Altering toluene 4-monooxygenase by active-site engineering for the synthesis of 3-methoxycatechol, methoxyhydroquinone, and methylhydroquinone

Abstract

Wild-type toluene 4-monooxygenase (T4MO) of Pseudomonas mendocina KR1 oxidizes toluene to p-cresol (96%) and oxidizes benzene sequentially to phenol, to catechol, and to 1,2,3-trihydroxybenzene. In this study T4MO was found to oxidize o-cresol to 3-methylcatechol (91%) and methylhydroquinone (9%), to oxidize m-cresol and p-cresol to 4-methylcatechol (100%), and to oxidize o-methoxyphenol to 4-methoxyresorcinol (87%), 3-methoxycatechol (11%), and methoxyhydroquinone (2%). Apparent Vmax values of 6.6 +/- 0.9 to 10.7 +/- 0.1 nmol/min/ mg of protein were obtained for o-, m-, and p-cresol oxidation by wild-type T4MO, which are comparable to the toluene oxidation rate (15.1 +/- 0.8 nmol/min/mg of protein). After these new reactions were discovered, saturation mutagenesis was performed near the diiron catalytic center at positions I100, G103, and A107 of the alpha subunit of the hydroxylase (TmoA) based on directed evolution of the related toluene o-monooxygenase of Burkholderia cepacia G4 (K. A. Canada, S. Iwashita, H. Shim, and T. K. Wood, J. Bacteriol. 184:344-349, 2002) and a previously reported T4MO G103L regiospecific mutant (K. H. Mitchell, J. M. Studts, and B. G. Fox, Biochemistry 41:3176-3188, 2002). By using o-cresol and o-methoxyphenol as model substrates, regiospecific mutants of T4MO were created; for example, TmoA variant G103A/A107S produced 3-methylcatechol (98%) from o-cresol twofold faster and produced 3-methoxycatechol (82%) from 1 mM o-methoxyphenol seven times faster than the wild-type T4MO (1.5 +/- 0.2 versus 0.21 +/- 0.01 nmol/min/mg of protein). Variant I100L produced 3-methoxycatechol from o-methoxyphenol four times faster than wild-type T4MO, and G103S/A107T produced methylhydroquinone (92%) from o-cresol fourfold faster than wild-type T4MO and there was 10 times more in terms of the percentage of the product. Variant G103S produced 40-fold more methoxyhydroquinone from o-methoxyphenol than the wild-type enzyme produced (80 versus 2%) and produced methylhydroquinone (80%) from o-cresol. Hence, the regiospecific oxidation of o-methoxyphenol and o-cresol was changed for significant synthesis of 3-methoxycatechol, methoxyhydroquinone, 3-methylcatechol, and methylhydroquinone. The enzyme variants also demonstrated altered monohydroxylation regiospecificity for toluene; for example, G103S/A107G formed 82% o-cresol, so saturation mutagenesis converted T4MO into an ortho-hydroxylating enzyme. Furthermore, G103S/A107T formed 100% p-cresol from toluene; hence, a better para-hydroxylating enzyme than wild-type T4MO was formed. Structure homology modeling suggested that hydrogen bonding interactions of the hydroxyl groups of altered residues S103, S107, and T107 influence the regiospecificity of the oxygenase reaction.

FIG. 1.

Positions of hydroxylation of toluene, o-methoxyphenol, and cresols by wild-type T4MO and the TmoA variants. The arrows indicate the sites of hydroxylation. The resulting products and the responsible enzymes are also indicated.

FIG. 2.

Active sites of the T4MO α-subunit TmoA, showing mutations (red) at positions G103 and A107. (A) Wild-type I100, G103, and A107; (B) G103A/A107S with putative hydrogen bonds to the carbonyl of A103 or E134 (dashed green lines); (C) G103S/A107G with a putative hydrogen bond to the carbonyl of I100 (dashed green line). Yellow residues (E104, E134, H137, E197, E231, and H234) are coordinate residues anchoring the diiron-binding sites (pink spheres). The four-helix bundle of TmoA (helix B, P87-F117; helix C, P121-K150; helix E, I186-E214; and helix F, F220-Q243) anchoring the diiron active site is shown in white in panel A, and portions of the four-helix bundle are shown in white terminating at L90-G112 (helix B), N125-F144 (helix C), I190-L208 (helix E), and F220-Q236 (helix F) in panels B and C.