Assays Based on NMR Spectroscopy

Traditionally, NMR measurements are considered to be slow processes, but recent advances in the design of flow-through cells have allowed the method to be applied in combinatorial chemistry [38]. These technological improvements were then ap- 

The exact ratio of the two pseudo enantiomers is accessible by simple integration of the respective peaks, which provides the ee value. Quantitative analysis can be accomplished automatically by suitable software such as AMIX (Bruker Biospin). The presence of naturally occurring 13C in the nonlabeled (R) substrate is automatically considered in the dataprocessing step. As demonstrated in control experiments, the agreement with the corresponding ee values obtained by independent GC analysis is excellent, the correlation coefficient amounting to 0.9998 [21]. 

The enzymatic reactions are performed in the wells of microtiter plates (96-format) in water (as in lipase-catalyzed hydrolytic reaction of (.S)-13C-4/(i )-4), which is followed by a standard automatic extraction step. Depending on the particular substrate to be assayed and the type of solvent used, it may be necessary to remove the solvent. However, this is often not necessary. For enzymatic reactions in organic medium, solvent extraction is not required. For NMR analysis such solvents as CDC13, Dg-DMSO, or D20 are used. A minimum of about 6 pmol of substrate/product per milliliter of solvent is needed. Although the flow-through cell system does not need too much solvent (about 1L in 24 h), the solvents can be mixed with the undeuterated form in 1 9 ratios to reduce costs. 

In one version, classical derivatization using a chiral reagent or NMR shift agent is simply parallelized and automated by the use of flow-through cells, with about 1400 ee measurements being possible per day with a precision of +5%. In the second embodiment, illustrated here in detail, a principle related to that of the MS system described in Section III.C is applied 98). Chiral or mexo-substrates are labeled to produce /. sewiio-enantiomers or psendo-meso-compo md that are then used in the actual screen. Application is thus restricted to kinetic resolution of racemates and 

In a typical application, C-labeling is used to distinguish between the (R)- and (S)-forms of a chiral compound. Essentially, any carbon atom in the compound of interest can be labeled (Fig. 11), but methyl groups in which the H signals are not split by coupling are preferred because the relevant peaks to be integrated are 

A typical example that illustrates the method concerns the lipase- or esterase-catalyzed hydrolytic kinetic resolution of rac-1-phenyl ethyl acetate, derived from rac-1-phenyl ethanol (20). However, the acetate of any chiral alcohol or the acetamide of any chiral amine can be used. A 1 1 mixture of labeled and non-labeled compounds (S)- C-19 and (f )-19 is prepared, which simulates a racemate. It is used in the actual catalytic hydrolytic kinetic resolution, which affords a mixture of true enantiomers (5)-20 and (J )-20 as well as labeled and non-labeled acetic acid C-21 and 21, respectively, together with non-reacted starting esters 19. At 50% conversion (or at any other point of the kinetic resolution), the ratio of (5)- C-19 to (1 )-19 correlates with the enantiomeric purity of the non-reacted ester, and the ratio of C-21 to 21 reveals the relative amounts of (5)-20 and (J )-20 (98). 

The ratio of the two pseudo-er a.r t ovasrs is accessible by simple integration of the respective peaks, which provides the ee and the E value (by use of an internal standard for conversion). The quantitative analysis can be accomplished automat- 

---