Developing 13C NMR quantitative spectrometric data-activity relationship (QSDAR) models of steroid binding to the corticosteroid binding globulin.

We have developed four quantitative spectrometric data-activity relationship (QSDAR) models for 30 steroids binding to corticosteroid binding globulin, based on comparative spectral analysis (CoSA) of simulated 13C nuclear magnetic resonance (NMR) data. A QSDAR model based on 3 spectral bins had an explained variance (r2) of 0.80 and a cross-validated variance (q2) of 0.78. Another QSDAR model using the 3 atoms from the comparative structurally assigned spectral analysis (CoSASA) of simulated 13C NMR on a steroid backbone template gave an explained variance (r2) of 0.80 and a cross-validated variance (q2) of 0.73. Positions 3 and 14 from the steroid backbone template have correlations with the relative binding activity to corticosteroid binding globulin that are greater than 0.52. The explained correlation and cross-validated correlation of these QSDAR models are as good as previously published quantitative structure-activity relationship (QSAR), self-organizing map (SOM) and electrotopological state (E-state) models. One reason that the cross-validated variance of QSDAR models were as good as the other models is that simulated 13C NMR spectral data are more accurate than the errors introduced by the assumptions and approximations used in calculated electrostatic potentials, E-states, HE-states, and the molecular alignment process of QSAR modeling. The QSDAR models developed provide a rapid, simple way to predict the binding activity of a steroid to corticosteroid binding globulin.

Models of polychlorinated dibenzodioxins, dibenzofurans, and biphenyls binding affinity to the aryl hydrocarbon receptor developed using (13)c NMR data.

Quantitative spectroscopic data-activity relationship (QSDAR) models for polychlorinated dibenzofurans (PCDFs), dibenzodioxins (PCDDs), and biphenyls (PCBs) binding to the aryl hydrocarbon receptor (AhR) have been developed based on simulated (13)C nuclear magnetic resonance (NMR) data. All the models were based on multiple linear regression of comparative spectral analysis (CoSA) between compounds. A 1.0 ppm resolution CoSA model for 26 PCDF compounds based on chemical shifts in five bins had an explained variance (r(2)) of 0.93 and a leave-one-out (LOO) cross-validated variance (q(2)) of 0.90. A 2.0 ppm resolution CoSA model for 14 PCDD compounds based on chemical shifts in five bins had an r(2) of 0.91 and a q(2) of 0.81. The 1.0 ppm resolution CoSA model for 12 PCB compounds based on chemical shifts in five bins had an r(2) of 0.87 and a q(2) of 0.45. The models with more compounds had a better q(2) because there are more multiple chemical shift populated bins available on which to base the linear regression. A 1.0 ppm resolution CoSA model for all 52 compounds that was based on chemical shifts in 12 bins had an r(2) of 0.85 and q(2) of 0.71. A canonical variance analysis of the 1.0 ppm CoSA model for all 52 compounds when they were separated into 27 strong binding and 25 weak binding compounds was 98% correct. Conventional quantitative structure-activity relationship (QSAR) modeling suffer from errors introduced by the assumptions and approximations involved in calculated electrostatic potentials and the molecular alignment process. QSDAR modeling is not limited by such errors since electrostatic potential calculations and molecular alignment are not done. The QSDAR models provide a rapid, simple and valid way to model the PCDF, PCDD, and PCB binding activity in relation to the aryl hydrocarbon receptor (AhR).

(13)C NMR quantitative spectrometric data-activity relationship (QSDAR) models of steroids binding the aromatase enzyme.

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.

Use of 13C NMR spectrometric data to produce a predictive model of estrogen receptor binding activity.

We have developed a spectroscopic data-activity relationship (SDAR) model based on 13C NMR spectral data for 30 estrogenic chemicals whose relative binding affinities (RBA) are available for the alpha (ERalpha) and beta (ERbeta) estrogen receptors. The SDAR models segregated the 30 compounds into strong and medium binding affinities. The SDAR model gave a leave-one-out (LOO) cross-validation of 90%. Two compounds that were classified incorrectly in the SDAR model were in the transition zone between classifications. Real and predicted 13C NMR chemical shifts were used with test compounds to evaluate the predictive behavior of the SDAR model. The 13C NMR SDAR model using predicted 13C NMR data for the test compounds provides a rapid, reliable, and simple way to screen whether a compound binds to the estrogen receptors.

Producing 13C NMR, infrared absorption, and electron ionization mass spectrometric data models of the monodechlorination of chlorobenzenes, chlorophenols, and chloroanilines.

We have developed four spectroscopic data-activity relationship (SDAR) models of monodechlorination of 32 chlorinated benzene compounds in anaerobic estuarine sediment. The SDAR models were based on combinations of 13C nuclear magnetic resonance (NMR), infrared absorption (IR), and electron ionization mass spectrometric (EI MS) data. The SDAR models segregated the 32 compounds into 17 readily monodechlorinated compounds and 15 not readily monodechlorinated compounds. The SDAR model based on 13C NMR, IR, and EI MS data gave a leave-one-out cross-validation of 93.8%. The SDAR model based on a composite of 13C NMR and IR data gave a leave-one-out cross-validation of 90.6%. The SDAR model based on a composite of IR and EI MS data gave a leave-one-out cross-validation of 84.4%. The SDAR model based on a composite of 13C NMR and EI MS data gave a leave-one-out cross-validation of 84.4%. These reliable SDAR models provide a rapid and simple way to predict whether a chlorinated benzene compound will readily go through monodechlorination. The FDA has filed a patent application on methods of using any combination of spectral data (NMR, MS, UV-vis, IR, and fluorescence, phosphorescence) to model a chemical, physical, or biological endpoint.

(13)C NMR and electron ionization mass spectrometric data-activity relationship model of estrogen receptor binding.

Two Spectroscopic Data-Activity Relationship (SDAR) models based on (13)C nuclear magnetic resonance (NMR) and electron ionization mass spectra (EI MS) data were developed for 108 compounds whose relative binding affinities (RBA) to the estrogen receptor are known. The (13)C NMR and EI MS data were used as spectrometric digital fingerprints to reflect the electronic and structural characteristics of the compounds. Both SDAR models segregated the 108 compounds into 20 strong, 15 medium, and 73 weak relative binding classifications. The first SDAR model, based on (13)C NMR data alone, gave a leave-one-out (LOO) cross-validation of 75.0%. The second SDAR model, based on a composite of (13)C NMR and EI MS data, gave a LOO cross-validation of 82.4%. Many of the misidentifications from the cross-validations were between medium and weak classifications, where there were fewer specific spectrometric characteristics to identify the relationship of spectra to estrogen receptor binding. Real and predicted (13)C NMR chemical shifts were used to test the predictive behavior of both SDAR models. The ease of use and speed of SDAR modeling may facilitate their use with other toxicological endpoints.

Sample preparation for the analysis of flavors and off-flavors in foods.

Off-flavors in foods may originate from environmental pollutants, the growth of microorganisms, oxidation of lipids, or endogenous enzymatic decomposition in the foods. The chromatographic analysis of flavors and off-flavors in foods usually requires that the samples first be processed to remove as many interfering compounds as possible. For analysis of foods by gas chromatography (GC), sample preparation may include mincing, homogenation, centrifugation, distillation, simple solvent extraction, supercritical fluid extraction, pressurized-fluid extraction, microwave-assisted extraction, Soxhlet extraction, or methylation. For high-performance liquid chromatography of amines in fish, cheese, sausage and olive oil or aldehydes in fruit juice, sample preparation may include solvent extraction and derivatization. Headspace GC analysis of orange juice, fish, dehydrated potatoes, and milk requires almost no sample preparation. Purge-and-trap GC analysis of dairy products, seafoods, and garlic may require heating, microwave-mediated distillation, purging the sample with inert gases and trapping the analytes with Tenax or C18, thermal desorption, cryofocusing, or elution with ethyl acetate. Solid-phase microextraction GC analysis of spices, milk and fish can involve microwave-mediated distillation, and usually requires adsorption on poly(dimethyl)siloxane or electrodeposition on fibers followed by thermal desorption. For short-path thermal desorption GC analysis of spices, herbs, coffee, peanuts, candy, mushrooms, beverages, olive oil, honey, and milk, samples are placed in a glass-lined stainless steel thermal desorption tube, which is purged with helium and then heated gradually to desorb the volatiles for analysis. Few of the methods that are available for analysis of food flavors and off-flavors can be described simultaneously as cheap, easy and good.

Sample preparation and high-resolution separation of mycotoxins possessing carboxyl groups.

The chromatographic analysis of carboxyl-containing mycotoxins, such as fumonisin B1, ochratoxin A, and citrinin, presents a continual challenge. Toxins must first be extracted from foods or tissues and then cleaned up before chromatographic separation and detection. Liquid-liquid extraction efficiencies for some carboxylic mycotoxins are marginal for spiked samples and uncertain for incurred residues. Immunoaffinity columns may be useful for concentrating mycotoxins from samples before chromatography. In almost every case, more than one analytical method must be used to confirm the identification of the mycotoxin. The fumonisins are especially troublesome to analyze because they are relatively insoluble in organic solvents, they are not separated easily by gas chromatography, and they do not respond to the usual absorbance or fluorescence detectors used in liquid chromatography. Fluorescence derivatization and electrospray liquid chromatography-mass spectrometry have now made it possible to detect trace levels of mycotoxins. The purity of mycotoxin standards for toxicological studies can be determined by liquid chromatography with either an evaporative light scattering detector or electrospray mass spectrometer. New developments in capillary electrophoresis, nonporous microsphere liquid chromatography, and detection methods for low-volatility compounds show promise for improving the analysis of mycotoxins in the future.

Rapid identification of intact whole bacteria based on spectral patterns using matrix-assisted laser desorption/ionization with time-of-flight mass spectrometry.

Matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) was investigated as a method for the rapid identification of whole bacteria, either by comparison with archived reference spectra or by co-analysis with cultures of known bacteria. Bacteria were sampled from colonies on an agar plate, mixed with the matrix, air-dried, and introduced in batches into the mass spectrometer for analysis. In the first experiment, both bacterial strains that had been previously analyzed to obtain reference spectra and other strains that had not been analyzed were blind-numbered and their spectra were obtained. Those strains that matched reference spectra were found to be correctly identified. A second experiment involved co-analysis of reference strains and bind-numbered strains under identical conditions; species-specific identification was demonstrated by comparison of spectra of the blind-numbered strains with those of the standards. In all of the spectra obtained in these experiments, each bacterial strain showed a few characteristic high-mass ions which are thought to be derived from bacterial proteins. This work represents the first reported instance of successful bacterial chemotaxonomy by MALDI-TOFMS analysis of whole cells. For the strains tested, the method is rapid and simple.

Determination of underivatized fumonisin B1 and related compounds by HPLC.

A method is presented for determining the purity of the mycotoxin fumonisin B1 (FB1) by high performance liquid chromatography (HPLC) with evaporative light scattering detection (ELSD). The ELSD is a universal HPLC detector that exhibits a non-linear relationship between analyte amount and the resulting response. A log-log plot of ELSD response with the mass of FB1 injected was used as a calibration curve for determining the quantities of both FB1 and also individual impurities present in samples. Assumptions related to the uniformity of ELSD response for different but related compounds and other issues implied in this use of ELSD data were examined. One potential error produced by use of this method for purity analysis comes from the ELSD's decreased sensitivity for low-concentration analytes. Because analytes become more dilute the longer they remain on a chromatographic column, this sensitivity discrimination can be related to the retention times at which they appear. The ELSD response for FB1 at retention time 15.5 minutes was used to construct a general purpose calibration curve. Whenever a peak appeared at any time other than 15.5 minutes, the discrimination effect was corrected using a an empirically determined weighting factor and a proportion calculated from the retention time difference compared to 15.5 minutes. Purities for two fumonisin samples were calculated using both the ELSD method described above and an electrospray/mass spectrometric method. The quantitative assumptions underlying each method were discussed in order to understand and reconcile differences between the two sets of purity values obtained.

Determination of fumonisins B1, B2, B3 and B4 by high-performance liquid chromatography with evaporative light-scattering detection.

Fumonisins B1, B2, B3 and B4 (FB1-FB4), a group of mycotoxins produced by the fungus Fusarium moniliforme, were separated by HPLC using an analytical-scale, base-deactivated C8 column and a gradient of trifluoroacetic acid buffer (pH 2.7) and acetonitrile. An evaporative light-scattering detector was used to detect the fumonisin peaks. A semi-preparative-scale, base-deactivated C8 column with a 1:14 mobile phase split facilitated the purification of analytical standards of FB.

Particle size distribution is not the major factor explaining variable analyte transmission efficiency in liquid chromatography/particle beam/mass spectrometry.

We characterized the particle size distribution and the analyte transmission efficiency of a liquid chromatography/particle beam/mass spectrometry (LC/PB/MS) system as a function of experimental variations normally used to optimize the LC/PB/MS system. The particle size distribution was evaluated using an electrical differential mobility particle sizer (DMPS) and both the DMPS and the mass spectrometer were used to evaluate transmission. The latter results were correlated to provide evidence related to mechanisms which contribute to poor sample transmission. Addition of ammonium acetate buffer did not increase the aerosol particle mean diameter. However, it did lead to significant increases in caffeine transmission efficiency observed in both the DMPS and the mass spectrometer. Our results were interpreted to suggest a possible electrostatic cause for quantitative anomalies in LC/PB/MS rather than simple mass discrimination in the particle beam momentum separator.

AC corona-discharge aerosol-neutralization device adapted to liquid chromatography/particle beam/mass spectrometry.

An AC corona-discharge device was inserted upstream of a thermospray vaporizer tip in a liquid chromatography/particle beam mass spectrometer to neutralize static aerosol charging. Response of a test analyte was measured with or without discharge initiation. If the solvent contained no ammonium acetate buffer, increased analyte signal was associated with the discharge. However, in the presence of ammonium acetate the benefit of AC discharge neutralization was either not observed or was more subtle. This led to the conclusion that the previously observed ammonium acetate "carrier" effect is attributable, at least in part, to neutralization of static electric charges produced spontaneously during the solvent nebulization process. In a second experiment, the pattern of particles issuing from the system momentum separator was examined by aiming the particle beam at a cold target located within a mass spectrometer ion source. Variations in particle density were observed depending on (i) whether or not the aerosol had been neutralized and (ii) the proximity of electron-beam-collimating magnets to the particle beam trajectory. These results are consistent with a hypothesis that electrostatic charging occurs spontaneously during the nebulization process in which an aerosol is formed from the high performance liquid chromatography effluent. Such electrostatic charging introduces a factor likely to degrade system performance by at least two modes: through interactions of the charged aerosol particles (i) with the walls of the aerosol transmission pathway, and, after they are accelerated into a particle beam and introduced into the mass spectrometer, (ii) with the magnets used for electron beam collimation in many mass spectrometer ion sources.