Discovery and early clinical development of 2-{6-[2-(3,5-dichloro-4- pyridyl)acetyl]-2,3-dimethoxyphenoxy}- N -propylacetamide (LEO 29102), a soft-drug inhibitor of phosphodiesterase 4 for topical treatment of atopic dermatitis

Article abstract of DOI:10.1021/jm500378a

Development of orally available phosphodiesterase 4 (PDE4) inhibitors as anti-inflammatory drugs has been going on for decades. However, only roflumilast has received FDA approval. One key challenge has been the low therapeutic window observed in the clin

Full text of DOI:10.1021/jm500378a

Drug Annotation
pubs.acs.org/jmc
Discovery and Early Clinical Development of 2‑{6-[2-(3,5-Dichloro-4-
pyridyl)acetyl]-2,3-dimethoxyphenoxy}‑N‑propylacetamide (LEO
29102), a Soft-Drug Inhibitor of Phosphodiesterase 4 for Topical
Treatment of Atopic Dermatitis
Jakob Felding, Morten D. Sørensen, Tina D. Poulsen,† Jens Larsen, Christina Andersson,‡ Pia Refer,
Karen Engell, Lotte G. Ladefoged, Thorsten Thormann, Anne Marie Vinggaard,§ Pontus Hegardt,
Anders Søhoel, and Simon Feldbæk Nielsen*
New Product Discovery and Global Development, LEO Pharma A/S, Industriparken 55, 2750 Ballerup, Denmark
ABSTRACT: Development of orally available phosphodiester-
ase 4 (PDE4) inhibitors as anti-inflammatory drugs has been
going on for decades. However, only roflumilast has received
FDA approval. One key challenge has been the low therapeutic
window observed in the clinic for PDE4 inhibitors, primarily
due to PDE4 mediated side effects. Here we describe our
approach to circumvent this issue by applying a soft-drug
concept in the design of a topically acting PDE4 inhibitor for
treatment of dermatological diseases. We used a fast follower
approach, starting from piclamilast. In particular, simultaneous
introduction of 2′-alkoxy substituents and changing an amide to
a keto linker proved to be beneficial when designing potential
soft-drug candidates. This effort culminated in identification of
LEO 29102 (20), a potent, selective, and soft-drug PDE4
inhibitor with properties suitable for patient-friendly formulations giving efficient drug delivery to the skin. Compound 20 has
reached phase 2 and demonstrated clinically relevant efficacy in the treatment of atopic dermatitis.
and cyclic guanosine monophosphate (cGMP).6 PDEs are
classified accordingly into three groups: those that are (i)
specific to cAMP and (ii) specific to cGMP and (iii) those that
INTRODUCTION
■
Atopic dermatitis (AD) is the most common childhood skin
disease and shows an increased prevalence in Western
European countries and North America.1,2 It is usually present
act on both cAMP and cGMP. Phosphodiesterase 4 (PDE4)
enzymes belong to the first group, as they only participate in
during early infancy and childhood but can also start in
adulthood and affects up to 20% of children and 3% of adults.
The disease is a relapsing chronic inflammatory skin disease
often associated with hyper-reactivity to environmental triggers
the regulation of cAMP levels. When PDE4 is inhibited, the
resultant elevation of intracellular cAMP levels leads to an
activation of specific protein phosphorylation cascades, which
and is characterized by eczematous skin plaques and intense
itch (pruritus). The pruritus has a major impact on the quality
of sleep and is associated with profound negative effects on the
quality of life for AD patients and their family unit.3
Furthermore, AD is believed to represent the initial step of
the “atopic march”, in which AD can develop into systemic
diseases like asthma and allergic rhinitis.4 Topical cortico-
steroids are still the mainstream therapy in the treatment of
AD, but topical calcineurin inhibitors have also brought
significant clinical benefit to AD patients.5 However, new
steroid-free anti-inflammatory agents that could provide a safe
and effective long-term treatment for patients with AD would
Figure 1. Selected PDE4 inhibitors in clinical trial at the outset of this
add significantly to the treatment options for this complex
disease.
program.17
Phosphodiesterases (PDEs) belong to a family of 11 different
isoenzymes that catalyze the hydrolysis of the two secondary
Received: March 11, 2014
signal messengers cyclic adenosine monophosphate (cAMP)
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Figure 2. Structures of the most advanced PDE4 inhibitors for topical treatment of AD.17
a
Scheme 1. Synthesis Route for Variations in the 2′-O Position Exemplified by the Synthesis of 20
a
Reagents and conditions: (a) methyl 2,3,4-trimethoxybenzoate, LiHMDS, THF, 0 °C (65%); (b) 1 M BCl3 (74%); (c) ethyl bromoacetate, K2CO3
(78%); (d) LiOH, H2O, MeOH (92%); (e) propylamine, HATU, DIPEA, DMF.
elicit a variety of functional responses in the inflammatory cells
such as suppression of TNFα production. Four subtypes of
PDE4 have been identified (PDE4A−PDE4D) that are
encoded by four distinct genes.
and binding interaction to PDE4. We also discuss the
preformulation characterization from a topical perspective,
preclinical emetic potential, and initial clinical data.
Of these, PDE4A, -B, and -D are primarily found in
inflammatory and immune cells, including T cells, B cells,
macrophages, monocytes, neutrophils, and eosinophils.7
Development of PDE4 inhibitors as a novel anti-inflammatory
therapy has been ongoing for several decades with asthma and
chronic obstructive pulmonary disease (COPD) as the primary
indications. However, although a large number of PDE4
inhibitors have been evaluated in the clinic, roflumilast (Figure
1) is currently the only approved PDE4 inhibitor. A primary
problem with PDE4 as a drug target is the narrow therapeutic
window of PDE4 inhibitors, owing to the on-target gastro-
intestinal adverse effects (primarily nausea and emesis), which
severely limits the dose that can be given. Indeed, for many oral
PDE4 inhibitors in the clinic it is likely that the maximum
tolerated dose is subtherapeutic.8
MEDICINAL CHEMISTRY STRATEGY
■
Several compounds were in clinical development at the outset
of the project (see Figure 1) and piclamilast caught our
attention, as it was being developed for asthma, which could be
considered a pseudotopical route of administration.9 In
addition, piclamilast contains two structural features with
proven benefits for PDE4 inhibitors: (i) the diether catechol
scaffold, which is a well described isostere of adenosine to
mimic the nucleotide part of cAMP, and (ii) the 2,6-
dichloropyridine moiety which represents a key motif for the
high selectivity toward PDE4 and is also present in roflumilast.
Importantly, the very low aqueous solubility of piclamilast,
which could be advantageous for an inhaled product, was
considered limiting for development of a dermatological
product and the physicochemical profile of piclamilast needed
to be optimized for our purpose. Taken together, a fast follower
approach starting from piclamilast was seen as an attractive
medicinal chemistry strategy to be able to rapidly progress the
project.
Chemistry. 3′5-Dichloropyridine based derivatives, with a
keto linker to connect the catechol moiety, were synthesized
starting from 4. The synthesis route for variations in the 2′-O
position is exemplified with the structure of 20 in Scheme 1.
Compound 4 was obtained by adding LiHMDS to a mixture of
3,5-dichloro-4-methylpyridine and methyl 2,3,4-trimethoxyben-
zoate in THF. When the benzylic anion of 3,5-dichloro-4-
methylpyridine forms, it reacts in situ with the methyl ester
group of methyl 2,3,4-trimethoxybenzoate to provide the
desired keto linker in 4. Derivative 3 was prepared under the
same conditions but using methyl 3,4-dimethoxybenzoate in
place of methyl 2,3,4-trimethoxybenzoate (not shown).
Although the vast majority of PDE4 inhibitors in the clinic
are aiming for oral dosing, a small number of PDE4 inhibitors
in late stage clinical testing are aiming for topical treatment of
skin diseases (see Figure 2).
On the basis of the potential of PDE4 inhibition as a novel
anti-inflammatory mechanism in the treatment of AD, we
decided to embark on developing a soft-drug PDE4 inhibitor
for topical treatment of AD. The soft drug concept is based on
a rapid metabolic inactivation by loss of PDE4 activity in the
resultant metabolites. Combining a soft-drug approach with
topical application was chosen to limit the systemic exposure of
the topically acting drug and thereby increase the therapeutic
window. In this article, we describe the discovery and early
development of the potent and selective soft-drug PDE4
inhibitor 20,16 which is being evaluated in phase 2 clinical trials
for topical treatment of atopic dermatitis. We review the
medicinal chemistry strategy, structure−activity relationships,
B
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
A selective 2′-demethylation of 4 was achieved by treatment
with BCl3 to provide 4a. The free 2′-OH phenol group of 4a
was alkylated with a variety of alkyl halides under basic
conditions (K2CO3) to provide derivatives 8−11 and 22. For
compound 20 methyl bromoacetate was used as the alkylating
reagent and the formed methyl ester was subsequently
hydrolyzed using LiOH in a mixture of MeOH−water to
provide derivative 14. The free carboxylic acid of 14 was
converted to a range of amide derivatives using a HATU-
mediated amide coupling (derivatives 15−20). When propyl-
amine was used in the final amide coupling, 20 was obtained.
An alternative synthesis route was used for variations in the
3′-O position as exemplified for the synthesis of derivative 7 in
Scheme 2. When 4 was treated with HI in glacial acetic acid at
80 °C, the 2′,3′-bis-demethylated intermediate 4b was
obtained.
Table 1. Impact of Linker and 2′-Substituents on Enzymatic
Potency (PDE4D) and Inhibition of TNFα Release from
Human Blood Mononuclear Cells Stimulated with
Lipopolysaccharide
b
c
PDE4D
a
TNFα IC50
QM energy
compd
X
Y
IC50 (nM)
(nM)
(kcal/mol)
1
2
3
NH
NH
CH2
H
17
19
0.3
OCH3
H
3790
13
>5000
11
6.0
0.5 (enol)
0.0 (keto)
0.0 (enol)
2.8 (keto)
Scheme 2. Synthesis Route for Variations in the 3′-O
a
Position Exemplified by the Synthesis of 7
4
CH2
OCH3
110
436
a
b
Mean standard deviation is 37%. Mean standard deviation is
c
60%. Relative energy of the conformation similar to the bioactive
conformation of piclamilast estimated by QM calculations.
to PDE410 (Figure 3A) where the amide carbonyl is orientated
toward the 2′-position, we speculated that the lower potency of
2 compared to 1 is due to an unfavorable intramolecular
interaction in the bioactive conformation between the 2′-O and
the carbonyl of the amide linker.
a
Reagents and conditions: (a) HI on AcOH, 80 °C (81%); (b) 1 equiv
of MeI, K2CO3, DMF (27%); (c) cyclopentyl bromide, K2CO3.
We were however gratified to see that the corresponding
two-atom keto linker had a very different profile. The 2′-
unsubstituted amide and keto analogues were equipotent
against PDE4D (1 vs 3), and 2′-OMe substitution only
resulted in a 10-fold loss of potency when having the keto
linker (3 vs 4) compared to a 200-fold loss for the amide linker
(1 vs 2). A possible explanation could be that the two-atom
keto linker, in the tautomeric enol form, has the possibility of
forming an intramolecular H-bond to the 2′-O in the bioactive
conformation. To further investigate this hypothesis, we
performed a force field based coordinate scan and QM energy
calculations. Both analyses supported a lower energy penalty for
the bioactive conformation of an enol linker (and in fact also a
keto linker) compared to the amide linker (Figure 3B, Table 1).
SAR of Substituents in the 2′- and 3′-Position.
Encouraged by the promising PDE4D data as well as the
inhibition of TNFα production in human inflammatory cells by
derivative 4, we sought to optimize potency by positioning the
3′-alkoxy group in the well-known lipophilic pocket of the
enzyme to gain potency. However, this strategy was not very
successful, as increasing the size of the 3′-alkoxy group beyond
3′-ethoxy provided less potent compounds (see Table 2).
Apparently, the positioning of the 3′-alkoxy group is negatively
influenced by the 2′-substituent and increasing the size to
cyclopentyl (7, similar to piclamilast) totally eliminated activity.
At this point we obtained a crystal structure of an
ethylisoxazole analogue of piclamilast (Figure 4A) which
showed that the extended aromatic 3′-substituent of this
derivative was involved in favorable hydrophobic interactions
not observed for piclamilast. On the basis of this observation,
we hypothesized that extended hydrophobic 2′-alkoxy sub-
stituents would have the potential to reach the ethylisoxazole
pocket (Figure 4B). In addition to the hydrophobic interactions
Upon treatment of 4b with 1 equiv of methyl iodide under
basic conditions (K2CO3) the 2′,4′-dimethylated intermediate
(4c) could be isolated, although in a modest yield (27%).
Alkylation of the free 3′-OH phenol group of 4c with different
alkyl halides under basic conditions (K2CO3) provided
derivatives 5−7. When cyclopentyl bromide was used in the
final alkylation step, derivative 7 was obtained.
MEDICINAL CHEMISTRY STRATEGY:
IDENTIFICATION OF SCAFFOLD
■
We initially focused on identifying a non-amide linker to
connect the diether catechol and 2,6-dichloropyridine moieties,
as we believed the secondary amide in piclamilast was
negatively influencing the solubility of the compound because
of the possible formation of a hydrogen bond network.
In the initial screening cascade we used inhibition of PDE4
(enzymatic assay) and inhibition of TNFα production in
human inflammatory cells (cellular assay) as the two main
drivers for our compound design. In addition, the most critical
part of our optimization was to identify a suitable region in the
molecule to introduce a soft spot, allowing a fast hepatic
metabolism to metabolites with reduced activity. The 3′ and 4′
positions in the piclamilast series have been extensively
explored,9 whereas the 2′-position has, to the best of our
knowledge, not been investigated. Our work started by
exploring the effect of introducing a methoxy group in the
2′-position of 1. Compound 1 was chosen as a model system
for piclamilast and displayed reasonable potency toward
PDE4D (see Table 1). However, the additional 2′-OMe
substituent in 2 proved detrimental to the potency (see Table
1). On the basis of the conformation of piclamilast when bound
C
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Figure 3. (A) Conformation of piclamilast taken from the crystal structure with PDE4 (1XM4). The dihedral C2−C1−C−N was measured after a
MM energy minimization. (B) Plot of the relative (with respect to the global minimum energy) MM conformational energy obtained from a scan of
the corresponding dihedral in compounds 1, 2, and 4 (enol form). The piclamilast “bioactive” dihedral is indicated by the dotted line. (The figures
were made using Maestro; Schrodinger, LLC, New York, NY, 2013.)
̈
Table 2. Impact of Varying 3′-Substituents on Enzymatic
Potency (PDE4D) and Inhibition of TNFα Release from
Human Blood Mononuclear Cells Stimulated with
Lipopolysaccharide
these metabolites often proved highly active (e.g., the primary
metabolite of 9 is the hydroxylation product 11).
Reexamining the docking model suggested that charged
groups positioned close to the core scaffold might be
detrimental to the PDE4 activity because of the hydrophobic
environment formed by Phe414, Met431, Phe446, and Ile450.
As expected, compound 14 containing a short carboxylic acid
substituent was inactive. Consequently, we prepared a range of
amide analogues that potentially could be cleaved in the liver to
14. Analogues containing a secondary amide moiety were all
highly potent (15, 16, 18−20), whereas further substituting the
amide moiety led to a drop in potency (16 vs 17).
To obtain further insight into the SAR of the amide series,
we obtained a crystal structure of 15 complexed with PDE4D
(Figure 5). Surprisingly, the crystal structure revealed that the
binding mode of 15 is quite different from the docking model
proposed for 9. The 2′-substituent folds back in a conformation
stabilized by an intramolecular H-bond between the 2′-amide
NH and the linker oxygen, strongly suggesting that 15 binds in
the keto form. Furthermore, there is a stabilizing hydrophobic
collapse of the terminal aryl groups. Even more surprising was
the observation (data not shown) of visible electron density for
a C-terminal helix which made several interactions with 15,
most notably a H-bond with the side chain of Gln507 while the
terminal benzyl forms π-stacking interactions with Phe506 and
hydrophobic interactions with Leu510 (electron density for
these residues could not be observed in the crystal structure
with 9). The crystal structure also readily explains the decrease
in potency observed for 17 where an intramolecular H-bond
cannot be formed.
As seen from Table 3, several compounds had satisfactory
cellular potency and were rapidly metabolized by human liver
microsomes. Next step was to identify the major metabolites
and validate the metabolic inactivation (by loss of PDE4
activity), which is a prerequisite for a true soft-drug profile of
the parent compound. The metabolism of 15 is shown in
Figure 6, the primary metabolite being the inactive compound
14 together with minor amounts of two active metabolites (21
and 22). A similar pattern was observed for other amides
including 20.
a
b
Mean standard deviation is 7%. Mean standard deviation is 55%.
observed for the ethylisoxazole group, it was also noteworthy
that the terminal ethyl part extended toward the solvent
accessible region (not shown).
We were pleased to find that 2′-O-benzyl substitution is
allowed and in particular that 2′-O-phenethyl and 2′-O-
phenylpropyl substitution brings sufficient potency to the
compounds (8−10) albeit at the expense of increased
lipophilicity (AlogP for 4 and 9 is 3.4 and 5.3, respectively) .
The binding model of 9 suggested that the para position of the
phenethyl would be solvent accessible and potentially could
accommodate polar substituents. Accordingly, we introduced
several substituents like hydroxy, amide, and even carboxylic
acid (11−13) and found that hydrophilic para substitution is
widely allowed in the 2′-O-phenethyl series. This was, however,
quite unfortunate from a soft-drug perspective as the
metabolism primarily occurs on the new phenyl ring, and
D
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Figure 4. (A) Overlay of piclamilast (1XM4) and an ethylisoxazole analog in the PDE4 binding site. Residues involved in hydrophobic interactions
with the ethylisoxazole substituent are shown. The crystal structure of the ethylisoxazole analogue complexed with PDE4D was solved to 2.2 Å. (B)
Docking model of the enol form of 9 overlaid with the ethylisoxazole analogue. (The figures were made using Maestro; Schrodinger, LLC, New
̈
York, NY, 2013).
and the main metabolite 14 showed less than 10% inhibition of
PDE1−PDE11 at 1 μM, PDE6 not tested).
Likewise a “lead profile screen” against 68 primary molecular
targets was performed, showing only PDE4 inhibition.11
Genotoxicity was evaluated using the GreenScreen HC
genotoxicity assay.12 Unfortunately, 18 showed indications of
in vitro genotoxicity at very high concentrations in the
GreenScreen assay, and 18 was therefore deselected.
Furthermore, compound 19 had, through metabolic demethy-
lation, a potential to form 18 and was accordingly deselected as
well (even though no genotoxicity was observed with 19). On
the basis of these findings, only compounds 15 and 20 were
profiled further.
Final selection criteria were physicochemical properties
suitable for the properties of topical formulations and a
significantly improved emetic profile in vivo compared to other
PDE4 inhibitors.
PREFORMULATION CHARACTERIZATION
■
The possibility for making an optimal dermal formulation is
closely related to the solubility of the compound in solvents
relevant for topical delivery and to the physical and chemical
attributes of the compound related to the ability to penetrate
the skin on its own. Since the release of the compound from
the formulations is related to the diffusion of the compound
from the vehicle to the skin, the maximum delivery from the
vehicle is obtained with a compound dissolved in the vehicle
close to saturation or supersaturated, resulting in a maximal
thermodynamic drive. If it is not possible to dissolve the
compound in any of the vehicle components, the compound
needs to be suspended in the vehicle and dissolution rate and
particle size will become determining steps for the drug release.
Assuming almost equal potency of 15 and 20, we aimed for a
concentration of compound in the final cream formulation of
0.25% (correlating to ∼2000 μg/mL). To assess the potential
for developing topical formulations, solubility in various
relevant vehicles were determined. As seen in Table 4, the
probability of dissolving the compound in the lipid part of the
vehicle is highest for compound 20 which can be dissolved in
different lipophilic solvents (e.g., MCT). Thus, formulating a
Figure 5. Crystal structure of compound 15 complexed with PDE4D
solved to 2.5 Å. The additional C-terminal helix (residues 503−510) is
shown in yellow highlighting the interactions with Phe506, Gln507,
and Leu510. Residue numbering as in 1XM4. (The figure was made
using Maestro; Schrodinger, LLC, New York, NY, 2013).
̈
Four amides (15, 18, 19, and 20) were selected for further
profiling based on the above data and promising in vivo efficacy
data. The compounds were tested in a topical allergic contact
dermatitis model in mouse, consistently showing high level of
efficacy after dermal dosing. Compound 20 demonstrated a
dose dependent inhibition of TNFα in the tissue: 33%, 91%,
and 100% inhibition at 0.001, 0.01, and 0.1 mg, respectively. In
comparison betamethasone 17-valerate (0.003 mg) showed a
75% inhibition of TNFα.
PROFILING OF PRECANDIDATES
■
The four compounds (15, 18, 19, and 20) were extensively
profiled. The compounds and their main metabolites were
profiled against 10 other human phosphodiesterases, showing
them to be highly selective toward PDE4 (e.g., compound 20
E
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Table 3. SAR of Substitution in the 2′-Position: Impact on Enzymatic Potency (PDE4D), Inhibition of TNFα Release from
Human Blood Mononuclear Cells Stimulated with Lipopolysaccharide and Human Metabolic Degradation in Liver Microsomes
(Expressed as Extraction Ratio, Eh)
a
b
Mean standard deviation is 36%. Mean standard deviation is 52%.
range of different prototypes of creams should be possible for
this compound. In addition a formulation with 20 dissolved in
the water phase is possible if propylene glycol is used as
cosolvent. Compound 15 is less soluble, and a suspension
F
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Figure 6. Hepatic metabolism of 15 in rat.
Table 4. Preformulation Data: Solubility in Vehicles Relevant for Topical Formulation
af
,
be
,
cf
,
df
,
f
f
f
aqueous, pH 7.4,
20% HPCD,
MCT,
PG,
PG−H2O, 25:75,
PG−H2O, 50:50,
PG−H2O, 75:25,
Compd
μg/mL
μg/mL
μg/mL
μg/mL
μg/mL
μg/mL
μg/mL
piclamilast
1
190
57
2000
580
3300
990
15
20
2
20
230
14
220
6700
24000
135
1300
5600
a
b
c
d
e
Phosphate buffer, 0.067 M, pH 7.4. Hydroxypropyl-β-cyclodextrin in water. mMedium chain triglyceride. Propylene glycol. The solubility was
determined by shaking the compound in the vehicle for 1 h at room temperature. After centrifugation the concentration of compound in the
f
supernatant was determined by HPLC. The solubility was determined by shaking the compound in the vehicle for 24 h at room temperature. After
centrifugation the concentration of compound in the supernatant was determined by HPLC.
cream would most likely be the outcome. Consequently
compound 20 was selected for the final assessment of emetic
risk.
EARLY CLINICAL DEVELOPMENT OF 20
■
The first study in man was a safety, tolerability, and
pharmacokinetic study of single ascending and multiple
cutaneous doses of 20 (2.5 mg/g cream).14 A topical product
for atopic dermatitis is likely to be used on large skin areas, and
the clinical development must reflect that. Dose escalation was
therefore performed by increasing the treated body surface area
(BSA) up to 53% of BSA. Promising safety results were
obtained, and no cases of emesis were observed in the study.
The systemic PK profile supported the soft-drug concept
obtaining very low systemic levels of active compound even if
treated at a large body surface area (see Table 5).
EMESIS EVALUATION
■
The emetic potential of compound 20 was evaluated after iv
dosing to ferrets. Although the PK profile of compound 20 may
differ between ferrets and man, compound 20 showed
comparable inhibition of human and ferret PDE4 enzyme.
The animals were observed for a period of 2 h, and any emetic
episode and/or associated prodromic signs (indications of
nausea) were recorded. Compound 20 was very well tolerated
up to the maximum feasible dose (1 mg/kg) with no emesis
and no signs of nausea observed (Figure 7). As a comparison,
cipamfylline (selective PDE4 inhibitor, evaluated topically in
phase 2)13 resulted in clear signals of emesis at 100 times lower
doses (0.01 mg/kg).
Table 5. Geometric Mean Cmax and AUC0−48h after Singles
Doses of 20−2.5 mg/g Cream to Healthy Subjects (n = 6 in
each group) at Different BSA with Constant Amount of
a
Cream per cm2
BSA dosed (%)
Compound 20 was selected for further development based
on the superior formulation properties and lack of emesis.
7
14
30
53
Cmax (ng/mL)
0.279
5.45
0.754
14.9
2.39
49.2
5.07
100
AUC0−48h (ng·h/mL)
a
Cmax and AUC values increased more than proportional to the dose
(BSA), and an 8-fold increase in dose resulted in Cmax and AUC values.
Subsequently 20 entered into a 4-week (phase 2) proof-of-
concept and dose finding study, investigating treatment efficacy
after dermal application of 20 cream, 20 cream vehicle, and
Elidel cream.15 The results were encouraging, and 20 entered
into substantial clinical testing.
CONCLUSION
■
We have described the medicinal chemistry that resulted in the
discovery of 20 (LEO 29102), which is a potent and highly
selective PDE4 inhibitor with promising pharmaceutical
properties that are suitable for effective topical delivery. The
compound can be synthesized in a few simple synthetic steps,
which is major advantage from a cost-of-goods perspective,
Figure 7. Cumulative emetic profile (number of episodes) of vehicle
(S1−S4), compound 20 (0.03−1.0 mg/kg), and cipamfylline (Ci)
(0.01−0.1 mg/kg) after iv dosing to ferrrets.
G
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
vacuo, and purified by flash chromatography using a gradient of EtOAc
in heptane as eluent. The title compound 3 was obtained as a white
since treatment of large body surfaces requires large amount of
formulation to be applied (i.e., relatively large amount of
compound) in a disease where treatment options are fairly low
priced.
The applied soft-drug concept eliminated the PDE4 driven
systemic side effects in both preclinical models and in the
clinical setting. Furthermore, the clinical efficacy of 20 was
clearly demonstrated in atopic dermatitis patients and further
clinical studies were initiated.
1
solid. UPLC−MS (ESI+) m/z 326.03 (MH+). H NMR (300 MHz,
DMSO-d6) δ 8.66 (s, 2H), 7.85 (dd, J = 8.5, 2.0 Hz, 1H), 7.55 (d, J =
2.0 Hz, 1H), 7.14 (d, J = 8.5 Hz, 1H), 4.78 (s, 2H), 3.88 (s, 3H), 3.85
(s, 3H). 13C NMR (75 MHz, DMSO) δ 191.90, 153.68, 148.67,
147.08, 141.59, 132.81, 128.41, 123.08, 111.00, 110.31, 55.79, 55.52,
40.50. Yield 120 mg (73%).
2-(3,5-Dichloropyridin-4-yl)-1-(2,3,4-trimethoxyphenyl)-
ethanone (4). Methyl 2,3,4-trimethoxybenzoate (2.7 g, 10 mmol)
and 3,5-dichloro-4-methylpyridine dissolved in dry THF (40 mL). The
mixture was cooled on ice and treated dropwise (during 5 min) with 1
M lithium bis(trimethylsilyl)amide (12 mL, 12 mmol). After 1.5 h at 0
°C the reaction was quenched with saturated NH4Cl (100 mL). The
organic products were extracted with EtOAc (2 × 100 mL) and the
combined organic phases washed with saturated NaCl (100 mL). The
organic phase was dried over Na2SO4, evaporated in vacuo, and
purified by flash chromatography using a gradient of EtOAc in heptane
as eluent. The title compound 4 was obtained as a white solid. UPLC−
EXPERIMENTAL SECTION
■
1H nuclear magnetic resonance (NMR) spectra were usually recorded
at 300 MHz. Chemical shift values (δ, in ppm) are quoted in the
specified solvent relative to internal tetramethylsilane (δ = 0.00) or
chloroform (δ = 7.25) or deuteriochloroform (δ = 76.81 for 13C
NMR) standards. The value of a multiplet, either defined (doublet (d),
triplet (t), quartet (q)) or not (m) at the approximate midpoint is
given unless a range is quoted. bs indicates a broad singlet. The organic
solvents used were usually anhydrous. Chromatography was
performed on Merck silica gel 60 (0.040−0.063 mm). Preparative
HPLC/MS was performed on a Dionex APS-system with two
Shimadzu PP150 preparative pumps and a Thermo MSQ Plus mass
spectrometer with the following parameters: column, Waters XTerra
C-18, 150 mm × 19 mm, 5 μm; solvent system, A = water (0.1%
formic acid) and B = acetonitrile (0.1% formic acid); flow rate = 18
mL/min; method (10 min), linear gradient method going from 10% B
to 100% B in 6 min and staying at 100% B for another 2 min. The
fractions were collected based on ion traces of relevant ions and PDA
signal (240−400 nm). All compounds were analyzed by UPLC−MS.
High resolution chromatography was performed using the Waters
Acquity UPLC and a 2.1 mm × 50 mm C18 RP column (1.8 μm
particles). The mobile phases consisted of 0.1% formic acid in an
aqueous 10 mM ammonium acetate solution for buffer A and 0.1%
formic acid in acetonitrile for buffer B. A standard binary gradient was
used for the reverse phase chromatography. All compounds had an
estimated UV purity of >95% by UPLC−MS.
1
MS (ESI+) m/z 356.05 (MH+). H NMR (600 MHz, DMSO-d6) δ
8.66 (s, 2H), 7.52 (d, J = 8.9 Hz, 1H), 6.98 (d, J = 8.9 Hz, 1H), 4.66
(s, 2H), 4.00 (s, 3H), 3.89 (s, 3H), 3.81 (s, 3H). 13C NMR (151 MHz,
DMSO) δ 192.37, 157.83, 153.70, 147.03, 141.59, 141.49, 132.76,
125.29, 123.54, 107.99, 61.45, 60.40, 56.13, 44.38.
2-(3,5-Dichloro-pyridin-4-yl)-1-(2-hydroxy-3,4-dimethoxy-
phenyl)-ethanone (4a). To a solution of 4 (20 g, 56 mmol) in DCM
(75 mL) was slowly added 1 M BCl3 (95 mL, 95 mmol) in DCM. The
reaction mixture was stirred under nitrogen at room temperature for 2
h. Water (50 mL) was added slowly followed by EtOH (200 mL). The
white precipitate was filtered and recrystallized from EtOH. The title
1
compound 4a was obtained as a white solid. H NMR (300 MHz,
CDCl3) δ 11.92 (1H, s), 8.54 (2H, s), 7.68 (1H, d), 6.60 (1H, d), 4.67
(2H, s), 3.98 (3H, s), 3.90 (3H, s). Yield 14.2 g (74%)
2-(3,5-Dichloropyridin-4-yl)-1-(2,3-dihydroxy-4-
methoxyphenyl)ethanone (4b). To a solution of 4 (1.25g, 3.5
mmol) in AcOH (100%, 9.25 mL) was added HI (55−58%, 4.55 mL),
and the reaction mixture was heated to 80 °C for 18 h. The reaction
mixture was cooled to room temperature and evaporated in vacuo.
NaHCO3 (saturated, 150 mL) was added slowly, followed by NaCl
(saturated, 50 mL). The organic products were extracted with EtOAc
(4 × 100 mL). The combined organic phases were washed with
saturated NaCl (2 × 100 mL) and 10% Na2S2O3 (100 mL) and dried
over Na2SO4 and evaporated in vacuo. The title compound 4b was
obtained as a brown solid. 1H NMR (300 MHz, CDCl3) δ 11.88 (1H,
s), 8.54 (2H, s), 7.51 (1H, d), 6.60 (1H, d), 5.62 (1H, bs), 4.67 (2H,
s), 4.00 (3H, s). Yield 0.935g (81%)
2-(3,5-Dichloropyridin-4-yl)-1-(2,4-dimethoxy-3-
hydroxyphenyl)ethanone (4c). To a solution of 4b (0.77 mmol) in
dry DMF (5.5 mL) was added K2CO3 (0.77 mmol) followed by
methyl iodide (0.77 mmol). The reaction mixture was stirred at room
temperature for 18 h. NH4Cl (saturated, 15 mL) was added, and the
organic products were extracted with EtOAc (3 × 15 mL). The
combined organic phases were washed with saturated NaCl (2 × 15
mL). The organic phase was dried over Na2SO4, evaporated in vacuo,
and purified by flash chromatography using a gradient of EtOAc in
toluene as eluent. The title compound 4c was obtained as a yellow
solid. 1H NMR (600 MHz, CDCl3) δ 8.50 (s, 2H), 7.43 (d, J = 8.8 Hz,
1H), 6.75 (d, J = 8.8 Hz, 1H), 5.73 (s, 1H), 4.70 (s, 2H), 4.08 (s, 3H),
3.98 (s, 3H). Yield 72 mg (27%).
General Protocol for Preparation of 1 and 2. To a solution of
an alkoxybenzoic acid (3 mmol) in toluene (10 mL) was added SOCl2
(1 mL), and the reaction mixture was heated to 100 °C for 6 h. The
reaction mixture was cooled to room temperature and evaporated. The
alkoxybenzoyl chloride was redissolved in toluene (10 mL),
evaporated, and then dissolved in THF (10 mL). 4-Amino-3,5-
dichloropyridine (6 mmol) in THF (10 mL) was treated with NaH (6
mmol) for 3 h at room temperature followed by dropwise addition of
the alkoxybenzoyl chloride solution (prepared above). After 4 h at
room temperature the reaction was quenched with saturated NaCl
(100 mL). The organic products were extracted with EtOAc (2 × 50
mL). The combined organic phases were dried over Na2SO4,
evaporated in vacuo, and purified by flash chromatography using a
gradient of EtOAc in heptane as eluent.
N-(3,5-Dichloro-4-pyridyl)-3,4-dimethoxybenzamide (1).
UPLC−MS (ESI+) m/z 327.03 (MH+). 1H NMR (300 MHz,
DMSO-d6) δ 10.43 (s, 1H), 8.74 (s, 2H), 7.68 (dd, J = 8.4, 2.1 Hz,
1H), 7.57 (d, J = 2.1 Hz, 1H), 7.12 (d, J = 8.4 Hz, 1H), 3.85 (s, 3H),
3.84 (s, 3H). Alkoxybenzoic acid: 3,4-dimethoxybenzoic acid.
N-(3,5-Dichloro-4-pyridyl)-2,3,4-trimethoxybenzamide (2).
UPLC−MS (ESI+) m/z 357.04 (MH+). 1H NMR (300 MHz,
DMSO-d6) δ 10.14 (s, 1H), 8.72 (s, 2H), 7.59 (d, J = 8.9 Hz, 1H),
6.99 (d, J = 8.9 Hz, 1H), 3.98 (s, 3H), 3.88 (s, 3H), 3.81 (s, 3H).
Alkoxybenzoic acid: 2,3,4-trimethoxybenzoic acid.
General Protocol for Preparation of Compound 5, 6, and 7.
To a solution of 4c (0.06 mmol) in dry diethyl ketone (0.6 mL) was
added K2CO3 (0.06 mmol) followed by catalytic amounts of KI (2 mg,
0.012 mmol) and an alkyl halide (0.06 mmol). The reaction mixture
was stirred at 80 °C for 24 h. The reaction mixture was cooled to room
temperature and evaporated in vacuo. The reaction mixture was
redissolved in EtOAc (5 mL) and washed with saturated NaCl (2 × 5
mL). The organic phase was dried over Na2SO4 and evaporated in
vacuo. The pure compounds were obtained by standard preparative
HPLC purification.
2-(3,5-Dichloropyridin-4-yl)-1-(3,4-dimethoxyphenyl)-
ethanone (3). Methyl 3,4-dimethoxybenzoate (98 mg, 0.5 mmol) and
3,5-dichloro-4-methylpyridine (105 mg, 0.65 mmol) was dissolved in
dry THF (2 mL). The mixture was cooled on ice and treated dropwise
(during 5 min) with 1 M lithium bis(trimethylsilyl)amide (1.5 mL, 1.5
mmol). After 1.5 h at 0 °C the reaction was quenched with saturated
NH4Cl (5 mL). The organic products were extracted with EtOAc (2 ×
10 mL) and the combined organic phases washed with saturated NaCl
(10 mL). The organic phase was dried over Na2SO4, evaporated in
H
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
1-(3-Ethoxy-2,4-dimethoxy-phenyl)-2-(3,5-dichloro-pyridin-
4-yl)-ethanone Compound (5). UPLC−MS (ESI+) m/z 370.06
(MH+). 1H NMR (300 MHz, CDCl3) δ 8.50 (s, 2H), 7.61 (d, J = 8.9
Hz, 1H), 6.76 (d, J = 8.9 Hz, 1H), 4.68 (s, 2H), 4.11 (q, J = 7.0 Hz,
2H), 1.43 (t, J = 7.0 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 192.68,
158.45, 154.75, 147.19, 141.71, 141.05, 133.55, 125.99, 124.39, 107.35,
69.32, 61.49, 56.18, 44.93, 15.66. Alkyl halide: ethyl iodide.
1-(3-Cyclopropylmethoxy-2,4-dimethoxyphenyl)-2-(3,5-di-
chloropyridin-4-yl)ethanone (6). UPLC−MS (ESI+) m/z 396.08
(MH+). 1H NMR (300 MHz, CDCl3) δ 8.50 (s, 2H), 7.61 (d, J = 8.9
Hz, 1H), 6.75 (d, J = 8.9 Hz, 1H), 4.68 (s, 2H), 4.12 (s, 3H), 3.92 (s,
3H), 3.85 (d, J = 7.2 Hz, 2H), 1.42−1.18 (m, 1H), 0.72−0.53 (m,
2H), 0.42−0.24 (m, 2H). 13C NMR (75 MHz, CDCl3) δ 192.68,
158.45, 154.80, 147.20, 141.69, 141.11, 133.54, 126.02, 124.32, 107.32,
78.49, 61.58, 56.17, 44.92, 11.01, 3.18. Alkyl halide: 1-bromomethyl-
cyclopropane.
1-(3-Cyclopentoxy-2,4-dimethoxyphenyl)-2-(3,5-dichloro-
pyridin-4-yl)ethanone (7). UPLC−MS (ESI+) m/z 410.10 (MH+).
1H NMR (300 MHz, CDCl3) δ 8.50 (s, 2H), 7.60 (d, J = 8.9 Hz, 1H),
6.76 (d, J = 8.9 Hz, 1H), 4.97−4.76 (m, 1H), 4.69 (s, 2H), 4.05 (s,
3H), 3.91 (s, 3H), 2.06−1.82 (m, 4H), 1.83−1.55 (m, 4H). 13C NMR
(75 MHz, CDCl3) δ 192.83, 158.68, 155.10, 147.19, 141.73, 140.17,
133.56, 125.78, 124.53, 107.42, 85.25, 61.18, 56.12, 44.94, 32.86,
23.63. Alkyl halide: cyclopentyl bromide.
2 - { 6 - [ 2 - ( 3 , 5 - D i c h l o r o p y r i d i n - 4 - y l ) a c e t y l ] - 2 , 3 -
dimethoxyphenoxy}acetamide (22). UPLC−MS (ESI+) m/z
399.06 (MH+). 1H NMR (300 MHz, DMSO-d6) δ 8.65 (s, 2H),
7.66 (s, 1H), 7.61 (d, J = 8.9 Hz, 1H), 7.39 (s, 1H), 6.99 (d, J = 8.9
Hz, 1H), 4.80 (s, 2H), 4.63 (s, 2H), 3.92 (s, 3H), 3.81 (s, 3H). Alkyl
halide: 2-chloroacetamide.
4-[2-[6-[2-(3, 5-Dichloro-4-pyridyl)acetyl]-2, 3-
dimethoxyphenoxy]ethyl]benzoic Acid (13). To a solution of 4a
(0.87 g, 2.5 mmol) in dry THF (25 mL) was added methyl 4-(2-
hydroxyethyl)benzoate (0.37 g, 2 mmol) followed by solid phase
bound triphenylphosphine (2 g with a loading of 1.48 mmol/g, 3
mmol). The reaction mixture was cooled to 0 °C, and DEAD (0.93
mL, 3 mmol) was added. The reaction mixture was gently vortexed at
room temperature for 24 h. The reaction mixture was filtered and the
solvent evaporated in vacuo. The organic products were redissolved in
EtOAc (50 mL) and washed with 2 N NaOH (4 × 50 mL) and
saturated NaCl (50 mL). The organic phase was dried over Na2SO4
and evaporated in vacuo to provide methyl 4-[2-[6-[2-(3,5-dichloro-4-
pyridyl)acetyl]-2,3-dimethoxyphenoxy]ethyl]benzoate as a white solid.
Yield 535 mg (52%). The compound (1 mmol) was dissolved in THF
(10 mL). MeOH (5 mL) was added followed by LiOH (0.22 g, 5
mmol) dissolved in water (5 mL). The reaction mixture was heated to
50 °C for 4 h and then cooled to room temperature. The reaction
mixture was acidified to pH 1 by addition of 2 N HCl (7 mL). The
white precipitate was filtered and washed with water (3 × 10 mL). The
precipitate was dried in vacuo to provide the title compound 13 as a
General Protocol for Preparation of Compound 8, 9, 10, 11
and 22. To a solution of 4a (0.035 mmol) in dry DMSO (0.25 mL)
was added 2 M K2CO3 (25 μL, 0.050 mmol) followed by an alkyl
halide (0.053 mmol) dissolved in DMSO (25 μL). The reaction
mixture was left at room temperature for 48 h and then filtered. The
pure compounds were obtained by standard preparative HPLC
purification.
1
white solid. UPLC−MS (ESI+) m/z 490.08 (MH+). H NMR (300
MHz, DMSO-d6) δ 12.69 (s, 1H), 8.58 (s, 2H), 7.85−7.74 (m, 2H),
7.52 (d, J = 8.9 Hz, 1H), 7.50−7.41 (m, 2H), 6.96 (d, J = 8.9 Hz, 1H),
4.47 (t, J = 6.2 Hz, 2H), 4.23 (s, 2H), 3.88 (s, 3H), 3.71 (s, 3H), 3.21
(t, J = 6.2 Hz, 2H). Yield 254 mg (49%).
1-(2-Benzyloxy-3,4-dimethoxyphenyl)-2-(3,5-dichloropyri-
{ 6 - [ 2 - ( 3 , 5 - D i c h l o r o p y r i d i n - 4 - y l ) a c e t y l ] - 2 , 3 -
dimethoxyphenoxy}acetic Acid (14). To a solution of 4a (1.37 g,
4.1 mmol) in dry NMP (20 mL) was added ethyl bromoacetate (660
μL, 6.1 mmol) followed by K2CO3 (830 mg, 6.1 mmol). The reaction
mixture was stirred at room temperature overnight. Water (100 mL)
was added, and the organic products were extracted with EtOAc (2 ×
100 mL). The combined organic phases were washed with water (100
mL), brine (100 mL) and dried over MgSO4. The solvent were
removed in vacuo and the pure product was obtained be standard silica
gel chromatography to provide {6-[2-(3,5-dichloropyridin-4-yl)-
acetyl]-2,3-dimethoxyphenoxy}acetic acid ethyl ester as a white solid.
Yield 990 mg (78%). The compound was redissolved in MeOH−water
(1:1, 50 mL) by heating. LiOH (490 mg, 12 mmol, 5 equiv) was
added, and the reaction mixture was stirred at room temperature for 1
h. The reaction mixture was acidified to pH 1 by addition of 1 N HCl
(15 mL) and extracted with EtOAc (2 × 100 mL). The combined
organic phases were washed with brine (100 mL) and dried over
MgSO4. The solvent was removed in vacuo to provide the title
compound 14 as a white solid. UPLC−MS (ESI+) m/z 489.10
1
din-4-yl)ethanone (8). UPLC−MS (ESI+) m/z 432.07 (MH+). H
NMR (600 MHz, CDCl3) δ 8.45 (s, 2H), 7.62 (d, J = 8.9 Hz, 1H),
7.51−7.47 (m, 2H), 7.41−7.37 (m, 2H), 7.37−7.33 (m, 1H), 6.79 (d, J
= 8.9 Hz, 1H), 5.28 (s, 2H), 4.57 (s, 2H), 3.96 (s, 3H), 3.94 (s, 3H).
13C NMR (151 MHz, CDCl3) δ 192.88, 158.12, 153.17, 147.12,
142.05, 141.62, 136.64, 133.46, 128.74, 128.61, 128.53, 126.10, 125.18,
107.48, 76.62, 61.13, 56.23, 45.28. Alkyl halide: benzyl chloride.
2-(3,5-Dichloropyridin-4-yl)-1-(3,4-dimethoxy-2-
phenethyloxyphenyl)ethanone (9). UPLC−MS (ESI+) m/z
446.09 (MH+). 1H NMR (600 MHz, DMSO-d6) δ 8.63 (s, 2H),
7.52 (d, J = 8.9 Hz, 1H), 7.36−7.31 (m, 2H), 7.26−7.21 (m, 2H),
7.11−7.05 (m, 1H), 6.97 (d, J = 8.9 Hz, 1H), 4.43 (t, J = 6.6 Hz, 2H),
4.40 (s, 2H), 3.88 (s, 3H), 3.69 (s, 3H), 3.13 (t, J = 6.6 Hz, 2H). 13C
NMR (151 MHz, CDCl3) δ 192.50, 158.13, 153.32, 147.12, 142.12,
141.70, 138.06, 133.43, 129.08, 128.48, 126.52, 126.23, 124.57, 107.37,
74.92, 60.96, 56.18, 44.95, 36.98. Alkyl halide: 1-bromo-2-phenyl-
ethane.
2-(3,5-Dichloropyridin-4-yl)-1-[3,4-dimethoxy-2-(3-
phenylpropoxy)phenyl]ethanone (10). UPLC−MS (ESI+) m/z
460.11 (MH+). 1H NMR (600 MHz, CDCl3) δ 8.50 (s, 2H), 7.60 (d, J
= 8.9 Hz, 1H), 7.29−7.25 (m, 2H), 7.24−7.20 (m, 2H), 7.20−7.16 (m,
1H), 6.75 (d, J = 8.9 Hz, 1H), 4.69 (s, 2H), 4.28 (t, J = 6.7 Hz, 2H),
3.93 (s, 3H), 3.88 (s, 3H), 2.85 (dd, J = 8.7, 6.7 Hz, 2H), 2.28−2.16
(m, 2H). 13C NMR (151 MHz, CDCl3) δ 192.81, 158.08, 153.49,
147.20, 142.02, 141.57, 141.31, 133.51, 128.47, 128.36, 126.08, 126.05,
124.91, 107.23, 74.28, 61.03, 56.17, 45.07, 32.30, 32.00. Alkyl halide: 1-
bromo-3-phenylpropane.
1
(MH+). H NMR (300 MHz, DMSO-d6) δ 8.63 (s, 2H), 7.49 (d, J =
9.0 Hz, 1H), 6.95 (d, J = 9.0 Hz, 1H), 4.90 (s, 2H), 4.83 (s, 2H), 3.89
(s, 3H), 3.78 (s, 3H). 13C NMR (151 MHz, CDCl3) δ 191.92, 168.54,
158.21, 152.17, 147.24, 141.79, 140.81, 137.93, 133.41, 128.50, 127.87,
127.33, 126.43, 122.84, 107.04, 73.10, 61.07, 56.27, 43.18, 43.09. Yield
0.85g (92%).
General Protocol for Preparing Compounds 12, 15, 16, 17,
18, 19, and 20. To a solution of a carboxylic acid (13 or 14, 0.05
mmol) in dry DMF (0.2 mL) was added an amine (0.15 mmol
dissolved in DMF (50 μL) followed by DIPEA (0.15 mmol) and
HATU (0.075 mmol) dissolved in DMF (50 μL). The reaction
mixture was stirred at room temperature for 48 h. The pure
compounds were obtained by standard preparative HPLC purification.
4-[2-[6-[2-(3, 5-Dichloro-4-pyridyl)acetyl]-2, 3-
dimethoxyphenoxy]ethyl]-N-methylbenzamide (12). UPLC−
MS (ESI+) m/z 503.11 (MH+). 1H NMR (300 MHz, CDCl3) δ
8.47 (s, 2H), 7.71−7.63 (m, 2H), 7.61 (d, J = 8.9 Hz, 1H), 7.42−7.31
(m, 2H), 6.75 (d, J = 8.9 Hz, 1H), 6.11 (q, J = 4.8 Hz, 1H), 4.48 (t, J =
6.6 Hz, 2H), 4.41 (s, 2H), 3.92 (s, 3H), 3.75 (s, 3H), 3.21 (t, J = 6.6
2-(3,5-Dichloro-4-pyridyl)-1-[2-[2-(4-hydroxyphenyl)-
ethoxy]-3,4-dimethoxyphenyl]ethanone (11). UPLC−MS (ESI
1
+) m/z 462.09 (MH+). H NMR (300 MHz, DMSO-d6) δ 9.06 (s,
1H), 8.62 (s, 2H), 7.51 (d, J = 9.0 Hz, 1H), 7.17−7.04 (m, 2H), 6.95
(d, J = 9.0 Hz, 1H), 6.66−6.52 (m, 2H), 4.40 (s, 2H), 4.36 (t, J = 6.7
Hz, 2H), 3.88 (s, 3H), 3.72 (s, 3H), 3.01 (t, J = 6.7 Hz, 2H). 13C NMR
(151 MHz, DMSO) δ 192.18, 157.87, 155.74, 152.63, 146.92, 141.55,
141.49, 132.60, 129.70, 127.97, 125.43, 123.51, 114.93, 107.91, 74.74,
60.38, 56.09, 44.46, 35.14. Alkyl halide: 1-bromo-3-(4-hydroxyphenyl)-
propane. NaH was used in place of K2CO3 and heating to 60 °C
overnight was applied to obtain the desired product.
I
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
Hz, 2H), 2.96 (d, J = 4.8 Hz, 3H). Carboxylic acid: 13. Amine:
methylamine hydrochloride.
ActiveSight, San Diego, CA, U.S., with their off-the-shelf protein
structure determination service. The structures were solved to a
resolution of 2.2 and 2.5 Å, respectively, using a PDE4D construct
with residues 153−512 (1XM4 residue numbering).
N-Benzyl-2-{6-[2-(3,5-dichloropyridin-4-yl)acetyl]-2,3-
dimethoxyphenoxy}acetamide (15). UPLC−MS (ESI+) m/z
1
400.03 (MH+). H NMR (400 MHz, DMSO-d6) δ 8.74 (t, J = 6.1
Hydrogens were added, H-bonds were optimized, and the complex
was minimized using the protein preparation wizard implemented in
Maestro.
Hz, 1H), 8.64 (s, 2H), 7.61 (d, J = 8.9 Hz, 1H), 7.36−7.14 (m, 5H),
7.00 (d, J = 8.9 Hz, 1H), 4.76 (s, 2H), 4.73 (s, 2H), 4.35 (d, J = 6.1
Hz, 2H), 3.90 (s, 3H), 3.77 (s, 3H). Carboxylic acid: 14. Amine:
benzylamine.
Modeling. The dihedral coordinate scans of 1, 2, and 4 (enol
form) were done with MacroModel (OPLS-2005, water, TNCG
minimization) using 5° increments. The QM energies of 1−4 listed in
Table 1 were obtained by taking the local minimum conformation
from the coordinate scan (see Figure 3B) with the dihedral closest to
that observed for piclamilast in the 1XM4 structure (144.5°).
Subsequently, a QM minimization was performed with Jaguar. DFT
was used with default settings (B3LYP, 6-31**). Solvent was set to
Water (PBF model). Finally, the QM energy of the minimized
conformation was estimated using DFT settings B3LYP-MM, cc-
pVDZ+2, SM8 solvent model. A similar procedure was performed for
the global minimum conformation from the coordinate scan, and the
relative QM energy of the bioactive conformation was taken as the
difference.
2-{6-[2-(3,5-Dichloropyridin-4-yl)acetyl]-2,3-dimethoxyphe-
noxy}-N-phenylacetamide (16). UPLC−MS (ESI+) m/z 475.07
1
(MH+). H NMR (300 MHz, CDCl3) δ 9.96 (s, 1H), 8.51 (s, 2H),
7.74 (d, J = 8.9 Hz, 1H), 7.64−7.50 (m, 2H), 7.34−7.18 (m, 2H),
7.14−7.00 (m, 1H), 6.81 (d, J = 8.9 Hz, 1H), 4.80 (s, 2H), 4.65 (s,
2H), 3.99 (s, 3H), 3.89 (s, 3H). 13C NMR (75 MHz, CDCl3) δ
192.74, 166.91, 158.56, 152.63, 147.30, 141.76, 140.90, 138.00, 133.50,
128.78, 126.69, 124.07, 122.28, 119.53, 106.83, 73.48, 61.10, 56.34,
42.80. Carboxylic acid: 14. Amine: aniline.
2-{6-[2-(3,5-Dichloropyridin-4-yl)acetyl]-2,3-dimethoxyphe-
noxy}-N-methyl-N-phenylacetamide (17). UPLC−MS (ESI+) m/
z 489.09 (MH+). 1H NMR (300 MHz, CDCl3) δ 8.48 (s, 2H), 7.59 (d,
J = 8.9 Hz, 1H), 7.48−7.39 (m, 2H), 7.38−7.31 (m, 1H), 7.25−7.19
(m, 2H), 6.71 (d, J = 8.9 Hz, 1H), 4.83 (s, 2H), 4.69 (s, 2H), 3.90 (s,
3H), 3.77 (s, 3H), 3.31 (s, 3H). 13C NMR (75 MHz, CDCl3) δ
193.06, 167.35, 157.86, 152.46, 147.14, 142.12, 142.06, 141.15, 133.66,
130.14, 128.54, 127.17, 126.29, 124.58, 107.21, 70.67, 61.16, 56.15,
45.41, 37.42. Carboxylic acid: 14. Amine: N-methylaniline.
Docking of the enol form of 9 to PDE4 was performed using the
PDE4−piclamilast ethylisoxazole analogue crystal structure (Figure
4A). Compound 9 was docked flexibly with Glide XP with 0.8 scaling
of the protein atoms and default settings for ligand scaling. The highest
scoring docking pose is shown in Figure 4B.
MacroModel, Jaguar, Glide, and Maestro were used as implemented
2-{6-[2-(3,5-Dichloropyridin-4-yl)acetyl]-2,3-dimethoxyphe-
noxy}-N-(2-hydroxyethyl)acetamide (18). UPLC−MS (ESI+) m/
z 443.08 (MH+). 1H NMR (600 MHz, DMSO-d6) δ 8.64 (s, 2H), 8.14
(t, J = 6.0 Hz, 1H), 7.59 (d, J = 8.9 Hz, 1H), 6.99 (d, J = 8.9 Hz, 1H),
4.77 (s, 2H), 4.65 (s, 2H), 3.90 (s, 3H), 3.79 (s, 3H), 3.41 (t, J = 6.0
Hz, 3H), 3.20 (q, J = 6.0 Hz, 2H). 13C NMR (151 MHz, CDCl3) δ
192.90, 169.85, 158.51, 152.42, 147.38, 141.82, 140.63, 133.37, 126.73,
122.43, 106.96, 73.09, 62.39, 61.14, 56.32, 43.05, 42.35. Carboxylic
acid: 14. Amine: 2-hydroxyethylamine.
in the 2013-2 software suite; Schrodinger, LLC, New York, NY, 2013.
̈
PDE4 Assay. Human recombinant PDE4D (Genbank accession
no. NM_006203) was incubated for 1 h with the test compound at
concentrations up to 10 μM, with cAMP (1 × 10−5 M) and with a low
amount (0.021 MBq) of radioactively labeled cAMP. At the end of the
incubation, the cleavage of the substrate was evaluated by the binding
of the AMP product to SPA beads, which generate chemo-
luminescence when bound to the radioactive tracer. The AMP
product inhibited the binding of the radioactive tracer to the beads,
and the luminescent signal was competed.
2-{6-[2-(3,5-Dichloropyridin-4-yl)acetyl]-2,3-dimethoxyphe-
noxy}-N-(2-methoxyethyl)acetamide (19). UPLC−MS (ESI+) m/
z 457.09 (MH+). 1H NMR (600 MHz, DMSO-d6) δ 8.65 (s, 2H), 8.23
(t, J = 5.6 Hz, 1H), 7.61 (d, J = 8.9 Hz, 1H), 6.99 (d, J = 8.9 Hz, 1H),
4.77 (s, 2H), 4.65 (s, 2H), 3.91 (s, 3H), 3.79 (s, 3H), 3.35 (t, J = 5.6
Hz, 2H), 3.29 (q, J = 5.6 Hz, 2H), 3.16 (s, 3H). 13C NMR (151 MHz,
CDCl3) δ 191.97, 168.72, 158.17, 152.13, 147.24, 141.82, 141.02,
133.54, 126.37, 123.22, 107.25, 73.05, 70.87, 61.14, 58.56, 56.27,
43.52, 38.82. Carboxylic acid: 14. Amine: 2-methoxyethylamine.
2-{6-[2-(3,5-Dichloropyridin-4-yl)acetyl]-2,3-dimethoxyphe-
noxy}-N-propylacetamide (20). UPLC−MS (ESI+) m/z 441.09
The results were calculated as the molar concentrations resulting in
50% inhibition of the substrate cleavage compared to controls samples
and are expressed as a range of IC50 (nM).
TNF-α Release. Human peripheral blood mononuclear cells
(PBMCs) were isolated from buffy coats. The blood is mixed with
saline at a ratio of 1:1, and the PBMCs were isolated using
Lymphoprep tubes (Nycomed, Norway). The PBMCs were
suspended in RPMI 1640 with 0.5% human serum albumin, pen/
strep, and 2 mM L-glutamine at a concentration of 5 × 105 cells/mL.
The cells were preincubated for 30 min with the test compounds in
96-well tissue culture plates and stimulated for 18 h with
lipopolysaccharide 1 mg/mL (Sigma). TNF-α concentration in the
supernatants was measured using homogeneous time-resolved
fluorescence resonance (TR-FRET). The assay is quantified by
measuring fluorescence at 665 nm (proportional to TNF-a
concentration) and 620 nm (control). Results are expressed as IC50
values calculated from inhibition curves using as positive controls the
secretion in LPS stimulated wells and as negative controls the
secretion in unstimulated cells.
1
(MH+). H NMR (600 MHz, CDCl3) δ 8.52 (s, 2H), 7.87−7.76 (m,
1H), 7.68 (d, J = 8.9 Hz, 1H), 6.79 (d, J = 8.9 Hz, 1H), 4.67 (s, 2H),
4.60 (s, 2H), 3.97 (s, 3H), 3.87 (s, 3H), 3.35−3.21 (m, 2H), 1.50 (h, J
= 7.3 Hz, 2H), 0.87 (t, J = 7.3 Hz, 3H). 13C NMR (151 MHz, CDCl3)
δ 192.13, 168.59, 158.24, 152.33, 147.25, 141.80, 140.93, 133.46,
126.40, 122.88, 106.92, 73.16, 61.08, 56.28, 43.15, 40.79, 22.50, 11.37.
Carboxylic acid: 14. Amine: propylamine.
N-Benzyl-2-{6-[2-(3,5-dichloro-1-oxypyridin-4-yl)acetyl]-2,3-
dimethoxyphenoxy}acetamide (21). To a solution of 15 (10 mg,
0.021 mmol) in dry DCM (0.25 mL) was added methyltrioxorhenium
(26 mg, 0.105 mmol) followed by 30% hydrogen peroxide (6.4 μL,
0.063 mmol). The reaction mixture was stirred at room temperature
overnight. MnO2 (5.5 mg, 0.063 mmol) was added, and after 2 min the
reaction mixture was filtered and the solvent evaporated in vacuo. The
title compound 21 was obtained by standard preparative HPLC
HLM (Human Liver Microsomes) Assay. Incubations of test
compounds in DMSO, diluted with phosphate buffer, pH 7.4, at 0.5
μM were carried out with human liver microsomes (0.5 mg/mL). The
percentage of organic solvent in the incubations was 1%. The human
liver microsomal suspension in phosphate buffer was mixed with
NADPH (1 mM) and preheated to 37 °C before test compound was
added. Aliquots were taken at 0, 5, 10, 20, and 30 min, and reactions
were terminated by addition of methanol containing analytical internal
standard (IS).
1
purification. UPLC−MS (ESI+) m/z 505.09 (MH+). H NMR (300
MHz, CDCl3) δ 8.15 (2H, s), 8.10 (1H, bs), 7.62 (1H, d), 7.30 (5H,
m), 6.76 (1H, d), 4.72 (2H, s), 4.50 (2H, d), 4.46 (2H, s), 3.95 (3H,
s), 3.80 (3H, s). 13C NMR (75 MHz, CDCl3) δ 191.57, 168.50,
158.44, 152.25, 141.81, 137.90, 137.45, 134.13, 131.63, 128.52, 127.97,
127.41, 126.43, 122.48, 107.13, 73.11, 61.09, 56.30, 43.21, 42.32.
Crystallography. The crystal structures of the ethylisoxazole
analogue of piclamilast and 15 complexed with PDE4D were solved at
The results were expressed as apparent clearance (Clapp) (mL min−1
kg−1) and hepatic extraction ratio (Eh) (%) calculated from the rate
constant (k) (min−1) of test compound depletion. The upper limit for
determination of Clapp is 200 mL min−1 kg−1 (corresponding to
J
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Drug Annotation
extraction ratio of 91%), and the lower limit is 10 mL min−1 kg−1
(corresponding to extraction ratio of 33%).
Structural basis for the activity of drugs that inhibit phosphodies-
terases. Structure 2004, 12, 2233−2247.
Emesis Study. Domestic ferret (Mustela putorius furo), 24 weeks,
1.0−1.3 kg were dosed iv with compound dissolved 2-hydroxypropyl-
β-cyclodextrin in sterile water (volume 5 mL/kg). After treatment,
animals were singly housed (polycarbonate type E, Charles River
Laboratories; 405 mm × 255 mm × 197 mm) and emetic episodes and
associated prodromic signs were counted for 2 h. Emesis parameters
were the following: number of ferrets showing retches and vomits;
latency to first retching (hh:mm); latency to first vomiting (hh:mm);
retching (number of retches); vomiting (number of vomits).
Prodromic signs were the following: licking; mouth scratching;
yawning; wet dog shakes and backward walking.
(11) Ricerca Biosciences LLC (formerly MDS Pharma Services).
http://www.ricerca.com/.
(12) GreenScreen HC. http://www.gentronix.co.uk/product/
greenscreen/.
(13) Griffiths, C. E.; Van Leent, E. J.; Gilbert, M.; Traulsen, J.;
Cipamyflline Study Group.. Randomized comparison of the type 4
phosphodiesterase inhibitor cipamfylline cream, cream vehicle and
hydrocortisone 17-butyrate cream for the treatment of atopic
dermatitis. Br J. Dermatol. 2002, 147, 299−307.
(14) LEO 29102 Single and Multiple Dose Study by Dermal
Application, NCT00891709. http://www.ClinicalTrials.gov.
(15) LEO 29102 Cream in the Treatment of Atopic Dermatitis,
NCT01037881. http://www.ClinicalTrials.gov.
AUTHOR INFORMATION
■
(16) Felding, J.; Nielsen, S. F.. Substituted Acetophenones Useful as
PDE4 Inhibitors. U.S. Patents US 8,148,537 B2 and US 8,324,394 B2,
2012.
Corresponding Author
*E-mail: simon.nielsen@leo-pharma.com. Phone: +45
72263703.
(17) R&D Insight Database. http://www.adis.com.
Present Addresses
†T.D.P.: DTU Admission Course, Lautrupvang 15, DK-2750
Ballerup, Denmark.
‡C.A.: Novo Nordisk A/S, Immunopharmacology 479, Novo
Nordisk Park, DK-2760 Maløv, Denmark.
̊
§A.M.V.: Division of Toxicology and Risk Assessment, National
Food Institute, Technical University of Denmark, Mørkhøj
Bygade 19, DK-2860 Søborg, Denmark.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank all the dedicated PDE4 people at LEO for making
the selection of LEO 29102 a reality.
REFERENCES
■
(1) Hanifin, J. M.; Reed, M. L. A population-based survey of eczema
prevalence in the United States. Dermatitis 2007, 18, 82−91.
(2) Watson, W.; Kapur, S. Atopic dermatitis. J. Allergy Clin. Immunol.
2011, 7 (Suppl. 1), S4.
(3) Lewis-Jones, S. Quality of life and childhood atopic dermatitis:
the misery of living with childhood eczema. Int. J. Clin. Pract. 2006, 60,
984−992.
(4) Zheng, T.; Yu, J.; Oh, M. H.; Zhu, Z. The atopic march:
progression from atopic dermatitis to allergic rhinitis and asthma. J.
Allergy Clin. Immunol. 2011, 3, 67−73.
(5) Thaci, D.; Salgo, R. Malignancy concerns of topical calcineurin
inhibitors for atopic dermatitis: facts and controversies. Clin. Dermatol.
2010, 28, 52−56.
(6) Bender, A. T.; Beavo, J. A. Cyclic nucleotide phosphodiesterases:
molecular regulation to clinical use. Pharmacol. Rev. 2006, 58, 488−
520.
(7) Kodimuthali, A.; Jabaris, S. S. L.; Pal, M. Cyclic nucleotide
phosphodiesterases: molecular regulation to clinical use. J. Med. Chem.
2008, 51, 5471−5489.
(8) Giembycz, M. A. Can the anti-inflammatory potential of PDE4
inhibitors be realized: guarded optimism or wishful thinking? Br. J.
Pharmacol. 2008, 155, 288−290.
(9) Ashton, M. J.; Cook, D. C.; Fenton, G.; Karlsson, J.-A.;
Palfreyman, M. N.; Raeburn, D.; Ratcliffe, A. J.; Souness, J. E.;
Thurairatnam, S.; Vicker, N. Selective type IV phosphodiesterase
inhibitors as antiasthmatic agents. The syntheses and biological
activities of 3-(cyclopentyloxy)-4-methoxybenzamides and analogues.
J. Med. Chem. 1994, 37, 1696−1703.
(10) Card, G. L.; England, B. P.; Suzuki, Y.; Fong, D.; Powell, B.; Lee,
B.; Luu, C.; Tabrizizad, M.; Gillette, S.; Ibrahim, P. N.; Artis, D. R.;
Bollag, G.; Milburn, M. V.; Kim, S.-H.; Schlessinger, J.; Zhang, K. Y. J.
K
dx.doi.org/10.1021/jm500378a | J. Med. Chem. XXXX, XXX, XXX−XXX