Font Size: a A A

Multi-objective Simulation Optimization And Experimental Study Of Lysozyme Crystallization

Posted on:2023-09-20Degree:MasterType:Thesis
Country:ChinaCandidate:H Y ZhouFull Text:PDF
GTID:2531306815497484Subject:Chemical engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of bio-technology in recent years,the number of macromolecular drugs is increasing.Although the majority of biopharmaceuticals are currently marketed as certain kind of liquid proteins rather than in crystalline form,it is found that crystalline proteins have many potential advantages,such as higher bioavailability and enhanced stability.The macromolecular structures of proteins are relatively complex,and there are still many technical difficulties in the optimization and control of industrial crystallization.In practices,there usually coexist many requirements on the product quality of protein crystals.The multi-objective optimization of protein crystallization is investigated in this study using egg white lysozyme as a model protein.The morphological population balance model(MPB)is coupled with the multi-objective genetic algorithm(NSGA-II)for the solution of this kind of optimization problems.Two objective functions about crystal shape distribution and crystal size distribution(CSD)are respectively proposed,and the corresponding optimum cooling temperatures are obtained based on the analysis of Pareto frontier.It is found that the relative supersaturation of protein solutions during the crystallization process protein is much higher than that of common small molecules.The two objective functions influence each other,and the crystal shape evolves from the initial rod-like shape into the plate-like shape,resulting from the dominant effect of shape objective.With the introduction of crystal yield as the third objective,the threeobjective optimization is further investigated for protein crystallization.It is found that the results of three-objective optimization are quite different from those of two-objective optimization.In contrast to the product ration of 11.88% in two-objective optimization,it is enhanced to 59.65% for three-objective optimization.The discrepancy in the evolution of relative supersaturation leads to different crystal shapes eventually.The crystal growth is determined by the trade-off among these three different objectives,while the shape objective and yield objective have larger priority than the CSD objective.These results provide some guidelines to industrial protein crystallization to some extent.Finally,the protein crystallization is performed by experiments using 96-well and 24-well crystallization plates,and crystals with maximum particle size of 800 μm are obtained.The phenomenon of helical dislocation is found by high magnification microscopy,which is briefly analyzed as a result of the continuous growth of crystals from protein solution at lower concentration.According to the screening results of the 24-well crystal plate,the crystal was amplified to a 50 m L jacket reactor for highconcentration amplification crystallization,and two batches of crystals with different particle sizes were obtained by stirring and unstirring.It is found that the crystal size obtained under mechanical stirring condition is relatively uniform and no twin phenomenon occurs.
Keywords/Search Tags:Multi-objective optimization, morphological population balance models, genetic algorithms, protein purification, industrial crystallization
PDF Full Text Request
Related items