- 4:45 PM

Process and Enzyme Engineering of Aminotransferases for Improved Activity and Thermostability

Abraham Rogelio Mártin-García1, David R. Shonnard1, and Sachin Pannuri2. (1) Chemical Engineering, Michigan technological University, 1400 Townsend Drive, Houghton, MI 49931-1295, (2) Research and Development, Cambrex North Brunswick, Technology Centre of New Jersey, 661 Highway One, North Brunswick, NJ 08902



Process and Enzyme Engineering of Aminotransferases for Improved Activity and Thermostability


Abraham Rogelio Mártin García(1)(a), David R. Shonnard(1)(b), and Sachin Pannuri(2)(c)

1. Department of Chemical Engineering, Michigan Technological University

 2. Cambrex, North Brunswick, NJ

(a) armartin@mtu.edu

(b) drshonna@mtu.edu 

(c) Sachin.Pannuri@Cambrex.com 



The production by biosynthesis of optically active amino acids and amines satisfies the pharmaceutical industry in its demand for chiral building blocks for the synthesis of various pharmaceuticals. Among several enzymatic methods that allow the synthesis of optically active aminoacids and amines, the use of aminotransferase is a promising one due to its broad substrate specificity and no requirement for external cofactor regeneration. The synthesis of chiral compounds by aminotransferases can be done either by asymmetric synthesis starting from keto acids or ketones, and by kinetic resolution starting from racemic aminoacids or amines.  

The asymmetric synthesis of substituted (S)-aminotetralin, an active pharmaceutical ingredient (API) has shown to have two major factors that contribute to increasing the cost of production. These factors are the raw material cost of biocatalyst used to produce it and product loss during biomass separation. To minimize the cost contribution of biocatalyst and to minimize the loss of product, two routes have been chosen in this research: 1. To engineer the biocatalyst have greater specific activity 2. Improve the engineering of the process by immobilization of biocatalyst in calcium alginate and addition of cosolvents.  

An (S)-aminotransferase was immobilized and used to produce substituted (S)-aminotetralin at 50 °C and pH 7 in experiments where the immobilized enzyme was recycled. Initial rate for cycle 1 (6 hr duration) was determined, for cycle 2 (20 hr duration) it decreased by ~50% compared to cycle 1, and for cycle 3 (20 hr duration) it decreased by ~90% compared to cycle 1 (immobilized preparation consisted of 50 mg of spray dried cells per gram of calcium alginate). Total product accumulation for each cycle decreased as well, from 100% in cycle 1, 80% in cycle 2, and 30% after cycle 3. This enzyme was determined to be deactivated at elevated temperatures during the reaction cycle and was not stable enough to allow multiple cycles in its immobilized form.

  A new enzyme was isolated by means of error-prone polymerase chain reaction (PCR) and screening. This enzyme showed a significant improvement in thermostability in comparison to previous enzyme. The new enzyme was immobilized and tested under similar conditions.  Initial rate remained fairly constant over four cycles (each cycle with a duration of about 20 hours) with the enzyme retaining almost 80% of initial rate in the fourth cycle.  The final product concentrations after each cycle did not decrease during recycle experiments. Thermostability of the new enzyme was much improved compared to the previous enzyme.   

Under the same conditions as stated above, the addition of co-solvents was studied. Toluene and sodium dodecyl-sulfate (SDS) were used.   SDS at 0.01% (w/v) allowed four recycles of the immobilized enzyme, always reaching higher product concentration than the system with toluene at 3% (v/v). The initial rate of immobilized enzyme in a system with SDS 0.01% (w/v) at 50 °C, pH 7 was retained for three cycles but dropped precipitously in the fourth cycle. The final product concentrations for each cycle also followed the same pattern although significant improvement of immobilized enzyme productivity and stability were observed by one round of mutagenesis; another observation demonstrated the limitations of an immobilization strategy on reducing process economics.  After analyzing the results of this experiment it was seen that a sudden drop occurred in activity after the third recycle. This was due to a byproduct imine accumulation inside the immobilized preparation (a reaction of product amine with reactant ketone). In order to improve the economics of the process research was focused on developing an enzyme with an even higher activity thus reducing raw material cost as well as improve biomass separation.  

A new enzyme was obtained using error-prone PCR and screening derived from the previous improved enzyme. This enzyme was determined to have three times the initial rate as the previous enzyme and had a higher temperature optimum. This new enzyme was able to reduce enzyme loading in the reaction by three-fold. In addition, the decrease in enzyme loading allows for a better biomass separation leading to smaller product loss.