ToxEvaluator: an integrated computational platform to aid the interpretation of toxicology study-related findings.

Attempts are frequently made to investigate adverse findings from preclinical toxicology studies in order to better understand underlying toxicity mechanisms. These efforts often begin with limited information, including a description of the adverse finding, knowledge of the structure of the chemical associated with its cause and the intended pharmacological target. ToxEvaluator was developed jointly by Pfizer and the Comparative Toxicogenomics Database (http://ctdbase.org) team at North Carolina State University as an in silico platform to facilitate interpretation of toxicity findings in light of prior knowledge. Through the integration of a diverse set of in silico tools that leverage a number of public and proprietary databases, ToxEvaluator streamlines the process of aggregating and interrogating diverse sources of information. The user enters compound and target identifiers, and selects adverse event descriptors from a safety lexicon and mapped MeSH disease terms. ToxEvaluator provides a summary report with multiple distinct areas organized according to what target or structural aspects have been linked to the adverse finding, including primary pharmacology, structurally similar proprietary compounds, structurally similar public domain compounds, predicted secondary (i.e. off-target) pharmacology and known secondary pharmacology. Similar proprietary compounds and their associated in vivo toxicity findings are reported, along with a link to relevant supporting documents. For similar public domain compounds and interacting targets, ToxEvaluator integrates relationships curated in Comparative Toxicogenomics Database, returning all direct and inferred linkages between them. As an example of its utility, we demonstrate how ToxEvaluator rapidly identified direct (primary pharmacology) and indirect (secondary pharmacology) linkages between cerivastatin and myopathy.

Polymer membrane-based ion-, gas- and bio-selective potentiometric sensors.

Recent progress in the design of new polymer membrane-based potentiometric ion-, gas- and bio-selective electrodes in chemistry laboratories at the University of Michigan (Ann Arbor) is reviewed. Emphasis is placed on describing the performance of devices for measuring anions (e.g., salicylate, thiocyanate, chloride and heparin) and gases (e.g., ammonia, carbon dioxide and oxygen) in biological samples, both in vitro and in vivo. Beyond direct measurement of key ions and gases in complex matrices, some of the new membrane electrode systems reported can serve as base transducers for the development of biosensors containing integrated biological reagents, including enzymes and antibodies. New approaches for mass fabricating solid-state ion and biosensor devices as well as future directions for research in the entire field of polymer membrane sensors are also described.

An integrated process for measuring the physicochemical properties of drug candidates in a preclinical discovery environment.

Automated log P, pK(a), solubility, and chemical stability systems comprise an integrated process that provides early stage physicochemical property data to the discovery research organization. Capillary electrophoresis (CE) techniques are used to experimentally determine pK(a) and log P. Solubility is determined using a quasi-equilibrium approach employing sample quantitation by flow injection analysis with ultraviolet (UV) detection at 256 nm. Chemical stability is assessed by challenging compounds with pH 2, pH 7, pH 12, and 3% hydrogen peroxide solutions overnight, and comparing the chromatographic profiles of the stability challenged solutions to that of a freshly prepared control. Validation of the log P method using a set of drug-like compounds demonstrates that the method yields log P values within +/-0.5 units of literature values. The log P method is valid over the range -0.5-5.0, and the technique is compatible with acidic, neutral, and basic compounds. The pK(a) technique yields results within +/-0.2 units of corresponding values obtained by potentiometric titration over a pK(a) range of 2 to 12. Solubility is reported in a 3-60 microg/mL range, and the results are generally within 20% of values measured by equilibrium solubility techniques. The current level of automation supports the measurement of the physicochemical properties of 100 compounds per week. Physicochemical property data for approximately 2000 compounds have been generated to date.

Quantitation of combinatorial libraries of small organic molecules by normal-phase HPLC with evaporative light-scattering detection.

The advantages of evaporative light-scattering detection over UV detection for the quantitation of combinatorial libraries composed of small organic compounds by HPLC are described. The detector's response is independent of the sample chromophore, which makes it well-suited to chromatographic analyses of mixtures of dissimilar solutes. Thus, HPLC with evaporative light-scattering detection offers to potential for reducing false positive or false negative results in screening assays, because of its ability to detect the presence of impurities that absorb poorly in the UV (e.g., those impurities originating from the polymeric support). Furthermore, the evaporative light-scattering detector exhibits a nearly equivalent response to compounds of similar structural class. Hence, rapid quantitation of compound libraries may be carried out with the use of a single external standard. For example, the quantitation errors, based on a single external standard, for a series of steroids, hydantoins, and BOC- and Fmoc-protected amino acids by normal-phase HPLC with evaporative light-scattering detection average approximately +/-10%. The application of the evaporative light-scattering detector to the quantitation of low-level sample impurities and the detector's compatibility with gradient elution are also described.

Shape-selective separation of polycyclic aromatic hydrocarbons by reversed-phase liquid chromatography on tetraphenylporphyrin-based stationary phases.

The reversed-phase chromatographic behavior of planar and non-planar polycyclic aromatic hydrocarbons (PAHs) is investigated on tetraphenylporphyrin and two metallotetraphenylporphyrin [Sn(IV), In(III)] bonded stationary phases using methanol-water and acetonitrile-water as mixed solvent mobile phases. Large differences in the capacity factors of aromatic solute pairs having the same number of carbon atoms, but differing in three-dimensional shape (e.g., triphenylene/o-terphenyl and perylene/alpha, alpha'-binaphthyl), suggest that the three tetraphenylporphyrin-based supports possess shape selectivity toward small planar aromatic solutes. Capacity factors for planar PAH solutes on these supports are significantly greater than for non-planar polyaryls having the same number of carbon atoms.

Preparation and characterization of covalently bound tetraphenylporphyrin-silica gel stationary phases for reversed-phase and anion-exchange chromatography.

A tetraphenylporphyrin-based stationary phase for high-performance liquid chromatography is prepared by attaching a p-carboxyphenyl derivative of the porphyrin to aminopropyl silica gel through an amide bond. This stationary-phase support is used for the reversed-phase separation of several polycyclic aromatic hydrocarbons and is shown to exhibit novel shape selectivity toward planar aromatic solutes. Subsequent metalation of the immobilized tetraphenylporphyrin with either tin(IV) or indium(III) results in a stationary phase with anion-exchange characteristics. Both the tin(IV) and indium(III) metalloporphyrin-based stationary phases exhibit useful selectivity for the separation of aromatic sulfonates and aromatic carboxylates.

Rapid method for estimating the octanol--water partition coefficient (log P ow) by microemulsion electrokinetic chromatography.

Several surfactant systems were evaluated based on their system constants determined by the solvation parameter model for the design of a surrogate chromatographic model for the rapid estimation of octanol-water partition coefficient (log Pow) by microemulsion electrokinetic chromatography. The system constant ratios responsible for the log Pow partition system are (nearly) the same as those for the microemulsion system containing sodium dodecyl sulfate (1.4% w/v), butan-1-ol (8% v/v) and heptane (1.2% v/v). Neutral and basic compounds are analyzed using a fused-silica capillary column with a 50 mM sodium phosphate-sodium borate (3:2) buffer at pH 10. Weakly acid compounds require the use of sulfonated silica capillary column and a 50 mM sodium phosphate buffer at pH 3. For 29 varied neutral and weakly basic compounds the average error between log Pow estimated using MEEKC and literature values was +/-0.12 over a log Pow range from 0.3 to 5.8.