The realization of steroid hydroxylation by biotransformations with native microorganisms is well established in the steroid industry (e.g. at Schering and Pfizer) and has been performed for many decades on a massive scale (hundreds of tonnes per year). In a number of cases, the microbial enzyme systems that catalyze these reactions were purified and characterized to some extent, and as far as we know today, they are all cytochrome P450-dependent. However, to the best of our knowledge, the corresponding genes have not been cloned with the notable exception of the steroid 15P-hydroxylase from B. megaterium ATCC 13368 (see Section 6.4). This might be at least partially due to the low homology of these genes with mammalian steroid hydroxylases, thereby procluding their use as hybridization probes.
In any case, several consequences arise from this situation: (1) The phenotypi-cal optimization of the currently used organisms has been carried out very thouroughly and has led to processes with impressive space-time yields. Accordingly, there remains only little potential for further strain improvement via this method. (2) Because of the lack of genetic information, it is not (yet) possible to improve the rate of product formation by overexpression of the microbial P450s (and their electron transport partner proteins) in their natural environment (i.e. in the strains already used by industry). Also, the appropriate genetic tools for the genetic engineering of these organisms are scarcely established. Thus, it cannot be conjectured at the moment when this will be the case, and whether enzyme activity is actually the bottleneck of today's production processes. (3) It follows that in genetically modified microorganisms designed to carry out these reactions, enzymes of other (usually mammalian) origin and at typically lower activity must be used instead of their microbial counterparts (see Section 6.4). Naturally, these recombinant microorganisms tend to be the pertinent favorites of molecular biologists (like E. coli and S. cerevisiae), which have not normally been used on a large scale in steroid production processes so far. In view of all these considerations, the particular constraints of the transfer of scientific achievements in this field into industrial applications become evident.
Was this article helpful?