Exploring the combinatorial genomic space in Escherichia coli for ethanol tolerance

Strain tolerance to toxic chemicals is desirable for biologically producing biofuels and chemicals. Standard genomic libraries can be screened to identify genes imparting tolerance, but cannot capture interactions among proximal or distant loci. In search of ethanol tolerance determinants, we expanded the genomic space combinatorially by screening coexisting genomic libraries (CoGeLs) of fosmids (large inserts) and plasmids (smaller inserts) under increasing ethanol concentrations. Such screening led to identification of interacting genetic loci imparting ethanol tolerance. One pair of fragments ([galT, galE] and [recA, pncC, mltB]) increased survival under 50 g/L ethanol by 38% when coexpressed, but individually the fragments had no effect. Coexpression of two genomic fragments ([sfsB, murA, yrbA, mlaB, mlaC, mlaD, mlaE, mlaF, yrbG] and [yrbA, mlaB, mlaC]) enhanced Escherichia coli survival to 50 g/L ethanol by up to 115%. A 35-kb fosmid fragment increased tolerance to 63 g/L ethanol by 160%. While the tolerance levels of these strains compare favorably to or exceed the performance of previously reported engineered strains, more significantly, this study demonstrates that combinatorial library screening and screening fosmid libraries offer new, previously unexplored tools for identifying genetic determinants of ethanol, and by extrapolation, other alcohol tolerance.