Front Matter

18.5 Further aspects and future prospects

The information provided in this article, and especially the examples given above,

demonstrate that P450 enzymes are indeed a versatile and useful class of enzymes.

Industrial processes based on whole-cell hydroxylation steps are already known for

example in progesterone modifications (Sallam et al., 1977), 3-a-hydroxyl-5-beta-

cholic acid (Kulprecha et al., 1985) orb-hydroxybutyric acid production (Evans et

al., 1979). However, the use of isolated and purified P450 enzymes in a preparative-

scale synthesis of fine chemicals has to our knowledge not yet been reported. Major

limitations in the use of P450 enzymes reside in their difficult handling, low turnover

numbers (often between 0.1 and 100 equiv min–1), lack of thermostability (Juchau,

1990; Yamazaki et al., 1997) and high sensitivity against organic solvents (Wade et

al., 1972; Erjomin and Metelitza, 1983; Kuhn-Velten, 1997). This is especially true

for membrane-bound P450s, which represent the largest population of P450 en-

zymes, because their activity depends strongly on the reconstitution conditions,

and the ratio of P450 to reductase (Scheller et al., 1996), as well as the phospholipid

composition used for reconstitution (Blanck et al., 1989; Kisselev et al., 1998). The

lack of simple and rapid activity assays required to characterize P450 enzymes or to

screen for improved P450 enzyme variants (Moore et al., 1997; Zhao et al., 1998), as

well as the need for cofactor recycling, pose further major handicaps on the way to

exploiting this class of catalysts for industrial applications.

However, much effort has been undertaken during recent years to overcome these

drawbacks. Nowadays, even mammalian P450 enzymes can be expressed afterN-

terminal modification at high levels inE. coli(Barnes et al., 1991; Fisher et al.,

1992a,b; Gillam et al., 1993; Barnes, 1996) or in yeast expression systems, yielding

10–100 mg P450 per liter of fermenter broth. To render membrane-associated P450

enzymes water-soluble, and to avoid reconstitution experiments, artificial fusion

proteins between the P450 domain and the reductase part have been successfully

constructed (Murakami et al., 1987; Fisher et al., 1992a,b; Yabusaki, 1995; Chun

et al., 1997). A promising method of bypassing the cofactor regeneration of NADPH

is, as shown for example for P450 BM-3, by using electrochemical means (Faulkner

et al., 1995; Estabrook et al., 1996a,b), or with cobalt(III)sepulchrate, which can be

reduced by cheap zinc dust (Schwaneberg et al., unpublished results). Alternatively,

the shunt pathway (Figure 1) has been successfully evolved to drive the P450cam

catalyst with hydrogen peroxide instead of NADPH (Joo et al., 1999a,b). In addition,

the tremendous development of directed evolution methods during the past few years

has resulted in powerful tools such as gene-shuffling (Stemmer, 1994) or the StEP-

method (Zhao et al., 1998). These tools have been applied successfully to improve a

wide range of different enzyme properties such as reactivity, solvent stability (Moore

and Arnold, 1996), thermal stability (Giver et al., 1998) and regio- and stereoselec-

tivity (Bornscheuer et al., 1999; Henke and Bornscheuer, 1999; May et al., 2000). It

is likely that directed evolution will be applied to P450 enzymes in the near future,

resulting in P450 catalysts which are more suitable for industrial applications.

18.5 Further aspects and future prospects 407