Preparation of Magnetic Cross-Linked Amyloglucosidase Aggregates: Solving Some Activity Problems

The preparation of Cross-Linked Enzyme Aggregates (CLEAs) is a simple and cost-effective technique capable of generating insoluble biocatalysts with high volumetric activity and improved stability. The standard CLEA preparation consists of the aggregation of the enzyme and its further crosslinking, usually with glutaraldehyde. However, some enzymes have too low a content of surface lysine groups to permit effective crosslinking with glutaraldehyde, requiring co-aggregation with feeders rich in amino groups to aid the formation of CLEAs. The co-aggregation with magnetic particles makes their handling easier. In this work, CLEAs of a commercial amyloglucosidase (AMG) produced by Aspergillus niger were prepared by co-aggregation in the presence of polyethyleneimine (PEI) or starch with aminated magnetic nanoparticles (MNPs) or bovine serum albumin (BSA). First, CLEAs were prepared only with MNPs at different glutaraldehyde concentrations, yielding a recovered activity of around 20%. The addition of starch during the precipitation and crosslinking steps nearly doubled the recovered activity. Similar recovered activity (around 40%) was achieved when changing starch by PEI. Moreover, under the same conditions, AMG co-aggregated with BSA was also synthesized, yielding CLEAs with very similar recovered activity. Both CLEAs (co-aggregated with MNPs or BSA) were four times more stable than the soluble enzyme. These CLEAs were evaluated in the hydrolysis of starch at typical industrial conditions, achieving more than 95% starch-to-glucose conversion, measured as Dextrose Equivalent (DE). Moreover, both CLEAS could be reused for five cycles, maintaining a DE of around 90%. Although both CLEAs had good properties, magnetic CLEAs could be more attractive for industrial purposes because of their easy separation by an external magnetic field, avoiding the formation of clusters during the filtration or centrifugation recovery methods usually used.This research was funded in part by the Conselho Nacional de Desenvolvimento Científico e Tecnológico, grant #142107/2015-8, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Finance Code 001, and Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), grant #2016/10636-8. We gratefully recognize financial support from Project CTQ2017-86170-R (MINECO, Spain).Peer reviewe