Glycerol as a by-product of biodiesel production is an attractive precursor for producing d-glyceric acid. Here, we demonstrate the successful production of d-glyceric acid based on glycerol via glyceraldehyde in a two-step enzyme reaction with the FAD-dependent alditol oxidase from Streptomyces coelicolor A3(2). The hydrogen peroxide generated in the reaction can be used in detergent, food, and paper industry. In order to apply the alditol oxidase in industry, the enzyme was subjected to protein engineering. Different strategies were used to enhance the substrate specificity towards glycerol. Initial attempts based on rational protein design in the active site region were found unsuccessful to increase activity. However, through directed evolution, an alditol oxidase double mutant (V125M/A244T) with 1.5-fold improved activity for glycerol was found by screening 8,000 clones. Further improvement of activity was achieved by combinatorial experiments, which led to a quadruple mutant (V125M/A244T/V133M/G399R) with 2.4-fold higher specific activity towards glycerol compared to the wild-type enzyme. Through studying the effects of mutations created, we were able to understand the importance of certain amino acids in the structure of alditol oxidase, not only for conferring enzymatic structural stability but also with respect to their influence on oxidative activity.

Chlamydiae contain a rough-type lipopolysaccharide (LPS) of 3-deoxy-alpha-d-manno-oct-2-ulopyranosonic acid residues (Kdo). Two Kdo trisaccharides, 2.8/2.4- and 2.4/2.4-linked, and a branched 2.4[2.8]2.4-linked Kdo tetrasaccharide occur in Chlamydiaceae. While the 2.8/2.4-linked trisaccharide contains a family-specific epitope, the branched Kdo oligosaccharide occurs only in Chlamydophila psittaci and antibodies against it will be useful in human and veterinarian diagnostics. To overcome the generation of cross-reactive antibodies that bind with high affinity to a dominant epitope formed by 2.4/2.4-linked Kdo, we immunized mice with a synthetic 2.4[2.8]-linked branched Kdo trisaccharide and used phage display of scFv to isolate recombinant antibody fragments (NH2240-31 and SAG506-01) that recognize the branched Kdo oligosaccharide with a K(D) of less than 10 nM. Importantly, although these antibodies used germline genes coding for an inherited Kdo recognition site, they were able clearly to distinguish between 2.4[2.8]2.4- and 2.4/2.4-linked Kdo. Sequence determination, binding data, and X-ray structural analysis revealed the basis for the improved discrimination between similar Kdo ligands and indicated that the alteration of a stacking interaction from a phenylalanine residue in the center of the combining site to a tyrosine residue facing away from the center favors recognition of branched 2.4[2.8]2.4-linked Kdo residues. Immunofluorescence tests of infected cell monolayers using this antibody show specific staining of C. psittaci elementary bodies that allow it to be distinguished from other pathogenic chlamydiae.

A difference of seven conservative amino-acid substitutions between two single-chain antibodies (scFvs) specific for chlamydial lipopolysaccharide does not significantly affect their molecular structures or packing contacts, but dramatically affects their crystallization. The structure of the variable domain (Fv) of SAG173-04 was solved to 1.86 A resolution and an R(cryst) of 18.9% in space group P2(1)2(1)2(1). Crystals of the homologous SAG506-01 diffracted to 1.95 A resolution and appeared at first to have Patterson symmetry I4/m or P4/mmm; however, no solution could be found in space groups belonging to the former and refinement in the only solution corresponding to the latter (in space group P4(3)2(1)2) stalled at R(free) = 30.0%. Detailed examination of the diffraction data revealed that the crystal was likely to be twinned and that the correct space group was P2(1)2(1)2(1). Both translational pseudo-symmetry and pseudo-merohedral twinning were observed in one crystal of SAG506-01 and pseudo-merohedral twinning was observed for a second crystal. The final R factor for SAG506-01 after refinement in P2(1)2(1)2(1) was 20.5%.

No abstract

No abstract