Para hydroxybenzoate hydroxylase PHBH

A recent study of para-hydroxybenzoate hydroxylase (PHBH) by Bidder et al. [49,50] provides an interesting example of the validation of QM/MM calculations on the enzyme mechanism by comparison with experimental data. The correlation found between calculated activation barriers and the logarithm of experimental rate constants for a series of alternative substrates also provides support for the proposed mechanism hydroxylation of hydroxylation. These studies are a good example of QM/MM reaction pathway calculations for an enzyme, including technical aspects of system set-up and practical considerations, and so will be outlined here in some detail. 

The hydroxylation step was described by a reaction coordinate consisting of the difference between the breaking (peroxide distal oxygen-proximal oxygen) 

Further insight into the mechanism was provided by QM/MM analysis of the roles of active site groups on the hydroxylation step in PHBH. Qualitative information on nearby groups which significantly affect the reaction energetics can be found by a simple decomposition procedure described as a first-order perturbation analysis [13,48]. In essence, this involves the calculation of the 

Another aspect of the insight provided by the QM/MM calculations is that they strongly support substrate deprotonation as important for reaction in PHBH. QM/MM activation barriers for hydroxylation of the protonated substrate, and for para-fluorobenzoate (which is known not to be hydroxylated by PHBH, although it binds and initiates the reaction cycle) were twice as high as that for the deprotonated substrate [50]. 

---

However, the activated, protonated C(4a)-hydroperoxyflavin is not by itself a powerful oxidizing species. From detailed biochemical studies on the catalytic mechanism of such enzymes, especially para-hydroxybenzoate hydroxylase (PHBH) and phenol hydroxylase (PH), it appears that not only the peroxyflavin needs to be activated through protonation of the distal oxygen, but also the phenolic substrates require activation in order to obtain substrate conversion. This substrate activation is achieved through deprotonation of the hydroxyl moiety of the phenolic substrate. The active site of PHBH, for example, shows a tyrosine network consisting of tyrosines 385 and 201 (Fig. 4.82) responsible for this deprotonation and activation of the substrate. 

Much of the work into the structure and mechanism of aromatic hydroxylation by Class A FPMOs has come from studies of para-hydroxybenzoate hydroxylase (PHBH), the structure of which was first determined in 1979 [44]. PHBH catalyses not only the hydroxylation of 4-hydroxybenzoate 24 but also 2,4-dihydroxybenzoate 26, 4-mercaptobenzoate 27, and fluorinated and chlorinated derivatives of the parent substrate (Figure 8.10). 

The structure of PHBH is shown in Figure 8.12. It is an enzyme of 394 amino acids, with a three domain structure composed of an N-terminal "Rossman fold" domain that binds the ADP of the FAD coenzyme, a second domain that binds the para-hydroxybenzoate substrate and, a third, largely helical domain that is involved in dimerization wdth another subunit. The tricyclic isoalloxa-zine ring of the flavin is bound at the interface between the first two major domains, and there is known to be movement of flavin associated with the accommodation of substrate [46]. The structure of PHBH in the presence of NADPH has not been determined, but it is thought that there is only a transient association of the two cofactors to effect flavin reduction after which the NADP+ dissociates to permit substrate binding [46]. Although PHBH itself has not been the focus of studies in applications in synthesis, other Class A FPMO aromatic hydroxylases, such as hydroxybiphenyl monooxygenase from Pseudomonas azelaica have been used for the preparation of hydroxylated compounds. 

---