyrrolidinyl]butanoic Acid (DG-051)
as a Novel Leukotriene A4 Hydrolase Inhibitor of Leukotriene B4 Biosynthesis†
Vincent Sandanayaka,‡ Bjorn Mamat,§ Rama K. Mishra,‡ Jennifer Winger,‡ Michael Krohn,‡ Li-Ming Zhou,‡
Monica Keyvan,‡ Livia Enache,‡ David Sullins,‡ Emmanuel Onua,‡ Jun Zhang,‡ Gudrun Halldorsdottir,§
Heida Sigthorsdottir,§ Audur Thorlaksdottir,§ Gudmundur Sigthorsson,§ Margret Thorsteinnsdottir,§ Douglas R. Davies,
Lance J. Stewart, David E. Zembower,‡ Thorkell Andresson,§ Alex S. Kiselyov,‡ Jasbir Singh,‡ and Mark E. Gurney*,‡,§,
‡Medicinal Chemistry, deCODE Chemistry, Inc., 2501 Davey Road, Woodridge, Illinois 60517, §Drug Discovery, deCODE Genetics, Sturlgata 8,
IS-101, Reykjavik, Iceland, and deCODE Biostructures, Inc., 7869 NE Day Road West, Bainbridge Island, Washington 98110
Received June 9, 2009
Both in-house human genetic and literature data have converged on the identification of leukotriene 4
hydrolase (LTA4H) as a key target for the treatment of cardiovascular disease. We combined fragment-
based crystallography screening with an iterative medicinal chemistry effort to optimize inhibitors of
LTA4H. Ligand efficiency was followed throughout our structure-activity studies. As applied within
the context of LTA4H inhibitor design, the chemistry team was able to design a potent compound 20
(DG-051) (Kd = 26 nM) with high aqueous solubility (>30 mg/mL) and high oral bioavailability
(>80% across species) that is currently undergoing clinical evaluation for the treatment of myocardial
infarction and stroke. The structural biology-chemistry interaction described in this paper provides a
sound alternative to conventional screening techniques. This is the first example of a gene-to-clinic
paradigm enabled by a fragment-based drug discovery effort.
Introduction
panyl)phenyl)propyl)sulfanyl)methyl)cyclopropyl)acetic acid,
montelukast,9 and cyclopentyl(3-(2-methoxy-4-(((2-methyl-
phenyl)sulfonyl)carbamoyl)benzyl)-1-methyl-1H-indol-5-yl)-
carbamate, zafirlukast (ICI 204,219).10 LTA4H is a key
enzyme in the leukotriene pathway, which catalyzes the final
and rate-determining step in the synthesis of LTB4.11 Earlier
studies12 of the potential role of LTA4H in disease have
focused on allergy,13 dermatitis,14 and arthritis15 with a
variety of inhibitors16 reported in the literature (Figure 1).
Several fragment-based drug discovery approaches for
screening low molecular weight (<300 Da) compounds have
been reported in the literature, and these include X-ray
crystallography, nuclear magnetic resonance (NMR), surface
plasmon resonance, differential thermal denaturation, fluore-
scence polarization, and other techniques.17-20 In our struc-
ture-based approach, we used X-ray crystallography for both
identification of hits and their subsequent structure-based
evolution into advanced molecules. In our effort to maintain
drug-like characteristics of the optimized molecules, in addi-
tion to obtaining their bound pose and tracking improvement
in enzyme inhibitory activity(IC50) for the intended molecular
target, wealsofollowed indices suchasBEI(binding efficiency
index), SEI (surface efficiency index), and LE (ligand
efficiency) that simultaneously track three crucial variables:
potency, polar surface area, and molecular weight.21,22 On the
basis of the analysis of over 120 marketed drugs, a range of
optimal values (SEI=18 ( 8.7, BEI=28 ( 7.9, and LE ∼ 0.4),
for these indices have been provided as a guide.23
We recently reported the identification of sequence variants
in the FLAPa and LTA4H genes that were associated with
enhanced capacity for the production of leukotriene B4
(LTB4) and increased risk for myocardial infarction (MI)
and stroke.1,2 These conditions feature inflammatory damage
to the arterial wall via the formation of atherosclerotic plaque
followed by its rupture and consequent thrombosis.3 Inflam-
matory cell infiltration, LTB4 biosynthetic enzymes, and
LTB4 production are increased in atherosclerotic plaque.4-6
Literature evidence further suggests that leukotrienes medi-
ate vascular and cellular immune responses in asthma, allergy,
and inflammatory diseases.7 Leukotriene synthesis inhibitors
and cysteinyl leukotriene receptor antagonists have shown
efficacy in the clinic, resulting in a number of launched
therapeutic products (e.g., (N-(1-benzo[b]thien-2-ylethyl)-
N-hydroxyurea), zileuton (A-64077),8 (1-((((1R)-1-(3-((E)-
2-(7-chloro-2-quinolinyl)vinyl)phenyl)-3-(2-(2-hydroxy-2-pro-
†Protein Data Bank coordinates have been deposited under the
following accession codes: compound 2 bound to LTA4H, 3FHE;
compound 5 bound to LTA4H, 3FH8; compound 8 bound to LTA4H,
3FTZ; compound 9 bound to LTA4H, 3FUL; compound 14 bound to
LTA4H, 3FH5; compound 20 bound to LTA4H, 3FH7.
*To whom correspondence should be addressed. Phone: 1-630-635-
0937. Fax: 1-630-783-4646. E-mail: mark.gurney@decode.is. Address:
Mark E. Gurney, Ph.D., Sr. VP, Drug Discovery and Development,
2501 Davey Road, Woodridge, IL 60517.
a Abbreviations: APA, aminopeptidase A; APP1, aminopeptidase P1;
APN, aminopeptidase N; AUC, area-under the curve; BEI, binding
efficiency index; BSA, bovine serum albumin; CVD, cardiovascular
diseases; FLAP, 5-lipoxygenase activating protein inhibitor; FOL,
fragments of life; LTA4H, human leukotriene A4 hydrolase; HWB,
human whole blood; SEI, surface efficiency index; LE, ligand efficiency;
LRAP, leukocyte-derived arginine aminopeptidase; PILS, puromycin-
insensitive leucyl-specific aminopeptidase; PK, pharmacokinetic.
Results and Discussion
Initially, crystals of LTA4H were grown (Supporting In-
formation for details) and soaked with our fragment library,
r
2009 American Chemical Society
Published on Web 12/01/2009
pubs.acs.org/jmc

574 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Sandanayaka et al.
evaluate linking of the two identified fragments, biaryl and
acetate. Considering both literature evidence25 and our struc-
tural insight, we endowed fragment 6 with an N-pyrrolidine
ethanol moiety (Figure 4) to result in analogue 9. In our
hands, the binding mode of 9 to LTA4H (Supporting Infor-
mation Figure SF1b) was consistent with the observed pose
for 6. However, analogue 9 contained two potentially dis-
ruptive interactions with the enzyme. These were between the
carbonyls of 9 and Trp311 and between the pyridinyl nitrogen
of 9 and the oxygen of water WAT-33, which were in close
˚
proximity (ligand CO-CO (Trp311) = 3.0 A and N-O
(WAT-33) = 3.12 A). Thus, we observed only limited im-
˚
provement in LTA4H inhibition for 9 vs 6 (IC50 =199 and
5380 μM, respectively). These limitations were addressed in
analogue 5, which did not feature any of these repulsive
interactions (Supporting Information Figure SF1c) and con-
sequently showed potent enzyme inhibition (IC50=180 nM).
The binding mode of compound 2 (Supporting Information
Figure SF1d) provided further insight into chemical design.
X-ray crystallographic data showed good complementarity
between the biaryl moiety and the hydrophobic surface of the
active site channel. We also observed an ionic interaction
of the carboxylate group with the Zn ion (Zn2þ-COO-
Figure 1. LTA4H inhibitors reported in the literature.13
Figure 2. Structures of bound biaryl/hHeteroaryl FOL fragments.
˚
distance=1.49 A). This -COO-Zn contact is contrary to
termed “fragments of life” (FOL).24 This customized set of
∼1300 lead-like small molecules20 is composed of natural
metabolites, their derivatives, and heterocyclic derivatives
that mimic protein-turn motifs (e.g., R-, β-, and γ-turns).
These compounds have high solubility and chemical features
allowing rapid synthetic elaboration. The FOL library, clus-
tered into 6-8 diverse fragments per pool, was soaked with
LTA4H crystals and hits were reconfirmed by X-ray crystallo-
graphy with a single fragment. This expedited approach
provided several diverse biaryl/heteroaryl fragments bound
to LTA4H as chemical leads for further elaboration. The
structures of three fragments are shown in Figure 2; their
binding modes at the hydrophobic L-shaped active site pocket
are shown in Figure 3. For each of the identifiedfragments, we
observed both hydrophobic packing as well as specific hydro-
gen bond interactions with either crystallographic water (ex.,
previously published data based on extended X-ray absorp-
tion fine structure (EXAFS) spectroscopy of LTA4H, which
concluded that there was a lack of interaction of carboxylate
containing inhibitors with the zinc atom of the enzyme.26 As
anticipated, molecule 5 also featured sound ligand efficiency
(0.44), although it displayed very high SEI value (54.1).
Combining the structural evidence accumulated thus far,
we explored a series of functional groups to tether the
identified biaryl fragments to the carboxylate. A number of
linkers were considered and evaluated, with a goal for the
optimized linker to capture additional interactions with
LTA4H active site residues. Representative examples of these
linkers are shown in Supporting Information Figure SF2: (a)
the pyridine analogue 10 is based on potentials for capturing
π-π, π-to-edge or π-cation interactions with Tyr267 and (b)
replacement of the ether oxygen of 5 with tertiary nitrogen,
yielding piperazine analogues 11 and 12, which were designed
to fill the available volume and to mediate interaction with
Gln134/Gln136. Structures of analogues 10-12 are shown in
Figure 4. Analogue 10 showed relatively weak inhibitory
activity (IC50=8.7 μM), while derivatives 11 and 12 displayed
better inhibitory potency (IC50 = 1.4 and 0.340 μM, res-
pectively). The bound poses of the pyridine analogue 10 and
the iodo-analogue 12 are shown in Supporting Information,
parts a and b of Figure SF3, respectively.
˚
ligand 6, N-O (water)=3.2 A), active site residues (ex. ligand
8 with CO of Pro374), or water mediated interactions with
active site residues (ex., ligand 7 with Gln136). The observed
inhibition of LTA4H hydrolysis (IC50=5.38 mM, 98 μM, and
>2 mM and for fragments 6, 7, and 8, respectively) by these
fragments prompted their further elaboration. The enzyme
assay used recombinant human LTA4H with LTA4 as the
substrate to determine IC50 values (see Experimental Section,
Biological Assays). In several instances, we found an acetate
anion recruited from the crystallization media to bind to the
catalytic zinc in the active site of LTA4H. An acetate bound to
zinc was observed in the absence (not shown) or presence of a
bound fragment in the hydrophobic pocket (Figure 3). This
mapping of two distinct binding sites by the biaryl fragments
and an acetate anion provided our initial hypothesis for the
overall trajectory of an LTA4H inhibitor to be pursued by
structure-based design. The SEI/BEI for fragments 6 and 7
were 7.6/11.3 and 7.4/21.1, respectively. A combination of
factors including active site fit, synthetic feasibility, and ligand
efficiency prompted us to select 6 for further optimization.
In the next series of experiments, we continued with our
iterative chemistry-structural biology studies. Our focus was
to (i) capture additional interactions in the hydrophilic pocket
by chemical elaboration of the biaryl fragments and (ii)
Despite a good fit into the binding pocket, moieties featu-
ring a tertiary amine group in compounds 2 and 5 showed no
direct interaction with Gln134 and Gln136 (Supporting In-
formation Figure SF1c,d). In a subsequent chemical modi-
fication, we replaced the aminoethylether chain of 5 with
either (S) or (R) prolinol to result in molecules 13 and 14
(Figure 5a). Both enantiomeric derivatives, (S)-isomer 13 and
(R)-isomer 14, provided improved activity compared to ana-
logue 5 (IC50=244 and 87 nM, respectively). Moreover, the
pyrrolidine nitrogen of 13 and 14 is positioned to exert strong
˚
˚
interactions with Gln134 (2.76 A) and Gln136 (2.83 A)
(Figure 5b,c), as evidenced from the structural studies. The
side chain of Gln134 underwent a ligand induced conforma-
tional change from its native position in the enzyme27
to accommodate the interactions with 14. This side chain

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 575
Figure 3. Bound pose of biaryl fragments at the hydrophobic pocket and acetate bound to catalytic zinc. (a) Fragment 6 with N-O (water) =
˚
˚
˚
˚
3.2 A and Zn-OCOCH3 = 1.98 A, resolution 1.8 A; (b) fragment 7 and acetate with Zn-OCOCH3 = 1.98A, resolution 2.0 A; (c) fragment 7
˚
˚
and acetate with Zn-OCOCH3 = 2.19 A, resolution 2.05 A. In each figure, acetate (labeled as ACT) is bound to active site zinc. The catalytic
Zn ion is shown in magenta and water molecules are shown as red spheres.
Figure 4. Structures of analogues 9-12.
rotamer led to the formation of a new hydrogen bond between
the amide nitrogen of Gln134 and the peptidyl oxygens of
˚
Gln136 (2.82 A) and Gln134 (intraresidue hydrogen bond,
˚
3.04 A) (Figure 5b). The X-ray cocrystal structure of the
corresponding (S)-enantiomer 13 revealed a weaker interac-
˚
tion with Gln136 (3.35 A), whereas Gln134 retained its
original orientation. In support of the observed 3-fold differ-
ence in potency, cocrystal structures of13lacked the hydrogen
bond network observed for its enantiomer 14 (Figure 5c). For
both ligands, the methyleneoxy chain between the biphenyl
group and the pyrrolidine ring afforded optimal positioning
of the ligand within the LTA4H binding pocket (Figure 5b,c).
Of the evaluated linkers, the prolinol-derived ligand impacted
Figure 5. (a) Shows the chemical structures for the (S) and (R)
prolinol-derived analogues 13 and 14, respectively, and (b) and (c)
shows bound orientations of (S) and (R) prolinol-derived analogues
13 and 14, respectively. (b) Hydrogen bond network involving
Gln134 and Gln136 in the complex with 14. (c) Identical view for
the corresponding (S) enantiomer, 13. Note: in the bound poses of
these two enantiomers, Gln134 has different orientation, with
orientation of G1n134 with 14 being identical to that in the
apoenzyme (not shown).

576 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Sandanayaka et al.
Table 1. Biological and Pharmacological Data for Selected Compounds
compd
14
17
20
IC50 (nM) (enzyme)
Kd (nM)
87
30
47
25
37
ND
449
ND
533
IC50 (nM) (HWB)†
pKa
microsome stabilityc
10.41
10.5
10.39
86.6
4.02, 8.35 (pI = 6.18)b
a
95.2
PK parametersd
species
rat
mouse
rat
10
dog
10
monkey
dose (mg/kg)
1/2 (h)e
10
0.67
10
10
t
3.59
1.02
2362
85
2.57
4.8
4.02
19.3
5.54
18.9
52410
186
C
max (μM)f
AUC (h ng/mL)
1.46 (iv)
471 (iv)
ND
12968 56045
3
%Fg
87
252
1564
7.6
Vd (mL/kg) (iv)h
Cl (mL/min/kg) (iv)I 71.0
6121
6324
61.0
2351
11.0
2073
6.0
a pKa values are calculated using ACD laboratories software for 14
and 17, and experimental values are reported for compound 20 (pKa1
=
4.02, pKa2 = 8.35). b Isoelectric constant. c Microsome stability indicates
parent remaining after 30 min incubation in human liver microsomes.
d Unless otherwise indicated, pharmacokinetic (PK) values are for oral
administration (po). e The plasma elimination half-life, t1/2, was deter-
mined by linear regression of the terminal phase of the logarithm plasma
concentration-time curve. f The maximum plasma concentration
(Cmax) is the observed peak plasma concentration after an oral dose.
g Oral bioavailability (F%) was determined using AUC (area under the
plasma concentration-time curve from time zero to infinity normalized
for dose) in po, (except when indicated for iv) h Volume of distribution.
I Predicted clearance. ND = not determined.
were found to be cytotoxic and relatively nonspecific in a
broad panel screen (Cerep) for off-target activity.28 Replace-
ment of the heterocycle with a p-chlorine substituent to yield
16 addressed the issue of cellular toxicity. As shown by
structural studies, the para-Cl group did not displace the
bound water molecules. Values for the three ligand efficiency
indices were: LE = 0.21 for compound 6 and 0.24 for
compound 10 vs LE=0.49 for compound 14, and SEI and
BEI = 7.6 and 11.3 for compound 6 vs 33.1 and 26.4 for
compound 14, respectively. Thus, the structure-guided ela-
boration of fragment 6 to compound 14 was on track for
simultaneously enhancing key ligand parameters (Supporting
Information Table ST1 and Figure SF4).
As reported above, we found that 14 inhibited LTA4H with
an IC50 of 87 nM. In a physiologically relevant secondary
assay, human whole blood (HWB) was used to assess inhibi-
tion (IC50) of LTB4 production after stimulation with Ca2þ
ionophore (see Experimental Section, Biological Assays). In
this assay, 14 showed an IC50 of 449 nM, a 5-fold decrease
compared to the in vitro data. The cocrystal structure of 14
bound to the enzyme is shown in Figure 6b.
Figure 6. Displacement of “structural” water molecule with para-
30-thienyl group of ligand 15. (a) X-ray cocrystal structure of
˚
LTA4H in complex with 14 at 1.63 A resolution showing three
crystallographic waters in the hydrophobic pocket and (b) X-ray
˚
cocrystal structure of 15 at 1.69 A resolution showing only two
crystallographic water molecules in the hydrophobic pocket. The
thiophene has displaced the center water molecule present in (a).
The catalytic Zn ion is shown in magenta and water molecules are
shown as red spheres.
the Gln134 side chain movement the most, securing it as a
tether of choice for subsequent ligand elaboration.
We attempted further optimization by varying the nature of
the biaryl group. Considering the hydrophobic nature of the
pocket occupied by this pharmacophore, we incorporated
lipophilic substituents onto the terminal phenyl ring. Substi-
tution at the para-position was projected to displace one or
more of the three conserved water molecules deep within the
active site pocket (Figure 6a), thus potentially improving
affinity of the molecule toward LTA4H and blocking the
anticipated cytochrome P450 (CYP) mediated metabolism
at this site. We found that relatively large groups including
thiophenes and thiazoles did in fact displace the central water
molecule (WAT-762) as in the cocrystal structure of 14
(Figure 6b), and these modifications provided potent LTA4H
inhibitors. Compound 15 showed a 21-fold improvement in
enzyme inhibitory activity (IC50=4 nM) vs the unsubstituted
analogue 14 (IC50=87 nM). Unfortunately, these derivatives
In an attempt to further optimize compound 14, we re-
placed the methylene bridge with an ether moiety providing
compound 17 with a 2-fold increase in enzyme inhibitory
potency (IC50=49 nM) and HWB assay (IC50=284 nM). The
X-ray analysis confirmed an essentially identical binding
mode for 17. This modification also afforded improvement
in key PK parameters, such as half-life and systemic exposure
(for 14 t1/2=0.67 h; AUC0-Inf=471 h ng/mLand for 17: t1/2=
3.6 h; AUCInf=2362 h ng/mL), presumably due to eliminat-
3
3
ing metabolism at the methylene site (Table 1).
Focusing on the second binding element revealed by frag-
ment screening, the acetate ion bound to the catalytic zinc
(Figure 3), we optimized the carboxylic acid pharmacophore
by introducing carbon chains of variable length between the

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 577
Figure 7. Advanced intermediate and the clinical candidate compound.
(WAT-545 in the 20 structure) of the solvent cluster (Figure
8). Compound 20 was a potent inhibitor in the enzyme assay
with an IC50=47 nM. Although the molecule gained binding
affinity through interaction with Zn2þ, in contrast to 14 it did
not participate in hydrogen-bond interactions with Gln136 or
Gln134, although the tertiary nitrogen (of 20) appears to show
water mediated interaction with Glu271. Inhibitor 20 provided
over 4-log improvement in activity compared with the biaryl
compound 6 that was identified through crystallographic
screening of our in-house fragment library.
Although useful for screening and rank ordering of com-
pounds for inhibitory potency, the in vitro enzyme assay was
formatted as an end-point assay. It required relatively high
concentrations of both the enzyme and bovine serum albumin
(BSA) in the reaction mixture (see Experimental Section, In
Vitro Enzyme Assay). Therefore, we used isothermal micro-
calorimetry to determine the equilibrium dissociation con-
stant, Kd, of 20 binding to LTA4H. The measured value of 25
nM compared favorably against both enzymatic and whole
blood assay data, suggesting lack of significant plasma bind-
ing and high cellular permeability for the molecule.
The introduction of the carboxylic acid group into 20
allowed for the optimization of other compound properties.
Specifically, aqueous solubility at the isoelectric point (pH ∼
6.18) was high (>30 mg/mL), streamlining formulation and
oral administration protocols for 20. The molecule showed
good absorption and low clearance across all species tested
with very high oral bioavailability (Table 1). The half-life in
dog and monkey predicted a similar value in human, sufficient
to allow once daily dosing. Furthermore, for compound 20,
we were able to lower the volume of distribution (Vd ∼2000
mL/kg) compared to early compounds (14 and 17), resulting
in higher plasma concentration. Considering our target indi-
cation, it is beneficial to have more drug in the systemic
circulation than in the tissues, as blood cells and the vessel
wall are the main target compartments in cardiovascular
diseases. Very high values of Cmax and AUC based on dose
across all species indicated the potential for good in vivo
potency. It is worth noting that compound 20 both displayed
excellent PK profile across various species as well as good
physicochemical properties. This molecule also featured ex-
cellent ligand efficiency [(LE = 0.38), BEI (18.8), and SEI
(12.4)], values similar to several successful marketed drugs.20
Thus, crystallography driven medicinal chemistry optimiza-
tion provided compound 20 with enhanced potency against its
molecular target (LTA4H).
Figure 8. X-ray cocrystal structure of 20 in complex with LTA4H
showing the displacement of two water molecules located near Zn
by the carboxylic acid group. The Fo - Fc electron density map is
contoured at 2σ.
pyrrolidine nitrogen and the carboxylate. Both two- and
three-carbon chains containing carboxylic acid derivatives
(18 and 19, Figure 7) showed comparable activities in the
enzyme assay and were potent in the HWB assay with IC50
values of 23 and 33 nM, respectively. A two-carbon linker
previously was reported in compound 2 to undergo metabolic
N-oxidation followed by Cope reaction to furnish acrylic acid
and N-hydroxyl amine leading to toxicity in vivo.29 On the
basis of our concern that 18 may undergo similar metabolic
conversion, the butyric acid derivative 19 featuring a three-
carbon linker was selected for further evaluation.
Given the relatively high cost of D-proline, the correspond-
ing (S)-enantiomer derived from L-proline was prepared
for comparison (19 and 20, respectively). These molecules
were equipotent in the HWB assay (IC50 = 33 and 37 nM,
respectively). Therefore, the synthetically more feasible
(S)-enantiomer 20 was identified as the lead compound for
further evaluation. The X-ray crystal structure of 20 bound to
LTA4H showed that the carboxylate displaced two of the three
water molecules near Zn2þ by chelating to the metal. One
oxygen of the carboxylate group was a direct ligand of Zn2þ
adopting a slightly distorted tetrahedral coordination geome-
try with His295, His299, and Glu318, while the second oxygen
interacts with Glu296 and the remaining water molecule
Off-Target Activity Profiling. The activity and selectivity
of 20 was assessed in a variety of in vitro assays. LTA4H is a

578 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Sandanayaka et al.
Scheme 1a
a (a) PPh3, DIAD, THF, 70 °C, 27%.
Scheme 2a
a (a) (nBu)3P, Pd(dba)2, Na-OtBu, toluene, reflux, BOC-piperazine, 25-75%; (b) 4 N HCl/p-dioxane, 95-98%.
Scheme 3
(a) Py/TsCl, 91-99%; (b) NaH, DMF, (21); (c) 4 N HCl/p-dioxane, 73%.
dual function enzyme with both hydrolase and aminopepti-
dase activities. Substrates for the two reactions occupy
overlapping regions of the active site and both reactions
are catalyzed by the zinc ion. Compound 20 was a potent
inhibitor of LTA4H aminopeptidase activity against L-ala-
nine p-nitroanilide (IC50 = 72 nM). In assessing potential
liability related to off-target activities, compound 20 was
screened in vitro against a panel of more than 50 targets
including aminopeptidases (APN, LRAP, PILS, APP1, and
APA), ion channels, human cardiac ERG channel, G-pro-
tein coupled receptors, and nuclear receptors known to
mediate side effects in the clinical setting. No significant
inhibition (defined as >50% at 10 μM) was observed for any
of the targets tested. We also assessed the potential liability
of 20 for drug-drug interaction through inhibition of cyto-
chrome P450 isozymes. The ability of 20 to interfere with in
vitro metabolic activity of major human CYPs was deter-
mined using heterologously expressed human CYP 1A2,
2C9, 2C19, 2D6, and 3A4.30 No inhibition greater than
50% was observed with any of the above CYPs against their
substrates using 20 at 10 μM concentration. These data,
along with low overall metabolism of 20 and the low
predicted plasma exposures to reach an IC90 for LTA4H
inhibition suggested that clinically relevant inhibition of
major human CYPs is unlikely.
Chemistry. Compound 9 was prepared as reported pre-
viously.21 Compound 10 was prepared by Mitsunobu reac-
tion31 of 4-benzylphenol and 4-pyridyl ethanol (Scheme 1).
Synthesis of the piperazine derivatives 11 and 12 was accom-
plished by reaction of mono-BOC protected piperazine
under Buchwald conditions,32 (using Bu3P, Pd(dba)2, and
NaOtBu as a base in refluxing toluene) from 1-iodo-4-
phenoxy-benzene (23) and 1-iodo-4-(4-iodophenoxy) ben-
zene (24), respectively, to provide 25 and 26. Subsequent
deprotection of the N-BOC group under acidic conditions
provided the desired compounds 11 and 12 (Scheme 2).
Synthesis of prolinol derived analogues 13 and 14 was
accomplished by the nucleophilic displacement of the tosyl
group of 29 and 30, respectively, with the phenolate derived
from 4-benzylphenol 21 (Scheme 3), followed by removal
of the BOC protecting group. Tosylates 29 and 30 were
prepared following the literature procedure33 from (S)- and
(R)-BOC-prolinol, respectively (Scheme 3).
The Friedel-Crafts acylation of anisole with benzoyl
chlorides 31a/31b followed by reduction of the ketones
(32a/32b) with triethylsilane/trifluoroacetic acid provided
phenyl methyl ethers 33a/33b. Removal of methyl ethers
33a/33b with boron tribromide in CH2Cl2 provided the
phenols 34a/34b, respectively. Suzuki-Miyaura coupling34
of 34b with thiophene-3-boronic acid using Pd/C provided
phenol 35. The sodium phenolate of 35, generated in situ
with NaH in DMF, was subsequently reacted with tosylate
30. Deprotection of the N-BOC intermediate 36 under acidic
conditions provided the compound 15. Similarly, phenol 34a
was elaborated to provide the BOC protected amine 37.
Deprotection of the BOC with HCl in p-dioxane, followed by
Compound 20 was profiled for activity in a number of in
vivo models of inflammation using reduction in LTB4 pro-
duction as a biomarker.2,3 These studies will be reported in
subsequent publications elsewhere. Compound 20 currently
is in human phase II clinical trials for the prevention of
myocardial infarction and stroke.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 579
Scheme 4a
a (a) AlCl3, nitrobenzene, 90%; (b) Et3SiH, TFA, 99%; (c) BBr3, CH2Cl2; (d) 3-thiophene-boronic acid, K2CO3, Pd/C, iPrOH/H2O, 99%; (e) NaH,
DMF, (30); (f) 4 N HCl/p-dioxane, 53% (from 35 to 15); (g) K2CO3, DMF, Br(CH2)3CO2CH3, rt, 35%; (h) NaOH, MeOH, 87%.
Scheme 5a
a (a) KO-tBu, DMF, (29) or (30), 92%; (b) N,N-dimethylglycine, Cs2CO3, Cu(I), 4-chlorophenol, 56%; (c) 4 N HCl, p-dioxane, 83%; (d) K2CO3,
DMF, (e) aq NaOH, MeOH, 36%; (f) HCl/Et2O, 93%; (g) (46),1:1 conc HCl/p-dioxane, 60 °C, 4 h, 68%.
alkylation of the secondary amine with methyl-4-bromo-
butyrate and subsequent base hydrolysis of the ester (38),
provided the analogue 16 (Scheme 4).
basic conditions provided the corresponding esters
(46-48). The hydrolysis of the esters under basic conditions
followed by acidification or under acid hydrolysis conditions
provided the desired analogues 19-20.
The general synthesis of the diphenyl ether derivatives
17-20 is shown in Scheme 5. Reaction of 4-iodophenol with
BOC-tosylates 29 or 30, using Cu(I) mediated Ullmann
coupling35 of the iodo-phenyl intermediate 39/40 with
4-chlorophenol, provided the corresponding diphenyl ether
derivatives 41/42. Deprotection of the N-BOC derivatives
under acidic conditions provided the amine derivatives 17
((R)-enantiomer) and 43 ((S)-enantiomer) as hydrochloride
salts. Alkylation of the secondary amine (17 or 43) with a
4-bromobuterate (45) or with 3-bromopropionate under
Conclusion
In summary, the development of 20 illustrates a number of
trends in modern drug discovery. These include (a) the use of
human population genetic studies to identify new therapeutic
targets for major common diseases, (b) the utility of a frag-
ment-based screening for the identification of novel chemo-
types empowering optimization studies, and (c) iterative
structural biology-medicinal chemistry process leading to

580 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Sandanayaka et al.
accelerated lead optimization. As a result of the structure-
assisted SAR and simultaneous optimization of pharmaco-
kinetics, pharmacodynamics, selectivity, and solubility along
with the minimized potential for toxicity, we have identified
compound 20 as the clinical candidate. Specifically, our
genetic studies linked LTA4H genes that were associated with
increased risk for myocardial infarction (MI) and stroke.1,2
To the best of our knowledge, this is the first example of a
gene-to-clinic paradigm enabled by the iterative use of frag-
ment-based drug discovery and structure-guided medicinal
chemistry, providing a candidate optimized for ligand para-
meters and imparting desirable drug characteristics. DG-051
recently entered human phase II clinical trials in CVD.
(d, J=8.9 Hz, 2 H) 6.95-7.07 (m, 4 H) 6.75 (d, J=8.9 Hz, 2 H)
3.30-3.36 (m, 4 H) 3.22 (br s, 4 H).
1-[4-(4-Iodophenoxy)phenyl]piperazine Hydrochloride (12). A
solution of 4N HCl in dioxane (15 mL) and 12b (1.1 g, 2.3 mmol)
were stirred at rt for 1 h. EtOAc (30 mL) and hexanes (30 mL)
were added and solid was collected by filtration and washed with
EtOAc (2 ꢀ 15 mL) and the solid was dried under vacuo to
provide the title compound 12 (1.11 g, 95%) as an off-white
solid. MS m/z: 381 (M þ H). LCMS: 98%. 1H NMR (400 MHz,
DMSO-d6) δ: 9.29 (br. s., 1 H) 7.27-7.41 (m, 2 H) 7.01
-7.11 (m, 3 H) 6.89-6.99 (m, 4 H) 3.30-3.36 (m, 4 H)
3.18-3.25 (m, 4 H).
(2S)-2-[(4-Benzylphenoxy)methyl]pyrrolidine (13). To a solu-
tion of 4-benzylphenol, 21 (0.10 g, 0.56 mmol) in anhydrous
DMF (1 mL) at 0 °C was added a 60% dispersion of NaH in
mineral oil (0.02 g, 0.75 mmol) portionwise over 5 min. The
resulting slurry was stirred at 0 °C for 10 min before a solution of
(S)-tosylate 29 (0.20 g, 0.56 mmol) in DMF (2 mL) was added
dropwise over 5 min. The mixture was stirred at 95 °C for 8 h and
then stirred at rt for 16 h. The reaction mixture was poured over
ice and then concentrated under reduced pressure. The residue
was extracted into EtOAc (20 mL) and sequentially washed with
water (10 mL), saturated aqueous NaHCO3 (2 ꢀ 10 mL), water
(2 ꢀ 10 mL), and brine (2 ꢀ 10 mL). The combined organic layer
was dried over anhydrous Na2SO4, filtered, and the solvent was
removed in vacuo. The residue was chromatographed on silica
gel (10 g) and eluted with 30% EtOAc/haxanes to afford
(S)-2-(4-benzylphenoxymethyl)-1-BOC-pyrrolidine (0.15 g,
73%) as a tan oil. To a solution of this product (0.12 g, 3.26
mmol) was added 4 M HCl in dioxane (15 mL) at rt . The
resulting mixture was stirred overnight. The solvent was
removed in vacuo to obtain an off-white solid. The solid was
triturated with Et2O to provide compound 13 (55 mg, 63%) as a
white solid. MS m/z: 268 (M þ H). 1H NMR (400 MHz, DMSO-
d6) δ: 7.24-7.33 (m, 2H), 7.13-7.23 (m, 5H), 6.91 (d, J=8.6 Hz,
2H), 4.16-4.26 (m, 1H), 4.09 (dd, J = 8.3, 10.47 Hz, 1H),
3.80-3.94 (m, 3H), 3.12-3.25 (m, 2H), 2.05-2.17 (m, 1H),
1.82-2.03 (m, 2H), 1.63-1.80 (m, 1H). HPLC-1: 99.4%,
tR=15.1 min.
Experimental Section
Chemistry. All reagents, starting materials, and anhydrous
solvents were obtained from commercial sources and used
without further purification unless otherwise noted. Concentra-
€
tion refers to evaporation under vacuum using a Buchi rotatory
evaporator. NMR spectra were recorded at 400 or 500 MHz
(Varian Instruments) in the solvent indicated, and TMS was
used an internal reference. ACDLabs NMR software was used
to process FIDs to generate spectral parameters (ppm/Hz).
Coupling constants (J) are given in Hz. Mass spectra were
obtained using either APCI or electrospray ionization (PE-
SCIEX single-quad or Agilent mixed-mode units). Elemental
analyses were carried out by Galbraith Laboratories, Inc.
(Knoxville, TN) or Midwest Microlab, LLC (Indianapolis,
IN). Compounds were purified by flash chromatography using
˚
silica gel (MP EcoChrom, 32-63D, 60 A). Typically, 1:10 to
1:20 ratio of the crude product to silica gel was used. Purity of all
final products was determined by analytical HPLC and/or by
LC-MS to be g95%. HPLC purity of compounds were mea-
sured with a reversed-phase HPLC (Phemomenex Prodigy C18
column, 4.6 mm ꢀ 150 mm, 5 μm, typically at 254 nm, unless
indicated) with one of the following two conditions. HPLC-1:
compounds were eluted using a gradient of 90/10 to 10/90 A/B
over 40 min at a flow rate of 1.0 mL/min, where solvent A was
aqueous 0.05% TFA and solvent B was CH3CN (0.05% TFA).
HPLC-2: compounds were eluted using a gradient of 95/5 to
5/95 A/B over 40 min at a flow rate of 1.0 mL/min, where solvent
A was aqueous 0.05% TFA and solvent B was CH3CN (0.05%
TFA). For HPLC data, (final products), peak area percent and
retention time (tR in minutes) are provided.
(R)-2-(4-Benzylphenoxymethyl)pyrrolidine (14). Similar to the
preparation of compound 13, the (R)-BOC-intermediate was
prepared from 4-benzylphenol (103 mg, 0.56 mmol), NaH
(19 mg, 0.75 mmol) in 1 mL DMF and (R)-tosylate 30 (200
mg, 0.56 mmol) in DMF (3 mL) to provide after aqueous
workup a crude product which following purification by silica
gel (10 g) using 30% EtOAc/hexanes as eluent provided the
BOC-protected intermediate (150 mg, 73%) as a solid. MS m/z:
4-[2-(4-Benzylphenoxy)ethyl]pyridine (10). The 4-pyridine
ethanol (123 mg, 1.0 mmol) was dissolved in THF (3 mL) and
solutions of 4-benzylphenol (240 mg, 1.3 mmol) in THF (1 mL)
and Ph3P (315 mg, 1.2 mmol) in THF (1 mL) were added. The
reaction was cooled to 0 °C, and a solution of diisopropyldi-
azodicarboxylate (242 mg, 1.2 mmol) in THF (1 mL) was added,
followed by stirring at rt for 4 h. The reaction was heated to
70 °C and maintained for 16 h. The solvent was removed in
vacuo, and the resulting material was purified by silica gel flash
chromatography (20-50% EtOAc/hexanes, gradient) and stir-
red with 2 M HCl in Et2O (excess) at rt to give the compound 10
as the hydrochloride salt (0.89 g, 27%). MS m/z: 290 (M þ H).
1H NMR (400 MHz, DMSO-d6) δ: 8.80 (d, J=6.8 Hz, 2H),
7.96 (d, J = 6.8 Hz, 2H), 7.26 (m, 2H), 7.17 (m, 3H), 7.12
(d, J=8.8 Hz, 2H), 6.84 (d, J=8.8 Hz, 2H), 4.30 (t, J=6 Hz,
2H), 3.86 (s, 2H), 3.31 (t, J = 6.3 Hz, 2H). HPLC-1: 99.7%,
tR=19.5 min.
1-(4-Phenoxyphenyl)piperazine Hydrochloride (11). A solu-
tion of 4N HCl in dioxane (30 mL) and 10b (3.0 g, 8.46 mmol)
were stirred at rt for 1 h. EtOAc (30 mL) was added and solid
was collected by filtration and washed with EtOAc (2 ꢀ 10 mL)
and the solid was dried under vacuo to provide the title
compound 11 (2.71 g, 98.2%) as a white solid. MS m/z: 255
(M þ H). 1H NMR (400 MHz, DMSO-d6) δ: 9.11 (br s, 1 H) 7.65
1
368 (M þ H). H NMR (400 MHz, CDCl3) δ: 1.46 (m, 9H),
1.84-2.03 (m, 4H), 3.39 (m, 2H), 3.73-3.93 (m, 3H), 4.10 (m,
2H), 6.84 (m, 2H), 7.08 (m, 2H), 7.17 (m, 2H), 7.26 (m, 3H).
HPLC-1: 99.8% To a solution of the BOC-protected intermedi-
ate (0.12 g) in dioxane (2 mL) was added 4N HCl in dioxane
(8 mL) at rt and the resulting mixture was stirred for 1 h at rt.
The solvent was removed in vacuo to obtain a thick oil. The oil
was triturated with Et2O to obtain a white solid (2.5 g). The solid
was recrystallized with toluene (20 mL) to obtain the title
product (58 g, 53%) as a crystalline solid (14). MS: m/z 268
(M þ H). 1H NMR (400 MHz, DMSO-d6) δ: 7.15-7.20 (m, 5H),
7.27 (m, 2H), 6.91 (d, 2H, J=6.8 Hz), 4.19 (dd, 1H, J1=8.8, J2=
3.2 Hz), 4.13 (dd, 1H, J1=7.2 Hz, J2=6.4 Hz), 3.88 (s, 2H), 3.87
(m, 1H), 3.18 (m, 2H), 2.09 (m, 1H), 1.98 (m, 1H), 1.89 (m, 1H),
1.73 (m, 1H). HPLC-1: 98.8%, tR=15.2 min.
(R)-2-[4-(4-Thiophen-3-yl-benzyl)phenoxymethyl]pyrrolidine
(15). To an argon purged solution of 35 (9.0 g, 34 mmol) and
(R)-2-(toluene-4-sulfonyloxymethyl)pyrrolidine-1-carboxylic
acid tert-butyl ester (12.6 g, 36 mmol) in DMF at 0 °C was added
NaH (60% dispersion in mineral oil, 1.62 g, 40 mmol). The
reaction was stirred at 0 °C for 10 min and warmed to 85 °C for
16 h. The reaction mixture was poured into aqueous NaHCO3
(500 mL) and extracted with EtOAc (3 ꢀ 150 mL). The organic

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 581
layer was washed with water (3 ꢀ 50 mL) and dried over
anhydrous MgSO4. The volatiles were removed and the residue
was stirred with 4N HCl in dioxane (40 mL) for 16 h. The
mixture was diluted with THF (450 mL). The resulting solids
were filtered and rinsed with Et2O to give the compound 15 (7.0
g, 53%). MS m/z: 351 (M þ H). 1H NMR (400 MHz, DMSO-d6)
δ: 7.79 (d, J=1.61 Hz, 1H), 7.58-7.66 (m, 3H), 7.51 (dd, J=1.0,
5.0 Hz, 1H), 7.15-7.27 (m, 4H), 6.92 (d, J=8.5 Hz, 2H), 4.20 (d,
J = 3.6 Hz, 1H), 4.08 (s, 1H), 3.90 (s, 3H), 3.12-3.25 (m,
2H), 2.05-2.19 (m, 1H), 1.82-2.04 (m, 2H), 1.64-1.80 (m,
1H). HPLC-1: 99.9%, tR=21.0 min. Anal. calcd (C22H23NOS):
C, H, N.
4-{(R)-2-[4-(4-Chlorobenzyl)phenoxymethyl]pyrrolidin-1-yl}-
butyric Acid (16). The methyl ester 38 (0.5 g, 1.25 mmol) was
dissolved in 20% water in MeOH (6 mL). Aqueous NaOH (1N,
3.5 mL) was added, and the reaction mixture was heated to 50 °C
for 16 h. The solvents were removed in vacuo and the residue
dissolved in water (15 mL). The pH of the solution was adjusted
to ∼6-7 with 1N aqueous HCl and extracted with EtOAc
(100 mL). The organic layer was separated, dried over an-
hydrous MgSO4, and the solvent was removed in vacuo to give
the compound 16 (0.42 g, 87%). MS m/z: 388 (M þ H). 1H NMR
(400 MHz, DMSO-d6) δ: 7.32 (d, J=8.3 Hz, 2H), 7.22 (d, J=8.3
Hz, 2H), 7.11 (d, J = 8.5 Hz, 2H), 6.84 (d, J = 8.5 Hz, 2H),
3.82-3.90 (m, 3H), 3.72 (dd, J=6.7, 9.5 Hz, 1H), 3.00-3.09 (m,
1H), 2.74-2.90 (m, 2H), 2.36 (ddd, J=5.2, 6.8, 12.0 Hz, 1H),
2.14-2.30 (m, 3H), 1.91 (dd, J=8.3, 12.0 Hz, 1H), 1.52-1.75
(m, 5H). HPLC-1: 98.2%, tR=20.2 min.
5.45 mmol) in MeOH (22 mL) was added 2N aqueous NaOH
(7.2 mL, 14.5 mmol). The resulting solution was stirred at rt
overnight. The solvent was removed in vacuo. The crude oil was
dissolved in water (45 mL), and the pH was adjusted to 7 with
2N HCI solution. The crude residue was extracted into EtOAc
(3 ꢀ 250 mL). The combined organic portions were washed with
brine, dried over anhydrous Na2SO4, filtered, and the solvent
was removed in vacuo. The crude product was purified by silica
gel flash chromatography using (0-20% MeOH/CH2Cl2,
gradient) to give the free base of the title compound as a yellow
oil. To the oil was added 2 M HCI in Et2O (35 mL). The resulting
mixture was stirred at rt for 2 h. After removal of the super-
natant solvent, the remaining white solid was triturated in Et2O
(50 mL) for 1 h. The slurry was filtered, washed with Et2O
(3 ꢀ 25 mL), and dried at 45 °C for 48 h to afford the title
1
compound 20 (0.84 g, 36%) as a white solid. H NMR (400
MHz, DMSO-d6) δ: 1.82-2.03 (m, 5H), 2.23 (m, 1H), 2.38 (t,
J=7.2 Hz, 2H), 3.14 (m, 2H), 3.48 (m, 1H), 3.63 (m, 1H), 3.90
(m, 1H), 4.28-4.37 (m, 2H), 6.94 (d, J=8.8 Hz, 2H), 7.06 (s,
4H), 7.40 (d, J=8.8 Hz, 2H), 10.55 (br, 1H), 12.30 (br, 1H).
HPLC-1: 99.5%, tR = 19.5 min. Anal. calcd (C21H24ClNO4
HCl): C, H, N.
3
4-(4-Phenoxyphenyl)piperazine-1-carboxylic Acid tert-Butyl
Ester (25). nBu3P (0.7 g . 3.0 mmol), Pd2(dba)3 (0.36 g, 0.3
mmol), and NaOtBu (1.26 g, 13.1 mmol) were added to a
solution of 23 (3.9 g, 13.1 mmol) and 1-BOC-piperazine (2.5 g,
13.1 mmol) in toluene (20 mL). The reaction mixture was stirred
at rt overnight and then filtered. The solvent was removed in
vacuo, and the residue was purified on silica gel (70 g) using 1%
MeOH in CH2Cl2 to yield the title compound 25 (3.5 g, 75.5%)
as a light-yellow solid. MS m/z: 355 (M þ H). LCMS 99%. 1H
NMR (400 MHz, CDCl3) δ: 7.57 (d, J=8.9 Hz, 2 H) 6.89-6.98
(m, 4 H) 6.71 (d, J=8.9 Hz, 2 H) 3.55-3.62 (m, 4 H) 3.05-3.11
(m, 4 H) 1.49 (s, 9H).
4-[4-(4-Iodophenoxy)phenyl]piperazine-1-carboxylic Acid tert-
Butyl Ester (26). To a solution of 24 (4.22 g 10 mmol), 1-BOC-
piperazine (1.86 g, 10 mmol) were added Pd2(dba)3 (0.270 g,
0.295 mmol), nBu3P (0.58 g, 2.8 mmol), and NaOtBu (0.95 g, 10
mmol). The reaction mixture was stirred at rt overnight and then
filtered. The solvent was removed in vacuo, and the residue was
purified on silica gel (100 g) using 1% MeOH in CH2Cl2 to yield
the title compound 26 (1.2 g, 25%) as a yellow solid. MS m/z: 481
(M þ H), LCMS 97%. 1H NMR (400 MHz, CDCl3) δ: 7.57 (d,
J=8.9 Hz, 2 H), 6.93 (dd, J=16.0, 9.4 Hz, 4 H), 6.71 (d, J=
8.7 Hz, 2 H), 3.55-3.61 (m, 4 H), 3.05-3.12 (m, 4 H), 1.49
(s, 9H).
(S)-2-(Toluene-4-sulfonyloxymethyl)pyrrolidine-1-carboxylic
Acid tert-Butyl Ester (29). To a solution of (S)-N-BOC-prolinol
(22 g, 110 mmol) in pyridine (56 mL) at 0 °C was added a
solution of p-toluenesulfonyl chloride (22.9 g, 120 mmol) in
pyridine (56 mL) portionwise over 5 min. The pale-yellow
reaction mixture was stirred at 0 °C for 2 h and then at rt
overnight. The solvent was removed in vacuo. The crude oil was
extracted into EtOAc (400 mL) and sequentially washed with 0.5
M HCl (100 mL), saturated aqueous NaHCO3 (100 mL), and
brine (100 mL). The combined organic layer was dried over
anhydrous Na2SO4, filtered, and the solvent was removed in
vacuo to give compound 27 (39 g, 99%) as a yellow oil. MS m/z:
356 (M þ H). 1H NMR (400 MHz, CDCl3) δ: 7.79 (d, J=8.1 Hz,
2H), 7.31-7.40 (m, 2H), 4.06-4.18 (m, 2H), 3.90 (br s, 1H), 3.30
(d, J=6.3 Hz, 2H), 2.45 (s, 3H), 1.74-2.00 (m, 4H), 1.39 (d,
J=16.4 Hz, 9H).
(R)-2-(Toluene-4-sulfonyloxymethyl)pyrrolidine-1-carboxylic
Acid tert-Butyl Ester (30). This compound was prepared ana-
logously to the preparation of 29 from (R)-BOC-prolinol 28
(500 mg, 2.48 mmol) and tosyl chloride (565 mg, 2.96 mmol) in
pyridine (2.5 mL) to provide, after aqueous workup and con-
centration of organic solvent in vacuo, the tosylate 30 (800 mg,
91%) as a thick oil. MS m/z: 378 (M þ Na). 1H NMR (400 MHz,
CDCl3) δ: 1.38 (m, 9H), 1.79 (m, 2H), 1.93 (m, 2H), 2.44 (s, 3H),
(R)-2-[4-(4-Chlorophenoxy)phenoxymethyl]pyrrolidine Hy-
drochloride (17). Compound 41 was prepared from 30 analogous
to the procedure described for 40 as a yellow oil (0.37 g). H
1
NMR (400 MHz, DMSO-d6) δ: 7.38 (d, J=8.8 Hz, 2H), 7.00
(s, 4H), 6.93 (d, J=8.8 Hz, 2H), 4.02 (br s, 2H), 3.83-3.93 (m,
1H), 3.27 (br s, 2H), 1.86-2.02 (m, 3H), 1.76-1.84 (m, 1H), 1.40
(s, 9H).
A solution of 4N HCl in p-dioxane (5 mL) was added to 41
(0.36 g, 0.89 mmol). The resulting mixture was stirred at rt for
2 h. The solvent was removed in vacuo. The crude product was
triturated with Et2O and dried in vacuo to afford the compound
1
17 (0.19 g, 70%) as a light-yellow solid. H NMR (400 MHz,
DMSO-d6) δ: 1.75 (m, 1H), 1.87-2.01 (m, 2H), 2.12 (m, 1H),
3.21 (m, 2H), 3.89 (m, 1H), 4.17 (m, 1H), 4.24 (dd, J=4.0, 10.8
Hz, 1H), 6.94 (d, J=9.2 Hz, 2H), 7.05 (s, 4H), 7.40 (d, J=8.8 Hz,
2H). LCMS m/z: 304.2 (M þ l). HPLC-1: (99.1%), tR=19.2 min.
Anal. calcd (C17H18ClNO2 HCl): C, H, N.
3
3-{(R)-2-[4-(4-Chlorophenoxy)phenoxymethyl]pyrrolidin-1-yl}-
propionic Acid Hydrochloride (18). A solution of 46 (0.62 g, 1.59
mmol) in 1:1 conc. HCl/p-dioxane (26 mL) was stirred at 60 °C
for 4 h. The solvent was removed in vacuo. The crude product
was triturated with ether and the solvent was removed in vacuo
to afford the title compound 18 (0.40 g, 68%) as a white solid.
MS m/z: 374.5 (M - 1). 1H NMR (400 MHz, DMSO-d6) δ: 1.83
(m, IH), 2.00 (m, 2H), 2.24 (m, IH), 2.86 (m, 2H), 3.18 (m, IH),
3.38 (m, IH), 3.63 (m, 2H), 3.96 (m, IH), 4.28-4.37 (m, 2H), 6.94
(d, J=9.2 Hz, 2H), 7.06 (s, 4H), 7.39 (d, J=8.8 Hz, 2H). HPLC-
1: 97.7%, tR=19.5 min.
4-{(R)-2-(4-(4-Chlorophenoxy)phenoxymethyl)pyrrolidin-1-yl}-
l-butyric Acid Hydrochloride (19). A solution of the methyl ester
47 (0.8 g, 1.99 mmol) in conc. HCl (18 mL) and p-dioxane
(18 mL) was stirred at 60 °C for 4 h. The solvent was removed in
vacuo to provide an oil. The oil was triturated with Et2O
(10 mL), and the resulting solid was filtered to yield the title
compound 19 (0.72 g, 93%) as a white solid. MS m/z: 388
(M - 1). 1H NMR (400 MHz, DMSO-d6) δ: 1.82-2.03 (m, 5H),
2.24 (m, 1H), 2.38 (t, J=7.2 Hz, 2H), 3.15 (m, 2H), 3.49 (m, 1H),
3.64 (m, 1H), 3.90 (m, 1H), 4.28-4.34 (m, 2H), 6.94 (d, J=9.2
Hz, 2H), 7.06 (s, 4H), 7.40 (d, J=8.8 Hz, 2H). HPLC-2: 98.1%,
tR=18.5 min.
4-{(S)-2-[4-(4-Chlorophenoxy)phenoxymethyl]pyrrolidin-1-yl}-
butyric Acid Hydrochloride (20). To a solution of 48 (2.2 g,

582 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Sandanayaka et al.
3.26-3.32 (m, 3H), 3.88-3.97 (m, 2H), 4.07-4.14 (m, 2H), 7.34
(br, 2H), 7.77 (d, 2H, J=8.0 Hz).
isolated by filtration to give the title product 35 (10.6 g, 99%).
1H NMR (400 MHz, CDCl3) δ: 7.51 (d, J = 8.1 Hz, 2H),
7.39-7.43 (m, 1H), 7.36 (d, J=2.2 Hz, 2H), 7.19 (d, J=8.1
Hz, 2H), 7.07 (d, J=8.3 Hz, 2H), 6.76 (d, J=8.5 Hz, 2H), 4.68 (s,
1H), 3.93 (s, 2H).
(R)-2-[4-(4-Chlorobenzyl)phenoxymethyl]pyrrolidine (37).
Synthesis was carried out analogous to procedure used for the
preparation of 15 starting with 34a. BOC protected intermedi-
ate, (R)-2-[4-(4-chlorobenzyl)-phenoxymethyl]-pyrrolidine-1-
carboxylic acid tert-butyl ester, was purified by flash column
chromatography (15% EtOAc/hexanes, isocratic). 1H NMR
(400 MHz, CDCl3) δ: 7.23 (d, J=8.2 Hz, 2H), 7.09 (d, J=8.3
Hz, 2H), 7.05 (d, J = 8.1 Hz, 2H), 6.85 (d, J = 6.5 Hz, 2H),
4.04-4.21 (m, 2H), 3.87 (s, 2H), 3.68-3.81 (m, 1H), 3.27-3.46
(m, 2H), 1.78-2.03 (m, 4H), 1.46 (s, 9H).
4-{(R)-2-[4-(4-Chlorobenzyl)phenoxymethyl]pyrrolidin-1-yl}-
butyric Acid Methyl Ester (38). A solution of 37 (1.6 g, 4 mmol)
was treated with 4 N HCl in p-dioxane (25 mL) for 6 h at rt, the
solvent was removed in vacuo, and residue dried under high
vacuum for 4 h to provide solid (1.4 g) that was used as such for
the next step. To solid (1.4 g, 4 mmol) and methyl-4-bromo-
butyrate (0.9 g, 5 mmol) in DMF (15 mL) was added K2CO3 (1.1
g, 8 mmol). The reaction mixture was stirred at rt for 16 h. The
solvent was removed in vacuo, and the residue was partitioned
between water (35 mL) and EtOAc (130 mL). The organic layer
was dried over anhydrous MgSO4, the solvent was removed in
vacuo, and residue was purified by flash chromatography (50%
EtOAc/hexanes, isocratic) to give the ester 38 (0.57 g, 35%). 1H
NMR (400 MHz, CDCl3) δ: 7.25 (d, J=6.6 Hz, 2H), 7.09 (d, J=
8.3 Hz, 2H), 7.05 (d, J=8.6 Hz, 2H), 6.82 (d, J=8.6 Hz, 2H),
3.86-3.92 (m, 3H), 3.74 (dd, J=6.9, 9.1 Hz, 1H), 3.64 (s, 3H),
3.15 (d, J = 4.0 Hz, 1H), 2.79-2.92 (m, 2H), 2.30-2.47 (m,
3H), 2.24 (q, J=8.3 Hz, 1H), 1.96 (d, J=11.8 Hz, 1H), 1.63-1.88
(m, 5H).
(S)-2-(4-Iodophenoxymethyl)pyrrolidine-1-carboxylic Acid
tert-Butyl Ester (40). To a solution of 4-iodophenol (24.2 g,
110 mmol) in DMF (200 mL) was added KOtBu (12.3 g,
110 mmol). The cloudy solution was stirred at rt for 30 min
before a solution of 29 (35.9 g, 100 mmol) in DMF (200 mL) was
added. The reaction was heated to 45 °C for 16 h and then
increased to 55 °C for 8 h. The solvent was removed in vacuo,
and the residue was partitioned between hexanes (500 mL) and
water (250 mL). The organic layer was washed with 0.2 N NaOH
(2 ꢀ 200 mL) and H2O (250 mL). The solvent was reduced to
20% and allowed to stand at rt for 1 h. The supernatant liquid
was decanted, and the remaining solids were washed with
hexane (50 mL). The solid was dried under high vacuum at rt
to give the title compound 40 (37.0 g, 92%). 1H NMR (400 MHz,
DMSO-d6) δ: 7.58 (d, J=8.7 Hz, 2H), 6.82 (d, J=8.8 Hz, 2H),
3.95-4.07 (m, 2H), 3.82-3.93 (m, 1H), 3.26 (br. s., 2H),
1.71-2.03 (m, 4H), 1.39 (s, 9H).
(4-Chlorophenyl)(4-methoxyphenyl)methanone (32a). Prepa-
red according to the procedure for 32b from 4-chlorobenzoyl
1
chloride. H NMR (400 MHz, CDCl3) δ: 7.80 (d, J=8.7 Hz,
2H), 7.71 (d, J=8.5 Hz, 2H), 7.45 (d, J=8.5 Hz, 2H), 6.97 (d, J=
8.7 Hz, 2H), 3.89 (s, 3H).
(4-Iodophenyl)(4-methoxyphenyl)methanone (32b). Cold ni-
trobenzene (45 mL) was treated portionwise with AlCl3 (13.5
g, 101 mmol). 4-Iodobenzoyl chloride 31b (25 g, 94 mmol) in
nitrobenzene (20 mL) was added, keeping the reaction tempera-
ture at <10 °C. The mixture was stirred at 0 °C for 10 min, and
anisole (9.5 g, 88 mmol) was added dropwise to maintain
temperature at <10 °C. The reaction was allowed to rt over
16 h and then poured into ice water (750 mL). The resulting
solids were removed by filtration, and the filtrate was extracted
with CH2Cl2 (2 L). The organic materials were washed with
NaHCO3 (2 ꢀ 150 mL), dried over anhydrous MgSO4, and the
solvent was removed in vacuo. The residue was triturated with
cyclohexane (200 mL) to obtain 32b as a solid (26.75 g, 90%). 1H
NMR (400 MHz, CDCl3) δ: 7.84 (d, J=8.3 Hz, 2H), 7.80 (d, J=
8.9 Hz, 2H), 7.48 (d, J=8.3 Hz, 2H), 6.96 (d, J=8.9 Hz, 2H), 3.89
(s, 3H).
4-(4-Chlorobenzyl)methoxybenzene (33b). A solution of 32b
(9.8 g, 39.7 mmol) in TFA (30 mL) was cooled to 0 °C. To this
solution was added triethylsilane (19.22 g, 164.7 mmol), and the
reaction was allowed to rt for 16 h. The reaction was poured into
water (150 mL) and was adjusted to pH 6-7 with 1N NaOH.
The solution was extracted with EtOAc (400 mL), and the
organic layer was washed with water (200 mL) and then dried
over anhydrous MgSO4. The solvent was removed in vacuo to
give the title product 33b (9.2 g, 99%), which was used without
further purification for the next step. 1H NMR (400 MHz,
CDCl3) δ: 7.25 (d, J=3.9 Hz, 2H), 7.09 (d, J=8.4 Hz, 2H),
7.07 (d, J=8.4 Hz, 2H), 6.83 (d, J=8.6 Hz, 2H), 3.88 (s, 2H), 3.78
(s, 3H).
4-(4-Chlorobenzyl)phenol (34a). Prepared according to the
procedure for 34b starting with 4-(4-chlorobenzyl)methoxy-
1
benzene. H NMR (400 MHz, CDCl3) δ: 7.30 (d, J=3.4 Hz,
2H), 7.13 (d, J=8.3 Hz, 2H), 7.06 (d, J=8.5 Hz, 2H), 6.80 (d, J=
8.5 Hz, 2H), 4.74 (s, 1H), 3.91 (s, 2H).
4-(4-Iodobenzyl)phenol (34b). To a solution of 32b (26.7 g,
79 mmol) in TFA (90 mL) at 0 °C, triethylsilane (29 mL, 181
mmol) was added slowly. The reaction was stirred at rt for
16 h. The volatiles were removed in vacuo. The residue was
dissolved in EtOAc (200 mL), washed with NaHCO3 (2 ꢀ
300 mL) and 6N HCl (2 ꢀ 50 mL), and then dried over
anhydrous MgSO4. Removal of solvent in vacuo gave an oil,
which was used without further purification. 1H NMR (400
MHz, CDCl3) δ: 7.59 (d, J=8.2 Hz, 2H), 7.06 (d, J=8.6 Hz, 2H),
6.92 (d, J=8.2 Hz, 2H), 6.82 (d, J=8.6 Hz, 2H), 3.85 (s, 2H), 3.78
(s, 3H). To a -78 °C. solution of crude 33b in CH2Cl2 (150 mL)
was added BBr3 (1 M in CH2Cl2, 158 mL, 158 mmol) while
maintaining the reaction temperature below -65 °C. The reac-
tion was allowed to come to rt, then quenched by pouring over
ice water (1 L). The aqueous material was extracted with CH2Cl2
(2 ꢀ 100 mL). The organic layer was washed with NaHCO3 (2 ꢀ
2000 mL) and brine (100 mL) and then dried over anhydrous
MgSO4. The solvent was removed in vacuo, and the residue
triturated with hexanes to obtain the product 34b (22.3 g, 91%).
1H NMR (400 MHz, CDCl3) δ: 7.59 (d, J=8.2 Hz, 2H), 7.01 (d,
J=8.3 Hz, 2H), 6.91 (d, J=8.2 Hz, 2H), 6.75 (d, J=8.5 Hz, 2H),
4.68 (s, 1H), 3.84 (s, 2H).
(S)-2-[4-(4-Chlorophenoxy)phenoxymethyl]pyrrolidine-1-car-
boxylic Acid tert-Butyl Ester (42). A mixture of 40 (11.8 g, 29.3
mmol), 4-chlorophenol (11.3 g, 87.9 mmol), copper(I) iodide
(1.4 g, 7.3 mmol), cesium carbonate (38.2 g, 117.2 mmol), and N,
N-dimethylglycine (1.5 g, 14.6 mmol) in p-dioxane (200 mL) was
heated to reflux under an N2 atmosphere for 16 h. The reaction
mixture was concentrated in vacuo, and residue was partitioned
between hexanes (500 mL) and water (250 mL). The organic
layer was washed with 0.2N NaOH (2 ꢀ 250 mL) and water
(250 mL) and then concentrated in vacuo to give a crude oil. The
oil was purified by silica gel flash chromatography (10%
acetone/hexanes, isocratic) to give the title compound 42 (6.6
g, 56%). MS m/z: 404 (M þ H). 1H NMR (400 MHz, DMSO-d6)
δ: 7.38 (d, J=8.8 Hz, 2H), 7.00 (s, 4H), 6.93 (d, J=8.8 Hz, 2H),
3.96-4.09 (m, 2H), 3.83-3.94 (m, 1H), 3.27 (br. s., 2H),
1.86-2.04 (m, 3H), 1.74-1.85 (m, 1H), 1.40 (s, 9H).
4-(4-Thiophen-3-yl-benzyl)phenol (35). A mixture of 34b (12.4
g, 40 mmol), 3-thienylboronic acid (6.15 g, 48 mmol), Pd/C
(2.12 g, 2 mmol), and K2CO3 (16.6 g, 120 mmol) was suspended
in IPA/H2O (5:1, 240 mL) under an argon purge. The stirred
mixture was heated at 85 °C for 16 h. After cooling to rt, the
reaction was passed through a pad of celite and concentrated in
vacuo. The residue was triturated with water and the solids were
(S)-2-[4-(4-Chlorophenoxy)phenoxymethyl]pyrrolidine (43).
Compound 42 (6.6 g, 16.4 mmol) was stirred in 4 N HCl in
p-dioxane at rt for 2 h. The solvent was removed in vacuo, and

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 583
the residue was partitioned between hexanes (200 mL) and water
(200 mL). The organic layer was extracted with 1 N HCl
(25 mL). The aqueous layers were combined, neutralized with
Na2CO3 (solution), and extracted with hexanes (2 ꢀ 100 mL).
The solvent was removed in vacuo at 40 °C until the solution
became cloudy. The mixture was allowed to stand at rt and then
refrigerated for 1 h. The supernatant liquid was decanted, and
the resulting solids were dried under vacuum at rt to give the title
LTA4 were kept frozen at -80 °C for a maximum of one week
prior to use.
Recombinant human LTA4H (200 nM) was incubated with
various concentrations of test compound for 10 min at rt in
assay buffer (0.1 M Tris-HCl, 0.1 M NaCl, 5 mg/mL fatty acid
free BSA, 10% DMSO, pH 8.0). Immediately before the assay,
LTA4 was diluted to a concentration of 10 μM in assay buffer
(without DMSO) and added to the reaction mixture to a final
concentration of 2 μM to initiate the enzyme reaction. Samples
were incubated for 2 min at rt, followed by the addition of 2
volumes of chilled quenching buffer (CH3CN with 1% HOAc
and 225 nM LTB4-d4 (BioMol)). The samples were then stored
at 4 °C overnight to complete protein precipitation and centri-
fuged for 15 min at 1800g. LTB4 formed was measured in the
supernatant by LC-MS/MS using LTB4-d4 as an internal stan-
dard and an external LTB4 standard (BioMol) for a calibration
curve. Briefly, the analyte was separated from LTB4 isomers
formed by spontaneous hydrolysis of LTA4 using isocratic
elution on an HPLC system (Waters, Milford, MA) and ana-
lyzed on a tandem quadrupole mass spectrometer (Waters
Micromass Quattro Premier). MRM transitions followed
on 2 channels were 335.2 f 195.3 (LTB4) and 339.2 f 197.3
(LTB4-d4). On the basis of the measured amounts of LTB4
formed at each inhibitor concentration, a dose-response curve
was fitted to the data using a sigmoidal 4-parameter function
(XLfit, model 205) and an IC50 value was calculated.
Human Whole Blood LTB4 Assay. Human blood (45 mL) was
collected in heparin-containing Vacutainer tubes (Greiner-Bio
One) with informed consent. Individual experiments were per-
formed with blood from a single subject. For each sample,
200 μL of blood were dispensed into a prewarmed 96-well plate
and 188 μL of RPMI-1640 medium (Invitrogen) containing
20 μg/mL Indomethacin (Sigma, St. Louis, MO) was added.
Then 4 μL of a series of compound dilutions (final DMSO
concentration of 1%) were added in triplicate, followed by a
15 min incubation at 37 °C with gentle shaking. After that, blood
samples were stimulated by adding 8 μL of ionomycin (from
Streptomyces conglobatus, Calbiochem) to a final concentration
of 36 μM. After another incubation at 37 °C for 30 min, samples
were centrifuged at 4 °C for 5 min at 1800g. LTB4 concentrations
in supernatants were determined using a commercially avail-
able enzyme-linked immunosorbent assay (R&D Systems,
Minneapolis, MN) according to the manufacturer’s instruc-
tions. On the basis of the measured amounts of LTB4 formed
at each inhibitor concentration, a dose-response curve was
fitted to the data using a sigmoidal three-parameter func-
tion (XLfit, model 203 with Hill slope fixed to -1) and an
IC50 value was calculated. Each assay plate contained eight
positive controls (no compound) and eight negative controls (no
ionomycin).
1
compound 43 (4.1 g, 83%). H NMR (400 MHz, DMSO-d6)
δ: 7.38 (d, J=8.7 Hz, 2H), 6.87-7.05 (m, 6H), 3.73-3.84 (m,
2H), 3.26-3.43 (m, 1H), 2.72-2.88 (m, 2H), 2.30-2.46 (m, 1H),
1.78-1.92 (m, 1H), 1.54-1.76 (m, 2H), 1.35-1.52 (m, 1H).
Methyl 3-[(2R)-2-{[4-(4-Chlorophenoxy)phenoxy]methyl}-1-
pyrrolidinyl]propanoate (46). To a solution of (17) (0.8 g, 2.63
mmol) in CH2Cl2 (7.6 mL) was added Et3N (0.78 mL, 5.60
mmol) and methyl 3-bromopropionate (0.32 mL, 2.93 mmol).
The resulting solution was stirred at 30 °C overnight. The
reaction mixture was poured into water/CH2Cl2 (50 mL/50
mL). The crude residue was extracted into CH2Cl2 (50 mL).
The organic portion was washed with water (50 mL), washed
with brine (50 mL), dried over anhydrous Na2SO4, filtered, and
concentrated in vacuo. The crude product was purified by silica
gel flash chromatography (0-50% EtOAc/hexanes, gradient) to
give the title compound 46 (0.79 g, 77%). 1H NMR (400 MHz,
DMSO-d6) δ: 7.34-7.42 (m, 2H), 6.90-7.04 (m, 6H), 3.90
(dd, J=5.1, 9.4 Hz, 1H), 3.74 (dd, J=6.5, 9.5 Hz, 1H), 3.57
(s, 3H), 3.15 (d, J=12.2 Hz, 1H), 2.98-3.06 (m, 1H), 2.83 (dd,
J=5.9, 8.1 Hz, 1H), 2.56-2.66 (m, 1H), 2.38-2.48 (m, 2H),
2.23 (d, J=7.9 Hz, 1H), 1.85-1.98 (m, 1H), 1.65-1.76 (m, 2H),
1.59 (s, 1H).
4-{(R)-2-(4-(4-Chlorophenoxy)phenoxymethyl)pyrrolidin-1-yl}-
l-butyric Acid Methyl Ester (47). To a solution of 17 (0.8 g,
2.63 mmol) in CH2Cl2 (7.6 mL) was added Et3N (0.78 mL,
5.60 mmol) and methyl 4-bromobutyate (0.35 mL, 3.04 mmol).
The resulting solution was stirred at 30 °C overnight. The
reaction mixture was poured into water/CH2Cl2 (50 mL/
50 mL). The crude residue was extracted into CH2Cl2. The
organic portion was washed with water (50 mL), brine (50 mL),
dried over anhydrous Na2SO4, filtered, and concentrated in
vacuo. The crude product was purified by silica gel flash
chromatography using (0-50% EtOAc/hexanes, gradient) to
1
give the methyl ester 47 (0.60 g, 57%). H NMR (400 MHz,
DMSO-d6) δ: 7.38 (d, J=8.8 Hz, 2H), 6.89-7.03 (m, 6H), 3.87
(dd, J=5.0, 9.5 Hz, 1H), 3.73 (dd, J=6.6, 9.4 Hz, 1H), 3.55 (s,
3H), 2.98-3.08 (m, J=3.6 Hz, 1H), 2.81-2.89 (m, J=12.0 Hz,
1H), 2.73-2.81 (m, J=1.9 Hz, 1H), 2.29-2.40 (m, 3H), 2.17 (q,
J = 7.9 Hz, 1H), 1.85-1.98 (m, 1H), 1.63-1.77 (m, 4H),
1.53-1.63 (m, 1H).
4-{(S)-2-[4-(4-Chlorophenoxy)phenoxymethyl]pyrrolidin-1-yl}-
butyric Acid Methyl Ester (48). To a solution of 43 (4.0 g, 13.2
mmol) in DMF (50 mL) was added K2CO3 (3.7 g, 26.5 mmol)
and methyl 4-bromobutyrate (3.6 g, 19.8 mmol). The resulting
suspension was stirred at 55 °C for 16 h. The solvent was
removed in vacuo, and the residue was partitioned between
hexanes (250 mL) and water (150 mL). The organic layer was
washed with water (150 mL), dried over anhydrous Na2SO4,
filtered, and concentrated in vacuo. The crude product was
purified by silica gel flash chromatography (1-2% iPrOH/
EtOAc þ 0.25% TEA, gradient) to give the title compound 48
(4.1 g, 77%). MS m/z: 404 (M þ H). 1H NMR (400 MHz,
DMSO-d6) δ: 7.38 (d, J=8.8 Hz, 2H), 6.89-7.03 (m, 6H), 3.87
(dd, J=5.1, 9.4 Hz, 1H), 3.73 (dd, J=6.7, 9.3 Hz, 1H), 3.55 (s,
3H), 3.00-3.08 (m, 1H), 2.73-2.90 (m, 2H), 2.30-2.40 (m, 3H),
2.17 (q, J=8.2 Hz, 1H), 1.85-2.01 (m, 1H), 1.64-1.76 (m, 4H),
1.60 (q, J=6.0 Hz, 1H).
X-ray Crystallography. With PEG8000 as precipitant, ytter-
bium as an additive, and the use of microseeding to control
nucleation rate, well-diffracting crystals of the enzyme in com-
plex with the micromolar inhibitor Bestatin were obtained
routinely. These crystals were subsequently used to soak in
various high-affinity lead compounds of interest, replacing
Bestatin. Using this strategy, a crystal structure of LTA4H in
complex with compounds 2, 5, 6, 7, 8, 9, 13, 14, 15, and 20 were
determined at 1.67, 2.16, 1.8, 2.0, 2.05, 2.4, 1.90, 1.6, 1.63, and
˚
2.5 A resolution, respectively. The carboxylate group of com-
pound 20 chelates the catalytic Zn ion, its oxygens replacing two
water oxygens in almost identical positions in the structure with
compound 14. The two phenyl rings of both compounds bind in
a very similar mode to the end of the bend, hydrophobic
substrate binding pocket, which has three conserved waters
tightly bound.
Oral Dosing of Compounds to Rats. The compound was
formulated in 10% Solutol HS15 (BASF AG, Ludwigshafen,
Germany), 30 mM NaH2PO4, pH 7.9. The same solution without
the compound was used for vehicle control experiments. The
Biological Assays. In Vitro Enzyme Assay. LTA4 substrate
was prepared from the methyl ester of LTA4 (BioMol, Plymouth
Meeting, PA, or Cayman Chemicals, Ann Arbor, MI) by
treatment under nitrogen with 100 mol equiv of NaOH in an
acetone:H2O (4:1) solution at rt for 40 min. Stock solutions of

584 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Sandanayaka et al.
solutions were given by orogastric gavage to male Sprague-Daw-
ley rats in a dosing volume of 10 mL/kg (1% of body weight) using
stainless steel feeding needles of 0.9 mm ꢀ 38 mm (B & K
Universal AB, Sollentuna, Sweden) and 1 mL Omnifix syringes
(B. Braun, Melsungen, Germany). The rats were fasting in the
single dose time course experiments and nonfasting in all the other
experiments.
Isolation of Whole Blood from Dosed Animals. Rats were put
in a special cylinder to hold them still. The tail vein was
punctured with 25G ꢀ 5/8 in. needle and the blood collected
into a heparin tube. Dosing in dog and monkey was carried out
at CROs sites.
responses. Am. Rev. Respir. Crit. Care Dis. 1987, 136, 985–998. (c)
Drazen, J. M. Pharmacology of leukotriene receptor antagonists and
5-lipoxygenase inhibitors in the management of asthma. Pharma-
cotherapy 1997, 17, S22–S30. (d) Marx, J. L. The leukotrienes in
allergy and inflammation. Science 1982, 215, 1380–1383. (e) Griffiths,
R. J.; Pettipher, E. R.; Koch, K.; Farrell, C. A.; Breslow, R.; Conklyn, M.
J.; Smith, M. A.; Hackman, B. C.; Wimberly, D. J.; Milici, A. J.
Leukotriene B4 plays a critical role in the progression of collagen-
induced arthritis. Proc. Natl. Acad. Sci. U.S.A. 1995, 92 (2), 517–521.
(f) Shao, W. H.; Del, Prete, A.; Bock, C. B.; Haribabu, B. Targeted
disruption of leukotriene B4 receptors BLT1 and BLT2: a critical role
for BLT1 in collagen-induced arthritis in mice. J. Immunol. 2006, 176
(10), 6254–6261. (g) Tsuji, F.; Oki, K.; Fujisawa, K.; Okahara, A.;
Horiuchi, M.; Mita, S. Involvement of leukotriene B4 in arthritis
models. Life Sci. 1999, 64 (3), PL51–PL56. (h) Barnes, P. J. New
treatments for COPD. Nat. Rev. Drug Discovery 2002, 1 (6), 437–446.
(i) Gompertz, S.; Stockley, R. A. A randomized, placebo-controlled trial
of a leukotriene synthesis inhibitor in patients with COPD. Chest 2002,
122 (1), 289–294. (j) Lemiere, C.; Pelissier, S.; Tremblay, C.;
Chaboillez, S.; Thivierge, M.; Stankova, J.; Rola-Pleszczynski, R.
Leukotrienes and isocyanate-induced asthma: a pilot study. Clin.
Exp. Allergy 2004, 34 (11), 1684–1689. (k) Luster, A. D.; Tager, A.
M. T-cell trafficking in asthma: lipid mediators grease the way. Nat.
Rev. Immunol. 2004, 4 (9), 711–724. (l) Terawaki, K.; Yokomizo, T.;
Nagase, T.; Toda, A.; Taniguchi, M.; Hashizume, K..; et al. Absence
of leukotriene B4 receptor 1 confers resistance to airway hyperrespon-
siveness and Th2?type immune responses. J. Immunol. 2005,
175 (7), 4217–4225. (m) Montuschi, P.; Sala, A.; Dahlen, S. E.;
Folco, G. Pharmacological modulation of the leukotriene pathway in
allergic airway disease. Drug Discovery Today 2007, 12 (9-10), 404–
412.
Acknowledgment. We wish to acknowledge technical con-
tributions of Jenny Lin for providing guidance and overall
analytical support, Denise Anderson for running mass spec-
tral analyses, Dr. Nelson Zhao for acquiring NMR spectra,
and Brian Peasefor generating analyticalHPLC data. We also
thank Dr. Kari Stefansson for helpful discussions and en-
couragement during the course of this work.
Supporting Information Available: Elaboration of fragments,
hydrophilic linkers for which bound-structures are shown,
hydrolase-bound fragments containing hydrophilic linkers, plot
of BEI vs. SEI, preparation of recombinant human LTA4
hydrolase, protein crystallization, preparation of compound-
soaked crystals and crystal freezing, and data collection, struc-
ture solution, and refinement. This material is available free of
charge via the Internet at http://pubs.acs.org.
(8) (a) Brooks, D. W.; Summers, J. B.; Gunn, B. P.; Rodriques, K. E.;
Martin, J. G.; Martin, M. B.; Mazdiyasni, H.; Holms, J. H.;
Stewart, A. O.; Moore, J. L.; Young, P. R.; Albert, D. H.; Bouska,
J. B.; Malo, P. E.; Dyer, R. D.; Bell, R. L.; Rubin, P.; Kesterson, J.;
Carter, G. W. The discovery of A-64077, a clinical candidate for
treating diseases involving leukotriene mediators. In Abstracts of
the International Chemical Congress of Pacific Basin Societies;
Honolulu, HI, 1989; BIOS 34 . (b) Sirois, P.; Borgeat, P.; Lauziere,
M.; Dube, L.; Rubin, P. R.; Kesterson, J. Effect of zileuton (A-64077) on
the 5-lipoxygenase activity of human whole blood ex vivo. Agents
Actions 1991, 34, 117–120.
(9) Labelle, M.; Belley, M.; Gareau, Y.; Gauthier, J. Y.; Guay, D.;
Gordon, R.; Grossman, S. G.; Jones, T. R.; Leblanc, Y.;
McAuliffe, M.; McFarlane, C.; Masson, P.; Metters, K. M.;
Ouimet, N.; Patrick, D. H.; Piechuta, H.; Rochette, C.; Sawyer,
N.; Xiang, Y. B.; Pickett, C. B.; Ford-Hutchinson, A. W.;
Zamboni, R. J.; Young, R. N. Discovery of MK-0476, a potent
and orally active leukotriene D4 receptor antagonist devoid of
peroxisomal enzyme induction. Bioorg. Med. Chem. Lett. 1995,
5 (3), 283–288.
(10) Matassa, V. G.; Maduskuie, T. P.; Shapiro, H. S.; Hesp, B.; Snyder,
D. W.; Aharony, D.; Krell, R. D.; Keith, R. A. Evolutionof a Series
of Peptidoleukotriene Antagonists: Synthesis and Structure/Acti-
vity Relationships of 1,3,5-Substituted Indoles and Indazoles.
J. Med. Chem. 1990, 33, 1781–1790.
(11) Samuelsson, B.; Funk, C. D. Enzymes involved in the biosynthesis
of leukotriene B4. J. Biol. Chem. 1989, 264, 19469–19472.
(12) (a) Lewis, R. A.; Austen, K. F.; Soberman, R. J. Leukotrienes and
other products of the 5-lipoxygenase pathway. Biochemistry and
relation to pathology in human diseases. N. Engl. J. Med. 1990, 232,
645–655. (b) Werz, O.; Steinhilber, D. Therapeutic options for
5-lipoxygenase inhibitors. Pharmacol. Ther. 2006, 112, 701–718.
(13) Mitsunobu, F.; Mifune, T.; Hosaki, Y.; Ashida, K.; Tsugeno, H.;
Okamoto, M.; Takata, S.; Tanizaki, Y. Enhanced production of
leukotrienes by peripheral leukocytes and specific IgE antibodies in
patients with chronic obstructive pulmonary disease. J. Allergy
Clin. Immunol. 2001, 107, 492–498.
References
(1) Helgadottir, A.; Manolescu, A.; Thorleifsson, G.; Jonsdottir, H.;
Thorsteinsdottir, U.; Samani, N. J.; Gretarsdottir, S.; Johannsson,
H.; Gudmundsson, G.; Grant, S. F. A.; Sveinbjornsdottir, S.;
Valdimarsson, E. M.; Matthiasson, S. E.; Gudmundsdottir, O;
Gurney, M.; Sainz, J.; Thorhallsdottir; Andresdottir, M.; Frigge,
M. L.; Gudnason, V.; Kong, A.; Topol, E. J.; Thorgeirsson, G.;
Gulcher, J. R.; Hakonarson, H.; Stefansson, K. The gene encoding
5-lipoxygenase activating protein confers risk of myocardial in-
farction and stroke. Nat. Genet. 2004, 36, 233–239.
(2) Helgadottir, A.; Manolescu, A.; Helgason, A.; Thorleifsson, G.;
Thorsteinsdottir, U.; Gudbjartsson, D. F.; Gretarsdottir, S.;
Magnusson, K. P.; Gudmundsson, G.; Hicks, A.; Jonsson, T.;
Grant, S. F. A.; Sainz, J.; O’Brien, S. J.; Sveinbjornsdottir, S.;
Valdimarsson, E. M.; Matthiasson, S. E.; Levey, A. I.; Abramson,
J. L.; Reilly, M.; Vaccarino, V.; Wolfe, M.; Gudnason, V.;
Quyyumi, A. A.; Topol, E. J.; Rader, D. J.; Thorgeirsson, G.;
Gulcher, J. R.; Hakonarson, H.; Kong, A.; Stefansson, K. A
variant of the gene encoding leukotriene A4 hydrolase confers
ethnicity-specific risk of myocardial infarction. Nat. Genet. 2006,
38, 68–74.
(3) Meng, C. Q. Atherosclerosis is an inflammatory disorder after all.
Curr. Top. Med. Chem 2006, 6, 93–102.
(4) Spanbroek, R.; Grabner, R.; Lotzer, K.; Hildner, M.; Urbach, A.;
Ruhling, K.; Moos, M. P.; Kaiser, B.; Cohnert, T. U.; Wahlers, T.;
Zieske, A.; Plenz, G.; Robenek, H.; Salbach, P.; Kuhn, H.;
Radmark, O.; Samuelsson, B.; Habenicht, A. J. Expanding expres-
sion of the 5-lipoxygenase pathway within the arterial wall during
human atherogenesis. Proc. Natl. Acad. Sci. U.S.A. 2003, 100 (3),
1238–1243.
(5) Cipollone, F.; Mezzetti, A.; Fazia, M. L.; Cuccurullo, C.; Iezzi, A.;
Ucchino, S.; Spigonardo, F.; Bucci, M.; Cuccurullo, F.; Prescott, S.
M.; Stafforini, D. M. Association between 5-lipoxygenase expres-
sion and plaque instability in humans. Arterioscler. Thromb. Vasc.
Biol. 2005, 25, 1665–1670.
(14) Okano-Mitani, H.; Ikai, K.; Imamura, S. Leukotriene A4 hydro-
lase in peripheral leukocytes of patients with atopic dermatitis.
Arch. Dermatol. Res. 1996, 288, 168–172.
(15) Klickstein, L. B.; Shapleigh, C.; Geotzl, E. J. Lipoxygenation of
arachidonic acid as a source of polymorphonuclear leukocyte
chemotactic factors in synovial fluid and tissue in rheumatoid
arthritis and spondyloarthritis. J. Clin. Invest. 1980, 66, 1166–
1170.
(16) (a) Penning, T. D.; Askonas, L. J.; Djuric, S. W.; Haack, R. A.; Yu,
S. S.; Michener, M. L.; Krivi, G. G.; Pyla, E. Y. Kelatorphan and
related analogs: Potent and selective inhibitors of leukotriene A4
hydrolase. Bioorg. Med. Chem. Lett. 1995, 5, 2517–2522. (b) Hogg,
(6) Qiu, H.; Gabrielsen, A.; Agardh, H. E.; Wan, M.; Wetterholm, A.;
Wong, C. H.; Hedin, U.; Swedenborg, J.; Hansson, G. K.;
€
Samuelsson, B.; Paulsson-Berne, G.; Haeggstrom, J. Z. Expression
of 5-lipoxygenase and leukotriene A4 hydrolase in human athero-
sclerotic lesions correlates with symptoms of plaque instability.
Proc. Natl. Acad. Sci. U.S.A. 2006, 103 (21), 8161–8166.
(7) (a) Lewis, R. A.; Austen, K. F.; Soberman, R. J. Leukotrienes and
other products of the 5-lipoxygenase pathway. Biochemistry and
relationto pathologyin human diseases. N. Engl. J. Med. 1990, 232,
645–655. (b) Drazen, J. M.; Austen, K. F. Leukotrienes and airway
€
J. H.; Ollmann, I. R.; Haeggstrom, J. Z.; Wetterholm, A.; Samuelsson,
B.; Wong, C. H. Amino hydroxamic acids as potent inhibitors of

Article
leukotriene A4 hydrolase. Bioorg. Med. Chem. Lett. 1995, 3, 1405–
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 585
J.; Zembower, D.; Kim, H.; Kiselyov, A.; Burgin, A.; Gurney,
M.; Stewart, L. Discovery of leukotriene A4 hydrolase inhibitors
using metabolomics biased fragment crystallograph. J. Med.
Chem. 2009, 52, 4694–4715.
1415. (c) Tsuji, F.; Miyake, Y.; Enomoto, H.; Horiuchi, M.; Mita, S.
Effects of SA6541, a leukotriene A4 hydrolase inhibitor, and indo-
methacin on carrageenan-induced murine dermatitis. Eur. J. Pharma-
col. 1998, 346, 81–85. (d) Grice, C. A.; Tays, K. L.; Savall, B. M.; Wei,
J.; Butler, C. R.; Axe, F. U.; Bembenek, S. D.; Fourie, A. M.; Dunford, P.
J.; Lundeen, K.; Coles, F.; Xue, X.; Riley, J. P.; Williams, K. N.;
Karlsson, L.; Edwards, J. P. J. Med. Chem. 2008, 51, 4150–4169. (e)
Rao, N. L.; Dunford, P. J.; Xue, X.; Jiang, X.; Lundeen, K. A.; Coles, F.;
Riley, J. P.; Williams, K. N.; Grice, C. A.; Edwards, J. P.; Karlsson, L.;
Fourie, A. M. J. Pharmacol. Exp. Ther. 2007, 321, 1154–1160. (f)
Khim, S. K.; Bauman, J.; Evans, J.; Freeman, B.; King, B.; Kirkland, T.;
Kochanny, M.; Lentz, D.; Liang, A.; Mendoza, L.; Phillips, G.; Tseng, J.
L.; Wei, R. G.; Ye, H.; Yu, L.; Parkinson, J.; Guilford, W. J. Bioorg.
Med. Chem. Lett. 2008, 18, 3895–3898. (g) Kirkland, T. A.; Adler, M.;
Bauman, J. G.; Chen, M.; Haeggstrom, J. Z.; King, B.; Kochanny, M. J.;
Liang, A. M.; Mendoza, L.; Phillips, G. B.; Thunnissen, M.; Trinh, L.;
Whitlow, M.; Ye, B.; Ye, H.; Parkinson, J.; Guilford, W. J. Bioorg. Med.
Chem. 2008, 16, 4963–4983. (h) Ye, B.; Bauman, J.; Chen, M.; Davey,
D.; Khim, S. K.; King, B.; Kirkland, T.; Kochanny, M.; Liang, A.;
Lentz, D.; May, K.; Mendoza, L.; Phillips, G.; Selchau, V.; Schlyer, S.;
Tseng, J. L.; Wei, R. G.; Ye, H.; Parkinson, J.; Guilford, W. J. Bioorg.
Med. Chem. Lett. 2008, 18, 3891–3894. (i) Enomoto, H.; Morikawa,
Y.; Miyake, Y.; Tsuji, F.; Mizuchi, M.; Suhara, H.; Fujimura, K.;
Horiuchi, M.; Ban, M. Bioorg. Med. Chem. Lett. 2009, 19, 442–446.
(17) Congreve, M.; Chessari, G.; Tisi, D.; Woodhead, A. J. Recent
developments in fragment-based drug discovery. J. Med. Chem.
2008, 51, 3661–3689.
(25) (a) Thunnissen, M. M.; Nordlund, P.; Haeggstrom, J. Z. Crystal
structure of human leukotriene A(4) hydrolase, a bifunctional
enzyme in inflammation. Nat. Struct. Biol. 2001, 8, 131–135. (b)
Thunnissen, M. M.; Andersson, G. M.; Samuelsson, B.; Wong, B.;
€
Haeggstrom, C. H.; Crystal, J. Z. Structures of leukotriene A4 hydrolase
in complex with captopril and two competitive tight-binding inhibitors.
FASEB J. 2002, 16, 1648–1650.
(26) Penning, T. D.; Chandrakumar, N. Z; Chen, B. B; Chen, H. Y.;
Desai, B. N.; Djuric, S. W.; Docter, S. H.; Gasiecki, A. F.; Haack,
R. A.; Miyashiro, J. M.; Russell, M. A. Structure-activity relation-
ship studies on 1-[2-(4-phenylphenoxy)ethyl]pyrrolidine (SC-
22716), a potent inhibitor of leukotriene A4 (LTA4) hydrolase.
J. Med. Chem. 2002, 43, 721–735.
˚
(27) A 1.9 A resolution structure, 3FUD.pdb that contains a fragment
bound on the surface of LTA4H far from the active site (see ref 21),
was used as reference for unligated active site topology.
(28) Cerep; http://www.cerep.fr/cerep/users/pages/catalog/Profiles/catalog.
asp.
(29) Such a transformation has in fact been reported for SC-57461.
Yuan, J. H.; Birkmeier, J.; Yang, D. C.; Hribar, J. D.; Liu, N.;
Bible, R.; Hajdu, E.; Rock, M.; Schoenhard, G. Isolation and
identification of metabolites of leukotriene A4 hydrolase inhibitor
SC-57461 in rats. Drug Metab. Dispos. 1996, 24, 1124–1133.
(30) Yan, Z.; Caldwell, G. W. Metabolism profiling and cytochrome
P450 inhibition and induction. Drug Discovery Curr. Top. Med.
Chem. 2001, 1, 403–425.
(18) Rees, D. C.; Congreve, M.; Murray, C. W.; Carr, R. Fragment-
based lead discovery. Nat. Rev. Drug Discovery 2004, 3, 660–672.
(19) Congreve, M.; Carr, R.; Murray, C.; Jhoti, H. A “rule of three” for
fragment-based lead discovery? Drug Discovery. Today 2003, 8,
876–877.
(31) (a) Mitsunobu, O.; Yamada, M. Bull. Chem. Soc. Jpn. 1967, 40,
2380–2382. (b) Mitsunobu, O. Synthesis 1981, 1–28.
(32) Wolfe, J. P.; Buchwald, S. L. Palladium-catalyzed amination of
(20) Nienaber, V. L.; Greer, J. Discovering novel ligands for macro-
molecules using X-ray crystallographic screening. Nat. Biotechnol.
2000, 18, 1105–1108.
(21) Hopkins, A. L.; Colin, R.; Groom, C. R.; Alex, A. Ligand
efficiency: a useful metric for lead selection. Drug Discovery Today
2004, 9, 430–431.
(22) Abad-Zapatero, C.; Metz, J. T. Ligand efficiency indices as guide-
posts for drug discovery. Drug Discovery Today 2005, 10, 464–469.
(23) Oprea, T.; Allu, T. R.; Fara, D. C.; Rad, R. F.; Ostopovici, L.;
Bologa, C. G. Lead-like, drug-like or “Pub-like”: how different are
they? J. Comput.-Aided Des. 2007, 21, 113–119.
aryl iodides. J. Org. Chem. 1996, 61, 1133–1135.
(33) Gandolfi, C. A.; Di-Domenico, R.; Spinelli, S.; Gallico, L.;
Fiocchi, L.; Lotto, A.; Menta, E.; Borghi, A.; Rosa, C. D.;
Tognella, S. N-Acyl-2-substituted-1,3-thiazolidineas, new class of
non-narcotic antitussive agents: Studies leading to the discovery of
ethyl 2-[(2-methoxyphenoxy)methyl]-3-oxothiazolidine-3-propa-
noate. J. Med. Chem. 1995, 38, 508–525.
(34) Tagat, T.; Nishida, M. Palladium charcol-catlysed Suzuki-
Miyaura coupling to obatin arylpyridines and arylquinolines.
J. Org. Chem. 2003, 68, 9412–9415.
(35) Ma, D.; Qian, C. N,N-Dimethyl glycine-promoted Ullmann cou-
pling reaction of phenols and aryl halides. Org. Lett. 2003, 5 (21),
3799–3802.
(24) Davies, D.; Mamat, B.; Magnusson, O.; Christensen, J.;
Haraldsson, M.; Mishra, R.; Pease, B.; Hansen, E.; Singh,