Subscribe Now Subscribe Today
Research Article
 

QSAR Studies on Some GSK-3α Inhibitory 6-aryl-pyrazolo (3,4-b) pyridines



Ashutosh Jamloki, C. Karthikeyan, S.K. Sharma, N.S. Hari Narayana Moorthy and P. Trivedi
 
ABSTRACT

A quantitative structure-activity relationship study on 6-aryl-pyrazolo (3,4-b) pyridines was performed to gain structural insight into the binding mode of the molecules to the glycogen synthase kinase -3α, an enzyme phosphorylate and inhibit Glycogen Synthase (GS) which is the rate limiting enzyme in the glycogen biosynthesis. The molecular modeling studies were performed using CS Chem. Office 2001 molecular modeling software version 6.0. Allinger`s MM2 force field by fixing Root Mean Square Gradient (RMS) to 0.1 Kcal mol-1 and semiemperical AM1 Hamiltonian method (MOPAC module) were used to minimize the energy and calculate descriptors. The thermodynamic and steric features of 6-aryl-pyrazolo (3,4-b) pyridines are highly correlated with GSK-3α inhibitory activity. The results of the study suggests that introduction of bulky groups at C-5 position of the pyrazolopyridine ring will increase the GSK-3α inhibitory potency as it may involve in hydrophobic interaction with the ATP binding site of the enzyme. The results of the study reveal that the conformational rigidity and orientation of molecule play significant role in the GSK-3α inhibitory activity. Additionally, electronic interactions between molecule and enzyme were found to be crucial for GSK-3α inhibitory activity.

Services
Related Articles in ASCI
Similar Articles in this Journal
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

Ashutosh Jamloki, C. Karthikeyan, S.K. Sharma, N.S. Hari Narayana Moorthy and P. Trivedi, 2006. QSAR Studies on Some GSK-3α Inhibitory 6-aryl-pyrazolo (3,4-b) pyridines. Asian Journal of Biochemistry, 1: 236-243.

DOI: 10.3923/ajb.2006.236.243

URL: https://scialert.net/abstract/?doi=ajb.2006.236.243

Introduction

Glycogen Synthase Kinase -3 (GSK-3) is a serine/threonine protein kinase that was first identified in late 1970’s, due to its ability to phosphorylate and inhibit Glycogen Synthase (GS) which is the rate limiting enzyme in the glycogen biosynthesis (Embi et al., 1980). Two isoforms of GSK-3 exists, GSK-3α and GSK-3β, both share a high homology at their catalytic site but the α form possess an extended N-terminus with respect to the β form (Ali et al., 2001). The phosphorylation of proteins by GSK-3 is an important link in signaling pathways, those regulate cell differentiation, cellular growth and proliferation, metabolic processes, apoptosis control, inflammation and mechanisms involved in neuronal function (Martinez et al., 2002; Elder-Finkelman, 2002). Therefore GSK-3 inhibiting agents have potential to interfere with many physiological or pathological situations by interference with these signaling processes. Recently a number of publications have appeared suggesting GSK-3 as a target for the treatment of type 2-diabetes (Cohen and Frame, 2001) and Alzheimer’s Disease (AD) (Anderton et al., 2000). Two characteristic neuropathological hallmarks of AD are Neurofibrillary Tangles (NFT’s) and increased production of amyloid beta (Aβ) peptides followed by their aggregation (Caricasole et al., 2003). NFT’s are composed of highly phosphorylated form of the microtubule associated protein tau (τ) (Lee et al., 2001). Total 25 sites of phosphorylation have so far been identified in tau protein, many of which are serine or threonine residues and studies have shown that GSK-3 is one of the main in vivo players of phosphorylation of tau protein (Flaherty et al., 2000).

Another protein aggregate, amyloid plaque contains mainly Aβ peptide fibrils, which are derived from amyloid precursor protein (APP) by sequential proteolysis catalyzed by an aspartyl protease, β-site APP cleaving enzyme (BACE) followed by presinelin-dependent γ-secretase cleavage (Wolfe, 2001). Recently it has been reported that Lithium, a GSK-3 inhibitor, block production of Aβ peptides by interfering with APP cleavage at γ-secretase step, where the target for Lithium is GSK-3α, which is required for maximal processing of APP (Phiel et al., 2003). Hence GSK-3α inhibitors offer a new therapeutic approach for AD treatment by reducing the formation of both amyloid plaques and NFT’s.

So far a few compounds are known with GSK-3 inhibitory potential such as Lithium (Jope, 2003), Azaindolylmaleimides (Shen et al., 2004) and Anilinomaleimides (Smith et al., 2001), Paullones (Leost et al., 2000), Indirubins (Polychonopoulos et al., 2004), Aloisines (Mettey et al., 2003) and Thiazolidinones (Martinez et al., 2002). Recently, Witherington et al. (2003) identified and synthesized 6-aryl-pyrazolo (3,4-b) pyridines as potent inhibitors of GSK-3α. A QSAR analysis was performed on the abovementioned series of compounds in order to gain insight into the physicochemical requirements of these inhibitors. The applicability of QSAR studies in mechanistic interpretation of ligand macromolecular interactions has been proved. With the abovementioned rationale, an attempt has been made to propose quantitative models for describing the factors influencing the affinity of the drug molecules towards the enzyme. Additionally efforts have been made to justify the findings of the QSAR study with the results of the molecular modeling studies reported in the research paper chosen for the QSAR study. The results of the study will help in rationalizing the design of 6-aryl-pyrazolo (3,4-b) pyridines as GSK-3α inhibitors and predicting the inhibitory activity of the newly designed analogues.

Materials and Methods

The data set of a total of 16 compounds was taken from published literature (Witherington et al., 2003), out of which one compound was omitted due to its numerically undefined biological activity. All the values of biological data were expressed as IC50 values in nanomolar units. In the present study biological activity data was first converted to its molar unit and subsequently to negative logarithmic units (pIC50) to reduce skewness of data set. The general structure of these analogues is shown in Fig. 1 and compounds with their biological activity data are shown in Table 1.

Fig. 1: General structure of 6-aryl-pyrazolo (3,4-b) pyridines

Table 1: 6-aryl-pyrazolo (3,4-b) pyridines and their GSK-3α inhibitor activities

Table 2: Descriptors used in present study

The molecular modeling studies were performed using CS Chem. office, 2001 molecular modeling software ver. 6.0, supplied by Cambridge Software Company (CS, Chemoffice, 2001). The structure of each compound was drawn in ChemDraw Ultra ver. 6.0 and copied to Chem3D ultra ver. 6.0 in order to create its 3-D model. Each model was cleaned up and energy minimization was performed using Allinger’s MM2 force field by fixing Root Mean Square Gradient (RMS) to 0.1 Kcal mol-1 D. Further geometry optimization was done using semiemperical AM1 (Austin Model 1) Hamiltonian method, closed shell restricted wave function available in the MOPAC module until the RMS value becomes smaller than 0.001 Kcal mol-1 D. Energy minimized geometry was used for calculation of various thermodynamic, steric and electronic descriptors which are mentioned in Table 2.

Stepwise multiple linear regression was performed using SYSTAT software (SYSTAT 10.2 version, 2003) in order to obtain QSAR models. Statistical quality of the models was judged by correlation co-efficient (r), squared correlation co-efficient (r2), which is relative measure of quality of fit, adjusted squared correlation co-efficient (r2Adj) i.e. explained variance in biological activity, Standard Error of Estimate (SEE) representing absolute measure of quality of fit and Fischer’s value (F), which represents F-ratio between the variance of calculated and observed activity. Chance statistics (evaluated as the ratio of the equivalent regression equations to the total number of randomized sets; a chance value of 0.001 corresponds to 0.1% chance of fortuitous correlation) was performed in order to assure that the results are not merely based upon chance correlation.

Studentized-Residual method was adopted for the detection of outliers. A compound was considered as an outlier for deriving a particular model when the residual value exceeded twice the standard error of estimate of the model. In order to validate the generated QSAR models Leave One Out (LOO) method was used. Squared cross- correlation co-efficient (q2), standard deviation of sum of square of difference between predicted and observed values (SPRESS) and standard deviation of error of prediction (SDEP) were also calculated for each model to estimate the predictive potential of models with the help of VALSTAT Software (VALSTAT, 2004).

Results and Discussion

GSK-3α inhibitory activity data and various molecular descriptors were taken as dependent and independent variables respectively and correlations were established between them by employing stepwise multiple linear regression method. The orthogonality of descriptors in the selected correlations were confirmed by the calculation of overall correlation matrix and tolerance level of each descriptor (Table 3 and 4). Tolerance is defined as 1- R squared, where R squared is the multiple R of a given independent variable regressed on all other independent variables. Tolerance value ranges from 0.0 to 1.0. If tolerance value of an independent variable equals 1.00, it is totally independent of the other predictor variables, whereas if it equals 0.0 it is totally collinear with other independent variables. Further the selected models were checked for any autocorrelation by calculating Durbin-Watson (D-W) Statistics.

Among many correlations generated, statistical significant models were selected for further discussion. The most significant biparametric QSAR model selected was:

Where, N is the number of data points, r is correlation coefficient, r2 is squared correlation coefficient, r2Adj is the Explained Variance (EV) calculated as r2Adj = r2 (1-1/F) that accounts in percentage when multiplied by 100 for the variance in activity, SEE is standard error of estimate and other terms bear the meaning as described earlier.

This biparametric model explains 91.9% variance in the GSK-3α inhibitory activity. The F-statistics of the model is significant at 99.9% level, as shown by greater calculated value of F than the tabulated F (2, 12 α = o.oo1) = 8.51. t-tests were performed to assess significance of individual coefficients. The calculated t-values as given in Table 4, exceeded tabulated t = 3.012, at 99.5% confidence interval, showing significance of regression coefficients at this level.

Table 3: Correlation matrix for descriptors in the selected models

Table 4: T-statistics and tolerance values for the descriptors in the selected models

Tolerance level of each descriptor is close to 1 as shown in Table 4, thus confirming absence of collinearity among predictor variables. Magnitude of D-W coefficient shows independence of observation. High q2 value and low SPRESS and low SDEP values show good predictive power of the model.

This model establishes the correlation between principal moment of inertia (PMI-Y), Van der Waals 1,4 Energy (VDWE) and GSK-3α inhibitory activity. PMI-Y is a spatial descriptor, which explains the significance of orientation and molecular distribution along Y-axis for GSK-3α inhibitory potency. The positive coefficient of PMI-Y in the model suggests the presence of bulky substituents oriented towards Y-axis of the molecule will enhance the GSK-3α inhibitory activity. The observation supports the hypothesis that the presence of the bulky substituents like bromine with inherent hydrophobic character may involve in nonspecific interaction with the ATP binding site. However phenyl substituents that possess good hydrophobicity exhibits lower inhibitory potency. This may be attributed to the strain exerted by the two adjacent phenyl rings on the planar pyrazolo (3,4-b) pyridine ring thereby partly disrupting the hydrogen bonding interaction between nitrogen in the pyrazolo group and the complementary group in the enzyme.

Van der Waals 1,4 energy, a thermodynamic descriptor, can be defined as the sum of pairwise Van der Waals interaction energy terms for atoms separated by exactly three chemical bonds. Negative coefficient of this descriptor in the model indicates the necessity of conformational flexibility to allow proper orientation of the pharmocophoric groups in the molecule for GSK-3 inhibition.

On addition of a thermodynamic descriptor Dipole-Dipole energy (Ed) to the model 1 significant improvement in the statistical parameters was observed.

This model accounts for 96.2% of GSK-3α inhibitory activity. F-statistics proves it to be statistically highly significant (more than 99.9 %), as the calculated variance ratio i.e. Fischer value (F) exceeds the tabulated F (3, 12 α = o.oo1) = 12.7. Low standard error of estimate suggests a high degree of confidence in the analysis. D-W statistics affirm absence of serial autocorrelation.

Table 5: Descriptors contributing to the GSK-3α inhibitory activity of 6-Aryl-Pyrazolo (3,4-b) pyridines

Table 6: Observed and predicted activities of 6-Aryl-Pyrazolo (3,4-b) pyridines

The values of intercorrelation coefficient indicate lack of multicollinearity (Table 3), which was further reaffirmed by checking the tolerance level of each descriptor (Table 4). t-statistics shows that regression coefficient are highly significant (Table 4). The model has remarkable predictive power, which is mirrored in its high q2 value and low SPRESS and SDEP values. None of the compound was found to be outlier in any model.

Higher correlation coefficient value and low standard error of estimation indicates that this model is better than the preceding one. Moreover the model exhibits more predictive power than abovementioned model, as suggested by its higher q2 value. In view of all the statistical parameters model 2 seems to be superior in theoretical prediction of GSK-3α inhibitory activity.

Dipole-Dipole energy, which is a thermodynamic descriptor that describes the sum of the electrostatic energy terms resulting from interaction of two dipoles, shows positive contribution to the inhibitory activity. Positive coefficient of this term in the model indicates that substitution with the functional moieties those are capable of Dipole-Dipole interaction with the binding site of enzyme will be conducive for activity which reaffirms the significance of hydrogen bonding interaction between the para hydroxyl group of the 6-aryl moiety and amino acid residues (Glu97 and Asp200) of the enzyme.

Calculated values of all the descriptors in the selected models are shown in Table 5. Predicted activity values were calculated for each compound using the correlations developed and a comparison was made with the observed values (Table 6, Fig. 2 and 3).

The QSAR analysis resulted in statistically significant quantitative models with good predictive ability. It was observed from generated QSAR models that GSK-3α inhibitory activity of 6-aryl-pyrazolo (3,4-b) pyridines is governed by thermodynamic and steric features of the molecules.

Fig. 2: Scatter plot between observed and predicted gsk-3α inhibitory activity (Model 1)

Fig. 3: Scatter plot between observed and predicted gsk-3α inhibitory activity (Model 2)

The results of the study suggests that introduction of bulky groups at C-5 position of the pyrazolopyridine ring will increase the GSK-3α inhibitory potency as it may involve in hydrophobic interaction with the ATP binding site of the enzyme. Moreover, the study also illustrates the significance of the para hydroxyl group of the 6-aryl moiety for the drug-enzyme interaction. The findings of the QSAR analysis are consistent with molecular modeling studies performed on these analogs. The results obtained from the QSAR study emphasize the combined utility of the QSAR analysis and the molecular modeling studies to optimize the design of potent GSK-3α inhibitory molecules.

Acknowledgments

One of the authors, Ashutosh Jamloki likes to thank All India Council for Technical Education (AICTE) for providing fellowship. The authors wish to thank the Director, Shri G.S. Institute of Technology and Science for providing the necessary facilities for undertaking this study.

REFERENCES
Ali, A., P.K. Hoeflich and J.R. Woodgett, 2001. Glycogen synthase kinase-3: Properties functions and regulation. Chem. Rev., 101: 2527-2540.
PubMed  |  

Anderton, H.B., R. Dayanandan, R. Killick and S. Lovestone, 2000. Does dysregulation of the Notch and wingless wnt pathways underlie the pathogenesis of alzheimer's disease. Mol. Med. Today, 6: 54-59.
PubMed  |  

Caricasole, A., A. Copani, A. Caruso, F. Caraci and L. Luisa et al., 2003. The Wnt pathway cell-cycle activation and beta-amyloid novel therapeutic strategies in alzheimer's disease. Trends Pharmacol. Sci., 24: 233-238.
PubMed  |  

Chemoffice, C.S., 2001. Software package supplied by Cambridge Software Company 100, Cambridge Park Drive, Cambridge, USA.

Cohen, P. and S. Frame, 2001. The renaissance of GSK3. Nat. Rev. Mol. Cell Biol., 2: 769-776.
PubMed  |  Direct Link  |  

Eldar-Finkelman, H., 2002. Glycogen synthase kinase 3: An emerging therapeutic target. Trends Mol. Med., 8: 126-132.
CrossRef  |  PubMed  |  Direct Link  |  

Embi, N., D.B. Rylatt and P. Cohen, 1980. Glycogen synthase kinase-3 from rabbit skeletal muscle separation from cyclic-amp-dependent protein kinase and phosphorylase kinase. Eur. J. Biochem., 107: 519-527.
PubMed  |  

Flaherty, D.B., P.J. Sorrea, G.H. Tomasienicz and G.J. Wood, 2000. Phosphorylation of human tau protein by microtubule-associated kinases gsk3 beta and cdk5 are key participants. J. Neuro. Sci. Res., 62: 463-472.
PubMed  |  

Jope, R.S., 2003. Lithium and gsk-3 one inhibitor two inhibitory actions multiple outcomes. Trends Pharmacol. Sci., 24: 441-443.
PubMed  |  

Lee, V.M., M. Goedert and J.Q. Trojanowski, 2001. Neurodegenerative tauopathies. Ann. Rev. Neurosci., 24: 1121-1159.
PubMed  |  

Leost, M., C. Schultz, A. Link, Y.Z. Wu and J. Biernat et al., 2000. Paullones are potent inhibitors of glycogen synthase kinase-3β and cyclin-dependent kinase 5/p25. Eur. J. Biochem., 267: 5983-5994.
CrossRef  |  PubMed  |  Direct Link  |  

Martinez, A., A. Castro, I. Dorronsoro and M. Alonso, 2002. Glycogen synthase kinase 3 gsk-3 inhibitors as new promising drugs for diabetes neurodegeneration cancer and inflammation. Med. Res. Rev., 22: 373-384.
CrossRef  |  PubMed  |  

Martinez, A., M. Alonso, A. Castro, C. Perez and F.J. Moreno, 2002. First non-atp competitive glycogen synthase kinase 3 beta gsk-3beta inhibitors thiadiazolidinones tdzd as potential drugs for the treatment of alzheimer's disease. J. Med. Chem., 45: 1292-1299.
PubMed  |  

Mettey, Y., M. Gompel, V. Thomas, M. Garnier and M. Leost et al., 2003. Aloisines a new family of cdk/gsk-3 inhibitors sar study crystal structure in complex with cdk2 enzyme selectivity and cellular effects. J. Med. Chem., 46: 222-236.
PubMed  |  

Phiel, C.J., C.A. Wilson, V.M. Lee and P.S. Kecin, 2003. Gsk-3alpha regulates production of alzheimer's disease amyloid-beta peptides. Nature, 423: 435-439.
PubMed  |  

Polychonopoulos, P., P. Magiatis, A.L. Skaltsounis, V. Myrianthopoulos and E. Mikros et al., 2004. Structural basis for the synthesis of indirubins as potent and selective inhibitors of glycogen synthase kinase-3 and cyclin-dependent kinases. J. Med. Chem., 47: 935-946.
Direct Link  |  

Shen, L., C. Prouty, B.R. Conway, L. Westover and J.Z. Xu et al., 2004. Synthesis and biological evaluation of novel macrocyclic bis-7-azaindolylmaleimides as potent and highly selective glycogen synthase kinase-3 beta (GSK-3 beta) inhibitors. Bioorg. Med. Chem., 12: 1239-1255.
Direct Link  |  

Smith, D.G., M. Buffet, A.E. Fenwick, D. Haigh and R.J. Ife et al., 2001. 3-Anilino-4-arylmaleimides potent and selective inhibitors of glycogen synthase kinase-3 GSK-3. Bioorg. Med. Chem. Lett., 11: 635-639.
PubMed  |  

Witherington, J., V. Bordas, A. Gaiba, N.S. Garton and A. Naylor et al., 2003. 6-aryl-pyrazolo 3,4-b pyridines potent inhibitors of glycogen synthase kinase-3 GSK-3. Bioorg. Med. Chem. Lett., 13: 3055-3057.
Direct Link  |  

Wolfe, M.S., 2001. Secretase targets for alzheimer's disease identification and therapeutic potential. J. Med. Chem., 44: 2039-2060.
CrossRef  |  

©  2019 Science Alert. All Rights Reserved