Abhishek Seminar Presentation

1. PROJECT NAMEPROJECT NAME
QSAR ANALYSIS ANDQSAR ANALYSIS AND
VALIDATION STUDIES ONVALIDATION STUDIES ON
AMINOQUINOLINE DERIVATIVESAMINOQUINOLINE DERIVATIVES
AS MELANIN CONCENTRATINGAS MELANIN CONCENTRATING
HORMONE-1R INHIBITORSHORMONE-1R INHIBITORS
BYBY
M.ABHISHEKM.ABHISHEK

2. INTRODUCTION TO DRUGINTRODUCTION TO DRUG
DESIGNINGDESIGNING
 DEFFINITIONDEFFINITION
 Designing of drug molecules basing upon theirDesigning of drug molecules basing upon their
biological targets.biological targets.
 It is mainly of three typesIt is mainly of three types
 Structure based drugdesignStructure based drugdesign
 Denovo based drugdesignDenovo based drugdesign
 Analog based drugdesignAnalog based drugdesign
 Analog based drug design alsi called QSARAnalog based drug design alsi called QSAR
analysis.analysis.

3. About QSAR STUDIESAbout QSAR STUDIES
QSAR is qualitative structure activityQSAR is qualitative structure activity
relation shiprelation ship
Uses of qsar analysisUses of qsar analysis
Mainly useful for the determinationMainly useful for the determination
physiochemical properties.physiochemical properties.
Useful to predict the biological value of theUseful to predict the biological value of the
molculesmolcules

4. CONTENTSCONTENTS
INTRODUCTIONINTRODUCTION
MATERIALS AND METHODSMATERIALS AND METHODS
RESULTS AND DISCUSSIONRESULTS AND DISCUSSION
CONCLUSIONCONCLUSION
REFERENCESREFERENCES

5. INTRODUCTIONINTRODUCTION
 ABOUT MCHABOUT MCH
 The underlying causes of obesity are poorly understood butThe underlying causes of obesity are poorly understood but
probably involve complex interactions between manyprobably involve complex interactions between many
neurotransmitter and neuropeptide systems involved in theneurotransmitter and neuropeptide systems involved in the
regulation of food intake and energy balance. Three pieces ofregulation of food intake and energy balance. Three pieces of
evidence indicate that the neuropeptide melanin-concentratingevidence indicate that the neuropeptide melanin-concentrating
hormone (MCH) is an important component of this system.hormone (MCH) is an important component of this system.
 Melanin-concentrating hormone (MCH) is a cyclic neuropeptideMelanin-concentrating hormone (MCH) is a cyclic neuropeptide
(human/rat 19 aa) that regulates a variety of functions in mammalian(human/rat 19 aa) that regulates a variety of functions in mammalian
brain, in particular feeding behavior .brain, in particular feeding behavior .
 MCH is synthesized in mainly in the lateral hypothalamus and zonaMCH is synthesized in mainly in the lateral hypothalamus and zona
incerta. MCH stimulates feeding,incerta. MCH stimulates feeding,
 Recently, an orphan G-protein coupled receptor (SLC-1, GPR24)Recently, an orphan G-protein coupled receptor (SLC-1, GPR24)
has been identified as the receptor of MCH. MCH receptor ishas been identified as the receptor of MCH. MCH receptor is
predicted to contain 7 transmembrane domains, a feature typical ofpredicted to contain 7 transmembrane domains, a feature typical of
G-protein coupled receptorsG-protein coupled receptors

6.  Recently, a novel second human MCH receptorRecently, a novel second human MCH receptor
(MCH2R) has been cloned and characterized. MCH2R(MCH2R) has been cloned and characterized. MCH2R
gene encodes a 340 aa protein with 38% identity withgene encodes a 340 aa protein with 38% identity with
MCH1RMCH1R
 MOLECULAR CHARACTERIZATIONMOLECULAR CHARACTERIZATION
 Orphan G-protein-coupled receptors (GPCRs) areOrphan G-protein-coupled receptors (GPCRs) are
cloned proteins with structural characteristics common tocloned proteins with structural characteristics common to
the GPCRs but that bind unidentified ligands. Orphanthe GPCRs but that bind unidentified ligands. Orphan
GPCRs have been used as targets to identify novelGPCRs have been used as targets to identify novel
transmitter moleculestransmitter molecules
 We demonstrate that nanomolar concentrationsWe demonstrate that nanomolar concentrations
of MCH strongly activate SLC-1-relatedof MCH strongly activate SLC-1-related
pathways through G(alpha)i and/or G(alpha)qpathways through G(alpha)i and/or G(alpha)q
proteinsproteins

7. STRUCTURE OF MCHSTRUCTURE OF MCH

8. FUNCTION OF MCHFUNCTION OF MCH
 Melanin-concentrating hormone (MCH) is a cyclic neuropeptide, whichMelanin-concentrating hormone (MCH) is a cyclic neuropeptide, which
centrally regulates food intake and stress. MCH induces food intake incentrally regulates food intake and stress. MCH induces food intake in
rodents and, more generally, acts as an anabolic signal in energyrodents and, more generally, acts as an anabolic signal in energy
regulation.regulation.
 Two receptors for MCH in humans have very recently been characterised,Two receptors for MCH in humans have very recently been characterised,
namely, MCH-R1 and MCH-R2. MCH-R1 has received considerablenamely, MCH-R1 and MCH-R2. MCH-R1 has received considerable
attention, as potent and selective antagonists acting at that receptor displayattention, as potent and selective antagonists acting at that receptor display
anxiolytic, antidepressant and/or anorectic properties.anxiolytic, antidepressant and/or anorectic properties.
 ACTIVE SITE AND INACTIVE SITE OF MCHACTIVE SITE AND INACTIVE SITE OF MCH
 Human melanin-concentrating hormone (hMCH) and many of its analoguesHuman melanin-concentrating hormone (hMCH) and many of its analogues
are potent but nonspecific ligands for human melanin-concentratingare potent but nonspecific ligands for human melanin-concentrating
hormone receptors 1 and 2 (hMCH-1R and hMCH-2R). To differentiatehormone receptors 1 and 2 (hMCH-1R and hMCH-2R). To differentiate
between the physiological functions of these receptors, selectivebetween the physiological functions of these receptors, selective
antagonists are needed. In this study, analogues of Ac-Arg(6)-cyclo(S-S)antagonists are needed. In this study, analogues of Ac-Arg(6)-cyclo(S-S)
(Cys(7)-Met(8)-Leu(9)-Gly(10)-Arg(11)-Val(12)-Tyr(13)-Arg(14)-Pro(15)-(Cys(7)-Met(8)-Leu(9)-Gly(10)-Arg(11)-Val(12)-Tyr(13)-Arg(14)-Pro(15)-
Cys(16))-NH(2), a high affinity but nonselective agonist at hMCH-1R andCys(16))-NH(2), a high affinity but nonselective agonist at hMCH-1R and
hMCH-2R, were prepared and tested in binding and functional assays onhMCH-2R, were prepared and tested in binding and functional assays on
cells expressing these receptorscells expressing these receptors

9. MATERIALS&METHODSMATERIALS&METHODS
 Tsar (Tools for Structure Activity Relationship) is a program used toTsar (Tools for Structure Activity Relationship) is a program used to
investigates quantitative structure activity relationships (QSAR).investigates quantitative structure activity relationships (QSAR).
Tsar is an integrated analysis package for interactive investigationTsar is an integrated analysis package for interactive investigation
of Quantitative Structure-Activity Relationship (QSARs )of Quantitative Structure-Activity Relationship (QSARs )
 The major functional areas of Tsar and their significance in theThe major functional areas of Tsar and their significance in the
investigation of quantitative structure-activity relationship (QSARs)investigation of quantitative structure-activity relationship (QSARs)
and is intended to provide all the function require to carry out anyand is intended to provide all the function require to carry out any
QSAR investigation,QSAR investigation,
 TSAR uses an integrated approach to provide all componentsTSAR uses an integrated approach to provide all components
together.together.
 It uses a chemically aware spreadsheet to store and manipulateIt uses a chemically aware spreadsheet to store and manipulate
different type of data, including:different type of data, including:
 Molecular descriptionMolecular description
 3D structures3D structures
 Activity dataActivity data
 Computed dataComputed data

10. SOFTWARE USED INSOFTWARE USED IN
ANALYSIS:ANALYSIS:
 The software are: TSAR software and ISIS/DRAW softwareThe software are: TSAR software and ISIS/DRAW software
 TSAR software: TSAR software of version 3.3 was used to study theTSAR software: TSAR software of version 3.3 was used to study the
QSAR derivatives. It has TSAR project window, to which molecularQSAR derivatives. It has TSAR project window, to which molecular
data is entered through import/export file system. Multipledata is entered through import/export file system. Multiple
regression analysis is done by taking physiochemical propertiesAregression analysis is done by taking physiochemical propertiesA
description of the basic operation of Tsar and fundamental aspectsdescription of the basic operation of Tsar and fundamental aspects
of the application with which you need to familiar, including the Tsarof the application with which you need to familiar, including the Tsar
interface, how to work with projects, data and views. When you startinterface, how to work with projects, data and views. When you start
with Tsar graphical interface, the first screen that is displayed is thewith Tsar graphical interface, the first screen that is displayed is the
main Tsar window and biological activity. Then a graph was plottedmain Tsar window and biological activity. Then a graph was plotted
in between actual values and predicted values.in between actual values and predicted values.
CORINACORINA: The 3D structure of a molecule is closely related to a: The 3D structure of a molecule is closely related to a
large variety of chemical, physical and biological propertlarge variety of chemical, physical and biological propert
 This introduction to CORINA contains the following topics:This introduction to CORINA contains the following topics:
Automatic generation of high quality 3D molecular models providesAutomatic generation of high quality 3D molecular models provides
an introduction to the use of predicting a 3D structurean introduction to the use of predicting a 3D structure

11.  ISIS/DRAWISIS/DRAW: This software has several tools,: This software has several tools,
which are used to draw the chemical structure ofwhich are used to draw the chemical structure of
QSAR derivatives. About 88 molecules wereQSAR derivatives. About 88 molecules were
drawn using ISIS Draw 2.3 software and thedrawn using ISIS Draw 2.3 software and the
descriptors were calculated using Tsar 3.3descriptors were calculated using Tsar 3.3
software.software.
 QSAR regression analysis for this set ofQSAR regression analysis for this set of
molecules was carried out by considering allmolecules was carried out by considering all
molecules as complete set and removing outliermolecules as complete set and removing outlier
component from complete set to generatecomponent from complete set to generate
training set and test set respectively.training set and test set respectively.

12. STRUCTURE OF SOMESTRUCTURE OF SOME
MOLECULESMOLECULES
N
H
O
F
F
F
N N
N
H
O
F
F
F
N N
N
H
O
F
F
F
N N
N
H
O
F
F
F
N N
N
H
O
F
F
F
N N
N
H
O
F
F
F
N N
N
H
O
F
F
F
N N
Compound s_11_9 Compound s_11_10
Compound s_11_11 Compound s_11_12
Compound s_11_13 Compound s_11_14
Compound s_11_15

13. RESULTS ANDRESULTS AND
DISCUSSIONDISCUSSION
 MOLECLE ANALYSISMOLECLE ANALYSIS::
 To cover the whole activity range, the data set was randomly divided into training setTo cover the whole activity range, the data set was randomly divided into training set
and test set .QSAR model was constructed based on training set and then validatedand test set .QSAR model was constructed based on training set and then validated
internally using Leave One Out (LOO) technique and extremely by predicting theinternally using Leave One Out (LOO) technique and extremely by predicting the
activity of test set. The relationship between dependent variable (-log 1/C) andactivity of test set. The relationship between dependent variable (-log 1/C) and
independent variable (physiochemical properties) was established by using linearindependent variable (physiochemical properties) was established by using linear
multiple regression analysis using TSAR 3.3 software. Then significant descriptorsmultiple regression analysis using TSAR 3.3 software. Then significant descriptors
are chosen based on the statistical data analysis.are chosen based on the statistical data analysis.
 COMPLETE SET:COMPLETE SET:
 70 molecules are appended to multiple regression analysis.70 molecules are appended to multiple regression analysis.
 EquationsEquationsOriginal Data : Y = 0.055812515*X6 - 2.7095358*X35 - 0.75705647*X39 -Original Data : Y = 0.055812515*X6 - 2.7095358*X35 - 0.75705647*X39 -
1.8342798*X44 - 2.0094008Standardized Data : Y = 0.81183285*S6 -1.8342798*X44 - 2.0094008Standardized Data : Y = 0.81183285*S6 -
0.35177225*S35 - 0.45452115*S39 - 0.61900699*S44 - 1.21242860.35177225*S35 - 0.45452115*S39 - 0.61900699*S44 - 1.2124286CalculationCalculation
InformationInformation70 rows included in model0 rows excluded because of missing data4970 rows included in model0 rows excluded because of missing data49
independent variables considered0 independent variables excluded because ofindependent variables considered0 independent variables excluded because of
missing data0 independent variables in initial model4 variables included in final modelmissing data0 independent variables in initial model4 variables included in final model
using F-test steppingStandardized by mean/SDCross validated leaving out one rowusing F-test steppingStandardized by mean/SDCross validated leaving out one row
randomly over 2 random trialsCorrelation limit of 0.9 applied4 steps to generate finalrandomly over 2 random trialsCorrelation limit of 0.9 applied4 steps to generate final
modelF to enter = 4, F to leave = 4modelF to enter = 4, F to leave = 4Variance AnalysisVariance AnalysisRegression: 4 degrees ofRegression: 4 degrees of
freedom, sum of squares = 63.947Residual: 65 degrees of freedom, sum of squaresfreedom, sum of squares = 63.947Residual: 65 degrees of freedom, sum of squares
= 13.901Total: 69 degrees of freedom, sum of squares = 77.847= 13.901Total: 69 degrees of freedom, sum of squares = 77.847Statistical TestsStatistical Tests

14.  QSAR EQUATION:QSAR EQUATION:
 log (1/IC50) =log (1/IC50) = + 0.055205099* Inertia Moment 1+ 0.055205099* Inertia Moment 1
LengthLength
 - 2.6556225* Balaban Topological index- 2.6556225* Balaban Topological index
- 0.7120384* ADME H-bond Acceptors- 0.7120384* ADME H-bond Acceptors
 - 1.8028219* VAMP LUMO- 1.8028219* VAMP LUMO
 - 2.0724609- 2.0724609
 r = 0.890, r2 = 0.793, cvr2 = 0.700, F = 50.7138, n = 58,r = 0.890, r2 = 0.793, cvr2 = 0.700, F = 50.7138, n = 58,
PRESS = 19.4898, Residual sum = 13.4604.PRESS = 19.4898, Residual sum = 13.4604.
 Once the multiple regression analysis is performed onOnce the multiple regression analysis is performed on
the complete set and a statistically significant result isthe complete set and a statistically significant result is
obtained, the next step is to perform multiple regressionobtained, the next step is to perform multiple regression
analysis on training set and test set data.analysis on training set and test set data.

15.  TEST SET:TEST SET:
 The test set consists of 12 compounds that are separated from theThe test set consists of 12 compounds that are separated from the
complete set of 58 compounds. The test set compounds arecomplete set of 58 compounds. The test set compounds are
selected based on the hierarchical clustering data so that the totalselected based on the hierarchical clustering data so that the total
biological activity range of the complete set is covered.biological activity range of the complete set is covered.
 The regression equation obtained from the training set is appendedThe regression equation obtained from the training set is appended
to the test set. Thus the activity of the test set is predicted. Theto the test set. Thus the activity of the test set is predicted. The
predictive ability of the model is estimated from the graph plottedpredictive ability of the model is estimated from the graph plotted
from these values. The predicted values and their correspondingfrom these values. The predicted values and their corresponding
actual value is given below in a table:actual value is given below in a table:
 Molecule No.Actual ValuePredicted ValueMolecule No.Actual ValuePredicted Value 1-2.38-2.1105-1-2.38-2.1105-
1.94-1.9768-1.25-1.12216-2.9-2.54118-1.08-0.95219-0.7-0.51022-1.94-1.9768-1.25-1.12216-2.9-2.54118-1.08-0.95219-0.7-0.51022-
0.48-0.33533-1.65-1.53944-3.6-3.19850-0.9-0.6930.48-0.33533-1.65-1.53944-3.6-3.19850-0.9-0.693

16. CONCLUSIONCONCLUSION
 QSAR analysis was performed on 70 aminoquinoline MCH 1RQSAR analysis was performed on 70 aminoquinoline MCH 1R
molecules.Training set (58 molecules) , test set (12 molecules) and outliersmolecules.Training set (58 molecules) , test set (12 molecules) and outliers
(18 molecule ) was generated from the complete set of 70 molecules, each(18 molecule ) was generated from the complete set of 70 molecules, each
containing a set of active ,moderately active and inactive molecules. Acontaining a set of active ,moderately active and inactive molecules. A
regression equation was generated using multiple regression analysis onregression equation was generated using multiple regression analysis on
training set. This regression equation was applied on the test set to predicttraining set. This regression equation was applied on the test set to predict
biological activity of test set molecules. The predicted activity was obtainedbiological activity of test set molecules. The predicted activity was obtained
through the regression equation. The QSAR equation generated bythrough the regression equation. The QSAR equation generated by
considering training set molecules resulted identifying Inertia Moment 1considering training set molecules resulted identifying Inertia Moment 1
Length , Balaban Topological Index , ADME H-bond acceptors , VAMPLength , Balaban Topological Index , ADME H-bond acceptors , VAMP
LUMO .LUMO .
 Eq. 1 accounts for the significant correlation of the descriptors withEq. 1 accounts for the significant correlation of the descriptors with
biological activity and displayed good internal predictivity as shown by q2biological activity and displayed good internal predictivity as shown by q2
value of 0.700 and was able to explain 79.3% variance of inhibitory activitiesvalue of 0.700 and was able to explain 79.3% variance of inhibitory activities
of MCH-1R inhibitors. The predictive ability of QSAR model illustrated theof MCH-1R inhibitors. The predictive ability of QSAR model illustrated the
accuracy and robustness of QSAR model on test set molecules. Therefore,accuracy and robustness of QSAR model on test set molecules. Therefore,
considering the contributions of these descriptors on aminoquinolineconsidering the contributions of these descriptors on aminoquinoline
derivatives would help in designing novel compounds that enhance MCH-1Rderivatives would help in designing novel compounds that enhance MCH-1R
inhibitioninhibition

17. REFERENCESREFERENCES
 REFERENCESREFERENCES
 Chambers J, Ames RS, Bergsma D, Muir A, Fitzgerald LR, HervieuChambers J, Ames RS, Bergsma D, Muir A, Fitzgerald LR, Hervieu
G, Dytko GM, Foley JJ, Martin J, Liu WS, Park J, Ellis C, GangulyG, Dytko GM, Foley JJ, Martin J, Liu WS, Park J, Ellis C, Ganguly
sS, Konchar S, Cluderay J, Leslie R, Wilson S, Sarau HM. Melanin-sS, Konchar S, Cluderay J, Leslie R, Wilson S, Sarau HM. Melanin-
concentrating hormone is the cognate ligand Nature. 1999 Julconcentrating hormone is the cognate ligand Nature. 1999 Jul
15;400(6741):261-515;400(6741):261-5
 http://www.4adi.com/flr/mchrflr.htmlhttp://www.4adi.com/flr/mchrflr.html
 Saito Y, Nothacker HP, Wang Z, Lin SH, Leslie F, Civelli O.Saito Y, Nothacker HP, Wang Z, Lin SH, Leslie F, Civelli O.
Molecular characterization of the melanin-concentrating-hormoneMolecular characterization of the melanin-concentrating-hormone
receptor. Nature. 1999 Jul 15;400(6741):265-9.receptor. Nature. 1999 Jul 15;400(6741):265-9.
 Kawauchi H, Kawazoe I, Tsubokawa M, Kishida M, Baker BI.Kawauchi H, Kawazoe I, Tsubokawa M, Kishida M, Baker BI.
Characterization of melanin-concentrating hormone in chum salmonCharacterization of melanin-concentrating hormone in chum salmon
pituitaries. Nature. 1983 Sep 22-28;305(5932):321-3.pituitaries. Nature. 1983 Sep 22-28;305(5932):321-3.
 Guillaume HervieuGuillaume Hervieu Melanin-concentrating hormone functions in theMelanin-concentrating hormone functions in the
nervous system: food intake and stressnervous system: food intake and stress
http://www.expertopin.com/doi/abs/10.1517/14728222.7.4.495http://www.expertopin.com/doi/abs/10.1517/14728222.7.4.495

18. REFERENCESREFERENCES
 Romain Goutagny , Pierre-Hervé Luppi , Denise Salvert , DamienRomain Goutagny , Pierre-Hervé Luppi , Denise Salvert , Damien
Gervasoni , Patrice Fort. GABAergic control of hypothalamicGervasoni , Patrice Fort. GABAergic control of hypothalamic
melanin-concentrating hormone-containing neurons across themelanin-concentrating hormone-containing neurons across the
sleep-waking cycle. Neuroreport. 2005 Jul 13;16 (10):1069-73sleep-waking cycle. Neuroreport. 2005 Jul 13;16 (10):1069-73
15973150. http://lib.bioinfo.pl/find?15973150. http://lib.bioinfo.pl/find?
field=Library&query=+Melanin+Concentrating+Hormone+active+sfield=Library&query=+Melanin+Concentrating+Hormone+active+s
ite+ite+
 G M Moldavkin , T A Voronina , G G Neznamov , O K Maletova ,G M Moldavkin , T A Voronina , G G Neznamov , O K Maletova ,
N V Eliava. Participation of GABA--benzodiazepine receptorN V Eliava. Participation of GABA--benzodiazepine receptor
complex in the anxiolytic effect of piracetam. Eksp Klincomplex in the anxiolytic effect of piracetam. Eksp Klin
Farmakol. ;69 (3):7-9 16878489. http://lib.bioinfo.pl/meid:24845Farmakol. ;69 (3):7-9 16878489. http://lib.bioinfo.pl/meid:24845
 Agnieszka Basta-Kaim , Monika Leśkiewicz , BogusławaAgnieszka Basta-Kaim , Monika Leśkiewicz , Bogusława
Budziszewska , Władysław Lasoń. The role of neurosteroids inBudziszewska , Władysław Lasoń. The role of neurosteroids in
the central nervous system function. Przegl Lek. 2005 ;62the central nervous system function. Przegl Lek. 2005 ;62
(11):1287-92 16512622. http://lib.bioinfo.pl/meid:(11):1287-92 16512622. http://lib.bioinfo.pl/meid: