reveal structural differences of the ob- tained LbADH variants and help to explain varying crystallization behavior on an atomistic level. Protein engineering strategies to im- prove protein crystallization are discussed from both experimental and theoretical perspectives. Latest experimental results of engineered LbADH variants with dis- tinct crystallization kinetics are presented and correlations to MD simulations are drawn. V10.07 Multiscale modeling to investigate catalytically active enzymatic aggregates for cascade bioreactions P. N. Depta 1) (E-Mail: nicolas.depta@tuhh.de), Dr. U. Jandt 2) , S. Ilhan 2) , C. Mu ¨ ller 2) , Prof. Dr. M. Dosta 1) , Prof. Dr. A.-P. Zeng 2) , Prof. Dr. S. Heinrich 1) 1) Hamburg University of Technology, Institute of Solids Process Engineering and Particle Technology, Denickestraße 15, 21073 Hamburg, Germany 2) Hamburg University of Technology, Institute of Bioprocess and Biosystems Engineering, Denickestraße 15, 21073 Hamburg, Germany DOI: 10.1002/cite.201855443 The dynamic formation and stability of catalytically active enzymatic agglomerates for cascade bioreactions has received in- creasing interest in recent years. However, these formation processes, even though experimentally observed, are not well un- derstood. Understanding these processes is nevertheless critical to enable targeted modification and possibly de novo crea- tion of bioreaction cascades. A multi-scale modeling methodology in combination with experimental validation was developed. As it can be seen in the fig- ure, multiple methods are combined and scales from atomistic (< O(10 1 ) MDa) are linked to the micrometer scale for model- ing agglomerate systems (~ O(10 5 ) MDa). The models are parameterized ‘‘bottom- up’’ and validated ‘‘top-down’’ by compari- son with experimental data, which is ob- tained from BLI, DLS, and activity assays. Components of the pyruvate dehydrogen- ase complex (PDC) are used as model enzymes, which incorporates many pro- cesses of interest. First results for single PDC components show that the dynamic formation and breakup of agglomerates can be predicted using a novel DEM diffusion model in combination with interaction forces de- rived from MD without experimental data fitting – yielding accurate scale-bridging diffusion kinetics and agglomerate sizes matching corresponding DLS data. Financial support by the DFG (SPP 1934) is gratefully acknowledged. V10.08 Struktur-Eigenschaftsbeziehung bei der Aufarbeitung und Formulierung von Proteinpartikeln M. Kubiak 1,2) (E-Mail: marta.kubiak@tu-bs.de), J. Solarczek 3) , J. Mayer 4) , Dr. I. Kampen 1,2) , Dr. R. Biedendieck 4) , Prof. Dr. A. Schallmey 3) , Prof. Dr. C. Schilde 1,2) 1) Technische Universita ¨ t Braunschweig, Institut fu ¨ r Partikeltechnik, Volkmaroder Straße 5, 38104 Braunschweig, Deutschland 2) Technische Universita ¨ t Braunschweig, Zentrum fu ¨ r Pharmaverfahrenstechnik, Franz-Liszt-Straße 35 A, 38106 Braunschweig, Deutschland 3) Technische Universita ¨ t Braunschweig, Institut fu ¨ r Biochemie, Spielmannstraße 7, 38106 Braunschweig, Deutschland 4) Technische Universita ¨ t Braunschweig, Institut fu ¨ r Mikrobiologie, Spielmannstraße 7, 38106 Braunschweig, Deutschland DOI: 10.1002/cite.201855444 www.cit-journal.com ª 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Ing. Tech. 2018, 90, No. 9, 1335–1343 Figure. Methodological scheme of the modeling-based engineering approach to modify and analyze protein crystallization. Figure. Multiscale modeling methodology. 1338 10 DFG SPP DiSPBiotech Chemie Ingenieur Technik