In order for the final process to be ultimately viable, a method to supply the aminotransferases in the quantities required and at an acceptable price was required. A procedure for the expression of these enzymes in Escherichia coli cells was developed. Optimization of the E. coli fermentation process has given rise to both a high level of expression within the cells as well as a high cell density. The final enzyme obtained after processing and isolation retains both specificity and high activity. While the yield of the isolated enzyme can vary slightly depending upon the exact enzyme required, the fermentation process itself remains essentially unchanged and scales-up very predictably (from 20 liters to several cubic meters volume). These combined factors ensure the supply of the enzyme at an acceptable cost.
With the ability to tailor the enzymes to be specific for molecules, to modify them to increase efficiency for transformations of interest, and to obtain them in sufficient quantity, all the critical items were in place to begin commercialization of these processes. Commercialization requires robust, cost effective processes, competing favorably with other technologies providing these amines, and in a larger sense, must yield an overall process to the API that competes with other routes. One of the additional advantages of the aminotransferase technology is that processes are simple, operating under mild conditions in mainly aqueous solution with simple unit operations in typical batch processing equipment. A further advantage is that processes are all very similar, even with different substrates and different enzymes, so implementation of many processes is simplified. At this point, there is a need to separate the synthesis mode from the resolution mode for further commercialization discussions.
The enzymatic synthesis mode is conceptually the simplest manner in which the desired chiral amine can be prepared. Here, the prochiral ketone is converted directly into the product. The first considerations with this approach are the availability and the cost of the starting prochiral ketone. Detailed discussions about these concerns are outside the scope of this article, but it is readily apparent that they will have a major impact. If the ketone cannot be made available at a reasonable cost, then other routes may be more advantageous. It should be recalled here that these are equilibrium reactions, and to obtain good conversion of ketone and yield of product, one frequently needs to drive the reactions toward products, this being accomplished by using large excess of an inexpensive donor amine. One then needs to identify an appropriate aminotransferase or undertake enzyme engineering. As indicated earlier, it is usually possible to obtain an enantiospecific enzyme that does not promote side-reactions. The amount of enzyme used in the process needs to be carefully evaluated as it has an impact on the process throughput, the fact that enzyme cost contribution will vary here is also readily apparent. To minimize equipment and labor costs, one would like to maximize the amount of product/volume/time, and this is where the tradeoffs typically enter into the process. Even with good enzyme activity, these synthesis reactions frequently require at lease several hours or tens of hours, and this for processes where amine product concentrations are normally on the order of only 1%. As one tries to increase the concentration of amine product, reaction completion typically requires more time, the equilibrium is shifted more toward reactants, and the enzyme may deactivate due to inherent stability or amine tolerance issues. Other problems such as ketone or amine solubility in these mainly aqueous systems may also be observed. A detailed knowledge about the interaction of all the components of the reaction process allows one to optimize the overall procedure. Workup procedures in these enzymatic transformations are composed of standard chemical processing unit operations. The enzyme biomass can be removed by filtration and/or by extraction procedures, the unreacted starting ketone and ketone by-product by extraction, and the unreacted donor amine by extraction or by distillation. Isolation of product amine depends upon the product itself and can be achieved by crystallization, precipitation as a salt, removal of solvent, and/or distillation.
The resolution mode would appear at first glance to be a less desirable choice, but there are instances in which this is the preferred option. Some conditions that would favor this approach would be unavailability of ketone or an extremely inefficient enzymatic synthesis process. One must be cautious in application of this mode since it competes directly with classical resolution and chiral separation, but there are cases in which it is quite viable. Similar to the synthesis mode above, the first concerns are availability and cost of the racemic amine. The issue here is to recover the maximum amount of the desired amine enantiomer. One must examine options for recycle of ketone byproduct through conversion back to racemic amine. The next concern, as with synthesis mode above, is identification or engineering of the enzyme. Here the specificity of the enzyme is of paramount importance. In order to obtain high ee, the reaction must be driven to completion, converting all of undesired enantiomer to the corresponding ketone. Removal of ketone from the process as it is formed can assist in forcing this equilibrium reaction in the desired direction. Suitability of the resolution mode in some cases becomes more apparent when it is noted that these reactions involving the aminotransferases are frequently much faster than the corresponding synthesis reaction, typically taking only a few hours, and operate at much higher concentrations, often several percent. Workup procedures are similar to those utilized in the synthesis mode, involving typical unit operations in standard chemical batch processing equipment. Given the proper relationship of cost factors and optimized procedures, enzymatic resolution can also provide a cost effective means of manufacturing certain chiral amines.
Example of improvement of Enzyme Performance Using Error-prone
In summary, biocatalytic processes utilizing aminotransferase technology have been developed for the cost effective production of a variety of chiral amines. While this approach may not be appropriate for all products, it is a valuable tool that can be applied to provide convenient access to important chiral molecules. Increasing the performance of these aminotransferases and the range of substrates that they show active toward is the subject of continuing efforts. |
