Novel LCK/FMS inhibitors based on phenoxypyrimidine scaffold as potential treatment for inflammatory disorders

Article abstract of DOI:10.1016/j.ejmech.2017.10.003

Tyrosine kinases including LCK and FMS are involved in inflammatory disorders as well as many types of cancer. Our team has designed and synthesized thirty novel pyrimidine based inhibitors targeting LCK, classified into four different series (amides, ureas, imines (Schiff base) and benzylamines). Twelve of them showed nanomolar IC₅₀ values. Compound 7g showed excellent selectivity profile and was selectively potent over FMS kinase (IC₅₀ value of 4.6 nM). Molecular docking study was performed to help us rationalize the obtained results and predict the possible binding mode for our compounds in both LCK and FMS. Based on the obtained biological assay data and modelling results, a detailed SAR study was discussed. As a further testing regarding the anti-inflammatory effect of the new compounds, in vitro cellular assay over RAW 264.7 macrophages was performed. Compound 7g exhibited excellent anti-inflammatory effect. Therefore, we report the design of novel phenoxypyrimidine derivatives as potent and selective LCK inhibitors and the discovery of 7g as potent and selective FMS/LCK dual inhibitor for the potential application in inflammatory disorders including rheumatoid arthritis (RA).

Full text of DOI:10.1016/j.ejmech.2017.10.003

Accepted Manuscript
Novel LCK/FMS inhibitors based on phenoxypyrimidine scaffold as potential
treatment for inflammatory disorders
Ahmed Karam Farag, Ahmed Elkamhawy, Ashwini M. Londhe, Kyung-Tae Lee, Ae
Nim Pae, Eun Joo Roh
PII:
S0223-5234(17)30795-X
10.1016/j.ejmech.2017.10.003
EJMECH 9795
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 20 January 2017
Revised Date: 19 May 2017
Accepted Date: 2 October 2017
Please cite this article as: A.K. Farag, A. Elkamhawy, A.M. Londhe, K.-T. Lee, A.N. Pae, E.J. Roh,
Novel LCK/FMS inhibitors based on phenoxypyrimidine scaffold as potential treatment for inflammatory
disorders, European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.10.003.
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

ACCEPTED MANUSCRIPT
Novel LCK/FMS inhibitors based on phenoxypyrimidine scaffold as
potential treatment for inflammatory disorders
a,b
a,c
b,d
e
Ahmed Karam Farag , Ahmed Elkamhawy , Ashwini M. Londhe , Kyung-Tae Lee , Ae
b,d
a,b*
Nim Pae , Eun Joo Roh
a
Chemical Kinomics Research Center, Korea Institute of Science and Technology (KIST), Seoul 02792, Republic of Korea.
Division of Bio-Medical Science &Technology, KIST School, Korea University of Science and Technology, Seoul 02792,
b
Republic of Korea.
c
Department of Pharmaceutical Organic Chemistry, Faculty of Pharmacy, Mansoura University, Mansoura 35516, Egypt.
d
Convergence Research Center for Diagnosis, Treatment and Care System of Dementia, Korea Institute of Science and
Technology (KIST), Seoul 02792, Republic of Korea.
e
Department of Life and Nanopharmaceutical Science, Kyung Hee University, Seoul 130-701, Republic of Korea.
*
Corresponding author: Eun Joo Roh
Email address: r8636@kist.re.kr
Postal address: Korea Institute of Science and Technology, Chemical Kinomics Research Center,
Future Convergence Research Division, Hwarang-ro 14-gil 5, Seongbuk-gu, Seoul 02792,
Republic of Korea.

ACCEPTED MANUSCRIPT
Abstract
Tyrosine kinases including LCK and FMS are involved in inflammatory disorders
as well as many types of cancer. Our team has designed and synthesized thirty
novel pyrimidine based inhibitors targeting LCK, classified into four different
series (amides, ureas, imines (Schiff base) and benzylamines). Twelve of them
showed nanomolar IC values. Compound 7g showed excellent selectivity profile
5
0
and was selectively potent over FMS kinase (IC value of 4.6 nM). Molecular
5
0
docking study was performed to help us rationalize the obtained results and predict
the possible binding mode for our compounds in both LCK and FMS. Based on the
obtained biological assay data and modelling results, a detailed SAR study was
discussed. As a further testing regarding the anti-inflammatory effect of the new
compounds, in vitro cellular assay over RAW 264.7 macrophages was performed.
Compound 7g exhibited excellent anti-inflammatory effect. Therefore, we report
the design of novel phenoxypyrimidine derivatives as potent and selective LCK
inhibitors and the discovery of 7g as potent and selective FMS/ LCK dual inhibitor
for the potential application in inflammatory disorders including rheumatoid
arthritis (RA).
Keywords

ACCEPTED MANUSCRIPT
LCK, FMS, Inflammation, Synthesis, molecular docking and RAW 264.7
macrophages.
1
. Introduction
Inflammatory diseases have been the concern of many research groups and
pharmaceutical companies over the past decades. Inflammation process results in
several disorders including multiple sclerosis, RA, inflammatory bowel disease
(IBD) and psoriasis [1]. The classical treatment strategy for these diseases depends
on the use of NSAIDs and corticosteroids. Despite the success that these old
strategies had, the risk of resistance, unaffordable side effects and cost-of-
treatment make new approaches in this area worthy to be considered [2]. The
growing understanding of cellular signaling cascades and immune system offers
new therapeutic avenues to a wide diversity of inflammatory disorders including
RA.
Protein kinases covalently modify proteins by phosphorylating specific residues,
thus modulating their activity. They are currently the most extensively explored
drug targets in cancer research for its importance in regulating several cellular
signaling pathways which control cellular proliferation, migration, apoptosis,
differentiation and metabolism [3, 4]. Protein tyrosine kinases (PTK) are
considered as important molecular targets on the way of treatment of inflammatory

ACCEPTED MANUSCRIPT
and autoimmune diseases [5]. To date thirty small molecule inhibitors have been
approved by the FDA for treatment of cancer and inflammation disorders with two
drugs in 2016; namely Cabozantinib and lenvatinib. Amongst these thirty approved
inhibitors, only two drugs accounts for inflammatory disorders; namely tofacitinib
[6] (JAK-3 selective inhibitor, approved in 2012) and nintedanib [7, 8] (multi-
kinase inhibitor, approved in 2014) for treatment of RA and idiopathic pulmonary
fibrosis respectively. Even though many other drug candidates are currently in
clinical trials for inflammatory disorders, the needs from this research area were
not yet met [9].
Since the first discovery of JAK-STAT signaling pathway and its relation to
inflammatory cytokines in the 1990s, the search for potential PTK targets for
treatment of inflammatory diseases was growing [10, 11]. Numerous kinases are
now considered as approved targets for treatment of RA such as CSF1R, LCK, Btk,
Syk and JAK3 [12]. Many research groups proved that inhibition of CSF1R can
slow the progression of amyotrophic lateral sclerosis [13], prevent erosions and
reduce symptoms of RA and bone osteolysis [14]. Others proved the use of LCK
inhibitors in treatment of T cell-mediated disorders, such as organ transplant
rejection [15, 16], RA or psoriasis [17], while others proved the usefulness of the
dual target inhibition of FMS and KIT in preclinical disease models for
inflammation and cancer [18].

ACCEPTED MANUSCRIPT
Lymphocyte-specific kinase (LCK) is a 56 kDa protein, a member of the src family
of cytoplasmic tyrosine kinases that include 8 more kinases namely; Blk, Fgr, Fyn,
HCK, Lyn, Src, Yes, Yrk [19]. It’s expressed exclusively in T cells and Natural
Killer (NK) cells and it has been proved, through genetic evidences in both mice
and human, that LCK is crucial for T cell receptor (TCR) mediated signaling
required for T cell development and activation [20-24]. Different potent and
selective LCK inhibitors have been developed recently [25-28] and some
representative examples are illustrated in Figure 1.
(Please insert Figure 1 here)
Colony-stimulating factor receptor (CSF1R) (namely; FMS) is exclusively
expressed in monocyte-macrophages and belongs to the PDGFR family of type III
receptor tyrosine kinase (RTK) which includes FLT-3, KIT, PDGFR
α and β and is
the exclusive receptor for IL-34 and CSF1. Binding of CSF1 to the extracellular
domain of CSF1R leads to receptor dimerization and activation of the intracellular
FMS kinase domain resulting in phosphorylation and activation of several
intracellular signaling molecules ultimately resulting in the survival, proliferation
and differentiation of the target cells [29]. Experiments using M-CSF blocking
antibodies showed that CSF1-R aggravates collagen-induced arthritis in mice and

ACCEPTED MANUSCRIPT
CSF1-R knockout mice were more resistant to this. These experiments and others
[30-33] proved that FMS is a suitable target for inflammatory diseases caused by
macrophages. Examples of known FMS inhibitors [34-36] are depicted in Figure 2.
(Please insert Figure 2 here)
1
.1.Rational design.
The rational design of the newly synthesized compounds is illustrated in Figure 3.
compound V; a potent LCK inhibitor of Roche pharmaceuticals [37, 38], was taken
as a lead compound.
(Please insert Figure 3 here)
6
Compound V is believed to be type I inhibitor with the 7-anilino NH and N of the
pyrimidopyrimidine core directed to the adenine-binding pocket as hinge contactor.
The side chain on position 7 is to project out of the receptor as solvent exposure
3
allowing possible variations to this moiety. The 2,6-dichlorophenyl on N is
believed to be directed toward the hydrophobic pocket adjacent to the ATP-binding
site.

ACCEPTED MANUSCRIPT
Guided by the structural features and binding mode of compound V, we adopted a
non-classical bioisosterism strategy through opening of the saturated pyrimidine
ring of the parent pyrimido[4,5-d]pyrimidine core to get a pyrimidine based type I
inhibitors. The pyrimidine was substituted on position 2 with different solvent
exposure moieties and on position 4 with different phenoxy moieties to explore the
effect of substituting this position on binding affinity and selectivity; and on
position 5 with different substituted phenyl groups connected through a linker
which is amide, urea, aminomethylene or aminomethyl groups.
Based on previous findings, our objective was to design and synthesize a novel
series of 4-phenoxypyrimidne-2-amine derivatives that could inhibit LCK and/or
FMS kinases with potential activity against inflammation disorders. Consequently,
herein we report the design, synthesis, molecular docking study and SAR analysis
of a series of 4-phenoxypyrimidine-2-amine derivatives and the success to obtain
selectivity for LCK over other Src family members represented by 7j and the
discovery of 7g as dual selective LCK/FMS inhibitor with anti-inflammatory
activity represented by its ability to inhibit FMS and LCK in vitro with low
nanomolar IC in the enzymatic assay while in the cell-based assay it was able to
5
0
reduce the production of cytokines including TNF-
α, IL-6 and PGE as well as
2
inhibit the expression of iNOS, TNF- and IL-6 mRNA.
α

ACCEPTED MANUSCRIPT
2
. Results and discussion
.1.Chemistry.
2
The targeted phenoxypyrimidine-like derivatives were prepared as represented in
Scheme 1. The intermediates 4a–f were synthesized in three steps from the
commercially available 2,4-dichloro-5-nitropyrimidine which was substituted at
position 4 with different phenols by S Ar reaction in acetone under basic condition
N
of aqueous sodium bicarbonate solution as reported in literature [39, 40] to afford
2
a–e. Refluxing compounds 2a–e with different anilines; namely 3,5-
dimethoxyaniline and 4-morpholinoaniline in THF and pyridine afforded the novel
nitro derivatives 3a–f in a good yield as yellow to deep red solids [41]. Compounds
3
a–f were reduced to the corresponding amines by catalytic hydrogenation in
methanol and DCM at room temperature using Pd/C as a catalyst to obtain 4a–f.
(Please insert Scheme 1 here)
Compounds 4a–f were then refluxed with: (i) acid chlorides to afford the amides
a–b; (ii) isocyanates in DCM to afford the corresponding ureas 6a–k; (iii)
aldehydes in dry isopropanol (IPA) under anhydrous conditions and under N to
5
2

ACCEPTED MANUSCRIPT
afford the Schiff bases 7a–m; (iv) aldehyde in dry IPA then NaBH CN to afford
3
the benzylamines 8a–d.
Several reported solvents and conditions were explored for the preparation of 7a–
m and consequently 8a–d, such as MeOH, DCM, toluene and neat reaction with
IPA being the best option with the highest yield and to complete reaction in our
case [42-45]. 7g was taken as illustrative example to confirm the E-configuration
of the Schiff bases 7a–m. Based on the 3D shape of the imines, we hypothesized
that the pyrimidine proton and the imine proton will be close to each other in space,
if the imine attain the E-configuration. To confirm that, COSY and NOESY 2D
NMR experiments were carried out. The cross peak between the aforementioned
protons was observed (see supporting material for the NMR charts). Together
with the well-known energy difference between the E-isomers and the Z-isomers
and the potential steric clash between the pyrimidine proton and the ortho-proton
on the phenyl ring bearing R substituents, confirms the E-configuration of the
3
1
imines 7a–m. All the intermediates and final compounds were characterized by
1
3
HNMR, CNMR and/or HRMS. Previously reported intermediates (2b and 2c)
1
[
39, 40, 46] were assigned by HNMR only (see Experimental section and
supporting material).
2
.2.In vitro enzymatic assay.

ACCEPTED MANUSCRIPT
Compound 6b (available in our kinase library) was screened over a panel of 53
kinases as a proof-of-design and showed high selectivity for Src family members
including c-Src, Lyn and LCK as depicted in Figure 4 (details can be found in
supporting material).
(Please insert Figure 4 here)
Therefore it was tested in IC mode over these 3 kinases (LCK, Lyn and c-Src)
5
0
and found to be 8-fold more potent for LCK than Lyn and 16-fold more potent than
c-Src as shown in Table 1. Therefore our objective was to improve the potency and
selectivity for LCK by adopting medicinal chemistry strategies.
(
Please insert Table 1 here)
After synthesis of Twenty-nine more derivatives, they were assayed for LCK
inhibitory activity and the results are summarized in Table 2 (assay conditions,
protocols and materials can be found in the supporting material).
(
Please insert Table 2 here)

ACCEPTED MANUSCRIPT
2
.3.Selectivity profile.
To evaluate the selectivity of our compounds, the most potent six derivatives
belonging to the Schiff base series (7e, 7g, 7h, 7j, 7k and 7l), were assayed as %
inhibition at 10 uM against selected kinase panel (Figure 5). All of derivatives
showed no inhibition against KDR, P38α and Tie2. Compound 7j and 7k showed
very good selectivity against the closely related Src family members (c-Src, Fyn
and Lyn) as well as Abl-1, c-FMS and showed very weak inhibition against JAK3.
Compound 7g showed excellent selectivity against all the tested kinases with only
strong inhibitory activity against c-FMS. On the other hand, 7h and 7l were the
least selective among the tested compounds and both of them showed high potency
over JAK3.
(Please insert Figure 5 here)
Since 7g is the only derivative that has the 3,5-dimethoxybenzylideneamino moiety
on position 5 amongst the tested compounds, we hypothesized that this moiety may
be responsible for the high FMS potency. Therefore 7b, 7c, 7f, 7g and 7i were

ACCEPTED MANUSCRIPT
tested for IC over FMS kinase. 7c, 7f, 7g and 7i showed low nanomolar activity
5
0
with 7g being the most potent with 4.6 nM IC value (Table 3).
5
0
(
Please insert Table 3 here)
2
.4.SAR analysis for LCK and FMS.
Our SAR study is based on investigating the substitution pattern on the pyrimidine
core on positions 2, 4 and 5 as illustrated on compound 6b in Figure 6.
(Please insert Figure 6 here)
The 2-aminopyrimidine core in our design was conserved for the essential hinge
anchorage. On position 2, the 3,5-dimethoxyaniline moiety was tested as a solvent
exposure in the amide series (5a) and urea series (6a and 6d), but wasn’t tolerated
(
5a vs 5b and 6a vs 6b) so its use as solvent exposure moiety was discontinued.
Learning the lesson from A-770041 of Abbott, the solvent exposure moiety should
be devoid from a basic amine that’s capable of forming a charge-reinforced
hydrogen bonding with Asp326 to be able to obtain selectivity amongst other Src
family members especially c-Src and Fyn. Based on that, the 4-morpholinoaniline

ACCEPTED MANUSCRIPT
moiety was used and showed generally better potency (5b vs 5a) so was conserved
in the Schiff base derivatives and the benzylamines.
All of our derivatives are based on a substituted phenoxy group on position 4. For
LCK, the 2-methoxy-4-methyl substituents generally showed optimum activity
over other tested substitutions. Introduction of 4-methoxy moiety as in 7g lead to
increased potency in comparison to 7b by 4-folds. However when 3-methoxy
moiety was introduced, as in 7c, the activity was less than 7b by 17-folds.
Introduction of 4-fluoro or 4-trifluoromethyl moieties, as in 7i and 7f respectively,
significantly compromised the activity. However for FMS, the 2-methoxy-4-
methyl group, represented by 7b wasn’t tolerated. With exception of 7b all the
other tested compounds namely; 7c, 7f, 7g and 7i, showed high potency. The
introduction of a hydrophobic group capable of Hydrogen bond (HB) formation on
position 4 was well tolerated in FMS than in LCK (7i and 7f). Therefore we
concluded that fine tuning of the substitution on position 4 of the pyrimidine ring
had a significant role in changing the activity for both LCK and FMS kinases.
On position 5, we tested the insertion of 4 linkers resulting in amides (5a–b), ureas
(
6a–k), Schiff bases (7a–m) or benzylamines (8a–d).
Amides were synthesized to mimic the 2 atoms linker in compound V, but the
activity was unsatisfactory. Urea derivatives were synthesized and generally
2
showed better activity (5b vs 6b). The Schiff bases were prepared to attain the sp

ACCEPTED MANUSCRIPT
configuration of the linker without having the carbonyl O atom (cf. amides) which
showed to be effective strategy in our case. In an attempt to improve the
physicochemical properties of the Schiff base derivatives, the azomethine group
was saturated to result in benzylamine derivatives (8a–d). However saturation of
this double bond resulted in dramatic decrease in LCK activity as 8a was less
active than 7m by > 1500-fold while 8d was 5-fold less active than 7h.
In the urea series, substituting the phenyl ring with hydrophobic groups at 3’,5’-
positions was not tolerated (6f and 6k), however substituting this position with
3
’,5’-dimethoxy was tolerated (6b and 6h) albeit with low potency. 2’,6’-dichloro
substituents (6e) and 2’-methoxy (6g) were still tolerated by the receptor.
In the Schiff base series, hydrophobic substituents were favored over hydrophilic
ones (7j vs 7b). Mono-substitution on positions 3’ or 4’ with hydrophobic
substituents resulted in 2-digit nanomolar active compounds (7e, 7h and 7k) while
disubstitution resulted in single digit nanomolar inhibitors (7j and 7m).
Substitution on position 2’ showed less activity (7d vs 7e or 7h). It should be noted
that 7g has 3’,5’-dimethoxy substitution on the phenyl ring yet shows 2-digit
nanomolar activity probably because of having different phenoxy part on position
4
of the pyrimidine ring. It was noted that the Schiff base derivatives (7a–m)
showed the most potent activity in our series with 7j and 7m being the most potent
LCK inhibitors.

ACCEPTED MANUSCRIPT
In the benzylamine series, the saturation of the N–C bond aimed at conferring
flexibility to this position and it was observed that position 4’ was the best for
substitution with small hydrophobic group (8b vs 8c and 8d vs 8c).
In summary, Schiff bases were the most potent derivatives among the newly
synthesized compounds. Compounds 7j and 7m were the most potent LCK
inhibitors in our series. Compound 7g had the highest potency for FMS, showed
high potency for LCK and had excellent selectivity profile.
2
.5.Molecular docking study.
In the pursuit of understanding the binding mode of our potent compounds with
LCK and FMS kinases, we performed a molecular docking study using
Schrodinger software suit (maestro) [47].
First of all, 7j; the most potent LCK inhibitor in our series, was docked in the
active crystal structure of LCK kinase domain. By analysis of the obtained docking
poses, we found that 7j showed 2 possible binding modes: one normal mode that
fits our design and the other is a flipped mode where the morpholino part in buried
deep inside the receptor and held in that position by a HB contact with Lys273 (see
3
supporting information). As expected from our design, the pyrimidine N together
with the anilino NH were able to make a pair of HB anchorage to the hinge
Met319 (Figure 7A). In the normal binding pose, the 3’,5’-dichlorophenyl was

ACCEPTED MANUSCRIPT
buried deep inside the hydrophobic pocket near the ATP binding site and
contributed into several van der waal interactions explaining the high potency of 7j.
Since the normal binding pose accounts for a better docking score of -7.707
compared to -7.298 for the flipped mode, we believe that this mode is more close
to the real binding mode and explains the enzyme assay data more clearly. It’s also
worth mentioning that the pyrimidine core formed a pi-pi stacking interaction with
Tyr318 in the hinge region (Figure 7B).
(Please insert Figure 7 here)
On the other hand, 7g showed a flipped binding mode with the 3’,5’-
dimethoxyphenyl pointing outside of the receptor, yet it was able to make a HB to
the hinge Met319 along with several van der waal interactions with Tyr318,
Val295 and Leu371 and HB with Lys273 probably contributed to its potency
(Figure 8).
(Please insert Figure 8 here)

ACCEPTED MANUSCRIPT
In an attempt to rationalize the high potency of our FMS inhibitors, Compound 7g
was docked in the active crystal structure of FMS kinase domain. In all the docking
poses, 7g showed a flipped binding mode with the morpholino moiety inserted
deep inside the binding pocket and projects into the solvent area interacting with
3
Leu558 with the GLIDE docking score -6.000. Pyrimidine N and the aniline NH
anchored the hinge part with pair of HB with residue TYR665. Another HB was
observed between oxygen of 4-methoxy on the phenoxy ring and Lys612 located
on β3 close to the hinge region of FMS kinase (Figure 9A and 9B). Similarly, 7b
was docked into FMS and showed a lower docking score of -5.758. It showed only
one HB interaction with residue Tyr665 in the hinge region which explains its
lower potency (Figure 9C).
(Please insert Figure 9 here)
Compound 7g was shown to be more potent than 7b by 190-fold while the only
difference is the substitution pattern on the phenoxy moiety. The p-methoxy in 7g
was replaced by p-methyl in 7b along with having a methoxy group on the 2-
position in case of 7b. We believe that having the methoxy group at position 2 lead
to steric bulkiness that forced the compound to shift from binding tightly with the
receptor along with losing its chance to bind with Lys612 due to having a methyl

ACCEPTED MANUSCRIPT
group. This comparison highlights the importance of carefully modulating the
substitution pattern on the phenoxy moiety to avoid bulkiness while keeping a
small HBA in position 4 for potential HB with Lys612 (Figure 10).
(Please insert Figure 10 here)
2
.6.Cellular assay for anti-inflammatory activity.
Stimulation of CSF1R by CSF1 is known to increase the production LPS as well as
other inflammatory cytokines such as TNF- and IL-1 [48, 49]. It’s also known
α
β
that stimulation of Src family kinases has a significant role in the production of
LPS and therefore pro-inflammatory cytokines [50]. Inhibition of CSF1R and LCK
by small molecules is known to inhibit the production of these cytokines resulting
in anti-inflammatory action. Therefore, to test the anti-inflammatory effect of our
compounds, LPS-induced RAW 264.7 macrophages were used. All the materials
and experimental methods can be found in the supporting material. The results for
each test are described in details as follow.
2
.6.1. Cell viability and NO production in RAW 264.7 macrophages.
Since it’s very important for our compounds to show anti-inflammatory effect
without significant cytotoxicity to the macrophages, our compounds were first

ACCEPTED MANUSCRIPT
tested for cell viability in vitro in raw 264.7 macrophages side by side to NO
production assay as a preliminary evaluation method and the results are shown in
Figures 11 and 12. A combined results for both tests are tabulated in Table 4. The
results showed that with exception of 7k, all of the tested compounds didn’t show
significant cytotoxicity to the raw 264.7 macrophages. It was also noted that 7g
was able to inhibit the LPS-induced NO production in RAW 264.7 macrophages in
dose-dependent manner and showed more than 2.5-fold better potency than the
positive control used ( -NIL [51]) therefore it was selected for further anti-
L
inflammatory screening assay.
(
Please insert Figure 11 here)
Please insert Figure 12 here)
(
(
Please insert Table 4 here)
2
.6.2. LPS-induced PGE2 production in RAW 264.7 macrophages.
It’s known that RAW 264.7 macrophages release PGE upon stimulation with LPS
2
so 7g was tested for LPS-induced PGE production inhibition. As shown in Figure
2

ACCEPTED MANUSCRIPT
1
3, 7g was able to inhibit the production of PGE in dose-dependent manner. NS-
2
3
98, a known selective and potent COX-2 inhibitor [52], was used as a positive
control agent.
(Please insert Figure 13 here)
2
.6.3. iNOS and COX-2 proteins and mRNA expressions in LPS-induced
RAW 264.7 macrophages.
Compound 7g was then tested to determine whether its anti-inflammatory effect
was attributed to inhibiting the expression of COX-2 and iNOS proteins. Western
plotting technique was used and the results showed that LPS upregulated the levels
of COX-2 and iNOS proteins and that pretreatment with 7g was able to inhibit the
upregulation of iNOS and iNOS mRNA, but not COX-2 enzyme as illustrated in
Figure 14. This results clarify that the inhibition of PGE production is not related
2
to COX-2 inhibition.
(Please insert Figure 14 here)

ACCEPTED MANUSCRIPT
2
.6.4. Productions of pro-inflammatory cytokines in LPS-induced RAW
2
64.7 macrophages.
The effect of 7g on LPS-induced production of TNF-
explored and the results are shown in Figure 15. 7g was able to inhibit the
production of TNF- and IL-6 at concentration of 12 uM, but was unable to inhibit
the production of IL-1 at the same concentration.
α, IL-1β and IL-6 was
α
β
(Please insert Figure 15 here)
2
.6.5. Expressions of TNF-α and IL-6 mRNA in LPS-induced RAW 264.7
macrophage.
The effect of 7g on the expression of TNF-
RAW 264.7 macrophages was explored to clarify the reason for the noticed
inhibition of TNF- and IL-6. 7g was able to inhibit the expression of TNF- and
IL-6 mRNA at 2–4 h as shown in Figure 16.
α and IL-6 mRNA in LPS-induced
α
α
(Please insert Figure 16 here)
3
. Conclusion

ACCEPTED MANUSCRIPT
In conclusion, the search for a potent and selective kinase inhibitors to modulate
the immune system and target inflammatory diseases and cancers where immunity
is involved, is still a growing research area and the interest of several research
groups and pharmaceutical companies. In this report we have designed and
synthesized four series of a novel inhibitors based on phenoxypyrimidine-like
scaffold for LCK and FMS kinases in an attempt to target inflammatory disorders.
In vitro enzymatic assay for both kinases, selectivity profile as a proof-of-design,
SAR study, molecular docking study and in vitro cell-based assay over LPS-
induced RAW 264.7 macrophages were all performed as part of this study. We
identified a novel LCK selective inhibitors (for example; 7j), LCK-FMS dual
inhibitor (7g) and FMS selective inhibitors (for example; 7f). The SAR and the
modelling results helped us to rationalize the obtained results. Hence we introduce
7
g as potential dual FMS/LCK inhibitor with excellent selectivity profile, for the
treatment of inflammatory disorders including RA. We believe that this report can
be a step forward in the search for dual kinase inhibitors, acting through different
mechanisms, to target inflammatory disorders. The obtained results proves the
hypothesis that this novel substitution and substitution pattern on the pyrimidine
ring worth to be deeply explored with more derivatives and more biological assays.
4
. Experimental section

ACCEPTED MANUSCRIPT
4
.1.Chemistry
General: All reactions and manipulations were performed using standard Schlenk
techniques. Starting materials, reagents and solvents were purchased from
commercial suppliers and were used without purification with the following
exception. Deuterated chloroform was pretreated according to literature method
[53] and was only used for the imine samples (7a-m). All melting points were
measured by Optimelt Automated Melting Point System (Stanford Research
Systems) and are uncorrected. Thin-layer chromatography was performed with
Merk silica gel 60 F254 pre-coated glass sheets. Column chromatography was
performed on Merck Silica Gel 60 (230-400 mesh) for all the samples except the
imines, and Aluminum oxide (activated, neutral, Brockmann I of Aldrich Co.) for
the imines and the eluting solvents are noted as mixed solvent with given volume-
1
13
to-volume ratios or as percentage. H & C NMR was measured on a 400 MHz
Bruker Avance NMR spectrometer. Chemical shifts and coupling constants are
presented in parts per million (ppm) relative to Me Si and hertz (Hz), respectively
4
and the following abbreviations are used: s, singlet; d, doublet; dd, doublet of
doublets; ddd, doublet of doublets of doublets; t, triplet; m, multiplet. NMR spectra
were obtained from DMSO-d or CDCl solutions as indicated. High-resolution
6
3
mass spectra were performed on Waters ACQUITY UPLC BEH C18 1.7µ−Q-TOF
SYNAPT G2-Si High Definition Mass Spectrometry.

ACCEPTED MANUSCRIPT
4
.2.General procedure for preparation of 2a-e [39, 40]
To a solution of 2,4-dichloro-5-nitro-pyrimidine (1.0 g; 5.15 mmol) in acetone (20
mL), cooled to 0° C., was added a solution of appropriate phenol (5.15 mmol) in a
mixture of 1 N NaHCO (aq) (5 mL) and H O (2 mL) dropwise. After completion
3
2
of addition, the reaction mixture was allowed to warm to ambient temperature, and
was then stirred at room temperature for 2 h (monitored by TLC). The resultant
mixture was evaporated in vacuo, and the residue was washed sequentially with
EtOAc, 1 N NaOH (aq) (3 x 100 mL). The organic phase was washed sequentially
with H O (3 x 200 mL) and brine (3 x 200 mL), and dried over anhydrous MgSO .
2
4
The mixture was filtered and the solvent was evaporated under reduced pressure to
give the crude material. The crude material was purified by flash column
chromatography using 10 – 20% EtOAc in hexane mixture as the mobile phase to
yield 2a-e as yellow oil solidifies on trituration with hexane and drying.
4
.2.1. 2-chloro-4-(2-methoxy-4-methylphenoxy)-5-nitropyrimidine (2a)
1
Yellowish white solid, yield 85%, mp: 118.8–119.6 °C. H NMR (400 MHz,
CDCl3):
δ 9.18 (s, 1H), 7.07 (d, 1H, J = 7.8 Hz), 6.86–6.84 (m, 2H), 3.78 (s, 3H),
13
2
.43 (s, 3H); C NMR (100 MHz, CDCl ):
δ
162.99, 162.14, 157.09, 150.07,
3
1
38.07, 137.81, 131.22, 121.59, 121.51, 113.78, 55.90, 21.56.

ACCEPTED MANUSCRIPT
4
.2.2. 2-chloro-4-(3-methoxyphenoxy)-5-nitropyrimidine (2b) (Reported
[
46])
Yellowish white solid, yield 59%, mp: 213.3–214.9 °C. H NMR (400 MHz,
CDCl3): 9.16 (s, 1H), 7.37 (t, 1H, J = 8.3 Hz), 6.89 (ddd, 1H, J = 8.4, 2.4, 0.7 Hz),
.78 (ddd, 1H, J = 8.1, 2.3, 0.7 Hz), 7.73 (t, 1H, J = 2.3 Hz), 3.84 (s, 3H).
.2.3. 2-chloro-4-(4-methoxyphenoxy)-5-nitropyrimidine (2c) (Reported
39, 40])
Yellowish white solid, yield 89%, mp: 106.1–108.1 °C. H NMR (400 MHz,
DMSO-d6): 9.10 (s, 1H), 7.16–7.12 (m, 2H), 7.02–6.99 (m, 2H), 3.78 (s, 3H).
1
δ
6
4
[
1
δ
4
.2.4. 2-chloro-4-(4-(trifluoromethyl)phenoxy)-5-nitropyrimidine (2d)
1
Yellowish white solid, yield 48%. H NMR (400 MHz, DMSO-d ):
δ 9.47 (s, 1H),
6
13
7
.95 (d, 2H, J = 8.6 Hz), 7.61 (d, 2H, J = 8.4 Hz); C NMR (100 MHz, CDCl ):
δ
3
1
60.72, 155.43, 149.69, 147.80, 126.88, 125.05, 119.53, 115.65.
4
.2.5. 2-chloro-4-(4-fluorophenoxy)-5-nitropyrimidine (2e)
1
Yellowish white solid, yield 82%, mp: 238.1–240.0 °C. H NMR (400 MHz,
13
CDCl3):
δ 9.17 (s, 1H), 7.18–7.16 (m, 4H); C NMR (100 MHz, CDCl ): δ 161.90,
3
1
61.28, 160.98, 158.86, 158.21, 147.07, 132.25, 123.46, 123.37, 116.82, 116.59.

ACCEPTED MANUSCRIPT
4
.3.General procedure for preparation of 3a-f [41]
To a solution of compounds 2a-e (1 mmol) in THF (5 mL), cooled to 0° C.,
was added a solution of appropriate aniline (1 mmol) in THF (3 mL) and pyridine
79.1 mg, 1 mmol). The reaction mixture was allowed to warm to ambient
(
temperature, and was then refluxed at 80 °C for 4 h (monitored by TLC). The
resultant mixture was evaporated in vacuo to give the crude material which was
purified by column chromatography using 30 – 50% EtOAc in hexane mixtures as
the mobile phase to yield 3a-f.
4
.3.1. 4-(2-methoxy-4-methylphenoxy)-N-(4-morpholinophenyl)-5-
nitropyrimidin-2-amine (3a)
1
Red orange solid, yield 73%, mp: 184.1–185.9 °C. H NMR (400 MHz, DMSO-d ):
6
δ
10.63 (s, 1H), 9.13 (s, 1H), 7.13–7.08 (m, 3H), 6.87 (d, 2H, J = 8.0 Hz), 6.57 (d,
H, J = 8.7 Hz), 3.72 (t, 4H, J = 4.6 Hz), 3.69 (s, 3H), 3.00 (t, 4H, J = 4.4 Hz),
2
+
2
.43 (s, 3H). HRMS (ES+): m/z calculated for C H N O : 460.1597 [M+Na] .
22
23
5
5
Found 460.1603.
4
.3.2. 4-(3-methoxyphenoxy)-N-(4-morpholinophenyl)-5-nitropyrimidin-2-
amine (3b)
1
Red solid, yield 61%. H NMR (400 MHz, DMSO-d ):
δ 10.65 (s, 1H), 9.14 (s,
6
1
H), 7.47–7.43 (m, 1H), 7.17 (d, 2H, J = 8.9 Hz), 7.01–6.87 (m, 3H), 6.58 (d, 2H,

ACCEPTED MANUSCRIPT
13
J = 9.0 Hz), 3.77 (s, 3H), 3.72 (t, 4H, J = 4.3 Hz), 2.99 (t, 4H, J = 4.1 Hz); C
NMR (100 MHz, DMSO-d6): 165.51, 159.36, 156.79, 153.72, 148.37, 145.76,
27.61, 124.24, 124.14, 122.54, 114.60, 114.46, 55.33, 54.39, 47.90, 45.63.
δ
1
4
.3.3. 4-(4-methoxyphenoxy)-N-(4-morpholinophenyl)-5-nitropyrimidin-2-
amine (3c)
1
Orange solid, yield 51%, mp: 226.0–227.1 °C. H NMR (400 MHz, DMSO-d ):
δ
6
1
0.56 (s, 1H), 9.11 (s, 1H), 7.21 (d, 2H, J = 8.1 Hz), 7.13 (d, 2H, J = 7.8 Hz), 7.07
(d, 2H, J = 8.0 Hz), 6.57 (d, 2H, J = 7.9 Hz), 3.83 (s, 3H), 3.72 (s, 4H), 2.99 (s,
1
3
4
1
H); C NMR (100 MHz, DMSO-d ): δ 163.14, 158.83, 158.11, 157.21, 147.17,
6
45.24, 130.08, 123.02, 122.69, 120.73, 114.69, 114.60, 65.94, 55.45, 48.54.
+
HRMS (ES+): m/z calculated for C H N O : 446.1441 [M+Na] . Found 446.1442.
2
1
21
5
5
4
.3.4. 4-(4-(trifluoromethyl)phenoxy)-N-(4-morpholinophenyl)-5-
nitropyrimidin-2-amine (3d)
1
Red orange solid, yield 53%, mp: 220.1–221.0 °C. H NMR (400 MHz, DMSO-d ):
6
δ
7
4
1
10.68 (s, 1H), 9.17 (s, 1H), 7.95 (d, 2H, J = 8.6 Hz), 7.58 (d, 2H, J = 8.4 Hz),
.03 (d, 2H, J = 9.0 Hz), 6.54 (d, 2H, J = 9.1 Hz), 3.71 (t, 4H, J = 4.4 Hz), 2.95 (t,
1
3
H, J = 4.6 Hz); C NMR (100 MHz, DMSO-d ):
δ
163.12, 159.32, 158.91,
6
55.62, 147.86, 130.33, 127.77, 127.52, 124.13, 123.48, 121.35, 115.56, 115.00,

ACCEPTED MANUSCRIPT
+
6
6.44, 48.93. HRMS (ES+): m/z calculated for C H F N O : 484.1209 [M+Na] .
21
18
3
5
4
Found 484.1208.
4
.3.5. 4-(4-fluorophenoxy)-N-(4-morpholinophenyl)-5-nitropyrimidin-2-
amine (3e)
1
Red orange solid, yield 73%. H NMR (400 MHz, DMSO-d ):
δ 10.65 (s, 1H),
6
9
3
δ
1
.14 (s, 1H), 7.41–7.37 (m, 4H), 7.11 (d, 2H, J = 9.0 Hz), 6.61 (d, 2H, J = 9.0 Hz),
13
.72 (t, 4H, J = 4.8 Hz), 2.98 (t, 4H, J = 4.7 Hz); C NMR (100 MHz, DMSO-d ):
6
163.38, 159.33, 158.78, 148.47, 147.93, 130.36, 124.66, 124.58, 123.52, 121.43,
21.34,
117.01,
116.77,
115.56,
115.20,
66.48, 49.09.
HRMS
+
(
ES+): m/z calculated for C H FN O : 434.1241 [M+Na] . Found 434.1241.
20
18
5
4
4
.3.6. N-(3,5-dimethoxyphenyl)-4-(2-methoxy-4-methylphenoxy)-5-
nitropyrimidin-2-amine (3f)
1
Yellow solid, yield 65%, mp: 131.6–133.5 °C. H NMR (400 MHz, DMSO-d ):
δ
6
1
7
0.59 (s, 1H), 9.19 (s, 1H), 7.13 (d, 1H, J = 8.0 Hz), 7.04 (s, 1H), 6.83 (d, 1H, J =
1
3
.9 Hz), 6.57 (s, 1H), 6.18 (s, 1H), 3.71 (s, 3H), 3.60 (s, 6H), 2.38 (s, 3H); C
161.15, 161.07, 159.82, 158.06, 150.48, 138.99,
38.50, 137.33, 122.03, 121.35, 113.85, 101.78, 98.28, 95.96, 55.87, 55.38, 55.30,
1.53.
NMR (100 MHz, CDCl3):
δ
1
2