Design, synthesis and biological evaluation of cobalt(II)-Schiff base complexes as ATP-noncompetitive MEK1 inhibitors

Article abstract of DOI:10.1016/j.jinorgbio.2019.03.022

In this report, we designed and synthesized a series of cobalt(II)-Schiff base complexes (CoSBC) with competent MEK1 (mitogen-activated protein kinase kinase?1) inhibitory activity. Based on our previous report, the CoSBC exhibited high binding affinity with MEK1 protein. To further explore metal complexes as MEK1 inhibitors, a series of transition metals and ligands were employed to build a library of various metal Schiff base complexes. The MEK inhibition assays revealed that only CoSBC exhibited obvious inhibitory activity, complex 2b showed the best inhibition both in BRaf (B-rapidly accelerated fibrosarcoma)/MEK1 and MEK1/ERK2 (extracellular signal-regulated kinases-2) cascading (IC₅₀ is 1.988 ± 0.14 μM and 1.589 ± 0.054 μM respectively). In addition, homogeneous time-resolved fluorescence test method was used to prove that CoSBC as ATP-noncompetitive MEK1 inhibitor. MEK kinase selectivity assay indicated that CoSBC can selectively inhibit MEK1/2 kinases rather than other MAPKs (mitogen-activated protein kinases) family kinases. Moreover, the interaction mode of 2b with MEK1 protein has been demonstrated by computer aided drug design.

Full text of DOI:10.1016/j.jinorgbio.2019.03.022

Accepted Manuscript
Design, synthesis and biological evaluation of cobalt(II)-Schiff
base complexes as ATP-noncompetitive MEK1 inhibitors
Hongyue Li, Dandan Xi, Yan Niu, Chao Wang, Fengrong Xu, Lei
Liang, Ping Xu
PII:
S0162-0134(18)30676-7
DOI:
Reference:
https://doi.org/10.1016/j.jinorgbio.2019.03.022
JIB 10682
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
16 November 2018
22 March 2019
27 March 2019
Please cite this article as: H. Li, D. Xi, Y. Niu, et al., Design, synthesis and biological
evaluation of cobalt(II)-Schiff base complexes as ATP-noncompetitive MEK1 inhibitors,
Journal of Inorganic Biochemistry, https://doi.org/10.1016/j.jinorgbio.2019.03.022
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
Design, synthesis and biological evaluation of cobalt(II)-Schiff base
complexes as ATP-noncompetitive MEK1 inhibitors
Hongyue Li, Dandan Xi, Yan Niu, Chao Wang, Fengrong Xu, Lei Liang* and Ping Xu*
Department of Medicinal Chemistry, School of Pharmaceutical Sciences, Peking University Health Science
Center, 38 Xueyuan Road, Beijing, 100191, China
Abstract
In this report, we designed and synthesized a series of cobalt(II)-Schiff base complexes
(CoSBC) with competent MEK1 (mitogen-activated protein kinase kinase-1) inhibitory
activity. Based on our previous report, the CoSBC exhibited high binding affinity with MEK1
protein. To further explore metal complexes as MEK1 inhibitors, a series of transition metals
and ligands were employed to build a library of various metal Schiff base complexes. The
MEK inhibition assays revealed that only CoSBC exhibited obvious inhibitory activity,
complex 2b showed the best inhibition both in BRaf (B-rapidly accelerated
fibrosarcoma)/MEK1 and MEK1/ERK2 (extracellular signal-regulated kinases-2) cascading
(IC50 is 1.988 ± 0.14 μM and 1.589 ± 0.054 μM respectively). In addition, homogeneous time-
resolved fluorescence test method was used to prove that CoSBC as ATP-noncompetitive
MEK1 inhibitor. MEK kinase selectivity assay indicated that CoSBC can selectively inhibit
MEK1/2 kinases rather than other MAPKs (mitogen-activated protein kinases) family kinases.
Moreover, the interaction mode of 2b with MEK1 protein has been demonstrated by computer
aided drug design.
Keywords: cobalt(II)-Schiff base complexes, ATP-noncompetitive MEK1 inhibitors,
molecular docking
1. Introduction
Mitogen-activated protein kinase (MAPK) is an important protein kinase found in the mid-
1980s and is an important core signaling pathway in cells. The MAPK pathway is mainly
divided into four subfamilies including extracellular signal-regulated kinase (ERK) pathway,
*Corresponding author. Tel: +86-10-82802632. Fax: +86-10-82801117. Email address: pingxu@bjmu.edu.cn;
leiliang@bjmu.edu.cn
1

ACCEPTED MANUSCRIPT
c-Jun NH2-terminal kinase (JNK) pathway, p38 pathway and ERK5 (extracellular signal-
regulated kinases-5) pathway [1]. The Ras (a class of protein called small
GTPase)/Raf/MEK/ERK signaling pathway (also known as MAPK/ERK pathway) (Fig. 1) is
responsible for the coordination and regulation of cell growth and differentiation in response
of extracellular stimulations [2-4]. MEK (mitogen-activated protein kinase kinase), as a
pivotal node in this pathway, is activated by Raf (rapidly accelerated fibrosarcoma), which
including A-, B- and C-rapidly accelerated fibrosarcoma (ARaf, BRaf and CRaf) three
subtypes, and located in the downstream of it. Complete the process from inactivity to
phosphorylation. MEK1 and MEK2 are dual specificity kinases and responsible for the
phosphorylation and activation of the ERK1 and ERK2. Subsequently, activated ERK1/2 can
translocate to the nucleus and phosphorylate additional transcription factors. These series of
activities regulate multiple cellular events including proliferation, survival and differentiation,
etc.
Figure. 1 Components and functions of ERK signaling pathway interfered by cobalt complex. P:
phosphorylation, Co: CoSBC, CTS: cytosolic substrates, NCS: nuclear substrate.
The abnormal activation of this pathway is closely related to the occurrence and
development of many diseases, such as cerebral injury, myocardial hypertrophy, diabetes,
inflammation and cancer, among which cancer research is most conducted in-depth [5]. The
2

ACCEPTED MANUSCRIPT
up-regulation of this pathway is observed in many kinds of tumors including myeloma [6],
metastatic biliary carcinoma [7], non-small cell lung cancer and prostate cancer [8], etc. The
development of MEK inhibitors has prompted investigations against both this important drug
target and the whole pathway [9]. In the past two decades, many small molecular MEK
inhibitors have been reported [10]. According to the binding site of inhibitor with MEK
protein, it can be classified as ATP competitive MEK inhibitors and ATP noncompetitive
MEK inhibitors. However, ATP competitive MEK inhibitors exhibit lower selectivity and
more serious toxic side effects than ATP noncompetitive inhibitors. Therefore, the latter one
is more popular.
To date, three of them have been approved by FDA (food and drug administration), namely
Trametinib [11] (in 2013), Cobimetinib [12] (in 2015) and Binimetinib (in 2018, combined
with Encorafenib) for the treatment of advanced melanoma. Besides, Selumetinib has been
granted Orphan Drug Designation for the treatment of differentiated thyroid cancer [13] (Figs.
S1). Therefore, finding MEK inhibitors has important significance and prospects for the
treatment of diseases.
To our knowledge, the successful clinical application of inorganic drugs, such as
platinum(II) chemotherapeutics and gold-containing antiarthritic agents, have significantly
promoted the development of transition metals in medicine [14]. Among transition metals,
cobalt has unique physical and chemical properties, it can adopt a wide variety of coordination
ligands, geometries and oxidation states. These characters make cobalt can be manipulated to
potential drug candidates [15]. Moreover, cobalt plays a key role in composing vitamin B12
[16] and cobalt nitrilehydratases [17]. It also displays diverse properties as cobalt complexes
in antitumor [18, 19], antiviral [20, 21] and antimicrobial [22-24] drug candidates. However,
cobalt(II)-Schiff base complexes (CoSBC) used as MEK inhibitors has been scarcely reported
[9].
According to our previous report, the CoSBC exhibited high binding affinity with MEK1
protein [9]. However, it is not clear that if this kind of compounds exhibit MEK1 kinase
3

ACCEPTED MANUSCRIPT
inhibition and the mode of action. Herein, we design and synthesize a series of metal
complexes to explore that question. Homogeneous time-resolved fluorescence assay (HTRF)
technical method was used to test molecular level MEK1 inhibitory activity. The mechanism
by which CoSBC inhibits MEK1 kinase was studied and docking study was used to mimic the
CoSBC-MEK1 protein binding mode. The finding of this study will promote the discovery of
novel MEK1 targeted inhibitors and investigation of the target ability of MEK1 allosteric
binding site.
2. Results and discussion
2.1 Chemistry
The structures and synthetic procedures of the target compounds were outlined in Scheme
1. Different diamines were dissolved in methanol and reacted with a series of substituted
salicylaldehydes or pyridine salicylaldehydes to prepare organic ligands [25-33]. Then the
metal complexes were prepared according to literatures [25-33].
4

ACCEPTED MANUSCRIPT
Scheme 1. Synthetic route of ligands and metal complexes.
2.2 Biological evaluation
2.2.1 In vitro MEK1 inhibition and structure activity relationship (SAR) analysis
The metal complexes and selected Schiff base ligands were tested with enzyme-based
assays for their MEK inhibitory properties. Trametinib (approved by FDA in 2013) and
U0126 which are recognized allosteric MEK1 inhibitors reported before were used as the
positive controls. Most of the CoSBC showed promising inhibitory potencies both in
BRaf/MEK1 and MEK1/ERK2 cascading and also better ligand efficiency (LE) than U0126
(Table 1). Meanwhile, none of the Schiff base ligands showed inhibitory potencies in either
BRaf/MEK1 and MEK1/ERK2 cascading (Tables S1). Continuous assays were then
performed to determine the IC50 values of the selected candidates (inhibition ratio > 90% at
10 μM). As listed in Table 1, most of the CoSBC exhibited inhibitory activity both in
5

ACCEPTED MANUSCRIPT
BRaf/MEK1 and MEK1/ERK2 cascading at micromolar range. Comparing 2a-2c,
introduction of hydroxyl (2d-f) group to the salicylaldehyde moiety showed no enhancement
of the MEK inhibition activity. Nevertheless, replacing cobalt(II) by other transition metals
such as copper(II), zinc(II), nickel(II), manganese(II), iron(II) (3a-c, 4a-c, 5a-c) resulted in
complete loss of activity. Most of the complexes containing halogen(s) (2g-s) showed low
activity, except 2s which exhibited good inhibitory activity in MEK1/ERK2 cascading.
Moreover, bulk diamine moiety showed negative effect to the activity. For instance, complex
(2t) showed less activity. In addition, when the Schiff base ligand was substituted by electron
withdrawing group (2u), the inhibitory activity also disappeared. It is noteworthy that
tetradentate NNOO moiety is vital as the central metal. For example, tetradentate NNNN
ligand was used to replace the NNOO analog (9a-b) and the MEK1 inhibitory activity
disappeared. Above all, compared 2b (IC50 = 1.589 ± 0.054 μM, LE is 0.32) and positive
control U0126 (IC50 = 24.03 ± 5.80 μM, LE is 0.24), the inhibitory activity of 2b is higher in
MEK1/ERK2 cascading and exhibited a better ligand efficiency. But is not higher than
positive drug Trametinib (IC50 = 0.069 ± 0.13 μM, LE is 0.27), however, the structural types
of this two compounds are completely different and 2b has a larger LE, indicating that CoSBC
also have a large structural modification space to enhance the inhibitory activity.
Table 1. Structure, in vitro enzymatic inhibitory activity and ligand efficiency of metal complexes
BRaf/MEK1a MEK1/ERK2a
BRaf/MEK1a MEK1/ERK2a
Compd.
Trametinib 0.00048±0.08
LEb Compd.
LEb
-IC50(μM)
-IC50(μM)
-IC50(μM)
-IC50(μM)
0.069±0.13
0.27
0.24
0.32
0.32
2r
2s
2t
> 10
> 10
n.t.
0.30
0.21
n.t.
U0126
2a
0.20±0.012
2.01±0.073
1.99±0.14
6.54±0.45
2.82±1.59
2.95±1.98
24.03±5.80
1.59±0.337
1.59±0.054
6.56±0.849
7.72±2.59
> 10
1.19±0.023
7.70±2.184
> 10
2b
2u
3a
3b
3c
2c
2.41±0.438 0.37
> 10
> 10
n.t.
2d
1.58±0.52
4.61±0.21
0.29
0.27
> 10
> 10
n.t.
2e
> 10
> 10
n.t.
6

ACCEPTED MANUSCRIPT
2f
2g
2h
2i
2.19±0.77
> 10
3.11±1.165
5.57±1.898
6.61±0.097
> 10
0.33
0.27
0.26
n.t.
4a
4b
4c
5a
5b
5c
6a
6b
6c
7a
9a
9b
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
> 10
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
n.t.
> 10
> 10
2j
> 10
> 10
n.t.
2k
2l
2.27±0.195
2.41±0.877
1.84±0.169
2.93±1.743
> 10
6.30±0.798
4.24±0.180
2.06±0.155
2.63±0.692
> 10
0.26
0.27
0.34
0.33
n.t.
2m
2n
2o
2p
2q
> 10
> 10
n.t.
> 10
> 10
n.t.
n.t.: not tested. a Values represent the mean ± SD determined in three independent experiments, each based
on three biological replicates. b LE: Ligand efficiency (in kcal mol−1 atom−1) was calculated as LE = 1.37
×(-logIC50)/N. IC50 in M and MEK/ERK cascading IC50 was used, N is the number of non-hydrogen atoms.
2.2.2 Effects of ATP concentrations on MEK1 inhibition
To our knowledge, ATP-noncompetitive inhibitors for MEK1 are superior to the ATP-
competitive analogs due to their less undesired adverse effects [34]. In order to explore the
inhibition mechanism of CoSBC, ATP competitive profiles of these inhibitors were evaluated
[35] using representative compound 2l and 2n in BRaf/MEK1 cascading inhibition assay by
changing concentrations of ATP (Fig. 2). The IC50 values were 3.450 μM, 1.921 μM and 1.749
μM for 2l, and 4.407 μM, 1.311 μM and 2.105 μM for 2n, respectively when the ATP
concentration varied from 33.33 μM to 100 μM and 300 μM. No increase in IC50s was noticed
even when ATP concentration was increased up to 300 μM, and it suggested that CoSBC
inhibited the BRaf/MEK1 assay in an ATP non-competitive manner rather than ATP-
competitive inhibitors.
7

ACCEPTED MANUSCRIPT
B
Complex 2n ATP competitve assay IC50 curve
A
Complex 2l ATP competitive assay IC50 Curves
1.5
1.0
0.5
0.0
1.5
1.0
300 μM
100 μM
33.33 μM
300 μM
100 μM
0.5
33.33 μM
0.0
-0.5
-0.5
0
2
4
6
0
2
4
6
Log(nM)
Log(nM)
ATP conc. 300 μM 100 μM 33.33 μM
IC50/nm 4407 1311 2105
ATP conc. 300 μM 100 μM 33.33 μM
IC50/nm 3450 1921 1749
Figure. 2 Dose-response curves of (A) complex 2l and (B) complex 2n in ATP competition assay.
2.2.3 Effects of ERK concentrations on MEK1 inhibition
Since 2s exhibited the most potent inhibition in MEK1/ERK2 cascading, it was further used
to examine the effects on phosphorylated MEK1 kinase in different concentrations of ERK2.
We tested the IC50 of 2s in presence of different concentrations of ERK2 (20 nM, 40 nM and
80 nM). As shown in Fig. 3, the IC50 values are 1.293 μM, 1.206 μM and 1.363 μM
respectively, did not show a difference of more than 10 μM. The results indicating MEK1
inhibition of the CoSBC may not mediated by binding to ERK and pointing to these complexes
probably located in the allosteric pocket of MEK1.
Complex 2s IC50 Curves
1.5
1.0
ERK 20 nM
ERK 40 nM
0.5
ERK 80 nM
0.0
-0.5
0
2
4
6
Conc. Log(nM)
ERK conc. 20 nM
40 nM
80 nM
IC50/nm
1293 nM 1206 nM 1363 nM
Figure. 3 Dose-response curves of complex 2s in ERK competition assay
2.3 Docking study
Computer aided drug design (CADD) was used to mimic the interaction of the CoSBC with
8

ACCEPTED MANUSCRIPT
MEK1 protein. As shown in Fig. 4, the docking results provides us some information as (1)
One benzene ring of the Schiff base ligand occupies the hydrophobic pocket surrounded by
Met 144, Phe 210 and Ile 142 snugly, stabilizing π-π stacking with Phe 210. (2) The Schiff
base ligand provides two oxygen anion, building a salt bridge with the key amino acid Lys 98.
It was calculated that the distance between O- and Lys 98 is 2.14 Å. (3) Besides, Co2+ has a
salt bridging effect with Asp 209 as the distance between Co2+ and oxygen of carbonyl of Asp
209 is 2.36 Å, indicating it may be coordinate covalent bonds. (4) In addition, the dihedral
angle between the salt bridge and the CoSBC is calculated as 79.8o and 81.4o, closed to the
right angle (Fig. 5). Since Co(II) favorably forms quadridentate and hexa-coordinate
coordination structure by its valence, the two axial coordination of cobalt(II) Schiff base
complex can be easily bounded to oxygen of the carbonyl of Asp 209 [25]. We hypothesis that
the tunable coordination of cobalt Schiff base complex and its flexibility based on the
appropriate radius of cobalt cation result in its unique MEK1 inhibitory activity rather than
other analogues.
Figure. 4 Cobalt(II)-Schiff base complex (2b) with MEK protein (3WIG) docking 2D diagram.
9

ACCEPTED MANUSCRIPT
Figure. 5 Location relation between cobalt(II)-Schiff base complex (2b) and residue ASP 209
2.4 MEK kinase selectivity study
If the designed compounds only produces inhibitory activity against the target pathway
kinase and has little effect on other pathways, it will reduce the side effects of the compounds.
Therefore, MEK kinase selectivity study is necessary. To elucidate the specificity for MEK
kinase of CoSBC, compound 2a and 2b were selected to kinase selectivity assay. MEK1/2
inhibitors U0126, PD0325901 and typical metal drug cisplatin were used as positive controls
(Table 2). All compounds were tested in 10 µM concentration.
12 kinases were selected to detect the inhibitory of MAPK family kinases and mTOR
(mammalian target of rapamycin) kinase in PIKK (phosphatidylinositol kinase-related kinase)
family associated with the MAPK family. Among the 12 MAPK family kinases, ERK pathway
kinases (BRaf, BRaf (V599E), MEK1/2, ERK1/2), JNK pathway kinases (JNK1α1, JNK2α2,
JNK3 and MKK4 (mitogen activated-protein kinase kinase-4)), MKK6 (mitogen activated-
protein kinase kinase-6) belongs to the p38 pathway and MEKK2 (MEK kinase-2) belongs to
the ERK5 pathway. The results show that compounds 2a and 2b, as well as U0126 and
PD0325901, can significantly inhibit BRaf, BRaf (V599E), MEK1 and MEK2 kinases. The
significant inhibitory activity of both the BRaf and BRaf (V599E) kinases is due to the demand
of inactive MEK1 as the substrate for the kinase inhibition test. This result is consistent with
10

ACCEPTED MANUSCRIPT
the results that CoSBC can inhibit both of the BRaf/MEK1 cascading and the MEK1/ERK2
cascading. Furthermore, the MEK1/2 kinase inhibitory activity of 2a and 2b are better than
U0126, and equivalent to PD0325901. The results show that 2a and 2b exhibit better MEK1/2
selectivity inhibition than other MAPKs family kinases. It is obvious that CoSBC can
selectively inhibit BRaf/MEK1 cascading and MEK1/ERK2 cascading (Table 2).
Table 2. Inhibition activities of selected compounds against 13 kinds of related kinases
Inhibition (%), 10 µM*
kinases
2a
100
104
34
56
40
14
80
94
97
25
62
-1
2b
106
103
49
54
35
36
67
84
89
6
cisplatin
U0126
99
100
-3
PD0325901
B-Raf(h)
B-Raf(V599E)(h)
JNK1α1(h)
JNK2α2(h)
JNK3(h)
43
5
107
103
1
-1
8
3
6
-5
1
2
-8
21
7
ERK1(h)
24
18
36
63
0
ERK2(h)
16
6
MEK1(h)
98
98
-1
-4
3
MEK2(h)
-9
-1
-10
-3
-6
MEKK2(h)
MKK4(m)
MKK6(h)
mTOR(h)
39
-2
-8
10
0
17
5
-7
*The data is the average of two independent replicates. (h) = Human, (m) = Mouse
3. Conclusion
In this study, we designed, synthesized a series of cobalt(II)-Schiff base complexes as novel
ATP-noncompetitive MEK1 inhibitors. Some candidates (2a-f, 2k-n, 2s-t) can inhibit both
inactive and phosphorylated MEK1 and then activate the downstream substrate ERK2. In
order to investigate the SAR of CoSBC, we chose diverse transition metals (Ni, Cu, Zn, Mn,
Fe) and/or ligands. However, none of these analogues exhibited MEK1 inhibitory activity. In
conclusion, CoSBC can selectively inhibit BRaf/MEK1 cascading and MEK1/ERK2
cascading among 13 kinases. ATP competitive assay and substrate ERK2 competitive assay
indicated that CoSBC are probably allosteric MEK1 inhibitors. The detailed mechanistic study
of CoSBC as antitumor and antimicrobial agents is ongoing in our laboratory.
4. Experimental section
11

ACCEPTED MANUSCRIPT
4.1 Chemicals and instrumentation
All the reagents were commercially available and used without further purification. The
positive drug U0126, PD0325901, cisplatin and Trametinib were purchased from Selleck
Chemicals.
Nuclear magnetic resonance (NMR) spectroscopy was performed on Bruker avance III
400spectrometers. Chemical shifts were reported in parts per million (ppm, δ) relative to
tetramethylsilane (0.00 ppm) as the internal standard unless otherwise noted. The coupling
constant (J) was reported in Hertz (Hz). Proton coupling patterns were abbreviated as follows;
s: singlet, d: doublet, m: multiplet. Data were presented as follows; chemical shift (multiplicity,
coupling constant, integration). Low- and high-resolution mass spectra (LRMS and HRMS)
were obtained with electrospray ionization (ESI) produced by Waters ACQ-SQDLC–MS
spectrometer and Bruker Apex IV FTMS ESI spectrometer, respectively.
4.2 Synthesis and characterization
4.2.1 General procedure for synthesis of Schiff base ligands
A solution of substituted salicylaldehyde (10 mmol) in methanol (20 mL) was added
dropwise to a solution of different substituted 1,2-diaminoethane (5 mmol) in methanol (20
mL). The mixture was refluxed for 2 h and the precipitate of Schiff base ligand was obtained
as solid product. Then the solid was separated by filtration, washed with cold methanol and
dried in vacuum. Finally, it was recrystallized from methanol to yield as pure ligand.
4.2.2 General procedure for synthesis of metal complexes
Corresponding metal salt solution (1 mmol) and Schiff base ligand (1 mmol) were dissolved
in methanol (10 mL) and reflux for 1 h. Crystals of the metal complexes were obtained by
cooling the reaction mixture.
4.2.2.1 2,2'-((1E,1'E)-(1,2-phenylenebis(azanylylidene)) bis (methanylylidene)) diphenol
cobalt(II) (2a)
Brown solid, yield 62 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1611; HRMS-ESI m/z [M]+ calcd
for C20H14CoN2O2: 373.0387, found: 373.0383.
4.2.2.2
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) diphenol cobalt(II) (2b)
12

ACCEPTED MANUSCRIPT
Red solid, yield 64 %, mp 300-301 ℃; IR (KBr, cm-1) v(C=N) 1604; HRMS-ESI m/z [M]+ calcd
for C20H20CoN2O2: 379.0857, found: 379.0851.
4.2.2.3 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis(methanylylidene)) diphenol
cobalt(II) (2c)
Green solid, yield 78 %, mp 312-313 ℃; IR (KBr, cm-1) v(C=N) 1625; HRMS-ESI m/z [M]+
calcd for C16H14CoN2O2: 325.0387, found: 325.0384.
4.2.2.4 4,4'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) bis (benzene-
1,3-diol) cobalt(II) (2d)
Brown solid, yield 38 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1610; HRMS-ESI m/z [M]+ calcd
for C20H14CoN2O4: 405.0286, found: 405.0287.
4.2.2.5
4,4'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) bis (benzene-1,3-diol) cobalt(II) (2e)
Brown solid, yield 89 %, mp 295-297 ℃; IR (KBr, cm-1) v(C=N) 1609; HRMS-ESI m/z [M]+
calcd for C20H20CoN2O4: 411.0755, found: 411.0750.
4.2.2.6 4,4'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis
(benzene-1,3-diol) cobalt(II) (2f)
Brown solid, yield 38 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1607; HRMS-ESI m/z [M]+ calcd
for C16H14CoN2O4: 357.0286, found: 357.0281.
4.2.2.7 2,2'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) bis (4-
chlorophenol) cobalt(II) (2g)
Black solid, yield 99 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1606; HRMS-ESI m/z [M]+ calcd
for C20H12Cl2CoN2O2: 440.9608, found: 440.9611.
4.2.2.8 2,2'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) bis (4-
bromophenol) cobalt(II) (2h)
Black solid, yield 91 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1598; HRMS-ESI m/z [M]+ calcd
for C20H12Br2CoN2O2: 528.8598, found: 528.8602.
4.2.2.9 6,6'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) bis (2,4-
dichlorophenol) cobalt(II) (2i)
Brown solid, yield 97 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1616; HRMS-ESI m/z [M+H]+
13

ACCEPTED MANUSCRIPT
calcd for C20H11Cl4CoN2O2: 509.8907, found: 509.8916.
4.2.2.10 6,6'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) bis (2,4-
dibromophenol) cobalt(II) (2j)
Black solid, yield 93 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1601; HRMS-ESI m/z [M+H]+
calcd for C20H11Br4CoN2O2: 685.6886, found: 685.6891.
4.2.2.11
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) bis (4-chlorophenol) cobalt(II) (2k)
Red solid, yield 82 %, mp 329-333 ℃; IR (KBr, cm-1) v(C=N) 1600; HRMS-ESI m/z [M]+ calcd
for C20H18Cl2CoN2O2: 447.0077, found: 447.0081.
4.2.2.12
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) bis (4-bromophenol) cobalt(II) (2l)
Red solid, yield 94 %, mp 318-321 ℃; IR (KBr, cm-1) v(C=N) 1597; HRMS-ESI m/z [M]+ calcd
for C20H18Br2CoN2O2: 534.9067, found: 534.9064.
4.2.2.13 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis (4-
chlorophenol) cobalt(II) (2m)
Brown solid, yield 54 %, mp 316-318 ℃; IR (KBr, cm-1) v(C=N) 1631; HRMS-ESI m/z [M]+
calcd for C16H12Cl2CoN2O2: 392.9608, found: 392.9606.
4.2.2.14 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis (4-
bromophenol) cobalt(II) (2n)
Brown solid, yield 80 %, mp 320-322 ℃; IR (KBr, cm-1) v(C=N) 1608; HRMS-ESI m/z [M]+
calcd for C16H12Br2CoN2O2: 480.8598, found: 480.8593.
4.2.2.15 6,6'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis (2,4-
dichlorophenol) cobalt(II) (2o)
Brown solid, yield 92 %, mp 289-292 ℃; IR (KBr, cm-1) v(C=N) 1602; HRMS-ESI m/z [M]+
calcd for C16H10Cl4CoN2O2: 460.8828, found: 460.8829.
4.2.2.16 6,6'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis (2,4-
dibromophenol) cobalt(II) (2p)
Brown solid, yield 91 %, mp > 350 ℃; IR (KBr, cm-1) v(C=N) 1597; LRMS-ESI
(C16H10Br4CoN2O2) m/z 663.6868 [M+Na]+.
14

ACCEPTED MANUSCRIPT
4.2.2.17 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis (4-
fluorophenol) cobalt(II) (2q)
Brown solid, yield 11 %, mp 321-325 ℃; IR (KBr, cm-1) v(C=N) 1632; HRMS-ESI m/z [M]+
calcd for C16H12CoF2N2O2: 361.0199, found: 361.0196.
4.2.2.18 6,6'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) bis (3-
fluorophenol) cobalt(II) (2r)
Brown solid, yield 3 %, mp 322-325 ℃; IR (KBr, cm-1) v(C=N) 1622; HRMS-ESI m/z [M]+
calcd for C16H12CoF2N2O2: 361.0199, found: 361.0197
4.2.2.19 6,6'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) bis (3-
fluorophenol) cobalt(II) (2s)
Brown solid, yield 38 %, mp 310-312 ℃; IR (KBr, cm-1) v(C=N) 1617; HRMS-ESI m/z [M]+
calcd for C20H12CoF2N2O2: 409.0199, found: 409.0198.
4.2.2.20
2,2'-((1E,1'E)-((1,2-diphenylethane-1,2-diyl)
bis
(azanylylidene))
bis
(methanylylidene)) diphenol cobalt(II) (2t)
Red solid, yield 42 %, mp 337-340 ℃; IR (KBr, cm-1) v(C=N) 1603; HRMS-ESI m/z [M]+ calcd
for C28H22CoN2O2: 477.1013, found: 477.1018.
4.2.2.21 3,4-bis((E)-(2-hydroxybenzylidene) amino) benzoic acid cobalt(II) (2u)
Brown solid, yield 50 %, mp 324-326 ℃; IR (KBr, cm-1) v(C=N) 1615; HRMS-ESI m/z [M]+
calcd for C21H14CoN2O4: 417.0286, found: 417.0282.
4.2.2.22 2,2'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) diphenol
copper(II) (3a)
Green solid, yield 44 %, HRMS-ESI m/z [M+H]+ calcd for C20H15CuN2O2: 378.0430, found:
378.0426.
4.2.2.23
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) diphenol copper(II) (3b)
Brown solid, yield 73 %, HRMS-ESI m/z [M+H]+ calcd for C20H21CuN2O2: 384.0899, found:
384.0898.
4.2.2.24 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) diphenol
copper(II) (3c)
15

ACCEPTED MANUSCRIPT
Brown solid, yield 86 %, HRMS-ESI m/z [M+H]+ calcd for C16H15CuN2O2: 330.0430, found:
330.0426.
4.2.2.25 2,2'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) diphenol
zinc(II) (4a)
Yellow solid, yield 99 %, 1H NMR (400 MHz, DMSO-d6): δ = 9.02 (s, 2H,N=CH), 6.51-7.92
13
ppm (m, 12H, Ph); C NMR (100 MHz, DMSO-d6): δ = 172.74, 163.29, 139.84, 136.69,
134.78, 127.71, 123.55, 119.90, 116.94, 113.42 ppm; HRMS-ESI m/z [M+H]+ calcd for
C20H15N2O2Zn: 379.0425, found: 379.0419.
4.2.2.26
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) diphenol zinc(II) (4b)
White solid, yield 83 %, 1H NMR (400 MHz, DMSO-d6): δ = 8.43 (s, 2H,N=CH), 6.41-7.21
(m, 8H, Ph), 3.72 (s, 2H, CH), 1.43-2.03 ppm (m, 8H, CH2); 13C NMR (100 MHz, DMSO-d6):
δ = 171.46, 167.27, 135.56, 133.23, 123.06, 119.93, 112.57, 62.58, 27.45, 21.62 ppm; HRMS-
ESI m/z [M+Cl]- calcd for C20H20N2O2ZnCl: 419.0505, found: 419.0506.
4.2.2.27 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) diphenol
zinc(II) (4c)
Yellow solid, yield 93 %, 1H NMR (400 MHz, DMSO-d6): δ = 8.44 (s, 2H,N=CH), 6.41-7.16
13
(m, 8H, Ph), 3.73 ppm (s, 4H, CH2); C NMR (100 MHz, DMSO-d6): δ = 168.46, 155.36,
135.19, 133.24, 123.15, 119.68, 112.63, 56.22 ppm; LRMS-ESI (C16H14N2O2Zn) m/z
330.8770 [M-H]-.
4.2.2.28 2,2'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) diphenol
nickel(II) (5a)
Red solid, yield 94 %, HRMS-ESI m/z [M+H]+ calcd for C20H15NiN2O2: 373.0487, found:
373.0488.
4.2.2.29
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) diphenol nickel(II) (5b)
Red solid, yield 95 %, HRMS-ESI m/z [M+H]+ calcd for C20H21NiN2O2: 379.0957, found:
379.0948.
4.2.2.30 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) diphenol
16

ACCEPTED MANUSCRIPT
nickel(II) (5c)
Red solid, yield 71 %, HRMS-ESI m/z [M+H]+ calcd for C16H15NiN2O2: 325.0487, found:
325.0484.
4.2.2.31 2,2'-((1E,1'E)-(1,2-phenylenebis (azanylylidene)) bis (methanylylidene)) diphenol
manganese(II) (6a)
Brown solid, yield 54 %, HRMS-ESI m/z [M]+ calcd for C20H14MnN2O2: 369.0436, found:
369.0437.
4.2.2.32
2,2'-((1E,1'E)-((1R,2S)-cyclohexane-1,2-diylbis
(azanylylidene))
bis
(methanylylidene)) diphenol manganese (II) (6b)
Green solid, yield 32 %, HRMS-ESI m/z [M]+ calcd for C20H20MnN2O2: 375.0905, found:
375.0900.
4.2.2.33 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) diphenol
manganese(II) (6c)
Green solid, yield 8 %, HRMS-ESI m/z [M]+ calcd for C16H14MnN2O2: 321.0436, found:
321.0431.
4.2.2.34 2,2'-((1E,1'E)-(ethane-1,2-diylbis (azanylylidene)) bis (methanylylidene)) diphenol
iron (II) (7a)
Black solid, yield 62 %, HRMS-ESI m/z [M]+ calcd for C16H14FeN2O2: 322.0405, found:
322.0400.
4.2.2.35 (N1E,N2E)-N1,N2-bis (pyridin-2-ylmethylene) benzene-1,2-diamine cobalt(II) (9a)
Pink solid, yield 20 %, HRMS-ESI m/z [M-Co+H]+ calcd for C18H15N4: 287.1297, found:
287.1293.
4.2.2.36 (1R,2S,N1E,N2E)-N1,N2-bis (pyridin-2-ylmethylene) cyclohexane-1,2-diamine
cobalt(II) (9b)
Black solid, yield 63 %, LRMS-ESI (C18H20CoN4) m/z 351.1104 [M]+.
4.3 Biological assays
Recombinant purified inactive GST-tagged MEK1 and ERK2 protein and recombinant
purified active GST-BRAF and GST-MEK1 kinase were purchased from Carna Biosciences
Inc., Japan. Polyclonal anti-phospho MEK1/2(Ser217/221)-cryptate, mAb anti-phospho
17

ACCEPTED MANUSCRIPT
p44/42 MAPK (Thr202/Tyr204)-cryptate and mAb anti GST-XL665 antibodies were
purchased from Cisbio Bioassays. All the synthesized derivatives were tested for MEK
inhibitory potency in BRaf/MEK1 and MEK1/ERK2 cascade assay. The positive control
Trametinib and U0126 were purchased from Selleck Chemicals.
All assays were conducted following the general protocol as below [35, 36]:
The homogeneous time resolved fluorescence (HTRF) assay was performed on a
Proxiplate-384 F plus solid white plate (Greiner) with 2 μL kinase (0.44 ng/μL BRaf or 0.5
ng/μL active MEK1), 2 μL substrate (30 nM inactive MEK1 or 40 nM ERK2), 2 μL (100 μM)
ATP and 4 μL test compound at a variety of concentrations, which were incubated at room
temperature for 2 hours. Then add 10 μL of the detection buffer which contain 200× polyclonal
anti-phospho MEK1/2 (Ser217/221)-cryptate (or mAb anti-phospho p44/42 MAPK
(Thr202/Tyr204)-cryptate) and 26 nM mAb anti GST-XL665 incubated at room temperature
for 3 hours. Finally, the fluorescence signals are detected with FlexStation 3 Multi-Mode
Microplate Reader (Molecular Devices, USA). This energy transfer is detected by an increase
in the fluorescence emission of the tracer at 668 nm and a decrease in the fluorescence
emission of europium at 620 nm. Using the following formula to calculate the inhibition rate:
Inhibition% = (Rn-Rc)/(Rn-Rb)×100%
in which Rn is negative control, Rc is compound signal and Rb is blank control. The curve-
fitting software GraphPad Prism was used to generate the curves and determine the IC50 values
for individual compounds tested. Values represent the mean ± SD determined in three
independent experiments, each based on three biological replicates.
4.4 Molecular docking
To explore the binding mode of active compound with MEK1, the module Glide of
Schrödinger (LLC, New York, NY, USA) software was used for docking [37]. The default
parameter settings were adopted. The three-dimensional (3D) structure of MEK1 (PDB ID:
3WIG) was downloaded from Protein Data Bank (PDB).
4.5 Kinase selectivity assay
The selectivity of representative compound 2a and 2b against MEK1 and related pathway
kinase were tested by Eurofins using KinaseProfiler™ Service Assay [38]. See the
18

ACCEPTED MANUSCRIPT
Supplementary Information for detailed protocols.
Abbreviations
CoSBC
DMSO
ERK
ESI
Cobalt(II)-Schiff base complexes
Dimethyl sulfoxide
Extracellular signal-regulated kinases
Electrospray ionization
FDA
HRMS
HTRF
Hz
Food and drug administration
High-resolution mass spectra
Homogeneous time resolved fluorescence
Hertz
JNK
c-Jun NH2-terminal kinase
Ligand efficiency
LE
LRMS
Low-resolution mass spectra
MAPKs Mitogen-activated protein kinases
MEK Mitogen-activated protein kinase kinase
MEKK2 MEK kinase-2
MKK
MS
Mitogen activated-protein kinase kinase
Mass spectrum
mTOR
NMR
PDB
Ph
Mammalian target of rapamycin
Nuclear magnetic resonance
Protein Data Bank
Phenyl
PIKK
ppm
Raf
Phosphatidylinositol kinase-related kinase
Parts per million
Rapidly accelerated fibrosarcoma
a class of protein called small GTPase
Structure activity relationship
Ras
SAR
Acknowledgments
19

ACCEPTED MANUSCRIPT
This work was supported by the Beijing Natural Science Foundation (7162110),
interdisciplinary medicine Seed Fund of Peking University (BMU2018MC004) and the
National Natural Science Foundation of China (21807006/ 81872731). The authors would also
acknowledge Prof. Zhenjun Yang and Prof. Liangren Zhang for kindly providing space and
suggestions on bio-experiments.
Conflict of Interest
The authors confirm that this article content has no conflict of interests.
References
[1] L. F. Chang, A. D. Sharrocks, A. Whitmarsh, Nature 410 (2001) 37-40.
[2] N. G. Ahn, R. Seger, R. L. Bratlien, C. D. Diltz, N. K. Tonks, E. G. Krebs, J. Biol. Chem.
266 (1991) 4220-7.
[3] W. Kolch, Expert Opin. Pharmacother. 3 (2002) 709-18.
[4] Schaeffer, H.J., Weber, M.J., Mol. Cell. Biol. 19 (1999) 2435-44.
[5] R. Roskoski, Pharmacol. Res. 66 (2012) 105-43.
[6] R. Pinto, S. Herold, et al, Antivir. Res. 92 (2011) 45-56.
[7] T. Bekaii-Saab, M. A. Phelps, et al, J. Clin. Oncol. 29 (2011) 2357-63.
[8] J. L. Bromberg-White, N. J. Andersen, et al, Brief. Funct. Genomics. 11 (2012) 300-10.
[9] H. Li, T. Zhou, H. Liu, et al, Bioorg. Med. Chem. Lett. 27 (2017) 2221-4.
[10] S. Price, Expert. Opin. Ther. Patents 18 (2008) 603-27.
[11] D. E. Uehling, P. A. Harris, Bioorg. Med.Chem. Lett. 25 (2015) 4047-56.
[12] K. D. Rice, N. Aay, et al, ACS Med. Chem. Lett. 3 (2012) 416-21.
20

ACCEPTED MANUSCRIPT
[13] Y. T. Patel, V. M. Daryani, et al, Pharmacometrics & Systems Pharmacology 6 (2017)
305-314.
[14] T. W. Hambley, Dalton Trans 21 (2007) 4929-37.
[15] M. C. Heffern, N. Yamamoto, R. J. Holbrook, et al, Curr. Opin. Chem. Biol. 17 (2013)
189-196.
[16] R. Banerjee, S. W. Ragsdale, Annu. Rev. Biochem. 72 (2003) 209-247.
[17] T. Yano, T. Ozawa, H. Masuda, Chem. Lett. 37 (2008) 672-7.
[18] D. C. Ware, B. D. Palmer, W .R. Wilson, et al, J. Med. Chem. 36 (1993) 1839-46.
[19] I. Ott, K. Schmidt, B. Kircher, et al, J. Med. Chem. 48 (2005) 622-29.
[20] V. Groupe, C. Engle, P.E. Gaffney, J. Bacteriol. 4 (1955) 623-4.
[21] J. B. Delehanty, J. E. Bongard, D. Thach, Bioorg. Med. Chem. 16 (2008) 830-7.
[22] M. N. Patel, M. R. Chhasatia, D. S. Gandhi, Bioorg. Med. Chem. Lett. 19 (2009) 2870-3.
[23] E. K. Efthimiadou, A. Karaliota, G. Psomas, Bioorg. Med. Chem. Lett. 18 (2008) 4033-
7.
[24] L. A. Saghatforoush, F. Chalabian, A. Aminkhan, et al, Eur. J. Med. Chem. 44 (2009)
4490-5.
[25] T. Joseph, S. B. Halligudi, C. Satyanarayan, et al, J. Mol. Catal. A. Chem. 168 (2001) 87-
97.
[26] S. Fu, Y. Liu, Y. Ding, et al, Chem. Commun. 50 (2014) 2167-9.
[27] R.C. Felicio, E. Cavalheiro, E. R. Dockal, Polyhedron 20 (2001) 261-8.
[28] R.H. Bailes, M. Calvin, J. Am. Chem. Soc. 69 (1947) 1886-93.
21

ACCEPTED MANUSCRIPT
[29] M. Hirotsu, M. Kojima, K. Nakajima, et al, Bull. Chem. Soc. Jpn. 69 (1996) 2549-57.
[30] L. N. Rusere, T. Shalumova, J. M. Tanski, et al, Polyhedron 28 (2009) 3804-10.
[31] J. Welby, L. N. Rusere, J. M. Tanski, et al, Inorg. Chim. Acta. 362 (2009) 1405-11.
[32] T. Uchida, T. Katsuki, K. Ito, et al, Helv. Chim. Acta. 85 (2002) 3078-89.
[33] Y. Shen, W. Duan, M. Shi, J Org Chem. 68 (2003) 1559-62.
[34] J. F. Ohren, H. Chen, A. Pavlovsky, et al, Nat. Struct. Mol. Biol. 11 (2004) 1192-7.
[35] J. Sun, Y. Niu, C. Wang, et al, Bioorg. Med. Chem. 24 (2016) 3472-82.
[36] C. Wang, H. Zhang, F. Xu, et al, Molecules 18 (2013) 6057-91.
[37] G. L. Reddy, S. K. Guru, M. Srinivas, et al, Eur. J. Med. Chem. 80 (2014) 201-8.
[38] B. Mao, S. Gao, et al, Eur. J. Med. Chem. 129 (2017) 135-150.
22

ACCEPTED MANUSCRIPT
Graphical abstract
23

ACCEPTED MANUSCRIPT
Highlights
1 Cobalt(II)-Schiff base complexes (CoSBC) are synthesized as MEK1(mitogen-activated
protein kinase kinase-1) inhibitors.
2 CoSBC can inhibit both inactive and phosphorylated MEK1 in ATP-noncompetitive
manner.
3 Cis-cyclohexanediamine-substituted CoSBC exhibits good MEK1 inhibitory activity.
4 CoSBC can selectively inhibit MEK kinases.
5 Binding mode of CoSBC with MEK1 plays a guiding role in the next step of research.
24