Optimization of genome shuffling for high-yield production of the antitumor deacetylmycoepoxydiene in an endophytic fungus of mangrove plants.

Abstract

As an accelerated evolutionary tool, genome shuffling is largely dependent on the high fusion frequency of different parental protoplasts. However, it was unclear how many types of parental protoplasts would afford the highest fusion frequency. Here, we applied the Monte Carlo method to simulate the simplified processes of protoplast fusion, to achieve maximal useful fusions in genome shuffling. The basic principle of this simulation is that valid fusions would take place when the minimum distance between two different types of parent protoplasts is smaller than that between two of the same types. Accordingly, simulations indicated that the highest fusion frequency would be achieved from eight to 12 different parental protoplasts. Based on the simulation results, eight parental protoplasts of the fungal endophyte Phomopsis sp. A123 were subjected to genome shuffling for yield improvement of deacetylmycoepoxydiene (DAM), an antitumor natural product with a novel chemical structure. After only two rounds of genome shuffling, four high-yield DAM-producing strains, namely G2-119, G2-448, G2-866, and G2-919, were obtained with the aid of activity screening and HPLC analysis. The results showed that the DAM yield in these four strains were 243-, 241-, 225-, and 275-fold, respectively, higher than that of the starting strain A123. This is the first time Monte Carlo simulation is introduced into the field of cell fusion and is also the first report on the optimization of genome shuffling focusing on the number of parental types in protoplast fusions.