2.1 VIRTUAL SCREENING:
We have identified novel and potential inhibitors of MEK1 by HTS of several huge databases and MD simulations were performed. The MEK1 protein offers two sites that can be targeted by the inhibitor, i.e. ATP binding site and allosteric site. Docking studies were performed on allosteric sites as it offers kinase selectivity. It was done using Glide program which is a grid based ligand docking with energetics system of Schrodinger software. The data generated from docking the small-molecule libraries to allosteric site showed that the range of gscore for top 10 compounds from NCI library was -13.86 kcal/mol to -12.19 kcal/mol. Other high gscores were from KinaseSARfari library and ranged from -14.69 kcal/mol to -12.88 kcal/mol for top 10 compounds. But since the compounds from NCI library and KinaseSARfari were not procurable they weren’t considered for further analysis. The two libraries which had procurable compounds were ChemDiv Kinase and ChemDiv Anticancer. The highest and the lowest gscores for top 10 compounds, observed in case of ChemDiv Anticancer library were -11.0;5 kcal/mol and -10.73 kcal/mol, respectively. Similarly, the gscores for ChemDiv Kinase library range from -11.35 kcal/mol to -10.73 kcal/mol for top 10 compounds. Results of docking performed using Glide module from Schrodinger, were evaluated using GOLD software and post docking analysis was performed using X-score and Ligplot software. X-score scored the binding of protein and highest scoring conformation of the selected top ligand for both Glide and GOLD output. The same compounds were next analysed using Ligplot for their hydrophobic and hydrogen bond interactions with the protein. The analysis results of selected compounds are summarized in Table 1. The two best compounds based on all the scores were selected from each of these libraries for further evaluations. From ChemDiv Kinase library C387-1559 and from ChemDiv Anticancer library G545-0469 was selected for further evaluation. From the protein-ligand interaction analysis of the selected compounds, it was observed that the selected compounds and the benchmark molecules occupy same binding pocket (Fig. 3). The selected compounds C387-1559 and G545-0469 have higher number of hydrogen bonds and hydrophobic interactions as compared to the benchmark molecules Cobimetinib and Refametinib (Table 2). C387-1559 forms 2 hydrogen bond and 67 hydrophobic interactions whereas G545-0469 forms 4 hydrogen bond and 60 hydrophobic interactions, both of these are higher than 1 hydrogen bond each between Cobimetinib and MEK1 protein and Refametinib and MEK1 protein and 56 and 49 hydrophobic interactions between Cobimetinib and MEK1 protein and Refametinib and MEK1 protein, respectively. Higher interaction suggest more stable binding of ligand in the binding site.
Table 1 Comparison of screening and binding results for top ranking potential inhibitors against allosteric site of MEK1.
Compound ID
|
Glide gscore kcal/mol
|
Xscore Glide kcal/mol
|
Gold Fitness Score
|
Xscore Gold kcal/mol
|
Benchmark
|
Cobimetinib
|
-9.374
|
-8.77
|
53.82
|
-9.08
|
Refametinib
|
-6.95
|
-8.53
|
59.92
|
-8.32
|
ChemDiv Kinase
|
J081-2484
|
-11.35
|
-11.15
|
67.55
|
-9.64
|
Y020-4048
|
-11.61
|
-10.32
|
75.31
|
-9.84
|
S915-0049
|
-11.31
|
-9.51
|
64.68
|
-8.97
|
C387-1559
|
-11.15
|
-10.67
|
83.46
|
-10.45
|
K788-4520
|
-10.98
|
-10.11
|
76.65
|
-9.87
|
SA47-0465
|
-10.98
|
-9.07
|
68.27
|
-9.06
|
S639-3390
|
-10.83
|
-8.91
|
67.04
|
-9.05
|
Y020-4043
|
-11.08
|
-10.07
|
74.62
|
-9.82
|
J081-0036
|
-10.76
|
-10.32
|
75.83
|
-10.20
|
J081-0046
|
-10.73
|
-10.73
|
85.23
|
-10.66
|
ChemDiv Anticancer
|
D146-0335
|
-11.05
|
-10.52
|
63.74
|
-9.12
|
F919-1152
|
-11.04
|
-10.42
|
69.86
|
-9.53
|
F919-1156
|
-10.98
|
-10.63
|
65.95
|
-8.96
|
F919-0810
|
-10.91
|
-10.35
|
68.08
|
-8.82
|
C741-0772
|
-10.87
|
-10.95
|
81.85
|
-10.58
|
F919-0804
|
-10.85
|
-10.37
|
61.40
|
-8.98
|
C741-0200
|
-11.47
|
-9.83
|
82.27
|
-9.44
|
F919-1183
|
-10.82
|
-10.51
|
74.89
|
-9.55
|
F919-1160
|
-10.76
|
-10.88
|
77.34
|
-9.89
|
G545-0469
|
-10.73
|
-11.38
|
87.34
|
-11.22
|
KinaseSARfari
|
CHEMBL157649
|
-14.69
|
-9.92
|
196.52
|
-9.81
|
CHEMBL641001
|
-14.5
|
-10.02
|
168.57
|
-9.33
|
CHEMBL503716
|
-14.22
|
-10.07
|
185.52
|
-8.84
|
CHEMBL129187
|
-14.18
|
-10.48
|
252.09
|
-9.68
|
CHEMBL462871
|
-13.89
|
-10.37
|
253.01
|
-10.06
|
CHEMBL503713
|
-13.70
|
-9.99
|
218.73
|
-9.8
|
CHEMBL1140796
|
-13.56
|
-10.96
|
24.46
|
-10.97
|
CHEMBL162834
|
-13.06
|
-8.51
|
174.97
|
-7.98
|
CHEMBL162833
|
-13.00
|
-8.9
|
211.8
|
-7.9
|
CHEMBL236507
|
-12.88
|
-9.91
|
11.65
|
-9.69
|
NCI
|
NSC729646
|
-13.86
|
-10.99
|
311.13
|
-10.43
|
NSC691615
|
-13.67
|
-9.34
|
65.99
|
-9.44
|
NSC302048
|
-13.41
|
-10.13
|
51.97
|
-10.25
|
NSC627856
|
-13.40
|
-10.53
|
72.60
|
-10.49
|
NSC302048
|
-13.11
|
-9.97
|
52.25
|
-9.61
|
NSC92944
|
-12.91
|
-7.73
|
283.35
|
-8.01
|
NSC696477
|
-12.36
|
-9.62
|
33.64
|
-8.02
|
NSC302048
|
-12.20
|
-10.07
|
68.68
|
-9.93
|
NSC680101
|
-12.15
|
-8.86
|
76.55
|
-8.88
|
NSC733506
|
-12.19
|
-11.11
|
237.98
|
-10.89
|
Table 2 Comparison of Ligplot results for top ranking potential inhibitors and benchmark molecules at allosteric site.
Compound Ids
|
No. of Hydrogen bonds
|
No. of Hydrophobic Interactions
|
Donor- Acceptor residues involved in HB formation
|
Benchmark
|
Cobimetinib
|
1
|
56
|
Donor Acceptor Distance LIG ASP190 2.50
|
Refametinib
|
1
|
49
|
Donor Acceptor Distance ASN78 LIG. 3.00
|
ChemDiv Kinase
|
J081-2484
|
2
|
81
|
Donor Acceptor Distance SER212 LIG 2.91 VAL211 LIG 2.91
|
Y020-4048
|
2
|
68
|
Donor Acceptor Distance SER212 LIG 2.91 VAL211 LIG 2.89
|
S915-0049
|
2
|
56
|
Donor Acceptor Distance SER212 LIG 3.22 VAL211 LIG 2.89
|
C387-1559
|
2
|
67
|
Donor Acceptor Distance LIG ATP1 3.02 LYS97 LIG 3.12
|
K788-4520
|
2
|
77
|
Donor Acceptor Distance LIG ATP1 2.98 LYS97 LIG 3.09
|
SA47-0465
|
3
|
58
|
Donor Acceptor Distance SER212 LIG 3.30 VAL211 LIG 3.00 ASP208 LIG 2.99
|
S639-3390
|
0
|
60
|
______
|
Y020-4043
|
2
|
65
|
Donor Acceptor Distance SER212 LIG 3.11 VAL211 LIG 2.91
|
J081-0036
|
2
|
59
|
Donor Acceptor Distance SER212 LIG 2.99 VAL211 LIG 2.88
|
J081-0046
|
3
|
64
|
Donor Acceptor Distance SER212 LIG 3.06 VAL211 LIG 2.88 LYS192 LIG 3.20
|
J081-0058
|
2
|
53
|
Donor Acceptor Distance SER212 LIG 3.00 VAL211 LIG 2.90
|
ChemDiv Anticancer
|
D146-0335
|
4
|
78
|
Donor Acceptor Distance SER212 LIG 3.30 VAL211 LIG 2.93 LIG PHE209 2.86 ASP208 LIG 3.10
|
F919-1152
|
4
|
78
|
Donor Acceptor Distance SER212 LIG 3.30 VAL211 LIG 2.93 LIG PHE209 2.86 ASP208 LIG 3.10
|
F919-1156
|
4
|
78
|
Donor Acceptor Distance SER212 LIG 3.30 VAL211 LIG 2.93 LIG PHE209 2.86 ASP208 LIG 3.10
|
F919-0810
|
2
|
76
|
Donor Acceptor Distance LIG ATP1 2.75 LIG PHE209 2.79
|
C741-0772
|
2
|
76
|
Donor Acceptor Distance LIG ATP1 2.75 LIG PHE209 2.79
|
F919-0804
|
2
|
76
|
Donor Acceptor Distance LIG ATP1 2.75 LIG PHE209 2.79
|
C741-0200
|
5
|
141
|
Donor Acceptor Distance LIG ATP1 2.95 LIG ATP1 2.75 VAL211 LIG 3.22 LIG PHE209 2.79 LIG MET143 3.10
|
F919-1183
|
2
|
59
|
Donor Acceptor Distance LIG ATP1 2.64 ARG234 LIG 2.98
|
F919-1160
|
2
|
82
|
Donor Acceptor Distance LIG ATP1 2.62 LIG PHE209 2.74
|
G545-0469
|
4
|
60
|
Donor Acceptor Distance LIG PHE209 3.14 LYS192 LIG 2.94 LIG ASP190 2.85 LIG ASP190 2.67
|
For the sake of understanding of position and orientation of ligand molecules relative to receptor as well as for understanding core and rotamer group conformations of ligands against ATP-binding site of receptor, the small-molecule libraries were also docked and analysed for ATP-binding site. The libraries considered in this case were ChemDiv Anticancer, ChemDiv Kinase and ZINC Natural Product. The highest and the lowest gscores for top 10 compounds, observed in case of ChemDiv Anticancer library were -11.04 kcal/mol and -9.15 kcal/mol, respectively. Similarly, the gscores for ChemDiv Kinase library range from -10.17 kcal/mol to -9.42 kcal/mol for top 10 compounds. The best scores were from ZINC Natural Product library and ranged from -16.38 kcal/mol to -15.40 kcal/mol for top 10 compounds. The top compounds from Glide docking were further evaluated using GOLD docking and X-score analysis. The analysis results of selected compounds are summarized in Table 3.
Table 3 Comparison of screening and binding results for top ranking potential inhibitors against ATP binding site of MEK1.
Compound Name
|
Glide gscore Kcal/mol
|
Xscore Kcal/mol
|
Gold Fitness Score
|
Gold_Xscore Kcal/mol
|
Benchmark
|
E6201
|
-7.16
|
-9.37
|
56.7
|
-8.54
|
ChemDiv Anticancer
|
N001-0004
|
-11.05
|
-10.03
|
63.01
|
-9.91
|
7599-1863
|
-10.39
|
-8.29
|
54.18
|
-8.19
|
V005-8486
|
-10.26
|
-10.51
|
72.26
|
-9.39
|
C260-1185
|
-10.10
|
-10.94
|
79.14
|
-9.94
|
V011-0320
|
-9.93
|
-10.51
|
74.06
|
-9.11
|
L604-0810
|
-9.73
|
-9.13
|
68.98
|
-9.37
|
L604-1265
|
-9.60
|
-9.06
|
72.59
|
-9.33
|
P168-0038
|
-9.48
|
-8.66
|
64.06
|
-8.61
|
P168-0142
|
-9.38
|
-8.37
|
59.92
|
-8.36
|
P633-0126
|
-9.15
|
-7.93
|
50.34
|
-7.99
|
ChemDiv Kinase
|
D588-0188
|
-10.18
|
-8.96
|
65.46
|
-9.24
|
P168-0038
|
-9.48
|
-8.66
|
62.94
|
-8.58
|
D526-0244
|
-9.68
|
-9.18
|
64.77
|
-9.11
|
L987-0068
|
-9.38
|
-8.93
|
67.98
|
-9.55
|
D526-0268
|
-9.60
|
-8.90
|
68.37
|
-8.89
|
D526-0269
|
-9.51
|
-9.21
|
70.59
|
-9.52
|
D526-0227
|
-9.51
|
-9.31
|
67.68
|
-9.62
|
P168-0145
|
-9.23
|
-7.90
|
59.18
|
-8.14
|
D526-0215
|
-9.44
|
-9.22
|
66.92
|
-9.40
|
D526-0205
|
-9.42
|
-8.98
|
65.79
|
-9.19
|
ZINC Database
|
ZINC67910269
|
-16.39
|
-10.26
|
91.71
|
-10.19
|
ZINC67910271
|
-16.39
|
-10.26
|
91.32
|
-10.05
|
ZINC67911489
|
-16.11
|
-10.52
|
81.24
|
-9.95
|
ZINC67902753
|
-15.76
|
-9.68
|
76.10
|
-9.10
|
ZINC67912156
|
-15.65
|
-9.87
|
79.23
|
-9.61
|
ZINC08234351
|
-15.58
|
-10.00
|
81.54
|
-9.76
|
ZINC08234345
|
-15.49
|
-9.80
|
76.81
|
-9.33
|
ZINC67911001
|
-15.41
|
-10.24
|
71.24
|
-10.12
|
ZINC67911002
|
-15.41
|
-10.24
|
69.18
|
-10.13
|
ZINC67911004
|
-15.41
|
-10.24
|
71.32
|
-10.24
|
2.2 MOLECULAR DYNAMICS AND BINDING FREE ENERGY:
To determine and compare the stability of the protein-ligand complex for the selected ligands a molecular dynamic simulation of 100 ns was performed. For carrying out the MD simulation of protein-ligand complex formed at allosteric site of the protein, the force field was selected to include parameters for Mg++ and ATP therefore we selected charmm36 force field and TIP3P water molecule for this setup. The MD simulation results of selected compounds were compared with benchmark molecules. To measure the average deviation of the backbone atoms of protein-ligand complex suggesting the stability of the complex, RMSD was calculated and identified compounds it was well within the allowed range of 2Å. Initially there are slight fluctuations observed but towards the final 20ns the RMSD can be seen to stabilize for both the complexes (Fig.4(A)). The overall deviation in complexes of selected compounds with MEK1 was observed to be less than the complexes of reference molecules with MEK1, suggesting that selected compounds form more stable complexes than reference molecules. To measure the displacement of a particular atom, or group of atoms, relative to the reference structure, averaged over the number of atoms the RMSF analysis was carried out. From the RMSF plot, it was observed that both the molecules follow same pattern as the reference molecules Cobimetinib and Refametinib (Fig.4(B)). However, the fluctuations for G545-0469, were higher than both the reference molecules at a few sites but these were not the part of binding pocket where the inhibitor molecule was housed. The analysis of H-bonds was done to analyze the specificity of protein-ligand interactions. The stability of docked complexes was validated by computing HB, paired with 0.35 nm amongst donor and acceptor (Fig.4(C)). At least two to three additional hydrogen interactions were observed than benchmark compound, except in the case of Refametinib where towards the end of the run it forms one additional HB than selected compounds. The result also indicates that selected compounds were able to make stable complexes through reasonable numbers of hydrogen bonds during the 100 ns MD simulation process. Finally, the Rg values are calculated which are the indicator of size and compactness of the protein and throughout the 100 ns run these do not vary drastically for any of the studied complexes. The overall Rg values for both the identified compounds was within 2Å range and converges towards the end of the 100 ns run (Fig.4(D)). This indicates that the protein compactness does not varies much on binding with these ligands, thus suggesting that protein stability is retained on forming complex with the selected compounds.
The inhibitors were also subjected to the binding energies calculations to further elaborate on the protein ligand binding. This was calculated using polar and apolar solvation criteria. The protein–ligand energies that given in Table 4. Though the binding energies of the benchmark are slightly higher than the selected compounds, but the final average binding energy of identified compounds suggest their strong binding affinity with the protein.
Table 4 Comparison of MM-PBSA results for selected compounds and reference compounds.
Compounds Name
|
Binding energy ( kcal/mol)
|
Van der Waal energy ( kcal/mol)
|
Electrostatic energy ( kcal/mol)
|
Polar solvation energy ( kcal/mol)
|
SASA energy ( kcal/mol)
|
Cobimetinib
|
-53.90
|
-37.04
|
-77.48
|
66.01
|
-5.43
|
Refametinib
|
-16.49
|
-30.26
|
-7.68
|
25.54
|
-4.09
|
G545-0469
|
-13.00
|
-24.35
|
-3.25
|
17.69
|
-3.08
|
C387-1559
|
-17.10
|
-29.47
|
-31.9
|
47.65
|
-3.40
|
2.3 PRINCIPAL COMPONENT AND FREE ENERGY LAND-SCAPE ANALYSIS:
The PCA was performed to study the collective motion of MEK1 occupied in the conformational subspace in bound state with selected compounds (C387-1559, G545-0469), during the 100ns simulation. By utilizing the GROMACS utilities, i.e., gmx anaeig and gmxcovar, covariance matrix of eigenvector (EV) was generated and diagonalized to outline the essential subspace and then eigenvalues were extracted corresponding to each EV. The dynamic motion of MEK1 in the bound state with C387-1559 and G545-0469 obtained through the projection of EV1 and EV2 by the MEK1 Cα atoms is illustrated in Fig. 5. The plot shows that complex of MEK1 with C387-1559 and G545-0469 occupied notably different conformations within the reduced subspace. The most stable conformational ensembles of a protein which are significant in understanding the conformational changes underlying protein−ligand interactions are precisely portrayed by the FELs. The first two EVs of the MEK1 complexes with C387-1559 and G545-0469 were used for constructing and analyzing the FEL plots. Fig.6 displays the FELs where the deeper purple indicates the stable conformational states having lower energy. In case of C387-1559 there were two main free energy basins and in G545-0469 overall a single global minimum with the lowest energy was observed.
2.4 ADME ANALYSIS:
To study the drug-likeliness of the selected compound SwissADME tool from Swiss Institute of Bioinformatics was used. The screened results of ADMET were summarized in Table 5, revealing various descriptors categorised under properties such as physicochemical properties, lipophilicity, water solubility, pharmacokinetics etc. Values of physiochemical properties of both the selected compounds were within given range. In-fact the molecular weight of the benchmark compounds (Refametinib and Cobimetinib) is greater than 500 KDa and lies outside the acceptable molecular weight range whereas both the selected compounds are within the acceptable range. Values of most of the other physiochemical properties are comparable to the benchmark molecule. C387-1559 show no violation to Lipinski rule, unlike benchmark molecules which show one violation due to higher molecular weight. G545-0469 also shows one violation to Lipinski rule, like benchmark molecule but it’s violation is due to higher MLOGP value (4.46) than the acceptable value of MLOGP>4.15. Although the difference in the MLOGP value for G545-0469 with respect to acceptable value (>4.15) was very low (0.31). Both the selected compounds have same bioavailability score as the benchmark compound and are positive for drug-likeness in 3 out of five different rule-based filters, like the benchmark molecules. The ADME analysis suggested that both the selected compounds were potent molecules and were capable to be effective drug, as their physiochemical properties are comparable with an already in-use drug molecule (Refametinib and Cobimetinib) and in some cases are even better than the benchmark compound. This suggests that these potent molecules can reach their target in the body in sufficient concentration, and stay there in a bioactive form long enough for the expected biologic events to occur.
Table 5 Comparison of ADME results for top ranking potential inhibitors and benchmark.
Molecule
|
Refametinib
|
Cobimetinib
|
C387-1559
|
G545-0469
|
Formula
|
C19H20F3IN2O5S
|
C21H21F3IN3O2
|
C26H34N4O
|
C28H31FN4O2
|
MW
|
572.34 g/mol
|
531.31
|
418.57
|
474.57
|
#Heavy atoms
|
31
|
30
|
31
|
35
|
#Aromatic heavy atoms
|
12
|
12
|
16
|
18
|
Fraction Csp3
|
0.37
|
0.38
|
0.38
|
0.29
|
#Rotatable bonds
|
9
|
5
|
11
|
10
|
#H-bond acceptors
|
8
|
6
|
3
|
3
|
#H-bond donors
|
4
|
3
|
2
|
3
|
MR
|
117.62
|
122.92
|
130.6
|
141.83
|
TPSA
|
116.27 Ų
|
64.6
|
57.26
|
73.47
|
iLOGP
|
2.12
|
3.62
|
4.36
|
4.51
|
XLOGP3
|
2.78
|
3.88
|
6.2
|
5.09
|
WLOGP
|
5.57
|
4.28
|
5.56
|
5.63
|
MLOGP
|
2.64
|
4.08
|
3.8
|
4.46
|
Silicos-IT Log P
|
3.91
|
4.71
|
5.09
|
4.84
|
Consensus Log P
|
3.4
|
4.12
|
5
|
4.91
|
ESOL Log S
|
-4.83
|
-5.54
|
-6
|
-5.71
|
ESOL Solubility (mg/ml)
|
8.42e-03
|
1.52e-03
|
4.21e-04
|
9.26e-04
|
ESOL Solubility (mol/l)
|
1.47e-05
|
2.85e-06
|
1.01e-06
|
1.95e-06
|
ESOL Class
|
Moderately soluble
|
Moderately soluble
|
Moderately soluble
|
Moderately soluble
|
Ali Log S
|
-4.88
|
-4.93
|
-7.19
|
-6.38
|
Ali Solubility (mg/ml)
|
7.59e-03
|
6.18e-03
|
2.72e-05
|
2.00e-04
|
Ali Solubility (mol/l)
|
1.33e-05
|
1.16e-05
|
6.50e-08
|
4.21e-07
|
Ali Class
|
Moderately soluble
|
Moderately soluble
|
Poorly soluble
|
Poorly soluble
|
Silicos-IT LogSw
|
-7.12
|
-7.28
|
-9.13
|
-9.29
|
Silicos-IT Solubility (mg/ml)
|
4.34e-05
|
2.81e-05
|
3.08e-07
|
2.46e-07
|
Silicos-IT Solubility (mol/l)
|
7.59e-08
|
5.28e-08
|
7.36e-10
|
5.18e-10
|
Silicos-IT class
|
Moderately soluble
|
Poorly soluble
|
Poorly soluble
|
Poorly soluble
|
GI absorption
|
Low
|
High
|
High
|
High
|
BBB permeant
|
No
|
Yes
|
Yes
|
No
|
Pgp substrate
|
Yes
|
Yes
|
No
|
Yes
|
CYP1A2 inhibitor
|
No
|
No
|
No
|
No
|
CYP2C19 inhibitor
|
Yes
|
No
|
Yes
|
Yes
|
CYP2C9 inhibitor
|
No
|
No
|
No
|
No
|
CYP2D6 inhibitor
|
Yes
|
Yes
|
Yes
|
Yes
|
CYP3A4 inhibitor
|
Yes
|
Yes
|
Yes
|
No
|
log Kp (cm/s)
|
-7.82 cm/s
|
-6.79
|
-4.45
|
-5.58
|
Lipinski #violations
|
1
|
1
|
0
|
1
|
Ghose #violations
|
1
|
1
|
1
|
2
|
Veber #violations
|
Yes
|
No
|
Yes
|
No
|
Egan #violations
|
Yes
|
No
|
No
|
No
|
Muegge #violations
|
Yes
|
No
|
Yes
|
Yes
|
Bioavailability Score
|
0.55
|
0.55
|
0.55
|
0.55
|
PAINS #alerts
|
0
|
0
|
0
|
0
|
Brenk #alerts
|
1
|
1
|
0
|
0
|
Leadlikeness #violations
|
2
|
2
|
3
|
3
|
Synthetic Accessibility
|
4.48
|
3.92
|
3.39
|
3.45
|
2.5 EVALUATION OF PROTEIN-LIGAND INTERACTION:
The selected compounds C387-1559 and G545-0469 were further analysed at different concentrations by BLI technique for their real-time interaction with the MEK1 protein. Both the selected compounds as well as benchmark molecule Refametinib had significant binding for MEK1 protein. The Binding Affinity constant (KD) for Refametinib, C387-1559 and G545-0469 were 1μM, 80mM and 720nM, respectively. Interestingly, Cobimetinib did not project ideal curves with significant binding, but we were able to calculate its low binding affinity constant, i.e. 4.5mM (due to concentration dependent increase during first 5 seconds of the association). The binding affinities for the inhibitors have been listed in Table 6. The binding curves of the compounds at different concentration are given in Fig. 7. From the data it was observed that G545-0469 has the strongest binding affinity for MEK1 among all the compounds considered for this analysis. Between the benchmark molecules, Refametinib was a stronger binder of MEK1 as compared to Cobimetinib. In case of C387-1559, the KD value was higher than Refametinib and Cobimetinib implying its weaker binding to MEK1 with respect to the benchmark molecules. The high KD value of C387-1559 suggests that for achieving maximal occupancy of the binding site of MEK1 a relatively high concentration of a C387-1559 is required.
Table 6: Binding Affinity constant (KD) of tested compounds.
Compounds
|
Affinity constant (KD)
|
Refametinib
|
1μM
|
Cobimetinib
|
4.5mM
|
C387-1559
|
80mM
|
G545-0469
|
720nM
|
2.6 EVALUATION OF CYTOTOXICITY OF SELECTED COMPOUNDS AGAINST MEK1:
To investigate the cytotoxicity of C387-1559 (from ChemDiv Kinase library), G545-0469 (from ChemDiv Anticancer library), and benchmark compounds Cobimetinib and Refametinib in head and neck cancer cell line (Cal27), and prostate cancer call line (DU145), MTT assay was performed after 72 hours of drug treatment. The IC50 values obtained for C387-1559 and G545-0469 were 5µM and 6 µM respectively and the IC50 value of both the benchmark compounds Cobimetinib and Refametinib was 1µM. The cytotoxicity of these compounds in prostate cancer cell line DU145 showed IC50 values of 14.77 µM, 54.38 µM, 18.73 µM, and 25.16 µM for C387-1559, G545-0469, Cobimetinib, and Refametinib were, respectively (Table 7; Figure 8). The IC50 values of both selected compounds are much lower in case of CAL27 cell line as compared to DU145, suggesting their higher efficiency in human head and cancer cell line over human prostate cancer. Mostly the cancer chemotherapeutic agents which are currently available in the market are administered to patients at very high doses, which leads to severe side effects, therefore compounds with lower IC50 are needed to reduce the toxicity. These IC50 values of both selected compounds are much lower than 100 µM on both the cell lines which make them potential drug candidates capable of reducing dose-dependent cytotoxicity.
Table 7: IC50 values of tested compounds at 72 hours
Compounds
|
IC50 72 hrs
|
DU145
|
Cal27
|
Refametinib
|
25.16 µM
|
1 µM
|
Cobimetinib
|
18.73 µM
|
1 µM
|
C387-1559
|
14.77 µM
|
5 µM
|
G545-0469
|
54.38 µM
|
6 µM
|