Introduction

The Baeyer-Villiger oxidation reaction was discovered more than 100 years ago by Adolf von Baeyer and Victor Villiger.1 By this reaction, ketones are converted into the corresponding esters. In organic chemistry, peracids are commonly used as catalyst to perform this atypical oxidation reaction that results in oxygen insertion into a carbon—carbon bond (Fig. 1).

Already in 1948 it was recognized that enzymes that catalyze Baeyer-Villiger reactions exist.2 This was concluded from the observation that a biological Baeyer-Villiger reaction occurred during the biotransformation of steroids by fungi. It took two decades for the first Baeyer-Villiger monooxygenases (BVMOs, EC 1.14.13.X), sometimes also referred to as Baeyer-Villigerases, to be isolated and characterized.34 From then on, a number of microbial BVMOs have been reported revealing several recurring biochemical characteristics (for a recent review see Ref. 5).

All characterized BVMOs contain a flavin cofactor that is crucial for catalysis while NADH or NADPH is needed as electron donor. An interesting observation is the fact that most reported BVMOs are soluble proteins. This is in contrast to many other monooxygenase systems that often are found to be membrane-bound or membrane-associated. In 1997, Willetts6 concluded from careful inspection of

O Baeyer-Villiger O

Figure 1: The Baeyer-Villiger oxidation reaction.

all available biochemical data on BVMOs that at least two classes of BVMOs exist. The type I BVMOs consist of only one polypeptide chain, contain FAD as tightly bound cofactor and are dependent on NADPH for activity. They contain two Rossmann sequence motifs, GxGxxG, indicating that these enzymes bind the two cofactors (FAD or NADPH) using separate dinucleotide-binding domains.7 The type II BVMOs use FMN as flavin cofactor and NADH as electron donor and are composed of two different subunits. At the time of the initial classification, the respective N-terminal sequences did not provide any clue concerning the structure of these two-component monooxygenases. However, recent sequence data suggest a sequence relationship with the flavin-dependent luciferases.8 Therefore, it is likely that type II BVMO oxygenase subunits will also have a TIM-barrel fold.

Recent findings hint to at least two other BVMO classes of which one also represents a group of flavoproteins. It was found that the bacterial flavoprotein monooxygenase MtmOIV is involved in the biosynthetic pathway of the antitumor drug mithramycin.9 Sequence analysis indicates that it is related to flavoprotein monooxygenases that typically perform hydroxylation or epoxidation reactions.8 Crystals of this novel monooxygenase have recently been reported.10 A crystal structure would reveal what structural features separate this BVMO from the sequence-related hydroxylases and epoxidases. Also a heme-containing BVMO has recently been reported belonging to the cytochrome P450 superfamily.11 This plant enzyme was shown to convert a specific plant steroid. Earlier studies already suggested Baeyer-Villiger activity of other eukaryotic P450s.12 Future studies will reveal whether these novel oxidative enzymes can be of use for biocat-alytic applications. However, the first results suggest that these newly identified BVMOs are dedicated to convert very specific and complex molecules suggesting a narrow substrate specificity. Nevertheless, the finding of these novel BVMO types indicates that during evolution several different enzymes have evolved into BVMOs. Therefore, more BVMO types may be discovered in the coming years.

Most biochemical and biocatalytic studies have been performed with type I BVMOs.5 This is partly because of the fact that they represent relatively uncomplicated monooxygenase systems. These monooxygenases are typically soluble and composed of only one polypeptide chain. Expression systems have been developed for a number of type I BVMOs while no recombinant expression has been reported for a type II BVMO. Cyclohexanone monooxygenase (CHMO) from an Acinetobacter sp. NCIMB9871 was the only recombinant available BVMO

Release of NADP+ Release of water

NADPH-binding flavin reduction

NADP+

NADP+

Reaction with oxygen Formation of peroxyflavin

Figure 2: Scheme of the catalytic mechanism of type I BVMOs. Phenylacetone is taken as example substrate.

Figure 2: Scheme of the catalytic mechanism of type I BVMOs. Phenylacetone is taken as example substrate.

for a long time as it was cloned and overexpressed already in 1988.13 CHMO has been subjected to several sophisticated kinetic studies which have revealed that in BVMOs also, catalysis is achieved by formation of a peracid catalyst: a peroxyflavin (Fig. 2).1415

Upon reaction with NADPH and molecular oxygen, the flavin cofactor is able to form this peroxygenated flavin intermediate. This reactive intermediate is equivalent to the peracids used in organic chemistry and will react with a ketone to form an ester. In fact, it is the ability of BVMOs to form and stabilize a negatively charged peroxyflavin intermediate that enables these enzymes to perform Baeyer-Villiger reactions and other oxygenation reactions. The formation of the reactive oxygenated enzyme intermediate is not regulated by substrate binding which sets these BVMOs mechanistically apart from other well-studied monooxygenase systems. For example, cytochrome P450s and flavin-containing hydroxylases will only form the equivalent reactive enzyme intermediate after binding of a substrate. The peroxyflavin in CHMO is stabilized by active-site residues and the bound NADP+ coenzyme. Structural details concerning this enzyme complex are lacking as no CHMO structure is available. After its formation, the peroxyflavin enzyme intermediate waits until a suitable substrate enters the active site upon which oxygenation will take place. In the case that no suitable substrate is present, the peroxyflavin will decay to form hydrogen peroxide. However, this NADPH oxidase function of BVMOs is very inefficient (< 0.1s-1) due to the effective stabilization of the peroxy intermediate. This prevents intracellular formation of toxic hydrogen peroxide. A striking feature of the catalytic mechanism of CHMO is the fact that the coenzyme NADPH/NADP+ remains bound to the enzyme throughout the catalytic cycle. Only when the oxygenation reaction and the decay of the hydroxyflavin into oxidized flavin have occurred, NADP+ is released (Fig. 2). This mechanistic feature was recently confirmed for another type I BVMO: 4-hydroxyacetophenone monooxygenase (HAPMO)16. By kinetic inhibition studies and ESI-MS experiments, it could be demonstrated that the coenzyme remains bound during the whole catalytic cycle. Also, with the newly identified phenylacetone monooxygenase (PAMO)17 we have found that NADP+ is a competitive inhibitor with respect to NADPH (M. W. Fraaije, personal communication), suggesting that NADP+ release is again the last step in the catalytic cycle. Another indirect proof for binding of NADP+ throughout the catalytic cycle came from a study where an artificial electron donor was tested.18 This revealed that NADP+ binding is essential to maintain the enantioselectivity of PAMO, indicating that the bound coenzyme forms parts of the active site determining the positioning of the substrate. It also indicates that, in vivo, BVMOs are virtually always occupied by NADP+ or NADPH. This is in line with the observation that BVMOs are highly stabilized when NADP+ is bound.16

Was this article helpful?

0 0

Post a comment