Combinatorial Chemistry & High Throughput Screening - Volume 3, Issue 6, 2000

The importance of reliable detection systems for enantiomeric assays increases with the necessity of high throughput screening analysis of raw materials for the pharmaceutical industry. The utilization of electrochemical sensors in enantioselective analysis is an accurate and precise alternative to chromatographic techniques. The reliability of the response characteristics as well as of the analytical information obtained by using electrochemical sensors is strictly correlated with the design of the sensors. The designs evaluated for sensors have been based on PVC, imprinting polymers and carbon paste matrices. Among these, carbon paste sensors have been the most reliable and have been utilized for the construction of potentiometric, enantioselective membrane electrodes as well as for amperometric biosensors, and immunosensors. There are two ways to use the electrochemical sensors in enantioselective screening analysis selective binding and catalyst selectivity. A molecule with a special chemical architecture is required for selective binding a lock for a key. The high reliability of analytical information obtained using these sensors has made possible the automation of potentiometric and amperometric techniques by integration of enantioselective sensors as detectors in flow injection analysis and sequential injection analysis techniques.

Diagnosis of inherited diseases or cancer predispositions frequently involves determination of specific mutations or polymorphisms. The number of characterized monogenetic and polygenetic diseases is significantly rising every year. As a result, an increasing number of patient samples with a rising complexity of genetic diseases require molecular diagnostics. In order to apply genetic analyses to large groups of patients or population screening, automation of a sensitive and precise method is highly desirable. Capillary electrophoresis (CE) facilitates the development of methods which can rapidly process large number of patient samples in an automated fashion. In contrast, conventional techniques including Southern blotting, sequencing or standard gel electrophoresis are time consuming, cost ineffective and require substantial amounts of each specimen. Robustness, ease of operation, good reproducibility and low cost are the main advantages of CE.Currently, most protocols adapted to automated CE represent (i) analyses of DNA fragment length or DNA restriction patterns (RFLP), (ii) analyses of single-strand conformation polymorphism (SSCP) and (iii) microsatellite analyses. Recently, automated detection of variations in the FRAXA (CGG)n region (fragile X syndrome), LDL receptor gene, p53 gene, MTHFR (methylenetetrahydrofolate reductase) gene, HFE gene and others has been established on CE systems. These applications clearly demonstrate the suitability of CE for high throughput screening in medical applications.

This paper reviews high throughput screening procedures for the simultaneous detection of several drug classes relevant to clinical and forensic toxicology or doping control in urine or blood using gas chromatography-mass spectrometry (GC-MS), liquid chromatography coupled with a diode-array detector (LC-DAD) or mass spectrometry (LC-MS). Basic information describing these systematic toxicological analysis (STA) procedures such as the analytes, the biosample, work-up, separation column, mobile phase or separation buffer, detection mode and detection limits are summarized in tables arranged according to the analytical method. Examples of typical applications are presented in 2 figures. Analysis of alternative matrices, like sweat, saliva, nails or hair, was not reviewed.

The use of Kohonen feature maps for the visualization of various aspects of molecular similarity is briefly reviewed and illustrated. It is shown that a specific feature of self-organizing maps (SOM) makes them of special interest for the screening of compounds. In particular, these methods were used to design candidates for new sweeteners, which were then synthesized.

The appropriate selection of the mobile phase facilitated the use cellulose tris(3,5-dichlorophenylcarbamate (CDCPC) as a chiral stationary phase (CSP) during high-performance liquid chromatography (HPLC). A preliminary evaluations of this material indicated its very high chiral resolving ability toward many analytes of different chemical and pharmacological groups. Some chemicals and drugs containing two centers of chirality were also successfully resolved into all possible stereoisomers. The applicability of CDCPC for enantioseparations in capillary liquid chromatography was also shown giving promising prospects for the screening of novel biologically active compounds (which may be also synthesized based on combinatorial strategies) for their enantiomeric composition.

Lipophilic macrocyclic hexaamines supported by a poly(vinyl chloride) PVC matrix were used for the construction of liquid membrane electrodes sensitive toward adenine nucleotide polyanions. The membrane potential strongly depended on the pH of the sample solution. This phenomenon occurs due to the ability of the ionophore to accept protons. Therefore, the optimum pH was determined based on potential-pH profile. The potential measurements were carried out at pH 6.0 in the presence of 10-2 M 2-( N-morpholino) ethanesulfonic acid (MES) buffer. The potential response of these electrodes toward ATP-4 and/or HATP-3 was close to the Nernstian slope. The selectivities against ADP-3 , AMP-2 , HPO4-2 , and monovalent inorganic anions were estimated using the matched potential method. Chloride ions slightly affected potential response of the electrodes toward ATP-4 /HATP-3 . The influence of ionophore chemical structure on the selectivity and the sensitivity of these electrodes is briefly discussed.

Approximately 5-10 percent of all colorectal carcinomas arise from cancer predisposition syndromes caused by heterozygote germline mutations in post-replicative DNA mismatch repair (MMR) genes. In contrast to gastrointestinal polyposis syndromes, carcinomas in these patients do not occur on the background of increased numbers of polyps and hence are refered to as hereditary non-polyposis colorectal cancers (HNPCC). Six different MMR genes, MSH2, MSH3, MSH6, MLH1, MLH3 and PMS2, have been identified in the human genome. In the majority of HNPCC patients, heterozygote germline mutations are present in the MSH2 or MLH1 gene. Detection of mutations by conventional sequencing technology is expensive and labor intensive due to the complex intron and/or exon structures. In this study, we therefore have explored whether capillary electrophoresis-based single strand conformation polymorphism (SSCP-CE) provides a reliable means for mutation screening. We have tested different MLH1 mutations in exons 9 and 16 and find that SSCP-CE produces reliable electrophoretic patterns that allow recognition of wild-type alleles, microdeletions and point mutations. In summary, SSCP-CE provides a rapid, automated, and cost-effective technology for MSH2 and MLH1 mutation screening and will facilitate genetic diagnostics for HNPCC patients.

The performance of artificial neural network (ANN) models in predicting pharmacological classification of structurally diverse drugs based on their theoretical chemical parameters was demonstrated. The classification coefficients for psychotropic agents, b-adrenolytic drugs, histamine H 1 receptor antagonists and drugs binding to a-adrenoceptors were 100, 100, 95 and 86 percent, respectively. A set of easily accessible non-empirical molecular parameters describing the structure of xenobiotics can provide information allowing the prediction of some pharmacological properties of drugs and drug candidates employing ANN models. Since ANN analysis can help cluster as well as segregate drugs and drug candidates according to their known and expected pharmacological properties, the number of routine biological assays might be reduced. The results presented here might be used to improve the efficiency of high throughput screening programs for new drug hits by demonstrating a promising procedure for diverse combinatorial library design and evaluation.

A virtual high throughput screening test to identify potentially CNS-active drugs has been developed. Discrimination was based on the knowledge available in databases containing CNS-active (Cipsline from Prous Science) and inactive compounds (Chemical Directory from Sigma-Aldrich). Molecular structures were represented using 2D Unity fingerprints and a feedforward neural network was trained to classify molecules regarding their CNS activity. The parameterized network was validated by reclassification of the training set elements, by the classification of a test set preselected from the Prous database, and also by the prediction of activity for known CNS drugs not used in the training set but available in the Medchem database (Daylight).These tests revealed that our neural net recognized at least 89% of CNS-active compounds and would be suitable for use in our virtual screening protocol.