Comparative QSTR studies for predicting mutagenicity of nitro compounds

https://doi.org/10.1016/j.jmgm.2007.06.006Get rights and content

Abstract

Mutagenicity and carcinogenicity are toxicological endpoints which pose a great concern being the major determinants of cancers and tumours. Nitroarenes possess genotoxic properties as they can form various electrophilic intermediates and adducts with biological systems. Different QSTR techniques were employed to develop models for the prediction of mutagenicity of nitroarenes using a diverse set of 197 nitro aromatic and hetero aromatic molecules. The 2D and 3D QSTR methods used for model development gave statistically significant results. The alignment for 3D methods was obtained by maximum common substructures (MCS) approach, by taking the most mutagenic molecule of the dataset as the template. All the QSTR models were developed with the same set of training and test set molecules. The 3D contours and 2D contribution maps along with molecular fingerprints provide useful information about the mutagenic potentials of the molecules. The GFA based model shows thermodynamic and topological descriptors play an important role in characterizing mutagenicity of nitroarenes. Atomic-level thermodynamic descriptor namely AlogP throws light on hydrophobic features and helps to understand the bilinear model. Topological aspects of these classes of compounds were depicted by the fragment fingerprints and Balaban indices obtained from HQSAR and GFA models, respectively. The predictive abilities of 2D and 3D QSTR models may be useful as a vibrant predictive tool to screen out mutagenic nitroarenes and design safer non-mutagenic nitro compounds.

Introduction

Over the past two decades, there has been a vast increase in the development of SAR techniques. The linear free-energy approaches developed by Hansch and Free-Wilson have provided a fundamental scientific framework for the study of structure activity relationships. The QSAR techniques involve the use of 2D or 3D descriptors such as electron distribution, spatial disposition molecular volume, hydrophobicity, steric feature, solubility and ionization constants for developing robust models [1], [2], [3]. The well recognized 3D-QSAR methods, CoMFA and CoMSIA have received much attention in drug discovery scenario [4], [5]. More recently, the QSAR techniques were extrapolated to predict the other end points like toxicity and pharmacokinetic properties based on the molecular structures and named as QSTR and QSPR analysis [6], [7]. Developing QSTR models for toxic end points are encouraging in particular to overcome the time and cost associated with the animal studies for screening the new compounds. However, there are many difficulties one has to face while developing QSTR models, one being the diversity of the molecules and other being the varied mechanisms by which the molecules cause the toxicological effect. In addition, QSAR methods do not necessarily model interactions with receptor and do not highlight the exact mechanism involved in complex enzymatic steps observed in a toxic scenario. But they do give some information of the involved mechanism based on the molecular similarity and this was recently pointed out by Gieleciak and Polanski [8]. There are also other challenges exist which include predicting test and new molecules. The toxicological properties like hepatotoxicity, carcinogenicity and reproductive effects are mechanistically unclear leading to added complexity for prediction. Owing to these reasons there are not many successful QSTR models available in the literature. However, there are more than a few QSTR models reported for predicting toxicity for individual chemical classes such as quinolines, nitroarenes, aromatic amines, triazenes, polycyclic aromatic hydrocarbons and lactones which undergo similar mechanisms to cause toxicity.

Mutagenicity and carcinogenicity are toxicological endpoints which pose a great concern being the major determinants of cancers and tumours [9]. Carcinogenic process involves one or more mutations showing a relationship with mutagenicity [10], [11], [12]. Zeiger performed an analysis on 224 chemicals and revealed that all nitro carcinogens were mutagenic [12] in behaviour. Several groups have used log revertants data from Ames test of mutagenicity for building QSTR models of mutagenic compounds [13], [14], [15], [16], [17], [18]. The Ames test (Salmonella typhimurium his reversion assay) is a simple and less costly method for finding out mutagenic potential of the compounds. It is performed on different strains of S. typhimurium depending on the mechanism and metabolic activation of different classes of carcinogens [19], [20], [21], [22]. Literature studies reveal that S. typhimurium TA98 undergoes mutations in presence of nitro compounds; therefore one can use the TA98 strain for the generation of QSTR models [2], [23]. Chemical carcinogens are either genotoxic or non-genotoxic based on their mechanism of action [24]. Genotoxic carcinogens directly cause damage to DNA and many known mutagens belong to this category [25]. Non-genotoxic carcinogens do not bind covalently to DNA, so do not directly cause damage to the DNA.

We report here the results of the QSTR studies carried out on a dataset of 197 nitroarene compounds. Nitroarenes are environmental pollutants released from automobile exhausts and industrial areas, proved to be potent mutagens or carcinogens [26], [27], [30]. There are also examples in the literature to show drugs and drug-like molecules have nitro substitutions which may cause toxicological influence [28], [29]. Nitroarenes possess genotoxic properties as they can form various electrophilic intermediates and adducts with DNA, tissue proteins, blood proteins albumin and haemoglobin [25]. The proposed mechanism for nitro compounds involves transformation of nitro into a hydroxylamine intermediate. This results in an electrophilic nitrogen species interacting with DNA which is mediated by cytosolic reductases. The objective of the present work is to develop comparative QSTR models using 2D and 3D QSAR methods to get better insight into the mutagenicity of nitro compounds. This study may help to understand the influence of fragments fingerprints and other physicochemical descriptors on the mutagenic potential of nitroarenes. This may also aid to analyse the effect of various fields of interactions of CoMFA and CoMSIA on mutagenicity of nitroarenes. Hansch et al. published a conventional QSAR study on series of nitroarenes; however, we have approached this problem with the more advanced 2D QSAR and 3D QSAR methods, viz. HQSAR, GFA, CoMFA and CoMSIA.

Section snippets

Dataset for analysis

Mutagenicity data on S. typhimurium TA98 strain in log revertants/nmol was used for the study. The average log revertants/nmol values for 197 nitroarenes from published data by Goldring et al., Rosenkranz et al., Zielinska et al. [31], [32], [33] and Hansch et al. where taken for developing QSAR model (Table 1) [34]. Reasons for considering average values were to reduce the experimental errors which could be because of variation in assay techniques or laboratory conditions. Dataset of molecules

Computational details

Molecular modeling studies were performed using the molecular modeling package SYBYL7.1 [35] and Cerius2 4.9 [36] installed on a Silicon Graphics Fuel Workstation. The structures were minimized using Powell's conjugate gradient method with Tripos force field and Gasteiger-Huckel partial atomic charges [37], [38]. The minimum energy difference of 0.05 kcal/mol was set as a convergence criterion.

CoMFA and CoMSIA

CoMFA and CoMSIA analysis were performed for a dataset comprising 197 nitroarenes. The dataset contains two to six fused member ring systems making manual alignment difficult and selection of template structure more critical. So the molecules were aligned using maximum common substructure (MCS), based on the template molecule 183. A statistically significant model was obtained for the dataset containing 142 and 55 molecules in the training and test set, respectively. Table 2 shows the

Comparison of QSTR models

The analysis of the QSTR model shows that some classes of nitroarenes are predicted better by 3D QSAR method and some by 2D methods. We studied the predictive ability of different models by analyzing the different classes of nitroarenes using the residual values which are the difference between the actual and predicted log revertants. We used the cut-off as residual >1 for this analysis. Considering the pyrene class of molecules which forms the largest subset, we found CoMFA and CoMSIA

Conclusions

The present work focuses on use of 2D (HQSAR and GFA) and 3D QSAR (CoMFA and CoMSIA) methods to develop predictive models for a structurally diverse set 197 nitroarene molecules. The models developed in the study are statistically significant despite diversity of compound series used for the analysis. The findings from the 3D models shows, to form a stable complex between the mutagen and the DNA, the mutagens must have suitable electronic feature, hydrogen acceptor character, steric bulk and

References (57)

  • H.S. Rosenkranz et al.

    Mutagenicity and genotoxicity of nitroarenes. All nitro-containing chemicals were not created equal

    Mutat. Res.

    (1983)
  • B. Zielinska et al.

    Mutagenic activities of selected nitrofluoranthene derivatives in Salmonella typhimurium strains TA98, TA98NR and TA98/1,8-DNP6

    Mutat. Res.

    (1988)
  • J. Gasteiger et al.

    Iterative partial equalization of orbital electro negativity—a rapid access to atomic charges

    Tetrahedron

    (1980)
  • G. Klebe

    The use of composite crystal-field environments in molecular recognition and the de novo design of protein ligands

    J. Mol. Biol.

    (1994)
  • C. Glende et al.

    Transformation of mutagenic aromatic amines into non-mutagenic species by alkyl substituents. Part II: alkylation far away from the amino function

    Mutat. Res.

    (2002)
  • M. Klein et al.

    From mutagenic to non-mutagenic nitroarenes: effect of bulky alkyl substituents on the mutagenic activity of 4-nitrobiphenyl in Salmonella typhimurium—part I. Substituents ortho to the nitro group and in 2′-position

    Mutat. Res.

    (2000)
  • M. Nohara et al.

    Relationship between mutagenic potency in Salmonella typhimurium and chemical structure of amino- and nitro-substituted biphenyls

    Mutat. Res.

    (1985)
  • A.K. Debnath et al.

    International Commission for Protection Against Environmental Mutagens and Carcinogens. The importance of the hydrophobic interaction in the mutagenicity of organic compounds

    Mutat. Res.

    (1994)
  • M. Shimizu et al.

    Structural specificity of aromatic compounds with special reference to mutagenic activity in Salmonella typhimurium—a series of chloro- or fluoro-nitrobenzene derivatives

    Mutat. Res.

    (1983)
  • R. Benigni

    Structure–activity relationship studies of chemical mutagens and carcinogens: mechanistic investigations and prediction approaches

    Chem. Rev.

    (2005)
  • S. Khanna et al.

    Additivity of molecular fields: CoMFA study on dual activators of PPARalpha and PPARgamma

    J. Med. Chem.

    (2005)
  • M. Bohm et al.

    Three-dimensional quantitative structure–activity relationship analyses using comparative molecular field analysis and comparative molecular similarity indices analysis to elucidate selectivity differences of inhibitors binding to trypsin, thrombin, and factor Xa

    J. Med. Chem.

    (1999)
  • M.L. Hanson et al.

    New technique for estimating thresholds of toxicity in ecological risk assessment

    Environ. Sci. Technol.

    (2002)
  • R. Gieleciak et al.

    Modeling robust QSAR. 2. Iterative variable elimination schemes for CoMSA: application for modeling benzoic acid pKa values

    J. Chem. Inf. Model.

    (2007)
  • L. Tomatis et al.

    Evolution of cancer etiology and primary prevention

    Environ. Health Perspect.

    (2001)
  • J. McCann et al.

    Detection of carcinogens as mutagens in the Salmonella/microsome test: assay of 300 chemicals

    Proc. Natl. Acad. Sci.

    (1975)
  • J. McCann et al.

    Detection of carcinogens as mutagens: bacterial tester strains with R factor plasmids

    Proc. Natl. Acad. Sci.

    (1975)
  • E. Zeiger

    Carcinogenicity of mutagens: predictive capability of the Salmonella mutagenesis assay for rodent carcinogenicity

    Cancer Res.

    (1987)
  • Cited by (22)

    • The occurrence, distribution and removal of adsorbable organic halogens (AOX) in a typical fine chemical industrial park

      2022, Environmental Pollution
      Citation Excerpt :

      Among these AOX pollutants, 40% are carcinogenic, 60% are mutagenic and 100% are toxic to reproduction. The presence of nitro groups on the benzene ring structure may lead to mutagenicity of AOX pollutants, which has also been confirmed in other literatures (Nair and Sobhia, 2008; Sabbioni, 1994). The United Nations created the Globally Harmonised System of Classification and Labelling of Chemicals (GHS) as an internationally agreed system to replace the various classification and labelling standards used in different countries (United Nations, 2003).

    • In silico prediction of the mutagenicity of nitroaromatic compounds using a novel two-QSAR approach

      2017, Toxicology in Vitro
      Citation Excerpt :

      In fact, various packages and theoretical models have been proposed to quantitatively predict mutagenicity of Ames test (Hillebrecht et al., 2011; Xu et al., 2012). Also, a number of comparative molecular field analysis (CoMFA), molecular similarity indices analysis (CoMSIA), genetic function approximation (GFA), hologram QSAR (HQSAR), and classical QSAR models have been developed to predict mutagenicity of NACs (Caliendo et al., 1995; Debnath et al., 1991; Fan et al., 1998; King et al., 1996; Klein et al., 2000c; Lopez de Compadre et al., 1998; Lopez de Compadre et al., 1990; Lopez De Compadre et al., 1988; Maynard et al., 1986; Nair and Sobhia, 2008; Wang et al., 2005). Linear predictive models, such as linear partial least square (PLS), are useful to interpret the relationship between descriptor and biological activity (Hasegawa and Funatsu, 2010).

    • In silico prediction of toxic action mechanisms of phenols for imbalanced data with random forest learner

      2012, Journal of Molecular Graphics and Modelling
      Citation Excerpt :

      The evaluation of phenol toxicity has become an area of active research in the past decades. Recently, in silico methods, such as quantitative structure–activity relationships (QSAR) and quantitative structure–toxicity relationships (QSTR), bring many benefits for toxicity exploration [4,5]. On the one hand, the high cost and time needed to conduct the safety testing for a large volume of chemicals are heavily reduced by in silico methods.

    View all citing articles on Scopus
    View full text