Skip Navigation


PEDS Advance Access originally published online on December 9, 2005
Protein Engineering Design and Selection 2006 19(2):47-54; doi:10.1093/protein/gzi074
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
19/2/47    most recent
gzi074v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Xiao, J.
Right arrow Articles by Sun, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xiao, J.
Right arrow Articles by Sun, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Inhibitory mode of N-phenyl-4-pyrazolo[1,5-b] pyridazin-3-ylpyrimidin-2-amine series derivatives against GSK-3: molecular docking and 3D-QSAR analyses

Jingfa Xiao1, Zongru Guo1,3, Yanshen Guo1, Fengming Chu1 and Piaoyang Sun2

1Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050, China and 2Jiangsu Hengrui Medicine Co., Ltd, Jiangsu, China

3 To whom correspondence should be addressed. E-mail: zrguo{at}imm.ac.cn


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results and discussion
 3D-QSAR models
 Conclusions
 References
 
Glycogen synthase kinase 3 (GSK-3) inhibition is an important research topic because of its wide range of associated health implications. The interaction mode of a series of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds with human GSK-3 has been studied using molecular docking and 3D-QSAR approaches. In the 3D-QSAR studies, the molecular alignment and conformation determination are so important that they affect the success of a model. Flexible docking (AutoDock3.0.5) was used for the determination of ‘active’ conformation and molecular alignment. Comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were used to develop 3D-QSAR models of 80 N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds. The r2 values were 0.870 and 0.861 for CoMFA and CoMSIA models, respectively. The predictive ability of these models was validated by 10 compounds of the test set. Mapping these models back to the topology of the active site of GSK-3 led to a better understanding of the vital N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines–GSK-3 interactions. The results demonstrate that combination of ligand-based and receptor-based modeling is a powerful approach to build 3D-QSAR models. The interaction mode from this study may be helpful for the design of a novel inhibitor and guide the selection of candidate sites for further experimental studies on site-directed mutagenesis.

Keywords: 3D-QSAR/CoMFA/CoMSIA/glycogen synthase kinase 3/molecular docking


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results and discussion
 3D-QSAR models
 Conclusions
 References
 
The quantitative structure–activity relationship (QSAR) approach (Hansch and Klein, 1991Go) has been widely used in pharmacology in attempts to optimize drug compounds (Kubinyi, 1997aGo,bGo; Grover et al., 2000aGo,bGo), and molecular docking can fit molecules together in a favorable configuration to form a complex system. By correlating the features determining the binding properties within a set of ligands, QSAR can indirectly map the structural properties of the binding site. The information from the QSAR and molecular docking studies can further guide the protein engineering and structure-based drug design. In the 3D-QSAR studies, the molecular alignment and conformation determination are so important that they affect the success of a model. Most of the methods correlating ligand structures require as a starting point an alignment or superimposition of the set of considered ligands. However, the ‘correct’ ligand alignment for such an analysis is often not easy to define. An ideal alignment should resemble the conformations and orientations that the ligands adopt at the binding site of the target protein. Several strategies have been used to determine the active conformation and align molecules. Of these, the docking method is an attractive way to align molecules for comparative molecular field analysis (CoMFA) (Cramer et al., 1988Go) and comparative molecular similarity indices analysis (CoMSIA) (Klebe et al., 1994Go). Molecular docking has been shown to be very effective in the study of protein–ligand interactions, and the structural information from the theoretically modeled complex may help us to clarify the mechanism of molecular recognition and may even suggest how the structure of the receptor or ligand may be changed in order to improve some biological function for the design of new compounds. We generate an alignment of ligands known to bind to the target protein by docking into the geometry of the binding site. Subsequently, this alignment can be utilized to derive a 3D-QSAR model.

Since human glycogen synthase kinase 3 (GSK-3) is believed to be an important target enzyme in terms of the treatment of type 2 diabetes and Alzheimer's disease (Alonso et al., 2001Go; Frame and Cohen, 2001Go; Wagmann and Nuss, 2001Go), in this paper we studied the binding mode of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine derivatives against GSK-3 using the combined approach of ligand-based and receptor-based modeling. The strategy of combining conformations and alignment obtained from AutoDock3.0.5 (Morris et al., 2001Go) with CoMFA and CoMSIA produces a natural and reasonable elucidation of activation from a 3D-QSAR calculation. The 3D-QSAR models and the interaction mode may be helpful for the design of a novel inhibitor and guide the selection of candidate sites for further experimental studies on site-directed mutagenesis.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results and discussion
 3D-QSAR models
 Conclusions
 References
 
The crystal structure of GSK-3 complexed with the inhibitor BRW-6-bromoindirubin-3'-oxime (PDB entry 1UV5) was used in this research. The molecular modeling software Sybyl 6.9 (Tripos Associates, St Louis, MO) was employed for CoMFA (Cramer et al., 1988Go) and CoMSIA (Klebe et al., 1994Go) analyses and visualization. All calculations were performed on SGI Fuel workstations.

Dataset

The dataset for the CoMSIA and CoMFA calculations consisted of compounds that have been published by Tavares et al. (2004)Go. From the original 92 compounds, 12 were discarded because they had either undefined activity or stereochemistry. So we ended up with a list of 80 compounds (Table I). The IC50 values for GSK-3 were converted to pIC50 (–log IC50) values and used as dependent variables in the CoMFA and CoMSIA calculations. The pIC50 values of the compounds from the training set and the test set covered an interval of >5 log units for the target enzymes.


View this table:
[in this window]
[in a new window]
  		Table I.. Structures and inhibitory potencies (IC50) versus human GSK-3 of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine analogues

 
Molecular modeling and docking

The crystal structure of GSK-3 in complex with its inhibitor BRW (PDB entry: 1UV5) was recovered from the Brookhaven Protein Database (PDB http://www.rcsb.org/pdb). The potential of the 3D structures of GSK-3 was assigned according to the Amber 4.1 forcefield with Kollman-united-atom (Weiner et al., 1984Go) charges encoded in Sybyl 6.9. The initial structures of 80 compounds (Table I; A, B and C) were built by Sybyl software. Partial atomic charges were calculated using the semiempirical program MOPAC 6.0, applying AM1 Hamiltonian (Stewart, 1990Go). Energy minimizations were performed using the Tripos force field (Clark et al., 1989Go) with a non-bond cutoff of 8 Å and the Powell conjugate gradient algorithm (Powell, 1977Go) with convergence criterion of 0.005 kcal/(mol Å) and a maximum of 10 000 iterations, and taking charges into account. For the purpose of tackling the interaction mode of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines (inhibitors) with GSK-3 (enzyme), the advanced docking program AutoDock 3.0.5 (Morris et al., 2001Go) was used to perform the automated molecular docking. The Lamarckian genetic algorithm (LGA) (Solis and West, 1981Go; Morris et al., 1998Go) was employed to deal with the inhibitor–enzyme interactions. All compounds of the training set and the test set were manually docked into the ATP binding pocket of the above-mentioned GSK-3 enzyme. The docked structures of the inhibitors were generated after a reasonable number of evaluations. Finally, the docked inhibitor–enzyme complexes were selected according to the criteria of interacting energy combined with geometrical matching quality.

3D-QSAR analyses

In 3D-QSAR studies the spatial alignment of the compounds is usually one of the key steps in order to obtain meaningful results. For the alignment in the CoMFA and CoMSIA analyses, the relative three-dimensional orientations of the compounds included in the training set and the test set resulting from the molecular docking were used. Partial atomic charges were recalculated as indicated above, using the AM1 as implemented in MOPAC.

CoMFA

For the CoMFA calculations, steric and electrostatic field energies were calculated using sp3 carbon as the steric probe atom and a +1 net charge as the electrostatic probe. Steric and electrostatic interactions were calculated using the Tripos forcefield with a distance-dependent dielectric constant at all intersections in a regular grid spacing (2 Å). The minimum {sigma} (column filtering) was set to 2.0 kcal/mol to improve the signal-to-noise ratio by omitting those lattice points whose energy variation is below this threshold. A cutoff of 30 kcal/mol was adopted, and the regression analysis was performed using the cross-validation of leave-one-out method (LOO). The final model (non-cross-validated conventional analysis) was produced with the optimal number of components equal to that yielding the highest q2, and the corresponding conventional correlation coefficient r2, its standard error s, and the F ratio were also computed.

CoMSIA

For the CoMSIA calculation, the five similarity index fields, namely steric, electrostatic, hydrophobic, and hydrogen bond donor and acceptor fields, were calculated using a C1+ probe of 1 Å radius and the default value of 0.3 as attenuation factor. The steric contribution was reflected by the third power of the atomic radii of the atoms. Electrostatic properties were introduced as atomic charges resulted from molecular docking. An atom-based hydrophobicity was assigned according to the parametrization developed by Viswanadhan et al. (1989)Go. The statistical evaluation for the CoMSIA analyses was performed in the same way as described for CoMFA.


   Results and discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results and discussion
3D-QSAR models
 Conclusions
 References
 
The binding features of the inhibitors

Alignment In 3D-QSAR studies the spatial alignment of the compounds is usually one of the key steps used in order to obtain meaningful results. The biologically active conformations of the structures should be aligned in a way that represents a similar binding mode. Since there is no X-ray crystallographic structure information on the biologically active conformation of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines, we used a model (Meijer et al., 2003Go) with bromoindirubin-3'-oxime docked into the ATP binding site of GSK-3 as a template for our docking. In this model, two hydrogen bonds are observed (BRW OH-Ile62 backbone carbonyl; BRW carbonyl-Val135 backbone NH) which fix the structures in a similar spatial position. After manually docking all structures of the training set and the test set into the GSK-3 binding site, the docked inhibitor–enzyme complexes were selected according to the criteria of interacting energy combined with geometrical matching. By this docking method, the two hydrogen bonds of the crystal structure model from Meijer et al. (2003)Go were nicely reproduced (Figure 1). The resulting relative conformations of all compounds in the training set and test set were used (Figure 2A and B) for the following CoMFA and CoMSIA calculations. From Figure 2A we can see that all the inhibitors share common binding features. The very similar binding conformations of these inhibitors demonstrated that they interact with GSK-3 in a very similar way.


Figure 1
View larger version (29K):
[in this window]
[in a new window]
  		Fig. 1.. Conformational comparison of bromoindirubin-3'-oxime from the crystal structure (green) and that from the Autodock result (white) in the binding site. Hydrogen bonds with Ile62 and Val135 are depicted by dotted lines.

 

Figure 2
View larger version (65K):
[in this window]
[in a new window]
  		Fig. 2.. (A) Surface of active site of GSK-3 and the binding mode of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine series derivatives to GSK-3. (B) View of alignment of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine series derivatives.

 
Interaction of subsites To illustrate the interaction mechanism, compound 1 was selected for a more detailed analysis. In the latter discussions, all the descriptions refer to compound 1 unless otherwise noted. Figure 3A describes the interacting model of compound 1 with GSK-3. In general, compound 1 can be separated into the pyrazolo[1,5-b]pyridazine, aminopyrimidine and phenyl moieties. The last moiety can be further divided into three parts, namely the R6 substituent group, phenyl ring and Y substituent (Table 1 B). The aminopyrimidine and phenyl moieties of all the 80 N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds of the training and test sets are situated at the same site (Figure 2A). It can be seen from Figure 3A that the aminopyrimidine moiety of compound 1 is surrounded by residues Val135, Tyr134, Ile62 and Ala83 mainly through hydrophobic interactions. In addition, the pyrimidine nitrogen and amino group form two hydrogen bonds with the backbone carbonyl and backbone amide group NH of Val135, respectively. The phenyl moiety of compound 1 is surrounded by residues Glu137, Pro136, Tyr134, Ile62, Thr138 and Arg141 mainly through hydrophobic interactions, and the guanidinium group of Arg141 appears to have a favorable electrostatic interaction with the phenyl ring. The plane composed by the heavy atoms of the guanidinium group of Arg141 is almost parallel to the phenyl ring. From modeling studies, the substituents at positions 3 and 4 of the phenyl ring are well placed for a charged interaction with positively charged Arg141 of GSK-3. This in addition to the electrostatic interaction with Arg141 explains the higher inhibitory activities of compounds 12–20 as compared with compound 1. Compared with the aminopyrimidine and phenyl moieties, the pyrazolo[1,5-b]pyridazine moiety seems more flexible. The pyrazolo[1,5-b]pyridazine moiety is surrounded by residues Val70, Ala83, Lys85, Glu97, Met101, Leu132, Cys199, Phe201 and Asp200 mainly through the hydrophobic interaction. In the pyrazolo[1,5-b] pyridazine portion there is only one hydrogen bond between the nitrogen of the pyrazolo[1,5-b]pyridazine ring and Asp200. The R6 substituent mainly interacts with residues Glu97, Met101, Phe201 and Asp200 through the hydrophobic interaction. The steric hindrance of large substituents in inhibitors 24, 25, 32, 39, 40 and 80 was conducive to inversion of the pyrazolo[1,5-b]pyridazine ring. To illuminate this visually, the interaction models of compounds 40 and 80 with GSK-3 are represented in Figure 3C and D, respectively.


Figure 3
View larger version (35K):
[in this window]
[in a new window]
  		Fig. 3.. The interacting modes of compounds 1 (A), 51 (B), 40 (C) and 80 (D) with GSK-3. The inhibitors and the important residues for inhibitor–protein interaction are represented by ball-and-stick and stick models, respectively. Hydrogen bonds are depicted by magenta lines.

 
Hydrogen bonding interactions The hydrogen bonding is one important characteristic of the interaction between the N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines and GSK-3. When bound to GSK-3, these compounds share the common features of forming three hydrogen bonds with GSK-3. It can be seen clearly from Figure 3A that the pyrazolo-[1,5-b]pyridazine nitrogen of compound 1 acts as a donor to form a hydrogen bond with the backbone amino group of Asp200. In the Table 1 A (compounds 24, 25, 32, 39 and 40) and compound 80, the large substituents in R6 position was conducive to inversion of the pyrazolo[1,5-b]pyridazine ring. The hydrogen bond is formed between the pyrazolyl nitrogen and the backbone amino group of Asp200. To illuminate this visually, the interaction models of compounds 40 and 80 with GSK-3 are represented in Figure 3C and D, respectively. Similarly, as a hydrogen donor, the pyrimidine ring nitrogen also forms a hydrogen bond with the backbone amide NH group of Val135. The third hydrogen bond is formed between the amino group next to the pyrimidine ring as the donor and the backbone carbonyl of Val135. The addition of the N-methyl substituent (1 versus 2) causes a dramatic decrease in activity against GSK-3, suggesting that the NH formed an important hydrogen bonding interaction with the backbone of the ATP binding loop of the enzyme (Denessiouk et al., 2001Go). The hydrogen bond acts as an ‘anchor’, intensely affecting the 3D space position of the pyrazolo[1,5-b]pyridazine ring and the pyrimidine ring in the binding pocket and facilitating the hydrophobic interaction of the aromatic and heterocyclic rings with the side chain of Val135, Ile62, Arg141, Tyr134, Pro136 and Ala83. Compounds bearing a 3,4-OCH2O- group, such as 40 and 51(Figure 3B and C), form an additional hydrogen bond with the hydroxyl of Tyr134 residues besides the three main hydrogen bonds described above.


   3D-QSAR models
Top
 Abstract
 Introduction
 Materials and methods
 Results and discussion
 3D-QSAR models
Conclusions
 References
 
CoMFA

The major objective of CoMFA analysis for the N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines was to find the best predictive model within the system. We picked up 70 N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine inhibitors for CoMFA analyses and the other 10 as test compounds for model validation. The PLS analysis results of CoMFA are summarized in Table II, which shows that all of the statistical indexes are reasonably high. The data for predicted values versus experimental results are listed in Table IV, and the relationship between experimental binding affinities (–log IC50) and predicted activities by the CoMFA model is presented in Figure 4A. As listed in Table II, the CoMFA model gives q2(leave-one-out) = 0.480, q2(cross-validated) = 0.491, r2 = 0.870, F = 85.30, five components and the estimated standard error of 0.419. These values indicate that the CoMFA model has a good conventional statistical correlation and a fair predictive ability.


View this table:
[in this window]
[in a new window]
  		Table II.. CoMFA results for N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines

 

Figure 4
View larger version (9K):
[in this window]
[in a new window]
  		Fig. 4.. Predicted activities (PA) by CoMFA (A) and CoMSIA (B) models versus experimental activities (EA) of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines. Filled circles indicate compounds of the training set; filled triangles indicate compounds of the test set.

 
CoMSIA

Using steric, electrostatic, hydrophobic, and hydrogen bond donor and acceptor properties as descriptors, CoMSIA analysis was performed. The results are listed in Table III. The best q2 was found using all five different descriptor variables. This demonstrates that these variables are necessary to describe the interaction mode of the N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine inhibitors with GSK-3, as well as the field properties around the inhibitors. The predicted binding affinities derived from CoMSIA analysis are also listed in Table IV and shown in Figure 4B. The CoMSIA model with a cross-validated q2 of 0.527 for six components and a conventional r2 of 0.861 was obtained. These data demonstrate that the CoMSIA model is also satisfactorily predictive. The corresponding field distributions (Table III) of these five descriptor variables were 15.5, 36.7, 25.1, 15.6 and 7.1%, respectively, which indicates that hydrogen bond interactions play a crucial role in locating the N-phenyl-4-pyrazolo[1,5-b] pyridazin-3-ylpyrimidin-2-amine compounds in the binding site of GSK-3.


View this table:
[in this window]
[in a new window]
  		Table III.. CoMSIA results for N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines

 

View this table:
[in this window]
[in a new window]
  		Table IV.. Predicted activities (PA) versus experimental activities (EA, –log IC50) and residues ({delta}) by CoMFA and CoMSIA

 
The ultimate test for the usefulness of a 3D-QSAR model in the drug design process is predicting the activity of new compounds that are not included in the dataset that is used to obtain the model. To validate the stability and predictive ability of our 3D-QSAR model, 10 N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds (compounds 3*, 11*, 23*, 26*, 35*, 37*, 55, 66*, 71* and 77* in Table I) that were not included in the construction of CoMFA and CoMSIA models are selected as the test set. The predicted results for the test set were also summarized in Table IV and Figure 4. The binding affinities of the test set molecules were predicted reasonably well. Consequently, we believe that both QSAR models can be used as tools in the rational design of further GSK-3 inhibitors of the N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine family.

Analysis of the 3D maps of CoMSIA

The contour maps of all five CoMSIA fields were produced and analyzed. Figure 5A shows the steric and electrostatic contour maps for the GSK-3 model with compound 18 as an example of a good inhibitor. Detrimental and beneficial steric interactions are displayed in yellow and green contours, respectively, while blue and red contours illustrate the regions of favorable positive and negative electrostatic interactions, respectively. Sterically favored regions (colored in green) appear at position 2 of the pyrazolo[1,5-b]pyridazine group and position 3 of the phenyl ring, which suggests that more bulky substituents in these positions will improve the biological activity. Since the substituents are located near the entrance to the binding pocket, a large region of red contour near positions 3 and 5 of the phenyl group suggests that negatively charged groups and electron-withdrawing substituents may increase the inhibitory activity by taking advantage of the electrostatic nature of the environment around the polar residue Arg141. A blue contour near position 6 of the pyrazolo[1,5-b]pyridazine system represents an area where positive charge is favored by forming electrostatic interaction with the side chain of Glu97.


Figure 5
View larger version (35K):
[in this window]
[in a new window]
  		Fig. 5.. Contour maps as compared with the topology of GSK-3 complex. Only residues relevant to the discussion are shown for clarity. (A) The steric and electrostatic field distributions of CoMSIA. Green contours indicate areas where sterical bulk is favored, while yellow contours indicate areas where it is not favored. Positive charge favored areas are in blue; positive charge unfavored areas are in red. (B) The hydrophobic, H-bond donor and acceptor field distributions of CoMSIA. Hydrophobic favored areas are in yellow; hydrophilic favored areas are in white. H-bond donor favored areas are in cyan; H-bond donor unfavored areas are in purple. H-bond acceptor favored areas are in magenta; H-bond acceptor unfavored areas are in red. The residues are represented as sticks, and the inhibitor is shown in ball-and-stick.

 
The contour maps of the hydrogen bond donor and acceptor fields describe the spatial arrangement of favorable and unfavorable hydrogen bond interactions to acceptor or donor groups of the target enzyme (Figure 5B). The cyan contours that are observed at the amino hydrogen next to the pyrimidine group indicate a favorable hydrogen bond to an acceptor group in the protein. For compound 2, a methyl group replaced the amino hydrogen, which should in principle not be able to form the hydrogen bond. Consequently, compound 2 is three orders of magnitude less active than compound 1. The magenta contours appear opposite to the pyrazolo[1,5-b]pyridazine nitrogen atoms at positions 1 and 7, and position 4 of the phenyl ring, which indicate favorable interactions between a ligand–hydrogen bond acceptor group and the protein (Figure 5B). This is in agreement with the inhibitor–protein binding model. Acting as hydrogen bond donors, Val135 and Tyr134 form hydrogen bonds with N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds, while Val135 and Asp200 form hydrogen bonds with N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds as hydrogen bond acceptors. In the hydrophobic maps of CoMSIA (Figure 5B), the yellow and white contours represent the regions of favorable and unfavorable hydrophobic interactions, respectively. As shown in Figure 5B, one large yellow contour located opposite to position 2 of the pyrazolo[1,5-b]pyridazine group shows that this structural moiety interacts with the side chains of Ile62 at the binding site of GSK-3 through hydrophobic interaction. The other yellow contour near position 6 illustrates that more hydrophobic substituents will increase the inhibitory activity of the N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds by forming hydrophobic interactions with the side chains of Met101 and Leu130. The white contour around the carbonyl group of the Y moiety of compound 18 suggests that more hydrophilic group substitutions will increase inhibitory potencies due to the hydrophilicity of the environmental residues Arg141 and Glu137. The CoMSIA contour maps of the GSK-3 model nicely reflect the situation within the ATP binding site. For example, the three hydrogen bonds that are assumed to render the N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amines in the binding pocket are also accurately visualized by the contour plots.


   Conclusions
Top
 Abstract
 Introduction
 Materials and methods
 Results and discussion
 3D-QSAR models
 Conclusions
References
 
The binding conformations of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine derivatives against GSK-3 were predicted by employing the LGA algorithm and LS Algorithm of the AutoDock 3.0.5 program. The modeling results provide a satisfactory explanation for the binding mode of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine compounds with GSK-3. On the basis of the binding conformations of N-phenyl-4-pyrazolo[1,5-b]pyridazin-3-ylpyrimidin-2-amine derivatives, stable and predictive 3D-QSAR models with acceptable q2 values were developed by using CoMFA and CoMSIA techniques. The predictive ability of these models was validated by predicting the activity of 10 compounds exclusive to the training set, and this indicate that the application of these models for quantitative prediction of inhibitory potencies against GSK-3 is possible within the structural space. The robust QSAR model and its three-dimensional contour map provide guidelines to design compounds with new scaffolds and optimize current molecules. The strategy of combining conformations and alignment obtained from the AutoDock3.0.5 with CoMFA and CoMSIA produces a natural and reasonable elucidation of activation from a 3D-QSAR calculation.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results and discussion
 3D-QSAR models
 Conclusions
 References
 
Alonso,A., Zaidi,T., Novak,M., Grundke-Iqbal,I. and Iqbal,K. (2001) Proc. Natl Acad. Sci. USA, 98, 6923–6928.[Abstract/Free Full Text]

Clark,M., Cramer,R.D. and Van Obdenbosch,N. (1989) J. Comput. Chem., 10, 982–1012.[CrossRef]

Cramer,M., Cramer,R.D. and Jones,D.M. (1988) J. Am. Chem. Soc., 110, 5959–5967.[CrossRef]

Denessiouk,K.A., Rantanen,V.V. and Johnson,M.S. (2001) Proteins, 44, 282–291.[CrossRef][Web of Science][Medline]

Frame,S. and Cohen,P. (2001) Biochem. J., 359, 1–16.[CrossRef][Web of Science][Medline]

Grover,M., Singh,B., Bakshi,M. and Singh,S. (2000a) Pharm. Sci. Technol. Today, 3, 28–34.[CrossRef][Medline]

Grover,M., Singh,B., Bakshi,M. and Singh,S. (2000b) Pharm. Sci. Technol. Today, 3, 50–57.[CrossRef][Medline]

Hansch,C. and Klein,T.E. (1991) Methods Enzymol., 202, 512–543.[Web of Science][Medline]

Klebe,G., Abraham,U. and Mietzner,T. (1994) J. Med. Chem., 37, 4130–4146.[CrossRef][Web of Science][Medline]

Kubinyi,H. (1997a) Drug Discov. Today, 2, 457–467.[CrossRef]

Kubinyi,H. (1997b) Drug Discov. Today, 2, 538–547.[CrossRef]

Meijer,L., Skaltsounis,A.-L., Magiatis,P., Polychronopoulous,P., Knockaert,M., Leost,M., Ryan,X.P., Vonica,C.A., Brivanlou,A., Dajani,R., Crovace,C., Tarricone,C., Musacchio,A., Roe,S.M., Pearl,L.H. and Greengard,P. (2003) Chem. Biol., 11, 1255–1266.

Morris,G.M., Goodsell,D.S., Halliday,R.S., Huey,R., Hart,W.E., Belew,R.K. and Olson,A.J. (1998) J. Comput. Chem., 19, 1639–1662.[CrossRef][Web of Science]

Morris,G.M., Goodsell,D.S., Huey,R., Hart,W.E., Halliday,S., Belew,R. and Olson,A.J. (2001) AutoDock Version 3.0.5, Molecular Graphics Laboratory, Department of Molecular Biology. The Scripps Research Institute, CA, USA.

Powell,M.J.D. (1977) Math. Program., 12, 241–254.[CrossRef]

Solis,F.J. and West,R.J.B. (1981) Mathes. Opera. Res., 6, 19–30.

Stewart,J.J. (1990) J. Comput. Aided Mol. Des., 4, 1–105.[CrossRef][Web of Science][Medline]

Tavares,F.X., Boucheron,J.A., Dickerson,S.H., Griffin,R.J., Preugschat,F., Thomson,S.A., Wang,T.Y. and Zhou,H.Q. (2004) J. Med. Chem., 47, 4716–4730.[CrossRef][Web of Science][Medline]

Viswanadhan,V.N., Ghose,A.K., Revankar,G.R. and Robins,R.K. (1989) J. Chem. Inf. Comput. Sci., 29, 163–172.

Wagmann,A.S. and Nuss,J.M. (2001) Curr. Pharm. Des., 7, 417–450.[CrossRef][Web of Science][Medline]

Weiner,S.J., Kollman,P.A., Case,D.A., Singh,C., Ghio,G., Alagona,S., Profeta,P. and Weiner,P.K. (1984) J. Am. Chem. Soc., 106, 765–784.[CrossRef]

Received January 3, 2005; revised August 1, 2005; accepted October 6, 2005.

Edited by Robert Stroud


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
19/2/47    most recent
gzi074v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Xiao, J.
Right arrow Articles by Sun, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xiao, J.
Right arrow Articles by Sun, P.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?