Expression of active human P450 3A4 on the cell surface of Escherichia coli by Autodisplay.

The cytochrome P450 enzyme system comprises a large group of enzymes catalyzing a broad diversity of reactions and an extensive substrate specificity, which makes them the most versatile known catalysts. CYP3A4 is one of the important human P450 enzymes and involved in the oxidation of a large range of substrates including toxins and pharmaceuticals. Bottlenecks in studying this enzyme include the difficulty in expressing it in a bacterial host, its need for membrane surroundings and the limited substrate accessibility of enzymes expressed within the cell. To circumvent these difficulties, human CYP3A4 was expressed on the outer membrane of Escherichia coli using Autodisplay. Transport of CYP3A4 to the cell surface was monitored by SDS-PAGE and Western blot analysis of outer membrane proteins. Localization on the cell envelope was determined by flow cytometry after immunolabeling, a whole cell ELISA and a protease accessibility assay. A HPLC assay confirmed the catalytic activity of displayed CYP3A4, using testosterone as a substrate. This activity required the external addition of electron supplying enzymes, however surprisingly, we found that the external addition of a heme group was not necessary. Our results indicate that human CYP3A4 can be recombinantly expressed by surface display in a gram-negative bacterium.

Autodisplay of functional CYP106A2 in Escherichia coli.

Cytochrome P450 enzymes catalyse a wide variety of reactions, including the hydroxylation and epoxidation of CC bonds, and dealkylation reactions. There is high interest in these reactions for biotechnology and pharmaceutical processes. Many P450s require membrane surroundings and have substrates that do not cross biological membranes. To circumvent these obstacles, CYP106A2 from Bacillus megaterium was expressed on the outer membrane of Escherichia coli cells by Autodisplay. Exposure on the surface was confirmed by a protease accessibility test and flow cytometry after immunolabelling. HPLC assays showed that 0.5 ml of cells displaying the enzyme (OD578 = 6) converted 9.13 µmol of deoxycorticosterone to 15ß-OH-deoxycorticosterone within 1h. Imipramine and abietic acid were also accepted as substrates. The number of active enzyme molecules per cell was calculated to be 20,000. Surprisingly, surface-exposed CYP106A2 was active in E. coli BL21 without the external addition of the heme group. However, when CYP106A2 was expressed on the surface of an E. coli strain lacking the TolC channel protein (JW5503), enzymatic activity was almost completely abolished. The activity of CYP106A2 on the surface of E. coli JW5503 could be restored by the external addition of the heme group. This suggests, as has been reported before, that E. coli uses a TolC-dependent mechanism to export heme into the growth media, where it can be scavenged by a surface-displayed apoenzyme. Our results indicate that Autodisplay enables the functional surface display of P450 enzymes and provides a new platform to access their synthetic potential.