PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx 1

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps 5 6

4

QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb)

7

G. Dicky John Davis a,⇑, A. Hannah Rachel Vasanthi b

3

8 9 10 11 1 2 3 8 14 15 16 17 18 19 20 21 22 23 24 25 26 27

a b

Department of Bioinformatics, Sri Ramachandra University, Porur, Chennai 600116, India Department of Biotechnology, Pondicherry University, Kalapet, Puducherry 605014, India

a r t i c l e

i n f o

Article history: Received 31 January 2015 Received in revised form 22 April 2015 Accepted 29 April 2015 Available online xxxx Keywords: Marine algae QSAR Anticancer activity Docking Protein kinase B In silico ADMET Callophycin A

a b s t r a c t Marine algae are prolific source of bioactive secondary metabolites and are found to be active against different cancer cell lines. QSAR studies will explicate the significance of a particular class of descriptor in eliciting anticancer activity against a cancer type. Marine algal compounds showing anticancer activity against six different cancer cell lines namely MCF-7, A431, HeLa, HT-29, P388 and A549 taken from Seaweed metabolite database were subjected to comprehensive QSAR modeling studies. A hybrid-GA (genetic algorithm) optimization technique for descriptor space reduction and multiple linear regression analysis (MLR) approach was used as fitness functions. Cell lines HeLa and MCF-7 showed good statistical quality (R2  0.75, Q2  0.65) followed by A431, HT29 and P388 cell lines with reasonable statistical values (R2  0.70, Q2  0.60). The models developed were interpretable, with good statistical and predictive significance. Molecular descriptor analyses revealed that Baumann’s alignment-independent topological descriptors had a major role in variation of activity along with other descriptors. Incidentally, earlier QSAR analysis on a variety of chemically diverse PKBa inhibitors revealed Baumann’s alignment-independent topological descriptors that differentiated the molecules binding to Protein kinase B (PKBa) kinase or PH domain, hence a docking study of two crystal structures of PKBb was performed for identification of novel ATP-competitive inhibitors of PKBb. Five compounds had a good docking score and Callophycin A showed better ligand efficiency than other PKBb inhibitors. Furthermore in silico pharmacokinetic and toxicity studies also showed that Callophycin A had a high drug score (0.85) compared to the other inhibitors. These results encourages discovering novel inhibitors for cancer therapeutic targets by screening metabolites from marine algae. Ó 2015 Published by Elsevier B.V.

29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

51 52

1. Introduction

53

Cancer is a complex disease of global concern. In 2008, about 70% of cancer deaths took place in countries with low- and middle income. The American Cancer Society estimated a total of 1,665,540 new cancer cases and 585,720 cancer deaths are projected to occur in the United States alone in 2014 (Siegel et al., 2014). In analogy, Cancer deaths continue to keep rising worldwide and in 2030 it is projected to be a cause for 13.1 million deaths (Ferlay et al., 2010). Cancer-relevant genes have been intensively studied and numerous therapeutic targets which play a vital role in cancer pharmacology have been elucidated. Protein kinase B (PKB/Akt) is a foremost protein mediating the proliferation and involving several pathways that are essential for the growth of cancer, which makes it a unique therapeutic

54 55 56 57 58 59 60 61 62 63 64 65

⇑ Corresponding author.

target for treating cancers of multiple origins. Akt consists of three different cellular isoforms, namely, Akt1 (PKBa), Akt2 (PKBb), and Akt3 (PKBc). All three Akt isoforms have the ability to transform into cancerous cells in vitro. However, Akt2 is the major isoform found to be amplified or over expressed in human cancer which has been observed in pancreatic tumours (10%), hepatocellular carcinomas (40%) and colorectal cancers (57%) (Hers et al., 2011). Several functionally important regions are present in the protein; specifically the central kinase domain has a classical kinase ATP-binding site that can be explored as binding pockets for small molecule inhibitors. Virtual screening have been employed in previous studies to identify potent inhibitors using the crystal structure of PKBb in complex with glycogen synthase kinase-3b peptide (GSK-3b) and 50 -adenylylimidodiphoshate (AMP–PNP) (Vyas et al., 2013). Medina-Franco et al. (2009) reported the discovery of a novel competitive inhibitor for ATP binding site by search in a different area of chemical space for selective and potent inhibitors.

http://dx.doi.org/10.1016/j.ejps.2015.04.026 0928-0987/Ó 2015 Published by Elsevier B.V.

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 2 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121

G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx

Marine organisms constitute an important source of novel molecules for new drug discovery and drug development research of which 25% are from algae. Seaweeds have a distinct evolution on their biosynthetic pathways that frequently yield complex molecules with no counterparts in the terrestrial environment. Secondary metabolites from these algae are predominantly sesquiterpenes, diterpenes, triterpenes and C15-acetogenins characterized by the presence of halogen atoms in their chemical structures. An online database – Seaweed Metabolite Database (SWMD) was developed by us to share organized information about marine algal secondary metabolites and their geographical location obtained by literature data mining (www.swmd.co.in). It presently has 1055 compounds from red, green and brown macroalgae describing the method of compound extraction, chemical description of the compound and biological activity of the compound i.e., antibacterial, antimalarial, antioxidant and cytotoxic activities (Davis and Vasanthi, 2011). Quantitative Structure–Activity Relationship (QSAR) modeling is a ligand-based drug design method for both exploring and exploiting the relationship between chemical structure and its biological action (Liao et al., 2011). To predict the activities of anticancer compounds, quantum chemical descriptors like molecular orbital, dipole moment, charge, etc. and molecular property descriptors like hydrophobic, steric coefficient, etc. have been applied to develop 2D QSAR models (Chen et al., 2007; Zhang et al., 2007). In-vitro evaluation of biological activity can be performed on a number of cell lines for a specific cancer type, but the results of the evaluation would vary based on the cell line employed for the assay. Consequently to design novel and potent anticancer compounds, all the experimental data has to be considered from all the cell lines, such analyses involving all the QSAR descriptors would pave way for predictive models highlighting the importance of a particular class of descriptor in modeling anticancer activity against a specific cancer type (Bohari et al., 2011). Such extensive QSAR studies against many diverse cancer cell lines which are statistically robust are warranted. Hence, a comprehensive 2D QSAR modeling study was performed in the present study using the compounds in SWMD that have cytotoxic activity.

122

2. Methodology

123

SWMD which has 1055 entries, 245 compounds (23%) has documented anticancer activity against 43 different cell lines. For continuous response variable (activity), the total minimum number of compounds should be no less than 40 as the number of compounds in the training set should be at least 20, and about 10 compounds should be in each of the test and external evaluation sets (Tropsha, 2010). Therefore, the dataset taken for the study has 157 compounds having cytotoxic activity against six different cancer cell lines namely MCF-7 (Human breast adenocarcinoma), A431 (Human epithelial carcinoma), HeLa (Human cervical adenocarcinoma), HT-29 (Human colon adenocarcinoma grade II), P388 (Murine leukemia) and A549 (Human lung epithelial adenocarcinoma) cells, each having more than 40 compounds (Table 1). The

124 125 126 127 128 129 130 131 132 133 134 135

Table 1 Cell lines against which their anticancer activity was reported in SWMD along with the number of molecules in each cell lines. S. No. 1 2 3 4 5 6

Cell lines

Cancer type

A549 MCF7 HT29 P388 HeLa A431

Human Human Human Murine Human Human

lung epithelial adenocarcinoma breast adenocarcinoma colon adenocarcinoma grade II leukemia cervical adenocarcinoma epithelial carcinoma

# of compounds 72 65 62 54 50 41

dataset consists of chemical diverse compounds which include sesquiterpenes, diterpenes, triterpenes, sterol and acetogenins that are usually characterized by the presence of one or more halogen atoms in their structures (Fig. 1). The structure of all the 157 compounds and their experimental cytotoxic activity against six cell lines listed in SWMD along with its accession numbers are presented in Additional file (Tables S1). Measurement of cytotoxic activity is expressed as half maximal (50%) inhibitory concentration of a substance (IC50) and values are expressed in nanomolar (nM – 109) and micromolar (lM – 106) levels. The values were converted to the pIC50 scale (log IC50) to predict the narrow value wherein higher values indicate exponentially greater potency. The formula for micromolar conversion of IC50 values to pIC50 values is

136

pIC50 ¼  logðIC50  106 Þ

152

The pIC50 values were used as the dependant variables to construct the QSAR model.

153

2.1. Calculation of molecular descriptor

155

QSAR modeling involves the use of software to sketch chemical structures and calculate the descriptors to build predictive models. The molecules to be used in the study were taken from SWMD and 3D structures were generated using MarvinSketch 2011 (v.5.4.1.1) from ChemAxon. The resultant structures are crudely optimized using a molecular mechanics method within MarvinSketch (Csizmadia, 2000). Since some molecular descriptors also require information about the electronic environment of the molecule, the molecules were also optimized for electronic properties with Molecular Orbital PACkage (MOPAC) using WinMopac 7.21 software (Stewart, 1990). Descriptors for the present study were obtained using Vlife MDS 3.5; 239 descriptors based on the physicochemical properties of the molecule and 391 alignment independent descriptors considering topology of the molecule was used (VLife sciences technologies).

156

2.2. Selection of relevant descriptors

171

To select the most important descriptors and to decrease the overfitting/overtraining risk, Feature selection techniques were applied to decrease the model complexity using open source software – WEKA (Waikato Environment for Knowledge Analysis) from The University of Waikato (v.3.6.6) (Hall et al., 2009). The dataset consists of 157 anticancer compounds from SWMD for six cancer cell lines models with 630 descriptors each. Filters were applied in the first step to remove compounds and descriptors with missing or null values. Either one of the correlated descriptors having more than 90% correlation in their values were identified and removed. To search the descriptor subspace, linear forward selection was used wherein; correlation-based feature selection (CFS) algorithm in WEKA was employed to evaluate the descriptors. Efforts taken to keep the descriptors employed in the models virtually orthogonal to each other include changing and refining the descriptors that were highly inter-correlated in a given model. This pre-screening gave a quality-assured dataset of compounds which are used for further analysis. Selection of descriptors was done using the hybrid optimization technique developed by Rogers and Hopfinger wherein a wrapper of genetic algorithms (GA) for feature selection and multiple linear regression (MLR) as regression technique was employed (Rogers and Hopfinger, 1994). The GA was employed for searching the descriptor subspace, whereas the MLR was used for fitness evaluation to determine the significant descriptors. GeneticSearch in

172

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

137 138 139 140 141 142 143 144 145 146 147 148 149

150

154

157 158 159 160 161 162 163 164 165 166 167 168 169 170

173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx

3

Fig. 1. Structurally diverse cytotoxic compounds in Seaweed Metabolite Database (SWMD).

198 199 200 201 202 203 204 205 206 207 208

WEKA uses a simple genetic algorithm (Goldberg et al., 1993) and default values were selected for the GA parameters, such as 20 generations, population size of 20, crossover probability of 0.6 and mutation probability of 0.033. The attribute selection in WEKA employed for MLR uses the M5’s method where the attributes are removed stepwise given by the Akaike information criterion; the ones with the smallest standardised coefficient is removed until no improvement in the estimate of the error is observed (Witten and Frank, 2005). The extent of the coefficients discloses the degree of influence of the corresponding molecular descriptors on the given property.

209

2.3. Validation of QSAR models

210

The regression models are developed by dividing the dataset into multiple chemically diverse training and test sets with a rational approach based on Sphere Exclusion (SE) algorithm (Golbraikh and Tropsha, 2002). To evaluate the performance of the QSAR model, Leave-one-out cross validation (Q2) is carried out to obtain the optimal number of components (N) and the correlation coefficient R2. To evaluate the performance without any bias, two independent test set was made and the remaining compounds were used for model development. Models were developed using the training set while the test set was not used but serves to test the extrapolative ability of final models. For each compound in the training set, a correlation equation was derived with descriptors. The observed and predicted activity with residuals and descriptor values for all the developed models are presented in Additional file (Tables S2–S13). The predicted biological activities of untested compounds from their molecular structures are also presented in the above said tables. The two independent test set are presented at the end of the table marked with asterisks. The MLR regression equations for each of the table are also presented in Additional file (Tables S2–S13).

211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229

2.4. Docking of protein kinase (PKB b) inhibitors

230

Structures of PKBb were taken from Protein databank that were quite similar; (PDB ID: 2UW9 and 2JDR). The backbone RMSD for the whole protein was only 0.36 Å and 0.3 Å for the binding site residues that are within 5 Å of ligand (MedinaFranco et al., 2009). Bioactive conformation was simulated for 2UW9 and 2JDR using Molegro Virtual Docker (MVD v3.0.0) which implements evolutionary algorithms for molecular docking simulations and docking was performed by Moldock function (Thomsen and Christensen, 2006). For both PDB structures, during docking water molecules were removed and co-crystal inhibitors were ignored. From the docking wizard, ligands were selected from SWMD; 157 cytotoxic compounds that were used in the QSAR study. The protein and ligands molecules were prepared at first and bonds, bond orders, explicit hydrogens, charges and flexible torsions, were assigned if they were missing using the MVD program. MVD was used for active site (pocket) detection on PKBb protein. The ATP binding site was defined as active site box having volumes of 359 Å3 and 388 Å3 for 2UW9 and 2JDR, respectively. Further, the binding site was delineated by choosing all atoms within 10 Å of the analogous crystallographic ligand with the cavity detection mode active and using default parameters. The Ignore distant atoms option was applied to ignore atoms far-off from the binding site. The search algorithm taken was Moldock SE and the number of runs taken as 10 and max iterations were 2000 with population size of 50 and with an energy threshold of 100. The energy penalty was set to 100, RMSD threshold was 2.00 and RMSD calculation by atom ID (fast) were set. After the docking simulation was completed the poses generated were sorted by rerank score (De Azevedo, 2010). Results of the top ligands whose rerank score >100 were selected.

231

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 4

G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx

2.5. In silico ADMET analysis

0.9

264

Successful drug discovery necessitate good pharmacokinetics and no toxicity of lead structures which would make it more drug-like. In silico tools on absorption, distribution, metabolism, excretion and toxicity (ADMET) screening of candidate drugs help to reduce the risk of late-stage attrition and optimize the most promising compounds. Therefore, the ligands with top docking score were further screened for druglikeliness, drug score and toxicity characteristics using the program OSIRIS property explorer (v.2.0) (Sander et al., 2009).

0.8

265 266 267 268 269 270 271 272

Correlation coefficient

263

0.7

A431

3. Results and discussion

274

The predictive QSAR models were built for six different cancer cell lines with experimental data from 157 compounds, using independent and with minimum number of descriptors. The distribution of IC50 values among the six cell lines was seen to differ from one compound to another and is shown Fig. 2. The criterion for selecting the best model was based on the correlation coefficient values taken from the correlation of approximately 630 descriptors in different combinations. To evaluate the effect of the number of descriptors on the correlation coefficient values, all the models were tested on training set by correlating 1–10 descriptors separately and presented in Fig. 3. It was observed that in various models, four descriptors were adequate for obtaining a good correlation and in most of the models using more than four descriptors made only a little variation on the statistical quality. While in most of the cases, seven or more descriptor-based models gave high correlation and crossvalidation coefficient values, nevertheless this may be false and so will not be accurate in prediction of IC50 values further. Various statistical measures were adapted in the present study to assess and compare the predictive power and the stability of QSAR models, which is also widely applied for evolution of a significant model. As four descriptors were sufficient for getting a good correlation, all the training set had more than 20 numbers of molecules. Square of the correlation coefficient (R2) represents the statistical significance of the model, herein all the models in the study were inferred significant if R2 is greater than 0.7. While Q2 is the cross-validated R2, a measure of the quality of the QSAR model was inferred significant if Q2 is greater than 0.5. Fischer statistics (F) is the ratio between explained and unexplained variance for a given number of degrees of freedom. This indicates a factual correlation or the significance level for QSAR models. Higher the F-test more significant is the model. The average of absolute difference between experimental and predicted IC50 values is average residual (AE). Lower the AE, more significant is the model.

276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307

500

A549

400

HeLa

350 300

HT29

250

MCF-7

200

P388

150 100 50 0

0

10

HT29 MCF7

0.4

P388 1

2

3

4

5

6

7

8

9

10

Number of Descriptors

A431

450

IC50 values(µM)

275

HeLa

0.5

0.3

273

A549

0.6

20

30

40

Number of Compounds Fig. 2. Distribution of IC50 value among cell lines.

50

60

Fig. 3. Effect of number of descriptors on the correlation coefficient.

Regression equation for the QSAR models is presented in the regression summary along with name of the cell lines and types of cancer (Table 2). To identify that the developed models are valuable, AE was calculated for both training and test set of compounds and it was found that all the models have a lower AE values (0.15–0.35), thus have the capability to establish the relationship between the structure and activity. The QSAR models for HeLa (Cervical) and MCF-7 (Breast) cell lines can be used for the prediction as it exhibits good statistical quality (R2  0.75, Q2  0.65) and considered valuable for the available class of compounds. The statistical quality of A431 (Epithelial), HT29 (Colon) and P388 (Leukemia) cell lines are also reasonable (R2  0.70, Q2  0.60), and extra care is required before utilizing these models for the prediction. However, the statistical quality of A549 (Lung) cell lines cannot be used for the prediction because of the insignificant statistical results obtained for this model (R2 = 0.67, Q2 = 0.50). The reason for poor result in A549 cell lines is probably due to the involvement of diverse compound types in this model. The increase in the number of descriptors for A549 does not improve the quality of the model (with 10 descriptors R2  0.78) and thus indicates that the currently used four descriptors are good enough for developing the structure–activity relationship for this model. Statistical quality of QSAR depends on the distribution of IC50 values of compounds among the cell lines where IC50 values are distributed in broad range would show good statistical quality and vice versa. IC50 values for HeLa and MCF-7 cell lines are for a broad range wherein the predictions showed good statistical quality (Fig. 2). The range of IC50 values for A431, HT29, P388 and A549 cell lines are reasonable and so also the QSAR prediction. The quality of the QSAR model was poor for A549 cell lines as the compounds had diverse structures as shown in Additional file (Table S14). Compounds that do not fit in the developed QSAR model are called as outliers. Compound RL018 (2-tridecyl-2-heptadecenal) is an outlier in A431, HeLa and MCF7 QSAR models as it has an aliphatic hydrocarbon whereas the others are aromatic compounds. The experimental and predicted IC50 values for training and test set (1) and the number compounds in both the sets along with their average residual for all QSAR models are represented in Fig. 4. This figure clearly demonstrates that the compounds of the test set are closer to the line compared with the compounds of training set. The applicability of generated QSAR models was rigorously validated by dividing another independent test set. The statistical performance of the second test set was similar to that of the first test set. Both the test sets revealed similar statistical performance indicating that the developed models were satisfactory. The observed and predicted activities with residuals and number compounds in training and test set are presented in

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 5

G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx Table 2 Regression summary for all the QSAR models . Regression equation

R2

Q2

AE

F

1

=0.2967 ⁄ 5ChainCount + 0.2105 ⁄ SsssCHE-index 0.1817 ⁄ T_2_Br_5 + 0.3143 ⁄ T_O_Br_4 + 4.1884

0.74 0.73

0.60 0.51

0.15 0.16

13.76 12.67

17

0

=0.1969 ⁄ SaaCHE-index + 1.1969 ⁄ T_2_Cl_1 + 0.5973 ⁄ chiV3Cluster 3.2011 ⁄ SAMostHydrophilic + 4.5436

0.67 0.65

0.50 0.47

0.23 0.25

21.16 18.95

7

8

1

=0.0039 ⁄ 5PathCount 0.0929 ⁄ Xlog P + 0.5874 ⁄ T_O_Br_6 + 0.2666 ⁄ 3ChainCount + 4.0602

0.79 0.78

0.67 0.71

0.21 0.19

27.83 25.74

42

8

12

0

=0.3328 ⁄ SdssCE-index 9.9514 ⁄ SAAverageHydrophilicity + 0.1929 ⁄ T_2_2_4 + 9.7350 ⁄ chiV6chain + 3.5298

0.73 0.78

0.58 0.61

0.35 0.32

25.39 32.46

MCF7 (Breast)

31

6

26

1

=4.8683 ⁄ chiV5chain + 0.2718 ⁄ 6ChainCount + 0.1071 ⁄ T_O_Br_4 + 0.0674 ⁄ T_2_2_5 + 3.9220

0.74 0.77

0.67 0.69

0.22 0.22

18.40 21.16

P388 (Leukemia)

39

9

6

0

=18.4947 ⁄ chi6chain 0.2150 ⁄ T_O_O_2 –0.0535 ⁄ T_2_O_4 –0.2514 ⁄ 3ChainCount + 3.4853

0.73 0.72

0.67 0.66

0.29 0.32

22.98 21.52

Cell line (Type)

# compounds TR

TS

PD

O

A431 (Epithelial)

24

4

12

A549 (Lung)

46

9

HeLa (Cervical)

34

HT29 (Colon)

Cell line with type of cancer in parenthesis, regression summary (regression equation, correlation coefficient (R2), cross validation coefficient (Q2), average residual (AE) and number of outliers (O) and number of compounds (training set (TR), test set (TS) and predicted set (PD)) in various cell lines based QSAR models for both the test set.

A431 Cells

R2 Q2 AETR AETS

4.6

0.74 0.60 0.15(24) 0.25(4)

5.9 5.4

6

A549 Cells

R2 Q2 AETR AETS

0.67 0.49 0.23(46) 0.40(9)

5.5 5

4.9

4.1

3.5

7

4

4.5

5

HT29 Cells

5.5

R2 0.73 0.58 Q2 AETR 0.35(42) AETS 0.42(8)

6.5 6 5.5

3.9

5

3.5

4.5

5.5

6.5

MCF7 Cells

3.5

7.6

R2 0.74 0.67 Q2 AETR 0.22(31) AETS 0.32(6)

7.1 6.6

3.3

3.8

4.3

4.8

5.3

5.8

6.3

P388 Cells

R2 0.73 0.67 Q2 AETR 0.29(39) AETS 0.50(9)

6.1 5.6

4.5

4

5.1

4 3.5

0.73 0.58 0.21(34) 0.35(7)

4

4.5

5

R2 Q2 AETR AETS

4.5

4.4 3.6

HeLa Cells

3

4

5

6

7

3.5

3.3

3.8

4.3

4.8

5.3

5.8

4.6

4.2

5.2

6.2

7.2

Fig. 4. Plot between experimental and predicted IC50 values for training and test set (1) QSAR models. Correlation coefficient (R2), Cross-validation coefficient (Q2), Average residual Training set (AETR), Average residual Test set (AETS) & number of compounds in training and test set are shown in brackets.

356 357 358 359 360 361 362 363 364 365 366 367 368 369

Fig. 5, for all the developed models for the second set of test compounds. In the developed QSAR models, 22 descriptors (14 Physicochemical and 8 Baumann’s Alignment independent) were used in different combinations. The particulars of all the 22 descriptors, its type and incidence in the models are depicted in Fig. 6. The details of the descriptors involved in the study and their occurrence in the QSAR models is shown in Additional file (Table S14). The inter-correlation of the descriptors appeared in all the developed models were taken into account, and the descriptors were found to be reasonably orthogonal and presented in Additional file (Table S15). All the models have identified alignment independent descriptors as vital descriptors. The ‘atom and bond count’ descriptors especially number of Oxygen, Bromine

and Chlorine atoms is chosen in most models. It is well known that the seaweed metabolites are biologically active due to the high degree of halogenations thereby exhibiting antibacterial, antifungal, antiviral, anti-inflammatory, antiproliferative, cytotoxic, antifouling, antifeedant, ichthyotoxic, and/or insecticidal activity (Lhullier et al., 2010). The same is reflected in the obtained descriptors as halogenation of molecules increases the cytotoxic activity and is also proved. Chain path count descriptors (such as 3ChainCount, 5ChainCount, 6ChainCount), retention index and atomic valence connectivity index also were identified in the models. The maximum six membered rings positively influence the activity and percentage contribution in most of the models (Fig. 7). The Hydrophobicity SlogpA descriptors which is the hydrophilic value on the Van der Waals surface of molecules decreases

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

370 371 372 373 374 375 376 377 378 379 380 381 382 383

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 6

G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx

A431 Cells

4.7

R2 Q2 AETR AETS

6.7 6.2 5.7

3.5

4

4.5

HT29 Cells

5 4.5

4.4

4

3.5

4.5

5.5

6.5

R2 0.76 0.69 Q2 AETR 0.22(31) AETS 0.34(6)

5

3.5

HeLa Cells

R2 Q2 AETR AETS

0.78 0.72 0.19(34) 0.41(7)

3.3

7.7

MCF7 Cells

5.5

R2 0.78 0.61 Q2 AETR 0.32(42) AETS 0.54(8)

5.5

4.9

3.9

5

4.3

5.3

6.3

P388 Cells

R2 0.71 0.66 Q2 AETR 0.32(39) AETS 0.44(9)

7.2 6.7 6.2

4.5

5.2

5.7

4.7

4

4.2 3.7

R2 0.65 0.46 Q2 AETR 0.25(46) AETS 0.30(9)

5.4

4.2

3.7

A549 Cells

5.9

0.73 0.51 0.16(24) 0.20(4)

3.4

4.4

5.4

6.4

7.4

3.5

5.2

3.3

3.8

4.3

4.8

5.3

5.8

4.7

4.2

5.2

6.2

7.2

Fig. 5. Plot between experimental and predicted IC50 values for training and test set (2) QSAR models. Correlation coefficient (R2), Cross-validation coefficient (Q2), Average residual Training set (AETR), Average residual Test set (AETS) & number of compounds in training and test set are shown in brackets.

36% ALIGNMENT INDEPENDENT T_2_2_4, T_2_2_5 T_2_Br_5 , T_2_Cl_1 T_2_O_4 , T_O_Br_4 T_O_Br_6 , T_O_O_2

14% Estate contributions SaaCHE-index SdssCE-index SsssCHE-index

4% Individual XlogP

14% Chain Path count 3ChainCount 5ChainCount 6ChainCount

4% Path Count 5PathCount

5% Chi Chain chi6chain 5% Cluster chiV3Cluster

9% Hydrophobicity SlogpA SAAverageHydrophilicity SAMostHydrophilic

9% Chiv Chain chiV5chain chiV6chain

PHYSICOCHEMICA Fig. 6. Classification of various descriptors involved in QSAR model. 384 385 386 387 388 389 390 391 392 393 394 395 396 397

the anticancer activity of compounds. The other descriptors include Estate contributions which are the Electrotopological state indices of valency of C atoms and bond order. QSAR analyses on a wide variety of structurally diverse Akt1 inhibitors revealed the central role of Baumann’s alignment independent topological descriptors beside additional descriptors such as the number of hydrogen bond donors, hydrogen bond acceptors, rotatable bonds and aromatic oxygen (SaaOcount) together with alkene carbon atom type (SdsCHE-index) and molecular branching (chi3Cluster), in accomplishing differential activity (Ajmani et al., 2010). Further in the same study, Group-based QSAR analyses showed that for achieving highly potent Akt1 inhibitors, chemical variations such as presence of hetero-aromatic ring, flexibility, polar surface area and fragment length present in the hinge

binding fragment are highly influential. In addition, the study also reported a three descriptors model using k-nearest neighbor classification to differentiate molecules binding to Akt1 kinase or PH domain and reported the main role of oxygen (SssOE-index) and aromatic carbon (SaaCHE-index and SaasCE-index) atoms electro-topological environment in binding to Akt1 kinase. Structure–activity analysis of Akt1 kinase reported by Ajmani et al. (2010) correlated well with the descriptor attributes of our present QSAR study of anticancer compounds from seaweeds. Wherein, the anticancer activity was contributed by Baumann’s alignment independent topological descriptors along with Oxygen, Bromine and Chlorine atoms and aromatic carbon (SaaCHE-index) atoms. Also as Akt isozymes are approximately 80 percent identical and have a high degree of overall homology,

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

398 399 400 401 402 403 404 405 406 407 408 409 410 411

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 7

G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx

A431 Cells 30.94

T_O_Br_4

26.77

-40

-20

-32.9 20

40

-50

-30

HT29 Cells

10.33

SAAverageHydroph ilicity

-42.61

-60

-40

-20

0

20

10

30

-20

T_2_2_5

37.86

T_O_Br_4

41

40

60

-60

-40

-20

-11.38

T_2_O_4

-19.97

T_O_O_2

-21.52

-37.38 0

20

3ChainCount

48.19

chiV5chain

0

40

60

80

P388 Cells

6ChainCount 38.79

SdssCE-index

69.77

5PathCount

MCF7 Cells 30.18

chiV6chain

-8.86

XlogP

-40.8

-10

64.32

T_O_Br_6

12.21

SaaCHE-index

0

T_2_2_4

15.61

T_2_Cl_1

39.44

3ChainCount

chiV3Cluster

SsssCHE-index 5ChainCount

-9.29

SAMostHydrophilic

-17.38

T_2_Br_5

HeLa Cells

A549 Cells

59.15

chi6chain

20

40

60

-40

-20

0

20

40

60

80

Fig. 7. Percentage contribution of each descriptor in developed QSAR model explaining variation in the activity.

Table 3 Docking results of PKBb inhibitors. Ligand name

PDB: 2UW9 RL378 RG009 RG004 PDB: 2JDR RL078 Laurenmariannol RC002 Callophycin A

Molecular formula

MolDock score

Rerank score

No of H bond

H bond energy

Ligand efficiency LE1

LE2

Interacting residues

C24H35BrO6 C27H42O4 C27H44O3

139.13 135.71 130.26

104.23 103.21 103.01

8 8 7

9.39 14.15 10.42

4.49 4.38 4.34

3.36 3.33 3.43

Ala232, Glu230, Glu230, Thr213, Thr292, Asp293, Lys181, Lys277 Ala232, Ala232, Glu230, Glu230, Thr213, Asn280, Asp293, Lys277 Ala232, Ala232, Glu230 Glu230, Thr213, Lys277, Glu279

C30H53BrO7

156.93

100.91

8

12.91

4.13

2.66

Ala232, Glu230, Glu230, Thr213, Thr292, Thr292, Asp293, Asp293

C19H18N2O3

123.00

106.14

5

9.81

5.13

4.42

Ala232, Glu230, Asp293, Glu200, Lys181

Ligand Efficiency 1 (LE1) – Moldock score divided by Heavy Atoms count and Ligand Efficiency 2 (LE2) – rerank Score divided by Heavy Atoms count. 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

the QSAR study was further extended to identify novel Akt2 inhibitors. The 157 cytotoxic compounds from SWMD were docked at the ATP binding site of 2JDR and 2UW9 wherein several molecules showed a better Moldock score than the co-crystal inhibitor. A low degree of consensus of Moldock score was observed with each crystal structure (2UW9 and 2JDR) between the top ranked scoring molecules. In general among the top 10 ranked compounds docked with Moldock only one molecule was found in common in both crystal structures. The selection of molecules were based on a high docking score with rerank score >100 and a characteristic feature observed in several PKBb inhibitors which is the ability to make hydrogen bonds with Glu230 and Ala232 (Saxty et al., 2007; Vyas et al., 2013; Medina-Franco et al., 2009). Docking results in the present study suggested that compound from marine red algae RL378, RG004 and RG009 had a good docking score for 2UW9 whereas RL078 and RC002 had a good docking score for 2JDR (Table 3). RC002 (Callophycin A), tetrahydro-b-carboline was isolated from the methanol extract of red algae Callophycus oppositifolius and was shown to mediate anticancer and cytotoxic effects on a series of human tumour cell lines

and a normal mammalian cell line (Ovenden et al., 2011). Five Hbonds were formed between ligand RC002 and protein PKBb (2JDR) with a rerank score of 106.14 (Fig. 8). The oxygen atom (hydroxyl) of ligand RC002 formed two H-bonds, one with –NH2 of Ala232 and another carbonyl oxygen atom of Glu230. The oxygen atom of carboxylic acid moiety formed two H-bond with NH2 of Lys181 and another with carboxylic acid moiety oxygen atom of Glu200. Oxygen atom (hydroxyl) of carboxylic acid formed H-bonds with –NH2 of Asp293. RC002 which has a good docking score and ligand efficiency better than the other ligands studied to be an active PKBb inhibitor hit and confirms the affinity with Glu230 and Ala232. Pharmacokinetic properties and toxicities were predicted for all the five ligands that showed good docking results using OSIRIS property explorer and are shown in Table 4. A compound’s hydrophilicity is measured as the logarithm of its partition coefficient between n-octanol and water log (coctanol/cwater), given as log P value of a compound. Calculated Log P (cLog P) is given by OSIRIS. Drug absorption and distribution characteristics depend on its aqueous solubility, estimated as log S value. The program calculates the drug score which is the compound’s overall potential to qualify

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 8

G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx

Fig. 8. Interaction of Callophycin A (RC002) in the ATP site of PKBb (PDB Code 2JDR). Docking studies showing 5 hydrogen bond interactions of Callophycin A with 2JDR at Ala232, Glu230, Asp293, Glu200, Lys181. Hydrogen bonds are shown in green. Atom colors: black, carbon; red, oxygen; blue, nitrogen. The diagram was created with the program LigPlot+ (v.1.4.5) Laskowski and Swindells, 2011. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Table 4 In silico ADMET prediction of PKBb inhibitors using OSIRIS property explorer. Physicochemical and ADMET parameters/properties

RL378

RG004

RG009

RL078

RC002

Mutagenic Tumorigenic Irritant Reproductive effective cLog P Solubility Molecular weight Drug likeness Drug score

No No Yes No 2.65 4.35 498 1.31 0.22

Yes Yes Yes No 5.98 5.89 416 2.52 0.04

Yes Yes Yes No 5.32 5.54 430 1.79 0.05

No No No No 3.22 5.58 604 10.96 0.19

No No No No 0.67 2.91 322 2.44 0.85

optimizing the molecule as PKBb inhibitor. Further development of Callophycin A will include exploring the structure–activity relationship required to obtain the desired PKB selectivity. These results further encourages discovering newer PKBb inhibitors for the treatment of cancer and screening metabolites of marine algae for particular beneficial biological effects which will undoubtedly pay off in the future. The present study has shown a roadmap for further exploiting the chemoinformatics approach in cancer drug discovery using various molecular targets for the development of novel anticancer agents from marine algae. The same approach can further be used for drug discovery and development from the ocean for other diseases as well.

465

Appendix A. Supplementary material

477

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejps.2015.04.026.

478

References

480

Ajmani, S., Agrawal, A., Kulkarni, S.A., 2010. A comprehensive structure-activity analysis of protein kinase B-alpha (Akt1) inhibitors. J. Mol. Graph. Model. 28, 683–694. Bohari, M.H., Srivastava, H.K., Sastry, G.N., 2011. Analogue-based approaches in anti-cancer compound modelling: the relevance of QSAR models. Org. Med. Chem. Lett. 1, 3–15. Chen, J., Shen, Y., Liao, S., Chen, L., Zheng, K., 2007. DFT-based QSAR study and molecular design of AHMA derivatives as potent anticancer agents. Int. J. Quantum Chem. 107, 1468–1478. Csizmadia, F., 2000. JChem: Java applets and modules supporting chemical database handling from web browsers. J. Chem. Inf. Comput. Sci. 40, 323–324.

481 482 483 484 485 486 487 488 489 490 491

466 467 468 469 470 471 472 473 474 475 476

Solubility measured in mol/liter is estimated as log S value.

460

for a drug by adding the total values of drug-likeliness, cLog P, log S, molecular weight, and toxicity risks. Thus the drug score judges the compound’s overall potential as a drug candidate wherein RC002 had the highest drug score (0.85) among the five compounds. Docking studies at the ATP-binding site of PKBb and the in silico ADMET properties suggest RC002 (Callophycin A) as a potential anticancer drug candidate.

461

4. Conclusion

462

A novel PKBb inhibitor with a unique scaffold compared to existing published PKBb inhibitors was identified by virtual screening from marine algae. This would be a starting point for an

454 455 456 457 458 459

463 464

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

479

PHASCI 3255

No. of Pages 9, Model 5G

29 April 2015 G. Dicky John Davis, A. Hannah Rachel Vasanthi / European Journal of Pharmaceutical Sciences xxx (2015) xxx–xxx 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522

Davis, G.D.J., Vasanthi, A.H.R., 2011. Seaweed metabolite database (SWMD): a database of natural compounds from marine algae. Bioinformation 5, 361–364. De Azevedo, W.F., 2010. MolDock applied to structure-based virtual screening. Curr. Drug Targets 11, 327–334. Ferlay, J., Shin, H.R., Bray, F., Forman, D., Mathers, C., Parkin, D.M., 2010. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int. J. Cancer 127, 2893– 2917. Golbraikh, A., Tropsha, A., 2002. Predictive QSAR modeling based on diversity sampling of experimental datasets for the training and test set selection. Mol. Divers. 5, 231–243. Goldberg, D.E., Deb, K., Kargupta, H., Harik, G., 1993. Rapid, accurate optimization of difficult problems using messy genetic algorithms. In: Proceedings of the Fifth International Conference on Genetic Algorithms, S. Forrest, Ed. pp. 56–64. Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H., 2009. The WEKA data mining software: an update. ACM SIGKDD Explor. Newslett. 11, 10– 18. Hers, I., Vincent, E.E., Tavaré, J.M., 2011. Akt signalling in health and disease. Cell. Signal. 23, 1515–1527. Laskowski, R.A., Swindells, M.B., 2011. LigPlot+: multiple ligand-protein interaction diagrams for drug discovery. J. Chem. Inf. Model. 51, 2778–2786. Lhullier, C., Falkenberg, M., Ioannou, E., Quesada, A., Papazafiri, P., Horta, P.A., Schenkel, E.P., Vagias, C., Roussis, V., 2010. Cytotoxic halogenated metabolites from the Brazilian red alga Laurencia catarinensis. J. Nat. Prod. 73, 27–32. Liao, C., Sitzmann, M., Pugliese, A., Nicklaus, M.C., 2011. Software and resources for computational medicinal chemistry. Future Med. Chem. 3, 1057–1085. Medina-Franco, J.L., Giulianotti, M.A., Yu, Y., Shen, L., Yao, L., Singh, N., 2009. Discovery of a novel protein kinase B inhibitor by structure-based virtual screening. Bioorg. Med. Chem. Lett. 19, 4634–4638. Ovenden, S.P., Nielson, J.L., Liptrot, C.H., Willis, R.H., Tapiolas, D.M., Wright, A.D., Motti, C.A., 2011. Callophycin A, a cytotoxic tetrahydro-b-carboline from the red alga Callophycus oppositifolius. Phytochem. Lett. 4, 69–71.

9

Rogers, D., Hopfinger, A.J., 1994. Application of genetic function approximation to quantitative structure-activity relationships and quantitative structureproperty relationships. J. Chem. Inf. Comput. Sci. 34, 854–866. Sander, T., Freyss, J., von Korff, M., Reich, J.R., Rufener, C., 2009. OSIRIS, an entirely in-house developed drug discovery informatics system. J. Chem. Inf. Model. 49, 232–246. Saxty, G., Woodhead, S.J., Berdini, V., Davies, T.G., Verdonk, M.L., Wyatt, P.G., Boyle, R.G., Barford, D., Downham, R., Garrett, M.D., Carr, R.A., 2007. Identification of inhibitors of protein kinase B using fragment-based lead discovery. J. Med. Chem. 50, 2293–2296. Siegel, R., Ma, J., Zou, Z., Jemal, A., 2014. Cancer statistics. Cancer J. Clin. 64, 9–29. Stewart, J.J.P., 1990. MOPAC: a semiempirical molecular orbital program. J. Comput. Aided Mol. Des. 4, 1–103. Thomsen, R., Christensen, M.H., 2006. MolDock: a new technique for high-accuracy molecular docking. J. Med. Chem. 49, 3315–3321. Tropsha, A., 2010. Best practices for QSAR model development, validation, and exploitation. Mol. Infor. 29, 476–488. VLife Sciences technologies: MDS 3.5 Molecular Design Suite. Pvt. Ltd., Pune, India. . Vyas, V.K., Ghate, M., Goel, A., 2013. Pharmacophore modeling, virtual screening, docking and in silico ADMET analysis of protein kinase B (PKB b) inhibitors. J. Mol. Graph. Model. 42, 17–25. Witten, I.H., Frank, E., 2005. Data Mining, Practical Machine Learning Tools and Techniques, second ed. Morgan Kaufmann. Zhang, S., Wei, L., Bastow, K., Zheng, W., Brossi, A., Lee, K.H., Tropsha, A., 2007. Antitumor agents 252. Application of validated QSAR models to database mining: discovery of novel tylophorine derivatives as potential anticancer agents. J. Comput. Aided Mol. Des. 21, 97–112.

Please cite this article in press as: Dicky John Davis, G., Hannah Rachel Vasanthi, A. QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBb). Eur. J. Pharm. Sci. (2015), http://dx.doi.org/10.1016/j.ejps.2015.04.026

523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551

QSAR based docking studies of marine algal anticancer compounds as inhibitors of protein kinase B (PKBβ).

Marine algae are prolific source of bioactive secondary metabolites and are found to be active against different cancer cell lines. QSAR studies will ...
2MB Sizes 5 Downloads 19 Views

Recommend Documents


QSAR and molecular docking studies on oxindole derivatives as VEGFR-2 tyrosine kinase inhibitors.
The three-dimensional quantitative structure-activity relationships (3D-QSAR) were established for 30 oxindole derivatives as vascular endothelial growth factor receptor-2 (VEGFR-2) tyrosine kinase inhibitors by using comparative molecular field anal

Molecular docking and pharmacophore studies of heterocyclic compounds as Heat shock protein 90 (Hsp90) Inhibitors.
Heat Shock Protein 90 was a key molecular chaperone involved in the proteome stability maintenance and its interference in many signaling networks associated with cancer progression, makes it of an important target for cancer therapeutics. The presen

QSAR and Molecular Docking Studies of Oxadiazole-Ligated Pyrrole Derivatives as Enoyl-ACP (CoA) Reductase Inhibitors.
A quantitative structure-activity relationship model was developed on a series of compounds containing oxadiazole-ligated pyrrole pharmacophore to identify key structural fragments required for anti-tubercular activity. Two-dimensional (2D) and three

3D-QSAR and docking studies of flavonoids as potent Escherichia coli inhibitors.
Flavonoids are potential antibacterial agents. However, key substituents and mechanism for their antibacterial activity have not been fully investigated. The quantitative structure-activity relationship (QSAR) and molecular docking of flavonoids rela

KDR Inhibitors Using Docking and 3D-QSAR Approaches.
Research and development of multi-target inhibitors has attracted increasing attention as anticancer therapeutics. B-RafV600E synergistically works with vascular endothelial growth factor receptor 2 (KDR) to promote the occurrence and progression of

Novel spirooxindole-pyrrolidine compounds: synthesis, anticancer and molecular docking studies.
Novel spirooxindole-pyrrolidine compounds have been synthesised through 1,3-dipolar cycloaddition of azomethine ylides generated from isatin and sarcosine or thioproline with the dipolarophile 3-(1H-imidazol-2-yl)-2-(1H-indole-3-carbonyl)acrylonitril

Biological activity and molecular docking studies of curcumin-related α,β-unsaturated carbonyl-based synthetic compounds as anticancer agents and mushroom tyrosinase inhibitors.
Hyperpigmentation in human skin and enzymatic browning in fruits, which are caused by tyrosinase enzyme, are not desirable. Investigations in the discovery of tyrosinase enzyme inhibitors and search for improved cytotoxic agents continue to be an imp

3D-QSAR, molecular docking and molecular dynamics studies of a series of RORγt inhibitors.
The discovery of clinically relevant inhibitors of retinoic acid receptor-related orphan receptor-gamma-t (RORγt) for autoimmune diseases therapy has proven to be a challenging task. In the present work, to find out the structural features required f

Cyclin-Dependent Kinase Inhibitors as Anticancer Therapeutics.
Cyclin-dependent kinases (CDKs) have been considered promising drug targets for a number of years, but most CDK inhibitors have failed rigorous clinical testing. Recent studies demonstrating clear anticancer efficacy and reduced toxicity of CDK4/6 in

3D-QSAR modeling and molecular docking study on Mer kinase inhibitors of pyridine-substituted pyrimidines.
Mer kinase is a novel therapeutic target for many cancers, and overexpression of Mer receptor tyrosine kinase has been observed in several kinds of tumors. To deeply understand the structure-activity correlation of a series of pyridine/pyrimidine ana