13 C NMR Quantitative Spectrometric Data-Activity Relationship (QSDAR) Models of Steroids Binding the Aromatase Enzyme Richard D. Beger,* Dan A. Buzatu, Jon G. Wilkes, and Jackson O. Lay, Jr. Division of Chemistry, National Center for Toxicological Research, Food and Drug Administration, Jefferson, Arkansas 72079 Received January 26, 2001 Five quantitative spectroscopic data-activity relationships (QSDAR) models for 50 steroidal inhibitors binding to aromatase enzyme have been developed based on simulated 13 C nuclear magnetic resonance (NMR) data. Three of the models were based on comparative spectral analysis (CoSA), and the two other models were based on comparative structurally assigned spectral analysis (CoSASA). A CoSA QSDAR model based on five principal components had an explained variance (r 2 ) of 0.78 and a leave-one-out (LOO) cross- validated variance (q 2 ) of 0.71. A CoSASA model that used the assigned 13 C NMR chemical shifts from a steroidal backbone at five selected positions gave an r 2 of 0.75 and a q 2 of 0.66. The 13 C NMR chemical shifts from atoms in the steroid template position 9, 6, 3, and 7 each had correlations greater than 0.6 with the relative binding activity to the aromatase enzyme. All five QSDAR models had explained and cross- validated variances that were better than the explained and cross-validated variances from a five structural parameter quantitative structure-activity relationship (QSAR) model of the same compounds. QSAR modeling suffers from errors introduced by the assumptions and approximations used in partial charges, dielectric constants, and the molecular alignment process of one structural conformation. One postulated reason that the variances of QSDAR models are better than the QSAR models is that 13 C NMR spectral data, based on quantum mechanical principles, are more reflective of binding than the QSAR model’s calculated electrostatic potentials and molecular alignment process. The QSDAR models provide a rapid, simple way to model the steroid inhibitor activity in relation to the aromatase enzyme. INTRODUCTION The aromatase enzyme catalyzes the conversion of test- osterone to estradiol by the aromatization of the A-ring in steroids. 1,2 Estrogen production from aromatase enzyme activity is important in the evolution and development of estrogen-dependent tumors. 3,4 Inhibition of the aromatase enzyme, a cytochrome P450 complex that converts androgens to estrogens, is therapeutically significant because it may control breast cancer. 4 Standard three-dimensional quantitative structure-activity relationship (3D-QSAR) models have been produced for 50 steroid inhibitors of the aromatase enzyme. 5 13 C NMR data has been used to produce a reliable classification for spectrometric data-activity relationship (SDAR) models of the estrogen receptor system 6 and monodechlorination rates. 7 We have developed a quantitative relationship between spectra and certain properties or activities for binding to the corticosterone binding globulin 8 and aryl hydrocarbon recep- tor. 9 Quantitative spectrometric data-activity relationships (QSDAR) is based on the spectral-activity leg in the triangular structure-spectrum-activity relationship. The bind- ing activity of 45 progestagen steroids to a steroid receptor have been quantitatively modeled with simulated 13 C NMR spectra by comparative spectral analysis (CoSA). 10 These CoSA models, using simulated 13 C NMR data, yielded better correlations and predictions than were seen with comparative molecular field analysis (CoMFA) methods. This paper demonstrates that CoSA of simulated 13 C NMR spectral data can be used to produce a reliable quantitative spectrometric data-activity relationship (QSDAR) model of steroids binding to the aromatase enzyme. QSAR is based on the assumption that there is a relation- ship between the structure and activity of a compound. QSAR modeling results have been able to show that receptor binding of a compound can be predicted from a combination of electrostatic potentials and geometrical structural analysis. 11-13 However, using a specific molecular structure for computer modeling of each compound dramatically extends the number of calculations required to define the model. Moreover, the selection of the most appropriate 3D structure for each molecule requires a number of assumptions. The necessary simplifying assumptions in some cases give results that are hard to replicate or are inaccurate. An advantage of QSDAR is that it is not necessary to solve any quantum mechanical calculations or use the structures of molecules for electrostatic calculations, as is done in QSAR techniques. 5,14-17 Using QSAR modeling results, receptor binding of a compound can be predicted based, in part, upon electrostatics and geometrical structure. The 13 C NMR spectrum of a compound contains frequencies that correspond directly to the quantum mechanical properties of the 13 C nuclear magnetic moment. The quantum mechanical description of magnetic moment, in turn, depends largely on its electrostatic features and geometry. 18 Therefore, we postulated that we could use 13 C NMR data in much the same way that QSAR uses comparative molecular field analysis (CoMFA) of * Corresponding author phone: (870)543-7080; fax: (870)543-7686; e-mail: rbeger@nctr.fda.gov. 1360 J. Chem. Inf. Comput. Sci. 2001, 41, 1360-1366 10.1021/ci010285e CCC: $20.00 © 2001 American Chemical Society Published on Web 08/03/2001