https://doi.org/10.13160/ricns.2017.10.4.209
Fragment based QSAR Analysis of CXCR-2 Inhibitors Using Topomer CoMFA Approach
Thirumurthy M†
Abstract
CXC chemokine receptor 2 (CXCR2) is a prominent chemokine receptor on neutrophils. CXCR2 antagonist may reduce the neutrophil chemotaxis and alter the inflammatory response because the neutrophilic inflammation in the lung diseases is found to be largely regulated through CXCR2 receptor. Hence, in the present study, Topomer based Comparative Molecular Field Analysis (Topomer CoMFA) was performed on a series of CXCR2 antagonist named pyrimidine-5- carbonitrile-6-alkyl derivatives. The best Topomer COMFA model was obtained with significant cross-validated correlation coefficient (q2 = 0.487) and non cross-validated correlation coefficients (r2 = 0.980). The model was evaluated with six external test compounds and its r2pred was found to be 0.616. The steric and electrostatic contribution map show that presence of bulkier and electropositive group around cyclopropyl ring may contribute more for improving the biological activities of these compounds. The generated Topomer CoMFA model could be helpful for future design of novel and structurally related CXCR2 antagonists.
Keywords: CXCR2; Topomer CoMFA
1. Introduction
Chemokines are G-protein-coupled receptors (GPCRs) which have long been implicated in the initiation and amplification of inflammatory responses by their role in leukocyte chemotaxis[1,2].Chemokines are small 8–
10 kDa proteins which act to regulate a variety of effects, including cell migration and inflammatory events. They are currently seven known CXCR receptor found in mammals named CXCR1-CXCR7. CXCR2 (also called CD182, IL8) plays a critical role in the regulation of neutrophil homeostasis[3] and is found on many cells including leukocytes, endothelial and epithelial cells[4,5]. CXCR2 plays an important role in chronic obstructive pulmonary disease (COPD), asthma, fibrotic pulmonary disorders[6-8]. CXCR2 receptor can be released from a number of inflammatory cell types and may have a broad functional role in number of acute and chronic diseases. It was found that neutrophilic inflammation in
the lung diseases is found to be largely regulated through CXCR2[9,2]. Therefore blockade of CXCR2 substan- tially reduces leukocyte recruitment, tissue damage and mortality. An antagonist of CXCR2 reduces neutro- philic chemotaxis and may alter the airway inflamma- tion. To date, there are no CXCR2 receptor antagonists approved for use in humans. However, several pharma- ceutical companies have disclosed CXCR2 antagonists and amongst these, navarixin and AZD-5069 are note- worthy.
Quantitative Structural Activity relationship (QSAR) models attempts to relate the chemical structure to bio- logical activity computationally or mathematically and to discover new compounds with improved biological activity. Topomer CoMFA includes alignment of struc- tural fragments called topomers. Structural fragments by definition contain a common feature, the “open valence”
or “attachment bond”. The Topomer methodology over- laps this common feature to provide an absolute orien- tation for any fragment. A Topomer is an invariant three-dimensional (3D) representation of molecular sub- unit generated from its two-dimensional (2D) topology by topomer alignment in topomer CoMFA[10]. In the present study, Topomer CoMFA has been employed to
Department of Genetic Engineering, School of Bioengineering, SRM University, SRM Nagar, Kattankulathur, Chennai 603203, India
†Corresponding author : thiru.murthyunom@gmail.com (Received : November 08, 2017, Revised : December 17, 2017,
Accepted : December 25, 2017)
study the activity of pyrimidine-5-carbonitrile-6-alkyl derivatives as CXCR2 antagonist. Topomer CoMFA models were generated with different R1 and R2 frag- ments and the best model was selected based on the cor- relation coefficient. The steric and electrostatic contribution map was analyzed to identify the important features of the compounds for improving the activity.
2. Materials and Methods
2.1. Data Set
The structure of the pyrimidine-5-carbonitrile-6-alkyl derivatives and their biological activities of 26 com- pounds were taken from the literature[11]. Biological activities i.e. IC50 values of each inhibitor was converted into pIC50 (-logIC50) and the dataset (26 compounds) were segregated into test (6 compounds) and training set (20 compounds). The training and test sets were clas- sified to ensure that both sets could completely cover the whole range of biological activity and structural diversity. The structures and their activity values are displayed in Table 1.
2.2. Topomer CoMFA
Topomer CoMFA, an alignment independent approach which merges the topomer technology and CoMFA was employed to overcome the alignment problem of CoMFA methodology[12,13]. A Topomer is an invariant three-dimensional (3D) representation of molecular sub- unit generated from its two-dimensional (2D) topol- ogy[10]. In Topomer CoMFA analysis, all molecules of dataset were divided into two fragments, shown as R1
(blue) and R2 (red) groups in Fig. 1. Each Topomer fragment was applied with topomer alignment to make a 3D invariant representation[14]. During Topomer CoMFA analysis, Gasteiger-Marsilli method was used to calcu- late the atomic charges and all molecules. were divided into two fragments whose contribution was calculated by multiplying each grid point at a regular space grid of 2 Å. The probe atom used for the steric and electro- static field calculation was sp3 hybridized carbon atom and oxygen atom respectively. The optimum number of the component obtained from cross validation analysis was used to calculate the r2 value.
2.3. Predictive Correlation Coefficient (r2pred) The predictive power of CoMFA model was deter- mined from six test molecules which were excluded during model development. The predictive correlation coefficient (r2pred) based on the test set molecules, is defined as:
where PRESS is the sum of the squared deviation between the predicted and actual activity of the test set molecules, and SD is defined as the sum of the square deviation between the biological activity of the test set compounds and the mean activity of the training set molecules[15].
3. Results and Discussion
3.1. Topomer CoMFA Analysis
In Topomer CoMFA analysis, each of the training set molecules was split into two fragments (such as R1 and R2). This was accomplished by specifying acyclic sin- gle bond to cut within each complete structure. Here we have used the highly active compound as the template for splitting of twenty training set molecules. The two fragments R1 and R2 is shown in the Fig. 1 where the blue color represents the R1 fragment and red color rep- resents the R2 fragment . Once fragmentation was com- pleted, topomers were generated. The model with good predictive ability in terms of q2= 0.487 and r2=0.980 with 5 components was obtained. The summary of Topomer CoMFA results was provided in Table 2. The predictive ability of the developed Topomer CoMFA
rpred2 (SD PRESS– ) ---SD
=
Fig. 1. Represents the Fragmentation pattern (R1 in blue and R2 in red) of highly active compound 17 obtained from topomer CoMFA analysis.
model has been assessed by predicting the activity of six test set compounds which were excluded during model generation. The predictive correlation coefficient r2pred
for Topomer CoMFA was 0.616 which indicates the developed model was more robust and reliable. The actual pIC50, predicted pIC50 and residual values for test Table 1. Structures and biological activities (pIC50) of CXCR2 inhibitors
Cmpd no Structure pIC50 values Cmpd no Structure pIC50 values
1
2
3
4
5
6
7
5.432
5.130
5.854
6.148
5.795
5.337
6.107
14
15
16
17
18
19
20
5.432
5.824
8.000
8.222
8.155
6.292
5.318
set (marked with *) and training set molecules were tab- ulated in Table 3 along with their R1 and R2 fragment contributions.
3.2. Contour Map Analysis
Topomer CoMFA steric and electrostatic contour maps for R1 and R2 fragments of the most active com- pound 17 are shown in Fig. 2. In case of steric contour
map, the green contours denote favorable steric inter- actions and the yellow contours shows the region where the steric group was not favored. The R1 and fragment of the most active compound 17 was superimposed with steric and electrostatic contour plots and shown in Fig. 2(a), (b) (c) and (d). In Fig. 2(a), the presence of bulky group in the cyclopropyl ring is covered by green steric contour which indicates the presence of bulky Table 1. Continued
Cmpd no Structure pIC50 values Cmpd no Structure pIC50 values
8
9
10
11
12
13
5.327
5.366
6.045
5.309
5.769
6.853
21
22
23
24
25
26
5.193
4.522
5.000
6.495
5.854
5.495
groups attached to pyrimidine ring makes the com- pound potent with higher activity. The blue colored regions show areas where more positively charged groups are favored, and red region highlight areas where groups with more negative charges are favored in the electrostatic contour map. The blue colored region in electrostatic contour plot indicates the presence of elec- tropositive groups in the cyclopropyl ring is most favor- able and hence the electro-positive group in these regions is very important for enhancing the biological activity.
4. Conclution
This study was conducted to rationalize the pyrimi- Table 2. Statistical results obtained from Topomer CoMFA
analysis
PLS Statistics Topomer CoMFA
q2 0.487
N 5
r2 0.980
StdErr 0.770
SEE 0.150
intercept 6.57
r2pred 0.616
q2= cross-validated correlation coefficient; N= number of statistical components; r2= non-cross validated correlation coefficient; SEE=standard estimated error; StdErr=standard error of prediction, F=Fisher value; r2predictive= predictive correlation coefficient for test set.
Table 3. Experimental and predicted pIC50 values of training and test set compounds
Compound no Actual pIC50 Field Contribution Topomer CoMFA
R1 R2 Predicted IC50 Residual values
1 5.432 -0.476 -0.332 5.758 -0.326
2* 5.131 -0.476 -0.080 6.011 -0.880
3 5.854 -0.476 -0.307 5.783 0.071
4 6.149 -0.476 -0.127 5.963 0.186
5 5.796 -0.476 -0.354 5.736 0.060
6 5.337 -0.476 -0.620 5.470 -0.133
7 6.108 -0.476 0.009 6.099 0.009
8 5.328 -0.476 -0.821 5.269 0.059
9* 5.367 0.158 0.006 6.730 -1.363
10 6.046 -0.476 -0.113 5.977 0.069
11 5.310 -0.476 -0.592 5.498 -0.188
12 5.770 -0.476 -0.165 5.925 -0.155
13* 6.854 0.076 0.401 7.044 -0.190
14 5.432 0.124 0.401 7.091 -1.659
15* 5.824 0.972 0.401 5.940 -2.116
16 8.000 1.007 0.401 7.974 0.026
17 8.222 1.253 0.401 8.221 0.001
18* 8.155 0.315 0.401 7.283 0.872
19 6.292 0.076 -0.390 6.253 0.039
20 5.319 0.076 -1.048 5.594 -0.275
21 5.194 0.076 -1.548 5.095 0.099
22 4.523 0.076 -2.116 4.527 -0.004
23 5.000 0.076 -1.724 4.919 0.081
24 6.495 0.076 -0.235 6.407 0.088
25 5.854 0.076 -0.703 5.940 -0.085
26* 5.495 0.076 -0.763 5.879 -0.384
dine-5-carbonitrile-6-alkyl derivatives by Topomer CoMFA analysis. The reliable Topomer model was generated and its predictive ability was determined using six test set compounds. The contour maps suggest the presence of bulky and electropositive group in the cyclopropyl ring attached to pyrimidine ring will help in improving the activity of the compound. Hence, this study is useful for the discovery of new antagonists for CXCR2 receptor.
References
[1] M. Uguccioni, M. D’Apuzzo, M. Loetscher, B. Dewald, and M. Baggiolini, “Actions of the chemotactic cytokines MCP-1, MCP-2, MCP-3, RANTES, MIP- 1a and MIP-1b on human monocytes”, Eur. J.
Immunol., Vol. 25, pp. 64-68, 1995.
[2] R. W. Chapman, J. E. Phillips, R. W. Hipkin, A. K.
Curran, D. Lundell, and J. S. Fine, “CXCR2 antag- onists for the treatment of pulmonary disease”, Pharmacol. Ther., Vol. 121, pp. 55-68, 2009.
[3] K. J. Eash, A. M. Greenbaum, P. K. Gopalan, and D. C. Link, “CXCR2 and CXCR4 antagonistically
regulate neutrophil trafficking from murine bone marrow”, J. Clin. Invest., Vol. 120, pp. 2423-2431, 2010.
[4] J. Reutershan, A. Basit, E. V. Galkina, and K. Ley,
“Sequential recruitment of neutrophils into lung and bronchoalveolar lavage fluid in LPS-induced lung injury”, Am. J. Physiol. Lung Cell. Mol. Physiol., Vol. 289, pp. L807-L815, 2005.
[5] S. L. Traves, S. J. Smith, P. J. Barnes, and L. E.
Donnelly, “Specific CXC but not CC chemokines cause elevated monocyte migration in COPD: a role for CXCR2”, J. Leukoc. Biol., Vol. 76, pp. 441-450, 2004.
[6] R. A. Pauwels and K. F. Rabe, “Burden and clinical features of chronic obstructive pulmonary disease (COPD)”, Lancet, Vol. 364, pp. 613-620, 2004.
[7] E. H. Bel, A. Sousa, L. Fleming, A. Bush, K. F.
Chung, J. Versnel, A. H. Wagener, S. S. Wagers, P.
J. Sterk, and C. H. Compton, “Diagnosis and defi- nition of severe refractory asthma: an international consensus statement from the Innovative Medicine Initiative (IMI)”, Thorax, Vol. 66, pp. 910-917, 2011.
[8] P. Anderson, “Emerging therapies in cystic fibrosis”, Ther. Adv. Respir. Dis., Vol. 4, pp. 177-185, 2010.
Fig. 2. Steric and electrostatic contour map for highly active compound 17 by Topomer CoMFA analysis. (a) and (b) Steric and Electrostatic contour map for R1 fragment. (c) and (d) Steric and Electrostatic contour map for R2 fragment. Sterically favoured areas are shown in green and unfavoured areas are shown in yellow contour, while blue contour depicts the favourable sites for positively charged groups and red contour depicts the favourable sites for negatively charged groups.
[9] P. M. Murphy, “Neutrophil receptors for interleukin- 8 and related CXC chemokines”, Semin. Hematol., Vol. 34, pp. 311-318, 1997.
[10] R. J. Jilek and R. D. Cramer III, “Topomers: a val- idated protocol for their self-consistent generation”, J. Chem. Inf. Model., Vol. 44, pp. 1221-1227, 2004.
[11] D. W. Porter, M. Bradley, Z. Brown, S. J. Charlton, B. Cox, P. Hunt, D. Janus, S. Lewis, P. Oakley, D.
O’Connor, J. Reilly, N. Smith, and N. J. Press, “The discovery of potent, orally bioavailable pyrimidine- 5-carbonitrile-6-alkyl CXCR2 receptor antago- nists”, Bioorg. Med. Chem. Lett., Vol. 24, pp. 3285- 3290, 2014.
[12] R. D. Cramer, D. E. Patterson, and J. D. Bunce,
“Comparative molecular field analysis (CoMFA). 1.
Effect of shape on binding of steroids to carrier pro- teins”, J. Am. Chem. Soc., Vol. 110, pp.5959-5967, 1998.
[13] R. D. Cramer, “Topomer CoMFA: a design meth- odology for rapid lead optimization”, J. Med. Chem., Vol. 46, pp. 374-388, 2003.
[14] R. D. Cramer, R. D. Clark, D. E. Patterson, and A.
M. Ferguson, “Bioisosterism as a molecular diver- sity descriptor: steric fields of single “topomeric”
conformers”, J. Med. Chem., Vol. 39, pp. 3060-3069, 1996.
[15] S. Wold, “Cross-validatory estimation of the number of components in factor and principal component model”, Technometrics, Vol. 20, pp. 397-405, 1978.
