In anion selectrodialysis (aSED) multivalent anions are entrapped in the so called "product stream" between a standard anion exchange membrane (SA membrane) and a monovalent anion selective membrane (MVA membrane). This study was carried out on nitrified and ultra-filtrated effluent of an upflow anaerobic sludge blanket reactor (UASB) of a potato processor. The selectivity for phosphate of various membranes that could act as SA membranes was tested on lab scale. From the calculated selectivities and current efficiencies for phosphate transport two membranes, the Fujifilm Type 1 membrane and the PC-Acid 100 OT membrane from PCA are good candidates to replace the previously used PC-SA membrane from PCA. Lowering the pH had a negative effect on the mobility of the phosphate ions and it is proposed that the HPO4 2is the ion that migrates through the SA membrane. Increasing the voltage resulted in a better phosphate mobility through the PC-Acid 100 OT membrane. Also, on pilot scale the PC-Acid 100 OT membrane performed much better than the PC-SA membrane: the current efficiency for phosphate transport was higher whereas the current efficiency for DIC transport was lower with the PC-Acid 100 OT membrane than with the PC-SA membrane. The PC ACID 100 OT is recommended as standard anion exchange membrane in pilot and full scale stacks for recovery of phosphate by aSED.

The Glycoside Hydrolase Family 65 (GH65) is an enzyme family of inverting a-glucoside phosphorylases and hydrolases that currently contains 10 characterized enzyme specificities. However, its sequence diversity has never been studied in detail. Here, an in-silico analysis of correlated mutations was performed, revealing specificity-determining positions that facilitate annotation of the family’s phylogenetic tree. By searching these positions for amino acid motifs that do not match those found in previously characterized enzymes from GH65, several clades that may harbor new functions could be identified. Three enzymes from across these regions were expressed in E. coli and their substrate profile was mapped. One of those enzymes, originating from the bacterium Mucilaginibacter mallensis, was found to hydrolyze kojibiose and a-1,2-oligoglucans with high specificity. We propose kojibiose glucohydrolase as the systematic name and kojibiose hydrolase or kojibiase as the short name for this new enzyme. This work illustrates a convenient strategy for mapping the natural diversity of enzyme families and smartly mining the ever-growing number of available sequences in the quest for novel specificities.