Experimental Testing of Quantum Mechanical Predictions of

Article pubs.acs.org/crt

Experimental Testing of Quantum Mechanical Predictions of Mutagenicity: Aminopyrazoles Andrew G. Leach,*,† William McCoull,* Andrew Bailey, Peter Barton, Christine Mee, and Eleanor Rosevere AstraZeneca, Alderley Park, Macclesfield, Cheshire SK10 4TG, United Kingdom S Supporting Information *

ABSTRACT: A computational method for predicting the likelihood of aromatic amines being active in the Ames test for mutagenicity was trialed on a set of aminopyrazoles. A virtual array of compounds was generated from the available sets of hydrazines and α-cyanoaldehydes (or ketones) and quantum mechanical calculations used to compute a probability of being active in the Ames test. The compounds selected for synthesis and testing were not based on the predictions and so spanned the range of predicted probabilities. The subsequently generated results of the Ames test were in good correspondence with the predictions and confirm this approach as a useful means of predicting likely mutagenic risk.

■

INTRODUCTION Aromatic amines are a commonly occurring motif in compounds of pharmaceutical interest. They may be present as the undecorated amine itself or embedded in features such as amides, sulfonamides, or alicyclic rings. While these moieties are often encountered in drug discovery programs, they can cause concern as aromatic amines are frequently mutagenic.1 Where the aromatic amine is embedded in a molecule, the risk that it causes is dependent upon two factors: whether the amine is liberated in vivo and whether the amine is mutagenic.1,2 A drug discovery project at AstraZeneca recently discovered a series that contained amides of amino pyrazoles such as 1. Amides are readily cleaved by a range of metabolic enzymes (Scheme 1), and so, aminopyrazole 2 was judged likely to be liberated to some degree or other. Hence, aminopyrazoles such as 2 that are nonmutagenic were sought. In order to identify aromatic amines that are mutagenic as early as possible in a drug discovery project, the Ames test is used.3−6 This test employs a number of strains of bacteria that have mutations which prevent them from synthesizing a

particular amino acid. If grown in a medium lacking that amino acid, they are unable to grow. When exposed to particular kinds of mutagens that cause reversion to a form that is once again able to synthesize the amino acid, they grow. This growth of colonies can be quantified and used to assess the mutagenic potential of compounds. In the assessment of aromatic amines, two strains (Salmonella typhimurium TA98 and TA100) are able to detect the normal mechanism of mutagenicity described below and lack the ability to make histidine. The tests described here use these two strains only. Many aromatic amines are not mutagenic themselves but are metabolized to generate species that are. This effect is mimicked by also conducting an assay in the presence of the S9 fraction of rat liver after treatment with Aroclor 1254, a broad spectrum enzyme inducer.5−7 Compounds that cause a 2-fold or larger increase in the number of revertant colonies in any of the tests are classed as active. It has been known for some time that a common chemical mechanism causing the mutagenicity of aromatic amines (3) is that shown in Scheme 2.8−15 Metabolic oxidation yields hydroxylamines (4) that are subsequently conjugated in such a way that the hydroxyl is converted into a leaving group (5). This activated species can react as an electrophile with DNA being the corresponding nucleophile. This chemical alteration of DNA is the mutagenic event. This reaction may be SN1 or SN2. The former would proceed through the transient formation of a nitrenium species 6 (Scheme 2). The latter would involve a transition state (7) in which the nitrogen bears a positive charge and would therefore be expected to show

Scheme 1. General Structure of the Amides of Pyrazole Amines Identified as Leads in a Drug Discovery Program and the Amines That Would Be Liberated from Them in Vivo

Received: December 19, 2012 Published: March 29, 2013 © 2013 American Chemical Society

703

dx.doi.org/10.1021/tx3005136 | Chem. Res. Toxicol. 2013, 26, 703−709

Chemical Research in Toxicology

■

Scheme 2. Normal Mechanism by Which Aromatic Amines Cause Covalent Modification of DNA

Article

MATERIALS AND METHODS

Ames Test. Standard plate incorporation assays were performed using Salmonella typhimurium strains TA98 and TA100 according to published methods.5,6 Tests were performed with amounts up to a maximum of 5000 μg/plate in the absence and presence of S9 from livers of rats pretreated with Aroclor 1254 (purchased form Moltox Inc., Boone, NC, U.S.A.). All test compounds were dissolved in DMSO. In all tests, there were three plates for the solvent control, for each level of test compound, and for the positive control groups. The S9-independent positive controls were sodium azide at 0.5 μg/plate for TA100 and 2-nitrofluorene at 0.5 μg/plate for TA98, except where otherwise indicated. The positive control in tests including S9 was 2aminoanthracene at 2 μg/plate for both strains. All of the compounds were tested at AstraZeneca, Alderley Park, U.K. Computational Methods. The molecules, 2, were generated virtually as SMILES strings. The calculated properties (e.g., lipophilicity, hydrogen bonding, and molecular weight) of the amides that could be generated from these amines were computed by the internal c-lab platform.22 These were filtered to appropriate ranges for the project. All minima were verified as having all positive frequencies. All quantum mechanical calculations were performed in the Gaussian 09 software suite.23 Initial geometries were generated from SMILES strings by CORINA.24 Two conformations about the Ar−N bond were generated for both Ar-NHOAc and Ar-NH+. Full coordinates and energies are provided in the Supporting Information. All calculations used the B3LYP/6-31G* method and included thermal corrections to 298K.25−27 Chemicals. Compounds were either purchased from commercial suppliers and repurified prior to Ames testing or synthesized. Generally, compounds 16 containing 4-cyano functionality were synthesized by condensation of (ethoxymethylene)malononitrile 14 with the corresponding hydrazine 15 in refluxing methanol (Scheme 4). Compound 19 was synthesized by reaction of 2-aminoprop-1-ene-

similar dependencies of activity upon structure as the SN1 reaction. This chemical reactivity can be studied using quantum mechanical calculations. Such calculations, performed by ourselves and others, show that for aromatic amines with molecular weight less than 250 Da the dissociation energy of a model activated conjugate ArNHOAc to form the nitrenium ArNH+ correlates with a likelihood of being active in the Ames test.13,14,16−20 In particular, the dissociation energy ΔE (shown in Scheme 3) computed at B3LYP/6-31G* including thermal Scheme 3. Chemical Reaction for Which the Computed Energy Change Is Found to Correlate Well with the Likelihood of Being Active in the Ames Test

corrections to energy at 298 K correlates with the proportion of compounds that are active in the Ames test.17 This permits a prospective probability of being active in the Ames test, P(active), to be generated using eq 1. 16 100 P(active) = (0.056ΔE − 8.012) (1 + e ) (1)

Scheme 4. General Synthesis of Compounds 16a

a

Previous studies of aminopyrazoles in the context of a program aimed at the discovery of glucose kinase activators found that the unsubstituted parent aminopyrazole 10 is not active in the Ames test and is computed to have a low probability of being active (Figure 1).21 When alkylated on the

Reagents and conditions: MeOH, reflux.

1,1,3-tricarbonitrile 17 with 2-hydrazinylpyridine 18 in refluxing ethanol (Scheme 5). All hydrazines were commercially available with

Scheme 5. Synthesis of Compound 19a

a

Figure 1. Four aminopyrazoles investigated previously using the same methods as those described in the current work.

Reagents and conditions: EtOH, reflux.

the exception of 2-hydrazinyl-3-methoxypyridine, which was synthesized by the reaction of hydrazine with 2-chloro-3-methoxypyridine. All solvents and chemicals used were of reagent grade. Anhydrous solvents were purchased from Sigma Aldrich. Flash column chromatography was carried out using Redisep or Crawford prepacked silica cartridges (4−330 g), and elution was with an Isco Companion system. Following isolation, compounds were purified to >95% purity (UV and NMR) by silica gel chromatography and reverse phase preparative high performance liquid chromatography (HPLC) purification carried out using a Waters XBridge Prep C18 OBD column, 5 μm silica, 50 mm diameter, and 150 mm length.

nitrogen distal to the amino group with Me, Et, or iPr, aminopyrazoles 11 to 13 are active in the Ames test and are computed to have higher probabilities of being active. The variation caused by changing R1, R2, and R3 in 2 has not been well studied, and it was our intention to predict the activity of a range of these compounds and then to conduct an Ames test on those that were subsequently prepared, without selecting only compounds less likely to be active. This should provide an unbiased test of the predictions. 704



Article

Mass spectrometry data were recorded by liquid chromatography− mass spectrometry (LCMS) on a Waters 2790 separations module with a Phenomenex 5 μm C18 50 mm × 2 mm column, a Waters 996 photodiode array detector, and Waters Micromass ZQ mass spectrometer, with detection by UV at 254 nm. Purity was >95% for all test compounds as determined by this HPLC/MS method. 1H NMR spectra were recorded on a Bruker Ultrashield 400 Plus at 400 MHz in the indicated deuterated solvent. Chemical shifts are reported in ppm relative to tetramethylsilane (TMS) (0.00 ppm) or solvent peaks as the internal reference, and coupling constant (J) values are reported in hertz (Hz). Merck precoated thin layer chromatography (TLC) plates were used for TLC analysis. For workup, solutions were dried over anhydrous magnesium sulfate, and the solvent was removed by rotary evaporation under reduced pressure. The general method by which all compounds were prepared is described for exemplar compound 35. Full experimental details for all other compounds synthesized are provided in the Supporting Information. General Method. 5-Amino-1-(6-methylpyrimidin-4-yl)-1H-pyrazole-4-carbonitrile (35). 2-(Ethoxymethylene)malononitrile (0.98 g, 8.1 mmol) was added to 4-hydrazinyl-6-methylpyrimidine (1 g, 8.1 mmol) in MeOH (45 mL) at room temperature (rt) under nitrogen. The resulting solution was stirred at 65 °C for 2 h during which time a precipitate formed. The reaction was cooled to rt and the solid was filtered off, washed with MeOH (30 mL), then dried under high vacuum to afford the title compound (1.08 g, 67%) as a pale brown solid. One hundred sixty milligrams of this material was recrystallized from EtOH to yield 138 mg of product as a yellow solid. 1H NMR (400 MHz, DMSO-d6) δ 2.60 (s, 3H), 7.80 (s, 1H), 8.04 (s, 1H), 8.33 (s, 2H), 8.99 (d, J = 1.1 Hz, 1H). HRMS (ESI) calculated for C9H9N6 ([M + H]+): 201.08832. Found: 201.08826. HPLC purity = 100%. 5-Amino-3-(cyanomethyl)-1-(pyridin-2-yl)-1H-pyrazole-4-carbonitrile (19). A suspension of 2-aminoprop-1-ene-1,1,3-tricarbonitrile (4.9 g, 37.1 mmol) in ethanol (40 mL) was heated to reflux for a few minutes until all of the solid had dissolved. The heat was removed, and a solution of 2-hydrazinylpyridine (4.45 g, 40.8 mmol) in ethanol (10 mL) was added dropwise to the warm solution over 10 min. The reaction mixture was refluxed for 1 h and then allowed to cool to rt and stand at rt for 3 days. The resultant crystals were filtered off and washed with fresh ethanol and then ether to afford the title compound (3.35 g, 40%). Three hundred milligrams of this material was recrystallized from acetonitrile to yield 114 mg of pure sample. 1H NMR (400 MHz, DMSO-d6) δ 4.14 (s, 2H), 7.36 (ddd, J = 0.9, 5.0, 7.4 Hz, 1H), 7.80 (d, J = 8.4 Hz, 1H), 8.03 (ddd, J = 1.9, 7.5, 8.4 Hz, 1H), 8.22 (s, 2H), 8.46 (ddd, J = 0.7, 1.8, 5.0 Hz, 1H). HRMS (ESI) calculated for C11H9N6 ([M + H]+): 225.08832. Found: 225.08830. HPLC purity = 100%.

(the full distribution is shown in Figure 2). McCarren et al. have shown that the approach described here is best applicable

Figure 2. Molecular weight distribution of the full enumerated set of pyrazole amines.

to aromatic amines with molecular weight less than 250 Da, and so, this set of 800 selected for study also coincides with those most likely to be well described by this model.18 The quantum mechanical calculations were performed in Gaussian 09 on geometries generated by CORINA.23,24 Rotations about the bond indicated that Ar-NHOAc and Ar-NH+ were generated and all other conformational degrees of freedom kept constant at the geometry created by CORINA. This set of 800 aromatic amines were treated with the quantum mechanical calculations, and these completed successfully in the allotted time for 753 examples. Six hundred and fifteen of the examples completed the optimization and frequency calculations for all four conformations, while the remaining examples completed only three out of the four. The distribution of the computed values of ΔE is shown in Figure 3a. A set of 24 examples were selected for synthesis or were already available within the AstraZeneca corporate collection, and the computed values of ΔE for these are shown in Figure 3b. The mean ± standard deviation is 154.7 ± 8.3 for the full set and 154.2 ± 9.2 for the selections. As can be seen from the distributions shown in Figure 3, the selections are representative of the full set while being skewed slightly toward compounds with a higher value of ΔE. This set of compounds were made and tested, and the results summarized in Table 1 alongside the computed values of ΔE and the corresponding P(active). The full experimental results are provided in the Supporting Information. In line with the mechanism for mutagenicity outlined in Scheme 2, all compounds are inactive in the absence of metabolic activation. In contrast to the aminobiphenyls recently described, these aminopyrazoles tend to be more active in the TA100 strain than in the TA98.16 Some SAR for activity in the Ames test can be deduced by inspection of the experimental data in Table 1. The R2 nitrile group generally causes the aminopyrazole to be negative in the Ames test. Even when the compound bears a nitrile at R2, the compound can be positive in the Ames test if an electron donating aromatic group is incorporated (see compound 21 compared to compound 20). More nuanced deduction is difficult but can be facilitated by the calculations, as described below. In addition to empirical observations about SAR, the data set in Table 1 also provides an assessment of the ability of the

■

RESULTS AND DISCUSSION The enumeration of possible compounds such as 16, based on commercially available and AstraZeneca proprietory hydrazines, resulted in a set of 3531 for study with quantum mechanics. These were filtered to remove compounds that included a second aromatic amine. Previous SAR had suggested that compounds in which R1 is an aromatic group are favored for on-target potency. Aromatic groups with an aza nitrogen adjacent to the point of connection to the pyrazole were further favored; this was believed to derive from the ability of such groups to form an internal hydrogen bond with the amide NH thereby fixing the conformation of both groups. The set was significantly biased in light of these considerations. Compounds containing the elements Br or I were also removed for computational convenience. This resulted in a set of 2719 aminopyrazoles. In order to focus on examples that were more likely to yield drug-like compounds when incorporated into amides 1, the lowest molecular weight 800 were selected for study. It is worth noting that the first 800 aromatic amines when ordered by molecular weight all weigh less than 280 Da 705



Article

Figure 3. (a) Values of ΔE computed for 753 lower molecular weight examples from the enumeration of pyrazole amines. (b) Values of ΔE computed for the set of pyrazole amines that were made and tested in an Ames test.

that by selecting compounds with lower probabilities of being active, a higher success rate is to be expected. It is not routine at AstraZeneca to analyze the output of the Ames test in more detail than to categorize compounds as positive or negative, but this can be done to provide further insight into the value of the calculations described here. Within the assay output that is provided in Supporting Information, all of the points in the assay including metabolic activation at which the difference between the treated and control is separated by more than 1.96 times the sum of the standard errors in the mean of two values were identified. These are points where the treated is different from the control at approximately 95% certainty. Some of the inactive compounds did not yield any such points. In each case, mean values arise from measurements made in triplicate such that the standard error in the mean is equal to the reported standard deviation divided by the square root of 3. The maximum fold increase observed at any of these points for each compound is extracted and their logarithms plotted against the ΔE values in Figure 6. These plots reveal that there is a general trend for compounds with lower values of ΔE to be more mutagenic for both TA98 and TA100 strains. They also show that compound 34 discussed above is only just detected as being mutagenic, and compound 19, which has a computed ΔE similar to that for 34, falls only just the other side of the experimetal cutoff of a 2-fold increase and is classified as inactive. The ΔE values provide a quantitative as well as a qualitative relationship with mutagenicity. Within the initial set of 753 compounds, there is a large amount of further SAR information. The set is dominated by compounds with R2 being cyano and R3 hydrogen, and 704 different groups at R1. A Free-Wilson type model can be built to evaluate group contributions to activity but the uneven sampling of groups, particularly the large number of R1 groups that are represented only once, make this an overfitted model with an R2 of 0.99 and RMSE of 4.88 kcal/mol.28 The computed contributions for groups at R2 and R3, shown in Table 2, do, however, provide some insight to complement the calculations described above. These contributions arise from the Free-Wilson modeling and represent the amount added to the final prediction for the groups shown at the relevant position. At both positions, it is clear that electron withdrawing groups (notably the cyano at R2) contribute positively: they increase ΔE and make the aromatic amine less likely to be active in the Ames test. Electron donating groups (notably

quantum mechanical energies to discriminate negative and positive aminopyrazoles. In Figure 4, the distribution of ΔEs from Figure 3b is redrawn with those compounds found to be positive (active) in the Ames test colored in red and those found to be negative (inactive) colored in green. It is clear that generally the distribution resembles that found in the original benchmarking exercise for these calculations.17 Compounds with low values of ΔE (150 kcal/mol) are likely to be negative, and those in the intermediate range (140−150 kcal/mol) are a mix. This confirms that this quantum mechanical approach is a useful way to guide the selection and design of aromatic amines that are less likely to be active in the Ames test and are therefore safer components of potential drug compounds. All of the values were computed before the compounds were made or tested; if a cut off of P(active) = 50% is taken as predicting active or inactive, it is possible to assess how good the predictions were. Of the 17 compounds found to be inactive, all 17 were predicted to be inactive. Of the 7 compounds found to be active, 4 were predicted to be active; 3 compounds predicted to be inactive were found to be active. Overall, 21 of the 24 (87.5%) predictions were correct. Finer detail on the SAR can be provided by calculating the value of ΔE for other compounds not tested (Figure 5). In particular, the values for compounds 43 and 44, when contrasted with those for 20 and 22 reveal that the CN at R2 tends to increase ΔE by 13−14 kcal/mol representing a significant decrease in the likelihood of activity in the Ames test. By contrast, the methyl at R3 as in compounds 38 and 39 tends to decrease ΔE by 3−4 kcal/mol, making compounds slightly more likely to be active. When these two are combined, as in compound 40, it is unsurprising that the resulting ΔE is 9 kcal/mol higher than that for 44 and high enough to give a compound, which is negative in the Ames test. There is one notable outlier in the distribution shown in Figure 4, corresponding to compound 34. This compound has a value of ΔE of 164.1 kcal/mol and a correspondingly low (23.6%) probability of being active in the Ames test. The test was repeated and confirmed this to be active although as noted in Table 1, the activity is weak at the highest dose. This example reiterates the challenge of definitively classifying compounds with predictive models. The outcome of the Ames bacterial mutagenicity test depends upon a complex interplay of factors, and the presentation of predictions as probabilities indicates that the outcome is not guaranteed but 706



Article

Table 1. Computational Predictions and Experimental Outcomes for a Set of Pyrazole Amines

In a previous test, compound 19 was found to be positive in the TA100 assay and gave a maximum increase over the control of 2.2-fold at 5000 μg/ plate. a

707



Article

Table 2. Group Contributions Computed from a FreeWilson Style Analysis of the Computed Values of ΔE for R2 and R3 Groups

Figure 4. Distribution of values of ΔE from selected compounds as shown in Figure 3b color coded by the subsequently determined activity in the Ames test.

Figure 5. Unsubstituted pyrazole compounds used to understand SAR.

NMe2 at R2) have the opposite effect. Groups make larger positive and negative contributions at R2 than at R3.

■

CONCLUSIONS

The quantum mechanical predictions previously described provide a good means to predict the likelihood of activity in the Ames test for aminopyrazoles. The SAR for this series in the Ames test indicates that variation at each of the positions explored can change the outcome of the test. Even compounds with relatively low probabilities of being active can be inactive and vice versa but selecting compounds using these probabilites allows the risk to be decreased. A range of nonmutagenic aminopyrazoles have been identified.

■

ASSOCIATED CONTENT

S Supporting Information *

Complete experimental details for Ames tests summarized in Table 1, geometries and energies from QM calculations, and synthesis details for other compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

Figure 6. Maximum fold increase in the number of revertant colonies observed for each compound in the Ames test run in the presence of S9 plotted against the corresponding ΔE value for each compound. Compounds found to be active are colored red and those that are inactive in green. Those found active overall despite being inactive in the indicated strain have a red outline. In each assay, several inactive compounds did not give any points distinguishable from the control and so do not contribute to the plot. 708



■

Article

(13) Borosky, G. L. (2007) Ultimate carcinogenic metabolites from aromatic and heterocyclic aromatic amines: a computational study in relation to their mutagenic potency. Chem. Res. Toxicol. 20, 171−180. (14) Borosky, G. L. (2008) Carcinogenic carbocyclic and heterocyclic aromatic amines: A DFT study concerning their mutagenic potency. J. Mol. Graphics Modell. 27, 459−465. (15) Himmelstein, M. W., Boogaard, P. J., Cadet, J., Farmer, P. B., Kim, J. H., Martin, E. A., Persaud, R., and Shuker, D. E. G. (2009) Creating context for the use of DNA adduct data in cancer risk assessment: II. Overview of methods of identification and quantitation of DNA damage. Crit. Rev. Toxicol. 39, 679−694. (16) Birch, A. M., Groombridge, S., Law, R., Leach, A. G., Mee, C. D., and Schramm, C. (2012) Rationally designing safer anilines: the challenging case of 4-aminobiphenyls. J. Med. Chem. 55, 3923−3933. (17) Leach, A. G., Cann, R., and Tomasi, S. (2009) Reaction energies computed with density functional theory correspond with a whole organism effect; modelling the Ames test for mutagenicity. Chem. Commun. (Cambridge, U.K.), 1094−1096. (18) McCarren, P., Bebernitz, G. R., Gedeck, P., Glowienke, S., Grondine, M. S., Kirman, L. C., Klickstein, J., Schuster, H. F., and Whitehead, L. (2011) Avoidance of the Ames test liability for arylamines via computation. Bioorg. Med. Chem. 19, 3173−3182. (19) Borosky, G. L. (2008) Quantum chemical studies on ultimate carcinogenic metabolites from polycyclic aromatic hydrocarbons. Curr. Med. Chem. 15, 2901−2920. (20) Bentzien, J., Hickey, E. R., Kemper, R. A., Brewer, M. L., Dyekjaer, J. D., East, S. P., and Whittaker, M. (2010) An in silico method for predicting ames activities of primary aromatic amines by calculating the stabilities of nitrenium ions. J. Chem. Inf. Model. 50, 274−297. (21) Waring, M. J., Clarke, D. S., Fenwick, M. D., Godfrey, L., Groombridge, S. D., Johnstone, C., McKerrecher, D., Pike, K. G., Rayner, J. W., Robb, G. R., and Wilson, I. (2012) Property based optimization of glucokinase activators - discovery of the phase IIb clinical candidate AZD1656. MedChemCommun 3, 1077−1081. (22) Bruneau, P. (2001) Search for predictive generic model of aqueous solubility using Bayesian neural nets. J. Chem. Inf. Comput. Sci. 41, 1605−1616. (23) Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E., Robb, M. A., Cheeseman, J. R., Scalmani, G., Barone, V., Mennucci, B., Petersson, G. A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H. P., Izmaylov, A. F., Bloino, J., Zheng, G., Sonnenberg, J. L., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, Jr., J. A., Peralta, J. E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin, K. N., Staroverov, V. N., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J. C., Iyengar, S. S., Tomasi, J., Cossi, M., Rega, N., Millam, J. M., Klene, M., Knox, J. E., Cross, J. B., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R. E., Yazyev, O., Austin, A. J., Cammi, R., Pomelli, C., Ochterski, J. W., Martin, R. L., Morokuma, K., Zakrzewski, V. G., Voth, G. A., Salvador, P., Dannenberg, J. J., Dapprich, S., Daniels, A. D., Farkas, Ö ., Foresman, J. B., Ortiz, J. V., Cioslowski, J., and Fox, D. J. (2009) Gaussian 09, revision A.1, Gaussian, Inc., Wallingford, CT. (24) CORINA, v3.4; Molecular Networks, GmbH Computer Chemie Langemarckplatz 1, Erlangen, Germany, 2006. (25) Becke, A. D. (1993) Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 98, 5648−5652. (26) Lee, C., Yang, W., and Parr, R. G. (1988) Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B: Condens. Matter 37, 785−789. (27) Hariharan, P. C., and Pople, J. A. (1973) Influence of polarization functions on MO hydrogenation energies. Theor. Chim. Acta 28, 213−222. (28) Free, S. M., and Wilson, J. W. (1964) A Mathematical Contribution to Structure-Activity Studies. J. Med. Chem. 7, 395−399.

AUTHOR INFORMATION

Corresponding Author

*Tel: +44 (0) 1625 519444 (W.M.). E-mail: [email protected]. uk (A.G.L.); [email protected] (W.M.). Present Address †

A.G.L.: School of Pharmacy and Biomolecular Sciences, Liverpool John Moores University, James Parsons Building, Byrom Street, Liverpool L3 3AF, U.K. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to Mike O’Donovan for the careful reading of and improvements to the manuscript, and thank Jane Moore, Shelley Collins, and Matthew Addie for synthetic contributions.

■

ABBREVIATIONS SAR, structure−activity relationship; DE, dissociation energy; DMSO, dimethylsulfoxide; UV, ultraviolet; NMR, nuclear magnetic resonance; HPLC, high performance liquid chromatography; rt, room temperature; MeOH, methanol; EtOH, ethanol

■

REFERENCES

(1) International Agency for Research on Cancer (2010) Some aromatic amines, organic dyes, and related exposures. IARC Monogr. Eval. Carcinog. Risks Hum. 99, 1−658. (2) ICH Expert Working Group (2011) Guidance on Genotoxicity Testing and Data Interpretation for Pharmaceuticals Intended for Human Use, International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use, Geneva, ICH S2(R1). (3) Ames, B. N., and Whitfield, H. J., Jr. (1966) Frameshift mutagenesis in Salmonella. Cold Spring Harb. Symp. Quant. Biol. 31, 221−225. (4) Ames, B. N., Lee, F. D., and Durston, W. E. (1973) Improved bacterial test system for the detection and classification of mutagens and carcinogens. Proc. Natl. Acad. Sci. U.S.A. 70, 782−786. (5) Mortelmans, K., and Zeiger, E. (2000) The Ames Salmonella/ microsome mutagenicity assay. Mutat. Res. 455, 29−60. (6) Maron, D. M., and Ames, B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res., Environ. Mutagen. Relat. Subj. 113, 173−215. (7) Easterbrook, J., Fackett, D., and Li, A. P. (2001) A comparison of aroclor 1254-induced and uninduced rat liver microsomes to human liver microsomes in phenytoin O-deethylation, coumarin 7-hydroxylation, tolbutamide 4-hydroxylation, S-mephenytoin 4′-hydroxylation, chloroxazone 6-hydroxylation and testosterone 6ß-hydroxylation. Chem.-Biol. Interact. 134, 243−249. (8) Beland, F. A., and Kadlubar, F. F. (1985) Formation and persistence of arylamine DNA adducts in vivo. Environ. Health Perspect. 62, 19−30. (9) Kadlubar, F. F., and Beland, F. A. (1985) Chemical properties of ultimate carcinogenic metabolites of arylamines and arylamides. ACS Symp. Ser. 283, 341−370. (10) Wild, D. (1990) A novel pathway to the ultimate mutagens of aromatic amino and nitro compounds. Environ. Health Perspect. 88, 27−31. (11) Turesky, R. J., Guengerich, F. P., Guillouzo, A., and Langouet, S. (2002) Metabolism of heterocyclic aromatic amines by human hepatocytes and cytochrome P4501A2. Mutat. Res., Fundam. Mol. Mech. Mutagen. 506−507, 187−195. (12) Novak, M., and Rajagopal, S. (2002) Correlations of nitrenium ion selectivities with quantitative mutagenicity and carcinogenicity of the corresponding amines. Chem. Res. Toxicol. 15, 1495−1503. 709


Experimental Testing of Quantum Mechanical Predictions of

Recommend Documents