Background The aryltetralin lignan (−)−podophyllotoxin is a potent antiviral and anti-neoplastic compound that is mainly found in Podophyllum plant species. Over the years, the commercial demand for this compound rose notably because of the high clinical importance of its semi-synthetic chemotherapeutic derivatives etoposide and teniposide. To satisfy this demand, (−)−podophyllotoxin is conventionally isolated from the roots and rhizomes of Sinopodophyllum hexandrum, which can only grow in few regions and is now endangered by overexploitation and environmental damage. For these reasons, targeting the biosynthesis of (−)−podophyllotoxin precursors or analogues is fundamental for the development of novel, more sustainable supply routes. Results We recently established a four-step multi-enzyme cascade to convert (+)−pinoresinol into (−)−matairesinol in E. coli. Herein, a five-step multi-enzyme biotransformation of (−)−matairesinol to (−)−deoxypodophyllotoxin was proven effective with 98 % yield at a concentration of 78 mg/L. Furthermore, the extension of this cascade to a sixth step leading to (−)−epipodophyllotoxin was evaluated. To this end, seven enzymes were combined in the reconstituted pathway involving inter alia three plant cytochrome P450 monooxygenases, with two of them being functionally expressed in E. coli for the first time. Conclusions Both, (−)−deoxypodophyllotoxin and (−)−epipodophyllotoxin, are direct precursors to etoposide and teniposide. Thus, the reconstitution of biosynthetic reactions of Sinopodophyllum hexandrum as an effective multi-enzyme cascade in E. coli represents a solid step forward towards a more sustainable production of these essential pharmaceuticals.

Cytochromes P450 catalyze oxidation of chemically diverse compounds and thus offer great potential for biocatalysis. Due to the complexity of these enzymes, their dependency of nicotinamide cofactors and redox partner proteins, recombinant microbial whole cells appear most appropriate for effective P450-mediated biocatalysis. However, some drawbacks exist that require individual solutions also when P450 whole-cell catalysts are used. Herein, we compared wet resting cells and lyophilized cells of recombinant E. coli regarding P450-catalyzed oxidation and found out that lyophilized cells are well-appropriate as P450-biocatalysts. E. coli harboring CYP105D from Streptomyces platensis DSM 40041 was used as model enzyme and testosterone as model substrate. Conversion was first enhanced by optimized handling of resting cells. Co-expression of the alcohol dehydrogenase from Rhodococcus erythropolis for cofactor regeneration did not affect P450 activity of wet resting cells (46% conversion) but was crucial to obtain sufficient P450 activity with lyophilized cells reaching a conversion of 72% under the same conditions. The use of recombinant lyophilized E. coli cells for P450 mediated oxidations is a promising starting point towards broader application of these enzymes.