Abstract
Laboratory Automation and High-Throughput Chemistry
COMPARISON TEST OF THREE AUTOMATED WORKSTATIONS FOR PARALLEL SYNTHESIS
One of the main objectives in current industrial chemical process research and development is increasing laboratory efficiency by high-throughput experimentation (HTE) for screening of reaction and process parameters and by high-speed experimentation (HSE) for optimization of process parameters. An important tool in reaching this second goal is the use of parallel-automated reactor systems.
A comparison test of three such automated workstations for parallel synthesis was described recently by M. E. Van Loo and P. E. Lengowski (Organic Process Research & Development, 2002, 6, 833), namely the Automate (HEL), Multimax (Mettler Toledo), and Flexylab (Systag). As requirements for the tests, the authors stated that the operation of experiments should be similar to that in pilot plant processes. The equipment should be able to perform the optimization of critical reaction parameters such as temperature, dosing rates, pH, and mixing. Much more information with respect to traditional methods should be gathered in order to create robust processes. Several chemical reactions were used as test systems (acetic anhydride hydrolysis, enolether preparation, oxidation with potassium permanganate, and crystallization of a steroid). The conclusion of this article is that each of these systems had its strong features and depending on the chemistry, one of the systems was preferable. A general lesson was that working with a system gives far more information than just reading about it; thus we recommend performing a trial run of your own chemistry prior to making any buying decision.
STREAMLINING PROCESS R&D USING MULTIDIMENSIONAL ANALYTICAL TECHNOLOGY
The combination of multidimensional analytical technology to form a single nth-order analyzer can produce reaction process data that are inaccessible using conventional techniques. J. R. McConnell et al. describe a third-order analyzer, which was assembled utilizing an automated batch reactor (Mettler RC1 calorimeter), Applied Systems React IR, and Hewlett-Packard mass selective detector (Organic Process Research & Development, 2002, 6, 700). This third-order analyzer was used to simultaneously perform reaction calorimetry, in situ infrared spectroscopy, and in situ mass spectrometry on the conversion of a vinyl sulfone to a vinyl tributylstannane. The goal of complete material and energy balance, coupled with knowledge of important scale-up issues, was achieved in a single experiment and in real time.
NMR-BASED HIGH-THROUGHPUT SCREENING ASSAY FOR ENANTIOSELECTIVE CATALYSTS
The enantioselective catalytic synthesis of chiral organic compounds is of substantial academic and industrial interest as evidenced e.g. by the award of the Nobel Prize for Chemistry for these types of reactions in 2001 to K. B. Sharpless, R. Noyori, and W. S. Knowles. Usually a large number of ee values (enantiomeric excess) for different reaction mixtures must be determined when studying methods for combinatorial asymmetric catalysis for either transition metal complexes or biocatalysts. Efficient methods for ee-determination are therefore crucial for rapid development of new asymmetric catalysis methods. M. Reetz et al. have developed a new practical NMR-based high-throughput assay for screening of enantioselective catalysts and biocatalysts which has the potential for at least 1400 ee determinations per day with an exceptionally high degree of accuracy (±2% to ± 5 %) (Adv. Synth. Catal. 2002, 344, 1008). For quantitative determinations, 1H-NMR spectroscopy of 13C-labeled compounds was used and the quantitative analysis of the spectra was accomplished automatically by AMIX™ software. To achieve high throughput, a Gilson 215 liquid handler was combined with a 300 MHz 1H-NMR instrument equipped with a flow-through cell system. The Gilson robot was used to transfer samples from the 96-microtiter plate to the flow-through cell and measuring time per sample was down to one minute. As NMR instrumentation is widely available in chemical research laboratories, it is very likely that this assay will be of considerable interest to researchers involved in the combinatorial search for asymmetric transition metal catalysts and in the directed evolution of enantioselective enzymes.
SECOND-GENERATION MS-BASED HIGH-THROUGHPUT SCREENING SYSTEM FOR ENANTIOSELECTIVE CATALYSTS AND BIOCATALYSTS
In another paper, Reetz et al. (Can. J. Chem. 2002, 80, 626) described a high- throughput method for determination of enantioselectivity of approximately 10,000 catalysts or biocatalysts per day. The method is based on electrospray mass spectrometric techniques using an eight-channel multiplexed (MUX) sprayer system connected to a time-of-flight mass spectrometer. The inlet of the ion source is controlled by a stepping rotor that is continuously moving from one sprayer to the next with a recording time of 100 ms for each channel and a delay time of 50 ms, thus allowing a spectrum to be obtained from each channel every 1.2 s. One cycle, where eight samples are being sprayed in parallel, requires around 70 s, which allows a 96-well microtiter plate to be screened in 14 minutes. This method for ee-determination is based on isotopical labeling of one enantiomer. Errors of approximately 2% for the ee-values indicate that in addition to the significant improvement in sample throughput, this is also a relatively precise method for high-throughput catalyst screening.
USE OF STATISTICAL DESIGN OF EXPERIMENTS IN THE OPTIMIZATION OF AMIDE FORMATION FOR LIBRARY PRODUCTION
Statistical Design of Experiments (DoE), in conjunction with automated synthesizers, is an important and well-established approach for optimizing chemical reactions during process development in the pharmaceutical and fine chemical industries. Another field where it is highly important to rapidly identify the best reaction parameters is in discovery research prior to analogue generation or library production. Here robust and general chemistry is required to reliably produce pure compounds with diverse structural features for activity screening. O. W. Gooding et al. demonstrated that DoE is a very useful tool to optimize the synthesis of amides employing resin-bound N-hydroxybenzotriazole active esters as intermediates (J. Comb. Chem. 2002, 4, 576). This methodology allowed the rapid development of an improved protocol for the parallel synthesis of amides by using an automated Trident synthesizer. By synthesizing a small but diverse 48-member library of amides with good to high yields and purities, it was demonstrated that this is a general and robust process.
SKELETAL DIVERSITY VIA A BRANCHED PATHWAY: MASSIVE PARALLEL SYNTHESIS OF 29400 DISCRETE POLYCYCLIC COMPOUNDS AND THEIR ARRAYING IN STOCK SOLUTIONS
What is possible in massive parallel synthesis was demonstrated recently by S. L. Schreiber et al. (J. Am. Chem. Soc. 2002, 124, 13402). They synthesized in a diversity-oriented approach by using a branching pathway 29400 discrete compounds comprising 10 distinct polycyclic skeletons. The six-step stereoselective synthesis which affords products having a central skeleton with between two and four rings and up to six stereocenters has been achieved using an inexpensive and accessible “one bead - one stock solution” technology platform. For automated preparation of stock solutions, single beads were arrayed into 384-well microtiter plates using a vacuum based bead arrayer and then a Hydra384 syringe-array robot was used to elute the reaction products from the beads and for further reformatting. Due to their great diversity, the compounds prepared might be effective probes for chemical genetic studies aimed at dissecting biology.
HIGH-THROUGHPUT LIQUID CHROMATOGRAPHY / MASS SPECTROMETRY PARALLEL PURIFICATION OF LARGE COMBINATORIAL LIBRARIES
Combinatorial chemistry is an important source of new chemical compounds to be tested in high-throughput in vitro screening against therapeutic targets. To minimize false positive and false negative hits, it is advantageous to assay only high-quality compounds. Therefore, great effort has been devoted to the development of automated purification technology designed to keep pace with the output of high-throughput combinatorial synthesis. D. B. Kassel et al. (J. Comb. Chem. 2002, 4, 600) report on the development of a validated, streamlined high-throughput process for the automated purification of parallel synthesis derived libraries. The steps involved in this library purification process include dissolution of dry films of crude synthetic material, dual-column LC/MS purification, dual-column post-purification analysis, quantitation, reformatting, and submission of pure compounds for registration. The robustness of this dual-column LC/MS system and of the overall purification process was assessed by purifying over 7,800 reaction mixtures over a period of seven months.
AUTOMATED ANALYSIS OF PROTON NMR SPECTRA FROM COMBINATORIAL RAPID PARALLEL SYNTHESIS USING SELF-ORGANIZING MAPS
The evolution of combinatorial chemistry as a drug discovery enabling technology over the past few years has resulted in a tremendous increase in the need for high-throughput analytical methods to keep pace with the number of samples required to be analyzed. It is now relatively simple to acquire proton NMR spectra of compounds in 96-well plates prepared in a rapid parallel synthesis fashion using a flow-NMR automation setup. However the manual analysis of 96 NMR spectra obtained in this manner is often laborious and painstakingly slow. S. Kalelkar et al. (J. Comb. Chem. 2002, 4, 622) have developed a new automated method for rapidly analyzing 96 NMR spectra of compounds synthesized in an 8 × 12 matrix using self-organizing maps. This unsupervised neural network is capable of clustering together NMR spectra containing a common pattern of R-groups and identifying outliers from within such clusters. Analysis of these outlier spectra can quickly help indicate the presence of undesired products, impurities, starting materials, and other unexpected errors in a 96-well plate synthesis by focusing the chemists' attention on the aberrant NMR spectra. Thus these self-organizing maps can be valuable tools in performing efficient quality control on combinatorial libraries.
HIGH-THROUGHPUT BIOASSAY-GUIDED FRACTIONATION TO INDIVIDUAL COMPONENTS OF COMBINATORIAL LIBRARIES
D. W. Phillipson et al. describe an automated, high-throughput analytical tool for the unambiguous characterization of the active components of a combinatorially derived reaction mixture (J. Comb. Chem. 2002, 4, 591). The authors name this technique high-throughput bioassay-guided fractionation (BGF); they describe the systematization of the BGF concept, the application of BGF to combinatorial chemistry, and the high-throughput nature of the identification technique. The identification of the active component in a well mixture is an essential step for subsequent resynthesis or isolation of the active component. UV and MS data generated from automated LC/MS analyses are combined with the biological assay data and therefore a rapid identification of the active component is made possible. Using this tool, more than 400 samples can be tested each month. The described BGF tool allows full utilization of highly diverse combinatorial libraries, thereby obviating costly up-front purification or extensive prescreening characterization efforts.
COMBINATORIAL PARALLEL SYNTHESIS AND AUTOMATED SCREENING OF A NOVEL CLASS OF LIQUID CRYSTALLINE MATERIALS
Combinatorial material research is gaining in popularity. The correlation of molecular structures with bulk properties in the solid state represents a general problem in materials research because these properties typically relate on numerous interdependent parameters. To date bulk properties of liquid crystals cannot be predicted by any current theory, but have to be evaluated empirically. P. Baeuerle et al. (Chem. Comm. 2002, 2762) report about a combinatorial parallel synthesis study which has led to the rapid generation of a single-compound library of novel fluorinated quater-phenyls by iterative Suzuki coupling steps. Subsequent automated screening of the phase transition temperatures and enthalpies of the individual quaterphenyls by using automated sequential differential scanning calorimetry (DSC) revealed liquid crystalline behavior and gave qualitative relationships of molecular structures and solid state properties. The use of this full combinatorial development process has led to the accelerated generation and screening of a novel class of liquid crystalline materials as potential candidates for LC displays.
Laboratory Automation in Analytics
There are at least three methods to automate analytical measurements including direct connection of analytical devices to processes, automated analysis of measured data, and automated method development. Each of these methods must be employed to enable higher throughput in complex systems.
INTEGRATING AUTOMATION AND LC/MS FOR DRUG DISCOVERY BIOANALYSIS
A novel integrated approach for automated sample handling in drug discovery bioanalysis has been described by D. R. Rossi (J. Autom. Methods & Management in Chem. 2002, 1). He developed a semi-automated liquid-liquid extraction technique in 96-well format for biological fluid sample preparation that has been used in conjunction with the integrated sample-handling approach. The preparation time per plate could thus be reduced to 1.5 h compared to the manual approach with about 4 h. The resulting plates are directly compatible with the LC/MS system. Numerous comparisons between the new system and conventional sample preparation approaches gave equivalent drug-quantitation results for different samples. The new sampling process doubled the efficiency of drug discovery bioanalysis and can also be used in combination with LC/TOF/MS to identify and quantify drug compounds in rats.
High-Throughput Screening
DNA ARRAYS WITHOUT AMPLIFICATION
High-throughput screening (HTS) technologies are widely used in many areas of drug discovery. Thus the development of new assays is one of the major goals in development today. S.J. Park et al. recently developed a new DNA array that does not require amplification or temperature dependent hybridization to identify targets (Science 295 2002, 1503). The new assay uses the strategy of labeling one end of the target DNA with gold nanoparticles. When the target DNA binds to probe DNA immobilized between two electrodes, the gold particles fill the gap between the electrodes. The amount of target DNA can be determined by conductivity measurements. The detection limits as low as 500 fmol of target DNA.
HIGH-THROUGHPUT METHOD FOR THE PRODUCTION AND ANALYSIS OF LARGE NATURAL PRODUCTS LIBRARIES FOR DRUG DISCOVERY
Natural products are a major resource for chemical diversity in the pharmaceutical industry. Since 40% of the structurally diverse compounds are found in the Dictionary of Natural Products, the development of libraries is necessary. L. Zeng et al. described the production, analysis, and characterization of libraries for natural products in order to accelerate the drug discovery process for high-throughput screening in the pharmaceutical industry (Anal. Chem. 74 2002, 3963). Automated flash chromatography, solid phase extraction, filtration, and HTS parallel preparative HPLC were combined to obtain the libraries in 96- or 384-well plates. The libraries consist of purified fractions with one to five compounds per well. They are analyzed prior to biological screening by LC/MS system for the determination of molecular weight, number, and quantity of compounds in the fraction. In vitro toxicity tests are done for biological screening. Active compounds are purified after biological screening and the structures of the active compounds are elucidated by NMR and MS.
Clinical Automation
Although developments for chemical and pharmaceutical processes today play a major role in laboratory automation, there are still some interesting developments in medicinal and clinical automation. Novel automated clinical analyzers are playing an increasing role in delivering diagnostic data.
DEVELOPMENT OF A MICROPROCESSOR-BASED BIOCHEMICAL SAMPLER
S. S. Randhawa et al. developed an inexpensive and portable microprocessor-based sampler containing a sample plate assembly, probe-drive linkage system, wash fluid receptacle, and a timing system (J. Autom. Methods & Management in Chem. 2002; the sampler is a set of mechanical and electronic controls coupled and synchronized to deliver the sample. Since its operation is automatic, the samples are automatically advanced by each step to the aspirating position, where a probe delivers a fixed quantity of sample each time. The sample probe is connected to peristaltic pump tubing, which delivers the sample fluid by the vacuum produced by the pump; 49). The developed sampler can be used in continuous flow analysis for the determination of biochemical parameters and can handle a capacity of 24 samples.
ANALYTICAL EVALUATION OF TWO AUTOMATIC METHODS TO MEASURE BLOOD CK-MB MASS AND TROPONIN I
For many years, the enzymatic profile including CK-MB isoenzyme activity and aspartate aminotransferase has been the biochemical method of choice in the diagnosis of acute myocardial infarction. In the 1990s, more sensitive and specific cardiac markers were developed. The lack of method standardization produces great problems for the comparison of results. Galan et al. (J. Autom. Methods & Management in Chem. 2002, 51) developed and evaluated two new immunoassays for the quantification of CK-MB mass and troponin as markers for myocardial infarction. The measurements were performed with a Dimension RxL automatic analyzer. The immunoassay for troponin is based on the sandwich principle with two monoclonal antibodies binding to a reagent composed with alkaline phosphatase. The resulting product absorbs at 510 nm. The immunoassay for CK-MB uses an antibody specific for MB isoenzyme; the detection system is ß-galactosidase. The absence of method cross-activity was demonstrated by using control material from different muscle tissue (e.g. myocardium, brain, skeletal).