Synthesis and molecular docking of new roflumilast analogues as preferential-selective potent PDE-4B inhibitors with improved pharmacokinetic profile

Article abstract of DOI:10.1016/j.ejmech.2018.02.038

In the present work, we designed and synthesized new roflumilast analogues with preferential-selective PDE-4B inhibition activity and improved pharmacokinetic properties. The unsubstituted benzo[d]thiazol-2-yl and -6-yl benzamide derivatives (4a and 6a) showed both good potency and preferential selectivity for PDE-4B. More remarkably, 6c revealed 6 times preferential PDE-4B/4D selectivity with a significant increase of in vitro cAMP and good % inhibition of TNF-α concentration. In addition, the in vitro pharmacokinetics of 6c showed good metabolic stability with in vitro CL_int (5.67 mL/min/kg) and moderate % plasma protein binding (53.71%). This was reflected onto increased in vivo exposure with a half-life greater than roflumilast by 3 folds (21 h) and a C_max value of 113.958 ng/mL. Molecular docking attributed its good activity to its key binding interactions in PDE-4B active site with additional hydrogen bonding with amino acids lining the metal pocket. Summing up, 6c can be considered as suitable candidate for further investigation for the treatment of COPD.

Full text of DOI:10.1016/j.ejmech.2018.02.038

Accepted Manuscript
Synthesis and molecular docking of new roflumilast analogues as preferential-
selective potent PDE-4B inhibitors with improved pharmacokinetic profile
Bahia A. Moussa, Asmaa A. El-Zaher, Mohamed K. El-Ashrey, Marwa A. Fouad
PII:
S0223-5234(18)30168-5
10.1016/j.ejmech.2018.02.038
EJMECH 10215
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 14 December 2017
Revised Date: 8 February 2018
Accepted Date: 12 February 2018
Please cite this article as: B.A. Moussa, A.A. El-Zaher, M.K. El-Ashrey, M.A. Fouad, Synthesis and
molecular docking of new roflumilast analogues as preferential-selective potent PDE-4B inhibitors
with improved pharmacokinetic profile, European Journal of Medicinal Chemistry (2018), doi: 10.1016/
j.ejmech.2018.02.038.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
8
7
6
5
4
3
2
1
0
6c
In vitro t1/2
247.55 min.
53.71
Rof
6c
%PPB
PDE-4B IC50 4D/4B Ratio TNF-α cAMP

ACCEPTED MANUSCRIPT
Synthesis and Molecular Docking of New Roflumilast Analogues as
Preferential-Selective Potent PDE-4B Inhibitors with Improved
Pharmacokinetic Profile
Bahia A. Moussa, Asmaa A. El-Zaher, Mohamed K. El-Ashrey and Marwa A. Fouad*
Pharmaceutical Chemistry Department, Faculty of Pharmacy, Cairo University, Kasr El-Eini
Street, P.O. Box 11562, Cairo, Egypt
______________________________________________________________________
*Corresponding author: ORCID Marwa Fouad: 0000-0003-3227-8683
E-mail address: marwa.fouad@pharma.cu.edu.eg
Tel: +201222441198
1

ACCEPTED MANUSCRIPT
Abstract
In the present work, we designed and synthesized new roflumilast analogues with
selective PDE-4B inhibition activity and improved pharmacokinetic properties. The
unsubstituted benzo[d]thiazol-2-yl and -6-yl benzamide derivatives (4a and 6a) showed both
good potency and preferential selectivity for PDE-4B. More remarkably, 6c revealed 6 times
preferential PDE-4B/4D selectivity with a significant increase of in vitro cAMP and good %
inhibition of TNF-α concentration. In addition, the in vitro pharmacokinetics of 6c showed good
metabolic stability with in vitro CLint (5.67 mL/min/kg) and moderate % plasma protein binding
(53.71%). This was reflected onto increased in vivo exposure with a half-life greater than
roflumilast by 3 folds (21 h) and a Cmax value of 113.958 µg/mL. Molecular docking attributed its
good activity to its key binding interactions in PDE-4B active site with additional hydrogen
bonding with amino acids lining the metal pocket. Summing up, 6c can be considered as suitable
candidate for further investigation for the treatment of COPD.
Keywords: Phosphodiesterase 4B inhibitor; 2-aminobenzothiazole; 6-aminobenzothiazole; 2-
aminothiazole; Chronic Obstructive Pulmonary disorder; Pharmacokinetics
2

ACCEPTED MANUSCRIPT
1
Introduction
Chronic obstructive pulmonary disease (COPD) has an increasing rate of morbidity and
mortality that it kills about 3 million people all over the world every year [1,2]. Many
inflammatory cells and mediators participate in COPD symptoms such as bronchitis and
emphysema, so prevention of their activation or antagonizing their receptors become a challenge
for a medicinal chemist to find new chemical entities (NCEs) that play this role [3–5].
Many treatments had been developed for COPD as inhaled corticosteroids [6–9] , inhibitors of
inducible nitric oxide synthase (i-NOS) [10], leukotriene inhibitors [11,12], adhesion-molecule
blockers [13], chemokine inhibitors [14], TNFα inhibitors [15], NFκB inhibitors [16], p38
MAPK inhibitors [17], Phosphoinositide 3-kinase (PI3K) inhibitors [18] and Peroxisome
proliferator-activated receptors (PPARs) activators [19–21].
The discovery of selective phosphodiesterase 4 (PDE-4) inhibitors as a way for treatment of
inflammation in COPD patients was prompted when it has been observed that raising the
intracellular levels of 3′5′-cyclic adenosine monophosphate (cAMP) can inhibit the functions of
inflammatory cells and also due to the wide distribution of PDE-4 enzymes in inflammatory cells
and in the lung [22–24],
There are 11 distinct enzymes of PDEs [25,26] in which PDE-4 specifically inactivates cAMP
which is highly expressed in inflammatory cells, and so, inhibitors of PDE-4 enzyme cause a
decrease in inflammatory response which is increased in COPD [27,28], PDE-4 enzymes are
classified into 4 subtypes (4A-D) which show a similarity of 78% in the active catalytic site [29].
It has been revealed that selective inhibition of PDE-4B showed a potent anti-inflammatory
effect not associated with emesis side effect accompanied with the inhibition of PDE-4D enzyme
[30–32].
3

ACCEPTED MANUSCRIPT
Roflumilast was marketed in the USA in 2012 for the treatment of COPD as a PDE-4 inhibitor
[33,34]. Though, its side effects, such as gastrointestinal upsets, headache, and weight loss
limited its therapeutic potential [35].
Roflumilast is rapidly metabolized to its active metabolite, roflumilast N-oxide, which has
twofold to threefold specificity and potency less than roflumilast [36]. The main cytochrome
P450 enzymes that are responsible for the conversion of roflumilast to its N-oxide metabolite are
CYP3A4 and 1A2 [37,38]. As the activity of these two enzymes can be affected by
interindividual variables, so it is can be predicted that these variables can affect roflumilast
pharmacokinetics [39–41].
As reported [42,43], “PDE-4B and PDE-4D enzymes are composed of a compact α-helical
structure that forms three subdomains. The PDE active site forms a deep pocket at the junction of
these three subdomains. The active site can be subdivided into three pockets: a metal binding
pocket which contains magnesium and zinc ions in addition to highly conserved hydrophobic
and polar residues that coordinates the metal ions (M pocket), a solvent filled side pocket which
consists mainly of hydrophilic amino acids and is filled with a network of water molecules in
most of the inhibitor complexes (S pocket), and a third pocket containing the invariant purine-
selective glutamine and a pair of conserved residues that form a hydrophobic clamp (Q pocket)”.
In an effort to discover PDE4B/4D selective inhibitors, a 2-arylpyrimidine derivative I was
identified as a selective and potent PDE4B inhibitor with an IC50 of sub-micromolar range (0.19
µM) and a potential 10-fold selectivity over PDE4D.[44] By high throughput screening,
tetrahydrobenzothiophene bisamide II was found to be a potent and modest PDE4B/4D-selective
inhibitor (PDE4B/4D pIC50 6.7/6.5).[45] The pyridazino[4,5-b]indolizine analogue III was
found to possess a 23-fold selectivity for PDE4B over PDE4D.[46] Compound IV was identified
4

ACCEPTED MANUSCRIPT
as an orally active PDE4B selective inhibitor over PDE4D both in humans (80-fold selective)
and mice (29-fold selective) (Figure 1).[47]
(I)
(II)
H
N
N
N
N
N
N
N
N
(III)
Figure 1. Structures of some reported PDE4B/4D selective inhibitors
(IV)
Reported research for the discovery of new PDE-4 inhibitors showed that some heterocyclic
groups, such as morpholine, phthalazinone, and benzothiazole could play the role of pyridine
moiety in roflumilast in affecting biological activity [48–52].
Accordingly, we can design a NCE through the hybridization of two reported pharmacophores
into one molecule [53–55]. Therefore, these previously mentioned facts urge the need for the
discovery of novel selective PDE-4B inhibitors with good potency and better metabolic profile.
This work deals with synthesis of novel series 4, 6 and 8 through hybridization of the
pharmacophores, 3-(cyclopropylmethoxy)-4-(difluoromethoxy)benzoic acid with amino-2-
thiazole, 2- or 6-benzothiazole via amide linkage (Chart. 1), hoping to obtain new PDE-4B
5

ACCEPTED MANUSCRIPT
inhibitors hits, followed by an investigation of the pharmacokinetics of inhibitors showing
promising potency. Eventually, lead compound 6c showed an improved pharmacokinetic profile
while retaining a PDE-4B inhibitory activity comparable to that of roflumilast with preferential
selectivity over PDE-4D.
Chart 1. Design of new compounds based on Roflumilast
2
Results and Discussion
2.1 Chemistry
The synthetic pathway for the preparation of the target compounds 4a-j, 6a-c and 8 is outlined
in Scheme 1. First, 2-aminobenzothiazoles 2c,d,f,g,i were prepared according to the reported
methods [56–59] from the corresponding substituted anilines 1. Many chemical approaches are
known for amide coupling.
N,N′-dicyclohexylcarbodiimide (DCC) can be utilized as water
absorbing agent [60] or phosphorous trichloride (PCl3) [61,62] as a dehydrating agent in the
condensation of aniline derivatives with carboxylic acids. Thionyl chloride is the most popular
reagent to activate carboxylic function because it is volatile and non-expensive [63]. The highest
6

ACCEPTED MANUSCRIPT
yield was obtained by using the one pot amide synthesis reaction by treating the starting acid 3
with thionyl chloride and the excess was then removed easily by distillation. Target compounds
were then obtained through coupling reaction [64] with the formed acid chloride by the addition
of the corresponding amine derivatives 2a-j, 5a-c and 7 in dry benzene in the presence of
anhydrous potassium carbonate.
Scheme 1. Preparation of Target Compounds
O
O
NH2
N
S
b
a
O
F
F
O
F
F
R
R
NH2
+
O
O
N
S
NH
R
HO
1
2a-j
3
4a-j
a: H
N
S
NH2
R =
N
NH2
b: 4-OCH3
c: 6-OCH3
d: 5,6-
b
S
b
R
7
5a-c
dimethoxy
e:
6-
OCH2CH3
f:
O
O
O
F
F
6-F
O
F
F
O
N
S
g: 6-Cl
h: 6-CF3
i: 6-CN
O
NH
NH
N
8
j:
6-NO2
S
R
6a-c
R =
a:
b:
CH3
c: SH
H
a: NH4SCN / HCl / Br2 / glacial acetic acid, b: SOCl2 / Anhyd K2CO3, Dry Benzene
2.2 Biological evaluation
All the newly synthesized compounds were evaluated for their ability to inhibit the PDE-4B
enzyme using Roflumilast as positive control. The IC50 values were determined by nonlinear
regression analysis of their inhibition curves and shown in table 1. All the synthesized
compounds showed potent PDE-4B inhibition activity in the nanomolar range. The three
7

ACCEPTED MANUSCRIPT
compounds,4a, 6a and 6c showed higher potency than Roflumilast with subnanomolar IC50
values.
Thorough study of the structure-activity relationship of the unsubstituted derivatives (4a, 6a
and 8) revealed that 4a and 6a showed good potency with no significant inhibition difference
from Roflumilast while replacement of the benzothiazole ring in either 4a or 6a with the
simplified thiazole ring (8) significantly decreased the potency of inhibition. Moreover,
substitution of the benzothiazole ring in series 4 with either an electron donating group such as
alkoxy or dialkoxy group or an electron withdrawing group such as halogens, trifluoromethyl,
cyano or nitro group (4b-j) significantly decreased the potency of the unsubstituted compound
(4a). The trifluoromethyl and the nitro substituted 2-aminobenzothiazole derivatives (4h and 4j,
respectively) exhibited the lowest potency in this series (p<0.05).
On the other hand, substitution of 6-aminobenzothiazole derivatives (6a-c) at position 2 with a
methyl group (6b) significantly increased the IC50 value of the unsubstituted derivative while
substitution with a mercapto group (6c) showed good potency with no significant difference with
the parent compound (6a) (p<0.05).
Compounds (4a, 6a and 6c) were selected to study their PDE-4B/4D selectivity (Table 2,
Figure 2). All the 3 compounds showed preferential selectivity towards PDE-4B isozyme
compared to the reference compound.
As 4a and 6c showed the highest selectivity, they were subjected to further investigations and
their effect on TNF-α and cAMP levels were studied and compared to those of Roflumilast.
Both, 4a and 6c, exhibited an increase in the level of cAMP higher than that of roflumilast. 6c
significantly increases cAMP level for more than 2 folds than roflumilast with a significant
difference at p<0.05. Decreasing TNF-α level assists in anti-inflammatory action of PDE-4
8

ACCEPTED MANUSCRIPT
inhibitors. The synthesized derivative 6c showed a higher level of inhibition of TNF-α than4a
derivative (Table 2, Figure 2).
Table 1. PDE-4B inhibitory data
Compd
R
IC50 [nM]
0.90
-
H
Roflumilast
0.57
4a
4b
4c
4d
4e
4f
4-OCH3
6-OCH3
5,6-Dimethoxy
6-OCH2CH3
6-F
1.06
2.28*
1.17
2.92*
1.87*
1.22
6-Cl
4g
4h
4i
6-CF3
6-CN
5.23*
1.28
6-NO2
H
8.67*
0.68
4j
6a
6b
6c
8
2-CH3
2-SH
1.26
0.75
-
1.60*
* IC50 is significantly different from that of Roflumilast at P<0.05
9

ACCEPTED MANUSCRIPT
Table 2. Biological evaluation results, S9 Stability and human plasma protein binding of
compounds: Roflumilast, 4a, 6a and 6c
Rof
4a
6a
6c
PDE-4B IC50
(nM)
0.90
0.57
0.68
0.75
PDE-4D IC50
(nM)
0.56
0.66
5.07
8.89
2.73
4.01
ND
4.54
6.05
4D/4B Ratio
TNF-α
(%Inhibition)
70.32
58.61
67.79
cAMP
1.898
12.29
2.676
42.27
ND
4.446
(pM/mL)
In vitro t1/2
63.01
247.55
(min.)
CLint
(mL/min/kg)
114.19
98.49
33.20
100
22.27
96.39
5.67
53.71
%Bound
*ND: not done
10

ACCEPTED MANUSCRIPT
10
9
8
7
6
5
4
3
2
1
0
PDE-4B IC50
4D/4B Ratio
TNF-α
cAMP
Rof
4a
6c
Figure 2. In vitro biological results of 4a and 6c compared to Roflumilast.
2.3 Molecular docking study
Many crystal structures for PDE-4B and PDE-4D isozymes co-crystallized with different
inhibitors are available in the protein data bank. To undergo the docking study, we retrieved the
X-ray crystal structure of PDE-4B (PDB ID: 1XMU, resolution 2.3 Å) and PDE-4D (1XOQ,
resolution 1.83 Å) [42] that are in complex with roflumilast.
Validation of the molecular docking setup was carried out by re-docking of roflumilast into the
PDE-4B and PDE-4D active sites. Nearly the same binding of the co-crystallized ligand was
reproduced in the validation step including key interactions [42] thus indicating that the used
setup is suitable for our docking study. As the co-crystallized ligand, the dialkoxyphenyl ring of
the docked roflumilast is inserted in between the hydrophobic clamp: Phe446 and Ile410 in PDE-
4B (Phe372 and Ile336 in PDE-4D) Q pocket. Also, the dichloropyridyl group forms one H bond
to a water molecule coordinated to Mg2+ in the M pocket (Supporting information, Figure S1).
11

ACCEPTED MANUSCRIPT
This was also confirmed by the small RMSD of 0.4555Å between the co-crystallized ligand and
the reproduced docked pose (energy score (S) = -25.1316 kcal/mol) for PDE4B (PDB ID:
1XMU) and RMSD of 0.7788Å (energy score (S) =-24.7497 kcal/mol) for PDE4D (PDB ID:
1XOQ).
By comparing docking of the new compounds with the reference compound in PDE-4B active
site, all the synthesized compounds interact through the oxygens of the dialkoxyphenyl moieties
with the NH of the γ-carboxamide group of Gln443 which is involved in nucleotide recognition,
through a single or bidentate H bonds accommodating the dialkoxyphenyl ring in the
hydrophobic clamp. The difluoromethoxy group is inserted into the small Q1 pocket and the
cyclopropylmethoxy group occupies the larger hydrophobic Q2 pocket and forms a hydrophobic
interaction with Phe446. Moreover, the primary aromatic ring exhibited CH-π interaction with
Ile410. These interactions play an important role in the stabilization of the inhibitor at the active
site (Supporting information, Figure S2-S8).
All the compounds are generally oriented such that their benzothiazole or thiazole moiety is
directed into the metal binding pocket. In contrast to compound 8, both 2-amino and 6-
aminobenzothiazole derivatives (series 4 and 6) perform H bond interaction with a water
molecule coordinated with Mg2+ cation through the π electrons of the benzothiazole moiety. This
orientation may interfere with the approach of cAMP to the catalytic domain.
Moreover, Compound 6c exhibited additional two H bonds with the α-carboxamido group of
Asn283 and the γ-carboxylate group of Glu304 through the 2-SH group as hydrogen acceptor
and hydrogen donor, respectively. These two residues are among those lining the metal binding
pocket (Figure 3).
12

ACCEPTED MANUSCRIPT
Figure 3. 3D Representation of the interaction of the docked 6c in the active site of PDE-4B
enzyme (PDB ID: 1XMU) using Molecular Operating Environment (MOE, 10.2008) software
In summary, this set of interactions ensures that the newly synthesized derivatives achieve a
strong binding to the PDE-4B enzyme comparable to that of the reference compound and thus,
good biological activity. The difference in experimental IC50 values among tested compounds is
well reflected in their docking scores (Table 3).
13

ACCEPTED MANUSCRIPT
Table 3. Docking energy scores (S) in kcal/mol for the newly synthesized compounds and
the reference compound in PDE-4B and PDE-4D enzymes.
Energy score (S)
kcal/mol (PDE4B)
-25.13
Energy score (S)
kcal/mol (PDE4D)
-24.75
Compd
Roflumilast
-26.35
-24.58
4a
4b
4c
4d
4e
4f
-22.18
-26.15
-20.53
-24.64
-22.93
-22.63
-20.89
-23.06
-23.90
-24.11
-22.66
-20.08
4g
4h
4i
-19.95
-20.09
-22.08
-20.32
-19.05
-22.20
4j
-25.89
-24.87
6a
6b
6c
8
-22.38
-25.95
-25.59
-23.68
-19.70
-21.96
Recent studies have indicated overall similarity of the catalytic domain in the PDE4
subfamilies with a high degree of sequence conservation [42,65]. Although the binding pattern of
the new compounds in PDE-4B and PDE-4D is comparable except for 6c (Supporting
information, Figure S9-S12), the difference between the docking scores of roflumilast, 4a, 6a
14

ACCEPTED MANUSCRIPT
and 6c when docked in PDE-4B and PDE-4D reflects, to a great extent, the in vitro experimental
IC50 values difference thus ascertaining their preferential binding affinity.
2.4 In vivo pharmacokinetic study
Pharmacokinetic parameters were estimated for the most potent synthesized derivatives 4a, 6a
and 6c in male Sprague-Dawley rats using Roflumilast as the reference compound. The three
compounds showed good pharmacokinetic parameters with a Cmax value of 95.491, 86.241 and
113.958
622.382
µ
g/mL comparable to that of roflumilast and an AUC value of 594.078, 892.133 and
g.h/mL, respectively, higher than that of roflumilast after a subcutaneous dose of 1
µ
mg/kg. In addition, they demonstrated relatively lower clearance rate (CL= 1.250, 0.619 and
0.960 mL/h.kg) than that of the reference drug. These data suggested that replacement of the
dichloropyridyl moiety of roflumilast, which is subjected to rapid N-oxidation, by a
benzothiazole ring might attribute to the improvement of the pharmacokinetic profile. Moreover,
6c demonstrated an excellent pharmacokinetic profile with the highest half-life > 21 h and the
highest maximum plasma concentration, Table 4, Figure 4.
Table 4. Sprague-Dawley rat PK Profile of Roflumilast and compounds 4a, 6a and 6c
Parameter
Rat no.
Rof
6
4a
6
6a
6
6c
6
Dose level (mg/kg)
t1/2 (h)
1
1
1
1
6.938
1
9.819
1
21.959
2
21.991
2
Tmax (h)
Cmax
AUC0-t (
AUC0-
(
µ
g/mL)
g.h/mL)
g.h/mL)
Vd (mL/kg)
108.751
449.426
499.506
20.039
95.491
594.078
799.823
17.712
86.241
892.133
1616.333
19.600*
113.958
622.382
1041.629
30.459
µ
(
µ
∞
15

ACCEPTED MANUSCRIPT
CL (mL/h.kg)
2.002
1.250
0.619
0.960
* Not significantly different from that of Roflumilast at P<0.05
140
120
100
80
ROF
4a
6a
60
6c
40
20
0
0
5
10
15
20
25
30
-20
Time (hr)
Figure 4. Time-dependent plasma concentrations of Rof, 4a, 6a and 6c after subcutaneous
administration to male Sprague-Dawley rats.
2.5 In vitro pharmacokinetic studies
2.5.1 Metabolic stability
The most promising inhibitors of PDE4B enzyme, 4a, 6a, 6c, were tested for their metabolic
stability in rat subcellular fraction and compared to roflumilast. This was carried out by
incubation of each inhibitor with the S9 fraction in the presence of the co-factor NADPH for
different time intervals to test their phase I metabolic stability. The in vitro half-life (t1/2) and the
intrinsic clearance (CLint) were calculated (Table 2). By comparing the 3 tested compounds to
roflumilast (t1/2= 12.29 min), it appears that they all have a better metabolic stability especially
the 2-mercapto-6-aminobenzothiazole derivative (6c) (t1/2= 247.55 min) which showed a half-life
time that is 20-fold greater than that of Roflumilast and an encouraging in vitro clearance of 5.67
mL/min/kg. Removal of the mercapto group decreased the half-life of the unsubstituted 6-
16

ACCEPTED MANUSCRIPT
aminobenzothiazole derivative (6a) (t1/2= 63.01 min) 4-folds. Moreover, the unsubstituted 2-
aminobenzothiazole derivative (4a) showed a moderate metabolic stability (t1/2= 42.27 min).
Thus, further investigation was needed in order to predict the possible metabolites of
compounds 4a and 6a and also to evaluate their metabolic fate. Incubation samples (60 min)
were treated and then injected onto UPLC-MS/MS. Metabolites were predicted according to the
reported roflumilast metabolism and expected benzothiazole biotransformations [66,67]. In case
of 4a, the detected metabolic product was derived from aromatic hydroxylation of the
benzothiazole moiety (405.07 m/z). For 6a, aromatic hydroxylated metabolite (405.11 m/z) in
addition to the cleaved benzothiazole (381.09 m/z) were detected (Supporting Information,
Figure S13-S15).
As 6c was the most metabolically stable analog with 84.2% intact after 60 min compared to 4a
and 6a (36.4% and 36.8%, respectively), it can be deduced that substitution at position 2 with the
mercapto group may play an important role in protecting the compound against phase I
metabolism.
2.5.2 Human plasma protein binding
Drug protein binding can impact both the pharmacokinetics and pharmacodynamics of a drug.
Thus, estimation of plasma protein binding is an important part of characterization of a new
chemical entity during its development into a drug [68]. Equilibrium dialysis is considered to be
the best method for the estimation of drug protein binding. Thus, compounds 4a, 6a and 6c were
selected to investigate their plasma protein binding in vitro using equilibrium dialysis method
and the %Bound for each compound was compared to that of reference compound Roflumilast
(Table 2). Compounds 4a and 6a showed very high plasma protein binding (more than 95%)
comparable to that of roflumilast while 6c showed a moderate percentage of plasma protein
17

ACCEPTED MANUSCRIPT
binding (53.71%) illustrating its availability for tissue distribution and avoiding drug-drug
protein displacement interaction.
3
Conclusion
New benzothiazole derivatives as PDE-4B inhibitors were developed by replacing the
dichloropyrid-4-yl amino moiety of roflumilast with either benzothiazol-2-yl, benzothiazol-6-yl
or thiazol-2-yl amino moiety. Most of the compounds showed potent PDE-4B inhibitory
activities. Moroever, 4a, 6a and 6c were found to be more potent than roflumilast and
subsequently selected for further investigation. One of the potent compounds, 6c, was found to
have preferential PDE-4B selectivity 6 times over PDE-4D with a significant increase of in vitro
cAMP concentration and good % inhibition of TNF-
α concentration. The introduction of
mercapto group at 2-position (6c) seems to stabilize the molecule against phase I metabolism
while retaining good PDE-4B IC50 value and potent PDE4B/4D selectivity. the increased
metabolic stability of 6c (in vitro t1/2= 247.55 min., Clint = 5.67 mL/min/kg) as compared to
roflumilast in rat subcellular fraction was reflected onto increased in vivo exposure with a half-
life greater than that of roflumilast by 3 folds (21 h). In addition, it showed a moderate % of
plasma protein binding (53.71%) less than that of roflumilast (98.5%). This new compound can
be considered as a good drug candidate for treating COPD.
4
Experimental
4.1 Chemistry
Solvents and chemicals were purchased from commercial sources and used without
purification. Melting points were recorded using Stuart SMP3 Digital Melting Point apparatus.
All the reactions were followed up by TLC using silica gel F254 plates (Merck), using
chloroform: methanol 9.5:0.5 or pure chloroform as eluting system and were visualized by UV-
18

ACCEPTED MANUSCRIPT
lamp. All final compounds had a percentage purity of at least 95%, and this was verified using
UPLC-MS/MS conducted on a Waters TQD LC–MS-MS system with a mobile phase consisting
of acetonitrile: 0.1% formic acid in water (95:5, v/v) pumped through an Acquity UPLC BEH
shield RP C18 column (1.7 µm, 2.1 × 150 mm) at a flow rate of 0.25 mL/min. The cone voltage
was set at +40 V and the collision energy was set at 30 V with 0.04s dwell time of all drugs.
Elemental Microanalyses were carried out at the Regional Center for Mycology and
Biotechnology, Al-Azhar University. IR spectra (KBr disc) were recorded on a Shimadzu FT-IR
8400S infrared spectrophotometer. NMR spectra were recorded on a Bruker Ascend 400/R (1H:
400, 13C: 100 MHz) spectrometer. The spectra were run at 400 MHz in deuterated
dimethylsulfoxide (DMSO-d6). Chemical shifts were expressed in δ units relative to TMS signal.
Chemical shifts of exchangeable protons were characterized in D2O solution. All coupling
constant (J) values are given in hertz. 6-(Un)substituted 2-aminobenzothiazoles 2a,b,e,h,j, 5a-c
and 7 were purchased from Sigma-Aldrich, Germany and Alfa Aesar, Great Britain while
2c,d,f,g,i were prepared according to the reported procedures [56–59].
4.1.1 General
procedure
for
the
preparation
of
3-(cyclopropylmethoxy)-4-
(difluoromethoxy)-N-(substituted benzo[d]thiazol-2-yl)benzamide 4a-j.
3-(cyclopropylmethoxy)-4-(difluoromethoxy)benzoic acid 3 (2 mmol., 0.52 gm) was refluxed
with thionyl chloride (10 mL) for 1 hr. Excess thionyl chloride was distilled off and then the
residue was washed with dry benzene followed by distillation (3x5 mL). The residue was
dissolved in dry benzene and the reaction was stirred on hot followed by the addition of a
solution of (un)substituted 2-aminobenzothiazole 2a-j (2 mmol.) in dry benzene (10 mL) and
anhydrous potassium carbonate (2 mmol., 0.28 gm). The reaction mixture was refluxed for 6 hr,
filtered while hot and then concentrated. The precipitate obtained on cooling was filtered and
recrystallized from ethanol.
19

ACCEPTED MANUSCRIPT
4.1.1.1 N-(benzo[d]thiazol-2-yl)-3-(cyclopropylmethoxy)-4-(difluoromethoxy)benzamide 4a.
Obtained from the reaction of 2a with 3, white solid, yield 85 % (0.64 gm), mp 98-99° C.
1HNMR (400 MHz, DMSO-d6
)
δ
12.45 (s, 1H, NH exchanged with D2O), 8.02 (dd,
J
= 2.28,
= 8.04,
7.76, 1H, Ar H), 7.93 (dd, = 1.80, 8.44, 1H, Ar H), 7.76-7.79 (m, 2H, Ar H), 7.47 (dd,
J
J
1.24, 1H, Ar H), 7.34 (m, 2H, Ar H), 7.26 (s, 1H, CHF2), 4.02 (d,
J = 7.00, 2H, CH2), 1.31 (m,
1H, CH), 0.62 (m, 2H, 2CH), 0.41 (m, 2H, 2CH). 13C NMR (100 MHz, DMSO-d6
)
δ 172.48,
165.31, 159.45, 149.91, 148.65, 143.74, 131.93, 129.93, 126.64, 124.16, 122.18, 120.59, 119.44,
116.87, 114.37, 114.30, 73.86, 10.38, 3.52. IR (KBr) υmax\cm-1: 3394 (NH), 1651 (C=O).
Elemental analysis calcd (%) for C19H16F2N2O3S (390.40): C, 58.45; H, 4.13; N, 7.18. Found: C,
58.79; H, 4.21; N, 7.43.
4.1.1.2 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(4-methoxybenzo[d]thiazol-2-
yl)benzamide 4b.
Obtained from the reaction of 2b with 3, buff solid, yield 89 % (0.72 gm), mp 132-133° C.
1HNMR (400 MHz, DMSO-d6
Ar H), 7.78(dd, = 1.8, 8.40, 1H, Ar H), 7.57 (dd, ,
1H, Ar H), 7.29 (d, = 8.40, 1H, Ar H), 7.22 (s, 1H, CHF2), 7.02 (d,
)
δ
13.00 (s, 1H, NH exchanged with D2O), 7.96 (d,
= 1.96, 8.36, 1H, Ar H), 7.35 (d,
= 8.2, 1H, Ar H), 4.01 (d,
J
= 1.8, 1H,
J
J
J
= 8.36,
J
J
,
J
= 7.00, 2H, CH2), 3.94 (s, 3H, CH3), 1.30 (m, 1H, CH), 0.61 (m, 2H, 2 CH), 0.39 (m, 2H, 2
CH). 13C NMR (100 MHz, DMSO-d6
)
δ 166.98, 164.92, 157.66, 152.36, 149.91, 143.75,
138.85, 129.75, 125.17, 120.60, 119.45, 115.19, 114.31, 113.88, 107.92, 73.72, 56.31, 10.38,
3.52. IR (KBr) υmax\cm-1: 3487 (NH) 1670 (C=O). Elemental analysis calcd (%) for
C20H18F2N2O4S (420.43): C, 57.14; H, 4.32; N, 6.66. Found: C, 56.88; H, 4.43; N, 6.87.
20

ACCEPTED MANUSCRIPT
4.1.1.3 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(6-methoxybenzo[d]thiazol-2-
yl)benzamide 4c.
Obtained from the reaction of 2c with 3, yellow solid, yield 45 % (0.37 gm), mp 122-123° C.
1HNMR (500 MHz, DMSO-d6)
8.36, 1H, Ar H), 7.49 (d, = 1.92, 1H, Ar H), 7.32 (d,
Ar H), 7.08 (s, 1H, CHF2), 3.75 (s, 3H, CH3), 7.02 (d,
δ
9.93 (s, 1H, NH exchanged with D2O), 7.52 (dd,
J
= 1.96,
= 1.84, 1H,
= 1.8, 8.2,
J
J
J
= 8.24, 1H, Ar H), 7.30 (d,
J
= 8.16, 1H, Ar H), 6.90 (dd,
J
1H, Ar H), 4.20 (s, 3H, OCH3), 3.86 (d,
J = 6.96, 2H, CH2), 1.00 (m, 1H, CH), 0.33 (m, 2H, 2
CH), 0.12 (m, 2H, 2 CH). 13C NMR (125 MHz, DMSO-d6)
δ
166.95, 161.95, 150.19, 149.88,
145.63, 143.83, 129.11, 125.95, 124.36, 122.76, 120.58, 116.89, 115.96, 115.19, 114.32, 73.59,
55.50, 10.37, 3.48. IR (KBr) υmax\cm-1: 3417 (NH),1670 (C=O). Elemental analysis calcd (%)
for C20H18F2N2O4S (420.43): C, 57.14; H, 4.32; N, 6.66. Found: C, 56.88; H, 4.43; N, 6.87.
4.1.1.4 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(5,6-dimethoxybenzo[d]thiazol-2-
yl)benzamide 4d.
Obtained from the reaction of 2d with 3, yellow solid, yield 82 % (072 gm), mp 155-156
° C.
1HNMR (400 MHz, DMSO-d6
)
δ 9.97 (s, 1H, NH exchanged with D2O), 7.72 (s, 1H, Ar H),
7.62 (d,
J = 1.8, 8.2, 1H, Ar H), 7.40 (s, 1H, Ar H), 7.31 (d, J = 1.8, 1H, Ar H), 7.22 (d, J = 8.2,
1H, Ar H), 7.02 (s, 1H, CHF2), 3.98 (d, ,
J = 6.92, 2H, CH2), 3.81 (s, 3H, OCH3), 3.76 (s, 3H,
OCH3), 1.30 (m, 1H, CH), 0.61 (m, 2H, 2 CH), 0.38 (m, 2H, 2 CH). 13C NMR (100 MHz,
DMSO-d6) δ 164.80, 149.97, 148.64, 148.45, 142.84, 132.44, 129.51, 120.94, 119.56, 116.99,
115.45, 114.42, 114.03, 113.10, 111.59, 73.72, 56.50, 56.31, 10.43, 3.52. IR (KBr) υmax\cm-1:
3259 (NH), 1647 (C=O). Elemental analysis calcd (%) for C21H20F2N2O5S (450.46): C, 55.99; H,
4.48; N, 6.22. Found: C, 56.23; H, 4.62; N, 6.39.
21

ACCEPTED MANUSCRIPT
4.1.1.5 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(6-ethoxybenzo[d]thiazol-2-
yl)benzamide 4e.
Obtained from the reaction of 2e with 3, yellow solid, yield 74 % (0.62 gm), mp 255-256° C.
1HNMR (400 MHz, DMSO-d6
Ar H), 7.75 (dd, = 1.68, 8.44, 1H, Ar H), 7.54 (d,
H), 7.34 (d, = 8.60, 1H, Ar H), 7.28 (s, 1H, CHF2), 7.00 (dd, = 2.20, 8.84, 1H, Ar H), 4.08 (d,
= 6.92, 2H, CH2), 4.02 (q, = 6.84, 2H, CH2CH3), 1.35 (t, J = 6.84, 3H, CH2CH3), 1.32 (m,
)
δ
10.11 (s, 1H, NH exchanged with D2O), 7.91 (d,
J = 1.64, 1H,
J
J
= 2.16, 1H, Ar H), 7.47 (d, = 8.84, 1H, Ar
J
J
J
J
J
13
1H, CH), 0.60 (m, 2H, 2 CH), 0.38 (m, 2H, 2 CH). C NMR (100 MHz, DMSO-d6
)
δ
168.63,
156.00, 149.86, 133.2, 132.28, 128.8, 125.47, 122.00, 120.56, 115.90, 115.61, 115.20, 114.40,
108.53, 105.79, 73.85, 64.27, 15.04, 10.39, 3.55. IR (KBr) υmax\cm-1: 3410 (NH), 1643 (C=O).
Elemental analysis calcd (%) for C21H20F2N2O4S (434.46): C, 58.06; H, 4.64; N, 6.45. Found: C,
56.88; H, 4.43; N, 6.87.
4.1.1.6 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(6-fluorobenzo[d]thiazol-2-
yl)benzamide 4f.
Obtained from the reaction of 2f with 3, white solid, yield 45 % (0.36 gm), mp 155-156°C.
1HNMR (400 MHz, DMSO-d6
)
δ
7.84 (s, 1H, NH exchanged with D2O), 7.44 (d,
J = 1.8, 1H, Ar
H), 7.33 (dd, = 1.88, 8.36, 1H, Ar H), 7.25 (d,
J
J
= 8.2, 1H, Ar H), 7.18 (s, 1H, CHF2), 7.06 (d,
J
= 8.32, 1H, Ar H), 7.00 (d,
J = 8.2, 1H, Ar H), 6.81 (d, J = 8.2, 1H, Ar H), 3.77 (d, J = 6.96,
2H, CH2), 0.98 (m, 1H, CH), 0.44 (m, 2H, 2 CH), 0.22 (m, 2H, 2 CH). 13C NMR (100 MHz,
DMSO-d6) δ 167.34, 149.83, 142.53, 142.50, 142.47, 132.72, 120.70, 120.61, 119.56, 116.99,
114.43, 113.98, 73.61, 10.40, 3.47. IR (KBr) υmax\cm-1: 3371 (NH), 1651 (C=O). Elemental
analysis calcd (%) for C19H15F3N2O3S (408.40): C, 55.88; H, 3.70; N, 686. Found: C, 55.98; H,
3.69; N, 6.98.
22

ACCEPTED MANUSCRIPT
4.1.1.7 N-(6-chlorobenzo[d]thiazol-2-yl)-3-(cyclopropylmethoxy)-4-
(difluoromethoxy)benzamide 4g.
Obtained from the reaction of 2g with 3, yellow solid, yield 35 % (0.29 gm), mp 168-169° C.
1HNMR (400 MHz, DMSO-d6
)
δ
12.90 (s, 1H, NH exchanged with D2O), 7.59 (d,
= 8.18, 1H, Ar H), 7.15 (s, 1H, CHF2), 7.10 (d,
J
= 8.2, 1H,
Ar H), 7.40 (m, 2H, Ar H), 7.22 (d,
J
J = 1.8, 1H,
Ar H), 7.04 (d,
J = 1.8, 1H, Ar H), 3.80 (d, J = 6.88, 2H, CH2), 1.08 (m, 1H, CH), 0.42 (m, 2H, 2
CH), 0.20 (m, 2H, 2 CH). 13C NMR (100 MHz, DMSO-d6
)
δ 166.98, 161.96, 149.88, 145.63,
143.84, 129.10, 125.96, 124.37, 122.77, 120.59, 119.47, 116.72, 115.97, 115.19, 114.14, 73.59,
10.38, 3.54. IR (KBr) υmax\cm-1: 3417 (NH), 1680 (C=O). Elemental analysis calcd (%) for
C19H15ClF2N2O3S (424.85): C, 53.72; H, 3.56; N, 6.59. Found: C, 53.98; H, 3.67; N, 6.89.
4.1.1.8 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(6-(trifluoromethyl)benzo[d]thiazol-
2-yl)benzamide 4h.
Obtained from the reaction of 2h with 3, white solid, yield 72 % (0.64 gm), mp 160-161°C.
1HNMR (400 MHz, DMSO-d6
) δ 10.34 (s, 1H, NH exchanged with D2O), 7.74 (m, 1H, Ar H),
7.54 (m, 1H, Ar H), 7.39 (s, 1H, CHF2), 7.26 (d,
J = 8.16, 1H, Ar H), 7.19 (d, J = 8.48, 1H, Ar
H), 3.93 (d,
J
= 6.92, 2H, CH2), 1.23 (m, 1H, CH), 0.55 (m, 2H, 2 CH), 0.36 (m, 2H, 2 CH). 13C
NMR (100 MHz, DMSO-d6) δ 170.53, 166.92, 149.86, 143.82, 133.98, 130.29, 129.11, 128.75,
126.10, 124.84, 122.75, 120.56, 119.44, 116.88, 115.18, 114.31, 73.58, 10.36, 3.48. IR (KBr)
υmax\cm-1: 3294 (NH), 1654 (C=O). Elemental analysis calcd (%) for C20H15F5N2O3S (458.40):
C, 52.40; H, 3.30; N, 6.11. Found: C, 552.67; H, 3.67; N, 6.39.
4.1.1.9 N-(6-cyanobenzo[d]thiazol-2-yl)-3-(cyclopropylmethoxy)-4-
(difluoromethoxy)benzamide 4i.
Obtained from the reaction of 2i with 3, yellow solid, yield 55 % (0.44 gm), mp 140-141°C.
1HNMR (400 MHz, DMSO-d6
) δ 10.06 (s, 1H, NH exchanged with D2O), 7.50 (d, J = 1.88, 1H,
23

ACCEPTED MANUSCRIPT
Ar H), 7.33 (m, 2H, Ar H), 7.16 (dd,
J
= 1.92, 8.2, 1H, Ar H), 7.07 (s, 1H, CHF2), 7.02 (d, J =
1.9, 1H, Ar H), 6.93 (d,
J
= 8.2, 1H, Ar H), 3.75 (d, ,
J
= 6.96, 2H, CH2), 1.02 (m, 1H, CH), 0.34
166.99, 161.92, 150.19,
(m, 2H, 2 CH), 0.13 (m, 2H, 2 CH). 13C NMR (100 MHz, DMSO-d6
) δ
148.80, 145.66, 143.84, 142.64, 130.29, 129.12, 125.97, 124.33, 122.76, 121.08, 120.58, 116.83,
115.19, 73.59, 10.28, 3.50. IR (KBr) υmax\cm-1: 3377 (NH), 2233 (C
≡N), 1671 (C=O). Elemental
analysis calcd (%) for C20H15F2N3O3S (415.41): C, 57.83; H, 3.64; N, 10.12. Found: C, 58.07; H,
3.72; N, 10.34.
4.1.1.10 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(6-nitrobenzo[d]thiazol-2-
yl)benzamide 4j.
Obtained from the reaction of 2j with 3, yellow solid, yield 76 % (0.64 gm), mp 144-145
1HNMR (400 MHz, DMSO-d6
°C.
)
δ
9.62 (s, 1H, NH exchanged with D2O), 8.80 (d,
J
= 2.44, 1H,
Ar H), 8.27 (dd,
J = 2.4, 8.92, 1H, Ar H), 8.19 (d, J = 8.88, 1H, Ar H), 7.91 (dd, J = 1.88, 8.92,
1H, Ar H), 7.76 (d,
J = 1.84, 1H, Ar H), 7.57 (d, J = 8.92, 1H, Ar H), 7.53 (s, 1H, CHF2), 4.03
(d, ,
J
= 7, 2H, CH2), 1.19 (m, 1H, CH), 0.59 (m, 2H, 2CH), 0.37 (m, 2H, 2CH).13C NMR (100
MHz, DMSO-d6) δ 171.78, 164.81, 150.44, 149.86, 142.69, 132.68, 129.27, 128.61, 123.23,
122.74, 120.89, 119.31, 116.83, 115.77, 114.49, 73.89, 10.37, 3.55. IR (KBr) υmax\cm-1: 3282
(NH), 1651 (C=O), 1527 & 1342 (NO2). Elemental analysis calcd (%) for C19H15F2N3O5S
(435.40): C, 52.41; H, 3.47; N, 9.65. Found: C, 52.68; H, 3.60; N, 9.84.
4.1.1.11 N-(benzo[d]thiazol-6-yl)-3-(cyclopropylmethoxy)-4-(difluoromethoxy)benzamide 6a.
Obtained from the reaction of 5a with 3, white solid, yield 85 % (0.64 gm), mp 95-96°C.
1HNMR (400 MHz, DMSO-d6
)
δ
10.48 (s, 1H, NH exchanged with D2O), 9.31 (s, 1H, Ar H),
= 1.88, 1H, Ar H), 8.08 (d, = 8.80, 1H, Ar H), 7.82 (dd, = 1.90, 8.84, 1H, Ar H),
= 1.8, 1H, Ar H), 7.63 (dd, = 1.8, 8.36, 1H, Ar H), 7.34 (d, = 8.32, 1H, Ar H),7.23
8.65 (d,
7.69 (d,
J
J
J
J
J
J
24

ACCEPTED MANUSCRIPT
(s, 1H, CHF2), 4.01 (d,
J
= 6.96, 2H, CH2), 1.30 (m, 1H, CH), 0.61 (m, 2H, 2CH), 0.39 (m, 2H,
2CH). 13C NMR (100 MHz, DMSO-d6
) δ 165.15, 155.49, 149.99, 142.85, 137.15, 134.55,
133.14, 123.29, 121.04, 120.78, 120.55, 119.53, 116.97, 114.23, 113.65, 73.78, 10.43, 3.52. IR
(KBr) υmax\cm-1: 3271 (NH), 1651 (C=O). Elemental analysis calcd (%) for C19H16F2N2O3S
(390.40): C, 58.45; H, 4.13; N, 7.18. Found: C, 58.68; H, 4.03; N, 7.47.
4.1.1.12 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(2-methylbenzo[d]thiazol-6-
yl)benzamide 6b.
Obtained from the reaction of 5b with 3, brown solid, yield 72 % (0.56 gm), mp 122-123
1HNMR (400 MHz, DMSO-d6
°C.
)
δ
10.42 (s, 1H, NH exchanged with D2O), 8.50 (d,
J
= 1.98, 1H,
Ar H), 7.89 (d, J = 8.80, 1H, Ar H), 7.75 (dd, J = 2.00, 8.76, 1H, Ar H), 7.69 (d, J = 1.88, 1H, Ar
H), 7.63 (dd,
J = 1.88, 8.32, 1H, Ar H), 7.34 (d, J = 8.3, 1H, Ar H),7.22 (s, 1H, CHF2), 4.00 (d, J
= 7, 2H, CH2), 2.78 (s, 3H, CH3), 1.28 (m, 1H, CH), 0.61 (m, 2H, 2CH), 0.39 (m, 2H, 2CH). 13C
NMR (100 MHz, DMSO-d6) δ 166.35, 165.05, 149.97, 149.88, 142.80, 136.45, 136.06, 133.18,
122.17, 121.01, 120.77, 119.53, 116.96, 114.19, 113.43, 73.77, 20.12, 10.43, 3.52. IR (KBr)
υmax\cm-1: 3298 (NH), 1670 (C=O). Elemental analysis calcd (%) for C20H18F2N2O3S (404.43):
C, 59.40; H, 4.49; N, 6.93. Found: C, 59.23; H, 4.68; N, 7.08.
4.1.1.13 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(2-mercaptobenzo[d]thiazol-6-
yl)benzamide 6c.
Obtained from the reaction of 5c with 3, yellow solid, yield 82 % (0.67 gm), mp 210-211
°C.
1HNMR (400 MHz, DMSO-d6
)
δ 12.15 (s, 1H, SH exchanged with D2O), 10.53 (s, 1H, NH
exchanged with D2O), 7.83 (d,
J = 1.62, 1H, Ar H), 7.65 (m, 2H, Ar H), 7.33 (3, 2H, Ar H), 7.2
(m, 2H, ArH, CHF2), 4.00 (d,
J = 6.2, 2H, CH2), 1.28 (m, 1H, CH), 0.58 (m, 2H, 2CH), 0.37 (m,
2H, 2CH). 13C NMR (100 MHz, DMSO-d6
)
δ 190.22, 164.91, 150.05, 149.70, 139.78, 136.2,
25

ACCEPTED MANUSCRIPT
132.96, 131.77, 131.06, 123.51, 121.4, 117.07, 115.12, 114.24, 113.89, 113.61, 73.85, 10.43,
3.53. IR (KBr) υmax\cm-1: 3367 (NH), 2580 (SH), 1690 (C=O). Elemental analysis calcd (%) for
C19H16F2N2O3S2 (422.46): C, 54.02; H, 3.82; N, 6.63. Found: C, 53.89; H, 3.63; N, 6.92.
4.1.1.14 3-(cyclopropylmethoxy)-4-(difluoromethoxy)-N-(thiazol-2-yl)benzamide 8.
°
Obtained from the reaction of 7 with 3, white solid, yield 55 % (0.36 gm), mp 97-98 C.
1HNMR (400 MHz, DMSO-d6
)
δ
12.85 (s, 1H, NH exchanged with D2O), 7.73 (d,
J = 1.2, 1H,
Ar H), 7.54 (m, 2H, Ar H), 7.37 (dd,
J
= 4.8 1H, Ar H), 7.23 (d, = 8.12, 1H, Ar H), 7.18 (s, 1H,
J
CHF2), 3.91 (d, ,
J = 6.88, 2H, CH2), 1.22 (m, 1H, CH), 0.55 (m, 2H, 2 CH), 0.34 (m, 2H, 2 CH).
13C NMR (100 MHz, DMSO-d6
)
δ 166.87, 161.84, 149.83, 143.79, 133.89, 129.02, 122.73,
120.55, 119.39, 116.75, 115.16, 73.67, 10.24, 3.46. IR (KBr) υmax\cm-1: 3336 (NH), 1697 (C=O).
Elemental analysis calcd (%) for C15H14F2N2O3S (340.34): C, 52.94; H, 4.15; N, 8.23. Found: C,
53.08; H, 4.34; N, 8.52.
4.2 Biological evaluation.
PDE-4B enzymatic assay was performed for roflumilast and all the synthesized compounds
using Human PDE4B ELISA Kit (MBS760982), BPS Bioscience®. PDE-4D enzymatic assay
was performed for roflumilast, 4a, 6a and 6c using Human PDE4D) ELISA Kit (MBS732623),
BPS Bioscience®. Intracellular cAMP was measured for roflumilast, 4a, and 6c using ab133038-
Direct cAMP ELISA Kit, Abcam®. The quantification of TNF-α was performed for roflumilast,
4a, and 6c using ab181421 –TNF-alpha Human SimpleStep® ELISA Kit obtained from
Abcam®. Manufacturer’s instructions were followed for each kit. Statistical significance
between the obtained biological values of roflumilast and the tested compounds was calculated
using one-way ANOVA followed by Bonferroni's multiple comparisons test (p<0.05). The tested
compounds were examined for known classes of assay interference compounds. According to
26

ACCEPTED MANUSCRIPT
SwissADME tool [69], none of the compounds contain substructural features recognized as pan
assay interference compounds (PAINS).
4.3 Molecular docking study
Molecular docking studies were carried out using Molecular Operating Environment (MOE,
10.2008) software. Minimization was performed with MOE until an RMSD gradient of 0.05
kcal·
mol−1Å−1 with MMFF94x force field and the partial charges were automatically calculated.
The X-ray crystallographic structures of PDE-4B and PDE-4D co-crystallized with roflumilast as
inhibitor were downloaded from the Protein Data Bank, https://www.rcsb.org/, (PDB ID: 1XMU
and 1XOQ, respectively) [42]. Preparation of the enzymes for docking study was done by
removing chain B of its dimer, ligands and water molecules that are not involved in the binding.
The enzymes were then protonated using Protonate 3D protocol in MOE with default options.
Triangle Matcher placement method and London dG scoring function were used for the docking
protocol and forcefield was used as refinement. The co-crystallized ligand (roflumilast) was re-
docked into the active site of each enzyme for validation of the docking setup. Then, the binding
mode and the binding interactions of the newly synthesized compounds at the active site of PDE-
4B and PDE-4D were predicted by using the validated setup. Roflumilast was used as reference
compound to compare its binding score to that of the newly synthesized compounds and so their
affinity to the target enzymes (PDE4B and PDE4D).
4.4 Determination of in vivo pharmacokinetic parameters in rats
Twenty-four male Sprague-Dawley rats (250-300 g), obtained from the Laboratory Animal
Center, Faculty of Pharmacy, Cairo University, were placed in cages, six per cage and had free
access to food and water. The temperature was set at 25˚C and maintained in a 12 h light/dark
cycle. The rats were randomly divided into four groups (n=6): reference group (roflumilast) and
27

ACCEPTED MANUSCRIPT
3 test groups (compounds 4a, 6a and 6c]. All experimental procedures were approved by and
conducted in accordance with the institutional guidelines of the Institutional Review Board of
Faculty of Pharmacy, Cairo University and the regulations of the local Animal Welfare
authorities. Each group was dosed with a subcutaneous dose (1 mg/kg). The compounds were
formulated in 5% tween 80 and normal saline. After single administration, blood samples were
withdrawn at the following time intervals: 0.25, 0.5, 1, 2, 4, 6, 8 and 24 h. All blood samples
were collected into heparinized tubes, centrifuged at 4000 rpm for 10 min at 4oC and
subsequently stored at -20°C prior to analysis. Plasma samples were extracted then analyzed by a
validated UPLC-MS/MS method for each individual compound. The pharmacokinetic
parameters were calculated using WinNonlin Software. Statistical significance between the
pharmacokinetic parameters (t1/2, Cmax, AUC0-t, AUC0- , Vd and CL) of roflumilast and the tested
∞
compounds was calculated using one-way ANOVA followed by Bonferroni's multiple
comparisons test (p<0.05).
4.5 In vitro pharmacokinetic studies
4.5.1 Metabolic stability and metabolite identification in rat S9 fraction
The assay was performed with Liver subcellular fraction (9,000× g S9 fraction) from male
Wistar rats, prepared as previously described [70,71]. All steps were performed at 0–4°C with
cold and sterile solutions and glassware. Compounds 4a, 6a and 6c in addition to roflumilast
were tested for their phase I metabolic stability. Twenty µM of each of the tested compounds
was incubated with 1 mg/mL prepared pooled mammalian liver S9 fraction, 10 mM NADPH,
o
100 mM MgCl2 and 100 mM Tris buffer (pH 7.4) 37 C for 0, 15, 30 and 60 min at a final
volume of 0.5 mL (except for roflumilast which was incubated for time intervals 0, 5, 10, 15 and
30 min.). Reactions were terminated by protein precipitation using 0.5 mL cold acetonitrile and
28