Abstract
Optically active epoxides are useful chiral synthons in organic synthesis. They have been used as intermediates in the synthesis of natural compounds and biologically active substances, because they are easily converted to various derivatives stereoselectively and regioselectively. But, hitherto the Sharpless oxidation has been the only practical method available for asymmetric epoxidation of olefinic compounds. On the other hand, microbial epoxidation of olefinic compounds has been investigated for the last 30 years. Although the epoxides produced by microorganisms have been shown to be optically active in most cases, they have not been employed in organic synthesis because of low productivity.
The authors have studied microbial epoxidation by Nocardia corallina B-276, which was isolated from soil with propylene as the sole source of carbon. This microorganism differs from other microorganisms because it grows on carbohydrates such as glucose and sucrose, and shows epoxidating activity while others exhibit their activities only when they grow on hydrocarbons such as methane, ethane, ethylene and higher alkanes and alkenes. This strain has a wide range of substrate specificity in epoxidation and produces epoxides from corresponding 1-alkenes (C2-C18), 2-alkenes, halogenated alkenes, allyl ether, styrene and related compounds. The optical purity of the products changed according to structure of substrate olefinic compounds. It ranged between 66 and 94%ee in aliphatic 1, 2-epoxides, while it ranged from 5 to 97%ee in aromatic ones. The optical purity of each product did not change by reaction conditions.
The productivity of epoxides by N. corallina B-276 has been improved in 3 ways: by carbon number of the substrates, degree of product inhibition and physical properties of the products. A two-liquid phase reaction system was applied to epoxidation of moderate chain length alkenes (C6-C12) by resting cell reaction with glucose-grown cells. Productivity of 1, 2-epoxyoctane was improved: by adding organic solvents, such as n-hexadecane, to the reaction mixture and the concentration of 1, 2-epoxyoctane in the reaction mixture reaching 30g/l after an optimization of reaction condition. Substrate inhibition was not observed in the epoxidation of 1-octene and the rate of epoxidation decreased when the product was added to reaction mixture. These results imply that the organic solvent reduces product inhibition by 1, 2-epoxyoctane. The same system enabled us to produce styrene oxide, phenyl glycidyl ethers and related compounds. In styrene oxide production, n-hexadecane reduced both product and substrate inhibition. The products accumulated in the solvent can be recovered by distillation.
N. corallina B-276 grows well on 1-alkenes above C13 and accumulates 1, 2-epoxides because of the weak product inhibition by epoxides with long chain length. In an optimized medium and cultivating conditions, the concentration of 1, 2-epoxytetradecane reached 80g/l in 6 days. This growing cell reaction is applicable to C13-C17 1-alkenes.
There are two difficulties in the production of epoxides from gaseous olefins. One is strong product inhibition and the other is recovery of the product present in exhaust gas in a very low concentration. In the production of propylene oxide, it was possible to raise production rate and maintain it by transferring the product to gas phase and keeping concentration of the product in reaction mixture low with high rate of aeration. After an optimization of reaction condition, the rate of propylene oxide production reached 0.4mol/l/day. The product in gas phase can be recovered at a favorable yield by a new recovery process employing absorption by solvents.
Among these optically active epoxides produced by N.
