Combinatorial libraries analytical characterization

The new field of molecular diversity raises three issues which need to be addressed by the organic analytical chemistry community (i) What tools can we use for following solid-phase reactions (ii) How can we analyze all these samples (iii) How much characterization of libraries is possible or appropriate This chapter deals with these problems and reviews the literature since a similar review written in June 1995 [2] (earlier seminal publications are described where appropriate). Other analytical issues such as decoding of combinatorial libraries or the applications of affinity separations and single-bead mass spectrometry for library deconvolution are dealt with in other chapters of this book. 

This issue highlights the characterization difference between parallel synthesis and combinatorial synthesis. Parallel synthesis is automated traditional organic chemistry. Each compound is made in a separate reactor, purified and characterized. There is no excuse for not fully characterizing compounds made by parallel synthesis. Jonathan Ellman s laboratory at UC Berkeley has been a pioneering academic center for solid-phase chemistry development. His philosophy is to synthesize libraries of discrete compounds in a spatially separate fashion, rather than libraries of compound mixtures, to allow for rigorous analytical characterization [48,49],... 

Sepetov N, Issakova O, Analytical characterization of synthetic organic libraries, in Combinatorial Chemistry and Combinatorial Technologies Methods and Applications (Eds. Fassina G, Miertus S), 237-268, 1998, 2005, Chapter 10, this volume. 

It is worth mentioning that a subject for analytical characterization is a library, but technically compounds are analyzed. What is the difference between analysis of a combinatorial library and analysis of a collection of many compounds There are at least two major differences related to quantitative and qualitative aspects of combinatorial libraries. The first is that combinatorial synthesis is a practically unlimited source of compounds. The number of compounds in a library may very easily reach the point at which none of the existing analytical methods will allow measurement of all compounds in a library. The second major difference between a combinatorial library and a... 

A. General Strategy of Analytical Characterization of Combinatorial Libraries... 

In this chapter we will discuss current approaches for analytical characterization of combinatorial libraries in a pharmaceutical industry environment. Recently, several analytical groups have presented very similar strategies for analysis of libraries [7-9]. As will be shown later, the key to successful analytical characterization of a combinatorial library is to perform analytical and chemical work in parallel with the library development. The accumulation of data and analytical experience during this process results in an assessment of library quality with a high level of confidence, even if as little as 5-10% of the library components are analyzed. Utilization of the strategy will be demonstrated using two examples analysis of a library synthesized on a robotic station in spatially addressed format and analysis of a library synthesized in accordance with split-and-mix technology. 

Any MS experiment begins with ionization of molecules of analyte. Numerous ionization techniques (electron ionization, fast atom bombardment, plasma desorption, electrospray ionization, etc.) allow MS analysis of a wide range of organic molecules. In most cases the characterization of combinatorial libraries means analysis of crude compounds i.e. one can expect not only the intended compound to be present in the analyte, but also products of side... 

Analytical characterization of OBOC library will be discussed for an example of a small-molecule combinatorial library with a complexity of 54,150 (57 x 25 x 38) compounds (Figure 10.9) [63]. In the result of an OBOC library analysis, identity, purity, and quantity of library components should be evaluated, but it is unknown which one out of thousands of compounds in a library is synthesized on the particular bead picked for analysis (Figure 10.10). General schemes of analytical characterization of combinatorial libraries (analysis of a small number of standard compounds, analysis of the model library, and analysis of the production-scale library) can be applied to analysis of OBOC libraries, but the issue of structure elucidation for OBOC libraries should be addressed through the scheme. 

Concluding this chapter, we would like to stress that the analytical characterization of combinatorial libraries is a very rapidly developing and changing area of analytical chemistry. The variety of synthetic strategies used in combinatorial chemistry continues to grow and requires development of new analytical methods. The authors would not be surprised if new ones replaced many of the approaches described in this chapter or radically... 

Issakova O, Nikolaev V, Ma N, Wade S, Sepetov N, Analytical characterization of one-bead-one-compound combinatorial libraries by LC/MS and LC/MS/ MS, Proc. 45th ASMS. Conf, p. 403, 1997. 

In addition to extending the utility of analytical NMR, these developments have had a direct impact on screening methodologies, where NMR spectroscopy may now be used directly as a rapid and sensitive method for the characterization of binding affinities of mixtures of compounds in combinatorial libraries.3,9 11 A powerful screening method based on observation of NMR signals from a protein target has been developed by researchers at Abbott laboratories.12 Termed SAR (structure-activity relationship) by NMR, this technique has proved to be extremely valuable for the identification of novel lead compounds. 

Mass spectrometric detection has also been directly interfaced with microchip separations for drug detection. These studies, detecting imipramine and desipramine in fortified human plasma, show analysis of spiked analytes in clinical sample matrices for drug detection [3]. These widely used tricyclic antidepressants inhibit the reuptake of the neurotransmitters serotonin and norepinephrine in the central nervous system. Unfortunately, the 5-mg/mL detection limit found for these antidepressants with this method is not low enough to detect typical clinical levels of the drugs. Combinatorial library characterization and preclinical drug delivery studies should benefit, however, since the concentra-... 

Why A primary library can be planned for many purposes. Sometimes the project goal is purely academic, that is, simply to open a new combinatorial synthetic route and to report it. This respectable option allows complete freedom to prepare any primary library, with no constraints dictated by parameters such as application, cost, or resources. We will not comment further upon this approach in this section. More often, though, a primary library is prepared as a source of relevant active molecules on various biological targets. This obviously necessitates the availability of several robust and reliable HTS assays for these targets, as well as the synthetic resources, the analytical resources, and the instrumentation to ensure the successful synthesis and analytical characterization of the library. 

Flow injection analysis mass spectrometry (FIA-MS) has been reported to be a fast method for the characterization of combinatorial libraries (55,56). The method verifies the presence of the molecular ions of the expected product and side products or impurities but does not provide information on the quality of the analyzed samples. Significant improvements related to the increased analytical throughput, obtained by reducing the time between each injection without increasing the intersample carry-over from each analysis, were recently reported (57, 58). When coupled with RP-HPLC, FIA-MS allows the separation and the determination of the molecular weight of the components of each sample. This is normally enough to unequivocally attribute the structure of the expected library component and of any side products from a library synthesis. 

L. Zeng, and D. B. Kassel, Developments of a fully automated parallel HPLC/mass spectrometry system for the analytical characterization and preparative puriflca-tion of combinatorial libraries, Anal. Chem. 70 (1998), 4380-4388. 

As for all synthetic products to be tested in biological systems, a careful analytical characterization of peptide libraries is crucial in order to confirm their identity and establish their quality. Compared to individual peptides, however, the analysis of peptide libraries is complicated due to the fact that the peptides are either bound to a solid support or arranged in highly complex mixtures. This poses certain restrictions on which analytical methods can be used to characterize combinatorial libraries. For example, analytical methods that are based on the separation of product components, such as high performance liquid chromatography (HPLC) and capillary electrophoresis (CE), are only of limited use for the analysis of peptide libraries, in particular of those made up of complex nnixtures (>100 peptides per mixture). The analytical methods beneficially applicable to peptide libraries include amino acid analysis, mass spectrometry, and sequencing. 

Characterization of combinatorial libraries (product structure determination) can be performed by direct analytical methods, while compounds are still attached to the beads. Analytical techniques such as NMRandMAS-NMR,1314ESI-MSandHPLC-ESI-MS,15-17... 

Because no separation is used, only crude information about the purity of the compounds can be obtained. For example, if unreacted, synthetic starting materials (Fig. 10b) are present in a sample of a combinatorial product (Fig. 10a), these can show up in the mass spectrum of the crude product (Fig. 10c). Because the starting materials are often structurally different from the finished product, a simple extraction can be used after the synthesis to remove much of the unused reactants and obtain a cleaner mass spectrum (Fig. lOd) for a solution-phase product. Conversely, washing the resin after solid-phase synthesis or using a scavenger resin in a solution-phase synthesis can also yield improved purity by removing these excess reactants. When direct flow injection is used to characterize combinatorial libraries, it is best to avoid dimethyl sulfoxide (DMSO) as a solvent, because it interferes with reliable ionization of the analytes. 

Zeng, L. Kassel, D.B. Developments of a Fully Automated Parallel HPLC/Mass Spectrometry System for the Analytical Characterization and Preparative Purification of Combinatorial Libraries, Anal. Chem., 70(20), 4380 388 (1998). 

The MS instrumentation is the most expensive part of the LC-MS system, hence efforts to improve the throughput of the LC-MS analysis often involve the use of parallel multiple columns that feed into a single mass spectrometer. Zeng and Kassel [99] developed an automated parallel analytical/preparative LC-MS workstation to increase the throughput for the characterization and purification of combinatorial libraries. The system incorporates two columns operated in parallel for both LC-MS analytical and preparative LC-MS purifications. A multiple-sprayer ESI interface was designed to support flows from multiple columns. The system is under complete software control and delivers the crude samples to the two HPLC columns from a single autosampler. The authors demonstrated characterization of more than 200 compounds per instrument per day, and purification of more than 200 compounds per instrument per night. De Biasi et al. [100] described a four-channel multiplexed... 

Capillary electrophoresis (CE) is a powerful separation technique. It is especially useful for separation of ionic compounds and chiral mixtures. Mass spectrometry has been coupled with CE to provide a powerful platform for separation and detection of complex mixtures such as combinatorial libraries. However, the full potential of CE in the application of routine analysis of samples has yet to be realized. This is in part due to perceived difficulty in the use of the CE technique compared to the more mature techniques of HPLC and even SFC. Dunayevskiy et al. [136] analyzed a library of 171 theoretically disubstituted xanthene derivatives with a CE/ESI-MS system. The method allowed the purity and makeup of the library to be determined 160 of the expected compounds were found to be present, and 12 side products were also detected in the mixture. Due to the ability of CE to separate analytes on the basis of charge, most of the xanthene derivatives could be resolved by simple CE-MS procedures even though 124 of the 171 theoretical compounds were isobaric with at least one other molecule in the mixture. Any remaining unresolved peaks were resolved by MS/MS experiments. The method shows promise for the analysis of small combinatorial libraries with fewer than 1000 components. Boutin et al. [137] used CE-MS along with NMR and MS/MS to characterize combinatorial peptide libraries that contain 3 variable positions. The CE-MS method was used to provide a rapid and routine method for initial assessment of the construction of the library. Simms et al. [138] developed a micellar electrokinetic chromatography method for the analysis of combinatorial libraries with an open-tube capillary and UV detection. The quick analysis time of the method made it suitable for the analysis of combinatorial library samples. CE-MS was also used in the analysis... 

---