- Research article
- Open Access
Interpretable correlation descriptors for quantitative structure-activity relationships
© Spowage et al; licensee BioMed Central Ltd. 2009
Received: 5 November 2009
Accepted: 24 December 2009
Published: 24 December 2009
The topological maximum cross correlation (TMACC) descriptors are alignment-independent 2D descriptors for the derivation of QSARs. TMACC descriptors are generated using atomic properties determined by molecular topology. Previous validation (J Chem Inf Model 2007, 47: 626-634) of the TMACC descriptor suggests it is competitive with the current state of the art.
Here, we illustrate the interpretability of the TMACC descriptors, through the analysis of the QSARs of inhibitors of angiotensin converting enzyme (ACE) and dihydrofolate reductase (DHFR). In the case of the ACE inhibitors, the TMACC interpretation shows features specific to C-domain inhibition, which have not been explicitly identified in previous QSAR studies.
The TMACC interpretation can provide new insight into the structure-activity relationships studied. Freely available, open source software for generating the TMACC descriptors can be downloaded from http://comp.chem.nottingham.ac.uk.
Quantitative structure-activity relationship (QSAR) models correlate molecular chemical structure to biological activity. The underlying principle for QSAR modelling is the similar property principle: molecules with similar chemical structures will exhibit similar biological properties . This principle can be explained by changes in chemical structure altering the electron distribution within a molecule, which is directly responsible for the activity of the molecule. QSARs can be used to elucidate a quantitative description of changes in biological activity arising from the exchange of the functional groups within a molecule. In general, QSAR modelling requires three main features: a data set of molecules, appropriate descriptors and an efficient statistical method for capturing correlation. Descriptors are characteristic properties of molecules, often represented as numerical values, which facilitate the analysis of chemical structure. A wide variety of molecular descriptors are available and descriptor selection is an integral process in QSAR modelling .
2D QSAR models are generated using descriptors derived from the two-dimensional graph representation of a molecule. In contrast, 3D QSAR models correlate activity with descriptors based on spatially localised features. Although 3D descriptors may allow more detailed descriptions of the molecular binding interactions between ligands and receptors, 3D methods are more time-consuming, due to the requirement of precise conformational detail on the molecule and exact alignment . In some cases, 2D QSAR methods can classify the biological activity molecules more efficiently than some more complex 3D QSAR methods . In many instances, the biologically active conformation of a molecule is unknown and 2D descriptors are useful, as they are not dependent upon spatial conformation.
Classic QSAR methods, developed by Hansch , provided a foundation on which numerous QSAR methods are now based: the correlation of physicochemical properties to activity using multivariable regression. Regression analysis models the activities of molecules through an equation constructed using a linear combination of physicochemical properties. The coefficient for each variable in the equation can, consequently, be examined to determine the extent to which each property contributes towards the activity of the molecule. Regression is central to many contemporary QSAR methods, although nowadays often the technique of partial least squares (PLS)  is used to cope with large numbers of descriptors. One of the appeals of regression is the relative ease with which models can be interpreted and this extends to approaches based on PLS . Sometimes an interpretable model might be favoured over a more accurate, but less transparent, QSAR .
Over the last decade, advances in computational technology combined with contemporary methodologies have led to a vast array of new descriptors . Topological maximum cross correlation (TMACC) descriptors were created  with the intention of developing an interpretable 2D descriptor for QSAR modelling. The TMACC descriptors are based on concepts derived from the grid-independent descriptors (GRIND) . GRIND are alignment-independent 3D molecular descriptors which represent a molecule using a grid on which the product of pairs of force field interactions is plotted against the distances between the pairs . This method is analogous to the autocorrelation descriptor, which represents atom pairs as a weighted histogram . GRIND are interpretable, as only one value is stored for each distance range: the maximum product of the two force field interactions. This method was termed maximum auto- and cross-correlation (MACC) . In a similar method, the TMACC descriptors use the topological bond distances and physicochemical properties of a molecule. Only the maximum value calculated as the product of pair combinations of physicochemical properties for each distance is used to generate the TMACC descriptors.
Previous validation of the TMACC descriptors was promising, with leave-one-out (LOO) cross-validated correlation coefficients comparable to those achieved by the state-of-the-art 2D QSAR method, hologram QSAR . An external test set is often used to estimate predictive accuracy . However, the external test set must be large to give results as reliable as cross validation [13–15]. We have previously shown , on the datasets in this study, that using a training/test set partition gives estimates of predictive accuracy that are qualitatively similar to those from cross validation. Thus, here we use cross validation only, as it makes more use of the data for model building. Whilst statistical validation is key, the interpretation and chemical significance of the structure-activity relationships generated are also important [2, 12]. To assess the interpretive ability of a QSAR model it is necessary to apply scientific rationale to the resultant interpretation . Interpretation of the TMACC descriptors is achieved through analysis of the regression model generated by PLS. The predicted activity of a molecule can be attributed to specific atoms that contribute towards the TMACC descriptors. Visualization of the resultant atom activity contributions is accomplished by atomic colour coding based on sign and magnitude of partial activity.
The present study aims to evaluate the ability to identify known structure-activity relationships using the TMACC descriptors. To exemplify the TMACC descriptors, we investigate two datasets, which were previously used in a comprehensive comparison of modern QSAR approaches . Models derived from the angiotensin converting enzyme (ACE) and dihydrofolate reductase (DHFR) inhibitor data sets were assessed to elucidate the encoded structure-activity relationships with the help of information in the literature and to evaluate the interpretive ability of TMACC models.
The data sets used in this study, 114 inhibitors of angiotensin converting enzyme  and 397 inhibitors of DHFR , have been widely used to investigate many QSAR methods . Each data set contained experimentally determined activity (pIC50) values for each molecule. TMACC descriptors were generated using the topological data of each molecule. All nonpolar hydrogen atoms were removed and their atomic value added to the heavy atom to which they were bonded. Polar hydrogen atoms were considered explicitly. Physicochemical properties were then assigned to each atom. Four parameters were used to represent these properties: Gasteiger partial charges , logS parameters , Crippen-Wildman molar refractivity parameters  and Crippen-Wildman partition coefficient (logP) parameters . Gasteiger partial charges were calculated using the method of partial equalization of orbital electronegativity . This procedure calculates atomic charges in σ-bonded and non-conjugated π-systems using only the topology of a molecule. LogS parameters were used to describe atomic contributions to aqueous solubility . Crippen-Wildman molar refractivity (MR) was used as a measure of the steric effect, which is determined through classification of atoms based on neighbouring atoms . Crippen-Wildman partition coefficients (logP) are assigned to each atom as a measure of atomic lipophilicity, determined in the same way as Crippen-Wildman molar refractivity.
Properties used in generation of TMACC descriptors
Gasteiger partial charge
(Solubility and solvation phenomena)
Crippen-Wildman logP (Hydrophobicity)
Crippen-Wildman Molar Refractivity
(Sterics and polarizability)
The activity bands used for TMACC descriptor interpretation
All calculations were performed using Nottingham Cheminformatics Workbench (NCW), a package which provides the function of generating the TMACC descriptors and the TMACC interpretation. NCW is a Java-based application, which is suitable for all major operating systems. It builds upon the original TMACC code (available from our website http://comp.chem.nottingham.ac.uk/download/tmacc). The original software only generates TMACC descriptors; there is no facility to interpret your model. NCW allows the user to start with a set of molecules and perform a full analysis upon them. The popular machine-learning workbench Weka is included to provide PLS modeling, as well as an in-house implementation. The in-house PLS algorithm was written before Weka included one. The results of the PLS analysis are used to determine the atomic contribution of each atom. The interpretation is visualized by a colour scheme depicting activity contribution by atom. The user is able to view molecules individually or tabulated. NCW will be available as open source software for download from our website, http://comp.chem.nottingham.ac.uk/download/ncw. NCW utilizes Marvin for drawing and manipulating chemical structures: Marvin 5.2.2, 2009 http://www.chemaxon.com. All molecular graphics were created using YASARA ("Yet Another Scientific Artificial Reality Application," http://www.yasara.com).
Results and Discussion
Angiotensin converting enzyme (ACE) inhibition
LOO cross-validation of the PLS models generated for the ACE and DHFR data sets gave q2 values of 0.70 and 0.53, respectively, consistent with those previously reported . TMACC descriptors were generated for the ACE data set of 114 inhibitor molecules, previously used to assess the ability of a 3D QSAR method, comparative molecular field analysis (CoMFA) . The data set contained a diverse range of structures and activities selected from literature . ACE is a zinc metallopeptidase, which functions as a dipeptidyl carboxypeptidase, hydrolysing a range of oligopeptide substrates . It acts to induce hypertension and is, consequently, a widely investigated target for antihypertensive drugs . Several methods used to determine the ACE inhibitory activities (IC50 values) involved the use of the substrate hippuryl-histidyl-leucine (HHL) [24, 25]. An early method for determining ACE inhibitor activity was to measure the rate of hippuric acid production from HHL catalyzed by ACE . However, it was subsequently discovered that HHL is a C-terminal domain specific substrate of ACE [27, 28]. Consequently, the structure-activity relationship shown by this data set is likely to reflect that of C-terminal domain specific ACE inhibition, rather than general ACE inhibition.
Frequency of activity of ACE inhibitor features as determined by the TMACC interpretation.
ACE inhibitor feature
C-terminal carboxylate carbonyl
C-terminal carboxylate hydroxyl
Zinc binding carboxylate - carbonyl
Zinc binding carboxylate - hydroxyl
Zinc binding sulfhydryl sulfur
Zinc binding phosphinate phosphorus
Zinc binding phosphinate carbonyl
Zinc binding phosphinate hydroxyl
P1' lysyl nitrogen
An essential feature of any ACE inhibitor is a zinc coordinating group. The catalytic zinc ion is coordinated by three highly conserved residues present in both somatic ACE (sACE) domains . The important functional role of the zinc ion in the active site domains of ACE has led to the development of peptide based inhibitors, such as enalaprilat, with additional zinc-binding functional groups, including thiol, carboxylate and phosphinate groups. The importance of zinc binding functional groups in ACE inhibition has been demonstrated in crystal structure and structure-activity studies [22, 30, 31].
Zinc binding groups were frequently recognized by the TMACC interpretation as positive for activity. All sulfhydryl sulfur atoms located in the optimal position for zinc-binding were identified as positive for activity. Analysis of phosphinate zinc binding groups showed all phosphorus atoms were identified as positive for activity. However, phosphinyl oxygen atoms were mostly shown as negative for activity. In contrast, the interpretation most frequently identified both carboxylate zinc binding group oxygen atoms to be positive for activity. Although the results do not fully capture the correlation between the type of zinc-ligand and inhibitor activity observed in structure-activity studies, (phosphinate > carboxylate > sulfhydryl) , perhaps the negative activity attributed to the phosphinate oxygen atoms reflects its weak zinc-binding ability in comparison to the other zinc binding groups.
The crystal structures of testicular ACE (tACE) in complex with various inhibitors (Figure 3) show the intermolecular interactions responsible for ACE inhibition in tACE and correspondingly the C-terminal domain of sACE [22, 30, 33]. In contrast to most zinc protease inhibitors, which primarily rely on the strength of their zinc binding groups for activity, domain-specific ACE inhibitors utilize weak zinc binding groups and exploit both primed and unprimed sides of the active site in order to mimic peptide substrates, thereby achieving domain selective inhibition . Domain-specific inhibition of ACE is important, as each domain possesses individual functions . This discovery has developed the number of applications of ACE inhibitors, extending from treating hypertension to protecting stem cells during chemotherapy . A recent study has also suggested ACE may be involved in many physiological processes other than blood pressure regulation .
Conserved ACE residues important for inhibitor interactions
Inhibitor carbonyl hydrogen bonding
Inhibitor carbonyl hydrogen bonding
Inhibitor carboxy terminal carboxylic ionic bonding
Inhibitor carboxy terminal carboxylic hydrogen bonding
Inhibitor carboxy terminal carboxylic hydrogen bonding
A C-terminal carboxylate is found in many ACE inhibitors. This feature interacts with several conserved residues in both domains of sACE, hydrogen bonding with tyrosine and glutamine residues, and also forms an electrostatic interaction with a lysine residue (Figure 3) . Both C-terminal carboxylate oxygen atoms were identified as positive by the TMACC interpretation (Table 3).
Interestingly, the TMACC interpretation identified all inhibitor P1' lysyl nitrogen atoms as favourably contributing towards activity (Table 3). The interpretation also identified inhibitor P1' methyl groups as positive for activity. Thus, the TMACC interpretation identified P1' groups important for C-domain specific ACE inhibition, as illustrated in Figure 1. This C-domain specific bias in the data set, reflected by the TMACC interpretation, has not been shown in previous QSAR investigations using this data set [17, 18].
Dihydrofoloate reductase (DHFR) inhibition
Analysis of the TMACC interpretation of the DHFR data set.
DHFR inhibitor feature
N8 Pyrido [2,3-d]pyrimidine
Benzylmethoxy - methyl
Important human DHFR residues for inhibitor binding
2-amino hydrogen bonding
4-amino hydrogen bonding
4-amino hydrogen bonding
N1 hydrogen bonding
Pteridine ring hydrophobic interactions
Pteridine ring hydrophobic interactions
Pteridine ring hydrophobic interactions
Benzene ring hydrophobic interactions
Benzene ring hydrophobic interactions
Benzene ring hydrophobic interactions
Benzene ring hydrophobic interactions
5-methyl hydrophobic interactions
9-methyl and 10-methyl hydrophobic interactions
An important feature for DHFR inhibition is a 2,4-diamino-substituted pyrimidine ring . The 2-amino and 4-amino groups form hydrogen bonds with highly conserved residues, which orientate the pteridine ring of the inhibitor accordingly. The orientation of inhibitor binding differs from natural folate binding. Although both involve 30Glu, the orientation of the inhibitor allows extensive hydrogen bonding with other DHFR residues, which is not possible in the folate binding orientation. Binding of inhibitors containing 2,4-diaminopyrimidine shows the 2-amino group forms a hydrogen bond with 30Glu . Additionally, the 4-amino group of the inhibitor forms strong hydrogen bonds with the carbonyl groups of 7Ile and 115Val [46, 50]. The protonated form of N1 forms an electrostatic interaction with 30Glu [47, 51].
The N8 nitrogen of the pteridine ring forms a conserved network with a water molecule, which also hydrogen bonds with 30Glu and 24Trp [46, 52]. However, comparison of pyrido [2,3-d]pyrimidines and quinazolines suggests that the presence of the N8 hydrogen bond in inhibition by pyrido [2,3-d]pyrimidines may restrict the position of any bridge substituent, such as 9 or 10 methyl groups, preventing hydrophobic interaction of these features with the DHFR active site [53, 54]. Therefore, in general quinazoline DHFR inhibitors are more potent than pyrido [2,3-d] pyrimidine analogues.
The TMACC interpretation showed variable activity for the 2-amino group, with the activity approximately evenly distributed between negative and positive activity. However, interpretation of the 4-amino group showed a greater frequency of positive activity. The interpretation showed the N1 nitrogen as generally negative for activity. This classification could possibly result from the fact that the N1 was unprotonated within most molecules of the data set. However, studies suggest the protonated form of N1 forms a salt bridge with the DHFR active site [51, 55].
Comparison of the activity of C8 atoms in quinazoline inhibitors to N8 atoms in pyrido [2,3-d]pyrimidine inhibitors assigned by the TMACC interpretation showed the C8 atom of quinazoline inhibitors to be identified as positive for activity, whilst the N8 of pyrido [2,3-d]pyrimidines was most frequently negative for activity. This interpretation reflects experimental evidence, which indicates the presence of a C8 (quinazoline) is more potent for DHFR inhibition than an N8 (pyrido [2,3-d]pyrimidine) .
The presence of a 5-methyl group increases inhibitor potency by forming hydrophobic interactions with 115Val in human DHFR . Methyl groups in the 9 and 10 positions of the bridge region also improve DHFR inhibitor potency, as the environment surrounding the bridge region is generally hydrophobic. Therefore, hydrophobic interactions with 22Leu may be formed by these groups . The benzyl ring featured in many antifolate inhibitors forms hydrophobic contacts with many residues within the DHFR active site [46, 51]. This feature is often substituted for other aromatic rings and methoxy groups to increase hydrophobic interactions.
The TMACC interpretation identified 5-methyl groups as positive for activity, consistent with experimental data, which suggests this feature is important for the inhibition of human DHFR. The interpretation also identified methyl groups in the 9 and 10 positions as positive for activity. This is consistent with the known hydrophobic interactions formed by these features within the DHFR active site . The TMACC interpretation identified the methyl of methoxybenzyl groups substituted around the benzoic acid moiety of inhibitor molecules as positive for activity. This classification is supported by the hydrophobic interactions in which methoxy groups have been shown to participate within the DHFR active site [46, 51]. The TMACC interpretation of the DHFR data set identified many key structural features for DHFR inhibition. The analysis suggests that the hydrophobic groups investigated were more frequently identified as positive for activity than the hydrogen bonding groups investigated (Figure 6).
Analyses of the TMACC QSARs modelled for the ACE and DHFR data sets have shown that the TMACC interpretation can identify distinctive features of a structure-activity relationship. The TMACC interpretation provided a clear and precise representation of the activity of specific groups. Amalgamation of the atomic activity values determined for such groups within a data set, showed strong correlation with experimental evidence, which shows the TMACC interpretation can produce models which accurately depict the features of a structure-activity relationship.
Overall, the TMACC interpretation modelled the ACE inhibitor structure-activity relationship highlighted important features for C-domain selective ACE inhibition. The TMACC interpretation provided a consistent representation of the structure-activity relationship present in the ACE data set. However, the insight into the structure-activity relationship of ACE inhibitors produced by the TMACC interpretation was limited by the size of the data set. To obtain a more detailed analysis of components important or detrimental to the ACE inhibitor structure-activity relationship, it would be necessary to investigate a data set which represents a comprehensive range of functional groups and structural components. Investigation of the activity of features important for C-terminal domain selective inhibition in comparison to features important for N-terminal domain selective inhibition would provide further insight into the interpretive ability of the TMACC descriptors.
An inherent weakness, due to the 2D nature of the TMACC descriptor, is insensitivity to chirality. However, the use of chirality descriptors derived from topological data may provide a solution to this limitation and may also improve the predictive ability of the QSAR models . Investigation of alternative or additional atomic properties used in the TMACC descriptor would provide an insight into the properties which contribute towards activity. The effect of implementing more sophisticated partial charge calculations would be interesting, as a recent study has suggested that the method used for partial charge calculations can affect QSAR predictive accuracy . Investigation of a wider range of data sets would provide further validation of the utility of the TMACC interpretation.
We thank GSK for partial funding of a PhD studentship for CLB. We thank Abrar Hussain for technical assistance and helpful discussion. We thank James Melville for the development of the TMACC descriptors and related software. We thank ChemAxon for a license for Marvin.
- Johnson MA, Maggiora GM: Concepts and applications of molecular similarity. 1990, New York, NY, USA: John Wiley and SonsGoogle Scholar
- Guha R: On the interpretation and interpretability of quantitative structure-activity relationships. J Comput Aided Mol Des. 2008, 22: 857-871. 10.1007/s10822-008-9240-5.View ArticleGoogle Scholar
- Gedeck P, Rohde B, Bartels C: QSAR - how good is it in practice? Comparison of descriptor sets on an unbiased cross section of corporate data sets. J Chem Inf Model. 2006, 46: 1924-1936. 10.1021/ci050413p.View ArticleGoogle Scholar
- Brown RD, Martin YC: Use of structure-activity data to compare structure-based clustering methods and descriptors for use in compound selection. J Chem Inf Comput Sci. 1996, 36: 572-584.View ArticleGoogle Scholar
- Hansch C: A quantitative approach to biological structure-activity relationships. Acct Chem Res. 1969, 2: 232-239. 10.1021/ar50020a002.View ArticleGoogle Scholar
- Wold H: Soft modelling: The basic design and some extension. Systems under indirect observations: Causality, structure, prediction. Edited by: Joerskog KG, Wold H. 1982, Amsterdam: North Holland, 1-54.Google Scholar
- Stanton DT: On the physical interpretation of QSAR models. J Chem Inf Comput Sci. 2003, 43: 1423-1433.View ArticleGoogle Scholar
- Guha R, Jurs PC: Development of QSAR models to predict and interpret the biological activity of artemisinin analogues. J Chem Inf Comput Sci. 2004, 44: 1440-1449.View ArticleGoogle Scholar
- Melville JL, Hirst JD: TMACC: Interpretable correlation descriptors for quantitative structure-activity relationships. J Chem Inf Model. 2007, 47: 626-634. 10.1021/ci6004178.View ArticleGoogle Scholar
- Pastor M, Cruciani G, Mclay I, Pickett S, Clementi S: Grid-independent descriptors (GRIND): A novel class of alignment-independent three-dimensional molecular descriptors. J Med Chem. 2000, 43: 3233-3243. 10.1021/jm000941m.View ArticleGoogle Scholar
- Moreau G, Broto P: The autocorrelation of a topological structure: A new molecular descriptor. Nouv J Chim. 1980, 4: 359-360.Google Scholar
- Golbraikh A, Tropsha A: Beware of q(2)!. J Mol Graphics Modell. 2002, 20: 269-276. 10.1016/S1093-3263(01)00123-1.View ArticleGoogle Scholar
- Hawkins DM, Basak SC, Mills D: Assessing model fit by cross-validation. J Chem Inf Comput Sci. 2003, 43: 579-586.View ArticleGoogle Scholar
- Hawkins DM: The problem of overfitting. J Chem Inf Comput Sci. 2004, 44: 1-12.View ArticleGoogle Scholar
- Faber NKM: Estimating the uncertainty in estimates of root mean square error of prediction: Application to determining the size of an adequate test set in multivariate calibration. Chemom Intell Lab Syst. 1999, 49: 79-89. 10.1016/S0169-7439(99)00027-1.View ArticleGoogle Scholar
- Johnson SR: The trouble with QSAR (or how I learned to stop worrying and embrace fallacy). J Chem Inf Model. 2008, 48: 25-26. 10.1021/ci700332k.View ArticleGoogle Scholar
- Sutherland JJ, O'Brein LA, Weaver DF: Comparison of methods for modelling quantitative structure-activity relationships. J Med Chem. 2004, 47: 5541-5554. 10.1021/jm0497141.View ArticleGoogle Scholar
- DePriest SA, Mayer D, Naylor CB, Marshall GR: 3D-QSAR of angiotensin-converting enzyme and thermolysin inhibitors - a comparison of CoMFA models based on deduced and experimentally determined active-site geometries. J Am Chem Soc. 1993, 115: 5372-5384. 10.1021/ja00066a004.View ArticleGoogle Scholar
- Gasteiger J, Marsili M: Iterative partial equalization of orbital electronegativity - a rapid access to atomic charges. Tetrahedron. 1980, 36: 3219-3228. 10.1016/0040-4020(80)80168-2.View ArticleGoogle Scholar
- Hou TJ, Xia K, Zhang W, Xu XJ: Adme evaluation in drug discovery. 4. Prediction of aqueous solubility based on atom contribution approach. J Chem Inf Comput Sci. 2004, 44: 266-275.View ArticleGoogle Scholar
- Wildman SA, Crippen GM: Prediction of physicochemical parameters by atomic contributions. J Chem Inf Comput Sci. 1999, 39: 868-873.View ArticleGoogle Scholar
- Natesh R, Schwager SL, Sturrock ED, Acharya KR: Crystal structure of the human ace-lisinopril complex. Nature. 2003, 421: 551-554. 10.1038/nature01370.View ArticleGoogle Scholar
- Borer JS: Angiotensin-converting enzyme inhibition: A landmark advance in treatment for cardiovascular disease. Eur Heart J Suppl. 2007, 9: E2-E9. 10.1093/eurheartj/sum037.View ArticleGoogle Scholar
- Patchett AA, Harris E, Tristram EW, Wyvratt MJ, Wu MT, Taub D, Peterson ER, Ikeler TJ, Tenbroeke J, Payne LG, Ondeyka DL, Thorsett ED, Greenlee WJ, Lohr NS, Hoffsommer RD, Joshua H, Ruyle WV, Rothrock JW, Aster SD, Maycock AL, Robinson FM, Hirschmann R, Sweet CS, Ulm EH, Gross DM, Vassil TC, Stone CA: A new class of angiotensin-converting enzyme-inhibitors. Nature. 1980, 288: 280-283. 10.1038/288280a0.View ArticleGoogle Scholar
- Ondetti MA, Cushman DW: Angiotensin-converting enzyme-inhibitors - biochemical properties and biological actions. CRC Critical Reviews in Biochemistry. 1984, 16: 381-411. 10.3109/10409238409108720.View ArticleGoogle Scholar
- Cushman DW, Cheung HS: Spectrophotometric assay and properties of angiotensin-converting enzyme of rabbit lung. Biochem Pharmacol. 1971, 20: 1637-1648. 10.1016/0006-2952(71)90292-9.View ArticleGoogle Scholar
- Fuchs S, Xiao HD, Hubert C, Michaud A, Campbell DJ, Adams JW, Capecchi MR, Corvol P, Bernstein KE: Angiotensin-converting enzyme C-terminal catalytic domain is the main site of angiotensin I cleavage in vivo. Hypertension. 2008, 51: 267-274. 10.1161/HYPERTENSIONAHA.107.097865.View ArticleGoogle Scholar
- Wei L, AlhencGelas F, Corvol P, Clauser E: The 2 homologous domains of human angiotensin-I-converting enzyme are both catalytically active. J Biol Chem. 1991, 266: 9002-9008.PubMedGoogle Scholar
- Corradi HR, Schwager SL, Nchinda AT, Sturrock ED, Acharya KR: Crystal structure of the N domain of human somatic angiotensin I-converting enzyme provides a structural basis for domain-specific inhibitor design. J Mol Biol. 2006, 357: 964-974. 10.1016/j.jmb.2006.01.048.View ArticleGoogle Scholar
- Mayer D, Naylor CB, Motoc I, Marshall GR: A unique geometry of the active site of angiotensin-converting enzyme consistent with structure-activity studies. J Comput Aided Mol Des. 1987, 1: 3-16. 10.1007/BF01680553.View ArticleGoogle Scholar
- Natesh R, Schwager SLU, Evans HR, Sturrock ED, Acharya KR: Structural details on the binding of antihypertensive drugs captopril and enalaprilat to human testicular angiotensin I-converting enzyme. Biochemistry. 2004, 43: 8718-8724. 10.1021/bi049480n.View ArticleGoogle Scholar
- Tzakos AG, Gerothanassis IP: Domain-selective ligand-binding modes and atomic level pharmacophore refinement in angiotensin I converting enzyme (ACE) inhibitors. Chem Biochem. 2005, 6: 1089-1103.Google Scholar
- Corradi HR, Chitapi I, Sewell BT, Georgiadis D, Dive V, Sturrock ED, Acharya KR: The structure of testis angiotensin-converting enzyme in complex with the C domain-specific inhibitor RXPA380. Biochemistry. 2007, 46: 5473-5478. 10.1021/bi700275e.View ArticleGoogle Scholar
- Fernandez M, Liu X, Wouters MS, Heyberger S, Husain A: Angiotensin I-converting enzyme transition state stabilization by His(1089): Evidence for a catalytic mechansim distinct from other gluzincin metalloproteinases. J Biol Chem. 2001, 276: 4998-5004. 10.1074/jbc.M009009200.View ArticleGoogle Scholar
- Dive V, Georgiadis D, Matziari M, Makaritis A, Beau F, Cuniasse P, Yiotakis A: Phosphinic peptides as zinc metalloproteinase inhibitors. Cell Mol Life Sci. 2004, 61: 2010-2019. 10.1007/s00018-004-4050-y.View ArticleGoogle Scholar
- van Esch JHM, Tom B, Dive V, Batenburg WW, Georgiadis D, Yiotakis A, van Gool JMG, de Bruijn RJA, de Vries R, Danser AHJ: Selective angiotensin-converting enzyme c-domain inhibition is sufficient to prevent angiotensin I-induced vasoconstriction. Hypertension. 2005, 45: 120-125.View ArticleGoogle Scholar
- Azizi M, Rousseau A, Ezan E, Guyene TT, Michelet S, Grognet JM, Lenfant M, Corvol P, Menard J: Acute angiotensin-converting enzyme inhibition increases the plasma level of the antural stem cell regulator N-acetyl-seryl-aspartyl-lysyl-proline. J Clin Invest. 1996, 97: 839-844. 10.1172/JCI118484.View ArticleGoogle Scholar
- Shen XZ, Xiao HD, Li P, Lin CX, Billet S, Okwan-Duodu D, Adams JW, Bernstein EA, Xu Y, Fuchs S, Bernstein KE: New insights into the role of angiotensin-converting enzyme obtained from the analysis of genetically modified mice. Journal of Molecular Medicine-Jmm. 2008, 86: 679-684. 10.1007/s00109-008-0325-3.View ArticleGoogle Scholar
- Kamen B: Folate and antifolate pharmacology. Semin Oncol. 1997, 24: 30-S18.Google Scholar
- Schweitzer BI, Dicker AP, Bertino JR: Dihydrofolate-reductase as a therapeutic target. FASEB J. 1990, 4: 2441-2452.View ArticleGoogle Scholar
- Kompis IM, Islam K, Then RL: DNA and RNA synthesis: Antifolates. Chem Rev. 2005, 105: 593-620. 10.1021/cr0301144.View ArticleGoogle Scholar
- Curtin NJ, Hughes AN: Pemetrexed disodium, a novel antifolate with multiple targets. Lancet Oncol. 2001, 2: 298-306. 10.1016/S1470-2045(00)00325-9.View ArticleGoogle Scholar
- Allegra CJ, Kovacs JA, Drake JC, Swan JC, Chabner BA, Masur H: Potent invitro and invivo antitoxoplasma activity of the lipid-soluble antifolate trimetrexate. J Clin Invest. 1987, 79: 478-482. 10.1172/JCI112837.View ArticleGoogle Scholar
- Lane HC, Laughon BE, Falloon J, Kovacs JA, Davey RT, Polis MA, Masur H: Recent advances in the management of AIDS-related opportunistic infections. Ann Intern Med. 1994, 120: 945-955.View ArticleGoogle Scholar
- Wang YH, Bruenn JA, Queener SF, Cody V: Isolation of rat dihydrofolate reductase gene and characterization of recombinant enzyme. Antimicrob Agents Chemother. 2001, 45: 2517-2523. 10.1128/AAC.45.9.2517-2523.2001.View ArticleGoogle Scholar
- Klon AE, Heroux A, Ross LJ, Pathak V, Johnson CA, Piper JR, Borhani DW: Atomic structures of human dihydrofolate reductase complexed with NADPH and two lipophilic antifolates at 1.09 angstrom and 1.05 angstrom resolution. J Mol Biol. 2002, 320: 677-693. 10.1016/S0022-2836(02)00469-2.View ArticleGoogle Scholar
- Cody V, Galitsky N, Luft JR, Pangborn W, Gangjee A, Devraj R, Queener SF, Blakley RL: Comparison of ternary complexes of pneumocystis carinii and wild-type human dihydrofolate reductase with coenzyme NADPH and a novel classical antitumor furo[2,3-d]pyrimidine antifolate. Acta Crystallographica Section D-Biological Crystallography. 1997, 53: 638-649. 10.1107/S090744499700509X.View ArticleGoogle Scholar
- Pignatello R, Guccione S, Forte S, Di Giacomo C, Sorrenti V, Vicari L, Barretta GU, Balzano F, Puglisi G: Lipophilic conjugates of methotrexate with short-chain alkylamino acids as DHFR inhibitors. Synthesis, biological evaluation, and molecular modeling. Bioorg Med Chem. 2004, 12: 2951-2964. 10.1016/j.bmc.2004.03.040.View ArticleGoogle Scholar
- Richter WE, McCormac JJ: Inhibition of mammalian dihydrofolate-reductase by selected 2,4-diaminoquinazolines and related compounds. J Med Chem. 1974, 17: 943-947. 10.1021/jm00255a007.View ArticleGoogle Scholar
- Oefner C, Darcy A, Winkler FK: Crystal-structure of human dihydrofolate-reductase complexed with folate. Eur J Biochem. 1988, 174: 377-385. 10.1111/j.1432-1033.1988.tb14108.x.View ArticleGoogle Scholar
- Kovalevskaya NV, Smurnyy YD, Polshakov VI, Birdsall B, Bradbury AF, Frenkiel T, Feeney J: Solution structure of human dihydrofolate reductase in its complex with trimethoprim and NADPH. J Biomol NMR. 2005, 33: 69-72. 10.1007/s10858-005-1475-z.View ArticleGoogle Scholar
- Bolin JT, Filman DJ, Matthews DA, Hamlin RC, Kraut J: Crystal-structures of escherichia-coli and lactobacillus-casei dihydrofolate-reductase refined at 1.7 Å resolution .1. General features and binding of methotrexate. J Biol Chem. 1982, 257 (22): 13650-13662.PubMedGoogle Scholar
- Gangjee A, Vidwans AP, Vasudevan A, Queener SF, Kisliuk RL, Cody V, Li RM, Galitsky N, Luft JR, Pangborn S: Structure-based design and synthesis of lipophilic 2,4-diamino-6-substituted quinazolines and their evaluation as inhibitors of dihydrofolate reductases and potential antitumor agents. J Med Chem. 1998, 41: 3426-3434. 10.1021/jm980081y.View ArticleGoogle Scholar
- Cody V, Galitsky N, Luft JR, Pangborn W, Queener SF, Gangjee A: Analysis of quinazoline and pyrido[2,3-d]-pyrimidine n9-c10 reversed-bridge antifolates in complex with NADP(+) and pneumocystis carinii dihydrofolate reductase. Acta Crystallographica Section D-Biological Crystallography. 2002, 58: 1393-1399. 10.1107/S0907444902010442.View ArticleGoogle Scholar
- Bennett B, Langan P, Coates L, Mustyakimov M, Schoenborn B, Howell EE, Dealwis C: Neutron diffraction studies of escherichia coli dihydrofolate reductase complexed with methotrexate. Proc Natl Acad Sci USA. 2006, 103: 18493-18498. 10.1073/pnas.0604977103.View ArticleGoogle Scholar
- Gangjee A, Adair O, Queener SF: Synthesis of 2,4-diamino-6-(thioarylmethyl)pyrido [2,3-d]pyrimidines as dihydrofolate reductase inhibitors. Bioorg Med Chem. 2001, 9: 2929-2935. 10.1016/S0968-0896(01)00223-1.View ArticleGoogle Scholar
- Golbraikh A, Tropsha A: QSAR modeling using chirality descriptors derived from molecular topology. J Chem Inf Comput Sci. 2003, 43: 144-154.View ArticleGoogle Scholar
- Mittal RR, Harris L, McKinnon RA, Sorich MJ: Partial charge calculation method affects CoMFA QSAR prediction accuracy. J Chem Inf Model. 2009, 49: 704-709. 10.1021/ci800390m.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.