Anke C. Terwisscha van Scheltinga

Structure of a cephalosporin synthase

Penicillins and cephalosporins are among the most widely used therapeutic agents. These antibiotics are produced from fermentation-derived materials as their chemical synthesis is not commercially viable. Unconventional steps in their biosynthesis are catalysed by Fe(II)-dependent oxidases/oxygenases; isopenicillin N synthase (IPNS), creates in one step the bicyclic nucleus of penicillins, and deacetoxycephalosporin C synthase (DAOCS) catalyses the expansion of the penicillin nucleus into the nucleus of cephalosporins. Both enzymes use dioxygen-derived ferryl intermediates in catalysis but, in contrast to IPNS, the ferryl form of DAOCS is produced by the oxidative splitting of a co-substrate, 2-oxoglutarate (α-ketoglutarate). This route of controlled ferryl formation and reaction is common to many mononuclear ferrous enzymes, which participate in a broader range of reactions than their well-characterized counterparts, the haem enzymes. Here we report the first crystal structure of a 2-oxoacid-dependent oxygenase. High-resolution structures for apo-DAOCS, the enzyme complexed with Fe(II), and with Fe(II) and 2-oxoglutarate, were obtained from merohedrally twinned crystals. Using a model based on these structures, we propose a mechanism for ferryl formation.

The Structural Basis of Cephalosporin Formation in a Mononuclear Ferrous Enzyme

Deacetoxycephalosporin-C synthase (DAOCS) is a mononuclear ferrous enzyme that transforms penicillins into cephalosporins by inserting a carbon atom into the penicillin nucleus. In the first half-reaction, dioxygen and 2-oxoglutarate produce a reactive iron-oxygen species, succinate and CO2. The oxidizing iron species subsequently reacts with penicillin to give cephalosporin and water. Here we describe high-resolution structures for ferrous DAOCS in complex with penicillins, the cephalosporin product, the cosubstrate and the coproduct. Steady-state kinetic data, quantum-chemical calculations and the new structures indicate a reaction sequence in which a booby-trapped oxidizing species is formed. This species is stabilized by the negative charge of succinate on the iron. The binding sites of succinate and penicillin overlap, and when penicillin replaces succinate, it removes the stabilizing charge, eliciting oxidative attack on itself. Requisite groups of penicillin are within 1 Å of the expected position of a ferryl oxygen in the enzyme–penicillin complex.

Structural Basis of the Substrate Range and Enantioselectivity of Two (S)-Selective ω-Transaminases

ω-Transaminases are enzymes that can introduce an amino group in industrially interesting compounds. We determined crystal structures of two (S)-selective ω-transaminases, one from Arthrobacter sp. (Ars-ωTA) and one from Bacillus megaterium (BM-ωTA), which have 95% identical sequences but somewhat different activity profiles. Substrate profiling measurements using a range of (R)- and (S)-substrates showed that both enzymes have a preference for substrates with large, flat cyclic side groups, for which the activity of BM-ωTA is generally somewhat higher. BM-ωTA has a preference for (S)-3,3-dimethyl-2-butylamine significantly stronger than that of Ars-ωTA, as well as a weaker enantiopreference for 1-cyclopropylethylamine. The crystal structures showed that, as expected for (S)-selective transaminases, both enzymes have the typical transaminase type I fold and have spacious active sites to accommodate largish substrates. A structure of BM-ωTA with bound (R)-α-methylbenzylamine explains the enzymes preference for (S)-substrates. Site-directed mutagenesis experiments revealed that the presence of a tyrosine, instead of a cysteine, at position 60 increases the relative activities on several small substrates. A structure of Ars-ωTA with bound l-Ala revealed that the Arg442 side chain has been repositioned to bind the l-Ala carboxylate. Compared to the arginine switch residue in other transaminases, Arg442 is shifted by six residues in the amino acid sequence, which appears to be a consequence of extra loops near the active site that narrow the entrance to the active site.

MIR phasing using merohedrally twinned crystals

Merohedral twinning is a crystal-growth disorder that seriously hinders the determination of macromolecular crystal structures by isomorphous replacement. The strategies used in the structures solved so far are discussed. Several methods can be used to determine the extent of twinning, the twin fraction and to detwin the data. Accurate determination of the twin fraction by analysing heavy-atom refinement statistics is possible, but only influences the resulting phases slightly. It seems more crucial to restrict the variation in twin fractions between data sets, either by making the twin fractions of some data sets artificially higher or by screening crystals to obtain data with a low twin fraction.