Discovery of LYS006, a Potent and Highly Selective Inhibitor of Leukotriene A4Hydrolase

Christian Markert,* Gebhard Thoma, Honnappa Srinivas, Birgit Bollbuck, Rainer M. Lu

Drug Annotation

Discovery of LYS006, a Potent and Highly Selective Inhibitor of

Leukotriene A Hydrolase

Wolfgang Miltz, Rudolf Walchli, Romain Wolf, Jurgen Hinrichs, Christian Bergsdorf, Kamal Azzaoui,

Carlos A. Penno, Kai Klein, Nathalie Wack, Petra Jager, Franziska Hasler, Christian Beerli,

Pius Loetscher, Janet Dawson, Grazyna Wieczorek, Shin Numao, Amanda Littlewood-Evans,

and Till A. Rohn

ond,

Cite This: J. Med. Chem. 2021, 64, 1889 −1903

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: The cytosolic metalloenzyme leukotriene A hydrolase (LTA4H) is the ﬁnal and

rate-limiting enzyme in the biosynthesis of pro-inﬂammatory leukotriene B (LTB ). Preclinical

studies have validated this enzyme as an attractive drug target in chronic inﬂammatory diseases.

Despite several attempts, no LTA4H inhibitor has reached the market, yet. Herein, we disclose the

discovery and preclinical proﬁle of LYS006, a highly potent and selective LTA4H inhibitor. A

focused fragment screen identiﬁed hits that could be cocrystallized with LTA4H and inspired a

fragment merging. Further optimization led to chiral amino acids and ultimately to LYS006, a

picomolar LTA4H inhibitor with exquisite whole blood potency and long-lasting pharmacody-

namic eﬀects. Due to its high selectivity and its ability to fully suppress LTB generation at low

exposures in vivo, LYS006 has the potential for a best-in-class LTA4H inhibitor and is currently

investigated in phase II clinical trials in inﬂammatory acne, hidradenitis suppurativa, ulcerative

colitis, and NASH.

INTRODUCTION

myeloid cells, especially neutrophils, macrophages, dendritic

cells, and monocytes, but also mast cells. It is an important

■

Leukotriene A4 hydrolase (LTA4H) is a monomeric 69 kDa,

cytosolic metalloenzyme with an active site, that contains a

inﬂammatory mediator capable of initiating and amplifying

innate and adaptive immune responses and has a particularly

strong eﬀect on the recruitment and activation of neutro-

Zn ion, located between two domains. This active site has

distinct but overlapping substrate binding pockets and exhibits

two enzymatic activities, an aminopeptidase and an epoxide

,10

phils.

It has been implicated in many acute and chronic

human inﬂammatory diseases and its inhibition proved

beneﬁcial in numerous preclinical models of inﬂammation

hydrolase function.

The ﬁrst pocket of the catalytic site, the aminopeptidase

pocket is hydrophilic and open to the solvent. It can cleave

tripeptides, and it has been implicated in the degradation of

certain opioid peptides as well as tripeptide proline-glycine-

proline (PGP). The physiological relevance of this capacity of

11,12

with a strong neutrophil component.

LTA4H is a particularly interesting target as its inhibition

allows for the selective repression of LTB generation without

aﬀecting the biosynthesis of lipids that depend on the upstream

−5

enzymes 5-LO and FLAP. These include the diﬀerent

families of anti-inﬂammatory and resolution promoting lipid

mediators such as D- and E-series of resolvins, maresins, and

LTA4H to degrade peptides, however, remains unknown.

The second pocket, the hydrolase pocket of the enzyme,

forms a long, L-shaped hydrophobic channel. Substrate

leukotriene A (LTA ) binds to this pocket. The hydrolase

13,14

protectins, all of which are 5-LO dependent.

Moreover,

inhibition of LTA4H can even lead to elevated levels of anti-

function catalyzes the Zn assisted vinylogous ring-opening of

inﬂammatory and anti-ﬁbrotic lipoxin A (LXA ) due to

the epoxide LTA to form pro-inﬂammatory leukotriene B

pathway shunting. LXA₄is formed by 12- and 15-

(

LTB ). LTA4H is the ﬁnal and rate limiting enzyme in the

biosynthesis of this pro-inﬂammatory lipid mediator.

Received: November 11, 2020

Published: February 16, 2021

Whereas LTA4H is expressed in the majority of mammalian

cells and tissues, the upstream enzymatic machinery consisting

of cytosolic phospholipase (cPLA2), 5-lipoxygenase (5-LO),

and 5-lipoxygenase activating protein (FLAP), responsible for

the generation of LTA , is present in only a few cell types,

particularly leukocytes. Hence, LTB is mainly produced by

2021 American Chemical Society

889

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

clinical stage compounds (Figure 1 ).

Drug Annotation

lipoxygenases from the common precursor LTA that becomes

more abundant upon LTA4H inhibition. Hence, inhibition of

allergic asthma is considered minor. Cysteinyl leukotrienes

(CysLTs), on the other hand, which are not inhibited by

LTA4H inhibitors and may even be elevated due to pathway

shunting as indicated by the trial data, are key to the

LTA4H not only reduces pro-inﬂammatory LTB but can also

induce a lipid mediator class-switch from pro-inﬂammatory to

pro-resolving lipids and thereby possibly contribute to the

pathophysiology of asthma. CysLTs cause bronchoconstric-

tion and impact the FEV1 response. In addition, JNJ-

resolution of chronic inﬂammation.

Numerous pharmaceutical companies have initiated pro-

grams to discover LTA4H inhibitors and described several

40929837 caused testicular toxicity in rats due to accumulation

of a metabolite and was not further investigated in

indications beyond asthma and eventually discontinued.

Celtaxsys CTX-4430 (Acebilustat) is the most advanced

16,17

LTA4H inhibitor in clinical development. It reduced sputum

LTB levels and, importantly, led to a clinical reduction of

sputum neutrophil counts in cystic ﬁbrosis (CF) patients. This

translated to a 19−35% reduction in inﬂammatory exacer-

bations, but similarly to the previous observation by Janssen in

asthma it did not change FEV1 in a 48 week phase 2b study in

22−24

133 patients.

While Celtaxsys announced preparations

toward a phase III study in CF based on these results, no

further development of Acebilustat has been reported to date.

Herein, we describe the discovery and preclinical proﬁle of

LYS006, a potent and highly selective LTA4H inhibitor.

LYS006 not only exhibits an exquisite selectivity proﬁle but

also is characterized by picomolar enzyme inhibition, low IC₉₀

values in human whole blood assays, and sustained PD eﬀects

in vivo, ensuring nearly complete target engagement at very low

blood and particularly plasma exposures. Furthermore, we

Figure 1. Examples of LTA4H inhibitors that have reached phase II

clinical trials but are no longer reported to be in active development.

could demonstrate that LYS006 elevates LXA biosynthesis in

PBMCs via pathway shunting and eﬃciently suppresses

neutrophil activation in skin inﬂammation and KRN arthritis

models under prophylactic and therapeutic conditions.

Boehringer Ingelheim’s BI 691751 (structure not disclosed)

for the treatment of atherosclerosis was discontinued after

phase I. Trial data suggest that BI 691751 may have caused an

increase in both heart rate and length of the QT interval at

RESULTS AND DISCUSSION

■

higher doses. DeCODE discovered LTA4H inhibitor DG-

51. The compound exhibited mediocre potency in

Hit Finding. The RCSB Protein Data Bank holds several

inhibiting LTB₄biosynthesis in human cells, and further

development of DG-051 was discontinued after a phase II PK/

PD study.

cocrystal structures of small ligands bound to LTA4H. This

encouraged us to start our hit ﬁnding with a focused screen on

∼1800 compounds of the Novartis fragment library including

∼350 fragment-like molecules with in silico predicted potential

for LTA4H binding. Diﬀerential scanning ﬂuorimetry (DSF),

which measures the change in temperature at which a puriﬁed

protein unfolds in the presence of a putative binder, was used

as a way of testing for compound binding. A solvatochromic

In contrast, compound JNJ-40929837 by Janssen was

promoted into a phase II eﬃcacy trial in asthma. Despite

target engagement, early and late asthmatic response measured

by forced expiratory volume (FEV1) was not reduced. This is

possibly owing to the fact that the role of LTB in human

Figure 2. Fragment hits 1 (7AV2) and 2 (7AV1) bound to LTA4H (left) and hydrogen bonding network for fragment 2 (right). For clarity, only

key interactions are depicted.

890

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

particularly attracted our attention (Figures 2 and 3 ).

Drug Annotation

dye with aﬃnity for exposed hydrophobic parts of the protein

indicates the progress of unfolding.

Both hits 1 and 2, as well as hybrid 3 showed a

27,28

concentration dependent shift in LTA4H T , supporting the

In total, 14 fragments induced signiﬁcant positive shifts in

notion that DSF is well suited to identify binders to the

substrate binding site of LTA4H (Table 1).

Whereas fragments 1 and 2 showed measurable, but weak

melting temperature (T ) and were subsequently studied by

X-ray crystallography. From these, diaryl ether 1 and amine 2

inhibition in our biochemical peptidase assay (IC of 4.5 μM/

8.8 μM), hybrids 3 and 4 inhibited LTA4H with remarkable

IC₅₀values of 0.004 μM biochemically, and 0.121 and 0.129

μM in a human whole blood assay, respectively (Figure 3).

Early Optimization: Head Group SAR. Amine 3 and its

chloro-analogue 4 constituted an excellent starting point for

optimization. However, the combination of a basic amine with

a lipophilic tail raised concerns with regard to metabolic

instability, hERG and cytochrome P450 (CYP) inhibition

(

Table 2). Varying the chain length (5) or capping the amine

by acetylation (6) led to signiﬁcant losses in potency.

Substitution of the aliphatic chain (7, 8) was tolerated to

some degree, but it did not solve the inherent liabilities of this

starting point.

We speculated that more profound modiﬁcations were

needed and explored a diverse set of linkers and functional

groups attached to the tetrazole, while keeping the diaryl ether

unchanged. The key molecule identiﬁed from this eﬀort was

butyric acid derivative 9, as it oﬀered a structural alternative to

the amine headgroup. Compound 9 achieved high biochemical

activity combined with good microsomal stability, without

inhibiting CYPs or the hERG channel. Compound 10 carrying

a shorter propionic acid achieved less biochemical potency.

Activity in blood, however, was not detectable for either of the

two, which we attributed to very high plasma protein binding

of >99% (human plasma, RED).

Figure 3. Identiﬁcation of highly potent starting point 3 by merging

of fragment hits 1 and 2. Under DSF screening conditions, 1 and 2

increased the melting temperature (T ) of LTA4H by 0.5 °C (at 500

μM). Hybrid 3 stabilized by 6.8 °C (at 100 μM, Table 1).

Once this second motif 9 was identiﬁed, we probed the

combination of the two functional groups in diﬀerent

arrangements. The α-amino acids 11 and 12 showed good

microsomal stability and no hERG inhibition. Even more

interesting was β-amino acid 13, which as well solved the

liabilities of amine 4, while preserving or even improving

biochemical and cellular activity in human blood. Zwitterionic

lead compound 13 carries a stereocenter and both enantiomers

performed equally well in our biochemical and whole blood

assays. In an attempt to diﬀerentiate the two, we extended our

proﬁling to a panel of >100 potential oﬀ-targets and observed

very high speciﬁcity. Only a moderate binding to a single oﬀ-

target, the dopamine transporter (DAT) was found. However,

diﬀerentiation was still not possible, as both enantiomers (S)-

Both fragments bind to the hydrophobic hydrolase substrate

pocket of the second lobe of LTA4H. Binding of diaryl ether 1

is characterized by π−π interactions of both aromatic rings to

Phe314. Additionally, fragment 2 shows such a π−π interaction

to Phe314 and a π-stacking of its tetrazole to Tyr267. The

amine function of 2 forms three H-bonds in an almost perfect

sp geometry. The ﬁrst bond of the protonated amine goes to

negatively charged Glu271, the second to Gln136, and a third

one is a water mediated hydrogen bonding to Tyr383.

Most notably, however, superposition of both fragments

bound to LTA4H revealed an almost perfect overlay of the

central phenyl ring inspiring testing of hybrid amines, such as 3

and 4 (Figure 3 and Tables 1, 2).

3 and (R)-13 act as antagonists of DAT with similar IC₅₀

values of 10 and 15 μM, respectively, as determined in a

radioligand binding assay. As even moderate antagonism of

Table 1. DSF Conﬁrmation Data on Selected Compounds

compd

biochem IC₅₀

concn [μM]

ΔT_m[°C]

this transporter might complicate the development program at

later stages, requiring drug abuse and behavioral studies, we set

out to identify compounds without this moderate ﬂag.

Late Optimization: Tail SAR. In addition to DAT

inhibition, we considered the low solubility (27 μM in a

HT-solubility assay ) as a possible development risk and

therefore focused further compound optimization on these two

parameters. The terminal chlorophenyl was rapidly conﬁrmed

as a promising site for eﬃcient ﬁne-tuning of properties (Table

IC₅₀= 4.5 μM

125

0.08 ± 0.02

0.2 ± 0.03

0.3 ± 0.04

0.2 ± 0.01

0.3 ± 0.01

0.6 ± 0.02

1.0 ± 0.03

4.2 ± 0.05

5.0 ± 0.03

5.9 ± 0.08

6.8 ± 0.09

1000

125

IC₅₀= 8.8 μM

000

12.5

). The very close ﬂuoro-analogue 14 addressed DAT

IC₅₀= 0.004 μM

inhibition and led to some increase in solubility. Likewise,

additional ﬂexibility introduced by aliphatic ethers 15−18

modulated both parameters. Sparring of DAT was most

consistently observed for ethers 19−23 carrying additional

heteroaromatic rings. The strongest increases in solubility were

Each experiment was executed in quadruplicate.

891

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

Drug Annotation

Table 2. Head Group Modiﬁcations

LTA4H enzyme inhibition assay using Arg-AMC as substrate with IC₅₀shown as the mean ± SD, in the case of n ≥ 2. Human whole blood

ionophore stimulated LTB release assay with IC shown as the mean ± SD, in the case of n ≥ 2. Metabolic stability in rat liver microsomes

supplemented with NADPH. Binding assay using radiolabeled H-dofetilide and HEK293 cell membranes expressing human recombinant HERG

K channels. Reversible Inhibition of CYP3A4 (midazolam), CYP2C9 (diclofenac) and CYP2D6 (bufuralol) in human liver microsomes.

achieved by reducing the degree of planarity (17), or the use of

small heteroaromatic ethers (21−23). Fluoro-chloropyridine

details of interactions between compounds (R )-13 and (S )-22

and protein are shown in Supplementary Figures S1 and S2.

Selection of Clinical Candidate (S)-22/LYS006 Based

on Mouse PK/PD Data. As LTA4H is highly expressed in

blood cells, particularly in neutrophils and monocytes, in vivo

target engagement can be easily assessed via determination of

2 ultimately provided the best option for maintaining high

potency in enzyme and human whole blood assays, while

increasing compound solubility and selectivity versus DAT.

However, (S)- and (R)-enantiomers could still not be

diﬀerentiated based on in vitro behavior.

LTB production in ex vivo stimulated blood (Table 3). Mouse

Cocrystal structures oﬀered an explanation for the

subordinate role of the stereoconﬁguration on in vitro potency.

Figure 4 shows an overlay of the two cocrystal structures

obtained for lead compound (R)-13 and the (S)-conﬁgured

candidate (S)-22. Their carboxylic acids both reach to the

catalytic zinc, an important additional interaction introduced

by the zwitterionic motif. Notably, despite opposite stereo-

conﬁgurations, both ligands locate their primary amines in

almost identical positions for the key H-bonding to Glu271,

providing a rationale for their similar biochemical behaviors.

The tail of (S)-22 superimposes almost perfectly with the

original lead structure, illustrating that the nature of the tail

optimization was a ﬁne-tuning of properties while preserving

an already favorable binding mode. Electron densities and

PK/PD studies at deﬁned time-points after oral dosing

conﬁrmed that several of our advanced candidates achieved

excellent LTB reduction in vivo. We ranked compounds

according to their in vivo potency and in particular according

to the longevity of the PD eﬀect by determining LTB release

14 h post oral dosing. Using these conditions, we observed that

for all three enantiomeric pairs tested, the (S)-enantiomer

showed superior performance in PK/PD experiments [Table 3:

(

S/R)-18, (S/R)-20, (S/R)-22], apparently due to faster

clearance of the (R)-enantiomers. Based on this assessment,

(S)-22 was given preference over (R)-22. It showed the best

overall proﬁle and was selected for in depth proﬁling and

became clinical development candidate LYS006.

892

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

Drug Annotation

Table 3. Tail Modiﬁcations

LTA4H enzyme inhibition assay in screening mode, using Arg-AMC as substrate with IC₅₀shown as the mean ± SD, in case of n ≥ 2. Lowest

determined biochemical potency is 3 nM in this setting. Human whole blood ionophore-stimulated LTB release assay with 15 min compound

preincubation. IC₅₀shown as the mean ± SD, in case of n ≥ 2. DAT radioligand binding assay on human recombinant DAT with IC shown as

32 e

the mean of 1−6 assessments.

Equilibrated solubility in aqueous buﬀer at pH 6.8. Inhibition of LTB release from ex vivo ionophore-

stimulated mouse blood 14 h after p.o. dose of compound compared to vehicle control given as the mean percent inhibition ± SEM (n ≥ 3).

Physicochemical Properties and In Vitro Proﬁle of

LYS006. LYS006 is a zwitterion (pK_a₁= 7.5; pK_a₂= 3.3),

showing pH dependent solubility with maxima at low (<2) and

high pH (>9) and a minimum between pH 5.4−6. Aqueous

solubility of crystalline material was conﬁrmed as 207 μM at 25

IC₅₀of 0.12 ± 0.01 nM (Table 4). Cellular potency was

determined in calcium ionophore stimulated human PBMCs.

The potency of LYS006 showed a time dependency, meaning

that increased incubation times resulted in lower IC and IC

values and steeper dose response curves (Figure 5), a

phenomenon we did not observe for LTA4H inhibitors from

other chemical series. Incubation for 4 h revealed an IC₅₀of

3.7 nM. LYS006 shows an unbound plasma fraction of 7.9% in

human plasma with comparable plasma protein binding across

all tested species. In human whole blood, we measured an IC₅₀

°C. Passive permeability of LYS006 is low to moderate (log P_e

pH 6.8 = −6).

Exact determination of the biochemical potency of LYS006

required a reduction in enzyme concentration compared to

screening conditions and delivered a remarkable biochemical

893

Journal of Medicinal Chemistry

J. Med. Chem. 2021, 64, 1889−1903

Figure 4. Optimized candidate (S)-22 (7AUZ) and lead compound

(R)-13 (7AV0) bound to LTA4H (overlay). Despite opposite

stereoconﬁgurations, the amine and acid groups occupy almost

identical positions, rationalizing similar biochemical behavior of the

optical antipodes.

of 53 nM and again noticed a time dependency (Figure 5).

Similar observations have recently been reported for poorly

Figure 5. Human PBMCs (upper panel) and human whole blood

(lower panel) were preincubated for diﬀerent times with increasing

concentrations of LYS006. Compound preincubation times ranged

from 15 to 240 min. Cells were then stimulated with ionophore and

permeable FLAP inhibitors. Whole blood IC₅₀values

determined in other species such as mouse, dog and rabbit

were very similar to those determined with human blood (data

not shown). Head to head comparison of LYS006 with

previously published clinical stage LTA4H inhibitors revealed

similar whole blood potency as JNJ-40929837 but clearly

superior potency to DG051 and acebilustat (Table 4).

Selectivity of LYS006. LYS006 shows an exquisite

selectivity proﬁle when tested at concentrations of ≥10 μM

in binding or functional mode on a panel of >150 GPCRs,

transporters, ion channels, nuclear receptors, and enzymes in

vitro. Likewise, in vitro investigations did not indicate any

inhibitory potential on hERG channel currents with IC₅₀> 30

μM for binding and Qpatch, and >100 μM in a manual patch

clamp assay. On Nav 1.5, an IC₅₀> 50 μM was also

established. LYS006 does not inhibit any CYP450 family

member up to tested concentrations of 20 μM, has no

phototoxic potential and is negative in genotoxic and

mutagenicity assays (Ames and in vitro and in vivo micro-

LTB release determined. Dose−response curves with mean LTB

levels ± SD are shown.

nucleus test). As LTA4H is a Zn -containing metal-

lopeptidase, we included cross-reactivity testing versus a

selection of other Zn containing metalloproteases testing

matrix metalloproteases 1, 2, 8, 9, 12, 13, 14 and TACE.

LYS006 did not inhibit any of these up to concentrations of 30

μM.

Another feature of LTA4H is that it carries an amino-

peptidase as well as a hydrolase function. Similarly to all

previous clinically tested LTA4H inhibitors, LYS006 inhibits

both functions, the hydrolytic opening of LTA leading to

LTB , and the degradation of tripeptides with similar IC

(data not shown).

Table 4. In Vitro Potency of LYS006 versus Clinical Stage LTA4H Inhibitors

potency [nM]

LYS006

DG051

JNJ-40929837

Acebilustat

biochemical IC₅₀

0.12 ± 0.01

3.7 ± 2.6

27 ± 13

53 ± 23

148 ± 67

54 ± 17

74 ± 10

1069 ± 759

510 ± 133

5054 ± 2731

1.9 ± 1.2

18 ± 8.5

92 ± 76

48 ± 5.5

104 ± 9.5

4.9 ± 1.3

24 ± 9.0

181 ± 91

142 ± 37

728 ± 225

human PBMC IC₅₀

human PBMC IC₉₀

human whole blood IC₅₀

human whole blood IC₉₀

LTA4H enzyme inhibition assay using Arg-AMC as substrate with <1 nM enzyme concentration. IC₅₀are mean values ± SD (n ≥ 3). Human

PBMC ionophore stimulated LTB release assay with 4 h compound preincubation. IC and IC₉₀shown as the mean ± SD (n ≥ 3). Human

whole blood ionophore stimulated LTB release assay with 4 h compound preincubation. IC and IC₉₀shown as the mean ± SD (n ≥ 3).

894

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

Data are shown for 30 min incubation time (Figure 7 ). At

Drug Annotation

Furthermore, we tested selectivity of LYS006 on eicosanoid

metabolizing enzymes and determined the concentration of a

panel of 45 lipid mediators released from stimulated human

PBMCs after treatment with rising concentrations of LYS006.

disposition (TMDD), where the distribution changes with the

degree of target saturation. LTA4H is expressed in nearly all

mammalian cells and expression in blood is high in

neutrophils, monocytes but also erythrocytes. One conse-

quence is a concentration-dependent blood-to-plasma distri-

bution (in vitro and in vivo) in all species. For concentrations

close to complete target (LTA4H) saturation and above (>250

nM), blood-to-plasma ratios decline to <1, whereas LYS006

was predominantly distributed into blood cells at concen-

trations below 250 nM reaching ratios of 7:1 in human blood.

Whereas formation of LTB was inhibited at concentrations in

agreement with the previously determined IC₅₀of LYS006 on

human PBMCs, and showed >95% inhibition at 1 μM, no

other lipid mediator was signiﬁcantly suppressed by more than

0%, demonstrating the excellent selectivity of LYS006 for

LTA4H (Figure 6 ). As expected, formation of the pro-

Figure 7. In vitro blood/plasma distribution ratios of LYS006. Rising

concentrations of [ H]LYS006 were incubated for 30 min at 37 °C

with blood from diﬀerent species. Total radioactivity in blood and

plasma was determined in triplicate. Values are given as the ratio of

[

H]LYS006 concentration in blood divided by plasma concentration

of [ H]LYS006.

longer preincubation times, blood/plasma ratios became even

larger (data not shown). No concentration-dependency was

observed in blood of LTA4H knockout mice (constant

distribution C /C_pof 0.7), conﬁrming that binding of

LYS006 to the target enzyme was mediating the concen-

tration-dependent blood distribution (Figure 7).

After i.v. dosing, clearance of LYS006 was low for mice, rats,

and dogs (Table 5). In all species, we observed nonlinear

pharmacokinetics with a concentration-dependent volume of

Table 5. In Vitro and In Vivo Pharmacokinetic Parameters

Of LYS006 in Mouse, Rat, and Dog

Figure 6. Lipidomics proﬁle of LYS006-treated, stimulated human

PBMCs. PBMCs were pretreated for 15 min with 1 μM LYS006 or

vehicle control prior to stimulation with calcium ionophore for 2 h.

Lipids were quantiﬁed by LCMS/MS. Concentrations are shown in

ng/mL as the mean ± SD. Depicted is a selection of lipid mediators

from diﬀerent eicosanoid metabolism pathways of a total of 45 tested

lipids. Statistical analysis Welch’s t test. *p < 0.05, **p < 0.01, and

in vitro

mouse

rat

dog

_C_L_i_n_t₍_μ_L_m_i_n−1 _m_g−1

)

9.3

3.4

9.4

10.5

hepatic extraction (%)

male Beagle,

fasted

**p < 0.001. Abbreviations: AA, arachidonic acid; 12-HETE, 12-

in vivo

female C57BL6

3.0 mg/kg i.v.

male Wistar

hydroxyeicosatetraenoic acid; 15-HETE, 15-hydroxyeicosotetraenoic

acid; PGE2, prostaglandin E2; PGD2, prostaglandin D2; TXB2 =

thromboxane B2.

dose

1.0 mg/kg i.v.

3.0 mg/kg p.o.

0.662 (n = 2)

0.1 mg/kg i.v.

0.3 mg/kg p.o.

0.945 ± 0.110

0 mg/kg p.o.

4.12 ± 0.86

(

mL min⁻kg⁻¹

)

resolution lipid LXA₄was elevated upon treatment with

LYS006 due to pathway shunting and conversion of elevated

LTA into LXA . A moderate increase of cysteinyl leukotrienes

V_s_s(L/kg)

T₁_/₂(h)

C_m_a_xd.n. (nM)

T_m_a_x(h)

2.15 ± 0.78

16.7 ± 5.3

689 ± 271

0.25 ± 0

3.29 (n = 2)

169 (n = 2)

1240 ± 232

4.0 ± 0

3.81 ± 0.460

48.8 ± 4.3

306 ± 60.3

5.0 ± 1.7

such as LTC could also be observed due to conversion of

LTA into LTC by LTC -synthase, as expected.

Preclinical PK Proﬁle of LYS006. The pharmacokinetic

proﬁle of LYS006 is characterized by target-mediated drug

Oral BA (%F)

18.2 ± 1.3

1940 ± 139

56.8 ± 9.0

38500 ± 6070

38.7 ± 3.4

17600 ± 1550

AUC p.o. d.n.

(

nM·h)

895

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

distribution and long terminal half-lives; both volume of

distribution and half-life increase with lower systemic

concentrations of LYS006 when compound is mainly

distributed into blood cells.

After oral administration, LYS006 is rapidly absorbed and

reaches C_m_a_xafter 0.25−5 h depending on the species.

Bioavailability based on blood concentrations was moderate

across species and best in rat (57%). These parameters,

however, may be aﬀected by the nonlinear PK properties.

Despite long terminal half-lives no relevant accumulation was

observed in high dose PK and toxicological studies (data not

shown).

Preclinical Pharmacodynamic Studies with LYS006.

We studied the detailed PK/PD relationship of LYS006 in

mice determining the release of LTB in ex vivo stimulated

blood as a proximal readout for target engagement. Mice were

given increasing doses of LYS006 or control compound

Acebilustat in a b.i.d. regimen for 3 days (altogether 5 p.o.

administrations per dose). Blood exposure, plasma exposure

and LTB release were determined at steady state. At an early

time point of 2 h after the last compound administration, doses

as low as 0.625 mg/kg still led to >90% suppression of LTB₄

release at exposures in agreement with the blood IC₉₀

determined in human and mouse whole blood assays.

Acebilustat showed similar levels of inhibition but at higher

blood and plasma exposures. At 14 h after the last dosing,

LYS006 maintained strong inhibition of LTB release despite

very low plasma exposures. Under these conditions, the

compound exhibited a very high blood/plasma ratio, which

was even more pronounced than that observed in the in vitro

blood/plasma distribution assay described before. IC₉₀was

achieved at plasma exposures of only 8 nM and blood exposure

of 130 nM, again in good agreement with the IC₉₀values

determined in human whole blood. Acebilustat did not reach

IC₉₀despite considerably higher blood and plasma exposures,

again in good agreement with previously determined blood

IC₉₀values (Figure 8).

Figure 8. Steady-state PK/PD relationship of LYS006 (blue) and

Acebilustat (green). Female C57BL/6 mice (n = 4 per group) were

dosed p.o. ﬁve times in a b.i.d. regimen. Compound exposures in

blood and plasma, and LTB inhibition in ex vivo stimulated blood

were determined 2 h (upper panel) and 14 h after the last dose (lower

panel). Shown are average exposures in blood (black numbers above

graphs) and plasma (green numbers above graph), and the means of

LTB₄inhibition ± SEM (PD). One way ANOVA followed by

Dunnett’s multiple comparison correction was applied with **p <

Preclinical Eﬃcacy in Neutrophilic Skin Inﬂammation

Models. Next, we set out to study how LTB inhibition

translates into suppression of LTB dependent neutrophil

inﬂux and activation. Therefore, we chose the arachidonic acid

skin inﬂammation model and investigated the correlation

0.01 and ***p < 0.001.

between LTB inhibition in ex vivo stimulated blood and

neutrophil counts in arachidonic acid stimulated skin (Figure

). LYS006 dose dependently inhibited ex vivo LTB₄

tion in this neutrophil and macrophage-driven disease model.

formation. At doses >1 mg/kg, LTB4 production was inhibited

Here, the model served as a mechanistic experiment to study

the pharmacology of LYS006 and the relevance of LTA4H

inhibition in resolving established neutrophilic inﬂammatory

conditions.

For the prophylactic setting, female mice were dosed orally

in a b.i.d. regimen (10, 14 h intervals) beginning on day

minus 1, with doses of 3, 10, or 30 mg/kg until day 6. On day

0, arthritis was induced by i.v. injection of KRN serum. Eﬃcacy

was evaluated via a scoring system, assessing swelling of all four

paws in LYS006 treated versus vehicle treated animals (Figure

10).

by at least 90%. The degree of LTB inhibition in ex vivo

stimulated blood correlated well with the level of MPO

reduction in skin homogenates, indicating that the peripheral

PD readout may be a good predictor for the anti-inﬂammatory

eﬀect of LTA4H inhibition in skin. The level of LYS006 in

blood required to achieve >90% inhibition of LTB in blood

and >90% reduction of MPO levels in the skin was

approximately 130 nM, in agreement with the values

determined in the PK/PD model and in in vitro whole blood

assays.

Preclinical Eﬃcacy in Prophylactic and Therapeutic

Setting in the KRN Arthritis Model. To extend our

preclinical pharmacology studies with LYS006, we investigated

the eﬃcacy of the drug in the mouse KRN passive serum-

induced inﬂammatory arthritis model in a prophylactic and a

The model showed eﬃcient and dose-dependent reduction

of composite paw swelling scores, leading to half-maximal

reduction in the 3 mg/kg group and maximal eﬃcacy at 10 and

30 mg/kg. The target proximal PD marker LTB was assessed

in blood 2 and 14 h after the ﬁnal dose and demonstrated

therapeutic setting. It has been demonstrated that LTB is a

partial target engagement with the 3 mg/kg dose (∼60% LTB

critical mediator for the onset and maintenance of inﬂamma-

reduction 14 h after dosing) and ≥90% reduction of LTB

896

Journal of Medicinal Chemistry

J. Med. Chem. 2021, 64, 1889−1903

Figure 10. Eﬃcacy of LYS006 administered prophylactically in the

mouse KRN passive serum-induced arthritis model. Upper panel: Paw

swelling scores of C57BL/6 mice (n = 6 per group) during arthritis

are shown as means ± SEM. Treatment with LYS006 was started in a

b.i.d. regimen 1 day before disease induction (day 0). Statistical

analysis (ANOVA, followed by Dunnett’s test) indicates *p < 0.05

Figure 9. Suppression of neutrophilic inﬂammation in arachidonic

acid induced skin inﬂammation model. Female 129S2/SvPasCrl mice

were dosed p.o. ﬁve times in a b.i.d. regimen. At 3 h after the last dose,

inﬂammation in the skin was induced by topical arachidonic acid

challenge for another 3 h. Ex vivo stimulated LTB levels in blood

(

upper panel) and tissue MPO levels (lower panel) were determined.

and **p < 0.01. Lower panel: LTB

whole blood taken 2 h (gray bars) and 14 h (blue bars) after the ﬁnal

dose at day 6. Shown are the mean LTB levels ± SEM. Numbers

generation in ex vivo stimulated

Shown are the means ± SEM, n = 5 per group. One way ANOVA

followed by Dunnett’s multiple comparison correction was applied

with *p < 0.05 and **p < 0.001.

above bars indicate compound exposures in blood at indicated dose

and time point.

throughout the entire dosing interval with the 10 and 30 mg/

kg doses. These results clearly indicate that a high target

engagement of at least 90% is required for the entire dosing

interval to achieve maximal eﬃcacy. A drop in target

engagement at trough directly translated into only partial

reduction of overall disease severity.

Given the good eﬃcacy of LYS006 in preventing joint

inﬂammation in the prophylactic setting of the KRN arthritis

model, a subsequent experiment was set up to assess its ability

to resolve already established neutrophilic joint inﬂammation

CHEMISTRY

■

The synthesis of LYS006 [(S)-22] is summarized in Scheme 1.

Diaryl ether 24 was obtained from ﬂuoropyridine 25 and

phenol 26 via S Ar reaction and further converted into

tetrazole 27 using a tin-catalyzed cycloaddition with TMS-

azide. The subsequent alkylation installed the headgroup and

occurred with high regioselectivity at the distal nitrogen of the

tetrazole. Enantiopure (S)-28, derived from natural L-aspartic

acid, serves as the source of chirality and can be coupled

directly using Mitsunobu conditions. On scale, we preferred

activation to electrophilic sulfamidate (S)-29, which signiﬁ-

cantly simpliﬁed puriﬁcation of protected (S)-30. Final

deprotection under acidic conditions aﬀorded zwitterionic

LYS006 as the HCl salt.

(Figure 11). LYS006, dosed p.o. at 10 mg/kg b.i.d starting at

day 4 immediately stopped the progression of paw swelling and

from day 7 onward even reduced the severity. In addition, we

observed signiﬁcantly reduced histological parameters of

inﬂammation and tissue destruction, as well as clearance of

neutrophils from the inﬂamed tissue. In contrast, the vehicle

group showed further increase in disease severity. The ability

of LYS006 to revert established inﬂammation in this model

CONCLUSION

■

underlines the importance of LTB in the perpetuation of the

Using a focused fragment-based DSF screen, followed by

cocrystallization and structure-guided fragment merging, we

identiﬁed a potent hit, which we succeeded in developing

further into our very potent and selective LTA4H inhibitor

LYS006. The key step in the hit to lead optimization was the

inﬂammatory response. Eﬃcient inhibition of LTA4H may

therefore possess the ability to resolve established chronic

inﬂammation, particularly in inﬂammatory diseases with a

strong neutrophilic component.

897

Journal of Medicinal Chemistry

J. Med. Chem. 2021, 64, 1889−1903

Figure 11. Eﬃcacy of LYS006 in a therapeutic KRN model: LYS006 reduces clinical scores, histological parameters, and numbers of neutrophils in

tissue. Upper left panel: Paw swelling scores of C57BL/6 mice (n = 6) during arthritis are shown as means ± SEM. Treatment with LYS006 was

started in a b.i.d. regimen with a 10 mg/kg dose on day 4 after disease induction. Statistical analysis (ANOVA, followed by Dunnett’s test) indicates

p < 0.05 and **p < 0.01. Upper right panel: Histology scores combined for the hind paws of all animals were assessed for cellularity, joint damage,

proteoglycan loss, and neutrophil numbers in the joint. Data are presented as the means ± SEM. Statistical analysis: Unpaired t test with *p < 0.05,

*p < 0.01, and ***p < 0.001. Lower panel: Photomicrographs of hind paw sections stained with H&E (left column), Safranin O (cartilage stained

red, middle column), and Ly-6B2 antibody (neutrophils stained brown, right column) illustrate joint inﬂammation and damage and preservation

after LYS006 treatment.

elaboration of a primary amine headgroup into a zwitterionic

β-amino acid. This signiﬁcantly improved human whole blood

potency and microsomal stability while eliminating inhibition

of cytochrome P450s or the hERG channel. Lead optimization

was instructed by early employment of an ex vivo LTB₄

stimulation assay for selection of compounds with long lasting

target engagement/PD eﬀect in vivo. This, together with the

optimization of physicochemical properties ultimately favored

LYS006 over related analogues. LYS006 exhibits exquisite

selectivity in comprehensive oﬀ-target proﬁling. Furthermore,

it does not inhibit eicosanoid metabolizing enzymes other than

LTA4H and elevates the biosynthesis of pro-resolving LXA₄

upon LTB inhibition. It shows excellent in vitro potency in

biochemical and cellular assays. Due to its high blood/plasma

ratio at low exposures, LYS006 achieves complete target

inhibition in vivo with plasma exposures in the single digit

898

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

Drug Annotation

Synthesis of LYS006 [(S)-22]. 4-((5-Chloro-3-ﬂuoropyridin-

-yl)oxy)benzonitrile (24). A colorless suspension of 5-chloro-2,3-

Scheme 1. Synthesis of LYS006 [(S)-22]

diﬂuoropyridine (25, 25.0 g, 167 mmol), 4-hydroxybenzonitrile (26,

9.0 g, 160 mmol), and K CO (26.5 g, 191 mmol) in DMF (100

mL) was heated to 50 °C for 16 h. The reaction mixture was

partitioned between heptane/EtOAc (2:1, 480 mL) and water (200

mL). The organic phase was washed with aq. NaHCO (8%, 200 mL)

and brine (200 mL), dried over MgSO , ﬁltered, and concentrated in

vacuo to aﬀord nitrile 24 (37.7 g, 93%) in 98% purity. H NMR (600

MHz, DMSO-d ): δ 8.31 (d, J = 9.8 Hz, 1H), 8.13 (s, 1H), 7.94 (d, J

8.2 Hz, 2H), 7.44 (d, J = 8.2 Hz, 2H). F NMR (376 MHz,

DMSO-d ): δ −133.65 (d, J = 11 Hz, 1F). MS (ESI ± ): m/z =

249.0/-247.1.

-(4-(Tetrazol-5-yl)phenoxy)-5-chloro-3-ﬂuoropyridine (27).

This reaction was executed in nine vials in parallel: To a solution of

nitrile 24 (2.24 g, 9.0 mmol) in dry toluene (9 mL) were added

dibutylstannanone (224 mg, 0.90 mmol) and azidotrimethylsilane

(2.4 mL, 18 mmol). The vials were sealed and heated to 90 °C for 18

h. After cooling to rt, all volatiles were removed in vacuo. and the

combined crudes were triturated with MeOH (100 mL) and

concentrated again. The material was suspended in MeOH (100

mL) and ﬁltered. The solids were washed with heptane (100 mL) and

Reagents and conditions: (a) K CO , DMF, 50 °C, 16 h, 93%; (b)

Bu SnO (10%), TMSN , toluene, 90 °C, 18 h, 77%; (c) (S)-29,

DMF, DIPEA, rt, 17 h, 77%; (d) HCl in dioxane, 45 °C, 26 h, 78%;

pentane (50 mL) and dried in vacuo to aﬀord tetrazole 27 as a

(

e) (i) SOCl , imidazole, 2-MeTHF, −78 °C to rt, 2 h; (ii) NaIO ,

colorless powder (18.5 g, 77%). H NMR (400 MHz, DMSO-d ): δ

cat. RuCl ·H O, MeCN-water, 0 °C, 2 h, 79% (two steps).

6.87 (s, 1H), 8.29 (dd, J = 9.9, 2.2 Hz, 1H), 8.15−8.06 (m, 3H),

.50−7.42 (m, 2H). F NMR (376 MHz, DMSO-d ): δ −133.98 (d,

nanomolar range, translating into nearly complete suppression

of inﬂammation in both, a neutrophil skin inﬂammation and

the KRN passive serum induced arthritis model. In summary,

in view of its excellent selectivity and in vitro/in vivo potency

proﬁle, LYS006 has potential for a best-in-class LTA4H

inhibitor.

LYS006 displayed a good preclinical tolerability and safety

proﬁle in two species tested, and was therefore progressed into

clinical studies where its good safety and tolerability were

conﬁrmed in single and multiple ascending dose phase I

clinical trials. The compound is currently being tested in

several inﬂammatory diseases in phase II clinical trials,

including inﬂammatory acne, hidradenitis suppurativa, ulcer-

ative colitis, and nonalcoholic steatohepatitis. Further details

on clinical properties of LYS006 and results of interventional

trials in disease will be reported in due course.

J = 7.5 Hz, 1F). MS (ESI ± ): m/z = +292.1/-290.1.

tert-Butyl (S)-4-(2-(tert-Butoxy)-2-oxoethyl)-1,2,3-oxathiazoli-

dine-3-carboxylate 2,2-dioxide [(S)-29] Step 1. A solution of

imidazole (29.7 g, 436 mmol) in 2-MeTHF (300 mL) was cooled

to −78 °C and treated dropwise with SOCl (7.95 mL, 109 mmol).

After 10 min, (S)-tert-butyl 3-((tert-butoxycarbonyl)amino)-4-hydrox-

ybutanoate (28, 10.0 g, 36.3 mmol) in 2-MeTHF (50 mL) was added

dropwise. Cooling was removed and the reaction mixture was stirred

at rt for 2 h, before it was ﬁltered over a pad of Celite. All volatiles

were removed in vacuo, and the residue was partitioned between

DCM (100 mL) and water (100 mL). The aqueous phase was

extracted with DCM (2 × 50 mL), and the combined organic layers

were washed with aq. HCl (10%, 20 mL) and brine (20 mL), dried

over MgSO , ﬁltered, and concentrated in vacuo to aﬀord a pale

yellow liquid (10.7 g) that was used directly in the next step. MS (ESI

): m/z = +322.2/-320.3.

Step 2. The residue was dissolved in MeCN (150 mL), cooled to 0

C, and treated with portions of solid RuCl monohydrate (328 mg,

.45 mmol) and NaIO (11.7 g, 54.5 mmol), followed by dropwise

EXPERIMENTAL SECTION

■

addition of water (75 mL). After stirring at 0 °C for 2 h, the reaction

mixture was partitioned between EtOAc (100 mL) and water (20

mL). The aqueous phase was extracted with EtOAc (2 × 50 mL), and

General. All chemicals were used directly as received from

commercial suppliers. Reference compounds DG-051, JNJ-40929837,

and Acebilustat were synthesized according to published proce-

the combined organic layers were washed with sat. NaHCO (50 mL)

36−38

dures.

NMR spectra were recorded on Bruker 400 or 600 MHz

and brine (50 mL). The gray organic phase was ﬁltered successively

spectrometers. Chemical shifts are expressed in δ ppm referenced to

the residual solvent peak (DMSO-d , δ = 2.50 ppm; CDCl , δ = 7.27

over plugs of Celite, Na SO , and silica until clear and colorless.

Removal of all volatiles in vacuo aﬀorded the title compound 29 (10.2

ppm; CD OD, δ = 3.31 ppm). Abbreviations used in describing peak

g, 79%) as a colorless solid. H NMR (400 MHz, CDCl ): δ 4.78

signals are br = broad signal, s = singlet, d = doublet, t = triplet, q =

quartet, m = multiplet. All ﬁnal compounds were puriﬁed to ≥95% by

reverse phase high performance liquid chromatography (HPLC),

supercritical ﬂuid chromatography (SFC), normal or reverse phase

column chromatography. Purities were assessed by reverse phase

UPLC with a gradient of 5−60% isopropanol in water (with either

acid or base modiﬁer) and monitored by UV absorption at 210−450

nm. Optical purity was determined by HPLC (Agilent 1100 COOK)

on chiral stationary phase. Low-resolution mass spectra were recorded

on a liquid chromatography−mass spectrometer in electrospray

positive (ESI+) and negative (ESI−) modes. HR-MS analyses were

performed by using electrospray ionization in positive ion mode after

separation by liquid chromatography (UltiMate 3000 from Dionex).

The elemental compositions were derived from the mass spectra

acquired at a high resolution of about 70000 on a QExactive Plus

mass spectrometer (Thermo Scientiﬁc). The high mass accuracy

below 1 ppm was obtained by using a lock mass.

(ddd, J = 9.6, 5.9, 1.3 Hz, 1H), 4.64−4.56 (m, 1H), 4.53 (dd, J = 9.6,

2.2 Hz, 1H), 3.02 (dd, J = 17.0, 3.4 Hz, 1H), 2.77 (dd, J = 16.9, 10.9

Hz, 1H), 1.58 (s, 9H), 1.48 (s, 9H).

tert-Butyl (S)-3-((tert-Butoxycarbonyl)amino)-4-(5-(4-((5-

chloro-3-ﬂuoropyridin-2-yl)oxy)phenyl)-2H-tetrazol-2-yl)-

butanoate [(S)-30]. A clear yellow solution of tetrazole 27 (73.0 g,

213 mmol) and sulfamidate 29 (75.0 g, 223 mmol) in DMF (620

mL) was treated with DIPEA (111 mL, 638 mmol) and stirred at rt

for 17 h. The reaction mixture was partitioned between EtOAc (1 L)

and 10% citric acid (300 mL). The aqueous phase was extracted with

EtOAc (500 mL), and the combined organic layers were washed with

brine (700 mL), dried over MgSO

vacuo. The crude material was puriﬁed by preparative MPLC (7 kg

SiO

, ﬁltered, and concentrated in

, 25−40 μm, heptane to EtOAc) to aﬀord (S)-30 as a colorless

powder (91.2 g, 77%). H NMR (600 MHz, DMSO-d ): δ 8.29 (dd, J

= 9.9, 2.2 Hz, 1H), 8.14−8.07 (m, 3H), 7.43−7.38 (m, 2H), 7.03 (d, J

= 9.0 Hz, 1H), 4.85 (dd, J = 13.6, 4.3 Hz, 1H), 4.64 (dd, J = 13.6, 8.5

899

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

Drug Annotation

Hz, 1H), 4.36−4.27 (m, 1H), 2.63 (dd, J = 15.7, 5.5 Hz, 1H), 2.44

using as ﬁlter settings λ_e_x_c_i_t_a_t_i_o_n= 380 nm and λ_e_m_i_s_s_i_o_n= 460 nm. For

accurate determination of IC₅₀values of key compounds, 12-point

serial dilutions of compound were tested in an assay format with

reduced enzyme concentration (0.5 nM) and otherwise identical assay

conditions as described above.

(

dd, J = 15.7, 9.1 Hz, 1H), 1.40 (s, 9H), 1.25 (s, 9H). ¹⁹F NMR (376

MHz, DMSO-d ): δ −134.02 (d, J = 7.5 Hz, 1F). MS (ESI+/−): m/z

+549.3/−547.4.

S)-3-Amino-4-(5-(4-((5-chloro-3-ﬂuoropyridin-2-yl)oxy)-

(

phenyl)-2H-tetrazol-2-yl)butanoic acid hydrochloride [(S)-22,

LYS006]. Solid (S)-30 (108 g, 197 mmol) was suspended in 4 M

HCl in dioxane (0.95 L) and stirred at 45 °C, until no more starting

material was detectable by UPLC-MS (26 h). All volatiles were

removed in vacuo. The crude material was suspended in MeCN,

stirred for 20 min at 45 °C, cooled to 10 °C, ﬁltered, and washed with

additional MeCN (2 × 250 mL). The colorless solid was dissolved in

Human PBMC Eicosanoid Release Assay. Fresh blood was

collected in heparinized vacutainers by venipuncture from healthy

volunteers in accordance with the guidelines of the Novartis Blood

Donor Center, Basel. Informed consent of all blood donors was

obtained and experimental protocols approved by the Novartis Blood

Donor Center. Mononuclear cells from human peripheral blood were

isolated by density gradient centrifugation from 1:1 diluted blood with

RPMI 1640 and layered onto Ficoll-Paque. Mononuclear cells were

carefully collected from the interphase. Erythrocytes were lysed with

red blood cell lysis buﬀer (Miltenyi). Cells were counted using

Trypan Blue staining on a Countess automated cell counter to

determine number of viable cells. A volume of 200 μL of PBMC (2 ×

water/MeCN (5:2, 700 mL) and lyophilized to aﬀord 68 g (78%) of

(

S)-22 as a colorless solid. H NMR (600 MHz, CD OD): δ 8.25−

.19 (m, 2H), 7.96 (d, J = 2.2 Hz, 1H), 7.90 (dd, J = 9.4, 2.2 Hz, 1H),

.36−7.31 (m, 2H), 5.16 (d, J = 5.7 Hz, 2H), 4.30−4.26 (m, 1H),

.95 (dd, J = 18.0, 5.9 Hz, 1H), 2.80 (dd, J = 17.9, 6.9 Hz, 1H).

NMR (151 MHz, CD OD): δ 170.82, 164.79, 155.29, 150.09 (d, J =

10 cells/mL) was transferred to 96-well cell culture plates. All

1.8 Hz, 1C), 147.22 (d, J = 265 Hz, 1C), 139.86 (d, J = 5.9 Hz, 1C),

27.85 (2C), 126.00, 125.31 (d, J = 19.1 Hz, 1C), 123.50, 120.85

compounds were tested in 11-point dose−response curves, and the

ﬁnal compound concentration range was from 5000 to 4.9 nM in a

1:2 dose titration. Cells and compounds were incubated for indicated

preincubation times at 37 °C in a humidiﬁed incubator. Cells were

then stimulated with 1 μg/mL calcium ionophore A23187 (Sigma) or

an equal volume of DMSO (control), mixed and incubated for an

additional 2 h at 37 °C in a humidiﬁed incubator. Cell supernatants

(

2C), 52.93, 47.03, 33.17. F NMR (376 MHz, CD OD): δ −135.63

d, J = 9.4 Hz, 1F). 99.7% ee (Agilent 1100 COOK, Chiracel IC,

heptane/EtOH/MeOH 80/10/10 + 0.1% HNEt + 0.1% TFA, 1.0

mL/min, 249 nm). HRMS (ESI+): m/z calcd for C H ClFN O [M

H] , 393.08727; found, 393.08728.

Diﬀerential Scanning Fluorimetry (DSF) Screening Assay. A

were taken for LTB determination by competitive immunoassays EIA

DSF assay was run in 384-well microtiter well plates on a CFX384

real-time PCR instrument (Bio-Rad, Hercules, CA). Heat induced

unfolding of LTA4H was monitored by measuring the ﬂuorescence

increase of the solvatochromic dye SYPRO Orange (ThermoFisher

Scientiﬁc, Waltham, MA) under the following assay conditions: 500

μM fragment concentration, 50 mM HEPES pH 7.4, 50 mM NaCl, 1

or 2% DMSO, 1.1 μM LTA4H, 1:1000 dilution of commercial

SYPRO Orange stock (5000×), temperature gradient 25−90 °C,

heating rate 1 °C/min, increment 0.5 °C, FRET channel (excitation

(

Assay designs). Concentrations of LTB in cell supernatants were

quantiﬁed according to the manufacturers manual. OD was

determined at 405 nm on a SpectraMax Paradigm (Molecular

Devices, USA).

Human Whole Blood Eicosanoid Release Assay. Compounds

were tested in 11-point dose response curves and the ﬁnal compound

concentration range was from 5000 to 4.9 nM in a 1:2 dose titration.

Fresh blood was collected in heparinized vacutainers by venipuncture

from healthy volunteers of the Basel Tissue Donor program. Blood

was diluted 1:2 with RPMI 1640 and incubated for the indicated

times with compounds at 37 °C in a humidiﬁed incubator. Then, the

blood was stimulated with 10 μg/mL calcium ionophore A23187

(Sigma) or equal volume DMSO (control), mixed well and incubated

for an additional 15 min at 37 °C in a humidiﬁed incubator. Plasma

supernatants were taken for eicosanoid determination by competitive

immunoassays EIA (Assay designs). Concentrations of LTB₄in

50−490 nM, emission 560−580 nm). Screening plates were

incubated for 30 min at room temperature prior to measurement.

Each screening plate contained 16 control wells (LTA4H with 1%

DMSO) to monitor assay quality. Melting temperatures (T ) were

extracted by plotting ﬂuorescence against temperature. The inﬂection

point of the ﬁrst derivative of the melting curve represents T_m.

Melting temperature shifts (ΔT ) were calculated as diﬀerence to

DMSO control. A ΔT > 3-fold the standard deviation of

plasma supernatants were quantiﬁed according to the manufacturers

manual. OD was determined at 405 nm on a SpectraMax Paradigm

(Molecular Devices, USA)

was classiﬁed as signiﬁcant. Standard deviations for

m(LTA4H/DMSO)

T_m₍_L_T_A₄_H_/_D_M_S_O₎were in the range of 0.1 °C for all experiments.

Screening hits identiﬁed to stabilize LTA4H were conﬁrmed by DSF

using a 4-point concentration series of 125, 250, 500, and 1000 μM

fragments.

Lipidomics Proﬁling of LYS006 Treated Human PBMCs by

LCMS/MS. Human PBMCs were isolated as described above. A

volume of 1100 μL of PBMCs (4 × 10 cells/mL) was transferred to

hERG Binding Assay. HEK293 cells stably transfected with the

HERG-1 gene (Swiss-Prot: Q12809) were obtained under a license

from the Wisconsin Alumni Research Foundation. Membranes

prepared from these cells were used in the binding assay using the

24-well ﬂat bottom cell culture plates. For LYS006 testing, a 1:10

serial dilution was prepared with concentrations ranging from 10 000

to 10 nM. Cells were incubated with LYS006 for 15 min at 37 °C in a

humidiﬁed incubator. Cells were then stimulated with 1 μg/mL

calcium ionophore A23187 (Sigma) or an equal volume of DMSO

(control), mixed, and incubated for an additional 2 h at 37 °C in a

humidiﬁed incubator. Incubation was terminated by centrifugation at

300g for 10 min at 22 °C, and supernatants (1000 μL) transferred

into 2 mL Eppendorf tubes. A volume of 800 μL of methanol

containing deuterium-labeled internal standards (ISTDs) at the ﬁnal

concentration of 5 nM was added to each sample and immediately

frozen on dry ice. Samples were stored at −80 °C until LCMS

analysis. For lipid extraction, samples were acidiﬁed with formic acid

(FA) at the ﬁnal concentration of 1% and centrifuged at 16 000g for

10 min at 4 °C. Samples were loaded and diluted (800 μL water) in a

previously methanol equilibrated SPE SOLA HRP plate (Thermo

Scientiﬁc). After loading, the SPE plate was washed with 2 mL of

water and lipids were eluted with 0.5 mL of acetonitrile. The eluted

fraction was evaporated under nitrogen ﬂow at room temperature for

approximately 1 h. Samples were then reconstituted in methanol/

water (50%/50%; v/v). The injection volume was 10 μL. The LCMS/

MS method used for data acquisition has been recently described in

radioligand [ H]dofetilide with speciﬁc activity of 46.5 Ci/mmol. The

assay contains in a 96-well Millipore GF/C ﬁlter plate 119 μL of

buﬀer, 1 μL of compound, 40 μL of 12.5 nM [ H]dofetilide, and 40

μL of crude membrane suspension (15 μg protein). After 90 min at

room temperature, incubations were terminated by rapid ﬁltration and

three washes with ice-cold buﬀer. Filters were partially dried at 40 °C

for 15 min, then 40 μL of scintillant was added, and the plate was

sealed. Plates were read in a Wallac MicroBeta Trilux β counter.

Enzymatic Screening Assay. In screening mode all compounds

were tested in 8-point serial dilutions in 384-well plates in duplicate

with compound concentrations starting at 50 000 nM. LTA4H (9

nM) was preincubated with compound for 15 min at room

temperature in assay buﬀer (50 mM Tris buﬀer, pH 7.5, 150 mM

NaCl, 10 mM CaCl , Sigma, Switzerland) prior to addition of

substrate (600 μM Arg-AMC (Bachem, Switzerland). Upon addition

of the substrate, the plate was immediately placed in a ﬂuorescence

reader (SpectraMax Paradigm, Molecular Devices) and the

ﬂuorescence was measured every 10 min for 300 min at 25 °C

900

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

The Supporting Information is available free of charge at

Drug Annotation

detail and employed in this study with a few modiﬁcations. Brieﬂy,

the analytical platform consisted of a Waters H Class UPLC system

coupled to a Waters XevoTQS triple quadrupole mass spectrometer

equipped with an electrospray ionization (ESI) source. Separation of

lipids was achieved using reverse-phase column (ACQUITY UPLC

BEH C18 1.7 μm, 2.1 × 50 mm) heated at 40 °C. The ﬂow rate was

set to 0.180 mL/min, and the mobile phase consisted of water/formic

acid (99.9/0.1; v/v) (A) and methanol/acetonitrile/formic acid (B)

hours after the ﬁnal compound application, mice were topically

treated on left ears with 10 μL of acetone and on right ears with 10 μL

of arachidonic acid (CalBiochem) solution (100 mg/mL). At 3 h post

sensitization and 6 h post drug administration, mice were terminated

and blood was collected in heparinized tubes for PD analysis and

EDTA coated tubes for PK analysis as described above. Ear punches

(6 mm) were collected from both ears, weighted and frozen at −80

°C for subsequent determination of tissue MPO levels. To this end,

(

50/50/0.05; v/v/v). The eluent gradient was set from 45%−65% of

ear punches were cut in 1 mm squares and immediately transferred

B during 0−16 min, from 65%−100% during 16−19 min and 100% of

B from 19 to 24 min. From 24 to 35 min the column was re-

equilibrated with initial conditions. Ionization was performed using an

ESI source operated in the negative ion mode. Fragmentation was

tuned for each compound individually using Intellistart (Waters). The

source parameters were set to source temperature, 150 °C; cone gas

ﬂow, 150 L/h; nebulizer gas ﬂow, 7 bar; desolvation temperature, 350

into homogenization tubes preﬁlled with 250 μL of lysis buﬀer (200

mM NaCl, 6.4 mM EDTA, 10 mM Tris-HCl pH 7.5, 11.5% v/v

glycerin, + protease inhibitor cocktail). Samples were homogenized

with “Precellys 24” 3 × 15 s at 5500 rpm, and homogenates were

centrifuged for 30 min at 4 °C at 13 000 rpm. Supernatants were

frozen at −80 °C until MPO quantiﬁcation with ELISA (Hycult

biotech), according to the manufacturer’s manual.

C; desolvation gas ﬂow, 600 L/h; capillary voltage, 2.5 kV; collision

KRN Passive Serum Induced Arthritis Model in Mouse.

Female 8 week old C57BL/6 mice (Janiver, France) were used for

these studies. To induce arthritis in these mice, 200 μL of serum from

KRN mice (pool, received from Genomics Institute of the Novartis

Research Foundation (GNF), San Diego, CA, USA) was injected i.v.

to recipient mice. Swelling of hind paws was scored three times a

week. Scoring was done on a composite scale of 0−24 per mouse,

evaluating each paw by visual inspection in the metatarsal region with

a score of 0−3 and in the ankle with a score of 0−3, thus obtaining a

maximal score of 6 per paw. The following scoring system used was: 0

= no detectable signs of inﬂammation; 1 = lightly swollen region; 2 =

more obviously swollen region; 3 = ankyloses or severely swollen

region. The paw scores were summed to obtain a score for each

individual animal. Prophylactic treatment of twice daily dosing started

on day −1 and continued until day 6 with either vehicle or compound

treatment at indicated doses. KRN serum was applied i.v. on day 0.

On day 6, animals were sacriﬁced. Hind paws were taken for histology

and blood for PK/PD analysis as described above. For therapeutic

treatment, 200 μL of KRN serum was administered via i.v. injection

on day 0. On day 4, mice were distributed to diﬀerent groups so that

each group had equal swelling prior to compound treatment.

Therapeutic treatment started on day 4 and continued until day 12

with b.i.d. dosing. On day 12, animals were sacriﬁced, blood was

collected for PK/PD analysis, and hind paws were collected for

histology.

gas ﬂow, 0.25 mL/min. During data acquisition, the MS program was

divided in diﬀerent functions in which deﬁned MRM transitions were

monitored according to the elution of lipid mediators. Quantiﬁcation

and data analysis of lipids was carried out essentially as described

previously.

In Vitro Blood/Plasma Distribution Assay. Blood was collected

from female mice (C57/BL6), female LTA4H KO mice (C57/BL6),

male rats (Hanover Wistar), male dogs (beagle), and male Caucasian

volunteers, spiked with raising concentrations of [ H]LYS006, and

incubated for 0.5 h at 37 °C. Plasma fractions were obtained after cell

separation by centrifugation (10 min, 1500g, 37 °C), and blood

samples were taken before centrifugation. Total radioactivity was

determined by liquid scintillation counting in triplicates (Packard

Tricarb LSC). Concentration of [ H]LYS006 was calculated based on

the determination of total radioactivity using the speciﬁc radioactivity.

Concentrations of radiolabeled LYS006 in blood distribution

experiments were determined in weighed samples. Data were

converted from Bq/g to ng/mL assuming a density of 1.00 g/mL

for all samples and using the speciﬁc radioactivity of the test

compound.

Animal Studies. All animal studies were conducted in accordance

with the Swiss Animal Welfare Regulations and speciﬁcally according

to animal licenses BS-2053, BS-2798, BS-2181, BS-2793, BS-2639,

and BS-2640 issued by the Basel-Stadt Cantonal Veterinary Oﬃce.

PK Studies. For intravenous administration compounds were

formulated at indicated doses and dissolved in PEG200:NMP (70:30

v/v), followed by application via the vena femoralis of female C57/

BL6 mice. For intravenous administration to male Wistar rats

compounds were prepared in NMP:PEG200:NaCl (0.9%) (5:30:65

v/v/v). For oral gavage compounds were formulated in propylengly-

col: VitE TPGS (20:10 v/v). The clear solution was diluted (30:70 v/

v) with 100 mM Na-citrate buﬀer pH 3. At selected time points (0.25,

Histopathology. Hind paw samples were ﬁxed in 10% buﬀered

formalin for 48 h, decalciﬁed over 6 days in Immunocal (Decal

Chemical Corp,), changed every second day, processed, and

embedded in paraﬃn. Sections of 3 μm thickness were stained with

hematoxylin and eosin (HE) and Safranin O (SafO). Histopatho-

logical changes were blindly scored on a scale of 0 (normal) to 3

(

severe changes). The following parameters were assessed: inﬂam-

matory cell inﬁltrate, joint damage, and proteoglycan loss.

Immunohistochemical staining for neutrophil-granulocytes was

performed on a Ventana Discovery XT stainer (Roche Diagnostics)

with monoclonal rat anti-Ly-6B2 (clone 7/4, AbD Serotec), followed

by goat anti-rat IgG (Jackson Immunoresearch) and DapMap

detection system (Roche Diagnostics).

.5, 1, 2, 4, 8, 24 h post dose for both routes), 20 μL whole blood was

collected with heparin coated minicaps via the vena saphena by

puncture. The blood was stored at −80 °C in EDTA tubes until PK

analysis by LCMS/MS. For intravenous application in dogs,

compounds were dissolved in PEG200:NMP (90:10 v/v) and applied

via the vena cephalica. For oral gavage the compound was

administered in 0.5% Tween80, 0.5% methyl cellulose in 99%

water. Blood was collected predose and at selected time points for i.v.

at 0.083, 0.25, 0.5, 1, 2, 4, 7, 24, 48, and 120 h and for p.o. at 0.25, 0.5,

ASSOCIATED CONTENT

sı Supporting Information

■

https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c01955.

, 2, 4, 7, 24, 48, and 120 h post dose, respectively.

PD Proﬁling. All experiments were performed with adult female

C57/BL6 mice (Charles River, France). Animals were dosed p.o. with

the indicated doses of compounds for the indicated times. For PD

analysis, heparinized animal blood was collected and diluted 1:2 with

RPMI medium. Blood was then stimulated and released LTB₄

quantiﬁed as described in the section “Human Whole Blood

Eicosanoid Release Assay”.

Murine Model of Arachidonic Acid Induced Skin Inﬂam-

mation. The experiment was performed with adult female 129S2/

SvPasCrl mice. Compounds were applied p.o. in a b.i.d regimen for 3

days (in total 5 doses) to achieve steady state drug exposures. Three

Synthesis of compounds 2−23; structural biology data

(PDF)

Molecular formula strings (CSV)

Accession Codes

PDB ID codes: 1, 7AV2; 2, 7AV1; (R)-13, 7AV0; (S)-22 =

LYS006, 7AUZ. Authors will release the atomic coordinates

and experimental data upon article publication.

901

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

orcid.org/0000-0003-1857-9885 ;

Drug Annotation

AUTHOR INFORMATION

Till A. Ro

Inﬂammation, Novartis Institutes for Biomedical Research,

002 Basel, Switzerland

hn − Autoimmunity, Transplantation and

■

Corresponding Author

Christian Markert − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland;

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.0c01955

Email: christian.markert@novartis.com

Notes

Authors

The authors declare the following competing ﬁnancial

interest(s): All work described was funded by Novartis

Institutes for BioMedical Research Inc. All authors are or

were employees of Novartis and may hold company stocks or

stock options with Novartis Pharma AG.

Gebhard Thoma − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

Honnappa Srinivas − Chemical Biology and Therapeutics,

Novartis Institutes for Biomedical Research, 4002 Basel,

Switzerland

Birgit Bollbuck − Global Discovery Chemistry, Novartis

ACKNOWLEDGMENTS

■

Institutes for Biomedical Research, 4002 Basel, Switzerland

We thank J. G. Ternois, L. Wohlgensinger, P. Nganga, J. Wiest,

N. Eckert, F. Gruber, O. Decoret, C. Rolando, E. Bucceri,

E. Aebischer, T. Hurni, E. Hugenschmidt, H. Lehmann,

T. Ruppen, and E. Schueller for their synthesis support,

F. Freuler, L. Leder, and E. Koch for cloning and protein

expression of LTA4H, O. Esser for his technical support of the

DSF screen, P. Ramseier, M. Vogelsanger, S. Sarret, and

R. Keller for their support of the in vivo studies, N. Caesar for

histological tissue processing, and C. Laguerre and B. Urban

for LCMS/MS analysis of lipid mediators. We are grateful to

the X10SA beamline team at the Swiss Light Source (PSI

Villigen, Switzerland) for outstanding support at the beamline

and Expose GmbH (Switzerland) for diﬀraction data

collection.

Rainer M. Luond − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

Wolfgang Miltz − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

Rudolf Walchli − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

Romain Wolf − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

rgen Hinrichs − Global Discovery Chemistry, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

Christian Bergsdorf − Chemical Biology and Therapeutics,

Novartis Institutes for Biomedical Research, 4002 Basel,

Switzerland

Kamal Azzaoui − Chemical Biology and Therapeutics,

Novartis Institutes for Biomedical Research, 4002 Basel,

Switzerland

Carlos A. Penno − Analytical Sciences & Imaging, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland

Kai Klein − PK Sciences, Novartis Institutes for Biomedical

Research, 4002 Basel, Switzerland

ABBREVIATIONS

■

b.i.d., (lat.) bis in die, twice a day; CL, clearance; DIPEA,

diisopropylethylamine; hERG, human Ether-a-

go-go-Related

Gene; i.v., intraveneous; LM, liver microsomes; 2-MeTHF, 2-

methyl tetrahydrofuran; MPO:, Myeloperoxidase; NASH:,

Nonalcoholic steatohepatitis; PD, pharmacodynamic; PK,

pharmacokinetic; RED, rapid equilibrium dialysis; rt, room

temperature; SAR, structure−activity relationship; SD, stand-

ard deviation; SEM, standard error of the mean; TACE, tumor

necrosis factor-α converting enzyme; TFA, triﬂuoroacetic acid;

THF, tetrahydrofuran

Nathalie Wack − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

002 Basel, Switzerland

Petra Jager − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

002 Basel, Switzerland

Franziska Hasler − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

REFERENCES

002 Basel, Switzerland

■

1) Haeggstrom, J. Z. Leukotriene A4 hydrolase and the committed

step in leukotriene B4 biosynthesis. Clin. Rev. Allergy Immunol. 1999,

Christian Beerli − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

2) Griffin, K. J.; Gierse, J.; Krivi, G.; Fitzpatrick, F. A. Opioid

7, 111−131.

002 Basel, Switzerland

Pius Loetscher − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

peptides are substrates for the bifunctional enzyme LTA4 hydrolase/

aminopeptidase. Prostaglandins 1992, 44, 251 −257.

3) Nissen, J. B.; Iversen, L.; Kragballe, K. Characterization of the

aminopeptidase activity of epidermal leukotriene A4 hydrolase against

the opioid dynorphin fragment 1 −7. Br. J. Dermatol. 1995, 133, 742−

749.

002 Basel, Switzerland

Miltz, W.; Rohn, T. A. Feasibility and physiological relevance of

Janet Dawson − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

002 Basel, Switzerland; orcid.org/0000-0003-4615-

091

(4) Numao, S.; Hasler, F.; Laguerre, C.; Srinivas, H.; Wack, N.;

Jager, P.; Schmid, A.; Osmont, A.; Rothlisberger, P.; Houguenade, J.;

Bergsdorf, C.; Dawson, J.; Carte, N.; Hofmann, A.; Markert, C.;

Hardaker, L.; Billich, A.; Wolf, R. M.; Penno, C. A.; Bollbuck, B.;

designing highly potent aminopeptidase-sparing leukotriene A4

Grazyna Wieczorek − Autoimmunity, Transplantation and

Inﬂammation, Novartis Institutes for Biomedical Research,

002 Basel, Switzerland

Shin Numao − Chemical Biology and Therapeutics, Novartis

Institutes for Biomedical Research, 4002 Basel, Switzerland;

orcid.org/0000-0002-6204-7117

Amanda Littlewood-Evans − Autoimmunity, Transplantation

and Inﬂammation, Novartis Institutes for Biomedical

Research, 4002 Basel, Switzerland

hydrolase inhibitors. Sci. Rep. 2017, 7, 13591.

(5) Snelgrove, R. J.; Jackson, P. L.; Hardison, M. T.; Noerager, B. D.;

Kinloch, A.; Gaggar, A.; Shastry, S.; Rowe, S. M.; Shim, Y. M.;

Hussell, T.; Blalock, J. E. A critical role for LTA4H in limiting chronic

pulmonary neutrophilic inflammation. Science 2010, 330, 90−94.

902

J. Med. Chem. 2021, 64, 1889−1903

Journal of Medicinal Chemistry

Drug Annotation

Biosynthesis of leukotriene B4. Semin. Immunol. 2017 , 33 , 3 −15.

6) Wan, M.; Tang, X.; Stsiapanava, A.; Haeggstrom, J. Z.

7) Haeggstrom, J. Z. Leukotriene A4 hydrolase/aminopeptidase,

(26) For details, see https://www.3ds.com/products-services/

biovia/products/data-science/pipeline-pilot/ (accessed 2020-12-2).

(27) Niesen, F. H.; Berglund, H.; Vedadi, M. The use of differential

scanning fluorimetry to detect ligand interactions that promote

protein stability. Nat. Protoc. 2007 , 2 , 2212 −2221.

(28) Bergsdorf, C.; Wright, S. K. A guide to run affinity screens using

the gatekeeper of chemotactic leukotriene B4 biosynthesis. J. Biol.

Chem. 2004, 279, 50639−50642.

(

8) Peters-Golden, M.; Henderson, W. R., Jr. Leukotrienes. N. Engl.

differential scanning fluorimetry and surface plasmon resonance

assays. Methods Enzymol. 2018, 610, 135−165.

J. Med. 2007, 357, 1841−1854.

9) Di Gennaro, A.; Haeggstrom, J. Z. The leukotrienes: immune-

modulating lipid mediators of disease. Adv. Immunol. 2012, 116, 51−

10) Di Gennaro, A.; Haeggstrom, J. Z. Targeting leukotriene B4 in

inflammation. Expert Opin. Ther. Targets 2014, 18, 79−93.

11) Fourie, A. M. Modulation of inflammatory disease by inhibitors

of leukotriene A4 hydrolase. Curr. Opin. Invest. Drugs 2009, 10, 1173−

182.

12) Haeggstrom, J. Z. Leukotriene biosynthetic enzymes as

therapeutic targets. J. Clin. Invest. 2018, 128, 2680−2690.

13) Serhan, C. N.; Chiang, N.; Van Dyke, T. E. Resolving

(30) Finlayson, K.; Turnbull, L.; January, C. T.; Sharkey, J.; Kelly, J.

(29) Pristupa, Z. B.; Wilson, J. M.; Hoffman, B. J.; Kish, S. J.; Niznik,

H. B. Pharmacological heterogeneity of the cloned and native human

dopamine transporter: disassociation of [3H]WIN 35,428 and

[3H]GBR 12,935 binding. Mol. Pharmacol. 1994, 45, 125−135.

S. [3H]Dofetilide binding to herg transfected membranes: a potential

high throughput preclinical screen. Eur. J. Pharmacol. 2001, 430, 147−

148.

(31) Zimmerlin, A.; Trunzer, M.; Faller, B. CYP3A time-dependent

inhibition risk assessment validated with 400 reference drugs. Drug

Metab. Dispos. 2011, 39, 1039−1046.

inflammation: dual anti-inflammatory and pro-resolution lipid

mediators. Nat. Rev. Immunol. 2008, 8, 349−361.

(32) Zhou, L.; Yang, L.; Tilton, S.; Wang, J. Development of a high

throughput equilibrium solubility assay using miniaturized shake-flask

method in early drug discovery. J. Pharm. Sci. 2007, 96, 3052−3071.

(14) Serhan, C. N.; Yacoubian, S.; Yang, R. Anti-inflammatory and

pro-resolving lipid mediators. Annu. Rev. Pathol.: Mech. Dis. 2008, 3,

79−312.

15) Rao, N. L.; Dunford, P. J.; Xue, X.; Jiang, X.; Lundeen, K. A.;

(33) Pettersen, D.; Broddefalk, J.; Emtenas, H.; Hayes, M. A.;

(

Lemurell, M.; Swanson, M.; Ulander, J.; Whatling, C.; Amilon, C.;

Ericsson, H.; Westin Eriksson, A.; Granberg, K.; Plowright, A. T.;

Shamovsky, I.; Dellsen, A.; Sundqvist, M.; Nagard, M.; Lindstedt, E.

L. Discovery and early clinical development of an inhibitor of 5-

lipoxygenase activating protein (AZD5718) for treatment of coronary

artery disease. J. Med. Chem. 2019 , 62 , 4312 −4324.

Coles, F.; Riley, J. P.; Williams, K. N.; Grice, C. A.; Edwards, J. P.;

Karlsson, L.; Fourie, A. M. Anti-inflammatory activity of a potent,

selective leukotriene A4 hydrolase inhibitor in comparison with the 5-

lipoxygenase inhibitor zileuton. J. Pharmacol. Exp. Ther. 2007, 321,

(34) McGee, J.; Fitzpatrick, F. Enzymatic hydration of leukotriene

154−1160.

16) Caliskan, B.; Banoglu, E. Overview of recent drug discovery

A4. purification and characterization of a novel epoxide hydrolase

approaches for new generation leukotriene A4 hydrolase inhibitors.

from human erythrocytes. J. Biol. Chem. 1985, 260, 12832−7.

Expert Opin. Drug Discovery 2013, 8, 49−63.

(35) Chen, M.; Lam, B. K.; Kanaoka, Y.; Nigrovic, P. A.; Audoly, L.

(17) Bhatt, L.; Roinestad, K.; Van, T.; Springman, E. B. Recent

P.; Austen, K. F.; Lee, D. M. Neutrophil-derived leukotriene B4 is

advances in clinical development of leukotriene B4 pathway drugs.

Semin. Immunol. 2017, 33, 65 −73.

18) For details, see https://trials.boehringer-ingelheim.com/

required for inflammatory arthritis. J. Exp. Med. 2006, 203, 837−842.

(36) Davies, D. R.; Mamat, B.; Magnusson, O. T.; Christensen, J.;

Haraldsson, M. H.; Mishra, R.; Pease, B.; Hansen, E.; Singh, J.;

Zembower, D.; Kim, H.; Kiselyov, A. S.; Burgin, A. B.; Gurney, M. E.;

Stewart, L. J. Discovery of leukotriene A4 hydrolase inhibitors using

metabolomics biased fragment crystallography. J. Med. Chem. 2009,

p u b l i c / t r i a l _ r e s u l t s _ d o c u m e n t s / 1 3 3 4 / _ e n g l i s h _

3341c2248492COpdf.pdf (accessed 2020-12-2).

19) Sandanayaka, V.; Mamat, B.; Mishra, R. K.; Winger, J.; Krohn,

(

M.; Zhou, L. M.; Keyvan, M.; Enache, L.; Sullins, D.; Onua, E.;

Zhang, J.; Halldorsdottir, G.; Sigthorsdottir, H.; Thorlaksdottir, A.;

Sigthorsson, G.; Thorsteinnsdottir, M.; Davies, D. R.; Stewart, L. J.;

Zembower, D. E.; Andresson, T.; Kiselyov, A. S.; Singh, J.; Gurney,

M. E. Discovery of 4-[(2 S )-2-{[4-(4-chlorophenoxy)phenoxy]-

methyl}-1-pyrrolidinyl]butanoic acid (DG-051) as a novel leukotriene

A4 hydrolase inhibitor of leukotriene B4 biosynthesis. J. Med. Chem.

(

2, 4694−4715.

37) Tanis, V. M.; Bacani, G. M.; Blevitt, J. M.; Chrovian, C. C.;

Crawford, S.; De Leon, A.; Fourie, A. M.; Gomez, L.; Grice, C. A.;

Herman, K.; Kearney, A. M.; Landry-Bayle, A. M.; Lee-Dutra, A.;

Nelson, J.; Riley, J. P.; Santillan, A., Jr.; Wiener, J. J.; Xue, X.; Young,

A. L. Azabenzthiazole inhibitors of leukotriene A4 hydrolase. Bioorg.

Med. Chem. Lett. 2012, 22, 7504−7511.

(

010, 53, 573−585.

(38) Diamine derivatives as inhibitors of leukotriene A4 hydrolase.

20) Barchuk, W.; Lambert, J.; Fuhr, R.; Jiang, J. Z.; Bertelsen, K.;

US 2007/155726, 2007.

39) Penno, C. A.; Jager, P.; Laguerre, C.; Hasler, F.; Hofmann, A.;

Gass, S. K.; Wettstein-Ling, B.; Schaefer, D. J.; Avrameas, A.; Raulf, F.;

Wieczorek, G.; Lehmann, J. C. U.; Loesche, C.; Roth, L.; Rohn, T. A.

Fourie, A.; Liu, X.; Silkoff, P. E.; Barnathan, E. S.; Thurmond, R.

Effects of JNJ-40929837, a leukotriene A4 hydrolase inhibitor, in a

bronchial allergen challenge model of asthma. Pulm. Pharmacol. Ther.

(

014, 29, 15−23.

21) Ward, P. D.; La, D. Testicular distribution and toxicity of a

novel LTA4H inhibitor in rats. Toxicol. Appl. Pharmacol. 2014, 278,

6−30.

22) Elborn, J. S.; Ahuja, S.; Springman, E.; Mershon, J.; Grosswald,

Lipidomics profiling of hidradenitis suppurativa skin lesions reveals

lipoxygenase pathway dysregulation and accumulation of proin-

flammatory leukotriene B4. J. Invest. Dermatol. 2020, 140, 2421−

(

2432.

R.; Rowe, S. M. EMPIRE-CF: a phase II randomized placebo-

controlled trial of once-daily, oral acebilustat in adult patients with

cystic fibrosis - study design and patient demographics. Contemp. Clin.

Trials 2018, 72, 86−94.

(23) Elborn, J. S.; Bhatt, L.; Grosswald, R.; Ahuja, S.; Springman, E.

B. Phase I studies of acebilustat: pharmacokinetics, pharmacody-

namics, food effect, and CYP3A induction. Clin. Transl. Sci. 2017, 10,

(

0−27.

24) Elborn, J. S.; Horsley, A.; MacGregor, G.; Bilton, D.;

Grosswald, R.; Ahuja, S.; Springman, E. B. Phase I studies of

acebilustat: biomarker response and safety in patients with cystic

fibrosis. Clin. Transl. Sci. 2017 , 10 , 28 −34.

(25) For details, see https://www.rcsb.org/ (accessed 2020-12-2).

903