Engineering a bifunctional cellobiose 2-epimerase into an isomerase capable of selective lactulose production

Lactulose is listed by the World Health Organization as an essential medicine and is also a prebiotic. Its industrial production from cheap and abundant lactose suffers from low yields and expensive purification procedures. Therefore, cellobiose 2-epimerases have received attention as they can catalyze the required isomerization of lactose to lactulose. The main challenge with these enzymes is that they are much faster at catalyzing the undesired epimerization of lactose to epilactose than its isomerisation to lactulose. To develop an enzyme that is suitable for lactulose production, we initially screened a panel of 19 wild-type and previously engineered cellobiose 2-epimerases. Using a 5-fold mutant of Caldicellulosiruptor saccharolyticus epimerase as the template, we performed saturation mutagenesis on the highly conserved active site residue His247, which based on mechanistic considerations we suspected to be critical for the epimerization reaction only. Several variants indeed displayed the desired higher lactulose:epilactose product ratios. Epilactose formation was even undetectable with seven of the mutants. Under industrially relevant conditions (400 g/L, 80°C, 1 h) the best variants CsCE-5M + H247E and CsCE-5M + H247D were characterized by high lactulose yield (64%_ts and 44%_ts, respectively) and by unprecedented selectivity for lactulose (<5%_ts epilactose). These findings open the possibility of an enzymatic production process for lactulose as an active pharmaceutical ingredient.