Molecular basis for hydroxylation during the biosynthesis of phosphonic acid antibiotics

Oxygenases have evolved to selectively activate atmospheric O₂ and use its oxidizing power for key biological processes. The non-heme iron /α-ketoglutarate (αKG) dependent dioxygenases constitute a group of enzymes that catalyze a wide range of oxidative reactions including hydroxylation, desaturation, ring closure as well as other reactions. The dioxygenase active site contains an Fe(II) atom that activates dioxygen and transfers one of the oxygens to α-ketogluterate (αKG) and the other to substrate. Independent of the reaction that they catalyze, all non-heme Fe(II)/αKG dependent dioxygenases share common structural features including a six-stranded b-sandwich core that anchors the iron, and surrounding flexible regions that provide the residues for binding of the substrate. The dioxygenases Orf10 and FzmG catalyze the α-hydroxylation of intermediates in the biosynthesis of phosphonate (C-P containing) natural products. The former hydroxylates 2-aminoethylphosphonate (2AEP), and the latter methyl-phosphonoacetate (MePnA), two compounds that share the phosphonic moiety, but differ at the other half of their structure. To identify the molecular basis of substrate selectivity and specificity, we determined the crystal structures of Orf10 complexed with Mn(II) and an analog of αKG, and of FzmG complexed with Mn(II), αKG and MePnA. The two enzymes share the same fold with an RMSD of 1.1 Å, and superposition of their structures provides insight into the preference of FzmG for MePnA and not 2AEP.